From 98d6e1b4ea4483baed5a3606de26441a96e86948 Mon Sep 17 00:00:00 2001 From: Marc Fiuczynski Date: Thu, 19 Oct 2006 03:31:43 +0000 Subject: [PATCH] Merge to Fedora kernel-2.6.17-1.2187_FC5-vs2.0.2.1 patched with stable patch-2.6.17.13-vs2.0.2.1 --- Documentation/networking/netdevices.txt | 8 +- MAINTAINERS | 8 + Makefile | 2 +- arch/i386/Kconfig | 95 +- arch/i386/Kconfig.cpu | 2 +- arch/i386/Kconfig.debug | 1 + arch/i386/Makefile | 24 + arch/i386/kernel/Makefile | 22 +- arch/i386/kernel/acpi/Makefile | 4 + arch/i386/kernel/asm-offsets.c | 9 +- arch/i386/kernel/cpu/Makefile | 5 + arch/i386/kernel/cpu/cpufreq/powernow-k7.c | 1 - arch/i386/kernel/cpu/cpufreq/powernow-k8.c | 2 +- arch/i386/kernel/cpu/mtrr/Makefile | 7 + arch/i386/kernel/entry.S | 10 +- arch/i386/kernel/irq.c | 4 +- arch/i386/kernel/reboot.c | 8 - arch/i386/kernel/sysenter.c | 32 +- arch/i386/kernel/traps.c | 9 +- arch/i386/kernel/vm86.c | 12 + arch/i386/mm/Makefile | 8 + arch/i386/mm/pageattr.c | 2 +- arch/i386/mm/pgtable.c | 12 + arch/i386/oprofile/Makefile | 4 + arch/i386/pci/Makefile | 9 + arch/i386/pci/mmconfig.c | 9 +- arch/i386/power/Makefile | 4 +- arch/ia64/Kconfig | 3 + arch/ia64/kernel/sys_ia64.c | 28 +- arch/ia64/sn/kernel/xpc_channel.c | 493 +- arch/ia64/sn/kernel/xpc_main.c | 28 +- arch/ia64/sn/kernel/xpc_partition.c | 527 +- arch/powerpc/Kconfig | 3 + arch/powerpc/kernel/cpu_setup_power4.S | 2 + arch/powerpc/kernel/crash.c | 13 + arch/powerpc/platforms/pseries/xics.c | 25 +- arch/sparc/kernel/sys_sparc.c | 27 +- arch/sparc64/kernel/sys_sparc.c | 36 +- arch/sparc64/mm/generic.c | 2 + arch/um/kernel/physmem.c | 4 +- arch/x86_64/Kconfig | 64 +- arch/x86_64/Makefile | 20 + arch/x86_64/ia32/Makefile | 20 +- arch/x86_64/ia32/vsyscall-sigreturn.S | 2 +- arch/x86_64/kernel/Makefile | 19 +- arch/x86_64/kernel/acpi/Makefile | 1 + arch/x86_64/kernel/asm-offsets.c | 2 + arch/x86_64/kernel/entry.S | 2 +- arch/x86_64/kernel/init_task.c | 3 + arch/x86_64/kernel/irq.c | 4 +- arch/x86_64/kernel/pmtimer.c | 32 +- arch/x86_64/kernel/setup.c | 2 +- arch/x86_64/kernel/setup64.c | 23 +- arch/x86_64/kernel/smp.c | 4 +- arch/x86_64/mm/Makefile | 10 + arch/x86_64/oprofile/Makefile | 5 +- arch/x86_64/pci/Makefile | 12 + arch/x86_64/pci/mmconfig.c | 13 +- block/elevator.c | 3 +- block/ll_rw_blk.c | 2 +- configs/kernel-2.6.17-i686-planetlab.config | 10 +- .../kernel-2.6.17-i686-smp-planetlab.config | 10 +- drivers/Makefile | 1 + drivers/acpi/Kconfig | 3 +- drivers/acpi/namespace/nsxfeval.c | 5 +- drivers/block/aoe/aoenet.c | 23 +- drivers/cdrom/cdrom.c | 2 +- drivers/char/mem.c | 38 +- drivers/char/tpm/tpm_tis.c | 1 + drivers/char/tty_io.c | 34 +- drivers/firmware/Kconfig | 2 +- drivers/i2c/busses/scx200_acb.c | 20 +- drivers/i2c/i2c-core.c | 4 +- drivers/ide/ide-lib.c | 8 +- drivers/ide/pci/via82cxxx.c | 3 +- drivers/ieee1394/ohci1394.c | 4 +- drivers/ieee1394/sbp2.c | 3 + drivers/infiniband/hw/mthca/mthca_reset.c | 68 +- .../infiniband/ulp/ipoib/ipoib_multicast.c | 280 +- drivers/md/dm-exception-store.c | 72 +- drivers/md/dm-ioctl.c | 34 +- drivers/md/dm-mpath.c | 99 +- drivers/md/dm-raid1.c | 140 +- drivers/md/dm-snap.c | 460 +- drivers/md/dm.c | 101 +- drivers/md/raid1.c | 4 +- drivers/media/dvb/bt8xx/dvb-bt8xx.c | 10 + drivers/media/dvb/bt8xx/dvb-bt8xx.h | 1 + drivers/media/dvb/dvb-core/dvb_frontend.c | 14 +- drivers/media/dvb/dvb-core/dvb_net.c | 4 +- drivers/media/dvb/frontends/dvb-pll.c | 333 +- drivers/media/dvb/ttpci/budget-av.c | 24 +- drivers/media/dvb/ttpci/budget.c | 6 - drivers/media/video/Kconfig | 6 +- drivers/media/video/stradis.c | 1 - drivers/net/8139cp.c | 2 +- drivers/net/bnx2.c | 1645 +++-- drivers/net/bonding/bond_main.c | 9 +- drivers/net/chelsio/sge.c | 4 +- drivers/net/e1000/e1000_hw.c | 1 + drivers/net/e1000/e1000_hw.h | 1 + drivers/net/e1000/e1000_main.c | 11 +- drivers/net/forcedeth.c | 26 +- drivers/net/hamradio/6pack.c | 8 +- drivers/net/hamradio/mkiss.c | 8 +- drivers/net/ifb.c | 10 +- drivers/net/irda/vlsi_ir.c | 2 +- drivers/net/ixgb/ixgb_main.c | 4 +- drivers/net/loopback.c | 4 +- drivers/net/mv643xx_eth.c | 2 +- drivers/net/natsemi.c | 4 +- drivers/net/r8169.c | 2 +- drivers/net/s2io.c | 16 +- drivers/net/sky2.c | 114 +- drivers/net/sky2.h | 17 +- drivers/net/tg3.c | 16 +- drivers/net/tg3.h | 1 + drivers/net/tulip/winbond-840.c | 9 +- drivers/net/typhoon.c | 6 +- drivers/net/via-velocity.c | 34 +- drivers/net/wireless/orinoco.c | 4 +- drivers/net/wireless/spectrum_cs.c | 2 +- drivers/oprofile/buffer_sync.c | 66 +- drivers/oprofile/cpu_buffer.c | 51 +- drivers/oprofile/cpu_buffer.h | 9 +- drivers/oprofile/event_buffer.h | 7 +- drivers/oprofile/oprof.c | 34 +- drivers/oprofile/oprof.h | 7 +- drivers/oprofile/oprofile_files.c | 205 +- drivers/pci/Kconfig | 1 + drivers/pci/quirks.c | 2 +- drivers/pnp/resource.c | 3 +- drivers/s390/net/qeth_eddp.c | 59 +- drivers/s390/net/qeth_main.c | 8 +- drivers/s390/net/qeth_tso.h | 12 +- drivers/scsi/Makefile | 2 +- drivers/scsi/iscsi_tcp.c | 3006 ++++++--- drivers/scsi/iscsi_tcp.h | 190 +- drivers/scsi/libiscsi.c | 1702 ----- drivers/scsi/scsi_ioctl.c | 5 +- drivers/scsi/scsi_transport_iscsi.c | 1403 +++- drivers/serial/8250.c | 42 +- drivers/serial/Kconfig | 2 + drivers/serial/serial_core.c | 12 +- drivers/usb/host/uhci-q.c | 1366 ++-- drivers/usb/serial/ftdi_sio.c | 84 +- drivers/usb/storage/unusual_devs.h | 12 +- drivers/usb/storage/usb.c | 18 +- drivers/video/Kconfig | 2 +- drivers/video/console/softcursor.c | 6 +- drivers/xen/Makefile | 17 + drivers/xen/balloon/Makefile | 2 + drivers/xen/balloon/balloon.c | 617 ++ drivers/xen/blkback/Makefile | 3 + drivers/xen/blkback/blkback.c | 566 ++ drivers/xen/blkback/common.h | 134 + drivers/xen/blkback/interface.c | 171 + drivers/xen/blkback/vbd.c | 119 + drivers/xen/blkfront/Kconfig | 6 + drivers/xen/blkfront/Makefile | 5 + drivers/xen/blkfront/blkfront.c | 841 +++ drivers/xen/blkfront/block.h | 156 + drivers/xen/blkfront/vbd.c | 318 + drivers/xen/blktap/Makefile | 3 + drivers/xen/blktap/blktap.c | 1450 +++++ drivers/xen/console/Makefile | 2 + drivers/xen/console/console.c | 644 ++ drivers/xen/evtchn/Makefile | 2 + drivers/xen/evtchn/evtchn.c | 458 ++ drivers/xen/netback/Makefile | 5 + drivers/xen/netback/common.h | 138 + drivers/xen/netback/interface.c | 339 + drivers/xen/netback/netback.c | 1345 ++++ drivers/xen/netfront/Kconfig | 6 + drivers/xen/netfront/Makefile | 4 + drivers/xen/netfront/netfront.c | 1841 ++++++ drivers/xen/privcmd/Makefile | 2 + drivers/xen/privcmd/privcmd.c | 286 + fs/Kconfig | 1 + fs/befs/linuxvfs.c | 11 +- fs/binfmt_elf.c | 15 +- fs/buffer.c | 7 +- fs/ext2/super.c | 41 + fs/ext3/inode.c | 19 +- fs/ext3/namei.c | 15 +- fs/ext3/super.c | 40 + fs/file.c | 4 +- fs/locks.c | 6 +- fs/namei.c | 8 +- fs/proc/base.c | 35 +- fs/splice.c | 230 +- fs/udf/super.c | 2 +- fs/udf/truncate.c | 64 +- fs/xfs/xfs_dir2_node.c | 2 +- include/asm-generic/mman.h | 6 + include/asm-i386/apic.h | 2 + include/asm-i386/desc.h | 2 +- include/asm-i386/fixmap.h | 4 +- include/asm-i386/mach-default/mach_traps.h | 12 + include/asm-i386/page.h | 2 +- include/asm-i386/pgtable-2level-defs.h | 2 + include/asm-i386/pgtable-3level-defs.h | 2 + include/asm-i386/spinlock.h | 2 +- include/asm-ia64/mman.h | 8 + include/asm-ia64/sn/xp.h | 44 +- include/asm-ia64/sn/xpc.h | 4 +- include/asm-s390/futex.h | 5 +- include/asm-sparc/mman.h | 6 + include/asm-sparc64/mman.h | 6 + include/asm-sparc64/sfp-machine.h | 2 +- include/asm-um/page.h | 2 +- include/asm-x86_64/apic.h | 2 + include/asm-x86_64/hw_irq.h | 2 +- include/asm-x86_64/ipi.h | 2 + include/asm-x86_64/pgtable.h | 2 +- include/linux/ethtool.h | 2 + include/linux/ext3_fs.h | 9 + include/linux/gfp.h | 6 +- include/linux/highmem.h | 6 + include/linux/idr.h | 1 + include/linux/interrupt.h | 6 + include/linux/mm.h | 10 + include/linux/netdevice.h | 74 +- include/linux/netfilter_bridge.h | 16 +- include/linux/oprofile.h | 11 + include/linux/pci_ids.h | 5 +- include/linux/pfn.h | 2 +- include/linux/skbuff.h | 97 +- include/linux/tty.h | 2 - include/net/pkt_sched.h | 7 +- include/net/protocol.h | 3 + include/net/sctp/sctp.h | 13 - include/net/sctp/sm.h | 3 +- include/net/sock.h | 4 + include/net/tcp.h | 7 +- include/scsi/iscsi_if.h | 128 +- include/scsi/iscsi_proto.h | 2 + include/scsi/libiscsi.h | 282 - include/scsi/scsi_transport_iscsi.h | 251 +- include/sound/initval.h | 3 +- kernel/Kconfig.preempt | 1 + kernel/fork.c | 3 + kernel/futex.c | 1 + kernel/irq/manage.c | 26 +- kernel/irq/spurious.c | 3 +- kernel/sched.c | 25 +- kernel/stop_machine.c | 1 - kernel/timer.c | 60 +- lib/Makefile | 3 + lib/idr.c | 43 + lib/spinlock_debug.c | 10 +- lib/ts_bm.c | 11 +- mm/Kconfig | 8 +- mm/filemap.c | 34 +- mm/filemap.h | 4 +- mm/highmem.c | 11 + mm/memory.c | 119 +- mm/mmap.c | 17 +- mm/page_alloc.c | 11 +- mm/pdflush.c | 15 +- mm/swapfile.c | 3 +- net/8021q/vlan.c | 8 +- net/atm/clip.c | 8 +- net/bridge/br_device.c | 8 +- net/bridge/br_forward.c | 12 +- net/bridge/br_if.c | 23 +- net/bridge/br_netfilter.c | 7 +- net/bridge/netfilter/ebt_ulog.c | 3 + net/core/dev.c | 289 +- net/core/dev_mcast.c | 28 +- net/core/dst.c | 3 +- net/core/ethtool.c | 39 +- net/core/netpoll.c | 9 +- net/core/pktgen.c | 12 +- net/core/rtnetlink.c | 15 +- net/core/skbuff.c | 285 +- net/decnet/dn_nsp_in.c | 3 +- net/decnet/dn_route.c | 3 +- net/decnet/dn_rules.c | 3 +- net/ieee80211/Kconfig | 1 + net/ipv4/af_inet.c | 87 + net/ipv4/fib_rules.c | 4 +- net/ipv4/fib_semantics.c | 12 +- net/ipv4/ip_output.c | 22 +- net/ipv4/ipcomp.c | 11 +- net/ipv4/netfilter/arp_tables.c | 3 +- net/ipv4/netfilter/ip_conntrack_helper_h323.c | 2 +- net/ipv4/netfilter/ip_nat_proto_tcp.c | 7 +- net/ipv4/netfilter/ip_nat_proto_udp.c | 10 +- net/ipv4/netfilter/ip_tables.c | 3 +- net/ipv4/netfilter/ipt_ULOG.c | 5 + net/ipv4/route.c | 2 +- net/ipv4/tcp.c | 86 +- net/ipv4/tcp_input.c | 2 +- net/ipv4/tcp_ipv4.c | 18 + net/ipv4/tcp_output.c | 48 +- net/ipv4/xfrm4_output.c | 133 +- net/ipv6/addrconf.c | 2 + net/ipv6/exthdrs.c | 29 +- net/ipv6/ip6_output.c | 9 +- net/ipv6/ipcomp6.c | 11 +- net/ipv6/xfrm6_output.c | 111 +- net/ipx/af_ipx.c | 3 +- net/netfilter/nfnetlink_log.c | 3 + net/sched/act_api.c | 60 +- net/sched/sch_generic.c | 66 +- net/sched/sch_teql.c | 9 +- net/sctp/sm_make_chunk.c | 30 +- net/sctp/sm_statefuns.c | 20 +- net/sctp/socket.c | 20 +- net/sunrpc/cache.c | 6 +- sound/core/oss/pcm_oss.c | 2 + sound/core/timer.c | 5 +- sound/isa/cs423x/Makefile | 1 + sound/pci/Kconfig | 14 +- sound/pci/au88x0/au88x0_mpu401.c | 2 +- sound/pci/fm801.c | 2 +- sound/pci/hda/hda_intel.c | 605 +- sound/pci/hda/patch_analog.c | 2536 +++++++- sound/pci/hda/patch_realtek.c | 5761 ++++++++++++++--- sound/pci/hda/patch_sigmatel.c | 8 +- sound/pci/rme9652/hdsp.c | 5 +- 322 files changed, 27687 insertions(+), 8158 deletions(-) delete mode 100644 drivers/scsi/libiscsi.c create mode 100644 drivers/xen/Makefile create mode 100644 drivers/xen/balloon/Makefile create mode 100644 drivers/xen/balloon/balloon.c create mode 100644 drivers/xen/blkback/Makefile create mode 100644 drivers/xen/blkback/blkback.c create mode 100644 drivers/xen/blkback/common.h create mode 100644 drivers/xen/blkback/interface.c create mode 100644 drivers/xen/blkback/vbd.c create mode 100644 drivers/xen/blkfront/Kconfig create mode 100644 drivers/xen/blkfront/Makefile create mode 100644 drivers/xen/blkfront/blkfront.c create mode 100644 drivers/xen/blkfront/block.h create mode 100644 drivers/xen/blkfront/vbd.c create mode 100644 drivers/xen/blktap/Makefile create mode 100644 drivers/xen/blktap/blktap.c create mode 100644 drivers/xen/console/Makefile create mode 100644 drivers/xen/console/console.c create mode 100644 drivers/xen/evtchn/Makefile create mode 100644 drivers/xen/evtchn/evtchn.c create mode 100644 drivers/xen/netback/Makefile create mode 100644 drivers/xen/netback/common.h create mode 100644 drivers/xen/netback/interface.c create mode 100644 drivers/xen/netback/netback.c create mode 100644 drivers/xen/netfront/Kconfig create mode 100644 drivers/xen/netfront/Makefile create mode 100644 drivers/xen/netfront/netfront.c create mode 100644 drivers/xen/privcmd/Makefile create mode 100644 drivers/xen/privcmd/privcmd.c delete mode 100644 include/scsi/libiscsi.h diff --git a/Documentation/networking/netdevices.txt b/Documentation/networking/netdevices.txt index 3c0a5ba61..847cedb23 100644 --- a/Documentation/networking/netdevices.txt +++ b/Documentation/networking/netdevices.txt @@ -42,9 +42,9 @@ dev->get_stats: Context: nominally process, but don't sleep inside an rwlock dev->hard_start_xmit: - Synchronization: dev->xmit_lock spinlock. + Synchronization: netif_tx_lock spinlock. When the driver sets NETIF_F_LLTX in dev->features this will be - called without holding xmit_lock. In this case the driver + called without holding netif_tx_lock. In this case the driver has to lock by itself when needed. It is recommended to use a try lock for this and return -1 when the spin lock fails. The locking there should also properly protect against @@ -62,12 +62,12 @@ dev->hard_start_xmit: Only valid when NETIF_F_LLTX is set. dev->tx_timeout: - Synchronization: dev->xmit_lock spinlock. + Synchronization: netif_tx_lock spinlock. Context: BHs disabled Notes: netif_queue_stopped() is guaranteed true dev->set_multicast_list: - Synchronization: dev->xmit_lock spinlock. + Synchronization: netif_tx_lock spinlock. Context: BHs disabled dev->poll: diff --git a/MAINTAINERS b/MAINTAINERS index c3c584240..6a1bb87d8 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -2572,6 +2572,14 @@ M: dbrownell@users.sourceforge.net L: spi-devel-general@lists.sourceforge.net S: Maintained +STABLE BRANCH: +P: Greg Kroah-Hartman +M: greg@kroah.com +P: Chris Wright +M: chrisw@sous-sol.org +L: stable@kernel.org +S: Maintained + TPM DEVICE DRIVER P: Kylene Hall M: kjhall@us.ibm.com diff --git a/Makefile b/Makefile index 4b9e53d83..bfefdc753 100644 --- a/Makefile +++ b/Makefile @@ -1,7 +1,7 @@ VERSION = 2 PATCHLEVEL = 6 SUBLEVEL = 17 -EXTRAVERSION = -1.2142_FC4.4 +EXTRAVERSION = -1.2142_FC4.3smp NAME=Crazed Snow-Weasel # *DOCUMENTATION* diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig index 03f53c27e..d75fa4047 100644 --- a/arch/i386/Kconfig +++ b/arch/i386/Kconfig @@ -91,6 +91,15 @@ config X86_PC help Choose this option if your computer is a standard PC or compatible. +config X86_XEN + bool "Xen-compatible" + select X86_UP_APIC if !SMP && XEN_PRIVILEGED_GUEST + select X86_UP_IOAPIC if !SMP && XEN_PRIVILEGED_GUEST + select SWIOTLB + help + Choose this option if you plan to run this kernel on top of the + Xen Hypervisor. + config X86_ELAN bool "AMD Elan" help @@ -193,6 +202,7 @@ source "arch/i386/Kconfig.cpu" config HPET_TIMER bool "HPET Timer Support" + depends on !X86_XEN help This enables the use of the HPET for the kernel's internal timer. HPET is the next generation timer replacing legacy 8254s. @@ -223,7 +233,7 @@ config NR_CPUS config SCHED_SMT bool "SMT (Hyperthreading) scheduler support" - depends on SMP + depends on SMP && !X86_XEN default off help SMT scheduler support improves the CPU scheduler's decision making @@ -233,7 +243,7 @@ config SCHED_SMT config SCHED_MC bool "Multi-core scheduler support" - depends on SMP + depends on SMP && !X86_XEN default y help Multi-core scheduler support improves the CPU scheduler's decision @@ -244,7 +254,7 @@ source "kernel/Kconfig.preempt" config X86_UP_APIC bool "Local APIC support on uniprocessors" - depends on !SMP && !(X86_VISWS || X86_VOYAGER) + depends on !SMP && !(X86_VISWS || X86_VOYAGER || XEN_UNPRIVILEGED_GUEST) help A local APIC (Advanced Programmable Interrupt Controller) is an integrated interrupt controller in the CPU. If you have a single-CPU @@ -269,12 +279,12 @@ config X86_UP_IOAPIC config X86_LOCAL_APIC bool - depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER) + depends on X86_UP_APIC || ((X86_VISWS || SMP) && !(X86_VOYAGER || XEN_UNPRIVILEGED_GUEST)) default y config X86_IO_APIC bool - depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER)) + depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER || XEN_UNPRIVILEGED_GUEST)) default y config X86_VISWS_APIC @@ -282,9 +292,14 @@ config X86_VISWS_APIC depends on X86_VISWS default y +config X86_TSC + bool + depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODEGX1) && !X86_NUMAQ && !X86_XEN + default y + config X86_MCE bool "Machine Check Exception" - depends on !X86_VOYAGER + depends on !(X86_VOYAGER || X86_XEN) ---help--- Machine Check Exception support allows the processor to notify the kernel if it detects a problem (e.g. overheating, component failure). @@ -374,6 +389,7 @@ config X86_REBOOTFIXUPS config MICROCODE tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support" + depends on !XEN_UNPRIVILEGED_GUEST ---help--- If you say Y here and also to "/dev file system support" in the 'File systems' section, you will be able to update the microcode on @@ -391,6 +407,7 @@ config MICROCODE config X86_MSR tristate "/dev/cpu/*/msr - Model-specific register support" + depends on !X86_XEN help This device gives privileged processes access to the x86 Model-Specific Registers (MSRs). It is a character device with @@ -406,6 +423,10 @@ config X86_CPUID with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to /dev/cpu/31/cpuid. +config SWIOTLB + bool + default n + source "drivers/firmware/Kconfig" choice @@ -578,7 +599,7 @@ config HAVE_ARCH_EARLY_PFN_TO_NID config HIGHPTE bool "Allocate 3rd-level pagetables from highmem" - depends on HIGHMEM4G || HIGHMEM64G + depends on (HIGHMEM4G || HIGHMEM64G) && !X86_XEN help The VM uses one page table entry for each page of physical memory. For systems with a lot of RAM, this can be wasteful of precious @@ -587,6 +608,7 @@ config HIGHPTE config MATH_EMULATION bool "Math emulation" + depends on !X86_XEN ---help--- Linux can emulate a math coprocessor (used for floating point operations) if you don't have one. 486DX and Pentium processors have @@ -612,6 +634,8 @@ config MATH_EMULATION config MTRR bool "MTRR (Memory Type Range Register) support" + depends on !XEN_UNPRIVILEGED_GUEST + default y if X86_XEN ---help--- On Intel P6 family processors (Pentium Pro, Pentium II and later) the Memory Type Range Registers (MTRRs) may be used to control @@ -646,7 +670,7 @@ config MTRR config EFI bool "Boot from EFI support (EXPERIMENTAL)" - depends on ACPI + depends on ACPI && !X86_XEN default n ---help--- This enables the the kernel to boot on EFI platforms using @@ -664,7 +688,7 @@ config EFI config IRQBALANCE bool "Enable kernel irq balancing" - depends on SMP && X86_IO_APIC + depends on SMP && X86_IO_APIC && !X86_XEN default y help The default yes will allow the kernel to do irq load balancing. @@ -712,7 +736,7 @@ source kernel/Kconfig.hz config KEXEC bool "kexec system call (EXPERIMENTAL)" - depends on EXPERIMENTAL + depends on EXPERIMENTAL && !X86_XEN help kexec is a system call that implements the ability to shutdown your current kernel, and to start another kernel. It is like a reboot @@ -765,20 +789,25 @@ config HOTPLUG_CPU endmenu +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y + depends on HIGHMEM menu "Power management options (ACPI, APM)" - depends on !X86_VOYAGER + depends on !(X86_VOYAGER || XEN_UNPRIVILEGED_GUEST) +if !X86_XEN source kernel/power/Kconfig +endif source "drivers/acpi/Kconfig" menu "APM (Advanced Power Management) BIOS Support" -depends on PM && !X86_VISWS +depends on PM && !(X86_VISWS || X86_XEN) config APM tristate "APM (Advanced Power Management) BIOS support" - depends on PM + depends on PM && PM_LEGACY ---help--- APM is a BIOS specification for saving power using several different techniques. This is mostly useful for battery powered laptops with @@ -963,6 +992,7 @@ choice config PCI_GOBIOS bool "BIOS" + depends on !X86_XEN config PCI_GOMMCONFIG bool "MMConfig" @@ -970,6 +1000,13 @@ config PCI_GOMMCONFIG config PCI_GODIRECT bool "Direct" +config PCI_GOXEN_FE + bool "Xen PCI Frontend" + depends on X86_XEN + help + The PCI device frontend driver allows the kernel to import arbitrary + PCI devices from a PCI backend to support PCI driver domains. + config PCI_GOANY bool "Any" @@ -977,7 +1014,7 @@ endchoice config PCI_BIOS bool - depends on !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY) + depends on !(X86_VISWS || X86_XEN) && PCI && (PCI_GOBIOS || PCI_GOANY) default y config PCI_DIRECT @@ -990,6 +1027,18 @@ config PCI_MMCONFIG depends on PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY) default y +config XEN_PCIDEV_FRONTEND + bool + depends on PCI && X86_XEN && (PCI_GOXEN_FE || PCI_GOANY) + default y + +config XEN_PCIDEV_FE_DEBUG + bool "Xen PCI Frontend Debugging" + depends on XEN_PCIDEV_FRONTEND + default n + help + Enables some debug statements within the PCI Frontend. + source "drivers/pci/pcie/Kconfig" source "drivers/pci/Kconfig" @@ -1000,7 +1049,7 @@ config ISA_DMA_API config ISA bool "ISA support" - depends on !(X86_VOYAGER || X86_VISWS) + depends on !(X86_VOYAGER || X86_VISWS || X86_XEN) help Find out whether you have ISA slots on your motherboard. ISA is the name of a bus system, i.e. the way the CPU talks to the other stuff @@ -1027,7 +1076,7 @@ config EISA source "drivers/eisa/Kconfig" config MCA - bool "MCA support" if !(X86_VISWS || X86_VOYAGER) + bool "MCA support" if !(X86_VISWS || X86_VOYAGER || X86_XEN) default y if X86_VOYAGER help MicroChannel Architecture is found in some IBM PS/2 machines and @@ -1091,6 +1140,8 @@ source "security/Kconfig" source "crypto/Kconfig" +source "drivers/xen/Kconfig" + source "lib/Kconfig" # @@ -1116,7 +1167,7 @@ config X86_SMP config X86_HT bool - depends on SMP && !(X86_VISWS || X86_VOYAGER) + depends on SMP && !(X86_VISWS || X86_VOYAGER || X86_XEN) default y config X86_BIOS_REBOOT @@ -1129,6 +1180,16 @@ config X86_TRAMPOLINE depends on X86_SMP || (X86_VOYAGER && SMP) default y +config X86_NO_TSS + bool + depends on X86_XEN + default y + +config X86_NO_IDT + bool + depends on X86_XEN + default y + config KTIME_SCALAR bool default y diff --git a/arch/i386/Kconfig.cpu b/arch/i386/Kconfig.cpu index eb130482b..f90d8d0df 100644 --- a/arch/i386/Kconfig.cpu +++ b/arch/i386/Kconfig.cpu @@ -251,7 +251,7 @@ config X86_PPRO_FENCE config X86_F00F_BUG bool - depends on M586MMX || M586TSC || M586 || M486 || M386 + depends on (M586MMX || M586TSC || M586 || M486 || M386) && !X86_NO_IDT default y config X86_WP_WORKS_OK diff --git a/arch/i386/Kconfig.debug b/arch/i386/Kconfig.debug index c92191b1f..fb28fe741 100644 --- a/arch/i386/Kconfig.debug +++ b/arch/i386/Kconfig.debug @@ -84,6 +84,7 @@ config X86_MPPARSE config DOUBLEFAULT default y bool "Enable doublefault exception handler" if EMBEDDED + depends on !X86_NO_TSS help This option allows trapping of rare doublefault exceptions that would otherwise cause a system to silently reboot. Disabling this diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 94e0fafb4..53b7343aa 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -48,6 +48,11 @@ CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-op CFLAGS += $(cflags-y) +cppflags-$(CONFIG_XEN) += \ + -D__XEN_INTERFACE_VERSION__=$(CONFIG_XEN_INTERFACE_VERSION) + +CPPFLAGS += $(cppflags-y) + # Default subarch .c files mcore-y := mach-default @@ -71,6 +76,10 @@ mcore-$(CONFIG_X86_BIGSMP) := mach-default mflags-$(CONFIG_X86_SUMMIT) := -Iinclude/asm-i386/mach-summit mcore-$(CONFIG_X86_SUMMIT) := mach-default +# Xen subarch support +mflags-$(CONFIG_X86_XEN) := -Iinclude/asm-i386/mach-xen +mcore-$(CONFIG_X86_XEN) := mach-xen + # generic subarchitecture mflags-$(CONFIG_X86_GENERICARCH) := -Iinclude/asm-i386/mach-generic mcore-$(CONFIG_X86_GENERICARCH) := mach-default @@ -105,6 +114,19 @@ boot := arch/i386/boot PHONY += zImage bzImage compressed zlilo bzlilo \ zdisk bzdisk fdimage fdimage144 fdimage288 isoimage install +ifdef CONFIG_XEN +CPPFLAGS := -Iinclude$(if $(KBUILD_SRC),2)/asm/mach-xen $(CPPFLAGS) +head-y := arch/i386/kernel/head-xen.o arch/i386/kernel/init_task-xen.o +boot := arch/i386/boot-xen +.PHONY: vmlinuz +all: vmlinuz + +vmlinuz: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $@ + +install: + $(Q)$(MAKE) $(build)=$(boot) XENGUEST=$(XENGUEST) $@ +else all: bzImage # KBUILD_IMAGE specify target image being built @@ -127,6 +149,7 @@ fdimage fdimage144 fdimage288 isoimage: vmlinux install: $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install +endif archclean: $(Q)$(MAKE) $(clean)=arch/i386/boot @@ -145,3 +168,4 @@ endef CLEAN_FILES += arch/$(ARCH)/boot/fdimage \ arch/$(ARCH)/boot/image.iso \ arch/$(ARCH)/boot/mtools.conf +CLEAN_FILES += vmlinuz vmlinux-stripped diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile index 96fb8a020..ddaa19d89 100644 --- a/arch/i386/kernel/Makefile +++ b/arch/i386/kernel/Makefile @@ -42,6 +42,12 @@ EXTRA_AFLAGS := -traditional obj-$(CONFIG_SCx200) += scx200.o +ifdef CONFIG_XEN +vsyscall_note := vsyscall-note-xen.o +else +vsyscall_note := vsyscall-note.o +endif + # vsyscall.o contains the vsyscall DSO images as __initdata. # We must build both images before we can assemble it. # Note: kbuild does not track this dependency due to usage of .incbin @@ -62,7 +68,7 @@ SYSCFLAGS_vsyscall-int80.so = $(vsyscall-flags) $(obj)/vsyscall-int80.so $(obj)/vsyscall-sysenter.so: \ $(obj)/vsyscall-%.so: $(src)/vsyscall.lds \ - $(obj)/vsyscall-%.o $(obj)/vsyscall-note.o FORCE + $(obj)/vsyscall-%.o $(obj)/$(vsyscall_note) FORCE $(call if_changed,syscall) # We also create a special relocatable object that should mirror the symbol @@ -74,5 +80,17 @@ $(obj)/built-in.o: ld_flags += -R $(obj)/vsyscall-syms.o SYSCFLAGS_vsyscall-syms.o = -r $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \ - $(obj)/vsyscall-sysenter.o $(obj)/vsyscall-note.o FORCE + $(obj)/vsyscall-sysenter.o $(obj)/$(vsyscall_note) FORCE $(call if_changed,syscall) + +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen + +obj-y += fixup.o +microcode-$(subst m,y,$(CONFIG_MICROCODE)) := microcode-xen.o +n-obj-xen := i8259.o timers/ reboot.o smpboot.o trampoline.o + +obj-y := $(call filterxen, $(obj-y), $(n-obj-xen)) +obj-y := $(call cherrypickxen, $(obj-y)) +extra-y := $(call cherrypickxen, $(extra-y)) +endif diff --git a/arch/i386/kernel/acpi/Makefile b/arch/i386/kernel/acpi/Makefile index 7e9ac9935..fa783e6f9 100644 --- a/arch/i386/kernel/acpi/Makefile +++ b/arch/i386/kernel/acpi/Makefile @@ -6,3 +6,7 @@ ifneq ($(CONFIG_ACPI_PROCESSOR),) obj-y += cstate.o processor.o endif +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen +obj-y := $(call cherrypickxen, $(obj-y), $(src)) +endif diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c index ac3f4e5ec..c6b52aa8d 100644 --- a/arch/i386/kernel/asm-offsets.c +++ b/arch/i386/kernel/asm-offsets.c @@ -13,6 +13,7 @@ #include #include #include +#include #define DEFINE(sym, val) \ asm volatile("\n->" #sym " %0 " #val : : "i" (val)) @@ -64,10 +65,14 @@ void foo(void) OFFSET(pbe_orig_address, pbe, orig_address); OFFSET(pbe_next, pbe, next); +#ifndef CONFIG_X86_NO_TSS /* Offset from the sysenter stack to tss.esp0 */ - DEFINE(TSS_sysenter_esp0, offsetof(struct tss_struct, esp0) - + DEFINE(SYSENTER_stack_esp0, offsetof(struct tss_struct, esp0) - sizeof(struct tss_struct)); +#else + /* sysenter stack points directly to esp0 */ + DEFINE(SYSENTER_stack_esp0, 0); +#endif DEFINE(PAGE_SIZE_asm, PAGE_SIZE); - DEFINE(VSYSCALL_BASE, __fix_to_virt(FIX_VSYSCALL)); } diff --git a/arch/i386/kernel/cpu/Makefile b/arch/i386/kernel/cpu/Makefile index 010aecfff..753f1d770 100644 --- a/arch/i386/kernel/cpu/Makefile +++ b/arch/i386/kernel/cpu/Makefile @@ -17,3 +17,8 @@ obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ obj-$(CONFIG_CPU_FREQ) += cpufreq/ + +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen +obj-y := $(call cherrypickxen, $(obj-y), $(src)) +endif diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c index da2db31d4..a748409b7 100644 --- a/arch/i386/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/i386/kernel/cpu/cpufreq/powernow-k7.c @@ -581,7 +581,6 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) rdmsrl (MSR_K7_FID_VID_STATUS, fidvidstatus.val); - /* recalibrate cpu_khz */ recalibrate_cpu_khz(); fsb = (10 * cpu_khz) / fid_codes[fidvidstatus.bits.CFID]; diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c index 71fffa174..2ea3c6cac 100644 --- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c @@ -1008,7 +1008,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) * an UP version, and is deprecated by AMD. */ - if ((num_online_cpus() != 1) || (num_possible_cpus() != 1)) { + if (num_online_cpus() != 1) { printk(KERN_ERR PFX "MP systems not supported by PSB BIOS structure\n"); kfree(data); return -ENODEV; diff --git a/arch/i386/kernel/cpu/mtrr/Makefile b/arch/i386/kernel/cpu/mtrr/Makefile index a25b701ab..06df4feee 100644 --- a/arch/i386/kernel/cpu/mtrr/Makefile +++ b/arch/i386/kernel/cpu/mtrr/Makefile @@ -3,3 +3,10 @@ obj-y += amd.o obj-y += cyrix.o obj-y += centaur.o +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen +n-obj-xen := generic.o state.o amd.o cyrix.o centaur.o + +obj-y := $(call filterxen, $(obj-y), $(n-obj-xen)) +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 840d990b8..eed41e6dd 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -177,7 +177,7 @@ need_resched: # sysenter call handler stub ENTRY(sysenter_entry) - movl TSS_sysenter_esp0(%esp),%esp + movl SYSENTER_stack_esp0(%esp),%esp sysenter_past_esp: sti pushl $(__USER_DS) @@ -414,7 +414,7 @@ vector=0 ENTRY(irq_entries_start) .rept NR_IRQS ALIGN -1: pushl $vector-256 +1: pushl $~(vector) jmp common_interrupt .data .long 1b @@ -431,7 +431,7 @@ common_interrupt: #define BUILD_INTERRUPT(name, nr) \ ENTRY(name) \ - pushl $nr-256; \ + pushl $~(nr); \ SAVE_ALL \ movl %esp,%eax; \ call smp_/**/name; \ @@ -500,7 +500,7 @@ device_not_available_emulate: * that sets up the real kernel stack. Check here, since we can't * allow the wrong stack to be used. * - * "TSS_sysenter_esp0+12" is because the NMI/debug handler will have + * "SYSENTER_stack_esp0+12" is because the NMI/debug handler will have * already pushed 3 words if it hits on the sysenter instruction: * eflags, cs and eip. * @@ -512,7 +512,7 @@ device_not_available_emulate: cmpw $__KERNEL_CS,4(%esp); \ jne ok; \ label: \ - movl TSS_sysenter_esp0+offset(%esp),%esp; \ + movl SYSENTER_stack_esp0+offset(%esp),%esp; \ pushfl; \ pushl $__KERNEL_CS; \ pushl $sysenter_past_esp diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c index f3a9c78c4..2904a6686 100644 --- a/arch/i386/kernel/irq.c +++ b/arch/i386/kernel/irq.c @@ -53,8 +53,8 @@ static union irq_ctx *softirq_ctx[NR_CPUS]; */ fastcall unsigned int do_IRQ(struct pt_regs *regs) { - /* high bits used in ret_from_ code */ - int irq = regs->orig_eax & 0xff; + /* high bit used in ret_from_ code */ + int irq = ~regs->orig_eax; #ifdef CONFIG_4KSTACKS union irq_ctx *curctx, *irqctx; u32 *isp; diff --git a/arch/i386/kernel/reboot.c b/arch/i386/kernel/reboot.c index 4163865dd..d20724297 100644 --- a/arch/i386/kernel/reboot.c +++ b/arch/i386/kernel/reboot.c @@ -120,14 +120,6 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"), }, }, - { /* HP laptops have weird reboot issues */ - .callback = set_bios_reboot, - .ident = "HP Laptop", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"), - DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"), - }, - }, { } }; diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c index ec4959f74..b97719aa4 100644 --- a/arch/i386/kernel/sysenter.c +++ b/arch/i386/kernel/sysenter.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -21,10 +22,15 @@ #include #include +#ifdef CONFIG_XEN +#include +#endif + extern asmlinkage void sysenter_entry(void); void enable_sep_cpu(void) { +#ifndef CONFIG_X86_NO_TSS int cpu = get_cpu(); struct tss_struct *tss = &per_cpu(init_tss, cpu); @@ -39,6 +45,7 @@ void enable_sep_cpu(void) wrmsr(MSR_IA32_SYSENTER_ESP, tss->esp1, 0); wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long) sysenter_entry, 0); put_cpu(); +#endif } /* @@ -47,7 +54,6 @@ void enable_sep_cpu(void) */ extern const char vsyscall_int80_start, vsyscall_int80_end; extern const char vsyscall_sysenter_start, vsyscall_sysenter_end; - static struct page *sysenter_pages[2]; int __init sysenter_setup(void) @@ -56,16 +62,28 @@ int __init sysenter_setup(void) sysenter_pages[0] = virt_to_page(page); - if (!boot_cpu_has(X86_FEATURE_SEP)) { +#ifdef CONFIG_XEN + if (boot_cpu_has(X86_FEATURE_SEP)) { + struct callback_register sysenter = { + .type = CALLBACKTYPE_sysenter, + .address = { __KERNEL_CS, (unsigned long)sysenter_entry }, + }; + + if (HYPERVISOR_callback_op(CALLBACKOP_register, &sysenter) < 0) + clear_bit(X86_FEATURE_SEP, boot_cpu_data.x86_capability); + } +#endif + + if (boot_cpu_has(X86_FEATURE_SEP)) { memcpy(page, - &vsyscall_int80_start, - &vsyscall_int80_end - &vsyscall_int80_start); + &vsyscall_sysenter_start, + &vsyscall_sysenter_end - &vsyscall_sysenter_start); return 0; } memcpy(page, - &vsyscall_sysenter_start, - &vsyscall_sysenter_end - &vsyscall_sysenter_start); + &vsyscall_int80_start, + &vsyscall_int80_end - &vsyscall_int80_start); return 0; } @@ -130,6 +148,7 @@ int arch_setup_additional_pages(struct linux_binprm *bprm, return err; } +#ifndef CONFIG_XEN int in_gate_area_no_task(unsigned long addr) { return 0; @@ -144,3 +163,4 @@ struct vm_area_struct *get_gate_vma(struct task_struct *tsk) { return NULL; } +#endif diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 0b9d741f8..5b65cb842 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -705,18 +705,11 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs) static void io_check_error(unsigned char reason, struct pt_regs * regs) { - unsigned long i; - printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n"); show_registers(regs); /* Re-enable the IOCK line, wait for a few seconds */ - reason = (reason & 0xf) | 8; - outb(reason, 0x61); - i = 2000; - while (--i) udelay(1000); - reason &= ~8; - outb(reason, 0x61); + clear_io_check_error(reason); } static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs) diff --git a/arch/i386/kernel/vm86.c b/arch/i386/kernel/vm86.c index 00e0118e7..c0adee5da 100644 --- a/arch/i386/kernel/vm86.c +++ b/arch/i386/kernel/vm86.c @@ -98,7 +98,9 @@ struct pt_regs * FASTCALL(save_v86_state(struct kernel_vm86_regs * regs)); struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) { +#ifndef CONFIG_X86_NO_TSS struct tss_struct *tss; +#endif struct pt_regs *ret; unsigned long tmp; @@ -123,12 +125,16 @@ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs) do_exit(SIGSEGV); } +#ifndef CONFIG_X86_NO_TSS tss = &per_cpu(init_tss, get_cpu()); +#endif current->thread.esp0 = current->thread.saved_esp0; current->thread.sysenter_cs = __KERNEL_CS; load_esp0(tss, ¤t->thread); current->thread.saved_esp0 = 0; +#ifndef CONFIG_X86_NO_TSS put_cpu(); +#endif loadsegment(fs, current->thread.saved_fs); loadsegment(gs, current->thread.saved_gs); @@ -252,7 +258,9 @@ out: static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk) { +#ifndef CONFIG_X86_NO_TSS struct tss_struct *tss; +#endif long eax; /* * make sure the vm86() system call doesn't try to do anything silly @@ -297,12 +305,16 @@ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk savesegment(fs, tsk->thread.saved_fs); savesegment(gs, tsk->thread.saved_gs); +#ifndef CONFIG_X86_NO_TSS tss = &per_cpu(init_tss, get_cpu()); +#endif tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0; if (cpu_has_sep) tsk->thread.sysenter_cs = 0; load_esp0(tss, &tsk->thread); +#ifndef CONFIG_X86_NO_TSS put_cpu(); +#endif tsk->thread.screen_bitmap = info->screen_bitmap; if (info->flags & VM86_SCREEN_BITMAP) diff --git a/arch/i386/mm/Makefile b/arch/i386/mm/Makefile index 80908b5aa..2b33b2003 100644 --- a/arch/i386/mm/Makefile +++ b/arch/i386/mm/Makefile @@ -8,3 +8,11 @@ obj-$(CONFIG_NUMA) += discontig.o obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o obj-$(CONFIG_HIGHMEM) += highmem.o obj-$(CONFIG_BOOT_IOREMAP) += boot_ioremap.o + +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen + +obj-y += hypervisor.o + +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/i386/mm/pageattr.c b/arch/i386/mm/pageattr.c index 92c3d9f0e..a00267a5b 100644 --- a/arch/i386/mm/pageattr.c +++ b/arch/i386/mm/pageattr.c @@ -85,7 +85,7 @@ static void set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte) unsigned long flags; set_pte_atomic(kpte, pte); /* change init_mm */ - if (PTRS_PER_PMD > 1) + if (HAVE_SHARED_KERNEL_PMD) return; spin_lock_irqsave(&pgd_lock, flags); diff --git a/arch/i386/mm/pgtable.c b/arch/i386/mm/pgtable.c index 2889567e2..de8d18774 100644 --- a/arch/i386/mm/pgtable.c +++ b/arch/i386/mm/pgtable.c @@ -13,6 +13,7 @@ #include #include #include +#include #include #include @@ -138,6 +139,10 @@ void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags) __flush_tlb_one(vaddr); } +static int nr_fixmaps = 0; +unsigned long __FIXADDR_TOP = 0xfffff000; +EXPORT_SYMBOL(__FIXADDR_TOP); + void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags) { unsigned long address = __fix_to_virt(idx); @@ -147,6 +152,13 @@ void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags) return; } set_pte_pfn(address, phys >> PAGE_SHIFT, flags); + nr_fixmaps++; +} + +void set_fixaddr_top(unsigned long top) +{ + BUG_ON(nr_fixmaps > 0); + __FIXADDR_TOP = top - PAGE_SIZE; } pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) diff --git a/arch/i386/oprofile/Makefile b/arch/i386/oprofile/Makefile index 30f3eb366..e596c39c0 100644 --- a/arch/i386/oprofile/Makefile +++ b/arch/i386/oprofile/Makefile @@ -6,7 +6,11 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \ oprofilefs.o oprofile_stats.o \ timer_int.o ) +ifdef CONFIG_XEN +oprofile-y := $(DRIVER_OBJS) xenoprof.o +else oprofile-y := $(DRIVER_OBJS) init.o backtrace.o oprofile-$(CONFIG_X86_LOCAL_APIC) += nmi_int.o op_model_athlon.o \ op_model_ppro.o op_model_p4.o oprofile-$(CONFIG_X86_IO_APIC) += nmi_timer_int.o +endif diff --git a/arch/i386/pci/Makefile b/arch/i386/pci/Makefile index 62ad75c57..4447f397f 100644 --- a/arch/i386/pci/Makefile +++ b/arch/i386/pci/Makefile @@ -4,6 +4,10 @@ obj-$(CONFIG_PCI_BIOS) += pcbios.o obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o direct.o obj-$(CONFIG_PCI_DIRECT) += direct.o +# pcifront should be after pcbios.o, mmconfig.o, and direct.o as it should only +# take over if direct access to the PCI bus is unavailable +obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront.o + pci-y := fixup.o pci-$(CONFIG_ACPI) += acpi.o pci-y += legacy.o irq.o @@ -12,3 +16,8 @@ pci-$(CONFIG_X86_VISWS) := visws.o fixup.o pci-$(CONFIG_X86_NUMAQ) := numa.o irq.o obj-y += $(pci-y) common.o + +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/i386/pci/mmconfig.c b/arch/i386/pci/mmconfig.c index 6b1ea0c9a..e545b0992 100644 --- a/arch/i386/pci/mmconfig.c +++ b/arch/i386/pci/mmconfig.c @@ -15,7 +15,9 @@ #include #include "pci.h" -#define MMCONFIG_APER_SIZE (256*1024*1024) +/* aperture is up to 256MB but BIOS may reserve less */ +#define MMCONFIG_APER_MIN (2 * 1024*1024) +#define MMCONFIG_APER_MAX (256 * 1024*1024) /* Assume systems with more busses have correct MCFG */ #define MAX_CHECK_BUS 16 @@ -197,9 +199,10 @@ void __init pci_mmcfg_init(void) return; if (!e820_all_mapped(pci_mmcfg_config[0].base_address, - pci_mmcfg_config[0].base_address + MMCONFIG_APER_SIZE, + pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN, E820_RESERVED)) { - printk(KERN_ERR "PCI: BIOS Bug: MCFG area is not E820-reserved\n"); + printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n", + pci_mmcfg_config[0].base_address); printk(KERN_ERR "PCI: Not using MMCONFIG.\n"); return; } diff --git a/arch/i386/power/Makefile b/arch/i386/power/Makefile index 8cfa4e8a7..e74fee667 100644 --- a/arch/i386/power/Makefile +++ b/arch/i386/power/Makefile @@ -1,2 +1,4 @@ -obj-$(CONFIG_PM) += cpu.o +obj-$(CONFIG_PM_LEGACY) += cpu.o +obj-$(CONFIG_SOFTWARE_SUSPEND) += cpu.o +obj-$(CONFIG_ACPI_SLEEP) += cpu.o obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig index 67a905304..1e3ad1ecc 100644 --- a/arch/ia64/Kconfig +++ b/arch/ia64/Kconfig @@ -270,6 +270,9 @@ config HOTPLUG_CPU can be controlled through /sys/devices/system/cpu/cpu#. Say N if you want to disable CPU hotplug. +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y + config SCHED_SMT bool "SMT scheduler support" depends on SMP diff --git a/arch/ia64/kernel/sys_ia64.c b/arch/ia64/kernel/sys_ia64.c index c7b943f10..3edefc8f6 100644 --- a/arch/ia64/kernel/sys_ia64.c +++ b/arch/ia64/kernel/sys_ia64.c @@ -164,10 +164,25 @@ sys_pipe (void) return retval; } +int ia64_map_check_rgn(unsigned long addr, unsigned long len, + unsigned long flags) +{ + unsigned long roff; + + /* + * Don't permit mappings into unmapped space, the virtual page table + * of a region, or across a region boundary. Note: RGN_MAP_LIMIT is + * equal to 2^n-PAGE_SIZE (for some integer n <= 61) and len > 0. + */ + roff = REGION_OFFSET(addr); + if ((len > RGN_MAP_LIMIT) || (roff > (RGN_MAP_LIMIT - len))) + return -EINVAL; + return 0; +} + static inline unsigned long do_mmap2 (unsigned long addr, unsigned long len, int prot, int flags, int fd, unsigned long pgoff) { - unsigned long roff; struct file *file = NULL; flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); @@ -189,17 +204,6 @@ do_mmap2 (unsigned long addr, unsigned long len, int prot, int flags, int fd, un goto out; } - /* - * Don't permit mappings into unmapped space, the virtual page table of a region, - * or across a region boundary. Note: RGN_MAP_LIMIT is equal to 2^n-PAGE_SIZE - * (for some integer n <= 61) and len > 0. - */ - roff = REGION_OFFSET(addr); - if ((len > RGN_MAP_LIMIT) || (roff > (RGN_MAP_LIMIT - len))) { - addr = -EINVAL; - goto out; - } - down_write(¤t->mm->mmap_sem); addr = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); up_write(¤t->mm->mmap_sem); diff --git a/arch/ia64/sn/kernel/xpc_channel.c b/arch/ia64/sn/kernel/xpc_channel.c index 0bf6fbcc4..7e8a4d1d9 100644 --- a/arch/ia64/sn/kernel/xpc_channel.c +++ b/arch/ia64/sn/kernel/xpc_channel.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2006 Silicon Graphics, Inc. All Rights Reserved. */ @@ -21,10 +21,36 @@ #include #include #include -#include +#include +#include #include #include -#include "xpc.h" +#include + + +/* + * Guarantee that the kzalloc'd memory is cacheline aligned. + */ +static void * +xpc_kzalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) +{ + /* see if kzalloc will give us cachline aligned memory by default */ + *base = kzalloc(size, flags); + if (*base == NULL) { + return NULL; + } + if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { + return *base; + } + kfree(*base); + + /* nope, we'll have to do it ourselves */ + *base = kzalloc(size + L1_CACHE_BYTES, flags); + if (*base == NULL) { + return NULL; + } + return (void *) L1_CACHE_ALIGN((u64) *base); +} /* @@ -56,7 +82,8 @@ xpc_initialize_channels(struct xpc_partition *part, partid_t partid) atomic_set(&ch->n_to_notify, 0); spin_lock_init(&ch->lock); - sema_init(&ch->msg_to_pull_sema, 1); /* mutex */ + mutex_init(&ch->msg_to_pull_mutex); + init_completion(&ch->wdisconnect_wait); atomic_set(&ch->n_on_msg_allocate_wq, 0); init_waitqueue_head(&ch->msg_allocate_wq); @@ -72,7 +99,7 @@ xpc_initialize_channels(struct xpc_partition *part, partid_t partid) enum xpc_retval xpc_setup_infrastructure(struct xpc_partition *part) { - int ret; + int ret, cpuid; struct timer_list *timer; partid_t partid = XPC_PARTID(part); @@ -90,20 +117,19 @@ xpc_setup_infrastructure(struct xpc_partition *part) * Allocate all of the channel structures as a contiguous chunk of * memory. */ - part->channels = kmalloc(sizeof(struct xpc_channel) * XPC_NCHANNELS, + part->channels = kzalloc(sizeof(struct xpc_channel) * XPC_NCHANNELS, GFP_KERNEL); if (part->channels == NULL) { dev_err(xpc_chan, "can't get memory for channels\n"); return xpcNoMemory; } - memset(part->channels, 0, sizeof(struct xpc_channel) * XPC_NCHANNELS); part->nchannels = XPC_NCHANNELS; /* allocate all the required GET/PUT values */ - part->local_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + part->local_GPs = xpc_kzalloc_cacheline_aligned(XPC_GP_SIZE, GFP_KERNEL, &part->local_GPs_base); if (part->local_GPs == NULL) { kfree(part->channels); @@ -112,60 +138,57 @@ xpc_setup_infrastructure(struct xpc_partition *part) "values\n"); return xpcNoMemory; } - memset(part->local_GPs, 0, XPC_GP_SIZE); - part->remote_GPs = xpc_kmalloc_cacheline_aligned(XPC_GP_SIZE, + part->remote_GPs = xpc_kzalloc_cacheline_aligned(XPC_GP_SIZE, GFP_KERNEL, &part->remote_GPs_base); if (part->remote_GPs == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; dev_err(xpc_chan, "can't get memory for remote get/put " "values\n"); + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->remote_GPs, 0, XPC_GP_SIZE); /* allocate all the required open and close args */ - part->local_openclose_args = xpc_kmalloc_cacheline_aligned( + part->local_openclose_args = xpc_kzalloc_cacheline_aligned( XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, &part->local_openclose_args_base); if (part->local_openclose_args == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; + dev_err(xpc_chan, "can't get memory for local connect args\n"); kfree(part->remote_GPs_base); part->remote_GPs = NULL; - dev_err(xpc_chan, "can't get memory for local connect args\n"); + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->local_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); - part->remote_openclose_args = xpc_kmalloc_cacheline_aligned( + part->remote_openclose_args = xpc_kzalloc_cacheline_aligned( XPC_OPENCLOSE_ARGS_SIZE, GFP_KERNEL, &part->remote_openclose_args_base); if (part->remote_openclose_args == NULL) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; - kfree(part->remote_GPs_base); - part->remote_GPs = NULL; + dev_err(xpc_chan, "can't get memory for remote connect args\n"); kfree(part->local_openclose_args_base); part->local_openclose_args = NULL; - dev_err(xpc_chan, "can't get memory for remote connect args\n"); + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcNoMemory; } - memset(part->remote_openclose_args, 0, XPC_OPENCLOSE_ARGS_SIZE); xpc_initialize_channels(part, partid); atomic_set(&part->nchannels_active, 0); + atomic_set(&part->nchannels_engaged, 0); /* local_IPI_amo were set to 0 by an earlier memset() */ @@ -182,18 +205,18 @@ xpc_setup_infrastructure(struct xpc_partition *part) ret = request_irq(SGI_XPC_NOTIFY, xpc_notify_IRQ_handler, SA_SHIRQ, part->IPI_owner, (void *) (u64) partid); if (ret != 0) { - kfree(part->channels); - part->channels = NULL; - kfree(part->local_GPs_base); - part->local_GPs = NULL; - kfree(part->remote_GPs_base); - part->remote_GPs = NULL; - kfree(part->local_openclose_args_base); - part->local_openclose_args = NULL; - kfree(part->remote_openclose_args_base); - part->remote_openclose_args = NULL; dev_err(xpc_chan, "can't register NOTIFY IRQ handler, " "errno=%d\n", -ret); + kfree(part->remote_openclose_args_base); + part->remote_openclose_args = NULL; + kfree(part->local_openclose_args_base); + part->local_openclose_args = NULL; + kfree(part->remote_GPs_base); + part->remote_GPs = NULL; + kfree(part->local_GPs_base); + part->local_GPs = NULL; + kfree(part->channels); + part->channels = NULL; return xpcLackOfResources; } @@ -209,7 +232,7 @@ xpc_setup_infrastructure(struct xpc_partition *part) * With the setting of the partition setup_state to XPC_P_SETUP, we're * declaring that this partition is ready to go. */ - (volatile u8) part->setup_state = XPC_P_SETUP; + part->setup_state = XPC_P_SETUP; /* @@ -223,11 +246,11 @@ xpc_setup_infrastructure(struct xpc_partition *part) xpc_vars_part[partid].openclose_args_pa = __pa(part->local_openclose_args); xpc_vars_part[partid].IPI_amo_pa = __pa(part->local_IPI_amo_va); - xpc_vars_part[partid].IPI_nasid = cpuid_to_nasid(smp_processor_id()); - xpc_vars_part[partid].IPI_phys_cpuid = - cpu_physical_id(smp_processor_id()); + cpuid = raw_smp_processor_id(); /* any CPU in this partition will do */ + xpc_vars_part[partid].IPI_nasid = cpuid_to_nasid(cpuid); + xpc_vars_part[partid].IPI_phys_cpuid = cpu_physical_id(cpuid); xpc_vars_part[partid].nchannels = part->nchannels; - (volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC1; + xpc_vars_part[partid].magic = XPC_VP_MAGIC1; return xpcSuccess; } @@ -256,8 +279,8 @@ xpc_pull_remote_cachelines(struct xpc_partition *part, void *dst, return part->reason; } - bte_ret = xp_bte_copy((u64) src, (u64) ia64_tpa((u64) dst), - (u64) cnt, (BTE_NORMAL | BTE_WACQUIRE), NULL); + bte_ret = xp_bte_copy((u64) src, (u64) dst, (u64) cnt, + (BTE_NORMAL | BTE_WACQUIRE), NULL); if (bte_ret == BTE_SUCCESS) { return xpcSuccess; } @@ -355,7 +378,7 @@ xpc_pull_remote_vars_part(struct xpc_partition *part) /* let the other side know that we've pulled their variables */ - (volatile u64) xpc_vars_part[partid].magic = XPC_VP_MAGIC2; + xpc_vars_part[partid].magic = XPC_VP_MAGIC2; } if (pulled_entry->magic == XPC_VP_MAGIC1) { @@ -442,22 +465,20 @@ xpc_allocate_local_msgqueue(struct xpc_channel *ch) for (nentries = ch->local_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; - ch->local_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, - (GFP_KERNEL | GFP_DMA), + ch->local_msgqueue = xpc_kzalloc_cacheline_aligned(nbytes, + GFP_KERNEL, &ch->local_msgqueue_base); if (ch->local_msgqueue == NULL) { continue; } - memset(ch->local_msgqueue, 0, nbytes); nbytes = nentries * sizeof(struct xpc_notify); - ch->notify_queue = kmalloc(nbytes, (GFP_KERNEL | GFP_DMA)); + ch->notify_queue = kzalloc(nbytes, GFP_KERNEL); if (ch->notify_queue == NULL) { kfree(ch->local_msgqueue_base); ch->local_msgqueue = NULL; continue; } - memset(ch->notify_queue, 0, nbytes); spin_lock_irqsave(&ch->lock, irq_flags); if (nentries < ch->local_nentries) { @@ -497,13 +518,12 @@ xpc_allocate_remote_msgqueue(struct xpc_channel *ch) for (nentries = ch->remote_nentries; nentries > 0; nentries--) { nbytes = nentries * ch->msg_size; - ch->remote_msgqueue = xpc_kmalloc_cacheline_aligned(nbytes, - (GFP_KERNEL | GFP_DMA), + ch->remote_msgqueue = xpc_kzalloc_cacheline_aligned(nbytes, + GFP_KERNEL, &ch->remote_msgqueue_base); if (ch->remote_msgqueue == NULL) { continue; } - memset(ch->remote_msgqueue, 0, nbytes); spin_lock_irqsave(&ch->lock, irq_flags); if (nentries < ch->remote_nentries) { @@ -532,7 +552,6 @@ static enum xpc_retval xpc_allocate_msgqueues(struct xpc_channel *ch) { unsigned long irq_flags; - int i; enum xpc_retval ret; @@ -550,13 +569,6 @@ xpc_allocate_msgqueues(struct xpc_channel *ch) return ret; } - for (i = 0; i < ch->local_nentries; i++) { - /* use a semaphore as an event wait queue */ - sema_init(&ch->notify_queue[i].sema, 0); - } - - sema_init(&ch->teardown_sema, 0); /* event wait */ - spin_lock_irqsave(&ch->lock, irq_flags); ch->flags |= XPC_C_SETUP; spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -625,6 +637,55 @@ xpc_process_connect(struct xpc_channel *ch, unsigned long *irq_flags) } +/* + * Notify those who wanted to be notified upon delivery of their message. + */ +static void +xpc_notify_senders(struct xpc_channel *ch, enum xpc_retval reason, s64 put) +{ + struct xpc_notify *notify; + u8 notify_type; + s64 get = ch->w_remote_GP.get - 1; + + + while (++get < put && atomic_read(&ch->n_to_notify) > 0) { + + notify = &ch->notify_queue[get % ch->local_nentries]; + + /* + * See if the notify entry indicates it was associated with + * a message who's sender wants to be notified. It is possible + * that it is, but someone else is doing or has done the + * notification. + */ + notify_type = notify->type; + if (notify_type == 0 || + cmpxchg(¬ify->type, notify_type, 0) != + notify_type) { + continue; + } + + DBUG_ON(notify_type != XPC_N_CALL); + + atomic_dec(&ch->n_to_notify); + + if (notify->func != NULL) { + dev_dbg(xpc_chan, "notify->func() called, notify=0x%p, " + "msg_number=%ld, partid=%d, channel=%d\n", + (void *) notify, get, ch->partid, ch->number); + + notify->func(reason, ch->partid, ch->number, + notify->key); + + dev_dbg(xpc_chan, "notify->func() returned, " + "notify=0x%p, msg_number=%ld, partid=%d, " + "channel=%d\n", (void *) notify, get, + ch->partid, ch->number); + } + } +} + + /* * Free up message queues and other stuff that were allocated for the specified * channel. @@ -669,9 +730,6 @@ xpc_free_msgqueues(struct xpc_channel *ch) ch->remote_msgqueue = NULL; kfree(ch->notify_queue); ch->notify_queue = NULL; - - /* in case someone is waiting for the teardown to complete */ - up(&ch->teardown_sema); } } @@ -683,7 +741,7 @@ static void xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags) { struct xpc_partition *part = &xpc_partitions[ch->partid]; - u32 ch_flags = ch->flags; + u32 channel_was_connected = (ch->flags & XPC_C_WASCONNECTED); DBUG_ON(!spin_is_locked(&ch->lock)); @@ -696,17 +754,20 @@ xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags) /* make sure all activity has settled down first */ - if (atomic_read(&ch->references) > 0) { + if (atomic_read(&ch->references) > 0 || + ((ch->flags & XPC_C_CONNECTEDCALLOUT_MADE) && + !(ch->flags & XPC_C_DISCONNECTINGCALLOUT_MADE))) { return; } DBUG_ON(atomic_read(&ch->kthreads_assigned) != 0); - /* it's now safe to free the channel's message queues */ - - xpc_free_msgqueues(ch); - DBUG_ON(ch->flags & XPC_C_SETUP); + if (part->act_state == XPC_P_DEACTIVATING) { + /* can't proceed until the other side disengages from us */ + if (xpc_partition_engaged(1UL << ch->partid)) { + return; + } - if (part->act_state != XPC_P_DEACTIVATING) { + } else { /* as long as the other side is up do the full protocol */ @@ -724,16 +785,46 @@ xpc_process_disconnect(struct xpc_channel *ch, unsigned long *irq_flags) } } + /* wake those waiting for notify completion */ + if (atomic_read(&ch->n_to_notify) > 0) { + /* >>> we do callout while holding ch->lock */ + xpc_notify_senders(ch, ch->reason, ch->w_local_GP.put); + } + /* both sides are disconnected now */ - ch->flags = XPC_C_DISCONNECTED; /* clear all flags, but this one */ + if (ch->flags & XPC_C_DISCONNECTINGCALLOUT_MADE) { + spin_unlock_irqrestore(&ch->lock, *irq_flags); + xpc_disconnect_callout(ch, xpcDisconnected); + spin_lock_irqsave(&ch->lock, *irq_flags); + } + + /* it's now safe to free the channel's message queues */ + xpc_free_msgqueues(ch); + + /* mark disconnected, clear all other flags except XPC_C_WDISCONNECT */ + ch->flags = (XPC_C_DISCONNECTED | (ch->flags & XPC_C_WDISCONNECT)); atomic_dec(&part->nchannels_active); - if (ch_flags & XPC_C_WASCONNECTED) { + if (channel_was_connected) { dev_info(xpc_chan, "channel %d to partition %d disconnected, " "reason=%d\n", ch->number, ch->partid, ch->reason); } + + if (ch->flags & XPC_C_WDISCONNECT) { + /* we won't lose the CPU since we're holding ch->lock */ + complete(&ch->wdisconnect_wait); + } else if (ch->delayed_IPI_flags) { + if (part->act_state != XPC_P_DEACTIVATING) { + /* time to take action on any delayed IPI flags */ + spin_lock(&part->IPI_lock); + XPC_SET_IPI_FLAGS(part->local_IPI_amo, ch->number, + ch->delayed_IPI_flags); + spin_unlock(&part->IPI_lock); + } + ch->delayed_IPI_flags = 0; + } } @@ -754,6 +845,19 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, spin_lock_irqsave(&ch->lock, irq_flags); +again: + + if ((ch->flags & XPC_C_DISCONNECTED) && + (ch->flags & XPC_C_WDISCONNECT)) { + /* + * Delay processing IPI flags until thread waiting disconnect + * has had a chance to see that the channel is disconnected. + */ + ch->delayed_IPI_flags |= IPI_flags; + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + if (IPI_flags & XPC_IPI_CLOSEREQUEST) { @@ -764,7 +868,7 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, /* * If RCLOSEREQUEST is set, we're probably waiting for * RCLOSEREPLY. We should find it and a ROPENREQUEST packed - * with this RCLOSEQREUQEST in the IPI_flags. + * with this RCLOSEREQUEST in the IPI_flags. */ if (ch->flags & XPC_C_RCLOSEREQUEST) { @@ -779,14 +883,22 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, /* both sides have finished disconnecting */ xpc_process_disconnect(ch, &irq_flags); + DBUG_ON(!(ch->flags & XPC_C_DISCONNECTED)); + goto again; } if (ch->flags & XPC_C_DISCONNECTED) { - // >>> explain this section - if (!(IPI_flags & XPC_IPI_OPENREQUEST)) { - DBUG_ON(part->act_state != - XPC_P_DEACTIVATING); + if ((XPC_GET_IPI_FLAGS(part->local_IPI_amo, + ch_number) & XPC_IPI_OPENREQUEST)) { + + DBUG_ON(ch->delayed_IPI_flags != 0); + spin_lock(&part->IPI_lock); + XPC_SET_IPI_FLAGS(part->local_IPI_amo, + ch_number, + XPC_IPI_CLOSEREQUEST); + spin_unlock(&part->IPI_lock); + } spin_unlock_irqrestore(&ch->lock, irq_flags); return; } @@ -816,9 +928,13 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, } XPC_DISCONNECT_CHANNEL(ch, reason, &irq_flags); - } else { - xpc_process_disconnect(ch, &irq_flags); + + DBUG_ON(IPI_flags & XPC_IPI_CLOSEREPLY); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; } + + xpc_process_disconnect(ch, &irq_flags); } @@ -834,7 +950,20 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, } DBUG_ON(!(ch->flags & XPC_C_CLOSEREQUEST)); - DBUG_ON(!(ch->flags & XPC_C_RCLOSEREQUEST)); + + if (!(ch->flags & XPC_C_RCLOSEREQUEST)) { + if ((XPC_GET_IPI_FLAGS(part->local_IPI_amo, ch_number) + & XPC_IPI_CLOSEREQUEST)) { + + DBUG_ON(ch->delayed_IPI_flags != 0); + spin_lock(&part->IPI_lock); + XPC_SET_IPI_FLAGS(part->local_IPI_amo, + ch_number, XPC_IPI_CLOSEREPLY); + spin_unlock(&part->IPI_lock); + } + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } ch->flags |= XPC_C_RCLOSEREPLY; @@ -852,8 +981,14 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, "channel=%d\n", args->msg_size, args->local_nentries, ch->partid, ch->number); - if ((ch->flags & XPC_C_DISCONNECTING) || - part->act_state == XPC_P_DEACTIVATING) { + if (part->act_state == XPC_P_DEACTIVATING || + (ch->flags & XPC_C_ROPENREQUEST)) { + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + + if (ch->flags & (XPC_C_DISCONNECTING | XPC_C_WDISCONNECT)) { + ch->delayed_IPI_flags |= XPC_IPI_OPENREQUEST; spin_unlock_irqrestore(&ch->lock, irq_flags); return; } @@ -867,8 +1002,11 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, * msg_size = size of channel's messages in bytes * local_nentries = remote partition's local_nentries */ - DBUG_ON(args->msg_size == 0); - DBUG_ON(args->local_nentries == 0); + if (args->msg_size == 0 || args->local_nentries == 0) { + /* assume OPENREQUEST was delayed by mistake */ + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } ch->flags |= (XPC_C_ROPENREQUEST | XPC_C_CONNECTING); ch->remote_nentries = args->local_nentries; @@ -906,7 +1044,13 @@ xpc_process_openclose_IPI(struct xpc_partition *part, int ch_number, spin_unlock_irqrestore(&ch->lock, irq_flags); return; } - DBUG_ON(!(ch->flags & XPC_C_OPENREQUEST)); + if (!(ch->flags & XPC_C_OPENREQUEST)) { + XPC_DISCONNECT_CHANNEL(ch, xpcOpenCloseError, + &irq_flags); + spin_unlock_irqrestore(&ch->lock, irq_flags); + return; + } + DBUG_ON(!(ch->flags & XPC_C_ROPENREQUEST)); DBUG_ON(ch->flags & XPC_C_CONNECTED); @@ -960,12 +1104,12 @@ xpc_connect_channel(struct xpc_channel *ch) struct xpc_registration *registration = &xpc_registrations[ch->number]; - if (down_interruptible(®istration->sema) != 0) { - return xpcInterrupted; + if (mutex_trylock(®istration->mutex) == 0) { + return xpcRetry; } if (!XPC_CHANNEL_REGISTERED(ch->number)) { - up(®istration->sema); + mutex_unlock(®istration->mutex); return xpcUnregistered; } @@ -976,7 +1120,7 @@ xpc_connect_channel(struct xpc_channel *ch) if (ch->flags & XPC_C_DISCONNECTING) { spin_unlock_irqrestore(&ch->lock, irq_flags); - up(®istration->sema); + mutex_unlock(®istration->mutex); return ch->reason; } @@ -1008,7 +1152,7 @@ xpc_connect_channel(struct xpc_channel *ch) * channel lock be locked and will unlock and relock * the channel lock as needed. */ - up(®istration->sema); + mutex_unlock(®istration->mutex); XPC_DISCONNECT_CHANNEL(ch, xpcUnequalMsgSizes, &irq_flags); spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -1023,7 +1167,7 @@ xpc_connect_channel(struct xpc_channel *ch) atomic_inc(&xpc_partitions[ch->partid].nchannels_active); } - up(®istration->sema); + mutex_unlock(®istration->mutex); /* initiate the connection */ @@ -1039,55 +1183,6 @@ xpc_connect_channel(struct xpc_channel *ch) } -/* - * Notify those who wanted to be notified upon delivery of their message. - */ -static void -xpc_notify_senders(struct xpc_channel *ch, enum xpc_retval reason, s64 put) -{ - struct xpc_notify *notify; - u8 notify_type; - s64 get = ch->w_remote_GP.get - 1; - - - while (++get < put && atomic_read(&ch->n_to_notify) > 0) { - - notify = &ch->notify_queue[get % ch->local_nentries]; - - /* - * See if the notify entry indicates it was associated with - * a message who's sender wants to be notified. It is possible - * that it is, but someone else is doing or has done the - * notification. - */ - notify_type = notify->type; - if (notify_type == 0 || - cmpxchg(¬ify->type, notify_type, 0) != - notify_type) { - continue; - } - - DBUG_ON(notify_type != XPC_N_CALL); - - atomic_dec(&ch->n_to_notify); - - if (notify->func != NULL) { - dev_dbg(xpc_chan, "notify->func() called, notify=0x%p, " - "msg_number=%ld, partid=%d, channel=%d\n", - (void *) notify, get, ch->partid, ch->number); - - notify->func(reason, ch->partid, ch->number, - notify->key); - - dev_dbg(xpc_chan, "notify->func() returned, " - "notify=0x%p, msg_number=%ld, partid=%d, " - "channel=%d\n", (void *) notify, get, - ch->partid, ch->number); - } - } -} - - /* * Clear some of the msg flags in the local message queue. */ @@ -1183,7 +1278,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number) */ xpc_clear_local_msgqueue_flags(ch); - (volatile s64) ch->w_remote_GP.get = ch->remote_GP.get; + ch->w_remote_GP.get = ch->remote_GP.get; dev_dbg(xpc_chan, "w_remote_GP.get changed to %ld, partid=%d, " "channel=%d\n", ch->w_remote_GP.get, ch->partid, @@ -1211,7 +1306,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number) */ xpc_clear_remote_msgqueue_flags(ch); - (volatile s64) ch->w_remote_GP.put = ch->remote_GP.put; + ch->w_remote_GP.put = ch->remote_GP.put; dev_dbg(xpc_chan, "w_remote_GP.put changed to %ld, partid=%d, " "channel=%d\n", ch->w_remote_GP.put, ch->partid, @@ -1223,7 +1318,7 @@ xpc_process_msg_IPI(struct xpc_partition *part, int ch_number) "delivered=%d, partid=%d, channel=%d\n", nmsgs_sent, ch->partid, ch->number); - if (ch->flags & XPC_C_CONNECTCALLOUT) { + if (ch->flags & XPC_C_CONNECTEDCALLOUT_MADE) { xpc_activate_kthreads(ch, nmsgs_sent); } } @@ -1240,6 +1335,7 @@ xpc_process_channel_activity(struct xpc_partition *part) u64 IPI_amo, IPI_flags; struct xpc_channel *ch; int ch_number; + u32 ch_flags; IPI_amo = xpc_get_IPI_flags(part); @@ -1266,8 +1362,9 @@ xpc_process_channel_activity(struct xpc_partition *part) xpc_process_openclose_IPI(part, ch_number, IPI_flags); } + ch_flags = ch->flags; /* need an atomic snapshot of flags */ - if (ch->flags & XPC_C_DISCONNECTING) { + if (ch_flags & XPC_C_DISCONNECTING) { spin_lock_irqsave(&ch->lock, irq_flags); xpc_process_disconnect(ch, &irq_flags); spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -1278,9 +1375,9 @@ xpc_process_channel_activity(struct xpc_partition *part) continue; } - if (!(ch->flags & XPC_C_CONNECTED)) { - if (!(ch->flags & XPC_C_OPENREQUEST)) { - DBUG_ON(ch->flags & XPC_C_SETUP); + if (!(ch_flags & XPC_C_CONNECTED)) { + if (!(ch_flags & XPC_C_OPENREQUEST)) { + DBUG_ON(ch_flags & XPC_C_SETUP); (void) xpc_connect_channel(ch); } else { spin_lock_irqsave(&ch->lock, irq_flags); @@ -1305,8 +1402,8 @@ xpc_process_channel_activity(struct xpc_partition *part) /* - * XPC's heartbeat code calls this function to inform XPC that a partition has - * gone down. XPC responds by tearing down the XPartition Communication + * XPC's heartbeat code calls this function to inform XPC that a partition is + * going down. XPC responds by tearing down the XPartition Communication * infrastructure used for the just downed partition. * * XPC's heartbeat code will never call this function and xpc_partition_up() @@ -1314,7 +1411,7 @@ xpc_process_channel_activity(struct xpc_partition *part) * at the same time. */ void -xpc_partition_down(struct xpc_partition *part, enum xpc_retval reason) +xpc_partition_going_down(struct xpc_partition *part, enum xpc_retval reason) { unsigned long irq_flags; int ch_number; @@ -1330,12 +1427,11 @@ xpc_partition_down(struct xpc_partition *part, enum xpc_retval reason) } - /* disconnect all channels associated with the downed partition */ + /* disconnect channels associated with the partition going down */ for (ch_number = 0; ch_number < part->nchannels; ch_number++) { ch = &part->channels[ch_number]; - xpc_msgqueue_ref(ch); spin_lock_irqsave(&ch->lock, irq_flags); @@ -1370,6 +1466,7 @@ xpc_teardown_infrastructure(struct xpc_partition *part) * this partition. */ + DBUG_ON(atomic_read(&part->nchannels_engaged) != 0); DBUG_ON(atomic_read(&part->nchannels_active) != 0); DBUG_ON(part->setup_state != XPC_P_SETUP); part->setup_state = XPC_P_WTEARDOWN; @@ -1428,19 +1525,11 @@ xpc_initiate_connect(int ch_number) if (xpc_part_ref(part)) { ch = &part->channels[ch_number]; - if (!(ch->flags & XPC_C_DISCONNECTING)) { - DBUG_ON(ch->flags & XPC_C_OPENREQUEST); - DBUG_ON(ch->flags & XPC_C_CONNECTED); - DBUG_ON(ch->flags & XPC_C_SETUP); - - /* - * Initiate the establishment of a connection - * on the newly registered channel to the - * remote partition. - */ - xpc_wakeup_channel_mgr(part); - } - + /* + * Initiate the establishment of a connection on the + * newly registered channel to the remote partition. + */ + xpc_wakeup_channel_mgr(part); xpc_part_deref(part); } } @@ -1450,9 +1539,6 @@ xpc_initiate_connect(int ch_number) void xpc_connected_callout(struct xpc_channel *ch) { - unsigned long irq_flags; - - /* let the registerer know that a connection has been established */ if (ch->func != NULL) { @@ -1465,10 +1551,6 @@ xpc_connected_callout(struct xpc_channel *ch) dev_dbg(xpc_chan, "ch->func() returned, reason=xpcConnected, " "partid=%d, channel=%d\n", ch->partid, ch->number); } - - spin_lock_irqsave(&ch->lock, irq_flags); - ch->flags |= XPC_C_CONNECTCALLOUT; - spin_unlock_irqrestore(&ch->lock, irq_flags); } @@ -1506,8 +1588,12 @@ xpc_initiate_disconnect(int ch_number) spin_lock_irqsave(&ch->lock, irq_flags); - XPC_DISCONNECT_CHANNEL(ch, xpcUnregistering, + if (!(ch->flags & XPC_C_DISCONNECTED)) { + ch->flags |= XPC_C_WDISCONNECT; + + XPC_DISCONNECT_CHANNEL(ch, xpcUnregistering, &irq_flags); + } spin_unlock_irqrestore(&ch->lock, irq_flags); @@ -1523,8 +1609,9 @@ xpc_initiate_disconnect(int ch_number) /* * To disconnect a channel, and reflect it back to all who may be waiting. * - * >>> An OPEN is not allowed until XPC_C_DISCONNECTING is cleared by - * >>> xpc_free_msgqueues(). + * An OPEN is not allowed until XPC_C_DISCONNECTING is cleared by + * xpc_process_disconnect(), and if set, XPC_C_WDISCONNECT is cleared by + * xpc_disconnect_wait(). * * THE CHANNEL IS TO BE LOCKED BY THE CALLER AND WILL REMAIN LOCKED UPON RETURN. */ @@ -1532,7 +1619,7 @@ void xpc_disconnect_channel(const int line, struct xpc_channel *ch, enum xpc_retval reason, unsigned long *irq_flags) { - u32 flags; + u32 channel_was_connected = (ch->flags & XPC_C_CONNECTED); DBUG_ON(!spin_is_locked(&ch->lock)); @@ -1547,61 +1634,51 @@ xpc_disconnect_channel(const int line, struct xpc_channel *ch, XPC_SET_REASON(ch, reason, line); - flags = ch->flags; + ch->flags |= (XPC_C_CLOSEREQUEST | XPC_C_DISCONNECTING); /* some of these may not have been set */ ch->flags &= ~(XPC_C_OPENREQUEST | XPC_C_OPENREPLY | XPC_C_ROPENREQUEST | XPC_C_ROPENREPLY | XPC_C_CONNECTING | XPC_C_CONNECTED); - ch->flags |= (XPC_C_CLOSEREQUEST | XPC_C_DISCONNECTING); xpc_IPI_send_closerequest(ch, irq_flags); - if (flags & XPC_C_CONNECTED) { + if (channel_was_connected) { ch->flags |= XPC_C_WASCONNECTED; } + spin_unlock_irqrestore(&ch->lock, *irq_flags); + + /* wake all idle kthreads so they can exit */ if (atomic_read(&ch->kthreads_idle) > 0) { - /* wake all idle kthreads so they can exit */ wake_up_all(&ch->idle_wq); } - spin_unlock_irqrestore(&ch->lock, *irq_flags); - - /* wake those waiting to allocate an entry from the local msg queue */ - if (atomic_read(&ch->n_on_msg_allocate_wq) > 0) { wake_up(&ch->msg_allocate_wq); } - /* wake those waiting for notify completion */ - - if (atomic_read(&ch->n_to_notify) > 0) { - xpc_notify_senders(ch, reason, ch->w_local_GP.put); - } - spin_lock_irqsave(&ch->lock, *irq_flags); } void -xpc_disconnected_callout(struct xpc_channel *ch) +xpc_disconnect_callout(struct xpc_channel *ch, enum xpc_retval reason) { /* - * Let the channel's registerer know that the channel is now + * Let the channel's registerer know that the channel is being * disconnected. We don't want to do this if the registerer was never - * informed of a connection being made, unless the disconnect was for - * abnormal reasons. + * informed of a connection being made. */ if (ch->func != NULL) { dev_dbg(xpc_chan, "ch->func() called, reason=%d, partid=%d, " - "channel=%d\n", ch->reason, ch->partid, ch->number); + "channel=%d\n", reason, ch->partid, ch->number); - ch->func(ch->reason, ch->partid, ch->number, NULL, ch->key); + ch->func(reason, ch->partid, ch->number, NULL, ch->key); dev_dbg(xpc_chan, "ch->func() returned, reason=%d, partid=%d, " - "channel=%d\n", ch->reason, ch->partid, ch->number); + "channel=%d\n", reason, ch->partid, ch->number); } } @@ -1754,7 +1831,7 @@ xpc_initiate_allocate(partid_t partid, int ch_number, u32 flags, void **payload) { struct xpc_partition *part = &xpc_partitions[partid]; enum xpc_retval ret = xpcUnknownReason; - struct xpc_msg *msg; + struct xpc_msg *msg = NULL; DBUG_ON(partid <= 0 || partid >= XP_MAX_PARTITIONS); @@ -1848,7 +1925,7 @@ xpc_send_msg(struct xpc_channel *ch, struct xpc_msg *msg, u8 notify_type, xpc_notify_func func, void *key) { enum xpc_retval ret = xpcSuccess; - struct xpc_notify *notify = NULL; // >>> to keep the compiler happy!! + struct xpc_notify *notify = notify; s64 put, msg_number = msg->number; @@ -1875,7 +1952,7 @@ xpc_send_msg(struct xpc_channel *ch, struct xpc_msg *msg, u8 notify_type, notify = &ch->notify_queue[msg_number % ch->local_nentries]; notify->func = func; notify->key = key; - (volatile u8) notify->type = notify_type; + notify->type = notify_type; // >>> is a mb() needed here? @@ -2024,7 +2101,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) enum xpc_retval ret; - if (down_interruptible(&ch->msg_to_pull_sema) != 0) { + if (mutex_lock_interruptible(&ch->msg_to_pull_mutex) != 0) { /* we were interrupted by a signal */ return NULL; } @@ -2060,7 +2137,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) XPC_DEACTIVATE_PARTITION(part, ret); - up(&ch->msg_to_pull_sema); + mutex_unlock(&ch->msg_to_pull_mutex); return NULL; } @@ -2069,7 +2146,7 @@ xpc_pull_remote_msg(struct xpc_channel *ch, s64 get) ch->next_msg_to_pull += nmsgs; } - up(&ch->msg_to_pull_sema); + mutex_unlock(&ch->msg_to_pull_mutex); /* return the message we were looking for */ msg_offset = (get % ch->remote_nentries) * ch->msg_size; diff --git a/arch/ia64/sn/kernel/xpc_main.c b/arch/ia64/sn/kernel/xpc_main.c index 1def91fc8..6b5d4c44f 100644 --- a/arch/ia64/sn/kernel/xpc_main.c +++ b/arch/ia64/sn/kernel/xpc_main.c @@ -1055,6 +1055,8 @@ xpc_do_exit(enum xpc_retval reason) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); } @@ -1215,24 +1217,20 @@ xpc_init(void) partid_t partid; struct xpc_partition *part; pid_t pid; + size_t buf_size; if (!ia64_platform_is("sn2")) { return -ENODEV; } - /* - * xpc_remote_copy_buffer is used as a temporary buffer for bte_copy'ng - * various portions of a partition's reserved page. Its size is based - * on the size of the reserved page header and part_nasids mask. So we - * need to ensure that the other items will fit as well. - */ - if (XPC_RP_VARS_SIZE > XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES) { - dev_err(xpc_part, "xpc_remote_copy_buffer is not big enough\n"); - return -EPERM; - } - DBUG_ON((u64) xpc_remote_copy_buffer != - L1_CACHE_ALIGN((u64) xpc_remote_copy_buffer)); + + buf_size = max(XPC_RP_VARS_SIZE, + XPC_RP_HEADER_SIZE + XP_NASID_MASK_BYTES); + xpc_remote_copy_buffer = xpc_kmalloc_cacheline_aligned(buf_size, + GFP_KERNEL, &xpc_remote_copy_buffer_base); + if (xpc_remote_copy_buffer == NULL) + return -ENOMEM; snprintf(xpc_part->bus_id, BUS_ID_SIZE, "part"); snprintf(xpc_chan->bus_id, BUS_ID_SIZE, "chan"); @@ -1296,6 +1294,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1314,6 +1314,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } @@ -1365,6 +1367,8 @@ xpc_init(void) if (xpc_sysctl) { unregister_sysctl_table(xpc_sysctl); } + + kfree(xpc_remote_copy_buffer_base); return -EBUSY; } diff --git a/arch/ia64/sn/kernel/xpc_partition.c b/arch/ia64/sn/kernel/xpc_partition.c index 2c3c4a8af..57c723f5c 100644 --- a/arch/ia64/sn/kernel/xpc_partition.c +++ b/arch/ia64/sn/kernel/xpc_partition.c @@ -3,7 +3,7 @@ * License. See the file "COPYING" in the main directory of this archive * for more details. * - * Copyright (c) 2004-2005 Silicon Graphics, Inc. All Rights Reserved. + * Copyright (c) 2004-2006 Silicon Graphics, Inc. All Rights Reserved. */ @@ -22,12 +22,13 @@ #include #include #include +#include #include #include #include #include #include -#include "xpc.h" +#include /* XPC is exiting flag */ @@ -43,16 +44,19 @@ static u64 xpc_sh2_IPI_access3; /* original protection values for each node */ -u64 xpc_prot_vec[MAX_COMPACT_NODES]; +u64 xpc_prot_vec[MAX_NUMNODES]; -/* this partition's reserved page */ +/* this partition's reserved page pointers */ struct xpc_rsvd_page *xpc_rsvd_page; - -/* this partition's XPC variables (within the reserved page) */ +static u64 *xpc_part_nasids; +static u64 *xpc_mach_nasids; struct xpc_vars *xpc_vars; struct xpc_vars_part *xpc_vars_part; +static int xp_nasid_mask_bytes; /* actual size in bytes of nasid mask */ +static int xp_nasid_mask_words; /* actual size in words of nasid mask */ + /* * For performance reasons, each entry of xpc_partitions[] is cacheline @@ -64,20 +68,37 @@ struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; /* - * Generic buffer used to store a local copy of the remote partitions - * reserved page or XPC variables. - * - * xpc_discovery runs only once and is a seperate thread that is - * very likely going to be processing in parallel with receiving - * interrupts. + * Generic buffer used to store a local copy of portions of a remote + * partition's reserved page (either its header and part_nasids mask, + * or its vars). */ -char ____cacheline_aligned - xpc_remote_copy_buffer[XPC_RSVD_PAGE_ALIGNED_SIZE]; +char *xpc_remote_copy_buffer; +void *xpc_remote_copy_buffer_base; + +/* + * Guarantee that the kmalloc'd memory is cacheline aligned. + */ +void * +xpc_kmalloc_cacheline_aligned(size_t size, gfp_t flags, void **base) +{ + /* see if kmalloc will give us cachline aligned memory by default */ + *base = kmalloc(size, flags); + if (*base == NULL) { + return NULL; + } + if ((u64) *base == L1_CACHE_ALIGN((u64) *base)) { + return *base; + } + kfree(*base); -/* systune related variables */ -int xpc_hb_interval = XPC_HB_DEFAULT_INTERVAL; -int xpc_hb_check_interval = XPC_HB_CHECK_DEFAULT_TIMEOUT; + /* nope, we'll have to do it ourselves */ + *base = kmalloc(size + L1_CACHE_BYTES, flags); + if (*base == NULL) { + return NULL; + } + return (void *) L1_CACHE_ALIGN((u64) *base); +} /* @@ -85,13 +106,16 @@ int xpc_hb_check_interval = XPC_HB_CHECK_DEFAULT_TIMEOUT; * for that nasid. This function returns 0 on any error. */ static u64 -xpc_get_rsvd_page_pa(int nasid, u64 buf, u64 buf_size) +xpc_get_rsvd_page_pa(int nasid) { bte_result_t bte_res; s64 status; u64 cookie = 0; u64 rp_pa = nasid; /* seed with nasid */ u64 len = 0; + u64 buf = buf; + u64 buf_len = 0; + void *buf_base = NULL; while (1) { @@ -107,13 +131,20 @@ xpc_get_rsvd_page_pa(int nasid, u64 buf, u64 buf_size) break; } - if (len > buf_size) { - dev_err(xpc_part, "len (=0x%016lx) > buf_size\n", len); - status = SALRET_ERROR; - break; + if (L1_CACHE_ALIGN(len) > buf_len) { + kfree(buf_base); + buf_len = L1_CACHE_ALIGN(len); + buf = (u64) xpc_kmalloc_cacheline_aligned(buf_len, + GFP_KERNEL, &buf_base); + if (buf_base == NULL) { + dev_err(xpc_part, "unable to kmalloc " + "len=0x%016lx\n", buf_len); + status = SALRET_ERROR; + break; + } } - bte_res = xp_bte_copy(rp_pa, ia64_tpa(buf), buf_size, + bte_res = xp_bte_copy(rp_pa, buf, buf_len, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bte_res != BTE_SUCCESS) { dev_dbg(xpc_part, "xp_bte_copy failed %i\n", bte_res); @@ -122,6 +153,8 @@ xpc_get_rsvd_page_pa(int nasid, u64 buf, u64 buf_size) } } + kfree(buf_base); + if (status != SALRET_OK) { rp_pa = 0; } @@ -140,15 +173,15 @@ xpc_rsvd_page_init(void) { struct xpc_rsvd_page *rp; AMO_t *amos_page; - u64 rp_pa, next_cl, nasid_array = 0; + u64 rp_pa, nasid_array = 0; int i, ret; /* get the local reserved page's address */ - rp_pa = xpc_get_rsvd_page_pa(cnodeid_to_nasid(0), - (u64) xpc_remote_copy_buffer, - XPC_RSVD_PAGE_ALIGNED_SIZE); + preempt_disable(); + rp_pa = xpc_get_rsvd_page_pa(cpuid_to_nasid(smp_processor_id())); + preempt_enable(); if (rp_pa == 0) { dev_err(xpc_part, "SAL failed to locate the reserved page\n"); return NULL; @@ -163,12 +196,19 @@ xpc_rsvd_page_init(void) rp->version = XPC_RP_VERSION; - /* - * Place the XPC variables on the cache line following the - * reserved page structure. - */ - next_cl = (u64) rp + XPC_RSVD_PAGE_ALIGNED_SIZE; - xpc_vars = (struct xpc_vars *) next_cl; + /* establish the actual sizes of the nasid masks */ + if (rp->SAL_version == 1) { + /* SAL_version 1 didn't set the nasids_size field */ + rp->nasids_size = 128; + } + xp_nasid_mask_bytes = rp->nasids_size; + xp_nasid_mask_words = xp_nasid_mask_bytes / 8; + + /* setup the pointers to the various items in the reserved page */ + xpc_part_nasids = XPC_RP_PART_NASIDS(rp); + xpc_mach_nasids = XPC_RP_MACH_NASIDS(rp); + xpc_vars = XPC_RP_VARS(rp); + xpc_vars_part = XPC_RP_VARS_PART(rp); /* * Before clearing xpc_vars, see if a page of AMOs had been previously @@ -183,7 +223,7 @@ xpc_rsvd_page_init(void) * memory protections are never restricted. */ if ((amos_page = xpc_vars->amos_page) == NULL) { - amos_page = (AMO_t *) mspec_kalloc_page(0); + amos_page = (AMO_t *) TO_AMO(uncached_alloc_page(0)); if (amos_page == NULL) { dev_err(xpc_part, "can't allocate page of AMOs\n"); return NULL; @@ -200,7 +240,8 @@ xpc_rsvd_page_init(void) if (ret != 0) { dev_err(xpc_part, "can't change memory " "protections\n"); - mspec_kfree_page((unsigned long) amos_page); + uncached_free_page(__IA64_UNCACHED_OFFSET | + TO_PHYS((u64) amos_page)); return NULL; } } @@ -219,39 +260,38 @@ xpc_rsvd_page_init(void) amos_page = (AMO_t *) TO_AMO((u64) amos_page); } + /* clear xpc_vars */ memset(xpc_vars, 0, sizeof(struct xpc_vars)); - /* - * Place the XPC per partition specific variables on the cache line - * following the XPC variables structure. - */ - next_cl += XPC_VARS_ALIGNED_SIZE; - memset((u64 *) next_cl, 0, sizeof(struct xpc_vars_part) * - XP_MAX_PARTITIONS); - xpc_vars_part = (struct xpc_vars_part *) next_cl; - xpc_vars->vars_part_pa = __pa(next_cl); - xpc_vars->version = XPC_V_VERSION; xpc_vars->act_nasid = cpuid_to_nasid(0); xpc_vars->act_phys_cpuid = cpu_physical_id(0); + xpc_vars->vars_part_pa = __pa(xpc_vars_part); + xpc_vars->amos_page_pa = ia64_tpa((u64) amos_page); xpc_vars->amos_page = amos_page; /* save for next load of XPC */ - /* - * Initialize the activation related AMO variables. - */ - xpc_vars->act_amos = xpc_IPI_init(XP_MAX_PARTITIONS); - for (i = 1; i < XP_NASID_MASK_WORDS; i++) { - xpc_IPI_init(i + XP_MAX_PARTITIONS); + /* clear xpc_vars_part */ + memset((u64 *) xpc_vars_part, 0, sizeof(struct xpc_vars_part) * + XP_MAX_PARTITIONS); + + /* initialize the activate IRQ related AMO variables */ + for (i = 0; i < xp_nasid_mask_words; i++) { + (void) xpc_IPI_init(XPC_ACTIVATE_IRQ_AMOS + i); } - /* export AMO page's physical address to other partitions */ - xpc_vars->amos_page_pa = ia64_tpa((u64) xpc_vars->amos_page); + + /* initialize the engaged remote partitions related AMO variables */ + (void) xpc_IPI_init(XPC_ENGAGED_PARTITIONS_AMO); + (void) xpc_IPI_init(XPC_DISENGAGE_REQUEST_AMO); + + /* timestamp of when reserved page was setup by XPC */ + rp->stamp = CURRENT_TIME; /* * This signifies to the remote partition that our reserved * page is initialized. */ - (volatile u64) rp->vars_pa = __pa(xpc_vars); + rp->vars_pa = __pa(xpc_vars); return rp; } @@ -385,6 +425,11 @@ xpc_check_remote_hb(void) remote_vars = (struct xpc_vars *) xpc_remote_copy_buffer; for (partid = 1; partid < XP_MAX_PARTITIONS; partid++) { + + if (xpc_exiting) { + break; + } + if (partid == sn_partition_id) { continue; } @@ -398,8 +443,8 @@ xpc_check_remote_hb(void) /* pull the remote_hb cache line */ bres = xp_bte_copy(part->remote_vars_pa, - ia64_tpa((u64) remote_vars), - XPC_VARS_ALIGNED_SIZE, + (u64) remote_vars, + XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { XPC_DEACTIVATE_PARTITION(part, @@ -408,14 +453,14 @@ xpc_check_remote_hb(void) } dev_dbg(xpc_part, "partid = %d, heartbeat = %ld, last_heartbeat" - " = %ld, kdb_status = %ld, HB_mask = 0x%lx\n", partid, - remote_vars->heartbeat, part->last_heartbeat, - remote_vars->kdb_status, + " = %ld, heartbeat_offline = %ld, HB_mask = 0x%lx\n", + partid, remote_vars->heartbeat, part->last_heartbeat, + remote_vars->heartbeat_offline, remote_vars->heartbeating_to_mask); if (((remote_vars->heartbeat == part->last_heartbeat) && - (remote_vars->kdb_status == 0)) || - !XPC_HB_ALLOWED(sn_partition_id, remote_vars)) { + (remote_vars->heartbeat_offline == 0)) || + !xpc_hb_allowed(sn_partition_id, remote_vars)) { XPC_DEACTIVATE_PARTITION(part, xpcNoHeartbeat); continue; @@ -427,31 +472,30 @@ xpc_check_remote_hb(void) /* - * Get a copy of the remote partition's rsvd page. + * Get a copy of a portion of the remote partition's rsvd page. * * remote_rp points to a buffer that is cacheline aligned for BTE copies and - * assumed to be of size XPC_RSVD_PAGE_ALIGNED_SIZE. + * is large enough to contain a copy of their reserved page header and + * part_nasids mask. */ static enum xpc_retval xpc_get_remote_rp(int nasid, u64 *discovered_nasids, - struct xpc_rsvd_page *remote_rp, u64 *remote_rsvd_page_pa) + struct xpc_rsvd_page *remote_rp, u64 *remote_rp_pa) { int bres, i; /* get the reserved page's physical address */ - *remote_rsvd_page_pa = xpc_get_rsvd_page_pa(nasid, (u64) remote_rp, - XPC_RSVD_PAGE_ALIGNED_SIZE); - if (*remote_rsvd_page_pa == 0) { + *remote_rp_pa = xpc_get_rsvd_page_pa(nasid); + if (*remote_rp_pa == 0) { return xpcNoRsvdPageAddr; } - /* pull over the reserved page structure */ - - bres = xp_bte_copy(*remote_rsvd_page_pa, ia64_tpa((u64) remote_rp), - XPC_RSVD_PAGE_ALIGNED_SIZE, + /* pull over the reserved page header and part_nasids mask */ + bres = xp_bte_copy(*remote_rp_pa, (u64) remote_rp, + XPC_RP_HEADER_SIZE + xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { return xpc_map_bte_errors(bres); @@ -459,8 +503,11 @@ xpc_get_remote_rp(int nasid, u64 *discovered_nasids, if (discovered_nasids != NULL) { - for (i = 0; i < XP_NASID_MASK_WORDS; i++) { - discovered_nasids[i] |= remote_rp->part_nasids[i]; + u64 *remote_part_nasids = XPC_RP_PART_NASIDS(remote_rp); + + + for (i = 0; i < xp_nasid_mask_words; i++) { + discovered_nasids[i] |= remote_part_nasids[i]; } } @@ -487,10 +534,10 @@ xpc_get_remote_rp(int nasid, u64 *discovered_nasids, /* - * Get a copy of the remote partition's XPC variables. + * Get a copy of the remote partition's XPC variables from the reserved page. * * remote_vars points to a buffer that is cacheline aligned for BTE copies and - * assumed to be of size XPC_VARS_ALIGNED_SIZE. + * assumed to be of size XPC_RP_VARS_SIZE. */ static enum xpc_retval xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) @@ -502,11 +549,8 @@ xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) return xpcVarsNotSet; } - /* pull over the cross partition variables */ - - bres = xp_bte_copy(remote_vars_pa, ia64_tpa((u64) remote_vars), - XPC_VARS_ALIGNED_SIZE, + bres = xp_bte_copy(remote_vars_pa, (u64) remote_vars, XPC_RP_VARS_SIZE, (BTE_NOTIFY | BTE_WACQUIRE), NULL); if (bres != BTE_SUCCESS) { return xpc_map_bte_errors(bres); @@ -522,7 +566,56 @@ xpc_get_remote_vars(u64 remote_vars_pa, struct xpc_vars *remote_vars) /* - * Prior code has determine the nasid which generated an IPI. Inspect + * Update the remote partition's info. + */ +static void +xpc_update_partition_info(struct xpc_partition *part, u8 remote_rp_version, + struct timespec *remote_rp_stamp, u64 remote_rp_pa, + u64 remote_vars_pa, struct xpc_vars *remote_vars) +{ + part->remote_rp_version = remote_rp_version; + dev_dbg(xpc_part, " remote_rp_version = 0x%016lx\n", + part->remote_rp_version); + + part->remote_rp_stamp = *remote_rp_stamp; + dev_dbg(xpc_part, " remote_rp_stamp (tv_sec = 0x%lx tv_nsec = 0x%lx\n", + part->remote_rp_stamp.tv_sec, part->remote_rp_stamp.tv_nsec); + + part->remote_rp_pa = remote_rp_pa; + dev_dbg(xpc_part, " remote_rp_pa = 0x%016lx\n", part->remote_rp_pa); + + part->remote_vars_pa = remote_vars_pa; + dev_dbg(xpc_part, " remote_vars_pa = 0x%016lx\n", + part->remote_vars_pa); + + part->last_heartbeat = remote_vars->heartbeat; + dev_dbg(xpc_part, " last_heartbeat = 0x%016lx\n", + part->last_heartbeat); + + part->remote_vars_part_pa = remote_vars->vars_part_pa; + dev_dbg(xpc_part, " remote_vars_part_pa = 0x%016lx\n", + part->remote_vars_part_pa); + + part->remote_act_nasid = remote_vars->act_nasid; + dev_dbg(xpc_part, " remote_act_nasid = 0x%x\n", + part->remote_act_nasid); + + part->remote_act_phys_cpuid = remote_vars->act_phys_cpuid; + dev_dbg(xpc_part, " remote_act_phys_cpuid = 0x%x\n", + part->remote_act_phys_cpuid); + + part->remote_amos_page_pa = remote_vars->amos_page_pa; + dev_dbg(xpc_part, " remote_amos_page_pa = 0x%lx\n", + part->remote_amos_page_pa); + + part->remote_vars_version = remote_vars->version; + dev_dbg(xpc_part, " remote_vars_version = 0x%x\n", + part->remote_vars_version); +} + + +/* + * Prior code has determined the nasid which generated an IPI. Inspect * that nasid to determine if its partition needs to be activated or * deactivated. * @@ -540,8 +633,12 @@ xpc_identify_act_IRQ_req(int nasid) { struct xpc_rsvd_page *remote_rp; struct xpc_vars *remote_vars; - u64 remote_rsvd_page_pa; + u64 remote_rp_pa; u64 remote_vars_pa; + int remote_rp_version; + int reactivate = 0; + int stamp_diff; + struct timespec remote_rp_stamp = { 0, 0 }; partid_t partid; struct xpc_partition *part; enum xpc_retval ret; @@ -551,7 +648,7 @@ xpc_identify_act_IRQ_req(int nasid) remote_rp = (struct xpc_rsvd_page *) xpc_remote_copy_buffer; - ret = xpc_get_remote_rp(nasid, NULL, remote_rp, &remote_rsvd_page_pa); + ret = xpc_get_remote_rp(nasid, NULL, remote_rp, &remote_rp_pa); if (ret != xpcSuccess) { dev_warn(xpc_part, "unable to get reserved page from nasid %d, " "which sent interrupt, reason=%d\n", nasid, ret); @@ -559,6 +656,10 @@ xpc_identify_act_IRQ_req(int nasid) } remote_vars_pa = remote_rp->vars_pa; + remote_rp_version = remote_rp->version; + if (XPC_SUPPORTS_RP_STAMP(remote_rp_version)) { + remote_rp_stamp = remote_rp->stamp; + } partid = remote_rp->partid; part = &xpc_partitions[partid]; @@ -584,44 +685,118 @@ xpc_identify_act_IRQ_req(int nasid) "%ld:0x%lx\n", (int) nasid, (int) partid, part->act_IRQ_rcvd, remote_vars->heartbeat, remote_vars->heartbeating_to_mask); + if (xpc_partition_disengaged(part) && + part->act_state == XPC_P_INACTIVE) { - if (part->act_state == XPC_P_INACTIVE) { + xpc_update_partition_info(part, remote_rp_version, + &remote_rp_stamp, remote_rp_pa, + remote_vars_pa, remote_vars); + + if (XPC_SUPPORTS_DISENGAGE_REQUEST(part->remote_vars_version)) { + if (xpc_partition_disengage_requested(1UL << partid)) { + /* + * Other side is waiting on us to disengage, + * even though we already have. + */ + return; + } + } else { + /* other side doesn't support disengage requests */ + xpc_clear_partition_disengage_request(1UL << partid); + } - part->remote_rp_pa = remote_rsvd_page_pa; - dev_dbg(xpc_part, " remote_rp_pa = 0x%016lx\n", - part->remote_rp_pa); + xpc_activate_partition(part); + return; + } - part->remote_vars_pa = remote_vars_pa; - dev_dbg(xpc_part, " remote_vars_pa = 0x%016lx\n", - part->remote_vars_pa); + DBUG_ON(part->remote_rp_version == 0); + DBUG_ON(part->remote_vars_version == 0); + + if (!XPC_SUPPORTS_RP_STAMP(part->remote_rp_version)) { + DBUG_ON(XPC_SUPPORTS_DISENGAGE_REQUEST(part-> + remote_vars_version)); + + if (!XPC_SUPPORTS_RP_STAMP(remote_rp_version)) { + DBUG_ON(XPC_SUPPORTS_DISENGAGE_REQUEST(remote_vars-> + version)); + /* see if the other side rebooted */ + if (part->remote_amos_page_pa == + remote_vars->amos_page_pa && + xpc_hb_allowed(sn_partition_id, + remote_vars)) { + /* doesn't look that way, so ignore the IPI */ + return; + } + } - part->last_heartbeat = remote_vars->heartbeat; - dev_dbg(xpc_part, " last_heartbeat = 0x%016lx\n", - part->last_heartbeat); + /* + * Other side rebooted and previous XPC didn't support the + * disengage request, so we don't need to do anything special. + */ - part->remote_vars_part_pa = remote_vars->vars_part_pa; - dev_dbg(xpc_part, " remote_vars_part_pa = 0x%016lx\n", - part->remote_vars_part_pa); + xpc_update_partition_info(part, remote_rp_version, + &remote_rp_stamp, remote_rp_pa, + remote_vars_pa, remote_vars); + part->reactivate_nasid = nasid; + XPC_DEACTIVATE_PARTITION(part, xpcReactivating); + return; + } - part->remote_act_nasid = remote_vars->act_nasid; - dev_dbg(xpc_part, " remote_act_nasid = 0x%x\n", - part->remote_act_nasid); + DBUG_ON(!XPC_SUPPORTS_DISENGAGE_REQUEST(part->remote_vars_version)); - part->remote_act_phys_cpuid = remote_vars->act_phys_cpuid; - dev_dbg(xpc_part, " remote_act_phys_cpuid = 0x%x\n", - part->remote_act_phys_cpuid); + if (!XPC_SUPPORTS_RP_STAMP(remote_rp_version)) { + DBUG_ON(!XPC_SUPPORTS_DISENGAGE_REQUEST(remote_vars->version)); - part->remote_amos_page_pa = remote_vars->amos_page_pa; - dev_dbg(xpc_part, " remote_amos_page_pa = 0x%lx\n", - part->remote_amos_page_pa); + /* + * Other side rebooted and previous XPC did support the + * disengage request, but the new one doesn't. + */ - xpc_activate_partition(part); + xpc_clear_partition_engaged(1UL << partid); + xpc_clear_partition_disengage_request(1UL << partid); + + xpc_update_partition_info(part, remote_rp_version, + &remote_rp_stamp, remote_rp_pa, + remote_vars_pa, remote_vars); + reactivate = 1; - } else if (part->remote_amos_page_pa != remote_vars->amos_page_pa || - !XPC_HB_ALLOWED(sn_partition_id, remote_vars)) { + } else { + DBUG_ON(!XPC_SUPPORTS_DISENGAGE_REQUEST(remote_vars->version)); + + stamp_diff = xpc_compare_stamps(&part->remote_rp_stamp, + &remote_rp_stamp); + if (stamp_diff != 0) { + DBUG_ON(stamp_diff >= 0); + + /* + * Other side rebooted and the previous XPC did support + * the disengage request, as does the new one. + */ + + DBUG_ON(xpc_partition_engaged(1UL << partid)); + DBUG_ON(xpc_partition_disengage_requested(1UL << + partid)); + + xpc_update_partition_info(part, remote_rp_version, + &remote_rp_stamp, remote_rp_pa, + remote_vars_pa, remote_vars); + reactivate = 1; + } + } + + if (part->disengage_request_timeout > 0 && + !xpc_partition_disengaged(part)) { + /* still waiting on other side to disengage from us */ + return; + } + if (reactivate) { part->reactivate_nasid = nasid; XPC_DEACTIVATE_PARTITION(part, xpcReactivating); + + } else if (XPC_SUPPORTS_DISENGAGE_REQUEST(part->remote_vars_version) && + xpc_partition_disengage_requested(1UL << partid)) { + XPC_DEACTIVATE_PARTITION(part, xpcOtherGoingDown); } } @@ -641,14 +816,17 @@ xpc_identify_act_IRQ_sender(void) u64 nasid; /* remote nasid */ int n_IRQs_detected = 0; AMO_t *act_amos; - struct xpc_rsvd_page *rp = (struct xpc_rsvd_page *) xpc_rsvd_page; - act_amos = xpc_vars->act_amos; + act_amos = xpc_vars->amos_page + XPC_ACTIVATE_IRQ_AMOS; /* scan through act AMO variable looking for non-zero entries */ - for (word = 0; word < XP_NASID_MASK_WORDS; word++) { + for (word = 0; word < xp_nasid_mask_words; word++) { + + if (xpc_exiting) { + break; + } nasid_mask = xpc_IPI_receive(&act_amos[word]); if (nasid_mask == 0) { @@ -666,7 +844,7 @@ xpc_identify_act_IRQ_sender(void) * remote nasid in our reserved pages machine mask. * This is used in the event of module reload. */ - rp->mach_nasids[word] |= nasid_mask; + xpc_mach_nasids[word] |= nasid_mask; /* locate the nasid(s) which sent interrupts */ @@ -685,6 +863,58 @@ xpc_identify_act_IRQ_sender(void) } +/* + * See if the other side has responded to a partition disengage request + * from us. + */ +int +xpc_partition_disengaged(struct xpc_partition *part) +{ + partid_t partid = XPC_PARTID(part); + int disengaged; + + + disengaged = (xpc_partition_engaged(1UL << partid) == 0); + if (part->disengage_request_timeout) { + if (!disengaged) { + if (jiffies < part->disengage_request_timeout) { + /* timelimit hasn't been reached yet */ + return 0; + } + + /* + * Other side hasn't responded to our disengage + * request in a timely fashion, so assume it's dead. + */ + + dev_info(xpc_part, "disengage from remote partition %d " + "timed out\n", partid); + xpc_disengage_request_timedout = 1; + xpc_clear_partition_engaged(1UL << partid); + disengaged = 1; + } + part->disengage_request_timeout = 0; + + /* cancel the timer function, provided it's not us */ + if (!in_interrupt()) { + del_singleshot_timer_sync(&part-> + disengage_request_timer); + } + + DBUG_ON(part->act_state != XPC_P_DEACTIVATING && + part->act_state != XPC_P_INACTIVE); + if (part->act_state != XPC_P_INACTIVE) { + xpc_wakeup_channel_mgr(part); + } + + if (XPC_SUPPORTS_DISENGAGE_REQUEST(part->remote_vars_version)) { + xpc_cancel_partition_disengage_request(part); + } + } + return disengaged; +} + + /* * Mark specified partition as active. */ @@ -719,7 +949,6 @@ xpc_deactivate_partition(const int line, struct xpc_partition *part, enum xpc_retval reason) { unsigned long irq_flags; - partid_t partid = XPC_PARTID(part); spin_lock_irqsave(&part->act_lock, irq_flags); @@ -747,17 +976,27 @@ xpc_deactivate_partition(const int line, struct xpc_partition *part, spin_unlock_irqrestore(&part->act_lock, irq_flags); - XPC_DISALLOW_HB(partid, xpc_vars); + if (XPC_SUPPORTS_DISENGAGE_REQUEST(part->remote_vars_version)) { + xpc_request_partition_disengage(part); + xpc_IPI_send_disengage(part); - dev_dbg(xpc_part, "bringing partition %d down, reason = %d\n", partid, - reason); + /* set a timelimit on the disengage request */ + part->disengage_request_timeout = jiffies + + (xpc_disengage_request_timelimit * HZ); + part->disengage_request_timer.expires = + part->disengage_request_timeout; + add_timer(&part->disengage_request_timer); + } + + dev_dbg(xpc_part, "bringing partition %d down, reason = %d\n", + XPC_PARTID(part), reason); - xpc_partition_down(part, reason); + xpc_partition_going_down(part, reason); } /* - * Mark specified partition as active. + * Mark specified partition as inactive. */ void xpc_mark_partition_inactive(struct xpc_partition *part) @@ -790,9 +1029,10 @@ xpc_discovery(void) void *remote_rp_base; struct xpc_rsvd_page *remote_rp; struct xpc_vars *remote_vars; - u64 remote_rsvd_page_pa; + u64 remote_rp_pa; u64 remote_vars_pa; int region; + int region_size; int max_regions; int nasid; struct xpc_rsvd_page *rp; @@ -802,7 +1042,8 @@ xpc_discovery(void) enum xpc_retval ret; - remote_rp = xpc_kmalloc_cacheline_aligned(XPC_RSVD_PAGE_ALIGNED_SIZE, + remote_rp = xpc_kmalloc_cacheline_aligned(XPC_RP_HEADER_SIZE + + xp_nasid_mask_bytes, GFP_KERNEL, &remote_rp_base); if (remote_rp == NULL) { return; @@ -810,13 +1051,12 @@ xpc_discovery(void) remote_vars = (struct xpc_vars *) remote_rp; - discovered_nasids = kmalloc(sizeof(u64) * XP_NASID_MASK_WORDS, + discovered_nasids = kzalloc(sizeof(u64) * xp_nasid_mask_words, GFP_KERNEL); if (discovered_nasids == NULL) { kfree(remote_rp_base); return; } - memset(discovered_nasids, 0, sizeof(u64) * XP_NASID_MASK_WORDS); rp = (struct xpc_rsvd_page *) xpc_rsvd_page; @@ -825,11 +1065,19 @@ xpc_discovery(void) * nodes that can comprise an access protection grouping. The access * protection is in regards to memory, IOI and IPI. */ -//>>> move the next two #defines into either include/asm-ia64/sn/arch.h or -//>>> include/asm-ia64/sn/addrs.h -#define SH1_MAX_REGIONS 64 -#define SH2_MAX_REGIONS 256 - max_regions = is_shub2() ? SH2_MAX_REGIONS : SH1_MAX_REGIONS; + max_regions = 64; + region_size = sn_region_size; + + switch (region_size) { + case 128: + max_regions *= 2; + case 64: + max_regions *= 2; + case 32: + max_regions *= 2; + region_size = 16; + DBUG_ON(!is_shub2()); + } for (region = 0; region < max_regions; region++) { @@ -839,8 +1087,8 @@ xpc_discovery(void) dev_dbg(xpc_part, "searching region %d\n", region); - for (nasid = (region * sn_region_size * 2); - nasid < ((region + 1) * sn_region_size * 2); + for (nasid = (region * region_size * 2); + nasid < ((region + 1) * region_size * 2); nasid += 2) { if ((volatile int) xpc_exiting) { @@ -850,14 +1098,14 @@ xpc_discovery(void) dev_dbg(xpc_part, "checking nasid %d\n", nasid); - if (XPC_NASID_IN_ARRAY(nasid, rp->part_nasids)) { + if (XPC_NASID_IN_ARRAY(nasid, xpc_part_nasids)) { dev_dbg(xpc_part, "PROM indicates Nasid %d is " "part of the local partition; skipping " "region\n", nasid); break; } - if (!(XPC_NASID_IN_ARRAY(nasid, rp->mach_nasids))) { + if (!(XPC_NASID_IN_ARRAY(nasid, xpc_mach_nasids))) { dev_dbg(xpc_part, "PROM indicates Nasid %d was " "not on Numa-Link network at reset\n", nasid); @@ -875,7 +1123,7 @@ xpc_discovery(void) /* pull over the reserved page structure */ ret = xpc_get_remote_rp(nasid, discovered_nasids, - remote_rp, &remote_rsvd_page_pa); + remote_rp, &remote_rp_pa); if (ret != xpcSuccess) { dev_dbg(xpc_part, "unable to get reserved page " "from nasid %d, reason=%d\n", nasid, @@ -946,6 +1194,13 @@ xpc_discovery(void) remote_vars->act_nasid, remote_vars->act_phys_cpuid); + if (XPC_SUPPORTS_DISENGAGE_REQUEST(remote_vars-> + version)) { + part->remote_amos_page_pa = + remote_vars->amos_page_pa; + xpc_mark_partition_disengaged(part); + xpc_cancel_partition_disengage_request(part); + } xpc_IPI_send_activate(remote_vars); } } @@ -972,12 +1227,12 @@ xpc_initiate_partid_to_nasids(partid_t partid, void *nasid_mask) return xpcPartitionDown; } - part_nasid_pa = part->remote_rp_pa + - (u64) &((struct xpc_rsvd_page *) 0)->part_nasids; + memset(nasid_mask, 0, XP_NASID_MASK_BYTES); - bte_res = xp_bte_copy(part_nasid_pa, ia64_tpa((u64) nasid_mask), - L1_CACHE_ALIGN(XP_NASID_MASK_BYTES), - (BTE_NOTIFY | BTE_WACQUIRE), NULL); + part_nasid_pa = (u64) XPC_RP_PART_NASIDS(part->remote_rp_pa); + + bte_res = xp_bte_copy(part_nasid_pa, (u64) nasid_mask, + xp_nasid_mask_bytes, (BTE_NOTIFY | BTE_WACQUIRE), NULL); return xpc_map_bte_errors(bte_res); } diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig index 747025987..f35dd7b70 100644 --- a/arch/powerpc/Kconfig +++ b/arch/powerpc/Kconfig @@ -620,6 +620,9 @@ config HOTPLUG_CPU Say N if you are unsure. +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y + config KEXEC bool "kexec system call (EXPERIMENTAL)" depends on PPC_MULTIPLATFORM && EXPERIMENTAL diff --git a/arch/powerpc/kernel/cpu_setup_power4.S b/arch/powerpc/kernel/cpu_setup_power4.S index b61d86e7c..55f367edd 100644 --- a/arch/powerpc/kernel/cpu_setup_power4.S +++ b/arch/powerpc/kernel/cpu_setup_power4.S @@ -94,6 +94,8 @@ _GLOBAL(__setup_cpu_ppc970) mfspr r0,SPRN_HID0 li r11,5 /* clear DOZE and SLEEP */ rldimi r0,r11,52,8 /* set NAP and DPM */ + li r11,0 + rldimi r0,r11,32,31 /* clear EN_ATTN */ mtspr SPRN_HID0,r0 mfspr r0,SPRN_HID0 mfspr r0,SPRN_HID0 diff --git a/arch/powerpc/kernel/crash.c b/arch/powerpc/kernel/crash.c index 778f22fd8..dbcb85994 100644 --- a/arch/powerpc/kernel/crash.c +++ b/arch/powerpc/kernel/crash.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include @@ -174,6 +175,8 @@ static void crash_kexec_prepare_cpus(void) void default_machine_crash_shutdown(struct pt_regs *regs) { + unsigned int irq; + /* * This function is only called after the system * has paniced or is otherwise in a critical state. @@ -186,6 +189,16 @@ void default_machine_crash_shutdown(struct pt_regs *regs) */ local_irq_disable(); + for_each_irq(irq) { + struct irq_desc *desc = irq_descp(irq); + + if (desc->status & IRQ_INPROGRESS) + desc->handler->end(irq); + + if (!(desc->status & IRQ_DISABLED)) + desc->handler->disable(irq); + } + if (ppc_md.kexec_cpu_down) ppc_md.kexec_cpu_down(1, 0); diff --git a/arch/powerpc/platforms/pseries/xics.c b/arch/powerpc/platforms/pseries/xics.c index 2d60ea30f..9ff7105bc 100644 --- a/arch/powerpc/platforms/pseries/xics.c +++ b/arch/powerpc/platforms/pseries/xics.c @@ -641,23 +641,28 @@ void xics_teardown_cpu(int secondary) ops->cppr_info(cpu, 0x00); iosync(); + /* + * Clear IPI + */ + ops->qirr_info(cpu, 0xff); + /* + * we need to EOI the IPI if we got here from kexec down IPI + * + * probably need to check all the other interrupts too + * should we be flagging idle loop instead? + * or creating some task to be scheduled? + */ + ops->xirr_info_set(cpu, XICS_IPI); + /* * Some machines need to have at least one cpu in the GIQ, * so leave the master cpu in the group. */ - if (secondary) { - /* - * we need to EOI the IPI if we got here from kexec down IPI - * - * probably need to check all the other interrupts too - * should we be flagging idle loop instead? - * or creating some task to be scheduled? - */ - ops->xirr_info_set(cpu, XICS_IPI); + if (secondary) rtas_set_indicator(GLOBAL_INTERRUPT_QUEUE, (1UL << interrupt_server_size) - 1 - default_distrib_server, 0); - } + } #ifdef CONFIG_HOTPLUG_CPU diff --git a/arch/sparc/kernel/sys_sparc.c b/arch/sparc/kernel/sys_sparc.c index 8fa541cc2..5a220d61b 100644 --- a/arch/sparc/kernel/sys_sparc.c +++ b/arch/sparc/kernel/sys_sparc.c @@ -220,6 +220,21 @@ out: return err; } +int sparc_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) +{ + if (ARCH_SUN4C_SUN4 && + (len > 0x20000000 || + ((flags & MAP_FIXED) && + addr < 0xe0000000 && addr + len > 0x20000000))) + return -EINVAL; + + /* See asm-sparc/uaccess.h */ + if (len > TASK_SIZE - PAGE_SIZE || addr + len > TASK_SIZE - PAGE_SIZE) + return -EINVAL; + + return 0; +} + /* Linux version of mmap */ static unsigned long do_mmap2(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, @@ -234,25 +249,13 @@ static unsigned long do_mmap2(unsigned long addr, unsigned long len, goto out; } - retval = -EINVAL; len = PAGE_ALIGN(len); - if (ARCH_SUN4C_SUN4 && - (len > 0x20000000 || - ((flags & MAP_FIXED) && - addr < 0xe0000000 && addr + len > 0x20000000))) - goto out_putf; - - /* See asm-sparc/uaccess.h */ - if (len > TASK_SIZE - PAGE_SIZE || addr + len > TASK_SIZE - PAGE_SIZE) - goto out_putf; - flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); down_write(¤t->mm->mmap_sem); retval = do_mmap_pgoff(file, addr, len, prot, flags, pgoff); up_write(¤t->mm->mmap_sem); -out_putf: if (file) fput(file); out: diff --git a/arch/sparc64/kernel/sys_sparc.c b/arch/sparc64/kernel/sys_sparc.c index 62d7497fb..ef419255c 100644 --- a/arch/sparc64/kernel/sys_sparc.c +++ b/arch/sparc64/kernel/sys_sparc.c @@ -550,6 +550,26 @@ asmlinkage long sparc64_personality(unsigned long personality) return ret; } +int sparc64_mmap_check(unsigned long addr, unsigned long len, + unsigned long flags) +{ + if (test_thread_flag(TIF_32BIT)) { + if (len >= STACK_TOP32) + return -EINVAL; + + if ((flags & MAP_FIXED) && addr > STACK_TOP32 - len) + return -EINVAL; + } else { + if (len >= VA_EXCLUDE_START) + return -EINVAL; + + if ((flags & MAP_FIXED) && invalid_64bit_range(addr, len)) + return -EINVAL; + } + + return 0; +} + /* Linux version of mmap */ asmlinkage unsigned long sys_mmap(unsigned long addr, unsigned long len, unsigned long prot, unsigned long flags, unsigned long fd, @@ -565,27 +585,11 @@ asmlinkage unsigned long sys_mmap(unsigned long addr, unsigned long len, } flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE); len = PAGE_ALIGN(len); - retval = -EINVAL; - - if (test_thread_flag(TIF_32BIT)) { - if (len >= STACK_TOP32) - goto out_putf; - - if ((flags & MAP_FIXED) && addr > STACK_TOP32 - len) - goto out_putf; - } else { - if (len >= VA_EXCLUDE_START) - goto out_putf; - - if ((flags & MAP_FIXED) && invalid_64bit_range(addr, len)) - goto out_putf; - } down_write(¤t->mm->mmap_sem); retval = do_mmap(file, addr, len, prot, flags, off); up_write(¤t->mm->mmap_sem); -out_putf: if (file) fput(file); out: diff --git a/arch/sparc64/mm/generic.c b/arch/sparc64/mm/generic.c index 8cb06205d..af9d81db0 100644 --- a/arch/sparc64/mm/generic.c +++ b/arch/sparc64/mm/generic.c @@ -69,6 +69,8 @@ static inline void io_remap_pte_range(struct mm_struct *mm, pte_t * pte, } else offset += PAGE_SIZE; + if (pte_write(entry)) + entry = pte_mkdirty(entry); do { BUG_ON(!pte_none(*pte)); set_pte_at(mm, address, pte, entry); diff --git a/arch/um/kernel/physmem.c b/arch/um/kernel/physmem.c index fc0f0b085..bd4627b4d 100644 --- a/arch/um/kernel/physmem.c +++ b/arch/um/kernel/physmem.c @@ -226,7 +226,7 @@ EXPORT_SYMBOL(physmem_forget_descriptor); EXPORT_SYMBOL(physmem_remove_mapping); EXPORT_SYMBOL(physmem_subst_mapping); -void arch_free_page(struct page *page, int order) +int arch_free_page(struct page *page, int order) { void *virt; int i; @@ -235,6 +235,8 @@ void arch_free_page(struct page *page, int order) virt = __va(page_to_phys(page + i)); physmem_remove_mapping(virt); } + + return 0; } int is_remapped(void *virt) diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig index b2cd1f245..0f4ed7159 100644 --- a/arch/x86_64/Kconfig +++ b/arch/x86_64/Kconfig @@ -123,6 +123,22 @@ config GENERIC_CPU endchoice +config X86_64_XEN + bool "Enable Xen compatible kernel" + select SWIOTLB + help + This option will compile a kernel compatible with Xen hypervisor + +config X86_NO_TSS + bool + depends on X86_64_XEN + default y + +config X86_NO_IDT + bool + depends on X86_64_XEN + default y + # # Define implied options from the CPU selection here # @@ -143,6 +159,7 @@ config X86_INTERNODE_CACHE_BYTES config X86_TSC bool + depends on !X86_64_XEN default y config X86_GOOD_APIC @@ -185,7 +202,7 @@ config X86_CPUID config X86_HT bool - depends on SMP && !MK8 + depends on SMP && !MK8 && !X86_64_XEN default y config MATH_EMULATION @@ -199,14 +216,22 @@ config EISA config X86_IO_APIC bool + depends !XEN_UNPRIVILEGED_GUEST default y +config X86_XEN_GENAPIC + bool + depends X86_64_XEN + default XEN_PRIVILEGED_GUEST || SMP + config X86_LOCAL_APIC bool + depends !XEN_UNPRIVILEGED_GUEST default y config MTRR bool "MTRR (Memory Type Range Register) support" + depends on !XEN_UNPRIVILEGED_GUEST ---help--- On Intel P6 family processors (Pentium Pro, Pentium II and later) the Memory Type Range Registers (MTRRs) may be used to control @@ -247,7 +272,7 @@ config SMP config SCHED_SMT bool "SMT (Hyperthreading) scheduler support" - depends on SMP + depends on SMP && !X86_64_XEN default n help SMT scheduler support improves the CPU scheduler's decision making @@ -257,7 +282,7 @@ config SCHED_SMT config SCHED_MC bool "Multi-core scheduler support" - depends on SMP + depends on SMP && !X86_64_XEN default y help Multi-core scheduler support improves the CPU scheduler's decision @@ -268,7 +293,7 @@ source "kernel/Kconfig.preempt" config NUMA bool "Non Uniform Memory Access (NUMA) Support" - depends on SMP + depends on SMP && !X86_64_XEN help Enable NUMA (Non Uniform Memory Access) support. The kernel will try to allocate memory used by a CPU on the local memory @@ -328,7 +353,7 @@ config ARCH_DISCONTIGMEM_DEFAULT config ARCH_SPARSEMEM_ENABLE def_bool y - depends on (NUMA || EXPERIMENTAL) + depends on (NUMA || EXPERIMENTAL) && !X86_64_XEN config ARCH_MEMORY_PROBE def_bool y @@ -352,6 +377,7 @@ config NR_CPUS int "Maximum number of CPUs (2-256)" range 2 255 depends on SMP + default "16" if X86_64_XEN default "8" help This allows you to specify the maximum number of CPUs which this @@ -369,9 +395,12 @@ config HOTPLUG_CPU can be controlled through /sys/devices/system/cpu/cpu#. Say N if you want to disable CPU hotplug. +config ARCH_ENABLE_MEMORY_HOTPLUG + def_bool y config HPET_TIMER bool + depends on !X86_64_XEN default y help Use the IA-PC HPET (High Precision Event Timer) to manage @@ -389,7 +418,7 @@ config GART_IOMMU bool "K8 GART IOMMU support" default y select SWIOTLB - depends on PCI + depends on PCI && !X86_64_XEN help Support for hardware IOMMU in AMD's Opteron/Athlon64 Processors and for the bounce buffering software IOMMU. @@ -409,6 +438,7 @@ config SWIOTLB config X86_MCE bool "Machine check support" if EMBEDDED + depends on !X86_64_XEN default y help Include a machine check error handler to report hardware errors. @@ -434,7 +464,7 @@ config X86_MCE_AMD config KEXEC bool "kexec system call (EXPERIMENTAL)" - depends on EXPERIMENTAL + depends on EXPERIMENTAL && !X86_64_XEN help kexec is a system call that implements the ability to shutdown your current kernel, and to start another kernel. It is like a reboot @@ -525,8 +555,11 @@ config GENERIC_PENDING_IRQ default y menu "Power management options" + depends on !XEN_UNPRIVILEGED_GUEST +if !X86_64_XEN source kernel/power/Kconfig +endif source "drivers/acpi/Kconfig" @@ -549,6 +582,21 @@ config PCI_MMCONFIG bool "Support mmconfig PCI config space access" depends on PCI && ACPI +config XEN_PCIDEV_FRONTEND + bool "Xen PCI Frontend" + depends on PCI && X86_64_XEN + default y + help + The PCI device frontend driver allows the kernel to import arbitrary + PCI devices from a PCI backend to support PCI driver domains. + +config XEN_PCIDEV_FE_DEBUG + bool "Xen PCI Frontend Debugging" + depends on XEN_PCIDEV_FRONTEND + default n + help + Enables some debug statements within the PCI Frontend. + source "drivers/pci/pcie/Kconfig" source "drivers/pci/Kconfig" @@ -621,4 +669,6 @@ source "security/Kconfig" source "crypto/Kconfig" +source "drivers/xen/Kconfig" + source "lib/Kconfig" diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index e573e2ab5..4b81d9f4c 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -31,6 +31,10 @@ cflags-$(CONFIG_MK8) += $(call cc-option,-march=k8) cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona) cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic) +cppflags-$(CONFIG_XEN) += \ + -D__XEN_INTERFACE_VERSION__=$(CONFIG_XEN_INTERFACE_VERSION) +CPPFLAGS += $(cppflags-y) + cflags-y += -m64 cflags-y += -mno-red-zone cflags-y += -mcmodel=kernel @@ -72,6 +76,21 @@ boot := arch/x86_64/boot PHONY += bzImage bzlilo install archmrproper \ fdimage fdimage144 fdimage288 isoimage archclean +ifdef CONFIG_XEN +CPPFLAGS := -Iinclude$(if $(KBUILD_SRC),2)/asm/mach-xen $(CPPFLAGS) +head-y := arch/x86_64/kernel/head-xen.o arch/x86_64/kernel/head64-xen.o arch/x86_64/kernel/init_task.o +LDFLAGS_vmlinux := -e _start +boot := arch/i386/boot-xen +.PHONY: vmlinuz +#Default target when executing "make" +all: vmlinuz + +vmlinuz: vmlinux + $(Q)$(MAKE) $(build)=$(boot) $@ + +install: + $(Q)$(MAKE) $(build)=$(boot) XENGUEST=$(XENGUEST) $@ +else #Default target when executing "make" all: bzImage @@ -92,6 +111,7 @@ fdimage fdimage144 fdimage288 isoimage: vmlinux install: $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(BOOTIMAGE) $@ +endif archclean: $(Q)$(MAKE) $(clean)=$(boot) diff --git a/arch/x86_64/ia32/Makefile b/arch/x86_64/ia32/Makefile index e9263b497..a84151e4c 100644 --- a/arch/x86_64/ia32/Makefile +++ b/arch/x86_64/ia32/Makefile @@ -23,9 +23,25 @@ quiet_cmd_syscall = SYSCALL $@ -Wl,-soname=linux-gate.so.1 -o $@ \ -Wl,-T,$(filter-out FORCE,$^) +$(obj)/vsyscall-int80.so \ $(obj)/vsyscall-sysenter.so $(obj)/vsyscall-syscall.so: \ $(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE $(call if_changed,syscall) -AFLAGS_vsyscall-sysenter.o = -m32 -Wa,-32 -AFLAGS_vsyscall-syscall.o = -m32 -Wa,-32 +AFLAGS_vsyscall-sysenter.o = -m32 -Wa,-32 -Iarch/i386/kernel +AFLAGS_vsyscall-syscall.o = -m32 -Wa,-32 -Iarch/i386/kernel + +ifdef CONFIG_XEN +AFLAGS_vsyscall-int80.o = -m32 -Wa,-32 -Iarch/i386/kernel +CFLAGS_syscall32-xen.o += -DUSE_INT80 +AFLAGS_syscall32_syscall-xen.o += -DUSE_INT80 + +$(obj)/syscall32_syscall-xen.o: \ + $(foreach F,int80 sysenter syscall,$(obj)/vsyscall-$F.so) + +targets := $(foreach F,int80 sysenter syscall,vsyscall-$F.o vsyscall-$F.so) + +include $(srctree)/scripts/Makefile.xen + +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/x86_64/ia32/vsyscall-sigreturn.S b/arch/x86_64/ia32/vsyscall-sigreturn.S index 1384367cd..6086ac6fe 100644 --- a/arch/x86_64/ia32/vsyscall-sigreturn.S +++ b/arch/x86_64/ia32/vsyscall-sigreturn.S @@ -139,5 +139,5 @@ __kernel_rt_sigreturn: .align 4 .LENDFDE3: -#include "../../i386/kernel/vsyscall-note.S" +#include diff --git a/arch/x86_64/kernel/Makefile b/arch/x86_64/kernel/Makefile index 059c88313..b0913f3ef 100644 --- a/arch/x86_64/kernel/Makefile +++ b/arch/x86_64/kernel/Makefile @@ -20,11 +20,13 @@ obj-$(CONFIG_MICROCODE) += microcode.o obj-$(CONFIG_X86_CPUID) += cpuid.o obj-$(CONFIG_SMP) += smp.o smpboot.o trampoline.o obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o +obj-$(CONFIG_X86_XEN_GENAPIC) += genapic.o genapic_xen.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o mpparse.o \ genapic.o genapic_cluster.o genapic_flat.o obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o obj-$(CONFIG_CRASH_DUMP) += crash_dump.o -obj-$(CONFIG_PM) += suspend.o +obj-$(CONFIG_SOFTWARE_SUSPEND) += suspend.o +obj-$(CONFIG_ACPI_SLEEP) += suspend.o obj-$(CONFIG_SOFTWARE_SUSPEND) += suspend_asm.o obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o @@ -49,3 +51,18 @@ intel_cacheinfo-y += ../../i386/kernel/cpu/intel_cacheinfo.o quirks-y += ../../i386/kernel/quirks.o i8237-y += ../../i386/kernel/i8237.o msr-$(subst m,y,$(CONFIG_X86_MSR)) += ../../i386/kernel/msr.o + +ifdef CONFIG_XEN +time-y += ../../i386/kernel/time-xen.o +pci-dma-y += ../../i386/kernel/pci-dma-xen.o +microcode-$(subst m,y,$(CONFIG_MICROCODE)) := ../../i386/kernel/microcode-xen.o +quirks-y := ../../i386/kernel/quirks-xen.o + +n-obj-xen := i8259.o reboot.o i8237.o smpboot.o trampoline.o + +include $(srctree)/scripts/Makefile.xen + +obj-y := $(call filterxen, $(obj-y), $(n-obj-xen)) +obj-y := $(call cherrypickxen, $(obj-y)) +extra-y := $(call cherrypickxen, $(extra-y)) +endif diff --git a/arch/x86_64/kernel/acpi/Makefile b/arch/x86_64/kernel/acpi/Makefile index 4fe97071f..aa84f6eb9 100644 --- a/arch/x86_64/kernel/acpi/Makefile +++ b/arch/x86_64/kernel/acpi/Makefile @@ -6,3 +6,4 @@ ifneq ($(CONFIG_ACPI_PROCESSOR),) obj-y += processor.o endif +boot-$(CONFIG_XEN) := ../../../i386/kernel/acpi/boot-xen.o diff --git a/arch/x86_64/kernel/asm-offsets.c b/arch/x86_64/kernel/asm-offsets.c index 38834bbba..06cf05867 100644 --- a/arch/x86_64/kernel/asm-offsets.c +++ b/arch/x86_64/kernel/asm-offsets.c @@ -66,7 +66,9 @@ int main(void) DEFINE(pbe_address, offsetof(struct pbe, address)); DEFINE(pbe_orig_address, offsetof(struct pbe, orig_address)); DEFINE(pbe_next, offsetof(struct pbe, next)); +#ifndef CONFIG_X86_NO_TSS BLANK(); DEFINE(TSS_ist, offsetof(struct tss_struct, ist)); +#endif return 0; } diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index 586b34c00..4c599eea1 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -596,7 +596,7 @@ retint_kernel: */ .macro apicinterrupt num,func INTR_FRAME - pushq $\num-256 + pushq $~(\num) CFI_ADJUST_CFA_OFFSET 8 interrupt \func jmp ret_from_intr diff --git a/arch/x86_64/kernel/init_task.c b/arch/x86_64/kernel/init_task.c index ce31d904d..9c141e1a6 100644 --- a/arch/x86_64/kernel/init_task.c +++ b/arch/x86_64/kernel/init_task.c @@ -37,6 +37,8 @@ union thread_union init_thread_union struct task_struct init_task = INIT_TASK(init_task); EXPORT_SYMBOL(init_task); + +#ifndef CONFIG_X86_NO_TSS /* * per-CPU TSS segments. Threads are completely 'soft' on Linux, * no more per-task TSS's. The TSS size is kept cacheline-aligned @@ -45,5 +47,6 @@ EXPORT_SYMBOL(init_task); * on exact cacheline boundaries, to eliminate cacheline ping-pong. */ DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS; +#endif #define ALIGN_TO_4K __attribute__((section(".data.init_task"))) diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c index d8bd0b345..c87488532 100644 --- a/arch/x86_64/kernel/irq.c +++ b/arch/x86_64/kernel/irq.c @@ -91,8 +91,8 @@ skip: */ asmlinkage unsigned int do_IRQ(struct pt_regs *regs) { - /* high bits used in ret_from_ code */ - unsigned irq = regs->orig_rax & 0xff; + /* high bit used in ret_from_ code */ + unsigned irq = ~regs->orig_rax; exit_idle(); irq_enter(); diff --git a/arch/x86_64/kernel/pmtimer.c b/arch/x86_64/kernel/pmtimer.c index feb5f108d..07340aecf 100644 --- a/arch/x86_64/kernel/pmtimer.c +++ b/arch/x86_64/kernel/pmtimer.c @@ -60,7 +60,9 @@ int pmtimer_mark_offset(void) delta = cyc2us((tick - last_pmtmr_tick) & ACPI_PM_MASK); last_pmtmr_tick = tick; +#ifndef CONFIG_XEN monotonic_base += delta * NSEC_PER_USEC; +#endif delta += offset_delay; @@ -68,7 +70,7 @@ int pmtimer_mark_offset(void) offset_delay = delta % (USEC_PER_SEC / HZ); rdtscll(tsc); - vxtime.last_tsc = tsc - offset_delay * cpu_khz; + vxtime.last_tsc = tsc - offset_delay * (u64)cpu_khz / 1000; /* don't calculate delay for first run, or if we've got less then a tick */ @@ -80,6 +82,32 @@ int pmtimer_mark_offset(void) return lost - 1; } +static unsigned pmtimer_wait_tick(void) +{ + u32 a, b; + for (a = b = inl(pmtmr_ioport) & ACPI_PM_MASK; + a == b; + b = inl(pmtmr_ioport) & ACPI_PM_MASK) + cpu_relax(); + return b; +} + +/* note: wait time is rounded up to one tick */ +void pmtimer_wait(unsigned us) +{ + u32 a, b; + a = pmtimer_wait_tick(); + do { + b = inl(pmtmr_ioport); + cpu_relax(); + } while (cyc2us(b - a) < us); +} + +void pmtimer_resume(void) +{ + last_pmtmr_tick = inl(pmtmr_ioport); +} + unsigned int do_gettimeoffset_pm(void) { u32 now, offset, delta = 0; @@ -95,7 +123,7 @@ unsigned int do_gettimeoffset_pm(void) static int __init nopmtimer_setup(char *s) { pmtmr_ioport = 0; - return 0; + return 1; } __setup("nopmtimer", nopmtimer_setup); diff --git a/arch/x86_64/kernel/setup.c b/arch/x86_64/kernel/setup.c index 655b9192e..e8d5f8441 100644 --- a/arch/x86_64/kernel/setup.c +++ b/arch/x86_64/kernel/setup.c @@ -1440,7 +1440,7 @@ struct seq_operations cpuinfo_op = { .show = show_cpuinfo, }; -#ifdef CONFIG_INPUT_PCSPKR +#if defined(CONFIG_INPUT_PCSPKR) || defined(CONFIG_INPUT_PCSPKR_MODULE) #include static __init int add_pcspkr(void) { diff --git a/arch/x86_64/kernel/setup64.c b/arch/x86_64/kernel/setup64.c index 8a691fa6d..189a7372e 100644 --- a/arch/x86_64/kernel/setup64.c +++ b/arch/x86_64/kernel/setup64.c @@ -46,7 +46,7 @@ Control non executable mappings for 64bit processes. on Enable(default) off Disable */ -int __init nonx_setup(char *str) +void __init nonx_setup(const char *str) { if (!strncmp(str, "on", 2)) { __supported_pte_mask |= _PAGE_NX; @@ -55,28 +55,7 @@ int __init nonx_setup(char *str) do_not_nx = 1; __supported_pte_mask &= ~_PAGE_NX; } - return 1; -} -__setup("noexec=", nonx_setup); /* parsed early actually */ - -int force_personality32 = 0; - -/* noexec32=on|off -Control non executable heap for 32bit processes. -To control the stack too use noexec=off - -on PROT_READ does not imply PROT_EXEC for 32bit processes -off PROT_READ implies PROT_EXEC (default) -*/ -static int __init nonx32_setup(char *str) -{ - if (!strcmp(str, "on")) - force_personality32 &= ~READ_IMPLIES_EXEC; - else if (!strcmp(str, "off")) - force_personality32 |= READ_IMPLIES_EXEC; - return 1; } -__setup("noexec32=", nonx32_setup); /* * Great future plan: diff --git a/arch/x86_64/kernel/smp.c b/arch/x86_64/kernel/smp.c index 4a6628b14..e13938d24 100644 --- a/arch/x86_64/kernel/smp.c +++ b/arch/x86_64/kernel/smp.c @@ -135,10 +135,10 @@ asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs) cpu = smp_processor_id(); /* - * orig_rax contains the interrupt vector - 256. + * orig_rax contains the negated interrupt vector. * Use that to determine where the sender put the data. */ - sender = regs->orig_rax + 256 - INVALIDATE_TLB_VECTOR_START; + sender = ~regs->orig_rax - INVALIDATE_TLB_VECTOR_START; f = &per_cpu(flush_state, sender); if (!cpu_isset(cpu, f->flush_cpumask)) diff --git a/arch/x86_64/mm/Makefile b/arch/x86_64/mm/Makefile index d25ac86fe..7d6ceaef4 100644 --- a/arch/x86_64/mm/Makefile +++ b/arch/x86_64/mm/Makefile @@ -9,3 +9,13 @@ obj-$(CONFIG_K8_NUMA) += k8topology.o obj-$(CONFIG_ACPI_NUMA) += srat.o hugetlbpage-y = ../../i386/mm/hugetlbpage.o + +ifdef CONFIG_XEN +include $(srctree)/scripts/Makefile.xen + +ioremap-y += ../../i386/mm/ioremap-xen.o +hypervisor-y += ../../i386/mm/hypervisor.o +obj-y += hypervisor.o + +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/x86_64/oprofile/Makefile b/arch/x86_64/oprofile/Makefile index 6be32683e..589a7966e 100644 --- a/arch/x86_64/oprofile/Makefile +++ b/arch/x86_64/oprofile/Makefile @@ -11,9 +11,12 @@ DRIVER_OBJS = $(addprefix ../../../drivers/oprofile/, \ oprofilefs.o oprofile_stats.o \ timer_int.o ) +ifdef CONFIG_XEN +OPROFILE-y := xenoprof.o +else OPROFILE-y := init.o backtrace.o OPROFILE-$(CONFIG_X86_LOCAL_APIC) += nmi_int.o op_model_athlon.o op_model_p4.o \ op_model_ppro.o OPROFILE-$(CONFIG_X86_IO_APIC) += nmi_timer_int.o - +endif oprofile-y = $(DRIVER_OBJS) $(addprefix ../../i386/oprofile/, $(OPROFILE-y)) diff --git a/arch/x86_64/pci/Makefile b/arch/x86_64/pci/Makefile index a3f6ad570..1ea6ea9b6 100644 --- a/arch/x86_64/pci/Makefile +++ b/arch/x86_64/pci/Makefile @@ -15,11 +15,23 @@ obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o direct.o obj-$(CONFIG_NUMA) += k8-bus.o +# pcifront should be after mmconfig.o and direct.o as it should only +# take over if direct access to the PCI bus is unavailable +obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront.o + direct-y += ../../i386/pci/direct.o acpi-y += ../../i386/pci/acpi.o +pcifront-y += ../../i386/pci/pcifront.o legacy-y += ../../i386/pci/legacy.o irq-y += ../../i386/pci/irq.o common-y += ../../i386/pci/common.o fixup-y += ../../i386/pci/fixup.o i386-y += ../../i386/pci/i386.o init-y += ../../i386/pci/init.o + +ifdef CONFIG_XEN +irq-y := ../../i386/pci/irq-xen.o +include $(srctree)/scripts/Makefile.xen + +obj-y := $(call cherrypickxen, $(obj-y)) +endif diff --git a/arch/x86_64/pci/mmconfig.c b/arch/x86_64/pci/mmconfig.c index a2060e4d5..3c55c76c6 100644 --- a/arch/x86_64/pci/mmconfig.c +++ b/arch/x86_64/pci/mmconfig.c @@ -13,7 +13,10 @@ #include "pci.h" -#define MMCONFIG_APER_SIZE (256*1024*1024) +/* aperture is up to 256MB but BIOS may reserve less */ +#define MMCONFIG_APER_MIN (2 * 1024*1024) +#define MMCONFIG_APER_MAX (256 * 1024*1024) + /* Verify the first 16 busses. We assume that systems with more busses get MCFG right. */ #define MAX_CHECK_BUS 16 @@ -175,9 +178,10 @@ void __init pci_mmcfg_init(void) return; if (!e820_all_mapped(pci_mmcfg_config[0].base_address, - pci_mmcfg_config[0].base_address + MMCONFIG_APER_SIZE, + pci_mmcfg_config[0].base_address + MMCONFIG_APER_MIN, E820_RESERVED)) { - printk(KERN_ERR "PCI: BIOS Bug: MCFG area is not E820-reserved\n"); + printk(KERN_ERR "PCI: BIOS Bug: MCFG area at %x is not E820-reserved\n", + pci_mmcfg_config[0].base_address); printk(KERN_ERR "PCI: Not using MMCONFIG.\n"); return; } @@ -190,7 +194,8 @@ void __init pci_mmcfg_init(void) } for (i = 0; i < pci_mmcfg_config_num; ++i) { pci_mmcfg_virt[i].cfg = &pci_mmcfg_config[i]; - pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, MMCONFIG_APER_SIZE); + pci_mmcfg_virt[i].virt = ioremap_nocache(pci_mmcfg_config[i].base_address, + MMCONFIG_APER_MAX); if (!pci_mmcfg_virt[i].virt) { printk("PCI: Cannot map mmconfig aperture for segment %d\n", pci_mmcfg_config[i].pci_segment_group_number); diff --git a/block/elevator.c b/block/elevator.c index a0afdd317..7be96bbe1 100644 --- a/block/elevator.c +++ b/block/elevator.c @@ -766,7 +766,8 @@ void elv_unregister(struct elevator_type *e) read_lock(&tasklist_lock); do_each_thread(g, p) { task_lock(p); - e->ops.trim(p->io_context); + if (p->io_context) + e->ops.trim(p->io_context); task_unlock(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); diff --git a/block/ll_rw_blk.c b/block/ll_rw_blk.c index 7eb36c53f..4293143c9 100644 --- a/block/ll_rw_blk.c +++ b/block/ll_rw_blk.c @@ -638,7 +638,7 @@ void blk_queue_bounce_limit(request_queue_t *q, u64 dma_addr) /* Assume anything <= 4GB can be handled by IOMMU. Actually some IOMMUs can handle everything, but I don't know of a way to test this here. */ - if (bounce_pfn < (0xffffffff>>PAGE_SHIFT)) + if (bounce_pfn < (min_t(u64,0xffffffff,BLK_BOUNCE_HIGH) >> PAGE_SHIFT)) dma = 1; q->bounce_pfn = max_low_pfn; #else diff --git a/configs/kernel-2.6.17-i686-planetlab.config b/configs/kernel-2.6.17-i686-planetlab.config index 3d0b18a22..2acdd7a0d 100644 --- a/configs/kernel-2.6.17-i686-planetlab.config +++ b/configs/kernel-2.6.17-i686-planetlab.config @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-1.2142_FC4-1.planetlab -# Mon Aug 21 16:39:45 2006 +# Linux kernel version: 2.6.17-1.2142_FC4.3 +# Wed Oct 18 22:35:27 2006 # CONFIG_X86_32=y CONFIG_SEMAPHORE_SLEEPERS=y @@ -93,6 +93,7 @@ CONFIG_DEFAULT_IOSCHED="cfq" # # CONFIG_SMP is not set CONFIG_X86_PC=y +# CONFIG_X86_XEN is not set # CONFIG_X86_ELAN is not set # CONFIG_X86_VOYAGER is not set # CONFIG_X86_NUMAQ is not set @@ -156,6 +157,7 @@ CONFIG_X86_MCE_P4THERMAL=y # CONFIG_MICROCODE is not set # CONFIG_X86_MSR is not set # CONFIG_X86_CPUID is not set +# CONFIG_SWIOTLB is not set # # Firmware Drivers @@ -194,6 +196,7 @@ CONFIG_HZ=1000 CONFIG_KEXEC=y CONFIG_CRASH_DUMP=y CONFIG_PHYSICAL_START=0x100000 +CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y # # Power management options (ACPI, APM) @@ -243,6 +246,7 @@ CONFIG_PCI=y # CONFIG_PCI_GOBIOS is not set # CONFIG_PCI_GOMMCONFIG is not set # CONFIG_PCI_GODIRECT is not set +# CONFIG_PCI_GOXEN_FE is not set CONFIG_PCI_GOANY=y CONFIG_PCI_BIOS=y CONFIG_PCI_DIRECT=y @@ -887,7 +891,7 @@ CONFIG_SK98LIN=m CONFIG_VIA_VELOCITY=m CONFIG_TIGON3=m CONFIG_BNX2=m -CONFIG_BCM5700=m +# CONFIG_BCM5700 is not set # # Ethernet (10000 Mbit) diff --git a/configs/kernel-2.6.17-i686-smp-planetlab.config b/configs/kernel-2.6.17-i686-smp-planetlab.config index e4887618c..3fe7f151a 100644 --- a/configs/kernel-2.6.17-i686-smp-planetlab.config +++ b/configs/kernel-2.6.17-i686-smp-planetlab.config @@ -1,7 +1,7 @@ # # Automatically generated make config: don't edit -# Linux kernel version: 2.6.17-1.2142_FC4-1.planetlab -# Mon Aug 21 16:40:40 2006 +# Linux kernel version: 2.6.17-1.2142_FC4.3smp +# Wed Oct 18 22:53:28 2006 # CONFIG_X86_32=y CONFIG_SEMAPHORE_SLEEPERS=y @@ -95,6 +95,7 @@ CONFIG_DEFAULT_IOSCHED="cfq" # CONFIG_SMP=y # CONFIG_X86_PC is not set +# CONFIG_X86_XEN is not set # CONFIG_X86_ELAN is not set # CONFIG_X86_VOYAGER is not set # CONFIG_X86_NUMAQ is not set @@ -161,6 +162,7 @@ CONFIG_X86_MCE_P4THERMAL=y # CONFIG_MICROCODE is not set # CONFIG_X86_MSR is not set # CONFIG_X86_CPUID is not set +# CONFIG_SWIOTLB is not set # # Firmware Drivers @@ -200,6 +202,7 @@ CONFIG_KEXEC=y CONFIG_CRASH_DUMP=y CONFIG_PHYSICAL_START=0x100000 CONFIG_HOTPLUG_CPU=y +CONFIG_ARCH_ENABLE_MEMORY_HOTPLUG=y # # Power management options (ACPI, APM) @@ -250,6 +253,7 @@ CONFIG_PCI=y # CONFIG_PCI_GOBIOS is not set # CONFIG_PCI_GOMMCONFIG is not set # CONFIG_PCI_GODIRECT is not set +# CONFIG_PCI_GOXEN_FE is not set CONFIG_PCI_GOANY=y CONFIG_PCI_BIOS=y CONFIG_PCI_DIRECT=y @@ -894,7 +898,7 @@ CONFIG_SK98LIN=m CONFIG_VIA_VELOCITY=m CONFIG_TIGON3=m CONFIG_BNX2=m -CONFIG_BCM5700=m +# CONFIG_BCM5700 is not set # # Ethernet (10000 Mbit) diff --git a/drivers/Makefile b/drivers/Makefile index 447d8e688..26b29563d 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -31,6 +31,7 @@ obj-y += base/ block/ misc/ mfd/ net/ media/ obj-$(CONFIG_NUBUS) += nubus/ obj-$(CONFIG_ATM) += atm/ obj-$(CONFIG_PPC_PMAC) += macintosh/ +obj-$(CONFIG_XEN) += xen/ obj-$(CONFIG_IDE) += ide/ obj-$(CONFIG_FC4) += fc4/ obj-$(CONFIG_SCSI) += scsi/ diff --git a/drivers/acpi/Kconfig b/drivers/acpi/Kconfig index c24652d31..f0c892613 100644 --- a/drivers/acpi/Kconfig +++ b/drivers/acpi/Kconfig @@ -46,7 +46,7 @@ if ACPI config ACPI_SLEEP bool "Sleep States" - depends on X86 && (!SMP || SUSPEND_SMP) + depends on X86 && (!SMP || SUSPEND_SMP) && !XEN depends on PM default y ---help--- @@ -300,6 +300,7 @@ config ACPI_SYSTEM config X86_PM_TIMER bool "Power Management Timer Support" if EMBEDDED depends on X86 + depends on !XEN default y help The Power Management Timer is available on all ACPI-capable, diff --git a/drivers/acpi/namespace/nsxfeval.c b/drivers/acpi/namespace/nsxfeval.c index a95f636dc..d7791fbc7 100644 --- a/drivers/acpi/namespace/nsxfeval.c +++ b/drivers/acpi/namespace/nsxfeval.c @@ -238,8 +238,9 @@ acpi_evaluate_object(acpi_handle handle, ACPI_ERROR((AE_INFO, "Both Handle and Pathname are NULL")); } else { - ACPI_ERROR((AE_INFO, - "Handle is NULL and Pathname is relative")); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, + "Null Handle with relative pathname [%s]", + pathname)); } status = AE_BAD_PARAMETER; diff --git a/drivers/block/aoe/aoenet.c b/drivers/block/aoe/aoenet.c index 9e6f51c52..c1434ed11 100644 --- a/drivers/block/aoe/aoenet.c +++ b/drivers/block/aoe/aoenet.c @@ -92,18 +92,6 @@ mac_addr(char addr[6]) return __be64_to_cpu(n); } -static struct sk_buff * -skb_check(struct sk_buff *skb) -{ - if (skb_is_nonlinear(skb)) - if ((skb = skb_share_check(skb, GFP_ATOMIC))) - if (skb_linearize(skb, GFP_ATOMIC) < 0) { - dev_kfree_skb(skb); - return NULL; - } - return skb; -} - void aoenet_xmit(struct sk_buff *sl) { @@ -120,19 +108,18 @@ aoenet_xmit(struct sk_buff *sl) * (1) len doesn't include the header by default. I want this. */ static int -aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt) +aoenet_rcv(struct sk_buff *skb, struct net_device *ifp, struct packet_type *pt, struct net_device *orig_dev) { struct aoe_hdr *h; u32 n; - skb = skb_check(skb); - if (!skb) + skb = skb_share_check(skb, GFP_ATOMIC); + if (skb == NULL) return 0; - + if (skb_linearize(skb)) + goto exit; if (!is_aoe_netif(ifp)) goto exit; - - //skb->len += ETH_HLEN; /* (1) */ skb_push(skb, ETH_HLEN); /* (1) */ h = (struct aoe_hdr *) skb->mac.raw; diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c index 3170eaa25..fdd37e407 100644 --- a/drivers/cdrom/cdrom.c +++ b/drivers/cdrom/cdrom.c @@ -1838,7 +1838,7 @@ static int dvd_read_bca(struct cdrom_device_info *cdi, dvd_struct *s) init_cdrom_command(&cgc, buf, sizeof(buf), CGC_DATA_READ); cgc.cmd[0] = GPCMD_READ_DVD_STRUCTURE; cgc.cmd[7] = s->type; - cgc.cmd[9] = cgc.buflen = 0xff; + cgc.cmd[9] = cgc.buflen & 0xff; if ((ret = cdo->generic_packet(cdi, &cgc))) return ret; diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 6a7aa9314..418da028c 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -36,6 +36,22 @@ # include #endif +static inline int range_is_allowed(unsigned long from, unsigned long to) +{ + unsigned long cursor; + + cursor = from >> PAGE_SHIFT; + while ((cursor << PAGE_SHIFT) < to) { + if (!devmem_is_allowed(cursor)) { + printk ("Program %s tried to read /dev/mem between %lx->%lx.\n", + current->comm, from, to); + return 0; + } + cursor++; + } + return 1; +} + /* * Architectures vary in how they handle caching for addresses * outside of main memory. @@ -103,22 +119,7 @@ static inline int valid_mmap_phys_addr_range(unsigned long addr, size_t size) } #endif -static inline int range_is_allowed(unsigned long from, unsigned long to) -{ - unsigned long cursor; - - cursor = from >> PAGE_SHIFT; - while ((cursor << PAGE_SHIFT) < to) { - if (!devmem_is_allowed(cursor)) { - printk ("Program %s tried to read /dev/mem between %lx->%lx.\n", - current->comm, from, to); - return 0; - } - cursor++; - } - return 1; -} - +#ifndef ARCH_HAS_DEV_MEM /* * This funcion reads the *physical* memory. The f_pos points directly to the * memory location. @@ -245,6 +246,7 @@ static ssize_t write_mem(struct file * file, const char __user * buf, *ppos += written; return written; } +#endif #ifndef __HAVE_PHYS_MEM_ACCESS_PROT static pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn, @@ -681,6 +683,7 @@ static int open_port(struct inode * inode, struct file * filp) #define open_kmem open_mem #define open_oldmem open_mem +#ifndef ARCH_HAS_DEV_MEM static struct file_operations mem_fops = { .llseek = memory_lseek, .read = read_mem, @@ -688,6 +691,9 @@ static struct file_operations mem_fops = { .mmap = mmap_mem, .open = open_mem, }; +#else +extern struct file_operations mem_fops; +#endif static struct file_operations kmem_fops = { .llseek = memory_lseek, diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index 8ea70625f..2678034be 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -424,6 +424,7 @@ static irqreturn_t tis_int_handler(int irq, void *dev_id, struct pt_regs *regs) iowrite32(interrupt, chip->vendor.iobase + TPM_INT_STATUS(chip->vendor.locality)); + ioread32(chip->vendor.iobase + TPM_INT_STATUS(chip->vendor.locality)); return IRQ_HANDLED; } diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c index 0b31cf25c..12c15826a 100644 --- a/drivers/char/tty_io.c +++ b/drivers/char/tty_io.c @@ -133,6 +133,8 @@ LIST_HEAD(tty_drivers); /* linked list of tty drivers */ vt.c for deeply disgusting hack reasons */ DEFINE_MUTEX(tty_mutex); +int console_use_vt = 1; + #ifdef CONFIG_UNIX98_PTYS extern struct tty_driver *ptm_driver; /* Unix98 pty masters; for /dev/ptmx */ extern int pty_limit; /* Config limit on Unix98 ptys */ @@ -1674,19 +1676,6 @@ release_mem_out: goto end_init; } -/* - * Get a copy of the termios structure for the driver/index - */ -void tty_get_termios(struct tty_driver *driver, int idx, struct termios *tio) -{ - lock_kernel(); - if (driver->termios[idx]) - *tio = *driver->termios[idx]; - else - *tio = driver->init_termios; - unlock_kernel(); -} - /* * Releases memory associated with a tty structure, and clears out the * driver table slots. @@ -2074,7 +2063,7 @@ retry_open: goto got_driver; } #ifdef CONFIG_VT - if (device == MKDEV(TTY_MAJOR,0)) { + if (console_use_vt && (device == MKDEV(TTY_MAJOR,0))) { extern struct tty_driver *console_driver; driver = console_driver; index = fg_console; @@ -2795,7 +2784,7 @@ static void flush_to_ldisc(void *private_) struct tty_struct *tty = (struct tty_struct *) private_; unsigned long flags; struct tty_ldisc *disc; - struct tty_buffer *tbuf; + struct tty_buffer *tbuf, *head; int count; char *char_buf; unsigned char *flag_buf; @@ -2812,7 +2801,9 @@ static void flush_to_ldisc(void *private_) goto out; } spin_lock_irqsave(&tty->buf.lock, flags); - while((tbuf = tty->buf.head) != NULL) { + head = tty->buf.head; + tty->buf.head = NULL; + while((tbuf = head) != NULL) { while ((count = tbuf->commit - tbuf->read) != 0) { char_buf = tbuf->char_buf_ptr + tbuf->read; flag_buf = tbuf->flag_buf_ptr + tbuf->read; @@ -2821,10 +2812,12 @@ static void flush_to_ldisc(void *private_) disc->receive_buf(tty, char_buf, flag_buf, count); spin_lock_irqsave(&tty->buf.lock, flags); } - if (tbuf->active) + if (tbuf->active) { + tty->buf.head = head; break; - tty->buf.head = tbuf->next; - if (tty->buf.head == NULL) + } + head = tbuf->next; + if (head == NULL) tty->buf.tail = NULL; tty_buffer_free(tty, tbuf); } @@ -3277,6 +3270,8 @@ static int __init tty_init(void) #endif #ifdef CONFIG_VT + if (!console_use_vt) + goto out_vt; cdev_init(&vc0_cdev, &console_fops); if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) || register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0) @@ -3285,6 +3280,7 @@ static int __init tty_init(void) class_device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0"); vty_init(); + out_vt: #endif return 0; } diff --git a/drivers/firmware/Kconfig b/drivers/firmware/Kconfig index 1e371a510..f0dff5ac1 100644 --- a/drivers/firmware/Kconfig +++ b/drivers/firmware/Kconfig @@ -8,7 +8,7 @@ menu "Firmware Drivers" config EDD tristate "BIOS Enhanced Disk Drive calls determine boot disk (EXPERIMENTAL)" depends on EXPERIMENTAL - depends on !IA64 + depends on !IA64 && !XEN help Say Y or M here if you want to enable BIOS Enhanced Disk Drive Services real mode BIOS calls to determine which disk diff --git a/drivers/i2c/busses/scx200_acb.c b/drivers/i2c/busses/scx200_acb.c index 766cc969c..9b4f4ee4f 100644 --- a/drivers/i2c/busses/scx200_acb.c +++ b/drivers/i2c/busses/scx200_acb.c @@ -181,21 +181,21 @@ static void scx200_acb_machine(struct scx200_acb_iface *iface, u8 status) break; case state_read: - /* Set ACK if receiving the last byte */ - if (iface->len == 1) + /* Set ACK if _next_ byte will be the last one */ + if (iface->len == 2) outb(inb(ACBCTL1) | ACBCTL1_ACK, ACBCTL1); else outb(inb(ACBCTL1) & ~ACBCTL1_ACK, ACBCTL1); - *iface->ptr++ = inb(ACBSDA); - --iface->len; - - if (iface->len == 0) { + if (iface->len == 1) { iface->result = 0; iface->state = state_idle; outb(inb(ACBCTL1) | ACBCTL1_STOP, ACBCTL1); } + *iface->ptr++ = inb(ACBSDA); + --iface->len; + break; case state_write: @@ -304,8 +304,12 @@ static s32 scx200_acb_smbus_xfer(struct i2c_adapter *adapter, buffer = (u8 *)&cur_word; break; - case I2C_SMBUS_BLOCK_DATA: + case I2C_SMBUS_I2C_BLOCK_DATA: + if (rw == I2C_SMBUS_READ) + data->block[0] = I2C_SMBUS_BLOCK_MAX; /* For now */ len = data->block[0]; + if (len == 0 || len > I2C_SMBUS_BLOCK_MAX) + return -EINVAL; buffer = &data->block[1]; break; @@ -369,7 +373,7 @@ static u32 scx200_acb_func(struct i2c_adapter *adapter) { return I2C_FUNC_SMBUS_QUICK | I2C_FUNC_SMBUS_BYTE | I2C_FUNC_SMBUS_BYTE_DATA | I2C_FUNC_SMBUS_WORD_DATA | - I2C_FUNC_SMBUS_BLOCK_DATA; + I2C_FUNC_SMBUS_I2C_BLOCK; } /* For now, we only handle combined mode (smbus) */ diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 45e2cdf54..2e79137dc 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -756,9 +756,9 @@ int i2c_probe(struct i2c_adapter *adapter, "parameter for adapter %d, " "addr 0x%02x\n", adap_id, address_data->ignore[j + 1]); + ignore = 1; + break; } - ignore = 1; - break; } if (ignore) continue; diff --git a/drivers/ide/ide-lib.c b/drivers/ide/ide-lib.c index 41d46dbe6..b09a6537c 100644 --- a/drivers/ide/ide-lib.c +++ b/drivers/ide/ide-lib.c @@ -410,10 +410,10 @@ void ide_toggle_bounce(ide_drive_t *drive, int on) { u64 addr = BLK_BOUNCE_HIGH; /* dma64_addr_t */ - if (!PCI_DMA_BUS_IS_PHYS) { - addr = BLK_BOUNCE_ANY; - } else if (on && drive->media == ide_disk) { - if (HWIF(drive)->pci_dev) + if (on && drive->media == ide_disk) { + if (!PCI_DMA_BUS_IS_PHYS) + addr = BLK_BOUNCE_ANY; + else if (HWIF(drive)->pci_dev) addr = HWIF(drive)->pci_dev->dma_mask; } diff --git a/drivers/ide/pci/via82cxxx.c b/drivers/ide/pci/via82cxxx.c index 3e677c4f8..9914a78a5 100644 --- a/drivers/ide/pci/via82cxxx.c +++ b/drivers/ide/pci/via82cxxx.c @@ -6,7 +6,7 @@ * * vt82c576, vt82c586, vt82c586a, vt82c586b, vt82c596a, vt82c596b, * vt82c686, vt82c686a, vt82c686b, vt8231, vt8233, vt8233c, vt8233a, - * vt8235, vt8237 + * vt8235, vt8237, vt8237a * * Copyright (c) 2000-2002 Vojtech Pavlik * @@ -82,6 +82,7 @@ static struct via_isa_bridge { { "vt6410", PCI_DEVICE_ID_VIA_6410, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8251", PCI_DEVICE_ID_VIA_8251, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8237", PCI_DEVICE_ID_VIA_8237, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, + { "vt8237a", PCI_DEVICE_ID_VIA_8237A, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8235", PCI_DEVICE_ID_VIA_8235, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8233a", PCI_DEVICE_ID_VIA_8233A, 0x00, 0x2f, VIA_UDMA_133 | VIA_BAD_AST }, { "vt8233c", PCI_DEVICE_ID_VIA_8233C_0, 0x00, 0x2f, VIA_UDMA_100 }, diff --git a/drivers/ieee1394/ohci1394.c b/drivers/ieee1394/ohci1394.c index 8f1292c36..d1788cdf0 100644 --- a/drivers/ieee1394/ohci1394.c +++ b/drivers/ieee1394/ohci1394.c @@ -3548,6 +3548,8 @@ static int ohci1394_pci_resume (struct pci_dev *pdev) static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state) { + pci_save_state(pdev); + #ifdef CONFIG_PPC_PMAC if (machine_is(powermac)) { struct device_node *of_node; @@ -3559,8 +3561,6 @@ static int ohci1394_pci_suspend (struct pci_dev *pdev, pm_message_t state) } #endif - pci_save_state(pdev); - return 0; } diff --git a/drivers/ieee1394/sbp2.c b/drivers/ieee1394/sbp2.c index 5413dc43b..e084dbf76 100644 --- a/drivers/ieee1394/sbp2.c +++ b/drivers/ieee1394/sbp2.c @@ -2541,6 +2541,9 @@ static int sbp2scsi_slave_configure(struct scsi_device *sdev) sdev->skip_ms_page_8 = 1; if (scsi_id->workarounds & SBP2_WORKAROUND_FIX_CAPACITY) sdev->fix_capacity = 1; + if (scsi_id->ne->guid_vendor_id == 0x0010b9 && /* Maxtor's OUI */ + (sdev->type == TYPE_DISK || sdev->type == TYPE_RBC)) + sdev->allow_restart = 1; return 0; } diff --git a/drivers/infiniband/hw/mthca/mthca_reset.c b/drivers/infiniband/hw/mthca/mthca_reset.c index 8ea801271..f4fddd532 100644 --- a/drivers/infiniband/hw/mthca/mthca_reset.c +++ b/drivers/infiniband/hw/mthca/mthca_reset.c @@ -37,6 +37,7 @@ #include #include #include +#include #include "mthca_dev.h" #include "mthca_cmd.h" @@ -48,6 +49,12 @@ int mthca_reset(struct mthca_dev *mdev) u32 *hca_header = NULL; u32 *bridge_header = NULL; struct pci_dev *bridge = NULL; + int bridge_pcix_cap = 0; + int hca_pcie_cap = 0; + int hca_pcix_cap = 0; + + u16 devctl; + u16 linkctl; #define MTHCA_RESET_OFFSET 0xf0010 #define MTHCA_RESET_VALUE swab32(1) @@ -71,8 +78,8 @@ int mthca_reset(struct mthca_dev *mdev) bridge)) != NULL) { if (bridge->hdr_type == PCI_HEADER_TYPE_BRIDGE && bridge->subordinate == mdev->pdev->bus) { - mthca_dbg(mdev, "Found bridge: %s (%s)\n", - pci_pretty_name(bridge), pci_name(bridge)); + mthca_dbg(mdev, "Found bridge: %s\n", + pci_name(bridge)); break; } } @@ -83,8 +90,8 @@ int mthca_reset(struct mthca_dev *mdev) * assume we're in no-bridge mode and hope for * the best. */ - mthca_warn(mdev, "No bridge found for %s (%s)\n", - pci_pretty_name(mdev->pdev), pci_name(mdev->pdev)); + mthca_warn(mdev, "No bridge found for %s\n", + pci_name(mdev->pdev)); } } @@ -109,6 +116,9 @@ int mthca_reset(struct mthca_dev *mdev) } } + hca_pcix_cap = pci_find_capability(mdev->pdev, PCI_CAP_ID_PCIX); + hca_pcie_cap = pci_find_capability(mdev->pdev, PCI_CAP_ID_EXP); + if (bridge) { bridge_header = kmalloc(256, GFP_KERNEL); if (!bridge_header) { @@ -128,6 +138,13 @@ int mthca_reset(struct mthca_dev *mdev) goto out; } } + bridge_pcix_cap = pci_find_capability(bridge, PCI_CAP_ID_PCIX); + if (!bridge_pcix_cap) { + err = -ENODEV; + mthca_err(mdev, "Couldn't locate HCA bridge " + "PCI-X capability, aborting.\n"); + goto out; + } } /* actually hit reset */ @@ -177,6 +194,20 @@ int mthca_reset(struct mthca_dev *mdev) good: /* Now restore the PCI headers */ if (bridge) { + if (pci_write_config_dword(bridge, bridge_pcix_cap + 0x8, + bridge_header[(bridge_pcix_cap + 0x8) / 4])) { + err = -ENODEV; + mthca_err(mdev, "Couldn't restore HCA bridge Upstream " + "split transaction control, aborting.\n"); + goto out; + } + if (pci_write_config_dword(bridge, bridge_pcix_cap + 0xc, + bridge_header[(bridge_pcix_cap + 0xc) / 4])) { + err = -ENODEV; + mthca_err(mdev, "Couldn't restore HCA bridge Downstream " + "split transaction control, aborting.\n"); + goto out; + } /* * Bridge control register is at 0x3e, so we'll * naturally restore it last in this loop. @@ -202,6 +233,35 @@ good: } } + if (hca_pcix_cap) { + if (pci_write_config_dword(mdev->pdev, hca_pcix_cap, + hca_header[hca_pcix_cap / 4])) { + err = -ENODEV; + mthca_err(mdev, "Couldn't restore HCA PCI-X " + "command register, aborting.\n"); + goto out; + } + } + + if (hca_pcie_cap) { + devctl = hca_header[(hca_pcie_cap + PCI_EXP_DEVCTL) / 4]; + if (pci_write_config_word(mdev->pdev, hca_pcie_cap + PCI_EXP_DEVCTL, + devctl)) { + err = -ENODEV; + mthca_err(mdev, "Couldn't restore HCA PCI Express " + "Device Control register, aborting.\n"); + goto out; + } + linkctl = hca_header[(hca_pcie_cap + PCI_EXP_LNKCTL) / 4]; + if (pci_write_config_word(mdev->pdev, hca_pcie_cap + PCI_EXP_LNKCTL, + linkctl)) { + err = -ENODEV; + mthca_err(mdev, "Couldn't restore HCA PCI Express " + "Link control register, aborting.\n"); + goto out; + } + } + for (i = 0; i < 16; ++i) { if (i * 4 == PCI_COMMAND) continue; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 70208c3d2..1d917edcf 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -1,5 +1,7 @@ /* * Copyright (c) 2004, 2005 Topspin Communications. All rights reserved. + * Copyright (c) 2005 Sun Microsystems, Inc. All rights reserved. + * Copyright (c) 2004 Voltaire, Inc. All rights reserved. * * This software is available to you under a choice of one of two * licenses. You may choose to be licensed under the terms of the GNU @@ -41,6 +43,8 @@ #include #include +#include + #include "ipoib.h" #ifdef CONFIG_INFINIBAND_IPOIB_DEBUG @@ -51,7 +55,7 @@ MODULE_PARM_DESC(mcast_debug_level, "Enable multicast debug tracing if > 0"); #endif -static DECLARE_MUTEX(mcast_mutex); +static DEFINE_MUTEX(mcast_mutex); /* Used for all multicast joins (broadcast, IPv4 mcast and IPv6 mcast) */ struct ipoib_mcast { @@ -93,8 +97,7 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast) struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_neigh *neigh, *tmp; unsigned long flags; - LIST_HEAD(ah_list); - struct ipoib_ah *ah, *tah; + int tx_dropped = 0; ipoib_dbg_mcast(netdev_priv(dev), "deleting multicast group " IPOIB_GID_FMT "\n", @@ -103,28 +106,31 @@ static void ipoib_mcast_free(struct ipoib_mcast *mcast) spin_lock_irqsave(&priv->lock, flags); list_for_each_entry_safe(neigh, tmp, &mcast->neigh_list, list) { + /* + * It's safe to call ipoib_put_ah() inside priv->lock + * here, because we know that mcast->ah will always + * hold one more reference, so ipoib_put_ah() will + * never do more than decrement the ref count. + */ if (neigh->ah) - list_add_tail(&neigh->ah->list, &ah_list); - *to_ipoib_neigh(neigh->neighbour) = NULL; - neigh->neighbour->ops->destructor = NULL; - kfree(neigh); + ipoib_put_ah(neigh->ah); + ipoib_neigh_free(neigh); } spin_unlock_irqrestore(&priv->lock, flags); - list_for_each_entry_safe(ah, tah, &ah_list, list) - ipoib_put_ah(ah); - if (mcast->ah) ipoib_put_ah(mcast->ah); while (!skb_queue_empty(&mcast->pkt_queue)) { - struct sk_buff *skb = skb_dequeue(&mcast->pkt_queue); - - skb->dev = dev; - dev_kfree_skb_any(skb); + ++tx_dropped; + dev_kfree_skb_any(skb_dequeue(&mcast->pkt_queue)); } + spin_lock_irqsave(&priv->tx_lock, flags); + priv->stats.tx_dropped += tx_dropped; + spin_unlock_irqrestore(&priv->tx_lock, flags); + kfree(mcast); } @@ -133,26 +139,18 @@ static struct ipoib_mcast *ipoib_mcast_alloc(struct net_device *dev, { struct ipoib_mcast *mcast; - mcast = kmalloc(sizeof (*mcast), can_sleep ? GFP_KERNEL : GFP_ATOMIC); + mcast = kzalloc(sizeof *mcast, can_sleep ? GFP_KERNEL : GFP_ATOMIC); if (!mcast) return NULL; - memset(mcast, 0, sizeof (*mcast)); - - init_completion(&mcast->done); - mcast->dev = dev; mcast->created = jiffies; - mcast->backoff = HZ; - mcast->logcount = 0; + mcast->backoff = 1; INIT_LIST_HEAD(&mcast->list); INIT_LIST_HEAD(&mcast->neigh_list); skb_queue_head_init(&mcast->pkt_queue); - mcast->ah = NULL; - mcast->query = NULL; - return mcast; } @@ -213,6 +211,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, { struct net_device *dev = mcast->dev; struct ipoib_dev_priv *priv = netdev_priv(dev); + struct ipoib_ah *ah; int ret; mcast->mcmember = *mcmember; @@ -251,6 +250,7 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, .port_num = priv->port, .sl = mcast->mcmember.sl, .ah_flags = IB_AH_GRH, + .static_rate = mcast->mcmember.rate, .grh = { .flow_label = be32_to_cpu(mcast->mcmember.flow_label), .hop_limit = mcast->mcmember.hop_limit, @@ -258,19 +258,10 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, .traffic_class = mcast->mcmember.traffic_class } }; - int path_rate = ib_sa_rate_enum_to_int(mcast->mcmember.rate); - av.grh.dgid = mcast->mcmember.mgid; - if (path_rate > 0 && priv->local_rate > path_rate) - av.static_rate = (priv->local_rate - 1) / path_rate; - - ipoib_dbg_mcast(priv, "static_rate %d for local port %dX, mcmember %dX\n", - av.static_rate, priv->local_rate, - ib_sa_rate_enum_to_int(mcast->mcmember.rate)); - - mcast->ah = ipoib_create_ah(dev, priv->pd, &av); - if (!mcast->ah) { + ah = ipoib_create_ah(dev, priv->pd, &av); + if (!ah) { ipoib_warn(priv, "ib_address_create failed\n"); } else { ipoib_dbg_mcast(priv, "MGID " IPOIB_GID_FMT @@ -280,11 +271,17 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, be16_to_cpu(mcast->mcmember.mlid), mcast->mcmember.sl); } + + spin_lock_irq(&priv->lock); + mcast->ah = ah; + spin_unlock_irq(&priv->lock); } /* actually send any queued packets */ + spin_lock_irq(&priv->tx_lock); while (!skb_queue_empty(&mcast->pkt_queue)) { struct sk_buff *skb = skb_dequeue(&mcast->pkt_queue); + spin_unlock_irq(&priv->tx_lock); skb->dev = dev; @@ -295,7 +292,9 @@ static int ipoib_mcast_join_finish(struct ipoib_mcast *mcast, if (dev_queue_xmit(skb)) ipoib_warn(priv, "dev_queue_xmit failed to requeue packet\n"); + spin_lock_irq(&priv->tx_lock); } + spin_unlock_irq(&priv->tx_lock); return 0; } @@ -307,6 +306,7 @@ ipoib_mcast_sendonly_join_complete(int status, { struct ipoib_mcast *mcast = mcast_ptr; struct net_device *dev = mcast->dev; + struct ipoib_dev_priv *priv = netdev_priv(dev); if (!status) ipoib_mcast_join_finish(mcast, mcmember); @@ -317,13 +317,12 @@ ipoib_mcast_sendonly_join_complete(int status, IPOIB_GID_ARG(mcast->mcmember.mgid), status); /* Flush out any queued packets */ + spin_lock_irq(&priv->tx_lock); while (!skb_queue_empty(&mcast->pkt_queue)) { - struct sk_buff *skb = skb_dequeue(&mcast->pkt_queue); - - skb->dev = dev; - - dev_kfree_skb_any(skb); + ++priv->stats.tx_dropped; + dev_kfree_skb_any(skb_dequeue(&mcast->pkt_queue)); } + spin_unlock_irq(&priv->tx_lock); /* Clear the busy flag so we try again */ clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags); @@ -357,7 +356,9 @@ static int ipoib_mcast_sendonly_join(struct ipoib_mcast *mcast) rec.mgid = mcast->mcmember.mgid; rec.port_gid = priv->local_gid; - rec.pkey = be16_to_cpu(priv->pkey); + rec.pkey = cpu_to_be16(priv->pkey); + + init_completion(&mcast->done); ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec, IB_SA_MCMEMBER_REC_MGID | @@ -394,11 +395,11 @@ static void ipoib_mcast_join_complete(int status, IPOIB_GID_ARG(mcast->mcmember.mgid), status); if (!status && !ipoib_mcast_join_finish(mcast, mcmember)) { - mcast->backoff = HZ; - down(&mcast_mutex); + mcast->backoff = 1; + mutex_lock(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_work(ipoib_workqueue, &priv->mcast_task); - up(&mcast_mutex); + mutex_unlock(&mcast_mutex); complete(&mcast->done); return; } @@ -426,9 +427,11 @@ static void ipoib_mcast_join_complete(int status, if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS) mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS; + mutex_lock(&mcast_mutex); + + spin_lock_irq(&priv->lock); mcast->query = NULL; - down(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) { if (status == -ETIMEDOUT) queue_work(ipoib_workqueue, &priv->mcast_task); @@ -437,7 +440,8 @@ static void ipoib_mcast_join_complete(int status, mcast->backoff * HZ); } else complete(&mcast->done); - up(&mcast_mutex); + spin_unlock_irq(&priv->lock); + mutex_unlock(&mcast_mutex); return; } @@ -457,7 +461,7 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, rec.mgid = mcast->mcmember.mgid; rec.port_gid = priv->local_gid; - rec.pkey = be16_to_cpu(priv->pkey); + rec.pkey = cpu_to_be16(priv->pkey); comp_mask = IB_SA_MCMEMBER_REC_MGID | @@ -478,6 +482,8 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, rec.traffic_class = priv->broadcast->mcmember.traffic_class; } + init_completion(&mcast->done); + ret = ib_sa_mcmember_rec_set(priv->ca, priv->port, &rec, comp_mask, mcast->backoff * 1000, GFP_ATOMIC, ipoib_mcast_join_complete, @@ -490,12 +496,12 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast, if (mcast->backoff > IPOIB_MAX_BACKOFF_SECONDS) mcast->backoff = IPOIB_MAX_BACKOFF_SECONDS; - down(&mcast_mutex); + mutex_lock(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_delayed_work(ipoib_workqueue, &priv->mcast_task, - mcast->backoff); - up(&mcast_mutex); + mcast->backoff * HZ); + mutex_unlock(&mcast_mutex); } else mcast->query_id = ret; } @@ -525,21 +531,24 @@ void ipoib_mcast_join_task(void *dev_ptr) } if (!priv->broadcast) { - priv->broadcast = ipoib_mcast_alloc(dev, 1); - if (!priv->broadcast) { + struct ipoib_mcast *broadcast; + + broadcast = ipoib_mcast_alloc(dev, 1); + if (!broadcast) { ipoib_warn(priv, "failed to allocate broadcast group\n"); - down(&mcast_mutex); + mutex_lock(&mcast_mutex); if (test_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_delayed_work(ipoib_workqueue, &priv->mcast_task, HZ); - up(&mcast_mutex); + mutex_unlock(&mcast_mutex); return; } - memcpy(priv->broadcast->mcmember.mgid.raw, priv->dev->broadcast + 4, + spin_lock_irq(&priv->lock); + memcpy(broadcast->mcmember.mgid.raw, priv->dev->broadcast + 4, sizeof (union ib_gid)); + priv->broadcast = broadcast; - spin_lock_irq(&priv->lock); __ipoib_mcast_add(dev, priv->broadcast); spin_unlock_irq(&priv->lock); } @@ -588,44 +597,57 @@ int ipoib_mcast_start_thread(struct net_device *dev) ipoib_dbg_mcast(priv, "starting multicast thread\n"); - down(&mcast_mutex); + mutex_lock(&mcast_mutex); if (!test_and_set_bit(IPOIB_MCAST_RUN, &priv->flags)) queue_work(ipoib_workqueue, &priv->mcast_task); - up(&mcast_mutex); + mutex_unlock(&mcast_mutex); + + spin_lock_irq(&priv->lock); + set_bit(IPOIB_MCAST_STARTED, &priv->flags); + spin_unlock_irq(&priv->lock); return 0; } -int ipoib_mcast_stop_thread(struct net_device *dev) +static void wait_for_mcast_join(struct ipoib_dev_priv *priv, + struct ipoib_mcast *mcast) +{ + spin_lock_irq(&priv->lock); + if (mcast && mcast->query) { + ib_sa_cancel_query(mcast->query_id, mcast->query); + mcast->query = NULL; + spin_unlock_irq(&priv->lock); + ipoib_dbg_mcast(priv, "waiting for MGID " IPOIB_GID_FMT "\n", + IPOIB_GID_ARG(mcast->mcmember.mgid)); + wait_for_completion(&mcast->done); + } + else + spin_unlock_irq(&priv->lock); +} + +int ipoib_mcast_stop_thread(struct net_device *dev, int flush) { struct ipoib_dev_priv *priv = netdev_priv(dev); struct ipoib_mcast *mcast; ipoib_dbg_mcast(priv, "stopping multicast thread\n"); - down(&mcast_mutex); + spin_lock_irq(&priv->lock); + clear_bit(IPOIB_MCAST_STARTED, &priv->flags); + spin_unlock_irq(&priv->lock); + + mutex_lock(&mcast_mutex); clear_bit(IPOIB_MCAST_RUN, &priv->flags); cancel_delayed_work(&priv->mcast_task); - up(&mcast_mutex); + mutex_unlock(&mcast_mutex); - flush_workqueue(ipoib_workqueue); + if (flush) + flush_workqueue(ipoib_workqueue); - if (priv->broadcast && priv->broadcast->query) { - ib_sa_cancel_query(priv->broadcast->query_id, priv->broadcast->query); - priv->broadcast->query = NULL; - ipoib_dbg_mcast(priv, "waiting for bcast\n"); - wait_for_completion(&priv->broadcast->done); - } + wait_for_mcast_join(priv, priv->broadcast); - list_for_each_entry(mcast, &priv->multicast_list, list) { - if (mcast->query) { - ib_sa_cancel_query(mcast->query_id, mcast->query); - mcast->query = NULL; - ipoib_dbg_mcast(priv, "waiting for MGID " IPOIB_GID_FMT "\n", - IPOIB_GID_ARG(mcast->mcmember.mgid)); - wait_for_completion(&mcast->done); - } - } + list_for_each_entry(mcast, &priv->multicast_list, list) + wait_for_mcast_join(priv, mcast); return 0; } @@ -646,7 +668,7 @@ static int ipoib_mcast_leave(struct net_device *dev, struct ipoib_mcast *mcast) rec.mgid = mcast->mcmember.mgid; rec.port_gid = priv->local_gid; - rec.pkey = be16_to_cpu(priv->pkey); + rec.pkey = cpu_to_be16(priv->pkey); /* Remove ourselves from the multicast group */ ret = ipoib_mcast_detach(dev, be16_to_cpu(mcast->mcmember.mlid), @@ -684,6 +706,14 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid, */ spin_lock(&priv->lock); + if (!test_bit(IPOIB_MCAST_STARTED, &priv->flags) || + !priv->broadcast || + !test_bit(IPOIB_MCAST_FLAG_ATTACHED, &priv->broadcast->flags)) { + ++priv->stats.tx_dropped; + dev_kfree_skb_any(skb); + goto unlock; + } + mcast = __ipoib_mcast_find(dev, mgid); if (!mcast) { /* Let's create a new send only group now */ @@ -694,6 +724,7 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid, if (!mcast) { ipoib_warn(priv, "unable to allocate memory for " "multicast structure\n"); + ++priv->stats.tx_dropped; dev_kfree_skb_any(skb); goto out; } @@ -707,8 +738,10 @@ void ipoib_mcast_send(struct net_device *dev, union ib_gid *mgid, if (!mcast->ah) { if (skb_queue_len(&mcast->pkt_queue) < IPOIB_MAX_MCAST_QUEUE) skb_queue_tail(&mcast->pkt_queue, skb); - else + else { + ++priv->stats.tx_dropped; dev_kfree_skb_any(skb); + } if (mcast->query) ipoib_dbg_mcast(priv, "no address vector, " @@ -728,13 +761,11 @@ out: if (skb->dst && skb->dst->neighbour && !*to_ipoib_neigh(skb->dst->neighbour)) { - struct ipoib_neigh *neigh = kmalloc(sizeof *neigh, GFP_ATOMIC); + struct ipoib_neigh *neigh = ipoib_neigh_alloc(skb->dst->neighbour); if (neigh) { kref_get(&mcast->ah->ref); neigh->ah = mcast->ah; - neigh->neighbour = skb->dst->neighbour; - *to_ipoib_neigh(skb->dst->neighbour) = neigh; list_add_tail(&neigh->list, &mcast->neigh_list); } } @@ -742,6 +773,7 @@ out: ipoib_send(dev, skb, mcast->ah, IB_MULTICAST_QPN); } +unlock: spin_unlock(&priv->lock); } @@ -749,48 +781,23 @@ void ipoib_mcast_dev_flush(struct net_device *dev) { struct ipoib_dev_priv *priv = netdev_priv(dev); LIST_HEAD(remove_list); - struct ipoib_mcast *mcast, *tmcast, *nmcast; + struct ipoib_mcast *mcast, *tmcast; unsigned long flags; ipoib_dbg_mcast(priv, "flushing multicast list\n"); spin_lock_irqsave(&priv->lock, flags); - list_for_each_entry_safe(mcast, tmcast, &priv->multicast_list, list) { - nmcast = ipoib_mcast_alloc(dev, 0); - if (nmcast) { - nmcast->flags = - mcast->flags & (1 << IPOIB_MCAST_FLAG_SENDONLY); - - nmcast->mcmember.mgid = mcast->mcmember.mgid; - - /* Add the new group in before the to-be-destroyed group */ - list_add_tail(&nmcast->list, &mcast->list); - list_del_init(&mcast->list); - rb_replace_node(&mcast->rb_node, &nmcast->rb_node, - &priv->multicast_tree); - - list_add_tail(&mcast->list, &remove_list); - } else { - ipoib_warn(priv, "could not reallocate multicast group " - IPOIB_GID_FMT "\n", - IPOIB_GID_ARG(mcast->mcmember.mgid)); - } + list_for_each_entry_safe(mcast, tmcast, &priv->multicast_list, list) { + list_del(&mcast->list); + rb_erase(&mcast->rb_node, &priv->multicast_tree); + list_add_tail(&mcast->list, &remove_list); } if (priv->broadcast) { - nmcast = ipoib_mcast_alloc(dev, 0); - if (nmcast) { - nmcast->mcmember.mgid = priv->broadcast->mcmember.mgid; - - rb_replace_node(&priv->broadcast->rb_node, - &nmcast->rb_node, - &priv->multicast_tree); - - list_add_tail(&priv->broadcast->list, &remove_list); - } - - priv->broadcast = nmcast; + rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree); + list_add_tail(&priv->broadcast->list, &remove_list); + priv->broadcast = NULL; } spin_unlock_irqrestore(&priv->lock, flags); @@ -801,24 +808,6 @@ void ipoib_mcast_dev_flush(struct net_device *dev) } } -void ipoib_mcast_dev_down(struct net_device *dev) -{ - struct ipoib_dev_priv *priv = netdev_priv(dev); - unsigned long flags; - - /* Delete broadcast since it will be recreated */ - if (priv->broadcast) { - ipoib_dbg_mcast(priv, "deleting broadcast group\n"); - - spin_lock_irqsave(&priv->lock, flags); - rb_erase(&priv->broadcast->rb_node, &priv->multicast_tree); - spin_unlock_irqrestore(&priv->lock, flags); - ipoib_mcast_leave(dev, priv->broadcast); - ipoib_mcast_free(priv->broadcast); - priv->broadcast = NULL; - } -} - void ipoib_mcast_restart_task(void *dev_ptr) { struct net_device *dev = dev_ptr; @@ -830,9 +819,11 @@ void ipoib_mcast_restart_task(void *dev_ptr) ipoib_dbg_mcast(priv, "restarting multicast task\n"); - ipoib_mcast_stop_thread(dev); + ipoib_mcast_stop_thread(dev, 0); - spin_lock_irqsave(&priv->lock, flags); + local_irq_save(flags); + netif_tx_lock(dev); + spin_lock(&priv->lock); /* * Unfortunately, the networking core only gives us a list of all of @@ -904,10 +895,14 @@ void ipoib_mcast_restart_task(void *dev_ptr) list_add_tail(&mcast->list, &remove_list); } } - spin_unlock_irqrestore(&priv->lock, flags); + + spin_unlock(&priv->lock); + netif_tx_unlock(dev); + local_irq_restore(flags); /* We have to cancel outside of the spinlock */ list_for_each_entry_safe(mcast, tmcast, &remove_list, list) { + wait_for_mcast_join(priv, mcast); ipoib_mcast_leave(mcast->dev, mcast); ipoib_mcast_free(mcast); } @@ -916,6 +911,8 @@ void ipoib_mcast_restart_task(void *dev_ptr) ipoib_mcast_start_thread(dev); } +#ifdef CONFIG_INFINIBAND_IPOIB_DEBUG + struct ipoib_mcast_iter *ipoib_mcast_iter_init(struct net_device *dev) { struct ipoib_mcast_iter *iter; @@ -925,21 +922,16 @@ struct ipoib_mcast_iter *ipoib_mcast_iter_init(struct net_device *dev) return NULL; iter->dev = dev; - memset(iter->mgid.raw, 0, sizeof iter->mgid); + memset(iter->mgid.raw, 0, 16); if (ipoib_mcast_iter_next(iter)) { - ipoib_mcast_iter_free(iter); + kfree(iter); return NULL; } return iter; } -void ipoib_mcast_iter_free(struct ipoib_mcast_iter *iter) -{ - kfree(iter); -} - int ipoib_mcast_iter_next(struct ipoib_mcast_iter *iter) { struct ipoib_dev_priv *priv = netdev_priv(iter->dev); @@ -988,3 +980,5 @@ void ipoib_mcast_iter_read(struct ipoib_mcast_iter *iter, *complete = iter->complete; *send_only = iter->send_only; } + +#endif /* CONFIG_INFINIBAND_IPOIB_DEBUG */ diff --git a/drivers/md/dm-exception-store.c b/drivers/md/dm-exception-store.c index 17212b420..34a75939a 100644 --- a/drivers/md/dm-exception-store.c +++ b/drivers/md/dm-exception-store.c @@ -91,7 +91,6 @@ struct pstore { struct dm_snapshot *snap; /* up pointer to my snapshot */ int version; int valid; - uint32_t chunk_size; uint32_t exceptions_per_area; /* @@ -133,7 +132,7 @@ static int alloc_area(struct pstore *ps) int r = -ENOMEM; size_t len; - len = ps->chunk_size << SECTOR_SHIFT; + len = ps->snap->chunk_size << SECTOR_SHIFT; /* * Allocate the chunk_size block of memory that will hold @@ -160,8 +159,8 @@ static int chunk_io(struct pstore *ps, uint32_t chunk, int rw) unsigned long bits; where.bdev = ps->snap->cow->bdev; - where.sector = ps->chunk_size * chunk; - where.count = ps->chunk_size; + where.sector = ps->snap->chunk_size * chunk; + where.count = ps->snap->chunk_size; return dm_io_sync_vm(1, &where, rw, ps->area, &bits); } @@ -188,7 +187,7 @@ static int area_io(struct pstore *ps, uint32_t area, int rw) static int zero_area(struct pstore *ps, uint32_t area) { - memset(ps->area, 0, ps->chunk_size << SECTOR_SHIFT); + memset(ps->area, 0, ps->snap->chunk_size << SECTOR_SHIFT); return area_io(ps, area, WRITE); } @@ -196,6 +195,7 @@ static int read_header(struct pstore *ps, int *new_snapshot) { int r; struct disk_header *dh; + chunk_t chunk_size; r = chunk_io(ps, 0, READ); if (r) @@ -210,8 +210,29 @@ static int read_header(struct pstore *ps, int *new_snapshot) *new_snapshot = 0; ps->valid = le32_to_cpu(dh->valid); ps->version = le32_to_cpu(dh->version); - ps->chunk_size = le32_to_cpu(dh->chunk_size); + chunk_size = le32_to_cpu(dh->chunk_size); + if (ps->snap->chunk_size != chunk_size) { + DMWARN("chunk size %llu in device metadata overrides " + "table chunk size of %llu.", + (unsigned long long)chunk_size, + (unsigned long long)ps->snap->chunk_size); + + /* We had a bogus chunk_size. Fix stuff up. */ + dm_io_put(sectors_to_pages(ps->snap->chunk_size)); + free_area(ps); + + ps->snap->chunk_size = chunk_size; + ps->snap->chunk_mask = chunk_size - 1; + ps->snap->chunk_shift = ffs(chunk_size) - 1; + r = alloc_area(ps); + if (r) + return r; + + r = dm_io_get(sectors_to_pages(chunk_size)); + if (r) + return r; + } } else { DMWARN("Invalid/corrupt snapshot"); r = -ENXIO; @@ -224,13 +245,13 @@ static int write_header(struct pstore *ps) { struct disk_header *dh; - memset(ps->area, 0, ps->chunk_size << SECTOR_SHIFT); + memset(ps->area, 0, ps->snap->chunk_size << SECTOR_SHIFT); dh = (struct disk_header *) ps->area; dh->magic = cpu_to_le32(SNAP_MAGIC); dh->valid = cpu_to_le32(ps->valid); dh->version = cpu_to_le32(ps->version); - dh->chunk_size = cpu_to_le32(ps->chunk_size); + dh->chunk_size = cpu_to_le32(ps->snap->chunk_size); return chunk_io(ps, 0, WRITE); } @@ -365,7 +386,7 @@ static void persistent_destroy(struct exception_store *store) { struct pstore *ps = get_info(store); - dm_io_put(sectors_to_pages(ps->chunk_size)); + dm_io_put(sectors_to_pages(ps->snap->chunk_size)); vfree(ps->callbacks); free_area(ps); kfree(ps); @@ -383,6 +404,16 @@ static int persistent_read_metadata(struct exception_store *store) if (r) return r; + /* + * Now we know correct chunk_size, complete the initialisation. + */ + ps->exceptions_per_area = (ps->snap->chunk_size << SECTOR_SHIFT) / + sizeof(struct disk_exception); + ps->callbacks = dm_vcalloc(ps->exceptions_per_area, + sizeof(*ps->callbacks)); + if (!ps->callbacks) + return -ENOMEM; + /* * Do we need to setup a new snapshot ? */ @@ -533,9 +564,6 @@ int dm_create_persistent(struct exception_store *store, uint32_t chunk_size) ps->snap = store->snap; ps->valid = 1; ps->version = SNAPSHOT_DISK_VERSION; - ps->chunk_size = chunk_size; - ps->exceptions_per_area = (chunk_size << SECTOR_SHIFT) / - sizeof(struct disk_exception); ps->next_free = 2; /* skipping the header and first area */ ps->current_committed = 0; @@ -543,18 +571,9 @@ int dm_create_persistent(struct exception_store *store, uint32_t chunk_size) if (r) goto bad; - /* - * Allocate space for all the callbacks. - */ ps->callback_count = 0; atomic_set(&ps->pending_count, 0); - ps->callbacks = dm_vcalloc(ps->exceptions_per_area, - sizeof(*ps->callbacks)); - - if (!ps->callbacks) { - r = -ENOMEM; - goto bad; - } + ps->callbacks = NULL; store->destroy = persistent_destroy; store->read_metadata = persistent_read_metadata; @@ -568,12 +587,9 @@ int dm_create_persistent(struct exception_store *store, uint32_t chunk_size) bad: dm_io_put(sectors_to_pages(chunk_size)); - if (ps) { - if (ps->area) - free_area(ps); - - kfree(ps); - } + if (ps && ps->area) + free_area(ps); + kfree(ps); return r; } diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c index 8edd64354..f7e743691 100644 --- a/drivers/md/dm-ioctl.c +++ b/drivers/md/dm-ioctl.c @@ -102,8 +102,10 @@ static struct hash_cell *__get_name_cell(const char *str) unsigned int h = hash_str(str); list_for_each_entry (hc, _name_buckets + h, name_list) - if (!strcmp(hc->name, str)) + if (!strcmp(hc->name, str)) { + dm_get(hc->md); return hc; + } return NULL; } @@ -114,8 +116,10 @@ static struct hash_cell *__get_uuid_cell(const char *str) unsigned int h = hash_str(str); list_for_each_entry (hc, _uuid_buckets + h, uuid_list) - if (!strcmp(hc->uuid, str)) + if (!strcmp(hc->uuid, str)) { + dm_get(hc->md); return hc; + } return NULL; } @@ -191,7 +195,7 @@ static int unregister_with_devfs(struct hash_cell *hc) */ static int dm_hash_insert(const char *name, const char *uuid, struct mapped_device *md) { - struct hash_cell *cell; + struct hash_cell *cell, *hc; /* * Allocate the new cells. @@ -204,14 +208,19 @@ static int dm_hash_insert(const char *name, const char *uuid, struct mapped_devi * Insert the cell into both hash tables. */ down_write(&_hash_lock); - if (__get_name_cell(name)) + hc = __get_name_cell(name); + if (hc) { + dm_put(hc->md); goto bad; + } list_add(&cell->name_list, _name_buckets + hash_str(name)); if (uuid) { - if (__get_uuid_cell(uuid)) { + hc = __get_uuid_cell(uuid); + if (hc) { list_del(&cell->name_list); + dm_put(hc->md); goto bad; } list_add(&cell->uuid_list, _uuid_buckets + hash_str(uuid)); @@ -289,6 +298,7 @@ static int dm_hash_rename(const char *old, const char *new) if (hc) { DMWARN("asked to rename to an already existing name %s -> %s", old, new); + dm_put(hc->md); up_write(&_hash_lock); kfree(new_name); return -EBUSY; @@ -328,6 +338,7 @@ static int dm_hash_rename(const char *old, const char *new) dm_table_put(table); } + dm_put(hc->md); up_write(&_hash_lock); kfree(old_name); return 0; @@ -611,10 +622,8 @@ static struct hash_cell *__find_device_hash_cell(struct dm_ioctl *param) return __get_name_cell(param->name); md = dm_get_md(huge_decode_dev(param->dev)); - if (md) { + if (md) mdptr = dm_get_mdptr(md); - dm_put(md); - } return mdptr; } @@ -628,7 +637,6 @@ static struct mapped_device *find_device(struct dm_ioctl *param) hc = __find_device_hash_cell(param); if (hc) { md = hc->md; - dm_get(md); /* * Sneakily write in both the name and the uuid @@ -653,6 +661,7 @@ static struct mapped_device *find_device(struct dm_ioctl *param) static int dev_remove(struct dm_ioctl *param, size_t param_size) { struct hash_cell *hc; + struct mapped_device *md; down_write(&_hash_lock); hc = __find_device_hash_cell(param); @@ -663,8 +672,11 @@ static int dev_remove(struct dm_ioctl *param, size_t param_size) return -ENXIO; } + md = hc->md; + __hash_remove(hc); up_write(&_hash_lock); + dm_put(md); param->data_size = 0; return 0; } @@ -790,7 +802,6 @@ static int do_resume(struct dm_ioctl *param) } md = hc->md; - dm_get(md); new_map = hc->new_map; hc->new_map = NULL; @@ -1078,6 +1089,7 @@ static int table_clear(struct dm_ioctl *param, size_t param_size) { int r; struct hash_cell *hc; + struct mapped_device *md; down_write(&_hash_lock); @@ -1096,7 +1108,9 @@ static int table_clear(struct dm_ioctl *param, size_t param_size) param->flags &= ~DM_INACTIVE_PRESENT_FLAG; r = __dev_status(hc->md, param); + md = hc->md; up_write(&_hash_lock); + dm_put(md); return r; } diff --git a/drivers/md/dm-mpath.c b/drivers/md/dm-mpath.c index 0c1b8520e..5af5265cf 100644 --- a/drivers/md/dm-mpath.c +++ b/drivers/md/dm-mpath.c @@ -63,6 +63,7 @@ struct multipath { unsigned nr_priority_groups; struct list_head priority_groups; unsigned pg_init_required; /* pg_init needs calling? */ + unsigned pg_init_in_progress; /* Only one pg_init allowed at once */ unsigned nr_valid_paths; /* Total number of usable paths */ struct pgpath *current_pgpath; @@ -72,7 +73,7 @@ struct multipath { unsigned queue_io; /* Must we queue all I/O? */ unsigned queue_if_no_path; /* Queue I/O if last path fails? */ - unsigned suspended; /* Has dm core suspended our I/O? */ + unsigned saved_queue_if_no_path;/* Saved state during suspension */ struct work_struct process_queued_ios; struct bio_list queued_ios; @@ -178,8 +179,7 @@ static struct multipath *alloc_multipath(void) m->queue_io = 1; INIT_WORK(&m->process_queued_ios, process_queued_ios, m); INIT_WORK(&m->trigger_event, trigger_event, m); - m->mpio_pool = mempool_create(MIN_IOS, mempool_alloc_slab, - mempool_free_slab, _mpio_cache); + m->mpio_pool = mempool_create_slab_pool(MIN_IOS, _mpio_cache); if (!m->mpio_pool) { kfree(m); return NULL; @@ -304,11 +304,12 @@ static int map_io(struct multipath *m, struct bio *bio, struct mpath_io *mpio, m->queue_size--; if ((pgpath && m->queue_io) || - (!pgpath && m->queue_if_no_path && !m->suspended)) { + (!pgpath && m->queue_if_no_path)) { /* Queue for the daemon to resubmit */ bio_list_add(&m->queued_ios, bio); m->queue_size++; - if (m->pg_init_required || !m->queue_io) + if ((m->pg_init_required && !m->pg_init_in_progress) || + !m->queue_io) queue_work(kmultipathd, &m->process_queued_ios); pgpath = NULL; r = 0; @@ -327,14 +328,19 @@ static int map_io(struct multipath *m, struct bio *bio, struct mpath_io *mpio, /* * If we run out of usable paths, should we queue I/O or error it? */ -static int queue_if_no_path(struct multipath *m, unsigned queue_if_no_path) +static int queue_if_no_path(struct multipath *m, unsigned queue_if_no_path, + unsigned save_old_value) { unsigned long flags; spin_lock_irqsave(&m->lock, flags); + if (save_old_value) + m->saved_queue_if_no_path = m->queue_if_no_path; + else + m->saved_queue_if_no_path = queue_if_no_path; m->queue_if_no_path = queue_if_no_path; - if (!m->queue_if_no_path) + if (!m->queue_if_no_path && m->queue_size) queue_work(kmultipathd, &m->process_queued_ios); spin_unlock_irqrestore(&m->lock, flags); @@ -379,25 +385,31 @@ static void process_queued_ios(void *data) { struct multipath *m = (struct multipath *) data; struct hw_handler *hwh = &m->hw_handler; - struct pgpath *pgpath; - unsigned init_required, must_queue = 0; + struct pgpath *pgpath = NULL; + unsigned init_required = 0, must_queue = 1; unsigned long flags; spin_lock_irqsave(&m->lock, flags); + if (!m->queue_size) + goto out; + if (!m->current_pgpath) __choose_pgpath(m); pgpath = m->current_pgpath; - if ((pgpath && m->queue_io) || - (!pgpath && m->queue_if_no_path && !m->suspended)) - must_queue = 1; + if ((pgpath && !m->queue_io) || + (!pgpath && !m->queue_if_no_path)) + must_queue = 0; - init_required = m->pg_init_required; - if (init_required) + if (m->pg_init_required && !m->pg_init_in_progress) { m->pg_init_required = 0; + m->pg_init_in_progress = 1; + init_required = 1; + } +out: spin_unlock_irqrestore(&m->lock, flags); if (init_required) @@ -668,7 +680,7 @@ static int parse_features(struct arg_set *as, struct multipath *m, return 0; if (!strnicmp(shift(as), MESG_STR("queue_if_no_path"))) - return queue_if_no_path(m, 1); + return queue_if_no_path(m, 1, 0); else { ti->error = "Unrecognised multipath feature request"; return -EINVAL; @@ -699,6 +711,8 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, return -EINVAL; } + m->ti = ti; + r = parse_features(&as, m, ti); if (r) goto bad; @@ -740,7 +754,6 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, } ti->private = m; - m->ti = ti; return 0; @@ -752,6 +765,8 @@ static int multipath_ctr(struct dm_target *ti, unsigned int argc, static void multipath_dtr(struct dm_target *ti) { struct multipath *m = (struct multipath *) ti->private; + + flush_workqueue(kmultipathd); free_multipath(m); } @@ -765,6 +780,9 @@ static int multipath_map(struct dm_target *ti, struct bio *bio, struct mpath_io *mpio; struct multipath *m = (struct multipath *) ti->private; + if (bio_barrier(bio)) + return -EOPNOTSUPP; + mpio = mempool_alloc(m->mpio_pool, GFP_NOIO); dm_bio_record(&mpio->details, bio); @@ -837,7 +855,7 @@ static int reinstate_path(struct pgpath *pgpath) pgpath->path.is_active = 1; m->current_pgpath = NULL; - if (!m->nr_valid_paths++) + if (!m->nr_valid_paths++ && m->queue_size) queue_work(kmultipathd, &m->process_queued_ios); queue_work(kmultipathd, &m->trigger_event); @@ -963,12 +981,13 @@ void dm_pg_init_complete(struct path *path, unsigned err_flags) bypass_pg(m, pg, 1); spin_lock_irqsave(&m->lock, flags); - if (!err_flags) - m->queue_io = 0; - else { + if (err_flags) { m->current_pgpath = NULL; m->current_pg = NULL; - } + } else if (!m->pg_init_required) + m->queue_io = 0; + + m->pg_init_in_progress = 0; queue_work(kmultipathd, &m->process_queued_ios); spin_unlock_irqrestore(&m->lock, flags); } @@ -981,6 +1000,7 @@ static int do_end_io(struct multipath *m, struct bio *bio, { struct hw_handler *hwh = &m->hw_handler; unsigned err_flags = MP_FAIL_PATH; /* Default behavior */ + unsigned long flags; if (!error) return 0; /* I/O complete */ @@ -988,17 +1008,20 @@ static int do_end_io(struct multipath *m, struct bio *bio, if ((error == -EWOULDBLOCK) && bio_rw_ahead(bio)) return error; - spin_lock(&m->lock); + if (error == -EOPNOTSUPP) + return error; + + spin_lock_irqsave(&m->lock, flags); if (!m->nr_valid_paths) { - if (!m->queue_if_no_path || m->suspended) { - spin_unlock(&m->lock); + if (!m->queue_if_no_path) { + spin_unlock_irqrestore(&m->lock, flags); return -EIO; } else { - spin_unlock(&m->lock); + spin_unlock_irqrestore(&m->lock, flags); goto requeue; } } - spin_unlock(&m->lock); + spin_unlock_irqrestore(&m->lock, flags); if (hwh->type && hwh->type->error) err_flags = hwh->type->error(hwh, bio); @@ -1018,12 +1041,12 @@ static int do_end_io(struct multipath *m, struct bio *bio, dm_bio_restore(&mpio->details, bio); /* queue for the daemon to resubmit or fail */ - spin_lock(&m->lock); + spin_lock_irqsave(&m->lock, flags); bio_list_add(&m->queued_ios, bio); m->queue_size++; if (!m->queue_io) queue_work(kmultipathd, &m->process_queued_ios); - spin_unlock(&m->lock); + spin_unlock_irqrestore(&m->lock, flags); return 1; /* io not complete */ } @@ -1051,27 +1074,27 @@ static int multipath_end_io(struct dm_target *ti, struct bio *bio, /* * Suspend can't complete until all the I/O is processed so if - * the last path failed we will now error any queued I/O. + * the last path fails we must error any remaining I/O. + * Note that if the freeze_bdev fails while suspending, the + * queue_if_no_path state is lost - userspace should reset it. */ static void multipath_presuspend(struct dm_target *ti) { struct multipath *m = (struct multipath *) ti->private; - unsigned long flags; - spin_lock_irqsave(&m->lock, flags); - m->suspended = 1; - if (m->queue_if_no_path) - queue_work(kmultipathd, &m->process_queued_ios); - spin_unlock_irqrestore(&m->lock, flags); + queue_if_no_path(m, 0, 1); } +/* + * Restore the queue_if_no_path setting. + */ static void multipath_resume(struct dm_target *ti) { struct multipath *m = (struct multipath *) ti->private; unsigned long flags; spin_lock_irqsave(&m->lock, flags); - m->suspended = 0; + m->queue_if_no_path = m->saved_queue_if_no_path; spin_unlock_irqrestore(&m->lock, flags); } @@ -1204,9 +1227,9 @@ static int multipath_message(struct dm_target *ti, unsigned argc, char **argv) if (argc == 1) { if (!strnicmp(argv[0], MESG_STR("queue_if_no_path"))) - return queue_if_no_path(m, 1); + return queue_if_no_path(m, 1, 0); else if (!strnicmp(argv[0], MESG_STR("fail_if_no_path"))) - return queue_if_no_path(m, 0); + return queue_if_no_path(m, 0, 0); } if (argc != 2) diff --git a/drivers/md/dm-raid1.c b/drivers/md/dm-raid1.c index 6e3cf7e13..2cab46ed1 100644 --- a/drivers/md/dm-raid1.c +++ b/drivers/md/dm-raid1.c @@ -106,12 +106,42 @@ struct region { struct bio_list delayed_bios; }; + +/*----------------------------------------------------------------- + * Mirror set structures. + *---------------------------------------------------------------*/ +struct mirror { + atomic_t error_count; + struct dm_dev *dev; + sector_t offset; +}; + +struct mirror_set { + struct dm_target *ti; + struct list_head list; + struct region_hash rh; + struct kcopyd_client *kcopyd_client; + + spinlock_t lock; /* protects the next two lists */ + struct bio_list reads; + struct bio_list writes; + + /* recovery */ + region_t nr_regions; + int in_sync; + + struct mirror *default_mirror; /* Default mirror */ + + unsigned int nr_mirrors; + struct mirror mirror[0]; +}; + /* * Conversion fns */ static inline region_t bio_to_region(struct region_hash *rh, struct bio *bio) { - return bio->bi_sector >> rh->region_shift; + return (bio->bi_sector - rh->ms->ti->begin) >> rh->region_shift; } static inline sector_t region_to_sector(struct region_hash *rh, region_t region) @@ -122,16 +152,6 @@ static inline sector_t region_to_sector(struct region_hash *rh, region_t region) /* FIXME move this */ static void queue_bio(struct mirror_set *ms, struct bio *bio, int rw); -static void *region_alloc(unsigned int __nocast gfp_mask, void *pool_data) -{ - return kmalloc(sizeof(struct region), gfp_mask); -} - -static void region_free(void *element, void *pool_data) -{ - kfree(element); -} - #define MIN_REGIONS 64 #define MAX_RECOVERY 1 static int rh_init(struct region_hash *rh, struct mirror_set *ms, @@ -173,8 +193,8 @@ static int rh_init(struct region_hash *rh, struct mirror_set *ms, INIT_LIST_HEAD(&rh->quiesced_regions); INIT_LIST_HEAD(&rh->recovered_regions); - rh->region_pool = mempool_create(MIN_REGIONS, region_alloc, - region_free, NULL); + rh->region_pool = mempool_create_kmalloc_pool(MIN_REGIONS, + sizeof(struct region)); if (!rh->region_pool) { vfree(rh->buckets); rh->buckets = NULL; @@ -233,7 +253,9 @@ static struct region *__rh_alloc(struct region_hash *rh, region_t region) struct region *reg, *nreg; read_unlock(&rh->hash_lock); - nreg = mempool_alloc(rh->region_pool, GFP_NOIO); + nreg = mempool_alloc(rh->region_pool, GFP_ATOMIC); + if (unlikely(!nreg)) + nreg = kmalloc(sizeof(struct region), GFP_NOIO); nreg->state = rh->log->type->in_sync(rh->log, region, 1) ? RH_CLEAN : RH_NOSYNC; nreg->rh = rh; @@ -375,16 +397,20 @@ static void rh_inc(struct region_hash *rh, region_t region) read_lock(&rh->hash_lock); reg = __rh_find(rh, region); - if (reg->state == RH_CLEAN) { - rh->log->type->mark_region(rh->log, reg->key); - spin_lock_irq(&rh->region_lock); + spin_lock_irq(&rh->region_lock); + atomic_inc(®->pending); + + if (reg->state == RH_CLEAN) { reg->state = RH_DIRTY; list_del_init(®->list); /* take off the clean list */ spin_unlock_irq(&rh->region_lock); - } - atomic_inc(®->pending); + rh->log->type->mark_region(rh->log, reg->key); + } else + spin_unlock_irq(&rh->region_lock); + + read_unlock(&rh->hash_lock); } @@ -406,17 +432,29 @@ static void rh_dec(struct region_hash *rh, region_t region) reg = __rh_lookup(rh, region); read_unlock(&rh->hash_lock); + spin_lock_irqsave(&rh->region_lock, flags); if (atomic_dec_and_test(®->pending)) { - spin_lock_irqsave(&rh->region_lock, flags); + /* + * There is no pending I/O for this region. + * We can move the region to corresponding list for next action. + * At this point, the region is not yet connected to any list. + * + * If the state is RH_NOSYNC, the region should be kept off + * from clean list. + * The hash entry for RH_NOSYNC will remain in memory + * until the region is recovered or the map is reloaded. + */ + + /* do nothing for RH_NOSYNC */ if (reg->state == RH_RECOVERING) { list_add_tail(®->list, &rh->quiesced_regions); - } else { + } else if (reg->state == RH_DIRTY) { reg->state = RH_CLEAN; list_add(®->list, &rh->clean_regions); } - spin_unlock_irqrestore(&rh->region_lock, flags); should_wake = 1; } + spin_unlock_irqrestore(&rh->region_lock, flags); if (should_wake) wake(); @@ -535,33 +573,6 @@ static void rh_start_recovery(struct region_hash *rh) wake(); } -/*----------------------------------------------------------------- - * Mirror set structures. - *---------------------------------------------------------------*/ -struct mirror { - atomic_t error_count; - struct dm_dev *dev; - sector_t offset; -}; - -struct mirror_set { - struct dm_target *ti; - struct list_head list; - struct region_hash rh; - struct kcopyd_client *kcopyd_client; - - spinlock_t lock; /* protects the next two lists */ - struct bio_list reads; - struct bio_list writes; - - /* recovery */ - region_t nr_regions; - int in_sync; - - unsigned int nr_mirrors; - struct mirror mirror[0]; -}; - /* * Every mirror should look like this one. */ @@ -607,7 +618,7 @@ static int recover(struct mirror_set *ms, struct region *reg) unsigned long flags = 0; /* fill in the source */ - m = ms->mirror + DEFAULT_MIRROR; + m = ms->default_mirror; from.bdev = m->dev->bdev; from.sector = m->offset + region_to_sector(reg->rh, reg->key); if (reg->key == (ms->nr_regions - 1)) { @@ -623,7 +634,7 @@ static int recover(struct mirror_set *ms, struct region *reg) /* fill in the destinations */ for (i = 0, dest = to; i < ms->nr_mirrors; i++) { - if (i == DEFAULT_MIRROR) + if (&ms->mirror[i] == ms->default_mirror) continue; m = ms->mirror + i; @@ -678,7 +689,7 @@ static void do_recovery(struct mirror_set *ms) static struct mirror *choose_mirror(struct mirror_set *ms, sector_t sector) { /* FIXME: add read balancing */ - return ms->mirror + DEFAULT_MIRROR; + return ms->default_mirror; } /* @@ -705,7 +716,7 @@ static void do_reads(struct mirror_set *ms, struct bio_list *reads) if (rh_in_sync(&ms->rh, region, 0)) m = choose_mirror(ms, bio->bi_sector); else - m = ms->mirror + DEFAULT_MIRROR; + m = ms->default_mirror; map_bio(ms, m, bio); generic_make_request(bio); @@ -829,7 +840,7 @@ static void do_writes(struct mirror_set *ms, struct bio_list *writes) rh_delay(&ms->rh, bio); while ((bio = bio_list_pop(&nosync))) { - map_bio(ms, ms->mirror + DEFAULT_MIRROR, bio); + map_bio(ms, ms->default_mirror, bio); generic_make_request(bio); } } @@ -896,6 +907,7 @@ static struct mirror_set *alloc_context(unsigned int nr_mirrors, ms->nr_mirrors = nr_mirrors; ms->nr_regions = dm_sector_div_up(ti->len, region_size); ms->in_sync = 0; + ms->default_mirror = &ms->mirror[DEFAULT_MIRROR]; if (rh_init(&ms->rh, ms, dl, region_size, ms->nr_regions)) { ti->error = "dm-mirror: Error creating dirty region hash"; @@ -925,9 +937,9 @@ static inline int _check_region_size(struct dm_target *ti, uint32_t size) static int get_mirror(struct mirror_set *ms, struct dm_target *ti, unsigned int mirror, char **argv) { - sector_t offset; + unsigned long long offset; - if (sscanf(argv[1], SECTOR_FORMAT, &offset) != 1) { + if (sscanf(argv[1], "%llu", &offset) != 1) { ti->error = "dm-mirror: Invalid offset"; return -EINVAL; } @@ -1060,6 +1072,7 @@ static int mirror_ctr(struct dm_target *ti, unsigned int argc, char **argv) } ti->private = ms; + ti->split_io = ms->rh.region_size; r = kcopyd_client_create(DM_IO_PAGES, &ms->kcopyd_client); if (r) { @@ -1105,7 +1118,7 @@ static int mirror_map(struct dm_target *ti, struct bio *bio, struct mirror *m; struct mirror_set *ms = ti->private; - map_context->ll = bio->bi_sector >> ms->rh.region_shift; + map_context->ll = bio_to_region(&ms->rh, bio); if (rw == WRITE) { queue_bio(ms, bio, rw); @@ -1193,16 +1206,17 @@ static int mirror_status(struct dm_target *ti, status_type_t type, for (m = 0; m < ms->nr_mirrors; m++) DMEMIT("%s ", ms->mirror[m].dev->name); - DMEMIT(SECTOR_FORMAT "/" SECTOR_FORMAT, - ms->rh.log->type->get_sync_count(ms->rh.log), - ms->nr_regions); + DMEMIT("%llu/%llu", + (unsigned long long)ms->rh.log->type-> + get_sync_count(ms->rh.log), + (unsigned long long)ms->nr_regions); break; case STATUSTYPE_TABLE: DMEMIT("%d ", ms->nr_mirrors); for (m = 0; m < ms->nr_mirrors; m++) - DMEMIT("%s " SECTOR_FORMAT " ", - ms->mirror[m].dev->name, ms->mirror[m].offset); + DMEMIT("%s %llu ", ms->mirror[m].dev->name, + (unsigned long long)ms->mirror[m].offset); } return 0; @@ -1229,7 +1243,7 @@ static int __init dm_mirror_init(void) if (r) return r; - _kmirrord_wq = create_workqueue("kmirrord"); + _kmirrord_wq = create_singlethread_workqueue("kmirrord"); if (!_kmirrord_wq) { DMERR("couldn't start kmirrord"); dm_dirty_log_exit(); diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c index 7e691ab9a..b84bc1aae 100644 --- a/drivers/md/dm-snap.c +++ b/drivers/md/dm-snap.c @@ -49,11 +49,26 @@ struct pending_exception { struct bio_list snapshot_bios; /* - * Other pending_exceptions that are processing this - * chunk. When this list is empty, we know we can - * complete the origins. + * Short-term queue of pending exceptions prior to submission. */ - struct list_head siblings; + struct list_head list; + + /* + * The primary pending_exception is the one that holds + * the sibling_count and the list of origin_bios for a + * group of pending_exceptions. It is always last to get freed. + * These fields get set up when writing to the origin. + */ + struct pending_exception *primary_pe; + + /* + * Number of pending_exceptions processing this chunk. + * When this drops to zero we must complete the origin bios. + * If incrementing or decrementing this, hold pe->snap->lock for + * the sibling concerned and not pe->primary_pe->snap->lock unless + * they are the same. + */ + atomic_t sibling_count; /* Pointer back to snapshot context */ struct dm_snapshot *snap; @@ -371,6 +386,17 @@ static inline ulong round_up(ulong n, ulong size) return (n + size) & ~size; } +static void read_snapshot_metadata(struct dm_snapshot *s) +{ + if (s->store.read_metadata(&s->store)) { + down_write(&s->lock); + s->valid = 0; + up_write(&s->lock); + + dm_table_event(s->table); + } +} + /* * Construct a snapshot mapping:

*/ @@ -457,7 +483,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) s->chunk_shift = ffs(chunk_size) - 1; s->valid = 1; - s->have_metadata = 0; + s->active = 0; s->last_percent = 0; init_rwsem(&s->lock); s->table = ti->table; @@ -492,7 +518,11 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) goto bad5; } + /* Metadata must only be loaded into one table at once */ + read_snapshot_metadata(s); + /* Add snapshot to the list of snapshots for this origin */ + /* Exceptions aren't triggered till snapshot_resume() is called */ if (register_snapshot(s)) { r = -EINVAL; ti->error = "Cannot register snapshot origin"; @@ -500,7 +530,7 @@ static int snapshot_ctr(struct dm_target *ti, unsigned int argc, char **argv) } ti->private = s; - ti->split_io = chunk_size; + ti->split_io = s->chunk_size; return 0; @@ -529,8 +559,12 @@ static void snapshot_dtr(struct dm_target *ti) { struct dm_snapshot *s = (struct dm_snapshot *) ti->private; + /* Prevent further origin writes from using this snapshot. */ + /* After this returns there can be no new kcopyd jobs. */ unregister_snapshot(s); + kcopyd_client_destroy(s->kcopyd_client); + exit_exception_table(&s->pending, pending_cache); exit_exception_table(&s->complete, exception_cache); @@ -539,7 +573,7 @@ static void snapshot_dtr(struct dm_target *ti) dm_put_device(ti, s->origin); dm_put_device(ti, s->cow); - kcopyd_client_destroy(s->kcopyd_client); + kfree(s); } @@ -573,78 +607,117 @@ static void error_bios(struct bio *bio) } } +static inline void error_snapshot_bios(struct pending_exception *pe) +{ + error_bios(bio_list_get(&pe->snapshot_bios)); +} + static struct bio *__flush_bios(struct pending_exception *pe) { - struct pending_exception *sibling; + /* + * If this pe is involved in a write to the origin and + * it is the last sibling to complete then release + * the bios for the original write to the origin. + */ - if (list_empty(&pe->siblings)) - return bio_list_get(&pe->origin_bios); + if (pe->primary_pe && + atomic_dec_and_test(&pe->primary_pe->sibling_count)) + return bio_list_get(&pe->primary_pe->origin_bios); - sibling = list_entry(pe->siblings.next, - struct pending_exception, siblings); + return NULL; +} - list_del(&pe->siblings); +static void __invalidate_snapshot(struct dm_snapshot *s, + struct pending_exception *pe, int err) +{ + if (!s->valid) + return; - /* This is fine as long as kcopyd is single-threaded. If kcopyd - * becomes multi-threaded, we'll need some locking here. - */ - bio_list_merge(&sibling->origin_bios, &pe->origin_bios); + if (err == -EIO) + DMERR("Invalidating snapshot: Error reading/writing."); + else if (err == -ENOMEM) + DMERR("Invalidating snapshot: Unable to allocate exception."); - return NULL; + if (pe) + remove_exception(&pe->e); + + if (s->store.drop_snapshot) + s->store.drop_snapshot(&s->store); + + s->valid = 0; + + dm_table_event(s->table); } static void pending_complete(struct pending_exception *pe, int success) { struct exception *e; + struct pending_exception *primary_pe; struct dm_snapshot *s = pe->snap; struct bio *flush = NULL; - if (success) { - e = alloc_exception(); - if (!e) { - DMWARN("Unable to allocate exception."); - down_write(&s->lock); - s->store.drop_snapshot(&s->store); - s->valid = 0; - flush = __flush_bios(pe); - up_write(&s->lock); - - error_bios(bio_list_get(&pe->snapshot_bios)); - goto out; - } - *e = pe->e; - - /* - * Add a proper exception, and remove the - * in-flight exception from the list. - */ + if (!success) { + /* Read/write error - snapshot is unusable */ down_write(&s->lock); - insert_exception(&s->complete, e); - remove_exception(&pe->e); + __invalidate_snapshot(s, pe, -EIO); flush = __flush_bios(pe); - - /* Submit any pending write bios */ up_write(&s->lock); - flush_bios(bio_list_get(&pe->snapshot_bios)); - } else { - /* Read/write error - snapshot is unusable */ + error_snapshot_bios(pe); + goto out; + } + + e = alloc_exception(); + if (!e) { down_write(&s->lock); - if (s->valid) - DMERR("Error reading/writing snapshot"); - s->store.drop_snapshot(&s->store); - s->valid = 0; - remove_exception(&pe->e); + __invalidate_snapshot(s, pe, -ENOMEM); flush = __flush_bios(pe); up_write(&s->lock); - error_bios(bio_list_get(&pe->snapshot_bios)); + error_snapshot_bios(pe); + goto out; + } + *e = pe->e; - dm_table_event(s->table); + /* + * Add a proper exception, and remove the + * in-flight exception from the list. + */ + down_write(&s->lock); + if (!s->valid) { + flush = __flush_bios(pe); + up_write(&s->lock); + + free_exception(e); + + error_snapshot_bios(pe); + goto out; } + insert_exception(&s->complete, e); + remove_exception(&pe->e); + flush = __flush_bios(pe); + + up_write(&s->lock); + + /* Submit any pending write bios */ + flush_bios(bio_list_get(&pe->snapshot_bios)); + out: - free_pending_exception(pe); + primary_pe = pe->primary_pe; + + /* + * Free the pe if it's not linked to an origin write or if + * it's not itself a primary pe. + */ + if (!primary_pe || primary_pe != pe) + free_pending_exception(pe); + + /* + * Free the primary pe if nothing references it. + */ + if (primary_pe && !atomic_read(&primary_pe->sibling_count)) + free_pending_exception(primary_pe); if (flush) flush_bios(flush); @@ -677,7 +750,7 @@ static void copy_callback(int read_err, unsigned int write_err, void *context) /* * Dispatches the copy operation to kcopyd. */ -static inline void start_copy(struct pending_exception *pe) +static void start_copy(struct pending_exception *pe) { struct dm_snapshot *s = pe->snap; struct io_region src, dest; @@ -721,38 +794,45 @@ __find_pending_exception(struct dm_snapshot *s, struct bio *bio) if (e) { /* cast the exception to a pending exception */ pe = container_of(e, struct pending_exception, e); + goto out; + } - } else { - /* - * Create a new pending exception, we don't want - * to hold the lock while we do this. - */ - up_write(&s->lock); - pe = alloc_pending_exception(); - down_write(&s->lock); + /* + * Create a new pending exception, we don't want + * to hold the lock while we do this. + */ + up_write(&s->lock); + pe = alloc_pending_exception(); + down_write(&s->lock); - e = lookup_exception(&s->pending, chunk); - if (e) { - free_pending_exception(pe); - pe = container_of(e, struct pending_exception, e); - } else { - pe->e.old_chunk = chunk; - bio_list_init(&pe->origin_bios); - bio_list_init(&pe->snapshot_bios); - INIT_LIST_HEAD(&pe->siblings); - pe->snap = s; - pe->started = 0; - - if (s->store.prepare_exception(&s->store, &pe->e)) { - free_pending_exception(pe); - s->valid = 0; - return NULL; - } + if (!s->valid) { + free_pending_exception(pe); + return NULL; + } - insert_exception(&s->pending, &pe->e); - } + e = lookup_exception(&s->pending, chunk); + if (e) { + free_pending_exception(pe); + pe = container_of(e, struct pending_exception, e); + goto out; + } + + pe->e.old_chunk = chunk; + bio_list_init(&pe->origin_bios); + bio_list_init(&pe->snapshot_bios); + pe->primary_pe = NULL; + atomic_set(&pe->sibling_count, 1); + pe->snap = s; + pe->started = 0; + + if (s->store.prepare_exception(&s->store, &pe->e)) { + free_pending_exception(pe); + return NULL; } + insert_exception(&s->pending, &pe->e); + + out: return pe; } @@ -769,15 +849,20 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio, { struct exception *e; struct dm_snapshot *s = (struct dm_snapshot *) ti->private; + int copy_needed = 0; int r = 1; chunk_t chunk; - struct pending_exception *pe; + struct pending_exception *pe = NULL; chunk = sector_to_chunk(s, bio->bi_sector); /* Full snapshots are not usable */ + /* To get here the table must be live so s->active is always set. */ if (!s->valid) - return -1; + return -EIO; + + if (unlikely(bio_barrier(bio))) + return -EOPNOTSUPP; /* * Write to snapshot - higher level takes care of RW/RO @@ -790,36 +875,41 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio, * to copy an exception */ down_write(&s->lock); + if (!s->valid) { + r = -EIO; + goto out_unlock; + } + /* If the block is already remapped - use that, else remap it */ e = lookup_exception(&s->complete, chunk); if (e) { remap_exception(s, e, bio); - up_write(&s->lock); - - } else { - pe = __find_pending_exception(s, bio); - - if (!pe) { - if (s->store.drop_snapshot) - s->store.drop_snapshot(&s->store); - s->valid = 0; - r = -EIO; - up_write(&s->lock); - } else { - remap_exception(s, &pe->e, bio); - bio_list_add(&pe->snapshot_bios, bio); - - if (!pe->started) { - /* this is protected by snap->lock */ - pe->started = 1; - up_write(&s->lock); - start_copy(pe); - } else - up_write(&s->lock); - r = 0; - } + goto out_unlock; + } + + pe = __find_pending_exception(s, bio); + if (!pe) { + __invalidate_snapshot(s, pe, -ENOMEM); + r = -EIO; + goto out_unlock; + } + + remap_exception(s, &pe->e, bio); + bio_list_add(&pe->snapshot_bios, bio); + + if (!pe->started) { + /* this is protected by snap->lock */ + pe->started = 1; + copy_needed = 1; } + r = 0; + + out_unlock: + up_write(&s->lock); + + if (copy_needed) + start_copy(pe); } else { /* * FIXME: this read path scares me because we @@ -831,6 +921,11 @@ static int snapshot_map(struct dm_target *ti, struct bio *bio, /* Do reads */ down_read(&s->lock); + if (!s->valid) { + up_read(&s->lock); + return -EIO; + } + /* See if it it has been remapped */ e = lookup_exception(&s->complete, chunk); if (e) @@ -848,16 +943,9 @@ static void snapshot_resume(struct dm_target *ti) { struct dm_snapshot *s = (struct dm_snapshot *) ti->private; - if (s->have_metadata) - return; - - if (s->store.read_metadata(&s->store)) { - down_write(&s->lock); - s->valid = 0; - up_write(&s->lock); - } - - s->have_metadata = 1; + down_write(&s->lock); + s->active = 1; + up_write(&s->lock); } static int snapshot_status(struct dm_target *ti, status_type_t type, @@ -875,9 +963,9 @@ static int snapshot_status(struct dm_target *ti, status_type_t type, snap->store.fraction_full(&snap->store, &numerator, &denominator); - snprintf(result, maxlen, - SECTOR_FORMAT "/" SECTOR_FORMAT, - numerator, denominator); + snprintf(result, maxlen, "%llu/%llu", + (unsigned long long)numerator, + (unsigned long long)denominator); } else snprintf(result, maxlen, "Unknown"); @@ -890,9 +978,10 @@ static int snapshot_status(struct dm_target *ti, status_type_t type, * to make private copies if the output is to * make sense. */ - snprintf(result, maxlen, "%s %s %c " SECTOR_FORMAT, + snprintf(result, maxlen, "%s %s %c %llu", snap->origin->name, snap->cow->name, - snap->type, snap->chunk_size); + snap->type, + (unsigned long long)snap->chunk_size); break; } @@ -902,37 +991,28 @@ static int snapshot_status(struct dm_target *ti, status_type_t type, /*----------------------------------------------------------------- * Origin methods *---------------------------------------------------------------*/ -static void list_merge(struct list_head *l1, struct list_head *l2) -{ - struct list_head *l1_n, *l2_p; - - l1_n = l1->next; - l2_p = l2->prev; - - l1->next = l2; - l2->prev = l1; - - l2_p->next = l1_n; - l1_n->prev = l2_p; -} - static int __origin_write(struct list_head *snapshots, struct bio *bio) { - int r = 1, first = 1; + int r = 1, first = 0; struct dm_snapshot *snap; struct exception *e; - struct pending_exception *pe, *last = NULL; + struct pending_exception *pe, *next_pe, *primary_pe = NULL; chunk_t chunk; + LIST_HEAD(pe_queue); /* Do all the snapshots on this origin */ list_for_each_entry (snap, snapshots, list) { - /* Only deal with valid snapshots */ - if (!snap->valid) - continue; - down_write(&snap->lock); + /* Only deal with valid and active snapshots */ + if (!snap->valid || !snap->active) + goto next_snapshot; + + /* Nothing to do if writing beyond end of snapshot */ + if (bio->bi_sector >= dm_table_get_size(snap->table)) + goto next_snapshot; + /* * Remember, different snapshots can have * different chunk sizes. @@ -943,49 +1023,75 @@ static int __origin_write(struct list_head *snapshots, struct bio *bio) * Check exception table to see if block * is already remapped in this snapshot * and trigger an exception if not. + * + * sibling_count is initialised to 1 so pending_complete() + * won't destroy the primary_pe while we're inside this loop. */ e = lookup_exception(&snap->complete, chunk); - if (!e) { - pe = __find_pending_exception(snap, bio); - if (!pe) { - snap->store.drop_snapshot(&snap->store); - snap->valid = 0; - - } else { - if (last) - list_merge(&pe->siblings, - &last->siblings); - - last = pe; - r = 0; + if (e) + goto next_snapshot; + + pe = __find_pending_exception(snap, bio); + if (!pe) { + __invalidate_snapshot(snap, pe, ENOMEM); + goto next_snapshot; + } + + if (!primary_pe) { + /* + * Either every pe here has same + * primary_pe or none has one yet. + */ + if (pe->primary_pe) + primary_pe = pe->primary_pe; + else { + primary_pe = pe; + first = 1; } + + bio_list_add(&primary_pe->origin_bios, bio); + + r = 0; } + if (!pe->primary_pe) { + atomic_inc(&primary_pe->sibling_count); + pe->primary_pe = primary_pe; + } + + if (!pe->started) { + pe->started = 1; + list_add_tail(&pe->list, &pe_queue); + } + + next_snapshot: up_write(&snap->lock); } + if (!primary_pe) + goto out; + /* - * Now that we have a complete pe list we can start the copying. + * If this is the first time we're processing this chunk and + * sibling_count is now 1 it means all the pending exceptions + * got completed while we were in the loop above, so it falls to + * us here to remove the primary_pe and submit any origin_bios. */ - if (last) { - pe = last; - do { - down_write(&pe->snap->lock); - if (first) - bio_list_add(&pe->origin_bios, bio); - if (!pe->started) { - pe->started = 1; - up_write(&pe->snap->lock); - start_copy(pe); - } else - up_write(&pe->snap->lock); - first = 0; - pe = list_entry(pe->siblings.next, - struct pending_exception, siblings); - - } while (pe != last); + + if (first && atomic_dec_and_test(&primary_pe->sibling_count)) { + flush_bios(bio_list_get(&primary_pe->origin_bios)); + free_pending_exception(primary_pe); + /* If we got here, pe_queue is necessarily empty. */ + goto out; } + /* + * Now that we have a complete pe list we can start the copying. + */ + list_for_each_entry_safe(pe, next_pe, &pe_queue, list) + start_copy(pe); + + out: return r; } @@ -1048,6 +1154,9 @@ static int origin_map(struct dm_target *ti, struct bio *bio, struct dm_dev *dev = (struct dm_dev *) ti->private; bio->bi_bdev = dev->bdev; + if (unlikely(bio_barrier(bio))) + return -EOPNOTSUPP; + /* Only tell snapshots if this is a write */ return (bio_rw(bio) == WRITE) ? do_origin(dev, bio) : 1; } @@ -1095,7 +1204,7 @@ static int origin_status(struct dm_target *ti, status_type_t type, char *result, static struct target_type origin_target = { .name = "snapshot-origin", - .version = {1, 0, 1}, + .version = {1, 4, 0}, .module = THIS_MODULE, .ctr = origin_ctr, .dtr = origin_dtr, @@ -1106,7 +1215,7 @@ static struct target_type origin_target = { static struct target_type snapshot_target = { .name = "snapshot", - .version = {1, 0, 1}, + .version = {1, 4, 0}, .module = THIS_MODULE, .ctr = snapshot_ctr, .dtr = snapshot_dtr, @@ -1158,8 +1267,7 @@ static int __init dm_snapshot_init(void) goto bad4; } - pending_pool = mempool_create(128, mempool_alloc_slab, - mempool_free_slab, pending_cache); + pending_pool = mempool_create_slab_pool(128, pending_cache); if (!pending_pool) { DMERR("Couldn't create pending pool."); r = -ENOMEM; diff --git a/drivers/md/dm.c b/drivers/md/dm.c index 4d710b7a1..dfd037858 100644 --- a/drivers/md/dm.c +++ b/drivers/md/dm.c @@ -26,6 +26,7 @@ static const char *_name = DM_NAME; static unsigned int major = 0; static unsigned int _major = 0; +static DEFINE_SPINLOCK(_minor_lock); /* * One of these is allocated per bio. */ @@ -54,12 +55,15 @@ union map_info *dm_get_mapinfo(struct bio *bio) return NULL; } +#define MINOR_ALLOCED ((void *)-1) + /* * Bits for the md->flags field. */ #define DMF_BLOCK_IO 0 #define DMF_SUSPENDED 1 #define DMF_FROZEN 2 +#define DMF_FREEING 3 struct mapped_device { struct rw_semaphore io_lock; @@ -218,9 +222,23 @@ static int dm_blk_open(struct inode *inode, struct file *file) { struct mapped_device *md; + spin_lock(&_minor_lock); + md = inode->i_bdev->bd_disk->private_data; + if (!md) + goto out; + + if (test_bit(DMF_FREEING, &md->flags)) { + md = NULL; + goto out; + } + dm_get(md); - return 0; + +out: + spin_unlock(&_minor_lock); + + return md ? 0 : -ENXIO; } static int dm_blk_close(struct inode *inode, struct file *file) @@ -744,14 +762,13 @@ static int dm_any_congested(void *congested_data, int bdi_bits) /*----------------------------------------------------------------- * An IDR is used to keep track of allocated minor numbers. *---------------------------------------------------------------*/ -static DEFINE_MUTEX(_minor_lock); static DEFINE_IDR(_minor_idr); static void free_minor(unsigned int minor) { - mutex_lock(&_minor_lock); + spin_lock(&_minor_lock); idr_remove(&_minor_idr, minor); - mutex_unlock(&_minor_lock); + spin_unlock(&_minor_lock); } /* @@ -764,23 +781,20 @@ static int specific_minor(struct mapped_device *md, unsigned int minor) if (minor >= (1 << MINORBITS)) return -EINVAL; - mutex_lock(&_minor_lock); + r = idr_pre_get(&_minor_idr, GFP_KERNEL); + if (!r) + return -ENOMEM; + + spin_lock(&_minor_lock); if (idr_find(&_minor_idr, minor)) { r = -EBUSY; goto out; } - r = idr_pre_get(&_minor_idr, GFP_KERNEL); - if (!r) { - r = -ENOMEM; - goto out; - } - - r = idr_get_new_above(&_minor_idr, md, minor, &m); - if (r) { + r = idr_get_new_above(&_minor_idr, MINOR_ALLOCED, minor, &m); + if (r) goto out; - } if (m != minor) { idr_remove(&_minor_idr, m); @@ -789,7 +803,7 @@ static int specific_minor(struct mapped_device *md, unsigned int minor) } out: - mutex_unlock(&_minor_lock); + spin_unlock(&_minor_lock); return r; } @@ -798,15 +812,13 @@ static int next_free_minor(struct mapped_device *md, unsigned int *minor) int r; unsigned int m; - mutex_lock(&_minor_lock); - r = idr_pre_get(&_minor_idr, GFP_KERNEL); - if (!r) { - r = -ENOMEM; - goto out; - } + if (!r) + return -ENOMEM; - r = idr_get_new(&_minor_idr, md, &m); + spin_lock(&_minor_lock); + + r = idr_get_new(&_minor_idr, MINOR_ALLOCED, &m); if (r) { goto out; } @@ -820,7 +832,7 @@ static int next_free_minor(struct mapped_device *md, unsigned int *minor) *minor = m; out: - mutex_unlock(&_minor_lock); + spin_unlock(&_minor_lock); return r; } @@ -833,12 +845,16 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent) { int r; struct mapped_device *md = kmalloc(sizeof(*md), GFP_KERNEL); + void *old_md; if (!md) { DMWARN("unable to allocate device, out of memory."); return NULL; } + if (!try_module_get(THIS_MODULE)) + goto bad0; + /* get a minor number for the dev */ r = persistent ? specific_minor(md, minor) : next_free_minor(md, &minor); if (r < 0) @@ -875,6 +891,10 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent) if (!md->disk) goto bad4; + atomic_set(&md->pending, 0); + init_waitqueue_head(&md->wait); + init_waitqueue_head(&md->eventq); + md->disk->major = _major; md->disk->first_minor = minor; md->disk->fops = &dm_blk_dops; @@ -884,9 +904,12 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent) add_disk(md->disk); format_dev_t(md->name, MKDEV(_major, minor)); - atomic_set(&md->pending, 0); - init_waitqueue_head(&md->wait); - init_waitqueue_head(&md->eventq); + /* Populate the mapping, nobody knows we exist yet */ + spin_lock(&_minor_lock); + old_md = idr_replace(&_minor_idr, md, minor); + spin_unlock(&_minor_lock); + + BUG_ON(old_md != MINOR_ALLOCED); return md; @@ -898,6 +921,8 @@ static struct mapped_device *alloc_dev(unsigned int minor, int persistent) blk_cleanup_queue(md->queue); free_minor(minor); bad1: + module_put(THIS_MODULE); + bad0: kfree(md); return NULL; } @@ -914,8 +939,14 @@ static void free_dev(struct mapped_device *md) mempool_destroy(md->io_pool); del_gendisk(md->disk); free_minor(minor); + + spin_lock(&_minor_lock); + md->disk->private_data = NULL; + spin_unlock(&_minor_lock); + put_disk(md->disk); blk_cleanup_queue(md->queue); + module_put(THIS_MODULE); kfree(md); } @@ -1015,13 +1046,18 @@ static struct mapped_device *dm_find_md(dev_t dev) if (MAJOR(dev) != _major || minor >= (1 << MINORBITS)) return NULL; - mutex_lock(&_minor_lock); + spin_lock(&_minor_lock); md = idr_find(&_minor_idr, minor); - if (!md || (dm_disk(md)->first_minor != minor)) + if (md && (md == MINOR_ALLOCED || + (dm_disk(md)->first_minor != minor) || + test_bit(DMF_FREEING, &md->flags))) { md = NULL; + goto out; + } - mutex_unlock(&_minor_lock); +out: + spin_unlock(&_minor_lock); return md; } @@ -1055,8 +1091,13 @@ void dm_put(struct mapped_device *md) { struct dm_table *map; - if (atomic_dec_and_test(&md->holders)) { + BUG_ON(test_bit(DMF_FREEING, &md->flags)); + + if (atomic_dec_and_lock(&md->holders, &_minor_lock)) { map = dm_get_table(md); + idr_replace(&_minor_idr, MINOR_ALLOCED, dm_disk(md)->first_minor); + set_bit(DMF_FREEING, &md->flags); + spin_unlock(&_minor_lock); if (!dm_suspended(md)) { dm_table_presuspend_targets(map); dm_table_postsuspend_targets(map); diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c index 4070eff6f..5a5449400 100644 --- a/drivers/md/raid1.c +++ b/drivers/md/raid1.c @@ -1486,7 +1486,6 @@ static void raid1d(mddev_t *mddev) d = conf->raid_disks; d--; rdev = conf->mirrors[d].rdev; - atomic_add(s, &rdev->corrected_errors); if (rdev && test_bit(In_sync, &rdev->flags)) { if (sync_page_io(rdev->bdev, @@ -1509,6 +1508,9 @@ static void raid1d(mddev_t *mddev) s<<9, conf->tmppage, READ) == 0) /* Well, this device is dead */ md_error(mddev, rdev); + else + atomic_add(s, &rdev->corrected_errors); + } } } else { diff --git a/drivers/media/dvb/bt8xx/dvb-bt8xx.c b/drivers/media/dvb/bt8xx/dvb-bt8xx.c index ccc7b2eb4..0bcaa358c 100644 --- a/drivers/media/dvb/bt8xx/dvb-bt8xx.c +++ b/drivers/media/dvb/bt8xx/dvb-bt8xx.c @@ -184,6 +184,11 @@ static struct mt352_config thomson_dtt7579_config = { .pll_set = thomson_dtt7579_pll_set, }; +static struct zl10353_config thomson_dtt7579_zl10353_config = { + .demod_address = 0x0f, + .pll_set = thomson_dtt7579_pll_set, +}; + static int cx24108_pll_set(struct dvb_frontend* fe, struct dvb_frontend_parameters* params) { u32 freq = params->frequency; @@ -617,6 +622,11 @@ static void frontend_init(struct dvb_bt8xx_card *card, u32 type) switch(type) { case BTTV_BOARD_DVICO_DVBT_LITE: card->fe = mt352_attach(&thomson_dtt7579_config, card->i2c_adapter); + + if (card->fe == NULL) + card->fe = zl10353_attach(&thomson_dtt7579_zl10353_config, + card->i2c_adapter); + if (card->fe != NULL) { card->fe->ops->info.frequency_min = 174000000; card->fe->ops->info.frequency_max = 862000000; diff --git a/drivers/media/dvb/bt8xx/dvb-bt8xx.h b/drivers/media/dvb/bt8xx/dvb-bt8xx.h index 00dd9fa54..e41066ae7 100644 --- a/drivers/media/dvb/bt8xx/dvb-bt8xx.h +++ b/drivers/media/dvb/bt8xx/dvb-bt8xx.h @@ -37,6 +37,7 @@ #include "cx24110.h" #include "or51211.h" #include "lgdt330x.h" +#include "zl10353.h" struct dvb_bt8xx_card { struct mutex lock; diff --git a/drivers/media/dvb/dvb-core/dvb_frontend.c b/drivers/media/dvb/dvb-core/dvb_frontend.c index a05179016..cb69372fe 100644 --- a/drivers/media/dvb/dvb-core/dvb_frontend.c +++ b/drivers/media/dvb/dvb-core/dvb_frontend.c @@ -519,7 +519,9 @@ static int dvb_frontend_thread(void *data) fepriv->delay = 3*HZ; fepriv->status = 0; fepriv->wakeup = 0; - fepriv->reinitialise = 1; + fepriv->reinitialise = 0; + + dvb_frontend_init(fe); while (1) { up(&fepriv->sem); /* is locked when we enter the thread... */ @@ -996,17 +998,17 @@ static int dvb_frontend_open(struct inode *inode, struct file *file) return ret; if ((file->f_flags & O_ACCMODE) != O_RDONLY) { + /* normal tune mode when opened R/W */ + fepriv->tune_mode_flags &= ~FE_TUNE_MODE_ONESHOT; + fepriv->tone = -1; + fepriv->voltage = -1; + ret = dvb_frontend_start (fe); if (ret) dvb_generic_release (inode, file); /* empty event queue */ fepriv->events.eventr = fepriv->events.eventw = 0; - - /* normal tune mode when opened R/W */ - fepriv->tune_mode_flags &= ~FE_TUNE_MODE_ONESHOT; - fepriv->tone = -1; - fepriv->voltage = -1; } return ret; diff --git a/drivers/media/dvb/dvb-core/dvb_net.c b/drivers/media/dvb/dvb-core/dvb_net.c index 2f0f35811..9fd87521a 100644 --- a/drivers/media/dvb/dvb-core/dvb_net.c +++ b/drivers/media/dvb/dvb-core/dvb_net.c @@ -1052,7 +1052,7 @@ static void wq_set_multicast_list (void *data) dvb_net_feed_stop(dev); priv->rx_mode = RX_MODE_UNI; - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); if (dev->flags & IFF_PROMISC) { dprintk("%s: promiscuous mode\n", dev->name); @@ -1077,7 +1077,7 @@ static void wq_set_multicast_list (void *data) } } - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); dvb_net_feed_start(dev); } diff --git a/drivers/media/dvb/frontends/dvb-pll.c b/drivers/media/dvb/frontends/dvb-pll.c index 2a3c2ce7b..40930a358 100644 --- a/drivers/media/dvb/frontends/dvb-pll.c +++ b/drivers/media/dvb/frontends/dvb-pll.c @@ -1,6 +1,4 @@ /* - * $Id: dvb-pll.c,v 1.7 2005/02/10 11:52:02 kraxel Exp $ - * * descriptions + helper functions for simple dvb plls. * * (c) 2004 Gerd Knorr [SuSE Labs] @@ -57,7 +55,7 @@ struct dvb_pll_desc dvb_pll_thomson_dtt7610 = { }; EXPORT_SYMBOL(dvb_pll_thomson_dtt7610); -static void thomson_dtt759x_bw(u8 *buf, int bandwidth) +static void thomson_dtt759x_bw(u8 *buf, u32 freq, int bandwidth) { if (BANDWIDTH_7_MHZ == bandwidth) buf[3] |= 0x10; @@ -84,17 +82,45 @@ struct dvb_pll_desc dvb_pll_lg_z201 = { .name = "LG z201", .min = 174000000, .max = 862000000, - .count = 5, + .count = 6, .entries = { { 0, 36166667, 166666, 0xbc, 0x03 }, - { 443250000, 36166667, 166666, 0xbc, 0x01 }, - { 542000000, 36166667, 166666, 0xbc, 0x02 }, - { 830000000, 36166667, 166666, 0xf4, 0x02 }, - { 999999999, 36166667, 166666, 0xfc, 0x02 }, + { 157500000, 36166667, 166666, 0xbc, 0x01 }, + { 443250000, 36166667, 166666, 0xbc, 0x02 }, + { 542000000, 36166667, 166666, 0xbc, 0x04 }, + { 830000000, 36166667, 166666, 0xf4, 0x04 }, + { 999999999, 36166667, 166666, 0xfc, 0x04 }, }, }; EXPORT_SYMBOL(dvb_pll_lg_z201); +struct dvb_pll_desc dvb_pll_microtune_4042 = { + .name = "Microtune 4042 FI5", + .min = 57000000, + .max = 858000000, + .count = 3, + .entries = { + { 162000000, 44000000, 62500, 0x8e, 0xa1 }, + { 457000000, 44000000, 62500, 0x8e, 0x91 }, + { 999999999, 44000000, 62500, 0x8e, 0x31 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_microtune_4042); + +struct dvb_pll_desc dvb_pll_thomson_dtt761x = { + /* DTT 7611 7611A 7612 7613 7613A 7614 7615 7615A */ + .name = "Thomson dtt761x", + .min = 57000000, + .max = 863000000, + .count = 3, + .entries = { + { 147000000, 44000000, 62500, 0x8e, 0x39 }, + { 417000000, 44000000, 62500, 0x8e, 0x3a }, + { 999999999, 44000000, 62500, 0x8e, 0x3c }, + }, +}; +EXPORT_SYMBOL(dvb_pll_thomson_dtt761x); + struct dvb_pll_desc dvb_pll_unknown_1 = { .name = "unknown 1", /* used by dntv live dvb-t */ .min = 174000000, @@ -114,6 +140,285 @@ struct dvb_pll_desc dvb_pll_unknown_1 = { }; EXPORT_SYMBOL(dvb_pll_unknown_1); +/* Infineon TUA6010XS + * used in Thomson Cable Tuner + */ +struct dvb_pll_desc dvb_pll_tua6010xs = { + .name = "Infineon TUA6010XS", + .min = 44250000, + .max = 858000000, + .count = 3, + .entries = { + { 115750000, 36125000, 62500, 0x8e, 0x03 }, + { 403250000, 36125000, 62500, 0x8e, 0x06 }, + { 999999999, 36125000, 62500, 0x8e, 0x85 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_tua6010xs); + +/* Panasonic env57h1xd5 (some Philips PLL ?) */ +struct dvb_pll_desc dvb_pll_env57h1xd5 = { + .name = "Panasonic ENV57H1XD5", + .min = 44250000, + .max = 858000000, + .count = 4, + .entries = { + { 153000000, 36291666, 166666, 0xc2, 0x41 }, + { 470000000, 36291666, 166666, 0xc2, 0x42 }, + { 526000000, 36291666, 166666, 0xc2, 0x84 }, + { 999999999, 36291666, 166666, 0xc2, 0xa4 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_env57h1xd5); + +/* Philips TDA6650/TDA6651 + * used in Panasonic ENV77H11D5 + */ +static void tda665x_bw(u8 *buf, u32 freq, int bandwidth) +{ + if (bandwidth == BANDWIDTH_8_MHZ) + buf[3] |= 0x08; +} + +struct dvb_pll_desc dvb_pll_tda665x = { + .name = "Philips TDA6650/TDA6651", + .min = 44250000, + .max = 858000000, + .setbw = tda665x_bw, + .count = 12, + .entries = { + { 93834000, 36249333, 166667, 0xca, 0x61 /* 011 0 0 0 01 */ }, + { 123834000, 36249333, 166667, 0xca, 0xa1 /* 101 0 0 0 01 */ }, + { 161000000, 36249333, 166667, 0xca, 0xa1 /* 101 0 0 0 01 */ }, + { 163834000, 36249333, 166667, 0xca, 0xc2 /* 110 0 0 0 10 */ }, + { 253834000, 36249333, 166667, 0xca, 0x62 /* 011 0 0 0 10 */ }, + { 383834000, 36249333, 166667, 0xca, 0xa2 /* 101 0 0 0 10 */ }, + { 443834000, 36249333, 166667, 0xca, 0xc2 /* 110 0 0 0 10 */ }, + { 444000000, 36249333, 166667, 0xca, 0xc4 /* 110 0 0 1 00 */ }, + { 583834000, 36249333, 166667, 0xca, 0x64 /* 011 0 0 1 00 */ }, + { 793834000, 36249333, 166667, 0xca, 0xa4 /* 101 0 0 1 00 */ }, + { 444834000, 36249333, 166667, 0xca, 0xc4 /* 110 0 0 1 00 */ }, + { 861000000, 36249333, 166667, 0xca, 0xe4 /* 111 0 0 1 00 */ }, + } +}; +EXPORT_SYMBOL(dvb_pll_tda665x); + +/* Infineon TUA6034 + * used in LG TDTP E102P + */ +static void tua6034_bw(u8 *buf, u32 freq, int bandwidth) +{ + if (BANDWIDTH_7_MHZ != bandwidth) + buf[3] |= 0x08; +} + +struct dvb_pll_desc dvb_pll_tua6034 = { + .name = "Infineon TUA6034", + .min = 44250000, + .max = 858000000, + .count = 3, + .setbw = tua6034_bw, + .entries = { + { 174500000, 36166667, 62500, 0xce, 0x01 }, + { 230000000, 36166667, 62500, 0xce, 0x02 }, + { 999999999, 36166667, 62500, 0xce, 0x04 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_tua6034); + +/* Infineon TUA6034 + * used in LG TDVS H061F and LG TDVS H062F + */ +struct dvb_pll_desc dvb_pll_tdvs_tua6034 = { + .name = "LG/Infineon TUA6034", + .min = 54000000, + .max = 863000000, + .count = 3, + .entries = { + { 165000000, 44000000, 62500, 0xce, 0x01 }, + { 450000000, 44000000, 62500, 0xce, 0x02 }, + { 999999999, 44000000, 62500, 0xce, 0x04 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_tdvs_tua6034); + +/* Philips FMD1216ME + * used in Medion Hybrid PCMCIA card and USB Box + */ +static void fmd1216me_bw(u8 *buf, u32 freq, int bandwidth) +{ + if (bandwidth == BANDWIDTH_8_MHZ && freq >= 158870000) + buf[3] |= 0x08; +} + +struct dvb_pll_desc dvb_pll_fmd1216me = { + .name = "Philips FMD1216ME", + .min = 50870000, + .max = 858000000, + .setbw = fmd1216me_bw, + .count = 7, + .entries = { + { 143870000, 36213333, 166667, 0xbc, 0x41 }, + { 158870000, 36213333, 166667, 0xf4, 0x41 }, + { 329870000, 36213333, 166667, 0xbc, 0x42 }, + { 441870000, 36213333, 166667, 0xf4, 0x42 }, + { 625870000, 36213333, 166667, 0xbc, 0x44 }, + { 803870000, 36213333, 166667, 0xf4, 0x44 }, + { 999999999, 36213333, 166667, 0xfc, 0x44 }, + } +}; +EXPORT_SYMBOL(dvb_pll_fmd1216me); + +/* ALPS TDED4 + * used in Nebula-Cards and USB boxes + */ +static void tded4_bw(u8 *buf, u32 freq, int bandwidth) +{ + if (bandwidth == BANDWIDTH_8_MHZ) + buf[3] |= 0x04; +} + +struct dvb_pll_desc dvb_pll_tded4 = { + .name = "ALPS TDED4", + .min = 47000000, + .max = 863000000, + .setbw = tded4_bw, + .count = 4, + .entries = { + { 153000000, 36166667, 166667, 0x85, 0x01 }, + { 470000000, 36166667, 166667, 0x85, 0x02 }, + { 823000000, 36166667, 166667, 0x85, 0x08 }, + { 999999999, 36166667, 166667, 0x85, 0x88 }, + } +}; +EXPORT_SYMBOL(dvb_pll_tded4); + +/* ALPS TDHU2 + * used in AverTVHD MCE A180 + */ +struct dvb_pll_desc dvb_pll_tdhu2 = { + .name = "ALPS TDHU2", + .min = 54000000, + .max = 864000000, + .count = 4, + .entries = { + { 162000000, 44000000, 62500, 0x85, 0x01 }, + { 426000000, 44000000, 62500, 0x85, 0x02 }, + { 782000000, 44000000, 62500, 0x85, 0x08 }, + { 999999999, 44000000, 62500, 0x85, 0x88 }, + } +}; +EXPORT_SYMBOL(dvb_pll_tdhu2); + +/* Philips TUV1236D + * used in ATI HDTV Wonder + */ +struct dvb_pll_desc dvb_pll_tuv1236d = { + .name = "Philips TUV1236D", + .min = 54000000, + .max = 864000000, + .count = 3, + .entries = { + { 157250000, 44000000, 62500, 0xc6, 0x41 }, + { 454000000, 44000000, 62500, 0xc6, 0x42 }, + { 999999999, 44000000, 62500, 0xc6, 0x44 }, + }, +}; +EXPORT_SYMBOL(dvb_pll_tuv1236d); + +/* Samsung TBMV30111IN / TBMV30712IN1 + * used in Air2PC ATSC - 2nd generation (nxt2002) + */ +struct dvb_pll_desc dvb_pll_samsung_tbmv = { + .name = "Samsung TBMV30111IN / TBMV30712IN1", + .min = 54000000, + .max = 860000000, + .count = 6, + .entries = { + { 172000000, 44000000, 166666, 0xb4, 0x01 }, + { 214000000, 44000000, 166666, 0xb4, 0x02 }, + { 467000000, 44000000, 166666, 0xbc, 0x02 }, + { 721000000, 44000000, 166666, 0xbc, 0x08 }, + { 841000000, 44000000, 166666, 0xf4, 0x08 }, + { 999999999, 44000000, 166666, 0xfc, 0x02 }, + } +}; +EXPORT_SYMBOL(dvb_pll_samsung_tbmv); + +/* + * Philips SD1878 Tuner. + */ +struct dvb_pll_desc dvb_pll_philips_sd1878_tda8261 = { + .name = "Philips SD1878", + .min = 950000, + .max = 2150000, + .count = 4, + .entries = { + { 1250000, 499, 500, 0xc4, 0x00}, + { 1550000, 499, 500, 0xc4, 0x40}, + { 2050000, 499, 500, 0xc4, 0x80}, + { 2150000, 499, 500, 0xc4, 0xc0}, + }, +}; +EXPORT_SYMBOL(dvb_pll_philips_sd1878_tda8261); + +/* + * Philips TD1316 Tuner. + */ +static void td1316_bw(u8 *buf, u32 freq, int bandwidth) +{ + u8 band; + + /* determine band */ + if (freq < 161000000) + band = 1; + else if (freq < 444000000) + band = 2; + else + band = 4; + + buf[3] |= band; + + /* setup PLL filter */ + if (bandwidth == BANDWIDTH_8_MHZ) + buf[3] |= 1 << 3; +} + +struct dvb_pll_desc dvb_pll_philips_td1316 = { + .name = "Philips TD1316", + .min = 87000000, + .max = 895000000, + .setbw = td1316_bw, + .count = 9, + .entries = { + { 93834000, 36166000, 166666, 0xca, 0x60}, + { 123834000, 36166000, 166666, 0xca, 0xa0}, + { 163834000, 36166000, 166666, 0xca, 0xc0}, + { 253834000, 36166000, 166666, 0xca, 0x60}, + { 383834000, 36166000, 166666, 0xca, 0xa0}, + { 443834000, 36166000, 166666, 0xca, 0xc0}, + { 583834000, 36166000, 166666, 0xca, 0x60}, + { 793834000, 36166000, 166666, 0xca, 0xa0}, + { 858834000, 36166000, 166666, 0xca, 0xe0}, + }, +}; +EXPORT_SYMBOL(dvb_pll_philips_td1316); + +/* FE6600 used on DViCO Hybrid */ +struct dvb_pll_desc dvb_pll_thomson_fe6600 = { + .name = "Thomson FE6600", + .min = 44250000, + .max = 858000000, + .count = 4, + .entries = { + { 250000000, 36213333, 166667, 0xb4, 0x12 }, + { 455000000, 36213333, 166667, 0xfe, 0x11 }, + { 775500000, 36213333, 166667, 0xbc, 0x18 }, + { 999999999, 36213333, 166667, 0xf4, 0x18 }, + } +}; +EXPORT_SYMBOL(dvb_pll_thomson_fe6600); + /* ----------------------------------------------------------- */ /* code */ @@ -143,11 +448,11 @@ int dvb_pll_configure(struct dvb_pll_desc *desc, u8 *buf, div = (freq + desc->entries[i].offset) / desc->entries[i].stepsize; buf[0] = div >> 8; buf[1] = div & 0xff; - buf[2] = desc->entries[i].cb1; - buf[3] = desc->entries[i].cb2; + buf[2] = desc->entries[i].config; + buf[3] = desc->entries[i].cb; if (desc->setbw) - desc->setbw(buf, bandwidth); + desc->setbw(buf, freq, bandwidth); if (debug) printk("pll: %s: div=%d | buf=0x%02x,0x%02x,0x%02x,0x%02x\n", @@ -160,9 +465,3 @@ EXPORT_SYMBOL(dvb_pll_configure); MODULE_DESCRIPTION("dvb pll library"); MODULE_AUTHOR("Gerd Knorr"); MODULE_LICENSE("GPL"); - -/* - * Local variables: - * c-basic-offset: 8 - * End: - */ diff --git a/drivers/media/dvb/ttpci/budget-av.c b/drivers/media/dvb/ttpci/budget-av.c index 8a7cd7d50..3f4aa083f 100644 --- a/drivers/media/dvb/ttpci/budget-av.c +++ b/drivers/media/dvb/ttpci/budget-av.c @@ -58,6 +58,7 @@ struct budget_av { struct tasklet_struct ciintf_irq_tasklet; int slot_status; struct dvb_ca_en50221 ca; + u8 reinitialise_demod:1; }; /* GPIO Connections: @@ -214,8 +215,9 @@ static int ciintf_slot_reset(struct dvb_ca_en50221 *ca, int slot) while (--timeout > 0 && ciintf_read_attribute_mem(ca, slot, 0) != 0x1d) msleep(100); - /* reinitialise the frontend */ - dvb_frontend_reinitialise(budget_av->budget.dvb_frontend); + /* reinitialise the frontend if necessary */ + if (budget_av->reinitialise_demod) + dvb_frontend_reinitialise(budget_av->budget.dvb_frontend); if (timeout <= 0) { @@ -1017,12 +1019,13 @@ static void frontend_init(struct budget_av *budget_av) struct saa7146_dev * saa = budget_av->budget.dev; struct dvb_frontend * fe = NULL; + /* Enable / PowerON Frontend */ + saa7146_setgpio(saa, 0, SAA7146_GPIO_OUTLO); + switch (saa->pci->subsystem_device) { case SUBID_DVBS_KNC1_PLUS: case SUBID_DVBC_KNC1_PLUS: case SUBID_DVBT_KNC1_PLUS: - // Enable / PowerON Frontend - saa7146_setgpio(saa, 0, SAA7146_GPIO_OUTLO); saa7146_setgpio(saa, 3, SAA7146_GPIO_OUTHI); break; } @@ -1059,7 +1062,14 @@ static void frontend_init(struct budget_av *budget_av) break; case SUBID_DVBC_KNC1: + budget_av->reinitialise_demod = 1; + fe = tda10021_attach(&philips_cu1216_config, + &budget_av->budget.i2c_adap, + read_pwm(budget_av)); + break; + case SUBID_DVBC_KNC1_PLUS: + budget_av->reinitialise_demod = 1; fe = tda10021_attach(&philips_cu1216_config, &budget_av->budget.i2c_adap, read_pwm(budget_av)); @@ -1208,11 +1218,7 @@ static int budget_av_attach(struct saa7146_dev *dev, struct saa7146_pci_extensio budget_av->budget.dvb_adapter.priv = budget_av; frontend_init(budget_av); - - if (!budget_av->has_saa7113) { - ciintf_init(budget_av); - } - + ciintf_init(budget_av); return 0; } diff --git a/drivers/media/dvb/ttpci/budget.c b/drivers/media/dvb/ttpci/budget.c index 689e97189..d6d37cf02 100644 --- a/drivers/media/dvb/ttpci/budget.c +++ b/drivers/media/dvb/ttpci/budget.c @@ -367,12 +367,6 @@ static void frontend_init(struct budget *budget) // try the ALPS BSRU6 now budget->dvb_frontend = stv0299_attach(&alps_bsru6_config, &budget->i2c_adap); - if (budget->dvb_frontend) { - budget->dvb_frontend->ops->diseqc_send_master_cmd = budget_diseqc_send_master_cmd; - budget->dvb_frontend->ops->diseqc_send_burst = budget_diseqc_send_burst; - budget->dvb_frontend->ops->set_tone = budget_set_tone; - break; - } break; case 0x1004: // Hauppauge/TT DVB-C budget (ves1820/ALPS TDBE2(sp5659)) diff --git a/drivers/media/video/Kconfig b/drivers/media/video/Kconfig index 6b4197018..67d729a84 100644 --- a/drivers/media/video/Kconfig +++ b/drivers/media/video/Kconfig @@ -380,10 +380,10 @@ config VIDEO_WM8739 source "drivers/media/video/cx25840/Kconfig" config VIDEO_SAA711X - tristate "Philips SAA7113/4/5 video decoders (OBSOLETED)" - depends on VIDEO_V4L1 && I2C && EXPERIMENTAL + tristate "Philips SAA7113/4/5 video decoders" + depends on VIDEO_DEV && I2C && EXPERIMENTAL ---help--- - Old support for the Philips SAA7113/4 video decoders. + Support for the Philips SAA7113/4/5 video decoders. To compile this driver as a module, choose M here: the module will be called saa7115. diff --git a/drivers/media/video/stradis.c b/drivers/media/video/stradis.c index 07476c711..af372dd25 100644 --- a/drivers/media/video/stradis.c +++ b/drivers/media/video/stradis.c @@ -2180,7 +2180,6 @@ static struct pci_device_id stradis_pci_tbl[] = { { 0 } }; -MODULE_DEVICE_TABLE(pci, stradis_pci_tbl); static struct pci_driver stradis_driver = { .name = "stradis", diff --git a/drivers/net/8139cp.c b/drivers/net/8139cp.c index 066e22b01..365a07494 100644 --- a/drivers/net/8139cp.c +++ b/drivers/net/8139cp.c @@ -792,7 +792,7 @@ static int cp_start_xmit (struct sk_buff *skb, struct net_device *dev) entry = cp->tx_head; eor = (entry == (CP_TX_RING_SIZE - 1)) ? RingEnd : 0; if (dev->features & NETIF_F_TSO) - mss = skb_shinfo(skb)->tso_size; + mss = skb_shinfo(skb)->gso_size; if (skb_shinfo(skb)->nr_frags == 0) { struct cp_desc *txd = &cp->tx_ring[entry]; diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c index 8acc655ec..a8f95e511 100644 --- a/drivers/net/bnx2.c +++ b/drivers/net/bnx2.c @@ -1,6 +1,6 @@ /* bnx2.c: Broadcom NX2 network driver. * - * Copyright (c) 2004, 2005 Broadcom Corporation + * Copyright (c) 2004, 2005, 2006 Broadcom Corporation * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -9,24 +9,65 @@ * Written by: Michael Chan (mchan@broadcom.com) */ +#include + +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#ifdef NETIF_F_HW_VLAN_TX +#include +#define BCM_VLAN 1 +#endif +#ifdef NETIF_F_TSO +#include +#include +#include +#define BCM_TSO 1 +#endif +#include +#include +#include +#include + #include "bnx2.h" #include "bnx2_fw.h" #define DRV_MODULE_NAME "bnx2" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "1.2.19" -#define DRV_MODULE_RELDATE "May 23, 2005" +#define DRV_MODULE_VERSION "1.4.40" +#define DRV_MODULE_RELDATE "May 22, 2006" #define RUN_AT(x) (jiffies + (x)) /* Time in jiffies before concluding the transmitter is hung. */ #define TX_TIMEOUT (5*HZ) -static char version[] __devinitdata = +static const char version[] __devinitdata = "Broadcom NetXtreme II Gigabit Ethernet Driver " DRV_MODULE_NAME " v" DRV_MODULE_VERSION " (" DRV_MODULE_RELDATE ")\n"; MODULE_AUTHOR("Michael Chan "); -MODULE_DESCRIPTION("Broadcom NetXtreme II BCM5706 Driver"); +MODULE_DESCRIPTION("Broadcom NetXtreme II BCM5706/5708 Driver"); MODULE_LICENSE("GPL"); MODULE_VERSION(DRV_MODULE_VERSION); @@ -41,10 +82,12 @@ typedef enum { NC370I, BCM5706S, NC370F, + BCM5708, + BCM5708S, } board_t; /* indexed by board_t, above */ -static struct { +static const struct { char *name; } board_info[] __devinitdata = { { "Broadcom NetXtreme II BCM5706 1000Base-T" }, @@ -52,7 +95,8 @@ static struct { { "HP NC370i Multifunction Gigabit Server Adapter" }, { "Broadcom NetXtreme II BCM5706 1000Base-SX" }, { "HP NC370F Multifunction Gigabit Server Adapter" }, - { 0 }, + { "Broadcom NetXtreme II BCM5708 1000Base-T" }, + { "Broadcom NetXtreme II BCM5708 1000Base-SX" }, }; static struct pci_device_id bnx2_pci_tbl[] = { @@ -62,52 +106,115 @@ static struct pci_device_id bnx2_pci_tbl[] = { PCI_VENDOR_ID_HP, 0x3106, 0, 0, NC370I }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706, PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706 }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5708, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5708 }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706S, PCI_VENDOR_ID_HP, 0x3102, 0, 0, NC370F }, { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5706S, PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5706S }, + { PCI_VENDOR_ID_BROADCOM, PCI_DEVICE_ID_NX2_5708S, + PCI_ANY_ID, PCI_ANY_ID, 0, 0, BCM5708S }, { 0, } }; static struct flash_spec flash_table[] = { /* Slow EEPROM */ - {0x00000000, 0x40030380, 0x009f0081, 0xa184a053, 0xaf000400, + {0x00000000, 0x40830380, 0x009f0081, 0xa184a053, 0xaf000400, 1, SEEPROM_PAGE_BITS, SEEPROM_PAGE_SIZE, SEEPROM_BYTE_ADDR_MASK, SEEPROM_TOTAL_SIZE, "EEPROM - slow"}, - /* Fast EEPROM */ - {0x02000000, 0x62008380, 0x009f0081, 0xa184a053, 0xaf000400, - 1, SEEPROM_PAGE_BITS, SEEPROM_PAGE_SIZE, - SEEPROM_BYTE_ADDR_MASK, SEEPROM_TOTAL_SIZE, - "EEPROM - fast"}, - /* ATMEL AT45DB011B (buffered flash) */ - {0x02000003, 0x6e008173, 0x00570081, 0x68848353, 0xaf000400, - 1, BUFFERED_FLASH_PAGE_BITS, BUFFERED_FLASH_PAGE_SIZE, - BUFFERED_FLASH_BYTE_ADDR_MASK, BUFFERED_FLASH_TOTAL_SIZE, - "Buffered flash"}, - /* Saifun SA25F005 (non-buffered flash) */ - /* strap, cfg1, & write1 need updates */ - {0x01000003, 0x5f008081, 0x00050081, 0x03840253, 0xaf020406, + /* Expansion entry 0001 */ + {0x08000002, 0x4b808201, 0x00050081, 0x03840253, 0xaf020406, 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, - SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE, - "Non-buffered flash (64kB)"}, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 0001"}, /* Saifun SA25F010 (non-buffered flash) */ /* strap, cfg1, & write1 need updates */ - {0x00000001, 0x47008081, 0x00050081, 0x03840253, 0xaf020406, + {0x04000001, 0x47808201, 0x00050081, 0x03840253, 0xaf020406, 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE*2, "Non-buffered flash (128kB)"}, /* Saifun SA25F020 (non-buffered flash) */ /* strap, cfg1, & write1 need updates */ - {0x00000003, 0x4f008081, 0x00050081, 0x03840253, 0xaf020406, + {0x0c000003, 0x4f808201, 0x00050081, 0x03840253, 0xaf020406, 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE*4, "Non-buffered flash (256kB)"}, + /* Expansion entry 0100 */ + {0x11000000, 0x53808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 0100"}, + /* Entry 0101: ST M45PE10 (non-buffered flash, TetonII B0) */ + {0x19000002, 0x5b808201, 0x000500db, 0x03840253, 0xaf020406, + 0, ST_MICRO_FLASH_PAGE_BITS, ST_MICRO_FLASH_PAGE_SIZE, + ST_MICRO_FLASH_BYTE_ADDR_MASK, ST_MICRO_FLASH_BASE_TOTAL_SIZE*2, + "Entry 0101: ST M45PE10 (128kB non-bufferred)"}, + /* Entry 0110: ST M45PE20 (non-buffered flash)*/ + {0x15000001, 0x57808201, 0x000500db, 0x03840253, 0xaf020406, + 0, ST_MICRO_FLASH_PAGE_BITS, ST_MICRO_FLASH_PAGE_SIZE, + ST_MICRO_FLASH_BYTE_ADDR_MASK, ST_MICRO_FLASH_BASE_TOTAL_SIZE*4, + "Entry 0110: ST M45PE20 (256kB non-bufferred)"}, + /* Saifun SA25F005 (non-buffered flash) */ + /* strap, cfg1, & write1 need updates */ + {0x1d000003, 0x5f808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, SAIFUN_FLASH_BASE_TOTAL_SIZE, + "Non-buffered flash (64kB)"}, + /* Fast EEPROM */ + {0x22000000, 0x62808380, 0x009f0081, 0xa184a053, 0xaf000400, + 1, SEEPROM_PAGE_BITS, SEEPROM_PAGE_SIZE, + SEEPROM_BYTE_ADDR_MASK, SEEPROM_TOTAL_SIZE, + "EEPROM - fast"}, + /* Expansion entry 1001 */ + {0x2a000002, 0x6b808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 1001"}, + /* Expansion entry 1010 */ + {0x26000001, 0x67808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 1010"}, + /* ATMEL AT45DB011B (buffered flash) */ + {0x2e000003, 0x6e808273, 0x00570081, 0x68848353, 0xaf000400, + 1, BUFFERED_FLASH_PAGE_BITS, BUFFERED_FLASH_PAGE_SIZE, + BUFFERED_FLASH_BYTE_ADDR_MASK, BUFFERED_FLASH_TOTAL_SIZE, + "Buffered flash (128kB)"}, + /* Expansion entry 1100 */ + {0x33000000, 0x73808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 1100"}, + /* Expansion entry 1101 */ + {0x3b000002, 0x7b808201, 0x00050081, 0x03840253, 0xaf020406, + 0, SAIFUN_FLASH_PAGE_BITS, SAIFUN_FLASH_PAGE_SIZE, + SAIFUN_FLASH_BYTE_ADDR_MASK, 0, + "Entry 1101"}, + /* Ateml Expansion entry 1110 */ + {0x37000001, 0x76808273, 0x00570081, 0x68848353, 0xaf000400, + 1, BUFFERED_FLASH_PAGE_BITS, BUFFERED_FLASH_PAGE_SIZE, + BUFFERED_FLASH_BYTE_ADDR_MASK, 0, + "Entry 1110 (Atmel)"}, + /* ATMEL AT45DB021B (buffered flash) */ + {0x3f000003, 0x7e808273, 0x00570081, 0x68848353, 0xaf000400, + 1, BUFFERED_FLASH_PAGE_BITS, BUFFERED_FLASH_PAGE_SIZE, + BUFFERED_FLASH_BYTE_ADDR_MASK, BUFFERED_FLASH_TOTAL_SIZE*2, + "Buffered flash (256kB)"}, }; MODULE_DEVICE_TABLE(pci, bnx2_pci_tbl); +static inline u32 bnx2_tx_avail(struct bnx2 *bp) +{ + u32 diff = TX_RING_IDX(bp->tx_prod) - TX_RING_IDX(bp->tx_cons); + + if (diff > MAX_TX_DESC_CNT) + diff = (diff & MAX_TX_DESC_CNT) - 1; + return (bp->tx_ring_size - diff); +} + static u32 bnx2_reg_rd_ind(struct bnx2 *bp, u32 offset) { @@ -247,13 +354,14 @@ bnx2_disable_int(struct bnx2 *bp) static void bnx2_enable_int(struct bnx2 *bp) { - u32 val; + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | + BNX2_PCICFG_INT_ACK_CMD_MASK_INT | bp->last_status_idx); REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | bp->last_status_idx); - val = REG_RD(bp, BNX2_HC_COMMAND); - REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW); + REG_WR(bp, BNX2_HC_COMMAND, bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW); } static void @@ -290,15 +398,13 @@ bnx2_netif_start(struct bnx2 *bp) static void bnx2_free_mem(struct bnx2 *bp) { - if (bp->stats_blk) { - pci_free_consistent(bp->pdev, sizeof(struct statistics_block), - bp->stats_blk, bp->stats_blk_mapping); - bp->stats_blk = NULL; - } + int i; + if (bp->status_blk) { - pci_free_consistent(bp->pdev, sizeof(struct status_block), + pci_free_consistent(bp->pdev, bp->status_stats_size, bp->status_blk, bp->status_blk_mapping); bp->status_blk = NULL; + bp->stats_blk = NULL; } if (bp->tx_desc_ring) { pci_free_consistent(bp->pdev, @@ -306,31 +412,30 @@ bnx2_free_mem(struct bnx2 *bp) bp->tx_desc_ring, bp->tx_desc_mapping); bp->tx_desc_ring = NULL; } - if (bp->tx_buf_ring) { - kfree(bp->tx_buf_ring); - bp->tx_buf_ring = NULL; - } - if (bp->rx_desc_ring) { - pci_free_consistent(bp->pdev, - sizeof(struct rx_bd) * RX_DESC_CNT, - bp->rx_desc_ring, bp->rx_desc_mapping); - bp->rx_desc_ring = NULL; - } - if (bp->rx_buf_ring) { - kfree(bp->rx_buf_ring); - bp->rx_buf_ring = NULL; - } + kfree(bp->tx_buf_ring); + bp->tx_buf_ring = NULL; + for (i = 0; i < bp->rx_max_ring; i++) { + if (bp->rx_desc_ring[i]) + pci_free_consistent(bp->pdev, + sizeof(struct rx_bd) * RX_DESC_CNT, + bp->rx_desc_ring[i], + bp->rx_desc_mapping[i]); + bp->rx_desc_ring[i] = NULL; + } + vfree(bp->rx_buf_ring); + bp->rx_buf_ring = NULL; } static int bnx2_alloc_mem(struct bnx2 *bp) { - bp->tx_buf_ring = kmalloc(sizeof(struct sw_bd) * TX_DESC_CNT, - GFP_KERNEL); + int i, status_blk_size; + + bp->tx_buf_ring = kzalloc(sizeof(struct sw_bd) * TX_DESC_CNT, + GFP_KERNEL); if (bp->tx_buf_ring == NULL) return -ENOMEM; - memset(bp->tx_buf_ring, 0, sizeof(struct sw_bd) * TX_DESC_CNT); bp->tx_desc_ring = pci_alloc_consistent(bp->pdev, sizeof(struct tx_bd) * TX_DESC_CNT, @@ -338,34 +443,40 @@ bnx2_alloc_mem(struct bnx2 *bp) if (bp->tx_desc_ring == NULL) goto alloc_mem_err; - bp->rx_buf_ring = kmalloc(sizeof(struct sw_bd) * RX_DESC_CNT, - GFP_KERNEL); + bp->rx_buf_ring = vmalloc(sizeof(struct sw_bd) * RX_DESC_CNT * + bp->rx_max_ring); if (bp->rx_buf_ring == NULL) goto alloc_mem_err; - memset(bp->rx_buf_ring, 0, sizeof(struct sw_bd) * RX_DESC_CNT); - bp->rx_desc_ring = pci_alloc_consistent(bp->pdev, - sizeof(struct rx_bd) * - RX_DESC_CNT, - &bp->rx_desc_mapping); - if (bp->rx_desc_ring == NULL) - goto alloc_mem_err; + memset(bp->rx_buf_ring, 0, sizeof(struct sw_bd) * RX_DESC_CNT * + bp->rx_max_ring); + + for (i = 0; i < bp->rx_max_ring; i++) { + bp->rx_desc_ring[i] = + pci_alloc_consistent(bp->pdev, + sizeof(struct rx_bd) * RX_DESC_CNT, + &bp->rx_desc_mapping[i]); + if (bp->rx_desc_ring[i] == NULL) + goto alloc_mem_err; + + } + + /* Combine status and statistics blocks into one allocation. */ + status_blk_size = L1_CACHE_ALIGN(sizeof(struct status_block)); + bp->status_stats_size = status_blk_size + + sizeof(struct statistics_block); - bp->status_blk = pci_alloc_consistent(bp->pdev, - sizeof(struct status_block), + bp->status_blk = pci_alloc_consistent(bp->pdev, bp->status_stats_size, &bp->status_blk_mapping); if (bp->status_blk == NULL) goto alloc_mem_err; - memset(bp->status_blk, 0, sizeof(struct status_block)); + memset(bp->status_blk, 0, bp->status_stats_size); - bp->stats_blk = pci_alloc_consistent(bp->pdev, - sizeof(struct statistics_block), - &bp->stats_blk_mapping); - if (bp->stats_blk == NULL) - goto alloc_mem_err; + bp->stats_blk = (void *) ((unsigned long) bp->status_blk + + status_blk_size); - memset(bp->stats_blk, 0, sizeof(struct statistics_block)); + bp->stats_blk_mapping = bp->status_blk_mapping + status_blk_size; return 0; @@ -374,6 +485,62 @@ alloc_mem_err: return -ENOMEM; } +static void +bnx2_report_fw_link(struct bnx2 *bp) +{ + u32 fw_link_status = 0; + + if (bp->link_up) { + u32 bmsr; + + switch (bp->line_speed) { + case SPEED_10: + if (bp->duplex == DUPLEX_HALF) + fw_link_status = BNX2_LINK_STATUS_10HALF; + else + fw_link_status = BNX2_LINK_STATUS_10FULL; + break; + case SPEED_100: + if (bp->duplex == DUPLEX_HALF) + fw_link_status = BNX2_LINK_STATUS_100HALF; + else + fw_link_status = BNX2_LINK_STATUS_100FULL; + break; + case SPEED_1000: + if (bp->duplex == DUPLEX_HALF) + fw_link_status = BNX2_LINK_STATUS_1000HALF; + else + fw_link_status = BNX2_LINK_STATUS_1000FULL; + break; + case SPEED_2500: + if (bp->duplex == DUPLEX_HALF) + fw_link_status = BNX2_LINK_STATUS_2500HALF; + else + fw_link_status = BNX2_LINK_STATUS_2500FULL; + break; + } + + fw_link_status |= BNX2_LINK_STATUS_LINK_UP; + + if (bp->autoneg) { + fw_link_status |= BNX2_LINK_STATUS_AN_ENABLED; + + bnx2_read_phy(bp, MII_BMSR, &bmsr); + bnx2_read_phy(bp, MII_BMSR, &bmsr); + + if (!(bmsr & BMSR_ANEGCOMPLETE) || + bp->phy_flags & PHY_PARALLEL_DETECT_FLAG) + fw_link_status |= BNX2_LINK_STATUS_PARALLEL_DET; + else + fw_link_status |= BNX2_LINK_STATUS_AN_COMPLETE; + } + } + else + fw_link_status = BNX2_LINK_STATUS_LINK_DOWN; + + REG_WR_IND(bp, bp->shmem_base + BNX2_LINK_STATUS, fw_link_status); +} + static void bnx2_report_link(struct bnx2 *bp) { @@ -405,6 +572,8 @@ bnx2_report_link(struct bnx2 *bp) netif_carrier_off(bp->dev); printk(KERN_ERR PFX "%s NIC Link is Down\n", bp->dev->name); } + + bnx2_report_fw_link(bp); } static void @@ -426,6 +595,18 @@ bnx2_resolve_flow_ctrl(struct bnx2 *bp) return; } + if ((bp->phy_flags & PHY_SERDES_FLAG) && + (CHIP_NUM(bp) == CHIP_NUM_5708)) { + u32 val; + + bnx2_read_phy(bp, BCM5708S_1000X_STAT1, &val); + if (val & BCM5708S_1000X_STAT1_TX_PAUSE) + bp->flow_ctrl |= FLOW_CTRL_TX; + if (val & BCM5708S_1000X_STAT1_RX_PAUSE) + bp->flow_ctrl |= FLOW_CTRL_RX; + return; + } + bnx2_read_phy(bp, MII_ADVERTISE, &local_adv); bnx2_read_phy(bp, MII_LPA, &remote_adv); @@ -472,7 +653,36 @@ bnx2_resolve_flow_ctrl(struct bnx2 *bp) } static int -bnx2_serdes_linkup(struct bnx2 *bp) +bnx2_5708s_linkup(struct bnx2 *bp) +{ + u32 val; + + bp->link_up = 1; + bnx2_read_phy(bp, BCM5708S_1000X_STAT1, &val); + switch (val & BCM5708S_1000X_STAT1_SPEED_MASK) { + case BCM5708S_1000X_STAT1_SPEED_10: + bp->line_speed = SPEED_10; + break; + case BCM5708S_1000X_STAT1_SPEED_100: + bp->line_speed = SPEED_100; + break; + case BCM5708S_1000X_STAT1_SPEED_1G: + bp->line_speed = SPEED_1000; + break; + case BCM5708S_1000X_STAT1_SPEED_2G5: + bp->line_speed = SPEED_2500; + break; + } + if (val & BCM5708S_1000X_STAT1_FD) + bp->duplex = DUPLEX_FULL; + else + bp->duplex = DUPLEX_HALF; + + return 0; +} + +static int +bnx2_5706s_linkup(struct bnx2 *bp) { u32 bmcr, local_adv, remote_adv, common; @@ -589,13 +799,27 @@ bnx2_set_mac_link(struct bnx2 *bp) val = REG_RD(bp, BNX2_EMAC_MODE); val &= ~(BNX2_EMAC_MODE_PORT | BNX2_EMAC_MODE_HALF_DUPLEX | - BNX2_EMAC_MODE_MAC_LOOP | BNX2_EMAC_MODE_FORCE_LINK); + BNX2_EMAC_MODE_MAC_LOOP | BNX2_EMAC_MODE_FORCE_LINK | + BNX2_EMAC_MODE_25G); if (bp->link_up) { - if (bp->line_speed != SPEED_1000) - val |= BNX2_EMAC_MODE_PORT_MII; - else - val |= BNX2_EMAC_MODE_PORT_GMII; + switch (bp->line_speed) { + case SPEED_10: + if (CHIP_NUM(bp) == CHIP_NUM_5708) { + val |= BNX2_EMAC_MODE_PORT_MII_10; + break; + } + /* fall through */ + case SPEED_100: + val |= BNX2_EMAC_MODE_PORT_MII; + break; + case SPEED_2500: + val |= BNX2_EMAC_MODE_25G; + /* fall through */ + case SPEED_1000: + val |= BNX2_EMAC_MODE_PORT_GMII; + break; + } } else { val |= BNX2_EMAC_MODE_PORT_GMII; @@ -658,7 +882,10 @@ bnx2_set_link(struct bnx2 *bp) bp->link_up = 1; if (bp->phy_flags & PHY_SERDES_FLAG) { - bnx2_serdes_linkup(bp); + if (CHIP_NUM(bp) == CHIP_NUM_5706) + bnx2_5706s_linkup(bp); + else if (CHIP_NUM(bp) == CHIP_NUM_5708) + bnx2_5708s_linkup(bp); } else { bnx2_copper_linkup(bp); @@ -751,39 +978,61 @@ bnx2_phy_get_pause_adv(struct bnx2 *bp) static int bnx2_setup_serdes_phy(struct bnx2 *bp) { - u32 adv, bmcr; + u32 adv, bmcr, up1; u32 new_adv = 0; if (!(bp->autoneg & AUTONEG_SPEED)) { u32 new_bmcr; + int force_link_down = 0; + + if (CHIP_NUM(bp) == CHIP_NUM_5708) { + bnx2_read_phy(bp, BCM5708S_UP1, &up1); + if (up1 & BCM5708S_UP1_2G5) { + up1 &= ~BCM5708S_UP1_2G5; + bnx2_write_phy(bp, BCM5708S_UP1, up1); + force_link_down = 1; + } + } + + bnx2_read_phy(bp, MII_ADVERTISE, &adv); + adv &= ~(ADVERTISE_1000XFULL | ADVERTISE_1000XHALF); bnx2_read_phy(bp, MII_BMCR, &bmcr); new_bmcr = bmcr & ~BMCR_ANENABLE; new_bmcr |= BMCR_SPEED1000; if (bp->req_duplex == DUPLEX_FULL) { + adv |= ADVERTISE_1000XFULL; new_bmcr |= BMCR_FULLDPLX; } else { + adv |= ADVERTISE_1000XHALF; new_bmcr &= ~BMCR_FULLDPLX; } - if (new_bmcr != bmcr) { + if ((new_bmcr != bmcr) || (force_link_down)) { /* Force a link down visible on the other side */ if (bp->link_up) { - bnx2_read_phy(bp, MII_ADVERTISE, &adv); - adv &= ~(ADVERTISE_1000XFULL | - ADVERTISE_1000XHALF); - bnx2_write_phy(bp, MII_ADVERTISE, adv); + bnx2_write_phy(bp, MII_ADVERTISE, adv & + ~(ADVERTISE_1000XFULL | + ADVERTISE_1000XHALF)); bnx2_write_phy(bp, MII_BMCR, bmcr | BMCR_ANRESTART | BMCR_ANENABLE); bp->link_up = 0; netif_carrier_off(bp->dev); + bnx2_write_phy(bp, MII_BMCR, new_bmcr); } + bnx2_write_phy(bp, MII_ADVERTISE, adv); bnx2_write_phy(bp, MII_BMCR, new_bmcr); } return 0; } + if (bp->phy_flags & PHY_2_5G_CAPABLE_FLAG) { + bnx2_read_phy(bp, BCM5708S_UP1, &up1); + up1 |= BCM5708S_UP1_2G5; + bnx2_write_phy(bp, BCM5708S_UP1, up1); + } + if (bp->advertising & ADVERTISED_1000baseT_Full) new_adv |= ADVERTISE_1000XFULL; @@ -807,7 +1056,19 @@ bnx2_setup_serdes_phy(struct bnx2 *bp) bnx2_write_phy(bp, MII_ADVERTISE, new_adv); bnx2_write_phy(bp, MII_BMCR, bmcr | BMCR_ANRESTART | BMCR_ANENABLE); - bp->serdes_an_pending = SERDES_AN_TIMEOUT / bp->timer_interval; + if (CHIP_NUM(bp) == CHIP_NUM_5706) { + /* Speed up link-up time when the link partner + * does not autonegotiate which is very common + * in blade servers. Some blade servers use + * IPMI for kerboard input and it's important + * to minimize link disruptions. Autoneg. involves + * exchanging base pages plus 3 next pages and + * normally completes in about 120 msec. + */ + bp->current_interval = SERDES_AN_TIMEOUT; + bp->serdes_an_pending = 1; + mod_timer(&bp->timer, jiffies + bp->current_interval); + } } return 0; @@ -936,7 +1197,61 @@ bnx2_setup_phy(struct bnx2 *bp) } static int -bnx2_init_serdes_phy(struct bnx2 *bp) +bnx2_init_5708s_phy(struct bnx2 *bp) +{ + u32 val; + + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, BCM5708S_BLK_ADDR_DIG3); + bnx2_write_phy(bp, BCM5708S_DIG_3_0, BCM5708S_DIG_3_0_USE_IEEE); + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, BCM5708S_BLK_ADDR_DIG); + + bnx2_read_phy(bp, BCM5708S_1000X_CTL1, &val); + val |= BCM5708S_1000X_CTL1_FIBER_MODE | BCM5708S_1000X_CTL1_AUTODET_EN; + bnx2_write_phy(bp, BCM5708S_1000X_CTL1, val); + + bnx2_read_phy(bp, BCM5708S_1000X_CTL2, &val); + val |= BCM5708S_1000X_CTL2_PLLEL_DET_EN; + bnx2_write_phy(bp, BCM5708S_1000X_CTL2, val); + + if (bp->phy_flags & PHY_2_5G_CAPABLE_FLAG) { + bnx2_read_phy(bp, BCM5708S_UP1, &val); + val |= BCM5708S_UP1_2G5; + bnx2_write_phy(bp, BCM5708S_UP1, val); + } + + if ((CHIP_ID(bp) == CHIP_ID_5708_A0) || + (CHIP_ID(bp) == CHIP_ID_5708_B0) || + (CHIP_ID(bp) == CHIP_ID_5708_B1)) { + /* increase tx signal amplitude */ + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, + BCM5708S_BLK_ADDR_TX_MISC); + bnx2_read_phy(bp, BCM5708S_TX_ACTL1, &val); + val &= ~BCM5708S_TX_ACTL1_DRIVER_VCM; + bnx2_write_phy(bp, BCM5708S_TX_ACTL1, val); + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, BCM5708S_BLK_ADDR_DIG); + } + + val = REG_RD_IND(bp, bp->shmem_base + BNX2_PORT_HW_CFG_CONFIG) & + BNX2_PORT_HW_CFG_CFG_TXCTL3_MASK; + + if (val) { + u32 is_backplane; + + is_backplane = REG_RD_IND(bp, bp->shmem_base + + BNX2_SHARED_HW_CFG_CONFIG); + if (is_backplane & BNX2_SHARED_HW_CFG_PHY_BACKPLANE) { + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, + BCM5708S_BLK_ADDR_TX_MISC); + bnx2_write_phy(bp, BCM5708S_TX_ACTL3, val); + bnx2_write_phy(bp, BCM5708S_BLK_ADDR, + BCM5708S_BLK_ADDR_DIG); + } + } + return 0; +} + +static int +bnx2_init_5706s_phy(struct bnx2 *bp) { bp->phy_flags &= ~PHY_PARALLEL_DETECT_FLAG; @@ -974,6 +1289,8 @@ bnx2_init_serdes_phy(struct bnx2 *bp) static int bnx2_init_copper_phy(struct bnx2 *bp) { + u32 val; + bp->phy_flags |= PHY_CRC_FIX_FLAG; if (bp->phy_flags & PHY_CRC_FIX_FLAG) { @@ -988,8 +1305,6 @@ bnx2_init_copper_phy(struct bnx2 *bp) } if (bp->dev->mtu > 1500) { - u32 val; - /* Set extended packet length bit */ bnx2_write_phy(bp, 0x18, 0x7); bnx2_read_phy(bp, 0x18, &val); @@ -999,8 +1314,6 @@ bnx2_init_copper_phy(struct bnx2 *bp) bnx2_write_phy(bp, 0x10, val | 0x1); } else { - u32 val; - bnx2_write_phy(bp, 0x18, 0x7); bnx2_read_phy(bp, 0x18, &val); bnx2_write_phy(bp, 0x18, val & ~0x4007); @@ -1009,6 +1322,10 @@ bnx2_init_copper_phy(struct bnx2 *bp) bnx2_write_phy(bp, 0x10, val & ~0x1); } + /* ethernet@wirespeed */ + bnx2_write_phy(bp, 0x18, 0x7007); + bnx2_read_phy(bp, 0x18, &val); + bnx2_write_phy(bp, 0x18, val | (1 << 15) | (1 << 4)); return 0; } @@ -1032,7 +1349,10 @@ bnx2_init_phy(struct bnx2 *bp) bp->phy_id |= val & 0xffff; if (bp->phy_flags & PHY_SERDES_FLAG) { - rc = bnx2_init_serdes_phy(bp); + if (CHIP_NUM(bp) == CHIP_NUM_5706) + rc = bnx2_init_5706s_phy(bp); + else if (CHIP_NUM(bp) == CHIP_NUM_5708) + rc = bnx2_init_5708s_phy(bp); } else { rc = bnx2_init_copper_phy(bp); @@ -1056,44 +1376,78 @@ bnx2_set_mac_loopback(struct bnx2 *bp) return 0; } +static int bnx2_test_link(struct bnx2 *); + +static int +bnx2_set_phy_loopback(struct bnx2 *bp) +{ + u32 mac_mode; + int rc, i; + + spin_lock_bh(&bp->phy_lock); + rc = bnx2_write_phy(bp, MII_BMCR, BMCR_LOOPBACK | BMCR_FULLDPLX | + BMCR_SPEED1000); + spin_unlock_bh(&bp->phy_lock); + if (rc) + return rc; + + for (i = 0; i < 10; i++) { + if (bnx2_test_link(bp) == 0) + break; + udelay(10); + } + + mac_mode = REG_RD(bp, BNX2_EMAC_MODE); + mac_mode &= ~(BNX2_EMAC_MODE_PORT | BNX2_EMAC_MODE_HALF_DUPLEX | + BNX2_EMAC_MODE_MAC_LOOP | BNX2_EMAC_MODE_FORCE_LINK | + BNX2_EMAC_MODE_25G); + + mac_mode |= BNX2_EMAC_MODE_PORT_GMII; + REG_WR(bp, BNX2_EMAC_MODE, mac_mode); + bp->link_up = 1; + return 0; +} + static int -bnx2_fw_sync(struct bnx2 *bp, u32 msg_data) +bnx2_fw_sync(struct bnx2 *bp, u32 msg_data, int silent) { int i; u32 val; - if (bp->fw_timed_out) - return -EBUSY; - bp->fw_wr_seq++; msg_data |= bp->fw_wr_seq; - REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_MB, msg_data); + REG_WR_IND(bp, bp->shmem_base + BNX2_DRV_MB, msg_data); /* wait for an acknowledgement. */ - for (i = 0; i < (FW_ACK_TIME_OUT_MS * 1000)/5; i++) { - udelay(5); + for (i = 0; i < (FW_ACK_TIME_OUT_MS / 10); i++) { + msleep(10); - val = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_FW_MB); + val = REG_RD_IND(bp, bp->shmem_base + BNX2_FW_MB); if ((val & BNX2_FW_MSG_ACK) == (msg_data & BNX2_DRV_MSG_SEQ)) break; } + if ((msg_data & BNX2_DRV_MSG_DATA) == BNX2_DRV_MSG_DATA_WAIT0) + return 0; /* If we timed out, inform the firmware that this is the case. */ - if (((val & BNX2_FW_MSG_ACK) != (msg_data & BNX2_DRV_MSG_SEQ)) && - ((msg_data & BNX2_DRV_MSG_DATA) != BNX2_DRV_MSG_DATA_WAIT0)) { + if ((val & BNX2_FW_MSG_ACK) != (msg_data & BNX2_DRV_MSG_SEQ)) { + if (!silent) + printk(KERN_ERR PFX "fw sync timeout, reset code = " + "%x\n", msg_data); msg_data &= ~BNX2_DRV_MSG_CODE; msg_data |= BNX2_DRV_MSG_CODE_FW_TIMEOUT; - REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_MB, msg_data); - - bp->fw_timed_out = 1; + REG_WR_IND(bp, bp->shmem_base + BNX2_DRV_MB, msg_data); return -EBUSY; } + if ((val & BNX2_FW_MSG_STATUS_MASK) != BNX2_FW_MSG_STATUS_OK) + return -EIO; + return 0; } @@ -1211,7 +1565,7 @@ bnx2_alloc_rx_skb(struct bnx2 *bp, u16 index) struct sk_buff *skb; struct sw_bd *rx_buf = &bp->rx_buf_ring[index]; dma_addr_t mapping; - struct rx_bd *rxbd = &bp->rx_desc_ring[index]; + struct rx_bd *rxbd = &bp->rx_desc_ring[RX_RING(index)][RX_IDX(index)]; unsigned long align; skb = dev_alloc_skb(bp->rx_buf_size); @@ -1263,10 +1617,11 @@ bnx2_phy_int(struct bnx2 *bp) static void bnx2_tx_int(struct bnx2 *bp) { + struct status_block *sblk = bp->status_blk; u16 hw_cons, sw_cons, sw_ring_cons; int tx_free_bd = 0; - hw_cons = bp->status_blk->status_tx_quick_consumer_index0; + hw_cons = bp->hw_tx_cons = sblk->status_tx_quick_consumer_index0; if ((hw_cons & MAX_TX_DESC_CNT) == MAX_TX_DESC_CNT) { hw_cons++; } @@ -1283,7 +1638,7 @@ bnx2_tx_int(struct bnx2 *bp) skb = tx_buf->skb; #ifdef BCM_TSO /* partial BD completions possible with TSO packets */ - if (skb_shinfo(skb)->tso_size) { + if (skb_is_gso(skb)) { u16 last_idx, last_ring_idx; last_idx = sw_cons + @@ -1321,61 +1676,66 @@ bnx2_tx_int(struct bnx2 *bp) dev_kfree_skb_irq(skb); - hw_cons = bp->status_blk->status_tx_quick_consumer_index0; + hw_cons = bp->hw_tx_cons = + sblk->status_tx_quick_consumer_index0; + if ((hw_cons & MAX_TX_DESC_CNT) == MAX_TX_DESC_CNT) { hw_cons++; } } - atomic_add(tx_free_bd, &bp->tx_avail_bd); + bp->tx_cons = sw_cons; if (unlikely(netif_queue_stopped(bp->dev))) { - unsigned long flags; - - spin_lock_irqsave(&bp->tx_lock, flags); + spin_lock(&bp->tx_lock); if ((netif_queue_stopped(bp->dev)) && - (atomic_read(&bp->tx_avail_bd) > MAX_SKB_FRAGS)) { + (bnx2_tx_avail(bp) > MAX_SKB_FRAGS)) { netif_wake_queue(bp->dev); } - spin_unlock_irqrestore(&bp->tx_lock, flags); + spin_unlock(&bp->tx_lock); } - - bp->tx_cons = sw_cons; - } static inline void bnx2_reuse_rx_skb(struct bnx2 *bp, struct sk_buff *skb, u16 cons, u16 prod) { - struct sw_bd *cons_rx_buf = &bp->rx_buf_ring[cons]; - struct sw_bd *prod_rx_buf = &bp->rx_buf_ring[prod]; - struct rx_bd *cons_bd = &bp->rx_desc_ring[cons]; - struct rx_bd *prod_bd = &bp->rx_desc_ring[prod]; + struct sw_bd *cons_rx_buf, *prod_rx_buf; + struct rx_bd *cons_bd, *prod_bd; + + cons_rx_buf = &bp->rx_buf_ring[cons]; + prod_rx_buf = &bp->rx_buf_ring[prod]; pci_dma_sync_single_for_device(bp->pdev, pci_unmap_addr(cons_rx_buf, mapping), bp->rx_offset + RX_COPY_THRESH, PCI_DMA_FROMDEVICE); - prod_rx_buf->skb = cons_rx_buf->skb; - pci_unmap_addr_set(prod_rx_buf, mapping, - pci_unmap_addr(cons_rx_buf, mapping)); + bp->rx_prod_bseq += bp->rx_buf_use_size; - memcpy(prod_bd, cons_bd, 8); + prod_rx_buf->skb = skb; - bp->rx_prod_bseq += bp->rx_buf_use_size; + if (cons == prod) + return; + + pci_unmap_addr_set(prod_rx_buf, mapping, + pci_unmap_addr(cons_rx_buf, mapping)); + cons_bd = &bp->rx_desc_ring[RX_RING(cons)][RX_IDX(cons)]; + prod_bd = &bp->rx_desc_ring[RX_RING(prod)][RX_IDX(prod)]; + prod_bd->rx_bd_haddr_hi = cons_bd->rx_bd_haddr_hi; + prod_bd->rx_bd_haddr_lo = cons_bd->rx_bd_haddr_lo; } static int bnx2_rx_int(struct bnx2 *bp, int budget) { + struct status_block *sblk = bp->status_blk; u16 hw_cons, sw_cons, sw_ring_cons, sw_prod, sw_ring_prod; struct l2_fhdr *rx_hdr; int rx_pkt = 0; - hw_cons = bp->status_blk->status_rx_quick_consumer_index0; + hw_cons = bp->hw_rx_cons = sblk->status_rx_quick_consumer_index0; if ((hw_cons & MAX_RX_DESC_CNT) == MAX_RX_DESC_CNT) { hw_cons++; } @@ -1388,23 +1748,28 @@ bnx2_rx_int(struct bnx2 *bp, int budget) rmb(); while (sw_cons != hw_cons) { unsigned int len; - u16 status; + u32 status; struct sw_bd *rx_buf; struct sk_buff *skb; + dma_addr_t dma_addr; sw_ring_cons = RX_RING_IDX(sw_cons); sw_ring_prod = RX_RING_IDX(sw_prod); rx_buf = &bp->rx_buf_ring[sw_ring_cons]; skb = rx_buf->skb; - pci_dma_sync_single_for_cpu(bp->pdev, - pci_unmap_addr(rx_buf, mapping), + + rx_buf->skb = NULL; + + dma_addr = pci_unmap_addr(rx_buf, mapping); + + pci_dma_sync_single_for_cpu(bp->pdev, dma_addr, bp->rx_offset + RX_COPY_THRESH, PCI_DMA_FROMDEVICE); rx_hdr = (struct l2_fhdr *) skb->data; len = rx_hdr->l2_fhdr_pkt_len - 4; - if (rx_hdr->l2_fhdr_errors & + if ((status = rx_hdr->l2_fhdr_status) & (L2_FHDR_ERRORS_BAD_CRC | L2_FHDR_ERRORS_PHY_DECODE | L2_FHDR_ERRORS_ALIGNMENT | @@ -1439,8 +1804,7 @@ bnx2_rx_int(struct bnx2 *bp, int budget) skb = new_skb; } else if (bnx2_alloc_rx_skb(bp, sw_ring_prod) == 0) { - pci_unmap_single(bp->pdev, - pci_unmap_addr(rx_buf, mapping), + pci_unmap_single(bp->pdev, dma_addr, bp->rx_buf_use_size, PCI_DMA_FROMDEVICE); skb_reserve(skb, bp->rx_offset); @@ -1463,15 +1827,13 @@ reuse_rx: } - status = rx_hdr->l2_fhdr_status; skb->ip_summed = CHECKSUM_NONE; if (bp->rx_csum && (status & (L2_FHDR_STATUS_TCP_SEGMENT | L2_FHDR_STATUS_UDP_DATAGRAM))) { - u16 cksum = rx_hdr->l2_fhdr_tcp_udp_xsum; - - if (cksum == 0xffff) + if (likely((status & (L2_FHDR_ERRORS_TCP_XSUM | + L2_FHDR_ERRORS_UDP_XSUM)) == 0)) skb->ip_summed = CHECKSUM_UNNECESSARY; } @@ -1488,13 +1850,20 @@ reuse_rx: rx_pkt++; next_rx: - rx_buf->skb = NULL; - sw_cons = NEXT_RX_BD(sw_cons); sw_prod = NEXT_RX_BD(sw_prod); if ((rx_pkt == budget)) break; + + /* Refresh hw_cons to see if there is new work */ + if (sw_cons == hw_cons) { + hw_cons = bp->hw_rx_cons = + sblk->status_rx_quick_consumer_index0; + if ((hw_cons & MAX_RX_DESC_CNT) == MAX_RX_DESC_CNT) + hw_cons++; + rmb(); + } } bp->rx_cons = sw_cons; bp->rx_prod = sw_prod; @@ -1516,29 +1885,27 @@ static irqreturn_t bnx2_msi(int irq, void *dev_instance, struct pt_regs *regs) { struct net_device *dev = dev_instance; - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); + prefetch(bp->status_blk); REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, BNX2_PCICFG_INT_ACK_CMD_USE_INT_HC_PARAM | BNX2_PCICFG_INT_ACK_CMD_MASK_INT); /* Return here if interrupt is disabled. */ - if (unlikely(atomic_read(&bp->intr_sem) != 0)) { - return IRQ_RETVAL(1); - } + if (unlikely(atomic_read(&bp->intr_sem) != 0)) + return IRQ_HANDLED; - if (netif_rx_schedule_prep(dev)) { - __netif_rx_schedule(dev); - } + netif_rx_schedule(dev); - return IRQ_RETVAL(1); + return IRQ_HANDLED; } static irqreturn_t bnx2_interrupt(int irq, void *dev_instance, struct pt_regs *regs) { struct net_device *dev = dev_instance; - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); /* When using INTx, it is possible for the interrupt to arrive * at the CPU before the status block posted prior to the @@ -1546,53 +1913,66 @@ bnx2_interrupt(int irq, void *dev_instance, struct pt_regs *regs) * When using MSI, the MSI message will always complete after * the status block write. */ - if ((bp->status_blk->status_idx == bp->last_status_idx) || + if ((bp->status_blk->status_idx == bp->last_status_idx) && (REG_RD(bp, BNX2_PCICFG_MISC_STATUS) & BNX2_PCICFG_MISC_STATUS_INTA_VALUE)) - return IRQ_RETVAL(0); + return IRQ_NONE; REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, BNX2_PCICFG_INT_ACK_CMD_USE_INT_HC_PARAM | BNX2_PCICFG_INT_ACK_CMD_MASK_INT); /* Return here if interrupt is shared and is disabled. */ - if (unlikely(atomic_read(&bp->intr_sem) != 0)) { - return IRQ_RETVAL(1); - } + if (unlikely(atomic_read(&bp->intr_sem) != 0)) + return IRQ_HANDLED; - if (netif_rx_schedule_prep(dev)) { - __netif_rx_schedule(dev); - } + netif_rx_schedule(dev); + + return IRQ_HANDLED; +} + +static inline int +bnx2_has_work(struct bnx2 *bp) +{ + struct status_block *sblk = bp->status_blk; - return IRQ_RETVAL(1); + if ((sblk->status_rx_quick_consumer_index0 != bp->hw_rx_cons) || + (sblk->status_tx_quick_consumer_index0 != bp->hw_tx_cons)) + return 1; + + if (((sblk->status_attn_bits & STATUS_ATTN_BITS_LINK_STATE) != 0) != + bp->link_up) + return 1; + + return 0; } static int bnx2_poll(struct net_device *dev, int *budget) { - struct bnx2 *bp = dev->priv; - int rx_done = 1; + struct bnx2 *bp = netdev_priv(dev); - bp->last_status_idx = bp->status_blk->status_idx; - - rmb(); if ((bp->status_blk->status_attn_bits & STATUS_ATTN_BITS_LINK_STATE) != (bp->status_blk->status_attn_bits_ack & STATUS_ATTN_BITS_LINK_STATE)) { - unsigned long flags; - - spin_lock_irqsave(&bp->phy_lock, flags); + spin_lock(&bp->phy_lock); bnx2_phy_int(bp); - spin_unlock_irqrestore(&bp->phy_lock, flags); + spin_unlock(&bp->phy_lock); + + /* This is needed to take care of transient status + * during link changes. + */ + REG_WR(bp, BNX2_HC_COMMAND, + bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_RD(bp, BNX2_HC_COMMAND); } - if (bp->status_blk->status_tx_quick_consumer_index0 != bp->tx_cons) { + if (bp->status_blk->status_tx_quick_consumer_index0 != bp->hw_tx_cons) bnx2_tx_int(bp); - } - if (bp->status_blk->status_rx_quick_consumer_index0 != bp->rx_cons) { + if (bp->status_blk->status_rx_quick_consumer_index0 != bp->hw_rx_cons) { int orig_budget = *budget; int work_done; @@ -1602,45 +1982,54 @@ bnx2_poll(struct net_device *dev, int *budget) work_done = bnx2_rx_int(bp, orig_budget); *budget -= work_done; dev->quota -= work_done; - - if (work_done >= orig_budget) { - rx_done = 0; - } } - if (rx_done) { + bp->last_status_idx = bp->status_blk->status_idx; + rmb(); + + if (!bnx2_has_work(bp)) { netif_rx_complete(dev); + if (likely(bp->flags & USING_MSI_FLAG)) { + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | + bp->last_status_idx); + return 0; + } + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | + BNX2_PCICFG_INT_ACK_CMD_MASK_INT | + bp->last_status_idx); + REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, - BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | - bp->last_status_idx); + BNX2_PCICFG_INT_ACK_CMD_INDEX_VALID | + bp->last_status_idx); return 0; } return 1; } -/* Called with rtnl_lock from vlan functions and also dev->xmit_lock +/* Called with rtnl_lock from vlan functions and also netif_tx_lock * from set_multicast. */ static void bnx2_set_rx_mode(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); u32 rx_mode, sort_mode; int i; - unsigned long flags; - spin_lock_irqsave(&bp->phy_lock, flags); + spin_lock_bh(&bp->phy_lock); rx_mode = bp->rx_mode & ~(BNX2_EMAC_RX_MODE_PROMISCUOUS | BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG); sort_mode = 1 | BNX2_RPM_SORT_USER0_BC_EN; #ifdef BCM_VLAN - if (!bp->vlgrp) { + if (!bp->vlgrp && !(bp->flags & ASF_ENABLE_FLAG)) rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG; - } #else - rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG; + if (!(bp->flags & ASF_ENABLE_FLAG)) + rx_mode |= BNX2_EMAC_RX_MODE_KEEP_VLAN_TAG; #endif if (dev->flags & IFF_PROMISC) { /* Promiscuous mode. */ @@ -1691,7 +2080,7 @@ bnx2_set_rx_mode(struct net_device *dev) REG_WR(bp, BNX2_RPM_SORT_USER0, sort_mode); REG_WR(bp, BNX2_RPM_SORT_USER0, sort_mode | BNX2_RPM_SORT_USER0_ENA); - spin_unlock_irqrestore(&bp->phy_lock, flags); + spin_unlock_bh(&bp->phy_lock); } static void @@ -1998,14 +2387,14 @@ bnx2_init_cpus(struct bnx2 *bp) } static int -bnx2_set_power_state(struct bnx2 *bp, int state) +bnx2_set_power_state(struct bnx2 *bp, pci_power_t state) { u16 pmcsr; pci_read_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, &pmcsr); switch (state) { - case 0: { + case PCI_D0: { u32 val; pci_write_config_word(bp->pdev, bp->pm_cap + PCI_PM_CTRL, @@ -2026,7 +2415,7 @@ bnx2_set_power_state(struct bnx2 *bp, int state) REG_WR(bp, BNX2_RPM_CONFIG, val); break; } - case 3: { + case PCI_D3hot: { int i; u32 val, wol_msg; @@ -2058,7 +2447,6 @@ bnx2_set_power_state(struct bnx2 *bp, int state) val |= BNX2_EMAC_MODE_PORT_MII | BNX2_EMAC_MODE_MPKT_RCVD | BNX2_EMAC_MODE_ACPI_RCVD | - BNX2_EMAC_MODE_FORCE_LINK | BNX2_EMAC_MODE_MPKT; REG_WR(bp, BNX2_EMAC_MODE, val); @@ -2094,7 +2482,8 @@ bnx2_set_power_state(struct bnx2 *bp, int state) wol_msg = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; } - bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT3 | wol_msg); + if (!(bp->flags & NO_WOL_FLAG)) + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT3 | wol_msg, 0); pmcsr &= ~PCI_PM_CTRL_STATE_MASK; if ((CHIP_ID(bp) == CHIP_ID_5706_A0) || @@ -2380,21 +2769,27 @@ bnx2_init_nvram(struct bnx2 *bp) /* Flash interface has been reconfigured */ for (j = 0, flash = &flash_table[0]; j < entry_count; - j++, flash++) { - - if (val == flash->config1) { + j++, flash++) { + if ((val & FLASH_BACKUP_STRAP_MASK) == + (flash->config1 & FLASH_BACKUP_STRAP_MASK)) { bp->flash_info = flash; break; } } } else { + u32 mask; /* Not yet been reconfigured */ + if (val & (1 << 23)) + mask = FLASH_BACKUP_STRAP_MASK; + else + mask = FLASH_STRAP_MASK; + for (j = 0, flash = &flash_table[0]; j < entry_count; j++, flash++) { - if ((val & FLASH_STRAP_MASK) == flash->strapping) { + if ((val & mask) == (flash->strapping & mask)) { bp->flash_info = flash; /* Request access to the flash interface. */ @@ -2421,10 +2816,17 @@ bnx2_init_nvram(struct bnx2 *bp) if (j == entry_count) { bp->flash_info = NULL; - printk(KERN_ALERT "Unknown flash/EEPROM type.\n"); - rc = -ENODEV; + printk(KERN_ALERT PFX "Unknown flash/EEPROM type.\n"); + return -ENODEV; } + val = REG_RD_IND(bp, bp->shmem_base + BNX2_SHARED_HW_CFG_CONFIG2); + val &= BNX2_SHARED_HW_CFG2_NVM_SIZE_MASK; + if (val) + bp->flash_size = val; + else + bp->flash_size = bp->flash_info->total_size; + return rc; } @@ -2543,7 +2945,7 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf, int buf_size) { u32 written, offset32, len32; - u8 *buf, start[4], end[4]; + u8 *buf, start[4], end[4], *flash_buffer = NULL; int rc = 0; int align_start, align_end; @@ -2583,12 +2985,19 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf, memcpy(buf + align_start, data_buf, buf_size); } + if (bp->flash_info->buffered == 0) { + flash_buffer = kmalloc(264, GFP_KERNEL); + if (flash_buffer == NULL) { + rc = -ENOMEM; + goto nvram_write_end; + } + } + written = 0; while ((written < len32) && (rc == 0)) { u32 page_start, page_end, data_start, data_end; u32 addr, cmd_flags; int i; - u8 flash_buffer[264]; /* Find the page_start addr */ page_start = offset32 + written; @@ -2659,7 +3068,7 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf, } /* Loop to write the new data from data_start to data_end */ - for (addr = data_start; addr < data_end; addr += 4, i++) { + for (addr = data_start; addr < data_end; addr += 4, i += 4) { if ((addr == page_end - 4) || ((bp->flash_info->buffered) && (addr == data_end - 4))) { @@ -2707,6 +3116,9 @@ bnx2_nvram_write(struct bnx2 *bp, u32 offset, u8 *data_buf, } nvram_write_end: + if (bp->flash_info->buffered == 0) + kfree(flash_buffer); + if (align_start || align_end) kfree(buf); return rc; @@ -2728,16 +3140,14 @@ bnx2_reset_chip(struct bnx2 *bp, u32 reset_code) val = REG_RD(bp, BNX2_MISC_ENABLE_CLR_BITS); udelay(5); + /* Wait for the firmware to tell us it is ok to issue a reset. */ + bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT0 | reset_code, 1); + /* Deposit a driver reset signature so the firmware knows that * this is a soft reset. */ - REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_RESET_SIGNATURE, + REG_WR_IND(bp, bp->shmem_base + BNX2_DRV_RESET_SIGNATURE, BNX2_DRV_RESET_SIGNATURE_MAGIC); - bp->fw_timed_out = 0; - - /* Wait for the firmware to tell us it is ok to issue a reset. */ - bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT0 | reset_code); - /* Do a dummy read to force the chip to complete all current transaction * before we issue a reset. */ val = REG_RD(bp, BNX2_MISC_ID); @@ -2776,10 +3186,10 @@ bnx2_reset_chip(struct bnx2 *bp, u32 reset_code) return -ENODEV; } - bp->fw_timed_out = 0; - /* Wait for the firmware to finish its initialization. */ - bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT1 | reset_code); + rc = bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT1 | reset_code, 0); + if (rc) + return rc; if (CHIP_ID(bp) == CHIP_ID_5706_A0) { /* Adjust the voltage regular to two steps lower. The default @@ -2797,6 +3207,7 @@ static int bnx2_init_chip(struct bnx2 *bp) { u32 val; + int rc; /* Make sure the interrupt is not active. */ REG_WR(bp, BNX2_PCICFG_INT_ACK_CMD, BNX2_PCICFG_INT_ACK_CMD_MASK_INT); @@ -2812,7 +3223,7 @@ bnx2_init_chip(struct bnx2 *bp) val |= (0x2 << 20) | (1 << 11); - if ((bp->flags & PCIX_FLAG) && (bp->bus_speed_mhz = 133)) + if ((bp->flags & PCIX_FLAG) && (bp->bus_speed_mhz == 133)) val |= (1 << 23); if ((CHIP_NUM(bp) == CHIP_NUM_5706) && @@ -2932,17 +3343,24 @@ bnx2_init_chip(struct bnx2 *bp) REG_WR(bp, BNX2_HC_ATTN_BITS_ENABLE, STATUS_ATTN_BITS_LINK_STATE); + if (REG_RD_IND(bp, bp->shmem_base + BNX2_PORT_FEATURE) & + BNX2_PORT_FEATURE_ASF_ENABLED) + bp->flags |= ASF_ENABLE_FLAG; + /* Initialize the receive filter. */ bnx2_set_rx_mode(bp->dev); - bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT2 | BNX2_DRV_MSG_CODE_RESET); + rc = bnx2_fw_sync(bp, BNX2_DRV_MSG_DATA_WAIT2 | BNX2_DRV_MSG_CODE_RESET, + 0); REG_WR(bp, BNX2_MISC_ENABLE_SET_BITS, 0x5ffffff); REG_RD(bp, BNX2_MISC_ENABLE_SET_BITS); udelay(20); - return 0; + bp->hc_cmd = REG_RD(bp, BNX2_HC_COMMAND); + + return rc; } @@ -2959,8 +3377,8 @@ bnx2_init_tx_ring(struct bnx2 *bp) bp->tx_prod = 0; bp->tx_cons = 0; + bp->hw_tx_cons = 0; bp->tx_prod_bseq = 0; - atomic_set(&bp->tx_avail_bd, bp->tx_ring_size); val = BNX2_L2CTX_TYPE_TYPE_L2; val |= BNX2_L2CTX_TYPE_SIZE_L2; @@ -2992,29 +3410,38 @@ bnx2_init_rx_ring(struct bnx2 *bp) ring_prod = prod = bp->rx_prod = 0; bp->rx_cons = 0; + bp->hw_rx_cons = 0; bp->rx_prod_bseq = 0; - rxbd = &bp->rx_desc_ring[0]; - for (i = 0; i < MAX_RX_DESC_CNT; i++, rxbd++) { - rxbd->rx_bd_len = bp->rx_buf_use_size; - rxbd->rx_bd_flags = RX_BD_FLAGS_START | RX_BD_FLAGS_END; - } + for (i = 0; i < bp->rx_max_ring; i++) { + int j; - rxbd->rx_bd_haddr_hi = (u64) bp->rx_desc_mapping >> 32; - rxbd->rx_bd_haddr_lo = (u64) bp->rx_desc_mapping & 0xffffffff; + rxbd = &bp->rx_desc_ring[i][0]; + for (j = 0; j < MAX_RX_DESC_CNT; j++, rxbd++) { + rxbd->rx_bd_len = bp->rx_buf_use_size; + rxbd->rx_bd_flags = RX_BD_FLAGS_START | RX_BD_FLAGS_END; + } + if (i == (bp->rx_max_ring - 1)) + j = 0; + else + j = i + 1; + rxbd->rx_bd_haddr_hi = (u64) bp->rx_desc_mapping[j] >> 32; + rxbd->rx_bd_haddr_lo = (u64) bp->rx_desc_mapping[j] & + 0xffffffff; + } val = BNX2_L2CTX_CTX_TYPE_CTX_BD_CHN_TYPE_VALUE; val |= BNX2_L2CTX_CTX_TYPE_SIZE_L2; val |= 0x02 << 8; CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_CTX_TYPE, val); - val = (u64) bp->rx_desc_mapping >> 32; + val = (u64) bp->rx_desc_mapping[0] >> 32; CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_NX_BDHADDR_HI, val); - val = (u64) bp->rx_desc_mapping & 0xffffffff; + val = (u64) bp->rx_desc_mapping[0] & 0xffffffff; CTX_WR(bp, GET_CID_ADDR(RX_CID), BNX2_L2CTX_NX_BDHADDR_LO, val); - for ( ;ring_prod < bp->rx_ring_size; ) { + for (i = 0; i < bp->rx_ring_size; i++) { if (bnx2_alloc_rx_skb(bp, ring_prod) < 0) { break; } @@ -3028,6 +3455,29 @@ bnx2_init_rx_ring(struct bnx2 *bp) REG_WR(bp, MB_RX_CID_ADDR + BNX2_L2CTX_HOST_BSEQ, bp->rx_prod_bseq); } +static void +bnx2_set_rx_ring_size(struct bnx2 *bp, u32 size) +{ + u32 num_rings, max; + + bp->rx_ring_size = size; + num_rings = 1; + while (size > MAX_RX_DESC_CNT) { + size -= MAX_RX_DESC_CNT; + num_rings++; + } + /* round to next power of 2 */ + max = MAX_RX_RINGS; + while ((max & num_rings) == 0) + max >>= 1; + + if (num_rings != max) + max <<= 1; + + bp->rx_max_ring = max; + bp->rx_max_ring_idx = (bp->rx_max_ring * RX_DESC_CNT) - 1; +} + static void bnx2_free_tx_skbs(struct bnx2 *bp) { @@ -3073,11 +3523,11 @@ bnx2_free_rx_skbs(struct bnx2 *bp) if (bp->rx_buf_ring == NULL) return; - for (i = 0; i < RX_DESC_CNT; i++) { + for (i = 0; i < bp->rx_max_ring_idx; i++) { struct sw_bd *rx_buf = &bp->rx_buf_ring[i]; struct sk_buff *skb = rx_buf->skb; - if (skb == 0) + if (skb == NULL) continue; pci_unmap_single(bp->pdev, pci_unmap_addr(rx_buf, mapping), @@ -3130,7 +3580,7 @@ bnx2_test_registers(struct bnx2 *bp) { int ret; int i; - static struct { + static const struct { u16 offset; u16 flags; u32 rw_mask; @@ -3160,190 +3610,23 @@ bnx2_test_registers(struct bnx2 *bp) { 0x0c00, 0, 0x00000000, 0x00000001 }, { 0x0c04, 0, 0x00000000, 0x03ff0001 }, { 0x0c08, 0, 0x0f0ff073, 0x00000000 }, - { 0x0c0c, 0, 0x00ffffff, 0x00000000 }, - { 0x0c30, 0, 0x00000000, 0xffffffff }, - { 0x0c34, 0, 0x00000000, 0xffffffff }, - { 0x0c38, 0, 0x00000000, 0xffffffff }, - { 0x0c3c, 0, 0x00000000, 0xffffffff }, - { 0x0c40, 0, 0x00000000, 0xffffffff }, - { 0x0c44, 0, 0x00000000, 0xffffffff }, - { 0x0c48, 0, 0x00000000, 0x0007ffff }, - { 0x0c4c, 0, 0x00000000, 0xffffffff }, - { 0x0c50, 0, 0x00000000, 0xffffffff }, - { 0x0c54, 0, 0x00000000, 0xffffffff }, - { 0x0c58, 0, 0x00000000, 0xffffffff }, - { 0x0c5c, 0, 0x00000000, 0xffffffff }, - { 0x0c60, 0, 0x00000000, 0xffffffff }, - { 0x0c64, 0, 0x00000000, 0xffffffff }, - { 0x0c68, 0, 0x00000000, 0xffffffff }, - { 0x0c6c, 0, 0x00000000, 0xffffffff }, - { 0x0c70, 0, 0x00000000, 0xffffffff }, - { 0x0c74, 0, 0x00000000, 0xffffffff }, - { 0x0c78, 0, 0x00000000, 0xffffffff }, - { 0x0c7c, 0, 0x00000000, 0xffffffff }, - { 0x0c80, 0, 0x00000000, 0xffffffff }, - { 0x0c84, 0, 0x00000000, 0xffffffff }, - { 0x0c88, 0, 0x00000000, 0xffffffff }, - { 0x0c8c, 0, 0x00000000, 0xffffffff }, - { 0x0c90, 0, 0x00000000, 0xffffffff }, - { 0x0c94, 0, 0x00000000, 0xffffffff }, - { 0x0c98, 0, 0x00000000, 0xffffffff }, - { 0x0c9c, 0, 0x00000000, 0xffffffff }, - { 0x0ca0, 0, 0x00000000, 0xffffffff }, - { 0x0ca4, 0, 0x00000000, 0xffffffff }, - { 0x0ca8, 0, 0x00000000, 0x0007ffff }, - { 0x0cac, 0, 0x00000000, 0xffffffff }, - { 0x0cb0, 0, 0x00000000, 0xffffffff }, - { 0x0cb4, 0, 0x00000000, 0xffffffff }, - { 0x0cb8, 0, 0x00000000, 0xffffffff }, - { 0x0cbc, 0, 0x00000000, 0xffffffff }, - { 0x0cc0, 0, 0x00000000, 0xffffffff }, - { 0x0cc4, 0, 0x00000000, 0xffffffff }, - { 0x0cc8, 0, 0x00000000, 0xffffffff }, - { 0x0ccc, 0, 0x00000000, 0xffffffff }, - { 0x0cd0, 0, 0x00000000, 0xffffffff }, - { 0x0cd4, 0, 0x00000000, 0xffffffff }, - { 0x0cd8, 0, 0x00000000, 0xffffffff }, - { 0x0cdc, 0, 0x00000000, 0xffffffff }, - { 0x0ce0, 0, 0x00000000, 0xffffffff }, - { 0x0ce4, 0, 0x00000000, 0xffffffff }, - { 0x0ce8, 0, 0x00000000, 0xffffffff }, - { 0x0cec, 0, 0x00000000, 0xffffffff }, - { 0x0cf0, 0, 0x00000000, 0xffffffff }, - { 0x0cf4, 0, 0x00000000, 0xffffffff }, - { 0x0cf8, 0, 0x00000000, 0xffffffff }, - { 0x0cfc, 0, 0x00000000, 0xffffffff }, - { 0x0d00, 0, 0x00000000, 0xffffffff }, - { 0x0d04, 0, 0x00000000, 0xffffffff }, { 0x1000, 0, 0x00000000, 0x00000001 }, { 0x1004, 0, 0x00000000, 0x000f0001 }, - { 0x1044, 0, 0x00000000, 0xffc003ff }, - { 0x1080, 0, 0x00000000, 0x0001ffff }, - { 0x1084, 0, 0x00000000, 0xffffffff }, - { 0x1088, 0, 0x00000000, 0xffffffff }, - { 0x108c, 0, 0x00000000, 0xffffffff }, - { 0x1090, 0, 0x00000000, 0xffffffff }, - { 0x1094, 0, 0x00000000, 0xffffffff }, - { 0x1098, 0, 0x00000000, 0xffffffff }, - { 0x109c, 0, 0x00000000, 0xffffffff }, - { 0x10a0, 0, 0x00000000, 0xffffffff }, { 0x1408, 0, 0x01c00800, 0x00000000 }, { 0x149c, 0, 0x8000ffff, 0x00000000 }, { 0x14a8, 0, 0x00000000, 0x000001ff }, - { 0x14ac, 0, 0x4fffffff, 0x10000000 }, + { 0x14ac, 0, 0x0fffffff, 0x10000000 }, { 0x14b0, 0, 0x00000002, 0x00000001 }, { 0x14b8, 0, 0x00000000, 0x00000000 }, { 0x14c0, 0, 0x00000000, 0x00000009 }, { 0x14c4, 0, 0x00003fff, 0x00000000 }, { 0x14cc, 0, 0x00000000, 0x00000001 }, { 0x14d0, 0, 0xffffffff, 0x00000000 }, - { 0x1500, 0, 0x00000000, 0xffffffff }, - { 0x1504, 0, 0x00000000, 0xffffffff }, - { 0x1508, 0, 0x00000000, 0xffffffff }, - { 0x150c, 0, 0x00000000, 0xffffffff }, - { 0x1510, 0, 0x00000000, 0xffffffff }, - { 0x1514, 0, 0x00000000, 0xffffffff }, - { 0x1518, 0, 0x00000000, 0xffffffff }, - { 0x151c, 0, 0x00000000, 0xffffffff }, - { 0x1520, 0, 0x00000000, 0xffffffff }, - { 0x1524, 0, 0x00000000, 0xffffffff }, - { 0x1528, 0, 0x00000000, 0xffffffff }, - { 0x152c, 0, 0x00000000, 0xffffffff }, - { 0x1530, 0, 0x00000000, 0xffffffff }, - { 0x1534, 0, 0x00000000, 0xffffffff }, - { 0x1538, 0, 0x00000000, 0xffffffff }, - { 0x153c, 0, 0x00000000, 0xffffffff }, - { 0x1540, 0, 0x00000000, 0xffffffff }, - { 0x1544, 0, 0x00000000, 0xffffffff }, - { 0x1548, 0, 0x00000000, 0xffffffff }, - { 0x154c, 0, 0x00000000, 0xffffffff }, - { 0x1550, 0, 0x00000000, 0xffffffff }, - { 0x1554, 0, 0x00000000, 0xffffffff }, - { 0x1558, 0, 0x00000000, 0xffffffff }, - { 0x1600, 0, 0x00000000, 0xffffffff }, - { 0x1604, 0, 0x00000000, 0xffffffff }, - { 0x1608, 0, 0x00000000, 0xffffffff }, - { 0x160c, 0, 0x00000000, 0xffffffff }, - { 0x1610, 0, 0x00000000, 0xffffffff }, - { 0x1614, 0, 0x00000000, 0xffffffff }, - { 0x1618, 0, 0x00000000, 0xffffffff }, - { 0x161c, 0, 0x00000000, 0xffffffff }, - { 0x1620, 0, 0x00000000, 0xffffffff }, - { 0x1624, 0, 0x00000000, 0xffffffff }, - { 0x1628, 0, 0x00000000, 0xffffffff }, - { 0x162c, 0, 0x00000000, 0xffffffff }, - { 0x1630, 0, 0x00000000, 0xffffffff }, - { 0x1634, 0, 0x00000000, 0xffffffff }, - { 0x1638, 0, 0x00000000, 0xffffffff }, - { 0x163c, 0, 0x00000000, 0xffffffff }, - { 0x1640, 0, 0x00000000, 0xffffffff }, - { 0x1644, 0, 0x00000000, 0xffffffff }, - { 0x1648, 0, 0x00000000, 0xffffffff }, - { 0x164c, 0, 0x00000000, 0xffffffff }, - { 0x1650, 0, 0x00000000, 0xffffffff }, - { 0x1654, 0, 0x00000000, 0xffffffff }, { 0x1800, 0, 0x00000000, 0x00000001 }, { 0x1804, 0, 0x00000000, 0x00000003 }, - { 0x1840, 0, 0x00000000, 0xffffffff }, - { 0x1844, 0, 0x00000000, 0xffffffff }, - { 0x1848, 0, 0x00000000, 0xffffffff }, - { 0x184c, 0, 0x00000000, 0xffffffff }, - { 0x1850, 0, 0x00000000, 0xffffffff }, - { 0x1900, 0, 0x7ffbffff, 0x00000000 }, - { 0x1904, 0, 0xffffffff, 0x00000000 }, - { 0x190c, 0, 0xffffffff, 0x00000000 }, - { 0x1914, 0, 0xffffffff, 0x00000000 }, - { 0x191c, 0, 0xffffffff, 0x00000000 }, - { 0x1924, 0, 0xffffffff, 0x00000000 }, - { 0x192c, 0, 0xffffffff, 0x00000000 }, - { 0x1934, 0, 0xffffffff, 0x00000000 }, - { 0x193c, 0, 0xffffffff, 0x00000000 }, - { 0x1944, 0, 0xffffffff, 0x00000000 }, - { 0x194c, 0, 0xffffffff, 0x00000000 }, - { 0x1954, 0, 0xffffffff, 0x00000000 }, - { 0x195c, 0, 0xffffffff, 0x00000000 }, - { 0x1964, 0, 0xffffffff, 0x00000000 }, - { 0x196c, 0, 0xffffffff, 0x00000000 }, - { 0x1974, 0, 0xffffffff, 0x00000000 }, - { 0x197c, 0, 0xffffffff, 0x00000000 }, - { 0x1980, 0, 0x0700ffff, 0x00000000 }, - - { 0x1c00, 0, 0x00000000, 0x00000001 }, - { 0x1c04, 0, 0x00000000, 0x00000003 }, - { 0x1c08, 0, 0x0000000f, 0x00000000 }, - { 0x1c40, 0, 0x00000000, 0xffffffff }, - { 0x1c44, 0, 0x00000000, 0xffffffff }, - { 0x1c48, 0, 0x00000000, 0xffffffff }, - { 0x1c4c, 0, 0x00000000, 0xffffffff }, - { 0x1c50, 0, 0x00000000, 0xffffffff }, - { 0x1d00, 0, 0x7ffbffff, 0x00000000 }, - { 0x1d04, 0, 0xffffffff, 0x00000000 }, - { 0x1d0c, 0, 0xffffffff, 0x00000000 }, - { 0x1d14, 0, 0xffffffff, 0x00000000 }, - { 0x1d1c, 0, 0xffffffff, 0x00000000 }, - { 0x1d24, 0, 0xffffffff, 0x00000000 }, - { 0x1d2c, 0, 0xffffffff, 0x00000000 }, - { 0x1d34, 0, 0xffffffff, 0x00000000 }, - { 0x1d3c, 0, 0xffffffff, 0x00000000 }, - { 0x1d44, 0, 0xffffffff, 0x00000000 }, - { 0x1d4c, 0, 0xffffffff, 0x00000000 }, - { 0x1d54, 0, 0xffffffff, 0x00000000 }, - { 0x1d5c, 0, 0xffffffff, 0x00000000 }, - { 0x1d64, 0, 0xffffffff, 0x00000000 }, - { 0x1d6c, 0, 0xffffffff, 0x00000000 }, - { 0x1d74, 0, 0xffffffff, 0x00000000 }, - { 0x1d7c, 0, 0xffffffff, 0x00000000 }, - { 0x1d80, 0, 0x0700ffff, 0x00000000 }, - - { 0x2004, 0, 0x00000000, 0x0337000f }, - { 0x2008, 0, 0xffffffff, 0x00000000 }, - { 0x200c, 0, 0xffffffff, 0x00000000 }, - { 0x2010, 0, 0xffffffff, 0x00000000 }, - { 0x2014, 0, 0x801fff80, 0x00000000 }, - { 0x2018, 0, 0x000003ff, 0x00000000 }, { 0x2800, 0, 0x00000000, 0x00000001 }, { 0x2804, 0, 0x00000000, 0x00003f01 }, @@ -3361,16 +3644,6 @@ bnx2_test_registers(struct bnx2 *bp) { 0x2c00, 0, 0x00000000, 0x00000011 }, { 0x2c04, 0, 0x00000000, 0x00030007 }, - { 0x3000, 0, 0x00000000, 0x00000001 }, - { 0x3004, 0, 0x00000000, 0x007007ff }, - { 0x3008, 0, 0x00000003, 0x00000000 }, - { 0x300c, 0, 0xffffffff, 0x00000000 }, - { 0x3010, 0, 0xffffffff, 0x00000000 }, - { 0x3014, 0, 0xffffffff, 0x00000000 }, - { 0x3034, 0, 0xffffffff, 0x00000000 }, - { 0x3038, 0, 0xffffffff, 0x00000000 }, - { 0x3050, 0, 0x00000001, 0x00000000 }, - { 0x3c00, 0, 0x00000000, 0x00000001 }, { 0x3c04, 0, 0x00000000, 0x00070000 }, { 0x3c08, 0, 0x00007f71, 0x07f00000 }, @@ -3380,88 +3653,11 @@ bnx2_test_registers(struct bnx2 *bp) { 0x3c18, 0, 0x00000000, 0xffffffff }, { 0x3c1c, 0, 0xfffff000, 0x00000000 }, { 0x3c20, 0, 0xffffff00, 0x00000000 }, - { 0x3c24, 0, 0xffffffff, 0x00000000 }, - { 0x3c28, 0, 0xffffffff, 0x00000000 }, - { 0x3c2c, 0, 0xffffffff, 0x00000000 }, - { 0x3c30, 0, 0xffffffff, 0x00000000 }, - { 0x3c34, 0, 0xffffffff, 0x00000000 }, - { 0x3c38, 0, 0xffffffff, 0x00000000 }, - { 0x3c3c, 0, 0xffffffff, 0x00000000 }, - { 0x3c40, 0, 0xffffffff, 0x00000000 }, - { 0x3c44, 0, 0xffffffff, 0x00000000 }, - { 0x3c48, 0, 0xffffffff, 0x00000000 }, - { 0x3c4c, 0, 0xffffffff, 0x00000000 }, - { 0x3c50, 0, 0xffffffff, 0x00000000 }, - { 0x3c54, 0, 0xffffffff, 0x00000000 }, - { 0x3c58, 0, 0xffffffff, 0x00000000 }, - { 0x3c5c, 0, 0xffffffff, 0x00000000 }, - { 0x3c60, 0, 0xffffffff, 0x00000000 }, - { 0x3c64, 0, 0xffffffff, 0x00000000 }, - { 0x3c68, 0, 0xffffffff, 0x00000000 }, - { 0x3c6c, 0, 0xffffffff, 0x00000000 }, - { 0x3c70, 0, 0xffffffff, 0x00000000 }, - { 0x3c74, 0, 0x0000003f, 0x00000000 }, - { 0x3c78, 0, 0x00000000, 0x00000000 }, - { 0x3c7c, 0, 0x00000000, 0x00000000 }, - { 0x3c80, 0, 0x3fffffff, 0x00000000 }, - { 0x3c84, 0, 0x0000003f, 0x00000000 }, - { 0x3c88, 0, 0x00000000, 0xffffffff }, - { 0x3c8c, 0, 0x00000000, 0xffffffff }, - - { 0x4000, 0, 0x00000000, 0x00000001 }, - { 0x4004, 0, 0x00000000, 0x00030000 }, - { 0x4008, 0, 0x00000ff0, 0x00000000 }, - { 0x400c, 0, 0xffffffff, 0x00000000 }, - { 0x4088, 0, 0x00000000, 0x00070303 }, - - { 0x4400, 0, 0x00000000, 0x00000001 }, - { 0x4404, 0, 0x00000000, 0x00003f01 }, - { 0x4408, 0, 0x7fff00ff, 0x00000000 }, - { 0x440c, 0, 0xffffffff, 0x00000000 }, - { 0x4410, 0, 0xffff, 0x0000 }, - { 0x4414, 0, 0xffff, 0x0000 }, - { 0x4418, 0, 0xffff, 0x0000 }, - { 0x441c, 0, 0xffff, 0x0000 }, - { 0x4428, 0, 0xffffffff, 0x00000000 }, - { 0x442c, 0, 0xffffffff, 0x00000000 }, - { 0x4430, 0, 0xffffffff, 0x00000000 }, - { 0x4434, 0, 0xffffffff, 0x00000000 }, - { 0x4438, 0, 0xffffffff, 0x00000000 }, - { 0x443c, 0, 0xffffffff, 0x00000000 }, - { 0x4440, 0, 0xffffffff, 0x00000000 }, - { 0x4444, 0, 0xffffffff, 0x00000000 }, - - { 0x4c00, 0, 0x00000000, 0x00000001 }, - { 0x4c04, 0, 0x00000000, 0x0000003f }, - { 0x4c08, 0, 0xffffffff, 0x00000000 }, - { 0x4c0c, 0, 0x0007fc00, 0x00000000 }, - { 0x4c10, 0, 0x80003fe0, 0x00000000 }, - { 0x4c14, 0, 0xffffffff, 0x00000000 }, - { 0x4c44, 0, 0x00000000, 0x9fff9fff }, - { 0x4c48, 0, 0x00000000, 0xb3009fff }, - { 0x4c4c, 0, 0x00000000, 0x77f33b30 }, - { 0x4c50, 0, 0x00000000, 0xffffffff }, { 0x5004, 0, 0x00000000, 0x0000007f }, { 0x5008, 0, 0x0f0007ff, 0x00000000 }, { 0x500c, 0, 0xf800f800, 0x07ff07ff }, - { 0x5400, 0, 0x00000008, 0x00000001 }, - { 0x5404, 0, 0x00000000, 0x0000003f }, - { 0x5408, 0, 0x0000001f, 0x00000000 }, - { 0x540c, 0, 0xffffffff, 0x00000000 }, - { 0x5410, 0, 0xffffffff, 0x00000000 }, - { 0x5414, 0, 0x0000ffff, 0x00000000 }, - { 0x5418, 0, 0x0000ffff, 0x00000000 }, - { 0x541c, 0, 0x0000ffff, 0x00000000 }, - { 0x5420, 0, 0x0000ffff, 0x00000000 }, - { 0x5428, 0, 0x000000ff, 0x00000000 }, - { 0x542c, 0, 0xff00ffff, 0x00000000 }, - { 0x5430, 0, 0x001fff80, 0x00000000 }, - { 0x5438, 0, 0xffffffff, 0x00000000 }, - { 0x543c, 0, 0xffffffff, 0x00000000 }, - { 0x5440, 0, 0xf800f800, 0x07ff07ff }, - { 0x5c00, 0, 0x00000000, 0x00000001 }, { 0x5c04, 0, 0x00000000, 0x0003000f }, { 0x5c08, 0, 0x00000003, 0x00000000 }, @@ -3507,11 +3703,11 @@ bnx2_test_registers(struct bnx2 *bp) rw_mask = reg_tbl[i].rw_mask; ro_mask = reg_tbl[i].ro_mask; - save_val = readl((u8 *) bp->regview + offset); + save_val = readl(bp->regview + offset); - writel(0, (u8 *) bp->regview + offset); + writel(0, bp->regview + offset); - val = readl((u8 *) bp->regview + offset); + val = readl(bp->regview + offset); if ((val & rw_mask) != 0) { goto reg_test_err; } @@ -3520,9 +3716,9 @@ bnx2_test_registers(struct bnx2 *bp) goto reg_test_err; } - writel(0xffffffff, (u8 *) bp->regview + offset); + writel(0xffffffff, bp->regview + offset); - val = readl((u8 *) bp->regview + offset); + val = readl(bp->regview + offset); if ((val & rw_mask) != rw_mask) { goto reg_test_err; } @@ -3531,11 +3727,11 @@ bnx2_test_registers(struct bnx2 *bp) goto reg_test_err; } - writel(save_val, (u8 *) bp->regview + offset); + writel(save_val, bp->regview + offset); continue; reg_test_err: - writel(save_val, (u8 *) bp->regview + offset); + writel(save_val, bp->regview + offset); ret = -ENODEV; break; } @@ -3545,7 +3741,7 @@ reg_test_err: static int bnx2_do_mem_test(struct bnx2 *bp, u32 start, u32 size) { - static u32 test_pattern[] = { 0x00000000, 0xffffffff, 0x55555555, + static const u32 test_pattern[] = { 0x00000000, 0xffffffff, 0x55555555, 0xaaaaaaaa , 0xaa55aa55, 0x55aa55aa }; int i; @@ -3570,12 +3766,12 @@ bnx2_test_memory(struct bnx2 *bp) { int ret = 0; int i; - static struct { + static const struct { u32 offset; u32 len; } mem_tbl[] = { { 0x60000, 0x4000 }, - { 0xa0000, 0x4000 }, + { 0xa0000, 0x3000 }, { 0xe0000, 0x4000 }, { 0x120000, 0x4000 }, { 0x1a0000, 0x4000 }, @@ -3593,29 +3789,37 @@ bnx2_test_memory(struct bnx2 *bp) return ret; } +#define BNX2_MAC_LOOPBACK 0 +#define BNX2_PHY_LOOPBACK 1 + static int -bnx2_test_loopback(struct bnx2 *bp) +bnx2_run_loopback(struct bnx2 *bp, int loopback_mode) { unsigned int pkt_size, num_pkts, i; struct sk_buff *skb, *rx_skb; unsigned char *packet; - u16 rx_start_idx, rx_idx, send_idx; - u32 send_bseq, val; + u16 rx_start_idx, rx_idx; dma_addr_t map; struct tx_bd *txbd; struct sw_bd *rx_buf; struct l2_fhdr *rx_hdr; int ret = -ENODEV; - if (!netif_running(bp->dev)) - return -ENODEV; - - bp->loopback = MAC_LOOPBACK; - bnx2_reset_nic(bp, BNX2_DRV_MSG_CODE_DIAG); - bnx2_set_mac_loopback(bp); + if (loopback_mode == BNX2_MAC_LOOPBACK) { + bp->loopback = MAC_LOOPBACK; + bnx2_set_mac_loopback(bp); + } + else if (loopback_mode == BNX2_PHY_LOOPBACK) { + bp->loopback = 0; + bnx2_set_phy_loopback(bp); + } + else + return -EINVAL; pkt_size = 1514; skb = dev_alloc_skb(pkt_size); + if (!skb) + return -ENOMEM; packet = skb_put(skb, pkt_size); memcpy(packet, bp->mac_addr, 6); memset(packet + 6, 0x0, 8); @@ -3625,18 +3829,17 @@ bnx2_test_loopback(struct bnx2 *bp) map = pci_map_single(bp->pdev, skb->data, pkt_size, PCI_DMA_TODEVICE); - val = REG_RD(bp, BNX2_HC_COMMAND); - REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_WR(bp, BNX2_HC_COMMAND, + bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_RD(bp, BNX2_HC_COMMAND); udelay(5); rx_start_idx = bp->status_blk->status_rx_quick_consumer_index0; - send_idx = 0; - send_bseq = 0; num_pkts = 0; - txbd = &bp->tx_desc_ring[send_idx]; + txbd = &bp->tx_desc_ring[TX_RING_IDX(bp->tx_prod)]; txbd->tx_bd_haddr_hi = (u64) map >> 32; txbd->tx_bd_haddr_lo = (u64) map & 0xffffffff; @@ -3644,18 +3847,17 @@ bnx2_test_loopback(struct bnx2 *bp) txbd->tx_bd_vlan_tag_flags = TX_BD_FLAGS_START | TX_BD_FLAGS_END; num_pkts++; - send_idx = NEXT_TX_BD(send_idx); - - send_bseq += pkt_size; - - REG_WR16(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BIDX, send_idx); - REG_WR(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BSEQ, send_bseq); + bp->tx_prod = NEXT_TX_BD(bp->tx_prod); + bp->tx_prod_bseq += pkt_size; + REG_WR16(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BIDX, bp->tx_prod); + REG_WR(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BSEQ, bp->tx_prod_bseq); udelay(100); - val = REG_RD(bp, BNX2_HC_COMMAND); - REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_WR(bp, BNX2_HC_COMMAND, + bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW_WO_INT); + REG_RD(bp, BNX2_HC_COMMAND); udelay(5); @@ -3663,7 +3865,7 @@ bnx2_test_loopback(struct bnx2 *bp) pci_unmap_single(bp->pdev, map, pkt_size, PCI_DMA_TODEVICE); dev_kfree_skb_irq(skb); - if (bp->status_blk->status_tx_quick_consumer_index0 != send_idx) { + if (bp->status_blk->status_tx_quick_consumer_index0 != bp->tx_prod) { goto loopback_test_done; } @@ -3682,7 +3884,7 @@ bnx2_test_loopback(struct bnx2 *bp) pci_unmap_addr(rx_buf, mapping), bp->rx_buf_size, PCI_DMA_FROMDEVICE); - if (rx_hdr->l2_fhdr_errors & + if (rx_hdr->l2_fhdr_status & (L2_FHDR_ERRORS_BAD_CRC | L2_FHDR_ERRORS_PHY_DECODE | L2_FHDR_ERRORS_ALIGNMENT | @@ -3709,6 +3911,30 @@ loopback_test_done: return ret; } +#define BNX2_MAC_LOOPBACK_FAILED 1 +#define BNX2_PHY_LOOPBACK_FAILED 2 +#define BNX2_LOOPBACK_FAILED (BNX2_MAC_LOOPBACK_FAILED | \ + BNX2_PHY_LOOPBACK_FAILED) + +static int +bnx2_test_loopback(struct bnx2 *bp) +{ + int rc = 0; + + if (!netif_running(bp->dev)) + return BNX2_LOOPBACK_FAILED; + + bnx2_reset_nic(bp, BNX2_DRV_MSG_CODE_RESET); + spin_lock_bh(&bp->phy_lock); + bnx2_init_phy(bp); + spin_unlock_bh(&bp->phy_lock); + if (bnx2_run_loopback(bp, BNX2_MAC_LOOPBACK)) + rc |= BNX2_MAC_LOOPBACK_FAILED; + if (bnx2_run_loopback(bp, BNX2_PHY_LOOPBACK)) + rc |= BNX2_PHY_LOOPBACK_FAILED; + return rc; +} + #define NVRAM_SIZE 0x200 #define CRC32_RESIDUAL 0xdebb20e3 @@ -3752,10 +3978,10 @@ bnx2_test_link(struct bnx2 *bp) { u32 bmsr; - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); bnx2_read_phy(bp, MII_BMSR, &bmsr); bnx2_read_phy(bp, MII_BMSR, &bmsr); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); if (bmsr & BMSR_LSTATUS) { return 0; @@ -3767,7 +3993,6 @@ static int bnx2_test_intr(struct bnx2 *bp) { int i; - u32 val; u16 status_idx; if (!netif_running(bp->dev)) @@ -3776,8 +4001,7 @@ bnx2_test_intr(struct bnx2 *bp) status_idx = REG_RD(bp, BNX2_PCICFG_INT_ACK_CMD) & 0xffff; /* This register is not touched during run-time. */ - val = REG_RD(bp, BNX2_HC_COMMAND); - REG_WR(bp, BNX2_HC_COMMAND, val | BNX2_HC_COMMAND_COAL_NOW); + REG_WR(bp, BNX2_HC_COMMAND, bp->hc_cmd | BNX2_HC_COMMAND_COAL_NOW); REG_RD(bp, BNX2_HC_COMMAND); for (i = 0; i < 10; i++) { @@ -3801,23 +4025,27 @@ bnx2_timer(unsigned long data) struct bnx2 *bp = (struct bnx2 *) data; u32 msg; + if (!netif_running(bp->dev)) + return; + if (atomic_read(&bp->intr_sem) != 0) goto bnx2_restart_timer; msg = (u32) ++bp->fw_drv_pulse_wr_seq; - REG_WR_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DRV_PULSE_MB, msg); + REG_WR_IND(bp, bp->shmem_base + BNX2_DRV_PULSE_MB, msg); if ((bp->phy_flags & PHY_SERDES_FLAG) && (CHIP_NUM(bp) == CHIP_NUM_5706)) { - unsigned long flags; - spin_lock_irqsave(&bp->phy_lock, flags); + spin_lock(&bp->phy_lock); if (bp->serdes_an_pending) { bp->serdes_an_pending--; } else if ((bp->link_up == 0) && (bp->autoneg & AUTONEG_SPEED)) { u32 bmcr; + bp->current_interval = bp->timer_interval; + bnx2_read_phy(bp, MII_BMCR, &bmcr); if (bmcr & BMCR_ANENABLE) { @@ -3860,24 +4088,24 @@ bnx2_timer(unsigned long data) } } + else + bp->current_interval = bp->timer_interval; - spin_unlock_irqrestore(&bp->phy_lock, flags); + spin_unlock(&bp->phy_lock); } bnx2_restart_timer: - bp->timer.expires = RUN_AT(bp->timer_interval); - - add_timer(&bp->timer); + mod_timer(&bp->timer, jiffies + bp->current_interval); } /* Called with rtnl_lock */ static int bnx2_open(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); int rc; - bnx2_set_power_state(bp, 0); + bnx2_set_power_state(bp, PCI_D0); bnx2_disable_int(bp); rc = bnx2_alloc_mem(bp); @@ -3920,12 +4148,7 @@ bnx2_open(struct net_device *dev) return rc; } - init_timer(&bp->timer); - - bp->timer.expires = RUN_AT(bp->timer_interval); - bp->timer.data = (unsigned long) bp; - bp->timer.function = bnx2_timer; - add_timer(&bp->timer); + mod_timer(&bp->timer, jiffies + bp->current_interval); atomic_set(&bp->intr_sem, 0); @@ -3976,18 +4199,23 @@ bnx2_reset_task(void *data) { struct bnx2 *bp = data; + if (!netif_running(bp->dev)) + return; + + bp->in_reset_task = 1; bnx2_netif_stop(bp); bnx2_init_nic(bp); atomic_set(&bp->intr_sem, 1); bnx2_netif_start(bp); + bp->in_reset_task = 0; } static void bnx2_tx_timeout(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); /* This allows the netif to be shutdown gracefully before resetting */ schedule_work(&bp->reset_task); @@ -3998,7 +4226,7 @@ bnx2_tx_timeout(struct net_device *dev) static void bnx2_vlan_rx_register(struct net_device *dev, struct vlan_group *vlgrp) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); bnx2_netif_stop(bp); @@ -4012,7 +4240,7 @@ bnx2_vlan_rx_register(struct net_device *dev, struct vlan_group *vlgrp) static void bnx2_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); bnx2_netif_stop(bp); @@ -4024,7 +4252,7 @@ bnx2_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid) } #endif -/* Called with dev->xmit_lock. +/* Called with netif_tx_lock. * hard_start_xmit is pseudo-lockless - a lock is only required when * the tx queue is full. This way, we get the benefit of lockless * operations most of the time without the complexities to handle @@ -4033,7 +4261,7 @@ bnx2_vlan_rx_kill_vid(struct net_device *dev, uint16_t vid) static int bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); dma_addr_t mapping; struct tx_bd *txbd; struct sw_bd *tx_buf; @@ -4041,9 +4269,7 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) u16 prod, ring_prod; int i; - if (unlikely(atomic_read(&bp->tx_avail_bd) < - (skb_shinfo(skb)->nr_frags + 1))) { - + if (unlikely(bnx2_tx_avail(bp) < (skb_shinfo(skb)->nr_frags + 1))) { netif_stop_queue(dev); printk(KERN_ERR PFX "%s: BUG! Tx ring full when queue awake!\n", dev->name); @@ -4064,7 +4290,7 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) (TX_BD_FLAGS_VLAN_TAG | (vlan_tx_tag_get(skb) << 16)); } #ifdef BCM_TSO - if ((mss = skb_shinfo(skb)->tso_size) && + if ((mss = skb_shinfo(skb)->gso_size) && (skb->len > (bp->dev->mtu + ETH_HLEN))) { u32 tcp_opt_len, ip_tcp_len; @@ -4140,8 +4366,6 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) prod = NEXT_TX_BD(prod); bp->tx_prod_bseq += skb->len; - atomic_sub(last_frag + 1, &bp->tx_avail_bd); - REG_WR16(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BIDX, prod); REG_WR(bp, MB_TX_CID_ADDR + BNX2_L2CTX_TX_HOST_BSEQ, bp->tx_prod_bseq); @@ -4150,17 +4374,13 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) bp->tx_prod = prod; dev->trans_start = jiffies; - if (unlikely(atomic_read(&bp->tx_avail_bd) <= MAX_SKB_FRAGS)) { - unsigned long flags; - - spin_lock_irqsave(&bp->tx_lock, flags); - if (atomic_read(&bp->tx_avail_bd) <= MAX_SKB_FRAGS) { - netif_stop_queue(dev); - - if (atomic_read(&bp->tx_avail_bd) > MAX_SKB_FRAGS) - netif_wake_queue(dev); - } - spin_unlock_irqrestore(&bp->tx_lock, flags); + if (unlikely(bnx2_tx_avail(bp) <= MAX_SKB_FRAGS)) { + spin_lock(&bp->tx_lock); + netif_stop_queue(dev); + + if (bnx2_tx_avail(bp) > MAX_SKB_FRAGS) + netif_wake_queue(dev); + spin_unlock(&bp->tx_lock); } return NETDEV_TX_OK; @@ -4170,13 +4390,21 @@ bnx2_start_xmit(struct sk_buff *skb, struct net_device *dev) static int bnx2_close(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); u32 reset_code; - flush_scheduled_work(); + /* Calling flush_scheduled_work() may deadlock because + * linkwatch_event() may be on the workqueue and it will try to get + * the rtnl_lock which we are holding. + */ + while (bp->in_reset_task) + msleep(1); + bnx2_netif_stop(bp); del_timer_sync(&bp->timer); - if (bp->wol) + if (bp->flags & NO_WOL_FLAG) + reset_code = BNX2_DRV_MSG_CODE_UNLOAD; + else if (bp->wol) reset_code = BNX2_DRV_MSG_CODE_SUSPEND_WOL; else reset_code = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; @@ -4190,7 +4418,7 @@ bnx2_close(struct net_device *dev) bnx2_free_mem(bp); bp->link_up = 0; netif_carrier_off(bp->dev); - bnx2_set_power_state(bp, 3); + bnx2_set_power_state(bp, PCI_D3hot); return 0; } @@ -4210,7 +4438,7 @@ bnx2_close(struct net_device *dev) static struct net_device_stats * bnx2_get_stats(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); struct statistics_block *stats_blk = bp->stats_blk; struct net_device_stats *net_stats = &bp->net_stats; @@ -4260,7 +4488,8 @@ bnx2_get_stats(struct net_device *dev) (unsigned long) (stats_blk->stat_Dot3StatsExcessiveCollisions + stats_blk->stat_Dot3StatsLateCollisions); - if (CHIP_NUM(bp) == CHIP_NUM_5706) + if ((CHIP_NUM(bp) == CHIP_NUM_5706) || + (CHIP_ID(bp) == CHIP_ID_5708_A0)) net_stats->tx_carrier_errors = 0; else { net_stats->tx_carrier_errors = @@ -4283,7 +4512,7 @@ bnx2_get_stats(struct net_device *dev) static int bnx2_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); cmd->supported = SUPPORTED_Autoneg; if (bp->phy_flags & PHY_SERDES_FLAG) { @@ -4330,7 +4559,7 @@ bnx2_get_settings(struct net_device *dev, struct ethtool_cmd *cmd) static int bnx2_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); u8 autoneg = bp->autoneg; u8 req_duplex = bp->req_duplex; u16 req_line_speed = bp->req_line_speed; @@ -4390,11 +4619,11 @@ bnx2_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) bp->req_line_speed = req_line_speed; bp->req_duplex = req_duplex; - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); bnx2_setup_phy(bp); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); return 0; } @@ -4402,7 +4631,7 @@ bnx2_set_settings(struct net_device *dev, struct ethtool_cmd *cmd) static void bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); strcpy(info->driver, DRV_MODULE_NAME); strcpy(info->version, DRV_MODULE_VERSION); @@ -4410,15 +4639,72 @@ bnx2_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info) info->fw_version[0] = ((bp->fw_ver & 0xff000000) >> 24) + '0'; info->fw_version[2] = ((bp->fw_ver & 0xff0000) >> 16) + '0'; info->fw_version[4] = ((bp->fw_ver & 0xff00) >> 8) + '0'; - info->fw_version[6] = (bp->fw_ver & 0xff) + '0'; - info->fw_version[1] = info->fw_version[3] = info->fw_version[5] = '.'; - info->fw_version[7] = 0; + info->fw_version[1] = info->fw_version[3] = '.'; + info->fw_version[5] = 0; +} + +#define BNX2_REGDUMP_LEN (32 * 1024) + +static int +bnx2_get_regs_len(struct net_device *dev) +{ + return BNX2_REGDUMP_LEN; +} + +static void +bnx2_get_regs(struct net_device *dev, struct ethtool_regs *regs, void *_p) +{ + u32 *p = _p, i, offset; + u8 *orig_p = _p; + struct bnx2 *bp = netdev_priv(dev); + u32 reg_boundaries[] = { 0x0000, 0x0098, 0x0400, 0x045c, + 0x0800, 0x0880, 0x0c00, 0x0c10, + 0x0c30, 0x0d08, 0x1000, 0x101c, + 0x1040, 0x1048, 0x1080, 0x10a4, + 0x1400, 0x1490, 0x1498, 0x14f0, + 0x1500, 0x155c, 0x1580, 0x15dc, + 0x1600, 0x1658, 0x1680, 0x16d8, + 0x1800, 0x1820, 0x1840, 0x1854, + 0x1880, 0x1894, 0x1900, 0x1984, + 0x1c00, 0x1c0c, 0x1c40, 0x1c54, + 0x1c80, 0x1c94, 0x1d00, 0x1d84, + 0x2000, 0x2030, 0x23c0, 0x2400, + 0x2800, 0x2820, 0x2830, 0x2850, + 0x2b40, 0x2c10, 0x2fc0, 0x3058, + 0x3c00, 0x3c94, 0x4000, 0x4010, + 0x4080, 0x4090, 0x43c0, 0x4458, + 0x4c00, 0x4c18, 0x4c40, 0x4c54, + 0x4fc0, 0x5010, 0x53c0, 0x5444, + 0x5c00, 0x5c18, 0x5c80, 0x5c90, + 0x5fc0, 0x6000, 0x6400, 0x6428, + 0x6800, 0x6848, 0x684c, 0x6860, + 0x6888, 0x6910, 0x8000 }; + + regs->version = 0; + + memset(p, 0, BNX2_REGDUMP_LEN); + + if (!netif_running(bp->dev)) + return; + + i = 0; + offset = reg_boundaries[0]; + p += offset; + while (offset < BNX2_REGDUMP_LEN) { + *p++ = REG_RD(bp, offset); + offset += 4; + if (offset == reg_boundaries[i + 1]) { + offset = reg_boundaries[i + 2]; + p = (u32 *) (orig_p + offset); + i += 2; + } + } } static void bnx2_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); if (bp->flags & NO_WOL_FLAG) { wol->supported = 0; @@ -4437,7 +4723,7 @@ bnx2_get_wol(struct net_device *dev, struct ethtool_wolinfo *wol) static int bnx2_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); if (wol->wolopts & ~WAKE_MAGIC) return -EINVAL; @@ -4457,26 +4743,27 @@ bnx2_set_wol(struct net_device *dev, struct ethtool_wolinfo *wol) static int bnx2_nway_reset(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); u32 bmcr; if (!(bp->autoneg & AUTONEG_SPEED)) { return -EINVAL; } - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); /* Force a link down visible on the other side */ if (bp->phy_flags & PHY_SERDES_FLAG) { bnx2_write_phy(bp, MII_BMCR, BMCR_LOOPBACK); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); msleep(20); - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); if (CHIP_NUM(bp) == CHIP_NUM_5706) { - bp->serdes_an_pending = SERDES_AN_TIMEOUT / - bp->timer_interval; + bp->current_interval = SERDES_AN_TIMEOUT; + bp->serdes_an_pending = 1; + mod_timer(&bp->timer, jiffies + bp->current_interval); } } @@ -4484,7 +4771,7 @@ bnx2_nway_reset(struct net_device *dev) bmcr &= ~BMCR_LOOPBACK; bnx2_write_phy(bp, MII_BMCR, bmcr | BMCR_ANRESTART | BMCR_ANENABLE); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); return 0; } @@ -4492,26 +4779,22 @@ bnx2_nway_reset(struct net_device *dev) static int bnx2_get_eeprom_len(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); - if (bp->flash_info == 0) + if (bp->flash_info == NULL) return 0; - return (int) bp->flash_info->total_size; + return (int) bp->flash_size; } static int bnx2_get_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *eebuf) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); int rc; - if (eeprom->offset > bp->flash_info->total_size) - return -EINVAL; - - if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) - eeprom->len = bp->flash_info->total_size - eeprom->offset; + /* parameters already validated in ethtool_get_eeprom */ rc = bnx2_nvram_read(bp, eeprom->offset, eebuf, eeprom->len); @@ -4522,14 +4805,10 @@ static int bnx2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, u8 *eebuf) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); int rc; - if (eeprom->offset > bp->flash_info->total_size) - return -EINVAL; - - if ((eeprom->offset + eeprom->len) > bp->flash_info->total_size) - eeprom->len = bp->flash_info->total_size - eeprom->offset; + /* parameters already validated in ethtool_set_eeprom */ rc = bnx2_nvram_write(bp, eeprom->offset, eebuf, eeprom->len); @@ -4539,7 +4818,7 @@ bnx2_set_eeprom(struct net_device *dev, struct ethtool_eeprom *eeprom, static int bnx2_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); memset(coal, 0, sizeof(struct ethtool_coalesce)); @@ -4561,7 +4840,7 @@ bnx2_get_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) static int bnx2_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); bp->rx_ticks = (u16) coal->rx_coalesce_usecs; if (bp->rx_ticks > 0x3ff) bp->rx_ticks = 0x3ff; @@ -4605,9 +4884,9 @@ bnx2_set_coalesce(struct net_device *dev, struct ethtool_coalesce *coal) static void bnx2_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); - ering->rx_max_pending = MAX_RX_DESC_CNT; + ering->rx_max_pending = MAX_TOTAL_RX_DESC_CNT; ering->rx_mini_max_pending = 0; ering->rx_jumbo_max_pending = 0; @@ -4622,19 +4901,30 @@ bnx2_get_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) static int bnx2_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); - if ((ering->rx_pending > MAX_RX_DESC_CNT) || + if ((ering->rx_pending > MAX_TOTAL_RX_DESC_CNT) || (ering->tx_pending > MAX_TX_DESC_CNT) || (ering->tx_pending <= MAX_SKB_FRAGS)) { return -EINVAL; } - bp->rx_ring_size = ering->rx_pending; + if (netif_running(bp->dev)) { + bnx2_netif_stop(bp); + bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_RESET); + bnx2_free_skbs(bp); + bnx2_free_mem(bp); + } + + bnx2_set_rx_ring_size(bp, ering->rx_pending); bp->tx_ring_size = ering->tx_pending; if (netif_running(bp->dev)) { - bnx2_netif_stop(bp); + int rc; + + rc = bnx2_alloc_mem(bp); + if (rc) + return rc; bnx2_init_nic(bp); bnx2_netif_start(bp); } @@ -4645,7 +4935,7 @@ bnx2_set_ringparam(struct net_device *dev, struct ethtool_ringparam *ering) static void bnx2_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); epause->autoneg = ((bp->autoneg & AUTONEG_FLOW_CTRL) != 0); epause->rx_pause = ((bp->flow_ctrl & FLOW_CTRL_RX) != 0); @@ -4655,7 +4945,7 @@ bnx2_get_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) static int bnx2_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); bp->req_flow_ctrl = 0; if (epause->rx_pause) @@ -4670,11 +4960,11 @@ bnx2_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) bp->autoneg &= ~AUTONEG_FLOW_CTRL; } - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); bnx2_setup_phy(bp); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); return 0; } @@ -4682,7 +4972,7 @@ bnx2_set_pauseparam(struct net_device *dev, struct ethtool_pauseparam *epause) static u32 bnx2_get_rx_csum(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); return bp->rx_csum; } @@ -4690,7 +4980,7 @@ bnx2_get_rx_csum(struct net_device *dev) static int bnx2_set_rx_csum(struct net_device *dev, u32 data) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); bp->rx_csum = data; return 0; @@ -4698,7 +4988,7 @@ bnx2_set_rx_csum(struct net_device *dev, u32 data) #define BNX2_NUM_STATS 45 -struct { +static struct { char string[ETH_GSTRING_LEN]; } bnx2_stats_str_arr[BNX2_NUM_STATS] = { { "rx_bytes" }, @@ -4750,7 +5040,7 @@ struct { #define STATS_OFFSET32(offset_name) (offsetof(struct statistics_block, offset_name) / 4) -unsigned long bnx2_stats_offset_arr[BNX2_NUM_STATS] = { +static const unsigned long bnx2_stats_offset_arr[BNX2_NUM_STATS] = { STATS_OFFSET32(stat_IfHCInOctets_hi), STATS_OFFSET32(stat_IfHCInBadOctets_hi), STATS_OFFSET32(stat_IfHCOutOctets_hi), @@ -4801,7 +5091,7 @@ unsigned long bnx2_stats_offset_arr[BNX2_NUM_STATS] = { /* stat_IfHCInBadOctets and stat_Dot3StatsCarrierSenseErrors are * skipped because of errata. */ -u8 bnx2_5706_stats_len_arr[BNX2_NUM_STATS] = { +static u8 bnx2_5706_stats_len_arr[BNX2_NUM_STATS] = { 8,0,8,8,8,8,8,8,8,8, 4,0,4,4,4,4,4,4,4,4, 4,4,4,4,4,4,4,4,4,4, @@ -4809,9 +5099,17 @@ u8 bnx2_5706_stats_len_arr[BNX2_NUM_STATS] = { 4,4,4,4,4, }; +static u8 bnx2_5708_stats_len_arr[BNX2_NUM_STATS] = { + 8,0,8,8,8,8,8,8,8,8, + 4,4,4,4,4,4,4,4,4,4, + 4,4,4,4,4,4,4,4,4,4, + 4,4,4,4,4,4,4,4,4,4, + 4,4,4,4,4, +}; + #define BNX2_NUM_TESTS 6 -struct { +static struct { char string[ETH_GSTRING_LEN]; } bnx2_tests_str_arr[BNX2_NUM_TESTS] = { { "register_test (offline)" }, @@ -4831,7 +5129,7 @@ bnx2_self_test_count(struct net_device *dev) static void bnx2_self_test(struct net_device *dev, struct ethtool_test *etest, u64 *buf) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); memset(buf, 0, sizeof(u64) * BNX2_NUM_TESTS); if (etest->flags & ETH_TEST_FL_OFFLINE) { @@ -4847,10 +5145,8 @@ bnx2_self_test(struct net_device *dev, struct ethtool_test *etest, u64 *buf) buf[1] = 1; etest->flags |= ETH_TEST_FL_FAILED; } - if (bnx2_test_loopback(bp) != 0) { - buf[2] = 1; + if ((buf[2] = bnx2_test_loopback(bp)) != 0) etest->flags |= ETH_TEST_FL_FAILED; - } if (!netif_running(bp->dev)) { bnx2_reset_chip(bp, BNX2_DRV_MSG_CODE_RESET); @@ -4907,18 +5203,23 @@ static void bnx2_get_ethtool_stats(struct net_device *dev, struct ethtool_stats *stats, u64 *buf) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); int i; u32 *hw_stats = (u32 *) bp->stats_blk; - u8 *stats_len_arr = 0; + u8 *stats_len_arr = NULL; if (hw_stats == NULL) { memset(buf, 0, sizeof(u64) * BNX2_NUM_STATS); return; } - if (CHIP_NUM(bp) == CHIP_NUM_5706) + if ((CHIP_ID(bp) == CHIP_ID_5706_A0) || + (CHIP_ID(bp) == CHIP_ID_5706_A1) || + (CHIP_ID(bp) == CHIP_ID_5706_A2) || + (CHIP_ID(bp) == CHIP_ID_5708_A0)) stats_len_arr = bnx2_5706_stats_len_arr; + else + stats_len_arr = bnx2_5708_stats_len_arr; for (i = 0; i < BNX2_NUM_STATS; i++) { if (stats_len_arr[i] == 0) { @@ -4942,7 +5243,7 @@ bnx2_get_ethtool_stats(struct net_device *dev, static int bnx2_phys_id(struct net_device *dev, u32 data) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); int i; u32 save; @@ -4977,6 +5278,8 @@ static struct ethtool_ops bnx2_ethtool_ops = { .get_settings = bnx2_get_settings, .set_settings = bnx2_set_settings, .get_drvinfo = bnx2_get_drvinfo, + .get_regs_len = bnx2_get_regs_len, + .get_regs = bnx2_get_regs, .get_wol = bnx2_get_wol, .set_wol = bnx2_set_wol, .nway_reset = bnx2_nway_reset, @@ -5006,14 +5309,15 @@ static struct ethtool_ops bnx2_ethtool_ops = { .phys_id = bnx2_phys_id, .get_stats_count = bnx2_get_stats_count, .get_ethtool_stats = bnx2_get_ethtool_stats, + .get_perm_addr = ethtool_op_get_perm_addr, }; /* Called with rtnl_lock */ static int bnx2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) { - struct mii_ioctl_data *data = (struct mii_ioctl_data *)&ifr->ifr_data; - struct bnx2 *bp = dev->priv; + struct mii_ioctl_data *data = if_mii(ifr); + struct bnx2 *bp = netdev_priv(dev); int err; switch(cmd) { @@ -5024,9 +5328,9 @@ bnx2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) case SIOCGMIIREG: { u32 mii_regval; - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); err = bnx2_read_phy(bp, data->reg_num & 0x1f, &mii_regval); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); data->val_out = mii_regval; @@ -5037,9 +5341,9 @@ bnx2_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd) if (!capable(CAP_NET_ADMIN)) return -EPERM; - spin_lock_irq(&bp->phy_lock); + spin_lock_bh(&bp->phy_lock); err = bnx2_write_phy(bp, data->reg_num & 0x1f, data->val_in); - spin_unlock_irq(&bp->phy_lock); + spin_unlock_bh(&bp->phy_lock); return err; @@ -5055,7 +5359,10 @@ static int bnx2_change_mac_addr(struct net_device *dev, void *p) { struct sockaddr *addr = p; - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); + + if (!is_valid_ether_addr(addr->sa_data)) + return -EINVAL; memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); if (netif_running(dev)) @@ -5068,7 +5375,7 @@ bnx2_change_mac_addr(struct net_device *dev, void *p) static int bnx2_change_mtu(struct net_device *dev, int new_mtu) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); if (((new_mtu + ETH_HLEN) > MAX_ETHERNET_JUMBO_PACKET_SIZE) || ((new_mtu + ETH_HLEN) < MIN_ETHERNET_PACKET_SIZE)) @@ -5089,7 +5396,7 @@ bnx2_change_mtu(struct net_device *dev, int new_mtu) static void poll_bnx2(struct net_device *dev) { - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); disable_irq(bp->pdev->irq); bnx2_interrupt(bp->pdev->irq, dev, NULL); @@ -5107,7 +5414,7 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) SET_MODULE_OWNER(dev); SET_NETDEV_DEV(dev, &pdev->dev); - bp = dev->priv; + bp = netdev_priv(dev); bp->flags = 0; bp->phy_flags = 0; @@ -5192,12 +5499,10 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) BNX2_PCICFG_MISC_CONFIG_REG_WINDOW_ENA | BNX2_PCICFG_MISC_CONFIG_TARGET_MB_WORD_SWAP); - bnx2_set_power_state(bp, 0); + bnx2_set_power_state(bp, PCI_D0); bp->chip_id = REG_RD(bp, BNX2_MISC_ID); - bp->phy_addr = 1; - /* Get bus information. */ reg = REG_RD(bp, BNX2_PCICFG_MISC_STATUS); if (reg & BNX2_PCICFG_MISC_STATUS_PCIX_DET) { @@ -5260,10 +5565,18 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bnx2_init_nvram(bp); + reg = REG_RD_IND(bp, BNX2_SHM_HDR_SIGNATURE); + + if ((reg & BNX2_SHM_HDR_SIGNATURE_SIG_MASK) == + BNX2_SHM_HDR_SIGNATURE_SIG) + bp->shmem_base = REG_RD_IND(bp, BNX2_SHM_HDR_ADDR_0); + else + bp->shmem_base = HOST_VIEW_SHMEM_BASE; + /* Get the permanent MAC address. First we need to make sure the * firmware is actually running. */ - reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_DEV_INFO_SIGNATURE); + reg = REG_RD_IND(bp, bp->shmem_base + BNX2_DEV_INFO_SIGNATURE); if ((reg & BNX2_DEV_INFO_SIGNATURE_MAGIC_MASK) != BNX2_DEV_INFO_SIGNATURE_MAGIC) { @@ -5272,21 +5585,20 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) goto err_out_unmap; } - bp->fw_ver = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + - BNX2_DEV_INFO_BC_REV); + bp->fw_ver = REG_RD_IND(bp, bp->shmem_base + BNX2_DEV_INFO_BC_REV); - reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_PORT_HW_CFG_MAC_UPPER); + reg = REG_RD_IND(bp, bp->shmem_base + BNX2_PORT_HW_CFG_MAC_UPPER); bp->mac_addr[0] = (u8) (reg >> 8); bp->mac_addr[1] = (u8) reg; - reg = REG_RD_IND(bp, HOST_VIEW_SHMEM_BASE + BNX2_PORT_HW_CFG_MAC_LOWER); + reg = REG_RD_IND(bp, bp->shmem_base + BNX2_PORT_HW_CFG_MAC_LOWER); bp->mac_addr[2] = (u8) (reg >> 24); bp->mac_addr[3] = (u8) (reg >> 16); bp->mac_addr[4] = (u8) (reg >> 8); bp->mac_addr[5] = (u8) reg; bp->tx_ring_size = MAX_TX_DESC_CNT; - bp->rx_ring_size = 100; + bnx2_set_rx_ring_size(bp, 100); bp->rx_csum = 1; @@ -5305,13 +5617,26 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->stats_ticks = 1000000 & 0xffff00; bp->timer_interval = HZ; + bp->current_interval = HZ; + + bp->phy_addr = 1; /* Disable WOL support if we are running on a SERDES chip. */ if (CHIP_BOND_ID(bp) & CHIP_BOND_ID_SERDES_BIT) { bp->phy_flags |= PHY_SERDES_FLAG; bp->flags |= NO_WOL_FLAG; + if (CHIP_NUM(bp) == CHIP_NUM_5708) { + bp->phy_addr = 2; + reg = REG_RD_IND(bp, bp->shmem_base + + BNX2_SHARED_HW_CFG_CONFIG); + if (reg & BNX2_SHARED_HW_CFG_PHY_2_5G) + bp->phy_flags |= PHY_2_5G_CAPABLE_FLAG; + } } + if (CHIP_NUM(bp) == CHIP_NUM_5708) + bp->flags |= NO_WOL_FLAG; + if (CHIP_ID(bp) == CHIP_ID_5706_A0) { bp->tx_quick_cons_trip_int = bp->tx_quick_cons_trip; @@ -5328,6 +5653,14 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->req_line_speed = 0; if (bp->phy_flags & PHY_SERDES_FLAG) { bp->advertising = ETHTOOL_ALL_FIBRE_SPEED | ADVERTISED_Autoneg; + + reg = REG_RD_IND(bp, bp->shmem_base + BNX2_PORT_HW_CFG_CONFIG); + reg &= BNX2_PORT_HW_CFG_CFG_DFLT_LINK_MASK; + if (reg == BNX2_PORT_HW_CFG_CFG_DFLT_LINK_1G) { + bp->autoneg = 0; + bp->req_line_speed = bp->line_speed = SPEED_1000; + bp->req_duplex = DUPLEX_FULL; + } } else { bp->advertising = ETHTOOL_ALL_COPPER_SPEED | ADVERTISED_Autoneg; @@ -5335,11 +5668,17 @@ bnx2_init_board(struct pci_dev *pdev, struct net_device *dev) bp->req_flow_ctrl = FLOW_CTRL_RX | FLOW_CTRL_TX; + init_timer(&bp->timer); + bp->timer.expires = RUN_AT(bp->timer_interval); + bp->timer.data = (unsigned long) bp; + bp->timer.function = bnx2_timer; + return 0; err_out_unmap: if (bp->regview) { iounmap(bp->regview); + bp->regview = NULL; } err_out_release: @@ -5394,7 +5733,7 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) dev->ethtool_ops = &bnx2_ethtool_ops; dev->weight = 64; - bp = dev->priv; + bp = netdev_priv(dev); #if defined(HAVE_POLL_CONTROLLER) || defined(CONFIG_NET_POLL_CONTROLLER) dev->poll_controller = poll_bnx2; @@ -5414,6 +5753,7 @@ bnx2_init_one(struct pci_dev *pdev, const struct pci_device_id *ent) pci_set_drvdata(pdev, dev); memcpy(dev->dev_addr, bp->mac_addr, 6); + memcpy(dev->perm_addr, bp->mac_addr, 6); bp->name = board_info[ent->driver_data].name, printk(KERN_INFO "%s: %s (%c%d) PCI%s %s %dMHz found at mem %lx, " "IRQ %d, ", @@ -5452,7 +5792,9 @@ static void __devexit bnx2_remove_one(struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); + + flush_scheduled_work(); unregister_netdev(dev); @@ -5466,25 +5808,28 @@ bnx2_remove_one(struct pci_dev *pdev) } static int -bnx2_suspend(struct pci_dev *pdev, u32 state) +bnx2_suspend(struct pci_dev *pdev, pm_message_t state) { struct net_device *dev = pci_get_drvdata(pdev); - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); u32 reset_code; if (!netif_running(dev)) return 0; + flush_scheduled_work(); bnx2_netif_stop(bp); netif_device_detach(dev); del_timer_sync(&bp->timer); - if (bp->wol) + if (bp->flags & NO_WOL_FLAG) + reset_code = BNX2_DRV_MSG_CODE_UNLOAD; + else if (bp->wol) reset_code = BNX2_DRV_MSG_CODE_SUSPEND_WOL; else reset_code = BNX2_DRV_MSG_CODE_SUSPEND_NO_WOL; bnx2_reset_chip(bp, reset_code); bnx2_free_skbs(bp); - bnx2_set_power_state(bp, state); + bnx2_set_power_state(bp, pci_choose_state(pdev, state)); return 0; } @@ -5492,12 +5837,12 @@ static int bnx2_resume(struct pci_dev *pdev) { struct net_device *dev = pci_get_drvdata(pdev); - struct bnx2 *bp = dev->priv; + struct bnx2 *bp = netdev_priv(dev); if (!netif_running(dev)) return 0; - bnx2_set_power_state(bp, 0); + bnx2_set_power_state(bp, PCI_D0); netif_device_attach(dev); bnx2_init_nic(bp); bnx2_netif_start(bp); @@ -5505,12 +5850,12 @@ bnx2_resume(struct pci_dev *pdev) } static struct pci_driver bnx2_pci_driver = { - name: DRV_MODULE_NAME, - id_table: bnx2_pci_tbl, - probe: bnx2_init_one, - remove: __devexit_p(bnx2_remove_one), - suspend: bnx2_suspend, - resume: bnx2_resume, + .name = DRV_MODULE_NAME, + .id_table = bnx2_pci_tbl, + .probe = bnx2_init_one, + .remove = __devexit_p(bnx2_remove_one), + .suspend = bnx2_suspend, + .resume = bnx2_resume, }; static int __init bnx2_init(void) diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c index 55d236726..8171cae06 100644 --- a/drivers/net/bonding/bond_main.c +++ b/drivers/net/bonding/bond_main.c @@ -1199,8 +1199,7 @@ int bond_sethwaddr(struct net_device *bond_dev, struct net_device *slave_dev) } #define BOND_INTERSECT_FEATURES \ - (NETIF_F_SG|NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM|\ - NETIF_F_TSO|NETIF_F_UFO) + (NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_TSO | NETIF_F_UFO) /* * Compute the common dev->feature set available to all slaves. Some @@ -1218,9 +1217,7 @@ static int bond_compute_features(struct bonding *bond) features &= (slave->dev->features & BOND_INTERSECT_FEATURES); if ((features & NETIF_F_SG) && - !(features & (NETIF_F_IP_CSUM | - NETIF_F_NO_CSUM | - NETIF_F_HW_CSUM))) + !(features & NETIF_F_ALL_CSUM)) features &= ~NETIF_F_SG; /* @@ -4191,7 +4188,7 @@ static int bond_init(struct net_device *bond_dev, struct bond_params *params) */ bond_dev->features |= NETIF_F_VLAN_CHALLENGED; - /* don't acquire bond device's xmit_lock when + /* don't acquire bond device's netif_tx_lock when * transmitting */ bond_dev->features |= NETIF_F_LLTX; diff --git a/drivers/net/chelsio/sge.c b/drivers/net/chelsio/sge.c index 4391bf4bf..722be62f3 100644 --- a/drivers/net/chelsio/sge.c +++ b/drivers/net/chelsio/sge.c @@ -1418,7 +1418,7 @@ int t1_start_xmit(struct sk_buff *skb, struct net_device *dev) struct cpl_tx_pkt *cpl; #ifdef NETIF_F_TSO - if (skb_shinfo(skb)->tso_size) { + if (skb_is_gso(skb)) { int eth_type; struct cpl_tx_pkt_lso *hdr; @@ -1433,7 +1433,7 @@ int t1_start_xmit(struct sk_buff *skb, struct net_device *dev) hdr->ip_hdr_words = skb->nh.iph->ihl; hdr->tcp_hdr_words = skb->h.th->doff; hdr->eth_type_mss = htons(MK_ETH_TYPE_MSS(eth_type, - skb_shinfo(skb)->tso_size)); + skb_shinfo(skb)->gso_size)); hdr->len = htonl(skb->len - sizeof(*hdr)); cpl = (struct cpl_tx_pkt *)hdr; sge->stats.tx_lso_pkts++; diff --git a/drivers/net/e1000/e1000_hw.c b/drivers/net/e1000/e1000_hw.c index 523c2c9fc..c5e702349 100644 --- a/drivers/net/e1000/e1000_hw.c +++ b/drivers/net/e1000/e1000_hw.c @@ -353,6 +353,7 @@ e1000_set_mac_type(struct e1000_hw *hw) case E1000_DEV_ID_82572EI_COPPER: case E1000_DEV_ID_82572EI_FIBER: case E1000_DEV_ID_82572EI_SERDES: + case E1000_DEV_ID_82572EI: hw->mac_type = e1000_82572; break; case E1000_DEV_ID_82573E: diff --git a/drivers/net/e1000/e1000_hw.h b/drivers/net/e1000/e1000_hw.h index 150e45e30..c01e5d2e5 100644 --- a/drivers/net/e1000/e1000_hw.h +++ b/drivers/net/e1000/e1000_hw.h @@ -462,6 +462,7 @@ int32_t e1000_check_phy_reset_block(struct e1000_hw *hw); #define E1000_DEV_ID_82572EI_COPPER 0x107D #define E1000_DEV_ID_82572EI_FIBER 0x107E #define E1000_DEV_ID_82572EI_SERDES 0x107F +#define E1000_DEV_ID_82572EI 0x10B9 #define E1000_DEV_ID_82573E 0x108B #define E1000_DEV_ID_82573E_IAMT 0x108C #define E1000_DEV_ID_82573L 0x109A diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c index 97e71a4fe..43e647284 100644 --- a/drivers/net/e1000/e1000_main.c +++ b/drivers/net/e1000/e1000_main.c @@ -2413,7 +2413,7 @@ e1000_tso(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, uint8_t ipcss, ipcso, tucss, tucso, hdr_len; int err; - if (skb_shinfo(skb)->tso_size) { + if (skb_is_gso(skb)) { if (skb_header_cloned(skb)) { err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC); if (err) @@ -2421,7 +2421,7 @@ e1000_tso(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, } hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2)); - mss = skb_shinfo(skb)->tso_size; + mss = skb_shinfo(skb)->gso_size; if (skb->protocol == ntohs(ETH_P_IP)) { skb->nh.iph->tot_len = 0; skb->nh.iph->check = 0; @@ -2538,7 +2538,7 @@ e1000_tx_map(struct e1000_adapter *adapter, struct e1000_tx_ring *tx_ring, * tso gets written back prematurely before the data is fully * DMA'd to the controller */ if (!skb->data_len && tx_ring->last_tx_tso && - !skb_shinfo(skb)->tso_size) { + !skb_is_gso(skb)) { tx_ring->last_tx_tso = 0; size -= 4; } @@ -2776,7 +2776,7 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) } #ifdef NETIF_F_TSO - mss = skb_shinfo(skb)->tso_size; + mss = skb_shinfo(skb)->gso_size; /* The controller does a simple calculation to * make sure there is enough room in the FIFO before * initiating the DMA for each buffer. The calc is: @@ -2825,8 +2825,7 @@ e1000_xmit_frame(struct sk_buff *skb, struct net_device *netdev) #ifdef NETIF_F_TSO /* Controller Erratum workaround */ - if (!skb->data_len && tx_ring->last_tx_tso && - !skb_shinfo(skb)->tso_size) + if (!skb->data_len && tx_ring->last_tx_tso && !skb_is_gso(skb)) count++; #endif diff --git a/drivers/net/forcedeth.c b/drivers/net/forcedeth.c index feb5b223c..1a556957d 100644 --- a/drivers/net/forcedeth.c +++ b/drivers/net/forcedeth.c @@ -533,9 +533,9 @@ typedef union _ring_type { * critical parts: * - rx is (pseudo-) lockless: it relies on the single-threading provided * by the arch code for interrupts. - * - tx setup is lockless: it relies on dev->xmit_lock. Actual submission + * - tx setup is lockless: it relies on netif_tx_lock. Actual submission * needs dev->priv->lock :-( - * - set_multicast_list: preparation lockless, relies on dev->xmit_lock. + * - set_multicast_list: preparation lockless, relies on netif_tx_lock. */ /* in dev: base, irq */ @@ -1213,7 +1213,7 @@ static void drain_ring(struct net_device *dev) /* * nv_start_xmit: dev->hard_start_xmit function - * Called with dev->xmit_lock held. + * Called with netif_tx_lock held. */ static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev) { @@ -1303,8 +1303,8 @@ static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev) np->tx_skbuff[nr] = skb; #ifdef NETIF_F_TSO - if (skb_shinfo(skb)->tso_size) - tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->tso_size << NV_TX2_TSO_SHIFT); + if (skb_is_gso(skb)) + tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->gso_size << NV_TX2_TSO_SHIFT); else #endif tx_flags_extra = (skb->ip_summed == CHECKSUM_HW ? (NV_TX2_CHECKSUM_L3|NV_TX2_CHECKSUM_L4) : 0); @@ -1407,7 +1407,7 @@ static void nv_tx_done(struct net_device *dev) /* * nv_tx_timeout: dev->tx_timeout function - * Called with dev->xmit_lock held. + * Called with netif_tx_lock held. */ static void nv_tx_timeout(struct net_device *dev) { @@ -1736,8 +1736,8 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu) * guessed, there is probably a simpler approach. * Changing the MTU is a rare event, it shouldn't matter. */ - nv_disable_irq(dev); - spin_lock_bh(&dev->xmit_lock); + disable_irq(dev->irq); + netif_tx_lock_bh(dev); spin_lock(&np->lock); /* stop engines */ nv_stop_rx(dev); @@ -1768,8 +1768,8 @@ static int nv_change_mtu(struct net_device *dev, int new_mtu) nv_start_rx(dev); nv_start_tx(dev); spin_unlock(&np->lock); - spin_unlock_bh(&dev->xmit_lock); - nv_enable_irq(dev); + netif_tx_unlock_bh(dev); + enable_irq(dev->irq); } return 0; } @@ -1803,7 +1803,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr) memcpy(dev->dev_addr, macaddr->sa_data, ETH_ALEN); if (netif_running(dev)) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); spin_lock_irq(&np->lock); /* stop rx engine */ @@ -1815,7 +1815,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr) /* restart rx engine */ nv_start_rx(dev); spin_unlock_irq(&np->lock); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } else { nv_copy_mac_to_hw(dev); } @@ -1824,7 +1824,7 @@ static int nv_set_mac_address(struct net_device *dev, void *addr) /* * nv_set_multicast: dev->set_multicast function - * Called with dev->xmit_lock held. + * Called with netif_tx_lock held. */ static void nv_set_multicast(struct net_device *dev) { diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c index 102c1f0b9..d12605f0a 100644 --- a/drivers/net/hamradio/6pack.c +++ b/drivers/net/hamradio/6pack.c @@ -308,9 +308,9 @@ static int sp_set_mac_address(struct net_device *dev, void *addr) { struct sockaddr_ax25 *sa = addr; - spin_lock_irq(&dev->xmit_lock); + netif_tx_lock_bh(dev); memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN); - spin_unlock_irq(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return 0; } @@ -767,9 +767,9 @@ static int sixpack_ioctl(struct tty_struct *tty, struct file *file, break; } - spin_lock_irq(&dev->xmit_lock); + netif_tx_lock_bh(dev); memcpy(dev->dev_addr, &addr, AX25_ADDR_LEN); - spin_unlock_irq(&dev->xmit_lock); + netif_tx_unlock_bh(dev); err = 0; break; diff --git a/drivers/net/hamradio/mkiss.c b/drivers/net/hamradio/mkiss.c index d81a8e1ee..3ebbbe56b 100644 --- a/drivers/net/hamradio/mkiss.c +++ b/drivers/net/hamradio/mkiss.c @@ -357,9 +357,9 @@ static int ax_set_mac_address(struct net_device *dev, void *addr) { struct sockaddr_ax25 *sa = addr; - spin_lock_irq(&dev->xmit_lock); + netif_tx_lock_bh(dev); memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN); - spin_unlock_irq(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return 0; } @@ -886,9 +886,9 @@ static int mkiss_ioctl(struct tty_struct *tty, struct file *file, break; } - spin_lock_irq(&dev->xmit_lock); + netif_tx_lock_bh(dev); memcpy(dev->dev_addr, addr, AX25_ADDR_LEN); - spin_unlock_irq(&dev->xmit_lock); + netif_tx_unlock_bh(dev); err = 0; break; diff --git a/drivers/net/ifb.c b/drivers/net/ifb.c index 31fb2d75d..2e222ef91 100644 --- a/drivers/net/ifb.c +++ b/drivers/net/ifb.c @@ -76,13 +76,13 @@ static void ri_tasklet(unsigned long dev) dp->st_task_enter++; if ((skb = skb_peek(&dp->tq)) == NULL) { dp->st_txq_refl_try++; - if (spin_trylock(&_dev->xmit_lock)) { + if (netif_tx_trylock(_dev)) { dp->st_rxq_enter++; while ((skb = skb_dequeue(&dp->rq)) != NULL) { skb_queue_tail(&dp->tq, skb); dp->st_rx2tx_tran++; } - spin_unlock(&_dev->xmit_lock); + netif_tx_unlock(_dev); } else { /* reschedule */ dp->st_rxq_notenter++; @@ -110,7 +110,7 @@ static void ri_tasklet(unsigned long dev) } } - if (spin_trylock(&_dev->xmit_lock)) { + if (netif_tx_trylock(_dev)) { dp->st_rxq_check++; if ((skb = skb_peek(&dp->rq)) == NULL) { dp->tasklet_pending = 0; @@ -118,10 +118,10 @@ static void ri_tasklet(unsigned long dev) netif_wake_queue(_dev); } else { dp->st_rxq_rsch++; - spin_unlock(&_dev->xmit_lock); + netif_tx_unlock(_dev); goto resched; } - spin_unlock(&_dev->xmit_lock); + netif_tx_unlock(_dev); } else { resched: dp->tasklet_pending = 1; diff --git a/drivers/net/irda/vlsi_ir.c b/drivers/net/irda/vlsi_ir.c index 97a49e0be..d70b9e8d6 100644 --- a/drivers/net/irda/vlsi_ir.c +++ b/drivers/net/irda/vlsi_ir.c @@ -959,7 +959,7 @@ static int vlsi_hard_start_xmit(struct sk_buff *skb, struct net_device *ndev) || (now.tv_sec==ready.tv_sec && now.tv_usec>=ready.tv_usec)) break; udelay(100); - /* must not sleep here - we are called under xmit_lock! */ + /* must not sleep here - called under netif_tx_lock! */ } } diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c index cfd67d812..83852450a 100644 --- a/drivers/net/ixgb/ixgb_main.c +++ b/drivers/net/ixgb/ixgb_main.c @@ -1168,7 +1168,7 @@ ixgb_tso(struct ixgb_adapter *adapter, struct sk_buff *skb) uint16_t ipcse, tucse, mss; int err; - if(likely(skb_shinfo(skb)->tso_size)) { + if (likely(skb_is_gso(skb))) { if (skb_header_cloned(skb)) { err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC); if (err) @@ -1176,7 +1176,7 @@ ixgb_tso(struct ixgb_adapter *adapter, struct sk_buff *skb) } hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2)); - mss = skb_shinfo(skb)->tso_size; + mss = skb_shinfo(skb)->gso_size; skb->nh.iph->tot_len = 0; skb->nh.iph->check = 0; skb->h.th->check = ~csum_tcpudp_magic(skb->nh.iph->saddr, diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c index b79d6e8d3..997cbce9a 100644 --- a/drivers/net/loopback.c +++ b/drivers/net/loopback.c @@ -74,7 +74,7 @@ static void emulate_large_send_offload(struct sk_buff *skb) struct iphdr *iph = skb->nh.iph; struct tcphdr *th = (struct tcphdr*)(skb->nh.raw + (iph->ihl * 4)); unsigned int doffset = (iph->ihl + th->doff) * 4; - unsigned int mtu = skb_shinfo(skb)->tso_size + doffset; + unsigned int mtu = skb_shinfo(skb)->gso_size + doffset; unsigned int offset = 0; u32 seq = ntohl(th->seq); u16 id = ntohs(iph->id); @@ -139,7 +139,7 @@ static int loopback_xmit(struct sk_buff *skb, struct net_device *dev) #endif #ifdef LOOPBACK_TSO - if (skb_shinfo(skb)->tso_size) { + if (skb_is_gso(skb)) { BUG_ON(skb->protocol != htons(ETH_P_IP)); BUG_ON(skb->nh.iph->protocol != IPPROTO_TCP); diff --git a/drivers/net/mv643xx_eth.c b/drivers/net/mv643xx_eth.c index 411f4d809..625ff61c9 100644 --- a/drivers/net/mv643xx_eth.c +++ b/drivers/net/mv643xx_eth.c @@ -1200,7 +1200,7 @@ static int mv643xx_eth_start_xmit(struct sk_buff *skb, struct net_device *dev) } if (has_tiny_unaligned_frags(skb)) { - if ((skb_linearize(skb, GFP_ATOMIC) != 0)) { + if (__skb_linearize(skb)) { stats->tx_dropped++; printk(KERN_DEBUG "%s: failed to linearize tiny " "unaligned fragment\n", dev->name); diff --git a/drivers/net/natsemi.c b/drivers/net/natsemi.c index 90627756d..2e4ecedba 100644 --- a/drivers/net/natsemi.c +++ b/drivers/net/natsemi.c @@ -318,12 +318,12 @@ performance critical codepaths: The rx process only runs in the interrupt handler. Access from outside the interrupt handler is only permitted after disable_irq(). -The rx process usually runs under the dev->xmit_lock. If np->intr_tx_reap +The rx process usually runs under the netif_tx_lock. If np->intr_tx_reap is set, then access is permitted under spin_lock_irq(&np->lock). Thus configuration functions that want to access everything must call disable_irq(dev->irq); - spin_lock_bh(dev->xmit_lock); + netif_tx_lock_bh(dev); spin_lock_irq(&np->lock); IV. Notes diff --git a/drivers/net/r8169.c b/drivers/net/r8169.c index 0ad331029..43641dd2c 100644 --- a/drivers/net/r8169.c +++ b/drivers/net/r8169.c @@ -2171,7 +2171,7 @@ static int rtl8169_xmit_frags(struct rtl8169_private *tp, struct sk_buff *skb, static inline u32 rtl8169_tso_csum(struct sk_buff *skb, struct net_device *dev) { if (dev->features & NETIF_F_TSO) { - u32 mss = skb_shinfo(skb)->tso_size; + u32 mss = skb_shinfo(skb)->gso_size; if (mss) return LargeSend | ((mss & MSSMask) << MSSShift); diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c index 79208f434..0ad01f07e 100644 --- a/drivers/net/s2io.c +++ b/drivers/net/s2io.c @@ -3564,8 +3564,8 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) txdp->Control_1 = 0; txdp->Control_2 = 0; #ifdef NETIF_F_TSO - mss = skb_shinfo(skb)->tso_size; - if (mss) { + mss = skb_shinfo(skb)->gso_size; + if (skb_shinfo(skb)->gso_type == SKB_GSO_TCPV4) { txdp->Control_1 |= TXD_TCP_LSO_EN; txdp->Control_1 |= TXD_TCP_LSO_MSS(mss); } @@ -3585,10 +3585,10 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) } frg_len = skb->len - skb->data_len; - if (skb_shinfo(skb)->ufo_size) { + if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) { int ufo_size; - ufo_size = skb_shinfo(skb)->ufo_size; + ufo_size = skb_shinfo(skb)->gso_size; ufo_size &= ~7; txdp->Control_1 |= TXD_UFO_EN; txdp->Control_1 |= TXD_UFO_MSS(ufo_size); @@ -3614,7 +3614,7 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) txdp->Host_Control = (unsigned long) skb; txdp->Control_1 |= TXD_BUFFER0_SIZE(frg_len); - if (skb_shinfo(skb)->ufo_size) + if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) txdp->Control_1 |= TXD_UFO_EN; frg_cnt = skb_shinfo(skb)->nr_frags; @@ -3629,12 +3629,12 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) (sp->pdev, frag->page, frag->page_offset, frag->size, PCI_DMA_TODEVICE); txdp->Control_1 = TXD_BUFFER0_SIZE(frag->size); - if (skb_shinfo(skb)->ufo_size) + if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) txdp->Control_1 |= TXD_UFO_EN; } txdp->Control_1 |= TXD_GATHER_CODE_LAST; - if (skb_shinfo(skb)->ufo_size) + if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) frg_cnt++; /* as Txd0 was used for inband header */ tx_fifo = mac_control->tx_FIFO_start[queue]; @@ -3648,7 +3648,7 @@ static int s2io_xmit(struct sk_buff *skb, struct net_device *dev) if (mss) val64 |= TX_FIFO_SPECIAL_FUNC; #endif - if (skb_shinfo(skb)->ufo_size) + if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) val64 |= TX_FIFO_SPECIAL_FUNC; writeq(val64, &tx_fifo->List_Control); diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c index fba1e4d4d..4103e304f 100644 --- a/drivers/net/sky2.c +++ b/drivers/net/sky2.c @@ -51,7 +51,7 @@ #include "sky2.h" #define DRV_NAME "sky2" -#define DRV_VERSION "1.4" +#define DRV_VERSION "1.6.1" #define PFX DRV_NAME " " /* @@ -233,6 +233,8 @@ static void sky2_set_power_state(struct sky2_hw *hw, pci_power_t state) if (hw->ports > 1) reg1 |= PCI_Y2_PHY2_COMA; } + sky2_pci_write32(hw, PCI_DEV_REG1, reg1); + udelay(100); if (hw->chip_id == CHIP_ID_YUKON_EC_U) { sky2_write16(hw, B0_CTST, Y2_HW_WOL_ON); @@ -243,8 +245,6 @@ static void sky2_set_power_state(struct sky2_hw *hw, pci_power_t state) sky2_pci_write32(hw, PCI_DEV_REG5, 0); } - sky2_pci_write32(hw, PCI_DEV_REG1, reg1); - break; case PCI_D3hot: @@ -321,7 +321,7 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) } ctrl = gm_phy_read(hw, port, PHY_MARV_PHY_CTRL); - if (hw->copper) { + if (sky2_is_copper(hw)) { if (hw->chip_id == CHIP_ID_YUKON_FE) { /* enable automatic crossover */ ctrl |= PHY_M_PC_MDI_XMODE(PHY_M_PC_ENA_AUTO) >> 1; @@ -338,25 +338,37 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) ctrl |= PHY_M_PC_DSC(2) | PHY_M_PC_DOWN_S_ENA; } } - gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); } else { /* workaround for deviation #4.88 (CRC errors) */ /* disable Automatic Crossover */ ctrl &= ~PHY_M_PC_MDIX_MSK; - gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); + } - if (hw->chip_id == CHIP_ID_YUKON_XL) { - /* Fiber: select 1000BASE-X only mode MAC Specific Ctrl Reg. */ - gm_phy_write(hw, port, PHY_MARV_EXT_ADR, 2); - ctrl = gm_phy_read(hw, port, PHY_MARV_PHY_CTRL); - ctrl &= ~PHY_M_MAC_MD_MSK; - ctrl |= PHY_M_MAC_MODE_SEL(PHY_M_MAC_MD_1000BX); - gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); + gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); + /* special setup for PHY 88E1112 Fiber */ + if (hw->chip_id == CHIP_ID_YUKON_XL && !sky2_is_copper(hw)) { + pg = gm_phy_read(hw, port, PHY_MARV_EXT_ADR); + + /* Fiber: select 1000BASE-X only mode MAC Specific Ctrl Reg. */ + gm_phy_write(hw, port, PHY_MARV_EXT_ADR, 2); + ctrl = gm_phy_read(hw, port, PHY_MARV_PHY_CTRL); + ctrl &= ~PHY_M_MAC_MD_MSK; + ctrl |= PHY_M_MAC_MODE_SEL(PHY_M_MAC_MD_1000BX); + gm_phy_write(hw, port, PHY_MARV_PHY_CTRL, ctrl); + + if (hw->pmd_type == 'P') { /* select page 1 to access Fiber registers */ gm_phy_write(hw, port, PHY_MARV_EXT_ADR, 1); + + /* for SFP-module set SIGDET polarity to low */ + ctrl = gm_phy_read(hw, port, PHY_MARV_PHY_CTRL); + ctrl |= PHY_M_FIB_SIGD_POL; + gm_phy_write(hw, port, PHY_MARV_CTRL, ctrl); } + + gm_phy_write(hw, port, PHY_MARV_EXT_ADR, pg); } ctrl = gm_phy_read(hw, port, PHY_MARV_CTRL); @@ -373,7 +385,7 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) adv = PHY_AN_CSMA; if (sky2->autoneg == AUTONEG_ENABLE) { - if (hw->copper) { + if (sky2_is_copper(hw)) { if (sky2->advertising & ADVERTISED_1000baseT_Full) ct1000 |= PHY_M_1000C_AFD; if (sky2->advertising & ADVERTISED_1000baseT_Half) @@ -386,8 +398,12 @@ static void sky2_phy_init(struct sky2_hw *hw, unsigned port) adv |= PHY_M_AN_10_FD; if (sky2->advertising & ADVERTISED_10baseT_Half) adv |= PHY_M_AN_10_HD; - } else /* special defines for FIBER (88E1011S only) */ - adv |= PHY_M_AN_1000X_AHD | PHY_M_AN_1000X_AFD; + } else { /* special defines for FIBER (88E1040S only) */ + if (sky2->advertising & ADVERTISED_1000baseT_Full) + adv |= PHY_M_AN_1000X_AFD; + if (sky2->advertising & ADVERTISED_1000baseT_Half) + adv |= PHY_M_AN_1000X_AHD; + } /* Set Flow-control capabilities */ if (sky2->tx_pause && sky2->rx_pause) @@ -949,14 +965,14 @@ static void sky2_vlan_rx_kill_vid(struct net_device *dev, unsigned short vid) /* * It appears the hardware has a bug in the FIFO logic that * cause it to hang if the FIFO gets overrun and the receive buffer - * is not aligned. ALso alloc_skb() won't align properly if slab + * is not aligned. Also dev_alloc_skb() won't align properly if slab * debugging is enabled. */ static inline struct sk_buff *sky2_alloc_skb(unsigned int size, gfp_t gfp_mask) { struct sk_buff *skb; - skb = alloc_skb(size + RX_SKB_ALIGN, gfp_mask); + skb = __dev_alloc_skb(size + RX_SKB_ALIGN, gfp_mask); if (likely(skb)) { unsigned long p = (unsigned long) skb->data; skb_reserve(skb, ALIGN(p, RX_SKB_ALIGN) - p); @@ -1160,7 +1176,7 @@ static unsigned tx_le_req(const struct sk_buff *skb) count = sizeof(dma_addr_t) / sizeof(u32); count += skb_shinfo(skb)->nr_frags * count; - if (skb_shinfo(skb)->tso_size) + if (skb_is_gso(skb)) ++count; if (skb->ip_summed == CHECKSUM_HW) @@ -1232,7 +1248,7 @@ static int sky2_xmit_frame(struct sk_buff *skb, struct net_device *dev) } /* Check for TCP Segmentation Offload */ - mss = skb_shinfo(skb)->tso_size; + mss = skb_shinfo(skb)->gso_size; if (mss != 0) { /* just drop the packet if non-linear expansion fails */ if (skb_header_cloned(skb) && @@ -1497,7 +1513,7 @@ static int sky2_down(struct net_device *dev) static u16 sky2_phy_speed(const struct sky2_hw *hw, u16 aux) { - if (!hw->copper) + if (!sky2_is_copper(hw)) return SPEED_1000; if (hw->chip_id == CHIP_ID_YUKON_FE) @@ -1855,7 +1871,7 @@ static struct sk_buff *sky2_receive(struct sky2_port *sky2, goto oversize; if (length < copybreak) { - skb = alloc_skb(length + 2, GFP_ATOMIC); + skb = dev_alloc_skb(length + 2); if (!skb) goto resubmit; @@ -2016,6 +2032,9 @@ static int sky2_status_intr(struct sky2_hw *hw, int to_do) } } + /* Fully processed status ring so clear irq */ + sky2_write32(hw, STAT_CTRL, SC_STAT_CLR_IRQ); + exit_loop: return work_done; } @@ -2187,9 +2206,6 @@ static int sky2_poll(struct net_device *dev0, int *budget) int work_done = 0; u32 status = sky2_read32(hw, B0_Y2_SP_EISR); - if (!~status) - goto out; - if (status & Y2_IS_HW_ERR) sky2_hw_intr(hw); @@ -2221,12 +2237,9 @@ static int sky2_poll(struct net_device *dev0, int *budget) *budget -= work_done; dev0->quota -= work_done; - if (status & Y2_IS_STAT_BMU) - sky2_write32(hw, STAT_CTRL, SC_STAT_CLR_IRQ); - if (sky2_more_work(hw)) return 1; -out: + netif_rx_complete(dev0); sky2_read32(hw, B0_Y2_SP_LISR); @@ -2290,7 +2303,7 @@ static inline u32 sky2_clk2us(const struct sky2_hw *hw, u32 clk) static int __devinit sky2_reset(struct sky2_hw *hw) { u16 status; - u8 t8, pmd_type; + u8 t8; int i; sky2_write8(hw, B0_CTST, CS_RST_CLR); @@ -2336,9 +2349,7 @@ static int __devinit sky2_reset(struct sky2_hw *hw) sky2_pci_write32(hw, PEX_UNC_ERR_STAT, 0xffffffffUL); - pmd_type = sky2_read8(hw, B2_PMD_TYP); - hw->copper = !(pmd_type == 'L' || pmd_type == 'S'); - + hw->pmd_type = sky2_read8(hw, B2_PMD_TYP); hw->ports = 1; t8 = sky2_read8(hw, B2_Y2_HW_RES); if ((t8 & CFG_DUAL_MAC_MSK) == CFG_DUAL_MAC_MSK) { @@ -2435,21 +2446,22 @@ static int __devinit sky2_reset(struct sky2_hw *hw) static u32 sky2_supported_modes(const struct sky2_hw *hw) { - u32 modes; - if (hw->copper) { - modes = SUPPORTED_10baseT_Half - | SUPPORTED_10baseT_Full - | SUPPORTED_100baseT_Half - | SUPPORTED_100baseT_Full - | SUPPORTED_Autoneg | SUPPORTED_TP; + if (sky2_is_copper(hw)) { + u32 modes = SUPPORTED_10baseT_Half + | SUPPORTED_10baseT_Full + | SUPPORTED_100baseT_Half + | SUPPORTED_100baseT_Full + | SUPPORTED_Autoneg | SUPPORTED_TP; if (hw->chip_id != CHIP_ID_YUKON_FE) modes |= SUPPORTED_1000baseT_Half - | SUPPORTED_1000baseT_Full; + | SUPPORTED_1000baseT_Full; + return modes; } else - modes = SUPPORTED_1000baseT_Full | SUPPORTED_FIBRE - | SUPPORTED_Autoneg; - return modes; + return SUPPORTED_1000baseT_Half + | SUPPORTED_1000baseT_Full + | SUPPORTED_Autoneg + | SUPPORTED_FIBRE; } static int sky2_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) @@ -2460,7 +2472,7 @@ static int sky2_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) ecmd->transceiver = XCVR_INTERNAL; ecmd->supported = sky2_supported_modes(hw); ecmd->phy_address = PHY_ADDR_MARV; - if (hw->copper) { + if (sky2_is_copper(hw)) { ecmd->supported = SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full | SUPPORTED_100baseT_Half @@ -2469,12 +2481,14 @@ static int sky2_get_settings(struct net_device *dev, struct ethtool_cmd *ecmd) | SUPPORTED_1000baseT_Full | SUPPORTED_Autoneg | SUPPORTED_TP; ecmd->port = PORT_TP; - } else + ecmd->speed = sky2->speed; + } else { + ecmd->speed = SPEED_1000; ecmd->port = PORT_FIBRE; + } ecmd->advertising = sky2->advertising; ecmd->autoneg = sky2->autoneg; - ecmd->speed = sky2->speed; ecmd->duplex = sky2->duplex; return 0; } @@ -3187,6 +3201,8 @@ static int __devinit sky2_test_msi(struct sky2_hw *hw) struct pci_dev *pdev = hw->pdev; int err; + init_waitqueue_head (&hw->msi_wait); + sky2_write32(hw, B0_IMSK, Y2_IS_IRQ_SW); err = request_irq(pdev->irq, sky2_test_intr, SA_SHIRQ, DRV_NAME, hw); @@ -3196,10 +3212,8 @@ static int __devinit sky2_test_msi(struct sky2_hw *hw) return err; } - init_waitqueue_head (&hw->msi_wait); - sky2_write8(hw, B0_CTST, CS_ST_SW_IRQ); - wmb(); + sky2_read8(hw, B0_CTST); wait_event_timeout(hw->msi_wait, hw->msi_detected, HZ/10); diff --git a/drivers/net/sky2.h b/drivers/net/sky2.h index 8a0bc5525..9516c1f39 100644 --- a/drivers/net/sky2.h +++ b/drivers/net/sky2.h @@ -1317,6 +1317,14 @@ enum { PHY_M_FESC_SEL_CL_A = 1<<0, /* Select Class A driver (100B-TX) */ }; +/* for Yukon-2 Gigabit Ethernet PHY (88E1112 only) */ +/***** PHY_MARV_PHY_CTRL (page 1) 16 bit r/w Fiber Specific Ctrl *****/ +enum { + PHY_M_FIB_FORCE_LNK = 1<<10,/* Force Link Good */ + PHY_M_FIB_SIGD_POL = 1<<9, /* SIGDET Polarity */ + PHY_M_FIB_TX_DIS = 1<<3, /* Transmitter Disable */ +}; + /* for Yukon-2 Gigabit Ethernet PHY (88E1112 only) */ /***** PHY_MARV_PHY_CTRL (page 2) 16 bit r/w MAC Specific Ctrl *****/ enum { @@ -1566,7 +1574,7 @@ enum { GMR_FS_ANY_ERR = GMR_FS_RX_FF_OV | GMR_FS_CRC_ERR | GMR_FS_FRAGMENT | GMR_FS_LONG_ERR | - GMR_FS_MII_ERR | GMR_FS_BAD_FC | GMR_FS_GOOD_FC | + GMR_FS_MII_ERR | GMR_FS_BAD_FC | GMR_FS_UN_SIZE | GMR_FS_JABBER, }; @@ -1879,7 +1887,7 @@ struct sky2_hw { int pm_cap; u8 chip_id; u8 chip_rev; - u8 copper; + u8 pmd_type; u8 ports; struct sky2_status_le *st_le; @@ -1891,6 +1899,11 @@ struct sky2_hw { wait_queue_head_t msi_wait; }; +static inline int sky2_is_copper(const struct sky2_hw *hw) +{ + return !(hw->pmd_type == 'L' || hw->pmd_type == 'S' || hw->pmd_type == 'P'); +} + /* Register accessor for memory mapped device */ static inline u32 sky2_read32(const struct sky2_hw *hw, unsigned reg) { diff --git a/drivers/net/tg3.c b/drivers/net/tg3.c index 862c226db..09a6b8e6d 100644 --- a/drivers/net/tg3.c +++ b/drivers/net/tg3.c @@ -69,8 +69,8 @@ #define DRV_MODULE_NAME "tg3" #define PFX DRV_MODULE_NAME ": " -#define DRV_MODULE_VERSION "3.59" -#define DRV_MODULE_RELDATE "June 8, 2006" +#define DRV_MODULE_VERSION "3.59.1" +#define DRV_MODULE_RELDATE "August 25, 2006" #define TG3_DEF_MAC_MODE 0 #define TG3_DEF_RX_MODE 0 @@ -3743,7 +3743,7 @@ static int tg3_start_xmit(struct sk_buff *skb, struct net_device *dev) #if TG3_TSO_SUPPORT != 0 mss = 0; if (skb->len > (tp->dev->mtu + ETH_HLEN) && - (mss = skb_shinfo(skb)->tso_size) != 0) { + (mss = skb_shinfo(skb)->gso_size) != 0) { int tcp_opt_len, ip_tcp_len; if (skb_header_cloned(skb) && @@ -3871,7 +3871,7 @@ static int tg3_start_xmit_dma_bug(struct sk_buff *skb, struct net_device *dev) #if TG3_TSO_SUPPORT != 0 mss = 0; if (skb->len > (tp->dev->mtu + ETH_HLEN) && - (mss = skb_shinfo(skb)->tso_size) != 0) { + (mss = skb_shinfo(skb)->gso_size) != 0) { int tcp_opt_len, ip_tcp_len; if (skb_header_cloned(skb) && @@ -11381,11 +11381,15 @@ static int __devinit tg3_init_one(struct pci_dev *pdev, tp->tg3_flags2 |= TG3_FLG2_TSO_CAPABLE; } - /* TSO is on by default on chips that support hardware TSO. + /* TSO is on by default on chips that support HW_TSO_2. + * Some HW_TSO_1 capable chips have bugs that can lead to + * tx timeouts in some cases when TSO is enabled. * Firmware TSO on older chips gives lower performance, so it * is off by default, but can be enabled using ethtool. */ - if (tp->tg3_flags2 & TG3_FLG2_HW_TSO) + if ((tp->tg3_flags2 & TG3_FLG2_HW_TSO_2) || + (GET_ASIC_REV(tp->pci_chip_rev_id) == ASIC_REV_5750 && + tp->pci_chip_rev_id >= CHIPREV_ID_5750_C2)) dev->features |= NETIF_F_TSO; #endif diff --git a/drivers/net/tg3.h b/drivers/net/tg3.h index ff0faab94..cd68f4688 100644 --- a/drivers/net/tg3.h +++ b/drivers/net/tg3.h @@ -125,6 +125,7 @@ #define CHIPREV_ID_5750_A0 0x4000 #define CHIPREV_ID_5750_A1 0x4001 #define CHIPREV_ID_5750_A3 0x4003 +#define CHIPREV_ID_5750_C2 0x4202 #define CHIPREV_ID_5752_A0_HW 0x5000 #define CHIPREV_ID_5752_A0 0x6000 #define CHIPREV_ID_5752_A1 0x6001 diff --git a/drivers/net/tulip/winbond-840.c b/drivers/net/tulip/winbond-840.c index 136a70c4d..56d86c7c0 100644 --- a/drivers/net/tulip/winbond-840.c +++ b/drivers/net/tulip/winbond-840.c @@ -1605,11 +1605,11 @@ static void __devexit w840_remove1 (struct pci_dev *pdev) * - get_stats: * spin_lock_irq(np->lock), doesn't touch hw if not present * - hard_start_xmit: - * netif_stop_queue + spin_unlock_wait(&dev->xmit_lock); + * synchronize_irq + netif_tx_disable; * - tx_timeout: - * netif_device_detach + spin_unlock_wait(&dev->xmit_lock); + * netif_device_detach + netif_tx_disable; * - set_multicast_list - * netif_device_detach + spin_unlock_wait(&dev->xmit_lock); + * netif_device_detach + netif_tx_disable; * - interrupt handler * doesn't touch hw if not present, synchronize_irq waits for * running instances of the interrupt handler. @@ -1635,11 +1635,10 @@ static int w840_suspend (struct pci_dev *pdev, pm_message_t state) netif_device_detach(dev); update_csr6(dev, 0); iowrite32(0, ioaddr + IntrEnable); - netif_stop_queue(dev); spin_unlock_irq(&np->lock); - spin_unlock_wait(&dev->xmit_lock); synchronize_irq(dev->irq); + netif_tx_disable(dev); np->stats.rx_missed_errors += ioread32(ioaddr + RxMissed) & 0xffff; diff --git a/drivers/net/typhoon.c b/drivers/net/typhoon.c index d9258d420..389d19fb5 100644 --- a/drivers/net/typhoon.c +++ b/drivers/net/typhoon.c @@ -340,7 +340,7 @@ enum state_values { #endif #if defined(NETIF_F_TSO) -#define skb_tso_size(x) (skb_shinfo(x)->tso_size) +#define skb_tso_size(x) (skb_shinfo(x)->gso_size) #define TSO_NUM_DESCRIPTORS 2 #define TSO_OFFLOAD_ON TYPHOON_OFFLOAD_TCP_SEGMENT #else @@ -805,7 +805,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) * If problems develop with TSO, check this first. */ numDesc = skb_shinfo(skb)->nr_frags + 1; - if(skb_tso_size(skb)) + if (skb_is_gso(skb)) numDesc++; /* When checking for free space in the ring, we need to also @@ -845,7 +845,7 @@ typhoon_start_tx(struct sk_buff *skb, struct net_device *dev) TYPHOON_TX_PF_VLAN_TAG_SHIFT); } - if(skb_tso_size(skb)) { + if (skb_is_gso(skb)) { first_txd->processFlags |= TYPHOON_TX_PF_TCP_SEGMENT; first_txd->numDesc++; diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c index 15e710283..09e05fe40 100644 --- a/drivers/net/via-velocity.c +++ b/drivers/net/via-velocity.c @@ -61,7 +61,6 @@ #include #include #include -#include #include #include #include @@ -249,6 +248,7 @@ static void velocity_free_rd_ring(struct velocity_info *vptr); static void velocity_free_tx_buf(struct velocity_info *vptr, struct velocity_td_info *); static int velocity_soft_reset(struct velocity_info *vptr); static void mii_init(struct velocity_info *vptr, u32 mii_status); +static u32 velocity_get_link(struct net_device *dev); static u32 velocity_get_opt_media_mode(struct velocity_info *vptr); static void velocity_print_link_status(struct velocity_info *vptr); static void safe_disable_mii_autopoll(struct mac_regs __iomem * regs); @@ -792,13 +792,16 @@ static int __devinit velocity_found1(struct pci_dev *pdev, const struct pci_devi #endif if (vptr->flags & VELOCITY_FLAGS_TX_CSUM) { - dev->features |= NETIF_F_HW_CSUM; + dev->features |= NETIF_F_IP_CSUM; } ret = register_netdev(dev); if (ret < 0) goto err_iounmap; + if (velocity_get_link(dev)) + netif_carrier_off(dev); + velocity_print_info(vptr); pci_set_drvdata(pdev, dev); @@ -1107,6 +1110,9 @@ static void velocity_free_rd_ring(struct velocity_info *vptr) for (i = 0; i < vptr->options.numrx; i++) { struct velocity_rd_info *rd_info = &(vptr->rd_info[i]); + struct rx_desc *rd = vptr->rd_ring + i; + + memset(rd, 0, sizeof(*rd)); if (!rd_info->skb) continue; @@ -1212,10 +1218,8 @@ static void velocity_free_td_ring(struct velocity_info *vptr) velocity_free_td_ring_entry(vptr, j, i); } - if (vptr->td_infos[j]) { - kfree(vptr->td_infos[j]); - vptr->td_infos[j] = NULL; - } + kfree(vptr->td_infos[j]); + vptr->td_infos[j] = NULL; } } @@ -1335,7 +1339,7 @@ static inline int velocity_rx_copy(struct sk_buff **rx_skb, int pkt_size, if (vptr->flags & VELOCITY_FLAGS_IP_ALIGN) skb_reserve(new_skb, 2); - memcpy(new_skb->data, rx_skb[0]->tail, pkt_size); + memcpy(new_skb->data, rx_skb[0]->data, pkt_size); *rx_skb = new_skb; ret = 0; } @@ -1456,9 +1460,9 @@ static int velocity_alloc_rx_buf(struct velocity_info *vptr, int idx) * Do the gymnastics to get the buffer head for data at * 64byte alignment. */ - skb_reserve(rd_info->skb, (unsigned long) rd_info->skb->tail & 63); + skb_reserve(rd_info->skb, (unsigned long) rd_info->skb->data & 63); rd_info->skb->dev = vptr->dev; - rd_info->skb_dma = pci_map_single(vptr->pdev, rd_info->skb->tail, vptr->rx_buf_sz, PCI_DMA_FROMDEVICE); + rd_info->skb_dma = pci_map_single(vptr->pdev, rd_info->skb->data, vptr->rx_buf_sz, PCI_DMA_FROMDEVICE); /* * Fill in the descriptor to match @@ -1653,8 +1657,10 @@ static void velocity_error(struct velocity_info *vptr, int status) if (linked) { vptr->mii_status &= ~VELOCITY_LINK_FAIL; + netif_carrier_on(vptr->dev); } else { vptr->mii_status |= VELOCITY_LINK_FAIL; + netif_carrier_off(vptr->dev); } velocity_print_link_status(vptr); @@ -1899,6 +1905,13 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev) int pktlen = skb->len; +#ifdef VELOCITY_ZERO_COPY_SUPPORT + if (skb_shinfo(skb)->nr_frags > 6 && __skb_linearize(skb)) { + kfree_skb(skb); + return 0; + } +#endif + spin_lock_irqsave(&vptr->lock, flags); index = vptr->td_curr[qnum]; @@ -1914,8 +1927,6 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev) */ if (pktlen < ETH_ZLEN) { /* Cannot occur until ZC support */ - if(skb_linearize(skb, GFP_ATOMIC)) - return 0; pktlen = ETH_ZLEN; memcpy(tdinfo->buf, skb->data, skb->len); memset(tdinfo->buf + skb->len, 0, ETH_ZLEN - skb->len); @@ -1933,7 +1944,6 @@ static int velocity_xmit(struct sk_buff *skb, struct net_device *dev) int nfrags = skb_shinfo(skb)->nr_frags; tdinfo->skb = skb; if (nfrags > 6) { - skb_linearize(skb, GFP_ATOMIC); memcpy(tdinfo->buf, skb->data, skb->len); tdinfo->skb_dma[0] = tdinfo->buf_dma; td_ptr->tdesc0.pktsize = diff --git a/drivers/net/wireless/orinoco.c b/drivers/net/wireless/orinoco.c index c2d0b09e0..a5fcfcde6 100644 --- a/drivers/net/wireless/orinoco.c +++ b/drivers/net/wireless/orinoco.c @@ -1833,7 +1833,9 @@ static int __orinoco_program_rids(struct net_device *dev) /* Set promiscuity / multicast*/ priv->promiscuous = 0; priv->mc_count = 0; - __orinoco_set_multicast_list(dev); /* FIXME: what about the xmit_lock */ + + /* FIXME: what about netif_tx_lock */ + __orinoco_set_multicast_list(dev); return 0; } diff --git a/drivers/net/wireless/spectrum_cs.c b/drivers/net/wireless/spectrum_cs.c index f7b77ce54..54fe4e4ea 100644 --- a/drivers/net/wireless/spectrum_cs.c +++ b/drivers/net/wireless/spectrum_cs.c @@ -245,7 +245,7 @@ spectrum_reset(struct pcmcia_device *link, int idle) u_int save_cor; /* Doing it if hardware is gone is guaranteed crash */ - if (pcmcia_dev_present(link)) + if (!pcmcia_dev_present(link)) return -ENODEV; /* Save original COR value */ diff --git a/drivers/oprofile/buffer_sync.c b/drivers/oprofile/buffer_sync.c index b2e8e49c8..a647d3982 100644 --- a/drivers/oprofile/buffer_sync.c +++ b/drivers/oprofile/buffer_sync.c @@ -6,6 +6,10 @@ * * @author John Levon * + * Modified by Aravind Menon for Xen + * These modifications are: + * Copyright (C) 2005 Hewlett-Packard Co. + * * This is the core of the buffer management. Each * CPU buffer is processed and entered into the * global event buffer. Such processing is necessary @@ -275,15 +279,31 @@ static void add_cpu_switch(int i) last_cookie = INVALID_COOKIE; } -static void add_kernel_ctx_switch(unsigned int in_kernel) +static void add_cpu_mode_switch(unsigned int cpu_mode) { add_event_entry(ESCAPE_CODE); - if (in_kernel) - add_event_entry(KERNEL_ENTER_SWITCH_CODE); - else - add_event_entry(KERNEL_EXIT_SWITCH_CODE); + switch (cpu_mode) { + case CPU_MODE_USER: + add_event_entry(USER_ENTER_SWITCH_CODE); + break; + case CPU_MODE_KERNEL: + add_event_entry(KERNEL_ENTER_SWITCH_CODE); + break; + case CPU_MODE_XEN: + add_event_entry(XEN_ENTER_SWITCH_CODE); + break; + default: + break; + } } - + +static void add_domain_switch(unsigned long domain_id) +{ + add_event_entry(ESCAPE_CODE); + add_event_entry(DOMAIN_SWITCH_CODE); + add_event_entry(domain_id); +} + static void add_user_ctx_switch(struct task_struct const * task, unsigned long cookie) { @@ -348,9 +368,9 @@ static int add_us_sample(struct mm_struct * mm, struct op_sample * s) * for later lookup from userspace. */ static int -add_sample(struct mm_struct * mm, struct op_sample * s, int in_kernel) +add_sample(struct mm_struct * mm, struct op_sample * s, int cpu_mode) { - if (in_kernel) { + if (cpu_mode >= CPU_MODE_KERNEL) { add_sample_entry(s->eip, s->event); return 1; } else if (mm) { @@ -496,10 +516,11 @@ void sync_buffer(int cpu) struct mm_struct *mm = NULL; struct task_struct * new; unsigned long cookie = 0; - int in_kernel = 1; + int cpu_mode = 1; unsigned int i; sync_buffer_state state = sb_buffer_start; unsigned long available; + int domain_switch = 0; down(&buffer_sem); @@ -512,16 +533,18 @@ void sync_buffer(int cpu) for (i = 0; i < available; ++i) { struct op_sample * s = &cpu_buf->buffer[cpu_buf->tail_pos]; - if (is_code(s->eip)) { - if (s->event <= CPU_IS_KERNEL) { - /* kernel/userspace switch */ - in_kernel = s->event; + if (is_code(s->eip) && !domain_switch) { + if (s->event <= CPU_MODE_XEN) { + /* xen/kernel/userspace switch */ + cpu_mode = s->event; if (state == sb_buffer_start) state = sb_sample_start; - add_kernel_ctx_switch(s->event); + add_cpu_mode_switch(s->event); } else if (s->event == CPU_TRACE_BEGIN) { state = sb_bt_start; add_trace_begin(); + } else if (s->event == CPU_DOMAIN_SWITCH) { + domain_switch = 1; } else { struct mm_struct * oldmm = mm; @@ -535,11 +558,16 @@ void sync_buffer(int cpu) add_user_ctx_switch(new, cookie); } } else { - if (state >= sb_bt_start && - !add_sample(mm, s, in_kernel)) { - if (state == sb_bt_start) { - state = sb_bt_ignore; - atomic_inc(&oprofile_stats.bt_lost_no_mapping); + if (domain_switch) { + add_domain_switch(s->eip); + domain_switch = 0; + } else { + if (state >= sb_bt_start && + !add_sample(mm, s, cpu_mode)) { + if (state == sb_bt_start) { + state = sb_bt_ignore; + atomic_inc(&oprofile_stats.bt_lost_no_mapping); + } } } } diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c index fc4bc9b94..a59878e14 100644 --- a/drivers/oprofile/cpu_buffer.c +++ b/drivers/oprofile/cpu_buffer.c @@ -6,6 +6,10 @@ * * @author John Levon * + * Modified by Aravind Menon for Xen + * These modifications are: + * Copyright (C) 2005 Hewlett-Packard Co. + * * Each CPU has a local buffer that stores PC value/event * pairs. We also log context switches when we notice them. * Eventually each CPU's buffer is processed into the global @@ -34,6 +38,8 @@ static void wq_sync_buffer(void *); #define DEFAULT_TIMER_EXPIRE (HZ / 10) static int work_enabled; +static int32_t current_domain = COORDINATOR_DOMAIN; + void free_cpu_buffers(void) { int i; @@ -57,7 +63,7 @@ int alloc_cpu_buffers(void) goto fail; b->last_task = NULL; - b->last_is_kernel = -1; + b->last_cpu_mode = -1; b->tracing = 0; b->buffer_size = buffer_size; b->tail_pos = 0; @@ -113,7 +119,7 @@ void cpu_buffer_reset(struct oprofile_cpu_buffer * cpu_buf) * collected will populate the buffer with proper * values to initialize the buffer */ - cpu_buf->last_is_kernel = -1; + cpu_buf->last_cpu_mode = -1; cpu_buf->last_task = NULL; } @@ -163,13 +169,13 @@ add_code(struct oprofile_cpu_buffer * buffer, unsigned long value) * because of the head/tail separation of the writer and reader * of the CPU buffer. * - * is_kernel is needed because on some architectures you cannot + * cpu_mode is needed because on some architectures you cannot * tell if you are in kernel or user space simply by looking at - * pc. We tag this in the buffer by generating kernel enter/exit - * events whenever is_kernel changes + * pc. We tag this in the buffer by generating kernel/user (and xen) + * enter events whenever cpu_mode changes */ static int log_sample(struct oprofile_cpu_buffer * cpu_buf, unsigned long pc, - int is_kernel, unsigned long event) + int cpu_mode, unsigned long event) { struct task_struct * task; @@ -180,18 +186,20 @@ static int log_sample(struct oprofile_cpu_buffer * cpu_buf, unsigned long pc, return 0; } - is_kernel = !!is_kernel; + WARN_ON(cpu_mode > CPU_MODE_XEN); task = current; /* notice a switch from user->kernel or vice versa */ - if (cpu_buf->last_is_kernel != is_kernel) { - cpu_buf->last_is_kernel = is_kernel; - add_code(cpu_buf, is_kernel); + if (cpu_buf->last_cpu_mode != cpu_mode) { + cpu_buf->last_cpu_mode = cpu_mode; + add_code(cpu_buf, cpu_mode); } - + /* notice a task switch */ - if (cpu_buf->last_task != task) { + /* if not processing other domain samples */ + if ((cpu_buf->last_task != task) && + (current_domain == COORDINATOR_DOMAIN)) { cpu_buf->last_task = task; add_code(cpu_buf, (unsigned long)task); } @@ -275,6 +283,25 @@ void oprofile_add_trace(unsigned long pc) add_sample(cpu_buf, pc, 0); } +int oprofile_add_domain_switch(int32_t domain_id) +{ + struct oprofile_cpu_buffer * cpu_buf = &cpu_buffer[smp_processor_id()]; + + /* should have space for switching into and out of domain + (2 slots each) plus one sample and one cpu mode switch */ + if (((nr_available_slots(cpu_buf) < 6) && + (domain_id != COORDINATOR_DOMAIN)) || + (nr_available_slots(cpu_buf) < 2)) + return 0; + + add_code(cpu_buf, CPU_DOMAIN_SWITCH); + add_sample(cpu_buf, domain_id, 0); + + current_domain = domain_id; + + return 1; +} + /* * This serves to avoid cpu buffer overflow, and makes sure * the task mortuary progresses diff --git a/drivers/oprofile/cpu_buffer.h b/drivers/oprofile/cpu_buffer.h index 09abb80e0..cd94735be 100644 --- a/drivers/oprofile/cpu_buffer.h +++ b/drivers/oprofile/cpu_buffer.h @@ -36,7 +36,7 @@ struct oprofile_cpu_buffer { volatile unsigned long tail_pos; unsigned long buffer_size; struct task_struct * last_task; - int last_is_kernel; + int last_cpu_mode; int tracing; struct op_sample * buffer; unsigned long sample_received; @@ -51,7 +51,10 @@ extern struct oprofile_cpu_buffer cpu_buffer[]; void cpu_buffer_reset(struct oprofile_cpu_buffer * cpu_buf); /* transient events for the CPU buffer -> event buffer */ -#define CPU_IS_KERNEL 1 -#define CPU_TRACE_BEGIN 2 +#define CPU_MODE_USER 0 +#define CPU_MODE_KERNEL 1 +#define CPU_MODE_XEN 2 +#define CPU_TRACE_BEGIN 3 +#define CPU_DOMAIN_SWITCH 4 #endif /* OPROFILE_CPU_BUFFER_H */ diff --git a/drivers/oprofile/event_buffer.h b/drivers/oprofile/event_buffer.h index 018023630..0c33ee71e 100644 --- a/drivers/oprofile/event_buffer.h +++ b/drivers/oprofile/event_buffer.h @@ -29,15 +29,20 @@ void wake_up_buffer_waiter(void); #define CPU_SWITCH_CODE 2 #define COOKIE_SWITCH_CODE 3 #define KERNEL_ENTER_SWITCH_CODE 4 -#define KERNEL_EXIT_SWITCH_CODE 5 +#define USER_ENTER_SWITCH_CODE 5 #define MODULE_LOADED_CODE 6 #define CTX_TGID_CODE 7 #define TRACE_BEGIN_CODE 8 #define TRACE_END_CODE 9 +#define XEN_ENTER_SWITCH_CODE 10 +#define DOMAIN_SWITCH_CODE 11 #define INVALID_COOKIE ~0UL #define NO_COOKIE 0UL +/* Constant used to refer to coordinator domain (Xen) */ +#define COORDINATOR_DOMAIN -1 + /* add data to the event buffer */ void add_event_entry(unsigned long data); diff --git a/drivers/oprofile/oprof.c b/drivers/oprofile/oprof.c index b3f1cd6a2..76bac8dfe 100644 --- a/drivers/oprofile/oprof.c +++ b/drivers/oprofile/oprof.c @@ -5,6 +5,10 @@ * @remark Read the file COPYING * * @author John Levon + * + * Modified by Aravind Menon for Xen + * These modifications are: + * Copyright (C) 2005 Hewlett-Packard Co. */ #include @@ -19,7 +23,7 @@ #include "cpu_buffer.h" #include "buffer_sync.h" #include "oprofile_stats.h" - + struct oprofile_operations oprofile_ops; unsigned long oprofile_started; @@ -33,6 +37,34 @@ static DECLARE_MUTEX(start_sem); */ static int timer = 0; +#ifdef CONFIG_XEN +int oprofile_set_active(int active_domains[], unsigned int adomains) +{ + int err; + + if (!oprofile_ops.set_active) + return -EINVAL; + + down(&start_sem); + err = oprofile_ops.set_active(active_domains, adomains); + up(&start_sem); + return err; +} + +int oprofile_set_passive(int passive_domains[], unsigned int pdomains) +{ + int err; + + if (!oprofile_ops.set_passive) + return -EINVAL; + + down(&start_sem); + err = oprofile_ops.set_passive(passive_domains, pdomains); + up(&start_sem); + return err; +} +#endif + int oprofile_setup(void) { int err; diff --git a/drivers/oprofile/oprof.h b/drivers/oprofile/oprof.h index 183236508..587db2b0a 100644 --- a/drivers/oprofile/oprof.h +++ b/drivers/oprofile/oprof.h @@ -35,5 +35,10 @@ void oprofile_create_files(struct super_block * sb, struct dentry * root); void oprofile_timer_init(struct oprofile_operations * ops); int oprofile_set_backtrace(unsigned long depth); - + +#ifdef CONFIG_XEN +int oprofile_set_active(int active_domains[], unsigned int adomains); +int oprofile_set_passive(int passive_domains[], unsigned int pdomains); +#endif + #endif /* OPROF_H */ diff --git a/drivers/oprofile/oprofile_files.c b/drivers/oprofile/oprofile_files.c index a72006c08..388846526 100644 --- a/drivers/oprofile/oprofile_files.c +++ b/drivers/oprofile/oprofile_files.c @@ -5,15 +5,21 @@ * @remark Read the file COPYING * * @author John Levon + * + * Modified by Aravind Menon for Xen + * These modifications are: + * Copyright (C) 2005 Hewlett-Packard Co. */ #include #include +#include +#include #include "event_buffer.h" #include "oprofile_stats.h" #include "oprof.h" - + unsigned long fs_buffer_size = 131072; unsigned long fs_cpu_buffer_size = 8192; unsigned long fs_buffer_watershed = 32768; /* FIXME: tune */ @@ -117,11 +123,206 @@ static ssize_t dump_write(struct file * file, char const __user * buf, size_t co static struct file_operations dump_fops = { .write = dump_write, }; - + +#define TMPBUFSIZE 512 + +#ifdef CONFIG_XEN +static unsigned int adomains = 0; +static int active_domains[MAX_OPROF_DOMAINS + 1]; +static DEFINE_MUTEX(adom_mutex); + +static ssize_t adomain_write(struct file * file, char const __user * buf, + size_t count, loff_t * offset) +{ + char *tmpbuf; + char *startp, *endp; + int i; + unsigned long val; + ssize_t retval = count; + + if (*offset) + return -EINVAL; + if (count > TMPBUFSIZE - 1) + return -EINVAL; + + if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL))) + return -ENOMEM; + + if (copy_from_user(tmpbuf, buf, count)) { + kfree(tmpbuf); + return -EFAULT; + } + tmpbuf[count] = 0; + + mutex_lock(&adom_mutex); + + startp = tmpbuf; + /* Parse one more than MAX_OPROF_DOMAINS, for easy error checking */ + for (i = 0; i <= MAX_OPROF_DOMAINS; i++) { + val = simple_strtoul(startp, &endp, 0); + if (endp == startp) + break; + while (ispunct(*endp) || isspace(*endp)) + endp++; + active_domains[i] = val; + if (active_domains[i] != val) + /* Overflow, force error below */ + i = MAX_OPROF_DOMAINS + 1; + startp = endp; + } + /* Force error on trailing junk */ + adomains = *startp ? MAX_OPROF_DOMAINS + 1 : i; + + kfree(tmpbuf); + + if (adomains > MAX_OPROF_DOMAINS + || oprofile_set_active(active_domains, adomains)) { + adomains = 0; + retval = -EINVAL; + } + + mutex_unlock(&adom_mutex); + return retval; +} + +static ssize_t adomain_read(struct file * file, char __user * buf, + size_t count, loff_t * offset) +{ + char * tmpbuf; + size_t len; + int i; + ssize_t retval; + + if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL))) + return -ENOMEM; + + mutex_lock(&adom_mutex); + + len = 0; + for (i = 0; i < adomains; i++) + len += snprintf(tmpbuf + len, + len < TMPBUFSIZE ? TMPBUFSIZE - len : 0, + "%u ", active_domains[i]); + WARN_ON(len > TMPBUFSIZE); + if (len != 0 && len <= TMPBUFSIZE) + tmpbuf[len-1] = '\n'; + + mutex_unlock(&adom_mutex); + + retval = simple_read_from_buffer(buf, count, offset, tmpbuf, len); + + kfree(tmpbuf); + return retval; +} + + +static struct file_operations active_domain_ops = { + .read = adomain_read, + .write = adomain_write, +}; + +static unsigned int pdomains = 0; +static int passive_domains[MAX_OPROF_DOMAINS]; +static DEFINE_MUTEX(pdom_mutex); + +static ssize_t pdomain_write(struct file * file, char const __user * buf, + size_t count, loff_t * offset) +{ + char *tmpbuf; + char *startp, *endp; + int i; + unsigned long val; + ssize_t retval = count; + + if (*offset) + return -EINVAL; + if (count > TMPBUFSIZE - 1) + return -EINVAL; + + if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL))) + return -ENOMEM; + + if (copy_from_user(tmpbuf, buf, count)) { + kfree(tmpbuf); + return -EFAULT; + } + tmpbuf[count] = 0; + + mutex_lock(&pdom_mutex); + + startp = tmpbuf; + /* Parse one more than MAX_OPROF_DOMAINS, for easy error checking */ + for (i = 0; i <= MAX_OPROF_DOMAINS; i++) { + val = simple_strtoul(startp, &endp, 0); + if (endp == startp) + break; + while (ispunct(*endp) || isspace(*endp)) + endp++; + passive_domains[i] = val; + if (passive_domains[i] != val) + /* Overflow, force error below */ + i = MAX_OPROF_DOMAINS + 1; + startp = endp; + } + /* Force error on trailing junk */ + pdomains = *startp ? MAX_OPROF_DOMAINS + 1 : i; + + kfree(tmpbuf); + + if (pdomains > MAX_OPROF_DOMAINS + || oprofile_set_passive(passive_domains, pdomains)) { + pdomains = 0; + retval = -EINVAL; + } + + mutex_unlock(&pdom_mutex); + return retval; +} + +static ssize_t pdomain_read(struct file * file, char __user * buf, + size_t count, loff_t * offset) +{ + char * tmpbuf; + size_t len; + int i; + ssize_t retval; + + if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL))) + return -ENOMEM; + + mutex_lock(&pdom_mutex); + + len = 0; + for (i = 0; i < pdomains; i++) + len += snprintf(tmpbuf + len, + len < TMPBUFSIZE ? TMPBUFSIZE - len : 0, + "%u ", passive_domains[i]); + WARN_ON(len > TMPBUFSIZE); + if (len != 0 && len <= TMPBUFSIZE) + tmpbuf[len-1] = '\n'; + + mutex_unlock(&pdom_mutex); + + retval = simple_read_from_buffer(buf, count, offset, tmpbuf, len); + + kfree(tmpbuf); + return retval; +} + +static struct file_operations passive_domain_ops = { + .read = pdomain_read, + .write = pdomain_write, +}; +#endif /* CONFIG_XEN */ + void oprofile_create_files(struct super_block * sb, struct dentry * root) { oprofilefs_create_file(sb, root, "enable", &enable_fops); oprofilefs_create_file_perm(sb, root, "dump", &dump_fops, 0666); +#ifdef CONFIG_XEN + oprofilefs_create_file(sb, root, "active_domains", &active_domain_ops); + oprofilefs_create_file(sb, root, "passive_domains", &passive_domain_ops); +#endif oprofilefs_create_file(sb, root, "buffer", &event_buffer_fops); oprofilefs_create_ulong(sb, root, "buffer_size", &fs_buffer_size); oprofilefs_create_ulong(sb, root, "buffer_watershed", &fs_buffer_watershed); diff --git a/drivers/pci/Kconfig b/drivers/pci/Kconfig index 4d762fc48..065ea43aa 100644 --- a/drivers/pci/Kconfig +++ b/drivers/pci/Kconfig @@ -5,6 +5,7 @@ config PCI_MSI bool "Message Signaled Interrupts (MSI and MSI-X)" depends on PCI depends on (X86_LOCAL_APIC && X86_IO_APIC) || IA64 + depends on !XEN help This allows device drivers to enable MSI (Message Signaled Interrupts). Message Signaled Interrupts enable a device to diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c index d37847861..6e3786f4d 100644 --- a/drivers/pci/quirks.c +++ b/drivers/pci/quirks.c @@ -427,6 +427,7 @@ static void __devinit quirk_ich6_lpc_acpi(struct pci_dev *dev) pci_read_config_dword(dev, 0x48, ®ion); quirk_io_region(dev, region, 64, PCI_BRIDGE_RESOURCES+1, "ICH6 GPIO"); } +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_0, quirk_ich6_lpc_acpi ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1, quirk_ich6_lpc_acpi ); /* @@ -1043,7 +1044,6 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801BA_0, asu DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801CA_12, asus_hides_smbus_lpc ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801DB_12, asus_hides_smbus_lpc ); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801EB_0, asus_hides_smbus_lpc ); -DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH6_1, asus_hides_smbus_lpc ); static void __init asus_hides_smbus_lpc_ich6(struct pci_dev *dev) { diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c index 6ded52716..688421de9 100644 --- a/drivers/pnp/resource.c +++ b/drivers/pnp/resource.c @@ -396,7 +396,8 @@ int pnp_check_irq(struct pnp_dev * dev, int idx) /* check if the resource is already in use, skip if the * device is active because it itself may be in use */ if(!dev->active) { - if (request_irq(*irq, pnp_test_handler, SA_INTERRUPT, "pnp", NULL)) + if (request_irq(*irq, pnp_test_handler, + SA_INTERRUPT|SA_PROBEIRQ, "pnp", NULL)) return 0; free_irq(*irq, NULL); } diff --git a/drivers/s390/net/qeth_eddp.c b/drivers/s390/net/qeth_eddp.c index f94f1f25e..38aad8321 100644 --- a/drivers/s390/net/qeth_eddp.c +++ b/drivers/s390/net/qeth_eddp.c @@ -1,6 +1,5 @@ /* - * - * linux/drivers/s390/net/qeth_eddp.c ($Revision: 1.13 $) + * linux/drivers/s390/net/qeth_eddp.c * * Enhanced Device Driver Packing (EDDP) support for the qeth driver. * @@ -8,8 +7,6 @@ * * Author(s): Thomas Spatzier * - * $Revision: 1.13 $ $Date: 2005/05/04 20:19:18 $ - * */ #include #include @@ -62,8 +59,7 @@ qeth_eddp_free_context(struct qeth_eddp_context *ctx) for (i = 0; i < ctx->num_pages; ++i) free_page((unsigned long)ctx->pages[i]); kfree(ctx->pages); - if (ctx->elements != NULL) - kfree(ctx->elements); + kfree(ctx->elements); kfree(ctx); } @@ -85,7 +81,7 @@ void qeth_eddp_buf_release_contexts(struct qeth_qdio_out_buffer *buf) { struct qeth_eddp_context_reference *ref; - + QETH_DBF_TEXT(trace, 6, "eddprctx"); while (!list_empty(&buf->ctx_list)){ ref = list_entry(buf->ctx_list.next, @@ -139,7 +135,7 @@ qeth_eddp_fill_buffer(struct qeth_qdio_out_q *queue, "buffer!\n"); goto out; } - } + } /* check if the whole next skb fits into current buffer */ if ((QETH_MAX_BUFFER_ELEMENTS(queue->card) - buf->next_element_to_fill) @@ -152,7 +148,7 @@ qeth_eddp_fill_buffer(struct qeth_qdio_out_q *queue, * and increment ctx's refcnt */ must_refcnt = 1; continue; - } + } if (must_refcnt){ must_refcnt = 0; if (qeth_eddp_buf_ref_context(buf, ctx)){ @@ -270,7 +266,7 @@ qeth_eddp_copy_data_tcp(char *dst, struct qeth_eddp_data *eddp, int len, int left_in_frag; int copy_len; u8 *src; - + QETH_DBF_TEXT(trace, 5, "eddpcdtc"); if (skb_shinfo(eddp->skb)->nr_frags == 0) { memcpy(dst, eddp->skb->data + eddp->skb_offset, len); @@ -393,9 +389,8 @@ qeth_eddp_create_eddp_data(struct qeth_hdr *qh, u8 *nh, u8 nhl, u8 *th, u8 thl) struct qeth_eddp_data *eddp; QETH_DBF_TEXT(trace, 5, "eddpcrda"); - eddp = kmalloc(sizeof(struct qeth_eddp_data), GFP_ATOMIC); + eddp = kzalloc(sizeof(struct qeth_eddp_data), GFP_ATOMIC); if (eddp){ - memset(eddp, 0, sizeof(struct qeth_eddp_data)); eddp->nhl = nhl; eddp->thl = thl; memcpy(&eddp->qh, qh, sizeof(struct qeth_hdr)); @@ -413,12 +408,19 @@ __qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx, struct tcphdr *tcph; int data_len; u32 hcsum; - + QETH_DBF_TEXT(trace, 5, "eddpftcp"); eddp->skb_offset = sizeof(struct qeth_hdr) + eddp->nhl + eddp->thl; + if (eddp->qh.hdr.l2.id == QETH_HEADER_TYPE_LAYER2) { + eddp->skb_offset += sizeof(struct ethhdr); +#ifdef CONFIG_QETH_VLAN + if (eddp->mac.h_proto == __constant_htons(ETH_P_8021Q)) + eddp->skb_offset += VLAN_HLEN; +#endif /* CONFIG_QETH_VLAN */ + } tcph = eddp->skb->h.th; while (eddp->skb_offset < eddp->skb->len) { - data_len = min((int)skb_shinfo(eddp->skb)->tso_size, + data_len = min((int)skb_shinfo(eddp->skb)->gso_size, (int)(eddp->skb->len - eddp->skb_offset)); /* prepare qdio hdr */ if (eddp->qh.hdr.l2.id == QETH_HEADER_TYPE_LAYER2){ @@ -463,13 +465,13 @@ __qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx, eddp->th.tcp.h.seq += data_len; } } - + static inline int qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx, struct sk_buff *skb, struct qeth_hdr *qhdr) { struct qeth_eddp_data *eddp = NULL; - + QETH_DBF_TEXT(trace, 5, "eddpficx"); /* create our segmentation headers and copy original headers */ if (skb->protocol == ETH_P_IP) @@ -486,6 +488,7 @@ qeth_eddp_fill_context_tcp(struct qeth_eddp_context *ctx, return -ENOMEM; } if (qhdr->hdr.l2.id == QETH_HEADER_TYPE_LAYER2) { + skb->mac.raw = (skb->data) + sizeof(struct qeth_hdr); memcpy(&eddp->mac, eth_hdr(skb), ETH_HLEN); #ifdef CONFIG_QETH_VLAN if (eddp->mac.h_proto == __constant_htons(ETH_P_8021Q)) { @@ -509,23 +512,23 @@ qeth_eddp_calc_num_pages(struct qeth_eddp_context *ctx, struct sk_buff *skb, int hdr_len) { int skbs_per_page; - + QETH_DBF_TEXT(trace, 5, "eddpcanp"); /* can we put multiple skbs in one page? */ - skbs_per_page = PAGE_SIZE / (skb_shinfo(skb)->tso_size + hdr_len); + skbs_per_page = PAGE_SIZE / (skb_shinfo(skb)->gso_size + hdr_len); if (skbs_per_page > 1){ - ctx->num_pages = (skb_shinfo(skb)->tso_segs + 1) / + ctx->num_pages = (skb_shinfo(skb)->gso_segs + 1) / skbs_per_page + 1; ctx->elements_per_skb = 1; } else { /* no -> how many elements per skb? */ - ctx->elements_per_skb = (skb_shinfo(skb)->tso_size + hdr_len + + ctx->elements_per_skb = (skb_shinfo(skb)->gso_size + hdr_len + PAGE_SIZE) >> PAGE_SHIFT; ctx->num_pages = ctx->elements_per_skb * - (skb_shinfo(skb)->tso_segs + 1); + (skb_shinfo(skb)->gso_segs + 1); } ctx->num_elements = ctx->elements_per_skb * - (skb_shinfo(skb)->tso_segs + 1); + (skb_shinfo(skb)->gso_segs + 1); } static inline struct qeth_eddp_context * @@ -538,12 +541,11 @@ qeth_eddp_create_context_generic(struct qeth_card *card, struct sk_buff *skb, QETH_DBF_TEXT(trace, 5, "creddpcg"); /* create the context and allocate pages */ - ctx = kmalloc(sizeof(struct qeth_eddp_context), GFP_ATOMIC); + ctx = kzalloc(sizeof(struct qeth_eddp_context), GFP_ATOMIC); if (ctx == NULL){ QETH_DBF_TEXT(trace, 2, "ceddpcn1"); return NULL; } - memset(ctx, 0, sizeof(struct qeth_eddp_context)); ctx->type = QETH_LARGE_SEND_EDDP; qeth_eddp_calc_num_pages(ctx, skb, hdr_len); if (ctx->elements_per_skb > QETH_MAX_BUFFER_ELEMENTS(card)){ @@ -551,13 +553,12 @@ qeth_eddp_create_context_generic(struct qeth_card *card, struct sk_buff *skb, kfree(ctx); return NULL; } - ctx->pages = kmalloc(ctx->num_pages * sizeof(u8 *), GFP_ATOMIC); + ctx->pages = kcalloc(ctx->num_pages, sizeof(u8 *), GFP_ATOMIC); if (ctx->pages == NULL){ QETH_DBF_TEXT(trace, 2, "ceddpcn2"); kfree(ctx); return NULL; } - memset(ctx->pages, 0, ctx->num_pages * sizeof(u8 *)); for (i = 0; i < ctx->num_pages; ++i){ addr = (u8 *)__get_free_page(GFP_ATOMIC); if (addr == NULL){ @@ -569,15 +570,13 @@ qeth_eddp_create_context_generic(struct qeth_card *card, struct sk_buff *skb, memset(addr, 0, PAGE_SIZE); ctx->pages[i] = addr; } - ctx->elements = kmalloc(ctx->num_elements * + ctx->elements = kcalloc(ctx->num_elements, sizeof(struct qeth_eddp_element), GFP_ATOMIC); if (ctx->elements == NULL){ QETH_DBF_TEXT(trace, 2, "ceddpcn4"); qeth_eddp_free_context(ctx); return NULL; } - memset(ctx->elements, 0, - ctx->num_elements * sizeof(struct qeth_eddp_element)); /* reset num_elements; will be incremented again in fill_buffer to * reflect number of actually used elements */ ctx->num_elements = 0; @@ -589,7 +588,7 @@ qeth_eddp_create_context_tcp(struct qeth_card *card, struct sk_buff *skb, struct qeth_hdr *qhdr) { struct qeth_eddp_context *ctx = NULL; - + QETH_DBF_TEXT(trace, 5, "creddpct"); if (skb->protocol == ETH_P_IP) ctx = qeth_eddp_create_context_generic(card, skb, diff --git a/drivers/s390/net/qeth_main.c b/drivers/s390/net/qeth_main.c index 9e671a48c..290866932 100644 --- a/drivers/s390/net/qeth_main.c +++ b/drivers/s390/net/qeth_main.c @@ -4417,7 +4417,6 @@ qeth_send_packet(struct qeth_card *card, struct sk_buff *skb) struct qeth_eddp_context *ctx = NULL; int tx_bytes = skb->len; unsigned short nr_frags = skb_shinfo(skb)->nr_frags; - unsigned short tso_size = skb_shinfo(skb)->tso_size; int rc; QETH_DBF_TEXT(trace, 6, "sendpkt"); @@ -4453,7 +4452,7 @@ qeth_send_packet(struct qeth_card *card, struct sk_buff *skb) queue = card->qdio.out_qs [qeth_get_priority_queue(card, skb, ipv, cast_type)]; - if (skb_shinfo(skb)->tso_size) + if (skb_is_gso(skb)) large_send = card->options.large_send; /*are we able to do TSO ? If so ,prepare and send it from here */ @@ -4500,9 +4499,8 @@ qeth_send_packet(struct qeth_card *card, struct sk_buff *skb) card->stats.tx_packets++; card->stats.tx_bytes += tx_bytes; #ifdef CONFIG_QETH_PERF_STATS - if (tso_size && - !(large_send == QETH_LARGE_SEND_NO)) { - card->perf_stats.large_send_bytes += tx_bytes; + if (skb_is_gso(skb) && !(large_send == QETH_LARGE_SEND_NO)) { + card->perf_stats.large_send_bytes += skb->len; card->perf_stats.large_send_cnt++; } if (nr_frags > 0){ diff --git a/drivers/s390/net/qeth_tso.h b/drivers/s390/net/qeth_tso.h index ad33e6f46..593f29814 100644 --- a/drivers/s390/net/qeth_tso.h +++ b/drivers/s390/net/qeth_tso.h @@ -1,13 +1,11 @@ /* - * linux/drivers/s390/net/qeth_tso.h ($Revision: 1.7 $) + * linux/drivers/s390/net/qeth_tso.h * * Header file for qeth TCP Segmentation Offload support. * * Copyright 2004 IBM Corporation * - * Author(s): Frank Pavlic - * - * $Revision: 1.7 $ $Date: 2005/05/04 20:19:18 $ + * Author(s): Frank Pavlic * */ #ifndef __QETH_TSO_H__ @@ -53,7 +51,7 @@ qeth_tso_fill_header(struct qeth_card *card, struct sk_buff *skb) hdr->ext.hdr_version = 1; hdr->ext.hdr_len = 28; /*insert non-fix values */ - hdr->ext.mss = skb_shinfo(skb)->tso_size; + hdr->ext.mss = skb_shinfo(skb)->gso_size; hdr->ext.dg_hdr_len = (__u16)(iph->ihl*4 + tcph->doff*4); hdr->ext.payload_len = (__u16)(skb->len - hdr->ext.dg_hdr_len - sizeof(struct qeth_hdr_tso)); @@ -119,11 +117,11 @@ __qeth_fill_buffer_frag(struct sk_buff *skb, struct qdio_buffer *buffer, int fragno; unsigned long addr; int element, cnt, dlen; - + fragno = skb_shinfo(skb)->nr_frags; element = *next_element_to_fill; dlen = 0; - + if (is_tso) buffer->element[element].flags = SBAL_FLAGS_MIDDLE_FRAG; diff --git a/drivers/scsi/Makefile b/drivers/scsi/Makefile index a262e384c..81803a16f 100644 --- a/drivers/scsi/Makefile +++ b/drivers/scsi/Makefile @@ -33,7 +33,7 @@ obj-$(CONFIG_SCSI_FC_ATTRS) += scsi_transport_fc.o obj-$(CONFIG_SCSI_ISCSI_ATTRS) += scsi_transport_iscsi.o obj-$(CONFIG_SCSI_SAS_ATTRS) += scsi_transport_sas.o -obj-$(CONFIG_ISCSI_TCP) += libiscsi.o iscsi_tcp.o +obj-$(CONFIG_ISCSI_TCP) += iscsi_tcp.o obj-$(CONFIG_SCSI_AMIGA7XX) += amiga7xx.o 53c7xx.o obj-$(CONFIG_A3000_SCSI) += a3000.o wd33c93.o obj-$(CONFIG_A2091_SCSI) += a2091.o wd33c93.o diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index b4743a9ec..2068b6682 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -3,8 +3,7 @@ * * Copyright (C) 2004 Dmitry Yusupov * Copyright (C) 2004 Alex Aizman - * Copyright (C) 2005 - 2006 Mike Christie - * Copyright (C) 2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2005 Mike Christie * maintained by open-iscsi@googlegroups.com * * This program is free software; you can redistribute it and/or modify @@ -37,28 +36,37 @@ #include #include #include +#include +#include +#include +#include #include #include #include #include "iscsi_tcp.h" -#define ISCSI_TCP_VERSION "1.0-595" - MODULE_AUTHOR("Dmitry Yusupov , " "Alex Aizman "); MODULE_DESCRIPTION("iSCSI/TCP data-path"); MODULE_LICENSE("GPL"); -MODULE_VERSION(ISCSI_TCP_VERSION); +MODULE_VERSION("0:4.445"); /* #define DEBUG_TCP */ +/* #define DEBUG_SCSI */ #define DEBUG_ASSERT #ifdef DEBUG_TCP -#define debug_tcp(fmt...) printk(KERN_INFO "tcp: " fmt) +#define debug_tcp(fmt...) printk(KERN_DEBUG "tcp: " fmt) #else #define debug_tcp(fmt...) #endif +#ifdef DEBUG_SCSI +#define debug_scsi(fmt...) printk(KERN_DEBUG "scsi: " fmt) +#else +#define debug_scsi(fmt...) +#endif + #ifndef DEBUG_ASSERT #ifdef BUG_ON #undef BUG_ON @@ -66,9 +74,22 @@ MODULE_VERSION(ISCSI_TCP_VERSION); #define BUG_ON(expr) #endif +#define INVALID_SN_DELTA 0xffff + static unsigned int iscsi_max_lun = 512; module_param_named(max_lun, iscsi_max_lun, uint, S_IRUGO); +/* global data */ +static kmem_cache_t *taskcache; + +static inline void +iscsi_buf_init_virt(struct iscsi_buf *ibuf, char *vbuf, int size) +{ + sg_init_one(&ibuf->sg, (u8 *)vbuf, size); + ibuf->sent = 0; + ibuf->use_sendmsg = 0; +} + static inline void iscsi_buf_init_iov(struct iscsi_buf *ibuf, char *vbuf, int size) { @@ -109,39 +130,68 @@ static inline void iscsi_hdr_digest(struct iscsi_conn *conn, struct iscsi_buf *buf, u8* crc) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - - crypto_digest_digest(tcp_conn->tx_tfm, &buf->sg, 1, crc); + crypto_digest_digest(conn->tx_tfm, &buf->sg, 1, crc); buf->sg.length += sizeof(uint32_t); } +static void +iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err) +{ + struct iscsi_session *session = conn->session; + unsigned long flags; + + spin_lock_irqsave(&session->lock, flags); + if (session->conn_cnt == 1 || session->leadconn == conn) + session->state = ISCSI_STATE_FAILED; + spin_unlock_irqrestore(&session->lock, flags); + set_bit(SUSPEND_BIT, &conn->suspend_tx); + set_bit(SUSPEND_BIT, &conn->suspend_rx); + iscsi_conn_error(conn->cls_conn, err); +} + static inline int -iscsi_hdr_extract(struct iscsi_tcp_conn *tcp_conn) +iscsi_check_assign_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr) { - struct sk_buff *skb = tcp_conn->in.skb; + uint32_t max_cmdsn = be32_to_cpu(hdr->max_cmdsn); + uint32_t exp_cmdsn = be32_to_cpu(hdr->exp_cmdsn); + + if (max_cmdsn < exp_cmdsn -1 && + max_cmdsn > exp_cmdsn - INVALID_SN_DELTA) + return ISCSI_ERR_MAX_CMDSN; + if (max_cmdsn > session->max_cmdsn || + max_cmdsn < session->max_cmdsn - INVALID_SN_DELTA) + session->max_cmdsn = max_cmdsn; + if (exp_cmdsn > session->exp_cmdsn || + exp_cmdsn < session->exp_cmdsn - INVALID_SN_DELTA) + session->exp_cmdsn = exp_cmdsn; + + return 0; +} - tcp_conn->in.zero_copy_hdr = 0; +static inline int +iscsi_hdr_extract(struct iscsi_conn *conn) +{ + struct sk_buff *skb = conn->in.skb; - if (tcp_conn->in.copy >= tcp_conn->hdr_size && - tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER) { + if (conn->in.copy >= conn->hdr_size && + conn->in_progress == IN_PROGRESS_WAIT_HEADER) { /* * Zero-copy PDU Header: using connection context * to store header pointer. */ if (skb_shinfo(skb)->frag_list == NULL && - !skb_shinfo(skb)->nr_frags) { - tcp_conn->in.hdr = (struct iscsi_hdr *) - ((char*)skb->data + tcp_conn->in.offset); - tcp_conn->in.zero_copy_hdr = 1; - } else { + !skb_shinfo(skb)->nr_frags) + conn->in.hdr = (struct iscsi_hdr *) + ((char*)skb->data + conn->in.offset); + else { /* ignoring return code since we checked * in.copy before */ - skb_copy_bits(skb, tcp_conn->in.offset, - &tcp_conn->hdr, tcp_conn->hdr_size); - tcp_conn->in.hdr = &tcp_conn->hdr; + skb_copy_bits(skb, conn->in.offset, + &conn->hdr, conn->hdr_size); + conn->in.hdr = &conn->hdr; } - tcp_conn->in.offset += tcp_conn->hdr_size; - tcp_conn->in.copy -= tcp_conn->hdr_size; + conn->in.offset += conn->hdr_size; + conn->in.copy -= conn->hdr_size; } else { int hdr_remains; int copylen; @@ -151,51 +201,118 @@ iscsi_hdr_extract(struct iscsi_tcp_conn *tcp_conn) * copying it... This'll happen quite rarely. */ - if (tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER) - tcp_conn->in.hdr_offset = 0; + if (conn->in_progress == IN_PROGRESS_WAIT_HEADER) + conn->in.hdr_offset = 0; - hdr_remains = tcp_conn->hdr_size - tcp_conn->in.hdr_offset; + hdr_remains = conn->hdr_size - conn->in.hdr_offset; BUG_ON(hdr_remains <= 0); - copylen = min(tcp_conn->in.copy, hdr_remains); - skb_copy_bits(skb, tcp_conn->in.offset, - (char*)&tcp_conn->hdr + tcp_conn->in.hdr_offset, - copylen); + copylen = min(conn->in.copy, hdr_remains); + skb_copy_bits(skb, conn->in.offset, + (char*)&conn->hdr + conn->in.hdr_offset, copylen); debug_tcp("PDU gather offset %d bytes %d in.offset %d " - "in.copy %d\n", tcp_conn->in.hdr_offset, copylen, - tcp_conn->in.offset, tcp_conn->in.copy); + "in.copy %d\n", conn->in.hdr_offset, copylen, + conn->in.offset, conn->in.copy); - tcp_conn->in.offset += copylen; - tcp_conn->in.copy -= copylen; + conn->in.offset += copylen; + conn->in.copy -= copylen; if (copylen < hdr_remains) { - tcp_conn->in_progress = IN_PROGRESS_HEADER_GATHER; - tcp_conn->in.hdr_offset += copylen; + conn->in_progress = IN_PROGRESS_HEADER_GATHER; + conn->in.hdr_offset += copylen; return -EAGAIN; } - tcp_conn->in.hdr = &tcp_conn->hdr; - tcp_conn->discontiguous_hdr_cnt++; - tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER; + conn->in.hdr = &conn->hdr; + conn->discontiguous_hdr_cnt++; + conn->in_progress = IN_PROGRESS_WAIT_HEADER; } return 0; } -/* - * must be called with session lock - */ -static void -__iscsi_ctask_cleanup(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) +static inline void +iscsi_ctask_cleanup(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct scsi_cmnd *sc; + struct scsi_cmnd *sc = ctask->sc; + struct iscsi_session *session = conn->session; - sc = ctask->sc; - if (unlikely(!sc)) + spin_lock(&session->lock); + if (unlikely(!sc)) { + spin_unlock(&session->lock); return; + } + if (sc->sc_data_direction == DMA_TO_DEVICE) { + struct iscsi_data_task *dtask, *n; + /* WRITE: cleanup Data-Out's if any */ + list_for_each_entry_safe(dtask, n, &ctask->dataqueue, item) { + list_del(&dtask->item); + mempool_free(dtask, ctask->datapool); + } + } + ctask->xmstate = XMSTATE_IDLE; + ctask->r2t = NULL; + ctask->sc = NULL; + __kfifo_put(session->cmdpool.queue, (void*)&ctask, sizeof(void*)); + spin_unlock(&session->lock); +} + +/** + * iscsi_cmd_rsp - SCSI Command Response processing + * @conn: iscsi connection + * @ctask: scsi command task + **/ +static int +iscsi_cmd_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) +{ + int rc; + struct iscsi_cmd_rsp *rhdr = (struct iscsi_cmd_rsp *)conn->in.hdr; + struct iscsi_session *session = conn->session; + struct scsi_cmnd *sc = ctask->sc; + + rc = iscsi_check_assign_cmdsn(session, (struct iscsi_nopin*)rhdr); + if (rc) { + sc->result = (DID_ERROR << 16); + goto out; + } + + conn->exp_statsn = be32_to_cpu(rhdr->statsn) + 1; + + sc->result = (DID_OK << 16) | rhdr->cmd_status; + + if (rhdr->response != ISCSI_STATUS_CMD_COMPLETED) { + sc->result = (DID_ERROR << 16); + goto out; + } + + if (rhdr->cmd_status == SAM_STAT_CHECK_CONDITION && conn->senselen) { + int sensecopy = min(conn->senselen, SCSI_SENSE_BUFFERSIZE); - tcp_ctask->xmstate = XMSTATE_IDLE; - tcp_ctask->r2t = NULL; + memcpy(sc->sense_buffer, conn->data + 2, sensecopy); + debug_scsi("copied %d bytes of sense\n", sensecopy); + } + + if (sc->sc_data_direction == DMA_TO_DEVICE) + goto out; + + if (rhdr->flags & ISCSI_FLAG_CMD_UNDERFLOW) { + int res_count = be32_to_cpu(rhdr->residual_count); + + if (res_count > 0 && res_count <= sc->request_bufflen) + sc->resid = res_count; + else + sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; + } else if (rhdr->flags & ISCSI_FLAG_CMD_BIDI_UNDERFLOW) + sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; + else if (rhdr->flags & ISCSI_FLAG_CMD_OVERFLOW) + sc->resid = be32_to_cpu(rhdr->residual_count); + +out: + debug_scsi("done [sc %lx res %d itt 0x%x]\n", + (long)sc, sc->result, ctask->itt); + conn->scsirsp_pdus_cnt++; + iscsi_ctask_cleanup(conn, ctask); + sc->scsi_done(sc); + return rc; } /** @@ -207,9 +324,7 @@ static int iscsi_data_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { int rc; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_data_rsp *rhdr = (struct iscsi_data_rsp *)tcp_conn->in.hdr; + struct iscsi_data_rsp *rhdr = (struct iscsi_data_rsp *)conn->in.hdr; struct iscsi_session *session = conn->session; int datasn = be32_to_cpu(rhdr->datasn); @@ -219,9 +334,9 @@ iscsi_data_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) /* * setup Data-In byte counter (gets decremented..) */ - ctask->data_count = tcp_conn->in.datalen; + ctask->data_count = conn->in.datalen; - if (tcp_conn->in.datalen == 0) + if (conn->in.datalen == 0) return 0; if (ctask->datasn != datasn) @@ -229,8 +344,8 @@ iscsi_data_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) ctask->datasn++; - tcp_ctask->data_offset = be32_to_cpu(rhdr->offset); - if (tcp_ctask->data_offset + tcp_conn->in.datalen > ctask->total_length) + ctask->data_offset = be32_to_cpu(rhdr->offset); + if (ctask->data_offset + conn->in.datalen > ctask->total_length) return ISCSI_ERR_DATA_OFFSET; if (rhdr->flags & ISCSI_FLAG_DATA_STATUS) { @@ -275,17 +390,19 @@ iscsi_solicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, struct iscsi_r2t_info *r2t) { struct iscsi_data *hdr; + struct iscsi_data_task *dtask; struct scsi_cmnd *sc = ctask->sc; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - hdr = &r2t->dtask.hdr; + dtask = mempool_alloc(ctask->datapool, GFP_ATOMIC); + BUG_ON(!dtask); + hdr = &dtask->hdr; memset(hdr, 0, sizeof(struct iscsi_data)); hdr->ttt = r2t->ttt; hdr->datasn = cpu_to_be32(r2t->solicit_datasn); r2t->solicit_datasn++; hdr->opcode = ISCSI_OP_SCSI_DATA_OUT; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - hdr->itt = ctask->hdr->itt; + memcpy(hdr->lun, ctask->hdr.lun, sizeof(hdr->lun)); + hdr->itt = ctask->hdr.itt; hdr->exp_statsn = r2t->exp_statsn; hdr->offset = cpu_to_be32(r2t->data_offset); if (r2t->data_length > conn->max_xmit_dlength) { @@ -301,9 +418,11 @@ iscsi_solicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, r2t->sent = 0; - iscsi_buf_init_iov(&r2t->headbuf, (char*)hdr, + iscsi_buf_init_virt(&r2t->headbuf, (char*)hdr, sizeof(struct iscsi_hdr)); + r2t->dtask = dtask; + if (sc->use_sg) { int i, sg_count = 0; struct scatterlist *sg = sc->request_buffer; @@ -332,9 +451,11 @@ iscsi_solicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, } BUG_ON(r2t->sg == NULL); } else - iscsi_buf_init_iov(&tcp_ctask->sendbuf, + iscsi_buf_init_iov(&ctask->sendbuf, (char*)sc->request_buffer + r2t->data_offset, r2t->data_count); + + list_add(&dtask->item, &ctask->dataqueue); } /** @@ -347,16 +468,17 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { struct iscsi_r2t_info *r2t; struct iscsi_session *session = conn->session; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_r2t_rsp *rhdr = (struct iscsi_r2t_rsp *)tcp_conn->in.hdr; + struct iscsi_r2t_rsp *rhdr = (struct iscsi_r2t_rsp *)conn->in.hdr; int r2tsn = be32_to_cpu(rhdr->r2tsn); int rc; - if (tcp_conn->in.datalen) + if (conn->in.ahslen) + return ISCSI_ERR_AHSLEN; + + if (conn->in.datalen) return ISCSI_ERR_DATALEN; - if (tcp_ctask->exp_r2tsn && tcp_ctask->exp_r2tsn != r2tsn) + if (ctask->exp_r2tsn && ctask->exp_r2tsn != r2tsn) return ISCSI_ERR_R2TSN; rc = iscsi_check_assign_cmdsn(session, (struct iscsi_nopin*)rhdr); @@ -374,7 +496,7 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) spin_unlock(&session->lock); return 0; } - rc = __kfifo_get(tcp_ctask->r2tpool.queue, (void*)&r2t, sizeof(void*)); + rc = __kfifo_get(ctask->r2tpool.queue, (void*)&r2t, sizeof(void*)); BUG_ON(!rc); r2t->exp_statsn = rhdr->statsn; @@ -396,10 +518,10 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) iscsi_solicit_data_init(conn, ctask, r2t); - tcp_ctask->exp_r2tsn = r2tsn + 1; - tcp_ctask->xmstate |= XMSTATE_SOL_HDR; - __kfifo_put(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*)); - __kfifo_put(conn->xmitqueue, (void*)&ctask, sizeof(void*)); + ctask->exp_r2tsn = r2tsn + 1; + ctask->xmstate |= XMSTATE_SOL_HDR; + __kfifo_put(ctask->r2tqueue, (void*)&r2t, sizeof(void*)); + __kfifo_put(conn->writequeue, (void*)&ctask, sizeof(void*)); scsi_queue_work(session->host, &conn->xmitwork); conn->r2t_pdus_cnt++; @@ -409,136 +531,258 @@ iscsi_r2t_rsp(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) } static int -iscsi_tcp_hdr_recv(struct iscsi_conn *conn) +iscsi_hdr_recv(struct iscsi_conn *conn) { - int rc = 0, opcode, ahslen; + int rc = 0; struct iscsi_hdr *hdr; + struct iscsi_cmd_task *ctask; struct iscsi_session *session = conn->session; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - uint32_t cdgst, rdgst = 0, itt; + uint32_t cdgst, rdgst = 0; - hdr = tcp_conn->in.hdr; + hdr = conn->in.hdr; /* verify PDU length */ - tcp_conn->in.datalen = ntoh24(hdr->dlength); - if (tcp_conn->in.datalen > conn->max_recv_dlength) { + conn->in.datalen = ntoh24(hdr->dlength); + if (conn->in.datalen > conn->max_recv_dlength) { printk(KERN_ERR "iscsi_tcp: datalen %d > %d\n", - tcp_conn->in.datalen, conn->max_recv_dlength); + conn->in.datalen, conn->max_recv_dlength); return ISCSI_ERR_DATALEN; } - tcp_conn->data_copied = 0; + conn->data_copied = 0; /* read AHS */ - ahslen = hdr->hlength << 2; - tcp_conn->in.offset += ahslen; - tcp_conn->in.copy -= ahslen; - if (tcp_conn->in.copy < 0) { + conn->in.ahslen = hdr->hlength * 4; + conn->in.offset += conn->in.ahslen; + conn->in.copy -= conn->in.ahslen; + if (conn->in.copy < 0) { printk(KERN_ERR "iscsi_tcp: can't handle AHS with length " - "%d bytes\n", ahslen); + "%d bytes\n", conn->in.ahslen); return ISCSI_ERR_AHSLEN; } /* calculate read padding */ - tcp_conn->in.padding = tcp_conn->in.datalen & (ISCSI_PAD_LEN-1); - if (tcp_conn->in.padding) { - tcp_conn->in.padding = ISCSI_PAD_LEN - tcp_conn->in.padding; - debug_scsi("read padding %d bytes\n", tcp_conn->in.padding); + conn->in.padding = conn->in.datalen & (ISCSI_PAD_LEN-1); + if (conn->in.padding) { + conn->in.padding = ISCSI_PAD_LEN - conn->in.padding; + debug_scsi("read padding %d bytes\n", conn->in.padding); } if (conn->hdrdgst_en) { struct scatterlist sg; sg_init_one(&sg, (u8 *)hdr, - sizeof(struct iscsi_hdr) + ahslen); - crypto_digest_digest(tcp_conn->rx_tfm, &sg, 1, (u8 *)&cdgst); + sizeof(struct iscsi_hdr) + conn->in.ahslen); + crypto_digest_digest(conn->rx_tfm, &sg, 1, (u8 *)&cdgst); rdgst = *(uint32_t*)((char*)hdr + sizeof(struct iscsi_hdr) + - ahslen); + conn->in.ahslen); if (cdgst != rdgst) { - printk(KERN_ERR "iscsi_tcp: hdrdgst error " - "recv 0x%x calc 0x%x\n", rdgst, cdgst); + printk(KERN_ERR "iscsi_tcp: itt %x: hdrdgst error " + "recv 0x%x calc 0x%x\n", conn->in.itt, rdgst, + cdgst); return ISCSI_ERR_HDR_DGST; } } - opcode = hdr->opcode & ISCSI_OPCODE_MASK; + /* save opcode for later */ + conn->in.opcode = hdr->opcode & ISCSI_OPCODE_MASK; + /* verify itt (itt encoding: age+cid+itt) */ - rc = iscsi_verify_itt(conn, hdr, &itt); - if (rc == ISCSI_ERR_NO_SCSI_CMD) { - tcp_conn->in.datalen = 0; /* force drop */ - return 0; - } else if (rc) - return rc; + if (hdr->itt != cpu_to_be32(ISCSI_RESERVED_TAG)) { + if ((hdr->itt & AGE_MASK) != + (session->age << AGE_SHIFT)) { + printk(KERN_ERR "iscsi_tcp: received itt %x expected " + "session age (%x)\n", hdr->itt, + session->age & AGE_MASK); + return ISCSI_ERR_BAD_ITT; + } + + if ((hdr->itt & CID_MASK) != (conn->id << CID_SHIFT)) { + printk(KERN_ERR "iscsi_tcp: received itt %x, expected " + "CID (%x)\n", hdr->itt, conn->id); + return ISCSI_ERR_BAD_ITT; + } + conn->in.itt = hdr->itt & ITT_MASK; + } else + conn->in.itt = hdr->itt; debug_tcp("opcode 0x%x offset %d copy %d ahslen %d datalen %d\n", - opcode, tcp_conn->in.offset, tcp_conn->in.copy, - ahslen, tcp_conn->in.datalen); + hdr->opcode, conn->in.offset, conn->in.copy, + conn->in.ahslen, conn->in.datalen); - switch(opcode) { - case ISCSI_OP_SCSI_DATA_IN: - tcp_conn->in.ctask = session->cmds[itt]; - rc = iscsi_data_rsp(conn, tcp_conn->in.ctask); - /* fall through */ - case ISCSI_OP_SCSI_CMD_RSP: - tcp_conn->in.ctask = session->cmds[itt]; - if (tcp_conn->in.datalen) - goto copy_hdr; - - spin_lock(&session->lock); - __iscsi_ctask_cleanup(conn, tcp_conn->in.ctask); - rc = __iscsi_complete_pdu(conn, hdr, NULL, 0); - spin_unlock(&session->lock); - break; - case ISCSI_OP_R2T: - tcp_conn->in.ctask = session->cmds[itt]; - if (ahslen) - rc = ISCSI_ERR_AHSLEN; - else if (tcp_conn->in.ctask->sc->sc_data_direction == - DMA_TO_DEVICE) - rc = iscsi_r2t_rsp(conn, tcp_conn->in.ctask); - else - rc = ISCSI_ERR_PROTO; - break; - case ISCSI_OP_LOGIN_RSP: - case ISCSI_OP_TEXT_RSP: - case ISCSI_OP_LOGOUT_RSP: - case ISCSI_OP_NOOP_IN: - case ISCSI_OP_REJECT: - case ISCSI_OP_ASYNC_EVENT: - if (tcp_conn->in.datalen) - goto copy_hdr; - /* fall through */ - case ISCSI_OP_SCSI_TMFUNC_RSP: - rc = iscsi_complete_pdu(conn, hdr, NULL, 0); - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } + if (conn->in.itt < session->cmds_max) { + ctask = (struct iscsi_cmd_task *)session->cmds[conn->in.itt]; - return rc; + if (!ctask->sc) { + printk(KERN_INFO "iscsi_tcp: dropping ctask with " + "itt 0x%x\n", ctask->itt); + conn->in.datalen = 0; /* force drop */ + return 0; + } -copy_hdr: - /* - * if we did zero copy for the header but we will need multiple - * skbs to complete the command then we have to copy the header - * for later use - */ - if (tcp_conn->in.zero_copy_hdr && tcp_conn->in.copy < - (tcp_conn->in.datalen + tcp_conn->in.padding + - (conn->datadgst_en ? 4 : 0))) { - debug_tcp("Copying header for later use. in.copy %d in.datalen" - " %d\n", tcp_conn->in.copy, tcp_conn->in.datalen); - memcpy(&tcp_conn->hdr, tcp_conn->in.hdr, - sizeof(struct iscsi_hdr)); - tcp_conn->in.hdr = &tcp_conn->hdr; - tcp_conn->in.zero_copy_hdr = 0; - } - return 0; + if (ctask->sc->SCp.phase != session->age) { + printk(KERN_ERR "iscsi_tcp: ctask's session age %d, " + "expected %d\n", ctask->sc->SCp.phase, + session->age); + return ISCSI_ERR_SESSION_FAILED; + } + + conn->in.ctask = ctask; + + debug_scsi("rsp [op 0x%x cid %d sc %lx itt 0x%x len %d]\n", + hdr->opcode, conn->id, (long)ctask->sc, + ctask->itt, conn->in.datalen); + + switch(conn->in.opcode) { + case ISCSI_OP_SCSI_CMD_RSP: + BUG_ON((void*)ctask != ctask->sc->SCp.ptr); + if (!conn->in.datalen) + rc = iscsi_cmd_rsp(conn, ctask); + else + /* + * got sense or response data; copying PDU + * Header to the connection's header + * placeholder + */ + memcpy(&conn->hdr, hdr, + sizeof(struct iscsi_hdr)); + break; + case ISCSI_OP_SCSI_DATA_IN: + BUG_ON((void*)ctask != ctask->sc->SCp.ptr); + /* save flags for non-exceptional status */ + conn->in.flags = hdr->flags; + /* save cmd_status for sense data */ + conn->in.cmd_status = + ((struct iscsi_data_rsp*)hdr)->cmd_status; + rc = iscsi_data_rsp(conn, ctask); + break; + case ISCSI_OP_R2T: + BUG_ON((void*)ctask != ctask->sc->SCp.ptr); + if (ctask->sc->sc_data_direction == DMA_TO_DEVICE) + rc = iscsi_r2t_rsp(conn, ctask); + else + rc = ISCSI_ERR_PROTO; + break; + default: + rc = ISCSI_ERR_BAD_OPCODE; + break; + } + } else if (conn->in.itt >= ISCSI_MGMT_ITT_OFFSET && + conn->in.itt < ISCSI_MGMT_ITT_OFFSET + + session->mgmtpool_max) { + struct iscsi_mgmt_task *mtask = (struct iscsi_mgmt_task *) + session->mgmt_cmds[conn->in.itt - + ISCSI_MGMT_ITT_OFFSET]; + + debug_scsi("immrsp [op 0x%x cid %d itt 0x%x len %d]\n", + conn->in.opcode, conn->id, mtask->itt, + conn->in.datalen); + + switch(conn->in.opcode) { + case ISCSI_OP_LOGIN_RSP: + case ISCSI_OP_TEXT_RSP: + case ISCSI_OP_LOGOUT_RSP: + rc = iscsi_check_assign_cmdsn(session, + (struct iscsi_nopin*)hdr); + if (rc) + break; + + if (!conn->in.datalen) { + rc = iscsi_recv_pdu(conn->cls_conn, hdr, + NULL, 0); + if (conn->login_mtask != mtask) { + spin_lock(&session->lock); + __kfifo_put(session->mgmtpool.queue, + (void*)&mtask, sizeof(void*)); + spin_unlock(&session->lock); + } + } + break; + case ISCSI_OP_SCSI_TMFUNC_RSP: + rc = iscsi_check_assign_cmdsn(session, + (struct iscsi_nopin*)hdr); + if (rc) + break; + + if (conn->in.datalen || conn->in.ahslen) { + rc = ISCSI_ERR_PROTO; + break; + } + conn->tmfrsp_pdus_cnt++; + spin_lock(&session->lock); + if (conn->tmabort_state == TMABORT_INITIAL) { + __kfifo_put(session->mgmtpool.queue, + (void*)&mtask, sizeof(void*)); + conn->tmabort_state = + ((struct iscsi_tm_rsp *)hdr)-> + response == ISCSI_TMF_RSP_COMPLETE ? + TMABORT_SUCCESS:TMABORT_FAILED; + /* unblock eh_abort() */ + wake_up(&conn->ehwait); + } + spin_unlock(&session->lock); + break; + case ISCSI_OP_NOOP_IN: + if (hdr->ttt != ISCSI_RESERVED_TAG) { + rc = ISCSI_ERR_PROTO; + break; + } + rc = iscsi_check_assign_cmdsn(session, + (struct iscsi_nopin*)hdr); + if (rc) + break; + conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; + + if (!conn->in.datalen) { + struct iscsi_mgmt_task *mtask; + + rc = iscsi_recv_pdu(conn->cls_conn, hdr, + NULL, 0); + mtask = (struct iscsi_mgmt_task *) + session->mgmt_cmds[conn->in.itt - + ISCSI_MGMT_ITT_OFFSET]; + if (conn->login_mtask != mtask) { + spin_lock(&session->lock); + __kfifo_put(session->mgmtpool.queue, + (void*)&mtask, sizeof(void*)); + spin_unlock(&session->lock); + } + } + break; + default: + rc = ISCSI_ERR_BAD_OPCODE; + break; + } + } else if (conn->in.itt == ISCSI_RESERVED_TAG) { + switch(conn->in.opcode) { + case ISCSI_OP_NOOP_IN: + if (!conn->in.datalen) { + rc = iscsi_check_assign_cmdsn(session, + (struct iscsi_nopin*)hdr); + if (!rc && hdr->ttt != ISCSI_RESERVED_TAG) + rc = iscsi_recv_pdu(conn->cls_conn, + hdr, NULL, 0); + } else + rc = ISCSI_ERR_PROTO; + break; + case ISCSI_OP_REJECT: + /* we need sth like iscsi_reject_rsp()*/ + case ISCSI_OP_ASYNC_EVENT: + /* we need sth like iscsi_async_event_rsp() */ + rc = ISCSI_ERR_BAD_OPCODE; + break; + default: + rc = ISCSI_ERR_BAD_OPCODE; + break; + } + } else + rc = ISCSI_ERR_BAD_ITT; + + return rc; } /** * iscsi_ctask_copy - copy skb bits to the destanation cmd task - * @conn: iscsi tcp connection + * @conn: iscsi connection * @ctask: scsi command task * @buf: buffer to copy to * @buf_size: size of buffer @@ -560,113 +804,110 @@ copy_hdr: * buf_left left to copy from in progress buffer **/ static inline int -iscsi_ctask_copy(struct iscsi_tcp_conn *tcp_conn, struct iscsi_cmd_task *ctask, +iscsi_ctask_copy(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, void *buf, int buf_size, int offset) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - int buf_left = buf_size - (tcp_conn->data_copied + offset); - int size = min(tcp_conn->in.copy, buf_left); + int buf_left = buf_size - (conn->data_copied + offset); + int size = min(conn->in.copy, buf_left); int rc; size = min(size, ctask->data_count); debug_tcp("ctask_copy %d bytes at offset %d copied %d\n", - size, tcp_conn->in.offset, tcp_conn->in.copied); + size, conn->in.offset, conn->in.copied); BUG_ON(size <= 0); - BUG_ON(tcp_ctask->sent + size > ctask->total_length); + BUG_ON(ctask->sent + size > ctask->total_length); - rc = skb_copy_bits(tcp_conn->in.skb, tcp_conn->in.offset, - (char*)buf + (offset + tcp_conn->data_copied), size); + rc = skb_copy_bits(conn->in.skb, conn->in.offset, + (char*)buf + (offset + conn->data_copied), size); /* must fit into skb->len */ BUG_ON(rc); - tcp_conn->in.offset += size; - tcp_conn->in.copy -= size; - tcp_conn->in.copied += size; - tcp_conn->data_copied += size; - tcp_ctask->sent += size; + conn->in.offset += size; + conn->in.copy -= size; + conn->in.copied += size; + conn->data_copied += size; + ctask->sent += size; ctask->data_count -= size; - BUG_ON(tcp_conn->in.copy < 0); + BUG_ON(conn->in.copy < 0); BUG_ON(ctask->data_count < 0); - if (buf_size != (tcp_conn->data_copied + offset)) { + if (buf_size != (conn->data_copied + offset)) { if (!ctask->data_count) { - BUG_ON(buf_size - tcp_conn->data_copied < 0); + BUG_ON(buf_size - conn->data_copied < 0); /* done with this PDU */ - return buf_size - tcp_conn->data_copied; + return buf_size - conn->data_copied; } return -EAGAIN; } /* done with this buffer or with both - PDU and buffer */ - tcp_conn->data_copied = 0; + conn->data_copied = 0; return 0; } /** * iscsi_tcp_copy - copy skb bits to the destanation buffer - * @conn: iscsi tcp connection + * @conn: iscsi connection + * @buf: buffer to copy to + * @buf_size: number of bytes to copy * * Notes: * The function calls skb_copy_bits() and updates per-connection * byte counters. **/ static inline int -iscsi_tcp_copy(struct iscsi_tcp_conn *tcp_conn) +iscsi_tcp_copy(struct iscsi_conn *conn, void *buf, int buf_size) { - void *buf = tcp_conn->data; - int buf_size = tcp_conn->in.datalen; - int buf_left = buf_size - tcp_conn->data_copied; - int size = min(tcp_conn->in.copy, buf_left); + int buf_left = buf_size - conn->data_copied; + int size = min(conn->in.copy, buf_left); int rc; debug_tcp("tcp_copy %d bytes at offset %d copied %d\n", - size, tcp_conn->in.offset, tcp_conn->data_copied); + size, conn->in.offset, conn->data_copied); BUG_ON(size <= 0); - rc = skb_copy_bits(tcp_conn->in.skb, tcp_conn->in.offset, - (char*)buf + tcp_conn->data_copied, size); + rc = skb_copy_bits(conn->in.skb, conn->in.offset, + (char*)buf + conn->data_copied, size); BUG_ON(rc); - tcp_conn->in.offset += size; - tcp_conn->in.copy -= size; - tcp_conn->in.copied += size; - tcp_conn->data_copied += size; + conn->in.offset += size; + conn->in.copy -= size; + conn->in.copied += size; + conn->data_copied += size; - if (buf_size != tcp_conn->data_copied) + if (buf_size != conn->data_copied) return -EAGAIN; return 0; } static inline void -partial_sg_digest_update(struct iscsi_tcp_conn *tcp_conn, - struct scatterlist *sg, int offset, int length) +partial_sg_digest_update(struct iscsi_conn *conn, struct scatterlist *sg, + int offset, int length) { struct scatterlist temp; memcpy(&temp, sg, sizeof(struct scatterlist)); temp.offset = offset; temp.length = length; - crypto_digest_update(tcp_conn->data_rx_tfm, &temp, 1); + crypto_digest_update(conn->data_rx_tfm, &temp, 1); } static void -iscsi_recv_digest_update(struct iscsi_tcp_conn *tcp_conn, char* buf, int len) +iscsi_recv_digest_update(struct iscsi_conn *conn, char* buf, int len) { struct scatterlist tmp; sg_init_one(&tmp, buf, len); - crypto_digest_update(tcp_conn->data_rx_tfm, &tmp, 1); + crypto_digest_update(conn->data_rx_tfm, &tmp, 1); } static int iscsi_scsi_data_in(struct iscsi_conn *conn) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_cmd_task *ctask = tcp_conn->in.ctask; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; + struct iscsi_cmd_task *ctask = conn->in.ctask; struct scsi_cmnd *sc = ctask->sc; struct scatterlist *sg; int i, offset, rc = 0; @@ -678,33 +919,31 @@ static int iscsi_scsi_data_in(struct iscsi_conn *conn) */ if (!sc->use_sg) { i = ctask->data_count; - rc = iscsi_ctask_copy(tcp_conn, ctask, sc->request_buffer, - sc->request_bufflen, - tcp_ctask->data_offset); + rc = iscsi_ctask_copy(conn, ctask, sc->request_buffer, + sc->request_bufflen, ctask->data_offset); if (rc == -EAGAIN) return rc; if (conn->datadgst_en) - iscsi_recv_digest_update(tcp_conn, sc->request_buffer, - i); + iscsi_recv_digest_update(conn, sc->request_buffer, i); rc = 0; goto done; } - offset = tcp_ctask->data_offset; + offset = ctask->data_offset; sg = sc->request_buffer; - if (tcp_ctask->data_offset) - for (i = 0; i < tcp_ctask->sg_count; i++) + if (ctask->data_offset) + for (i = 0; i < ctask->sg_count; i++) offset -= sg[i].length; /* we've passed through partial sg*/ if (offset < 0) offset = 0; - for (i = tcp_ctask->sg_count; i < sc->use_sg; i++) { + for (i = ctask->sg_count; i < sc->use_sg; i++) { char *dest; dest = kmap_atomic(sg[i].page, KM_SOFTIRQ0); - rc = iscsi_ctask_copy(tcp_conn, ctask, dest + sg[i].offset, + rc = iscsi_ctask_copy(conn, ctask, dest + sg[i].offset, sg[i].length, offset); kunmap_atomic(dest, KM_SOFTIRQ0); if (rc == -EAGAIN) @@ -713,17 +952,15 @@ static int iscsi_scsi_data_in(struct iscsi_conn *conn) if (!rc) { if (conn->datadgst_en) { if (!offset) - crypto_digest_update( - tcp_conn->data_rx_tfm, - &sg[i], 1); + crypto_digest_update(conn->data_rx_tfm, + &sg[i], 1); else - partial_sg_digest_update(tcp_conn, - &sg[i], + partial_sg_digest_update(conn, &sg[i], sg[i].offset + offset, sg[i].length - offset); } offset = 0; - tcp_ctask->sg_count++; + ctask->sg_count++; } if (!ctask->data_count) { @@ -731,26 +968,25 @@ static int iscsi_scsi_data_in(struct iscsi_conn *conn) /* * data-in is complete, but buffer not... */ - partial_sg_digest_update(tcp_conn, &sg[i], + partial_sg_digest_update(conn, &sg[i], sg[i].offset, sg[i].length-rc); rc = 0; break; } - if (!tcp_conn->in.copy) + if (!conn->in.copy) return -EAGAIN; } BUG_ON(ctask->data_count); done: /* check for non-exceptional status */ - if (tcp_conn->in.hdr->flags & ISCSI_FLAG_DATA_STATUS) { + if (conn->in.flags & ISCSI_FLAG_DATA_STATUS) { debug_scsi("done [sc %lx res %d itt 0x%x]\n", (long)sc, sc->result, ctask->itt); - spin_lock(&conn->session->lock); - __iscsi_ctask_cleanup(conn, ctask); - __iscsi_complete_pdu(conn, tcp_conn->in.hdr, NULL, 0); - spin_unlock(&conn->session->lock); + conn->scsirsp_pdus_cnt++; + iscsi_ctask_cleanup(conn, ctask); + sc->scsi_done(sc); } return rc; @@ -759,38 +995,71 @@ done: static int iscsi_data_recv(struct iscsi_conn *conn) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - int rc = 0, opcode; + struct iscsi_session *session = conn->session; + int rc = 0; - opcode = tcp_conn->in.hdr->opcode & ISCSI_OPCODE_MASK; - switch (opcode) { + switch(conn->in.opcode) { case ISCSI_OP_SCSI_DATA_IN: rc = iscsi_scsi_data_in(conn); break; - case ISCSI_OP_SCSI_CMD_RSP: - spin_lock(&conn->session->lock); - __iscsi_ctask_cleanup(conn, tcp_conn->in.ctask); - spin_unlock(&conn->session->lock); + case ISCSI_OP_SCSI_CMD_RSP: { + /* + * SCSI Sense Data: + * copying the entire Data Segment. + */ + if (iscsi_tcp_copy(conn, conn->data, conn->in.datalen)) { + rc = -EAGAIN; + goto exit; + } + + /* + * check for sense + */ + conn->in.hdr = &conn->hdr; + conn->senselen = (conn->data[0] << 8) | conn->data[1]; + rc = iscsi_cmd_rsp(conn, conn->in.ctask); + if (!rc && conn->datadgst_en) + iscsi_recv_digest_update(conn, conn->data, + conn->in.datalen); + } + break; case ISCSI_OP_TEXT_RSP: case ISCSI_OP_LOGIN_RSP: - case ISCSI_OP_NOOP_IN: - case ISCSI_OP_ASYNC_EVENT: - case ISCSI_OP_REJECT: + case ISCSI_OP_NOOP_IN: { + struct iscsi_mgmt_task *mtask = NULL; + + if (conn->in.itt != ISCSI_RESERVED_TAG) + mtask = (struct iscsi_mgmt_task *) + session->mgmt_cmds[conn->in.itt - + ISCSI_MGMT_ITT_OFFSET]; + /* * Collect data segment to the connection's data * placeholder */ - if (iscsi_tcp_copy(tcp_conn)) { + if (iscsi_tcp_copy(conn, conn->data, conn->in.datalen)) { rc = -EAGAIN; goto exit; } - rc = iscsi_complete_pdu(conn, tcp_conn->in.hdr, tcp_conn->data, - tcp_conn->in.datalen); - if (!rc && conn->datadgst_en && opcode != ISCSI_OP_LOGIN_RSP) - iscsi_recv_digest_update(tcp_conn, tcp_conn->data, - tcp_conn->in.datalen); - break; + rc = iscsi_recv_pdu(conn->cls_conn, conn->in.hdr, + conn->data, conn->in.datalen); + + if (!rc && conn->datadgst_en && + conn->in.opcode != ISCSI_OP_LOGIN_RSP) + iscsi_recv_digest_update(conn, conn->data, + conn->in.datalen); + + if (mtask && conn->login_mtask != mtask) { + spin_lock(&session->lock); + __kfifo_put(session->mgmtpool.queue, (void*)&mtask, + sizeof(void*)); + spin_unlock(&session->lock); + } + } + break; + case ISCSI_OP_ASYNC_EVENT: + case ISCSI_OP_REJECT: default: BUG_ON(1); } @@ -811,7 +1080,6 @@ iscsi_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, { int rc; struct iscsi_conn *conn = rd_desc->arg.data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; int processed; char pad[ISCSI_PAD_LEN]; struct scatterlist sg; @@ -820,15 +1088,15 @@ iscsi_tcp_data_recv(read_descriptor_t *rd_desc, struct sk_buff *skb, * Save current SKB and its offset in the corresponding * connection context. */ - tcp_conn->in.copy = skb->len - offset; - tcp_conn->in.offset = offset; - tcp_conn->in.skb = skb; - tcp_conn->in.len = tcp_conn->in.copy; - BUG_ON(tcp_conn->in.copy <= 0); - debug_tcp("in %d bytes\n", tcp_conn->in.copy); + conn->in.copy = skb->len - offset; + conn->in.offset = offset; + conn->in.skb = skb; + conn->in.len = conn->in.copy; + BUG_ON(conn->in.copy <= 0); + debug_tcp("in %d bytes\n", conn->in.copy); more: - tcp_conn->in.copied = 0; + conn->in.copied = 0; rc = 0; if (unlikely(conn->suspend_rx)) { @@ -836,9 +1104,9 @@ more: return 0; } - if (tcp_conn->in_progress == IN_PROGRESS_WAIT_HEADER || - tcp_conn->in_progress == IN_PROGRESS_HEADER_GATHER) { - rc = iscsi_hdr_extract(tcp_conn); + if (conn->in_progress == IN_PROGRESS_WAIT_HEADER || + conn->in_progress == IN_PROGRESS_HEADER_GATHER) { + rc = iscsi_hdr_extract(conn); if (rc) { if (rc == -EAGAIN) goto nomore; @@ -851,91 +1119,90 @@ more: /* * Verify and process incoming PDU header. */ - rc = iscsi_tcp_hdr_recv(conn); - if (!rc && tcp_conn->in.datalen) { + rc = iscsi_hdr_recv(conn); + if (!rc && conn->in.datalen) { if (conn->datadgst_en) { - BUG_ON(!tcp_conn->data_rx_tfm); - crypto_digest_init(tcp_conn->data_rx_tfm); + BUG_ON(!conn->data_rx_tfm); + crypto_digest_init(conn->data_rx_tfm); } - tcp_conn->in_progress = IN_PROGRESS_DATA_RECV; + conn->in_progress = IN_PROGRESS_DATA_RECV; } else if (rc) { iscsi_conn_failure(conn, rc); return 0; } } - if (tcp_conn->in_progress == IN_PROGRESS_DDIGEST_RECV) { + if (conn->in_progress == IN_PROGRESS_DDIGEST_RECV) { uint32_t recv_digest; - debug_tcp("extra data_recv offset %d copy %d\n", - tcp_conn->in.offset, tcp_conn->in.copy); - skb_copy_bits(tcp_conn->in.skb, tcp_conn->in.offset, + conn->in.offset, conn->in.copy); + skb_copy_bits(conn->in.skb, conn->in.offset, &recv_digest, 4); - tcp_conn->in.offset += 4; - tcp_conn->in.copy -= 4; - if (recv_digest != tcp_conn->in.datadgst) { + conn->in.offset += 4; + conn->in.copy -= 4; + if (recv_digest != conn->in.datadgst) { debug_tcp("iscsi_tcp: data digest error!" "0x%x != 0x%x\n", recv_digest, - tcp_conn->in.datadgst); + conn->in.datadgst); iscsi_conn_failure(conn, ISCSI_ERR_DATA_DGST); return 0; } else { debug_tcp("iscsi_tcp: data digest match!" "0x%x == 0x%x\n", recv_digest, - tcp_conn->in.datadgst); - tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER; + conn->in.datadgst); + conn->in_progress = IN_PROGRESS_WAIT_HEADER; } } - if (tcp_conn->in_progress == IN_PROGRESS_DATA_RECV && - tcp_conn->in.copy) { + if (conn->in_progress == IN_PROGRESS_DATA_RECV && conn->in.copy) { debug_tcp("data_recv offset %d copy %d\n", - tcp_conn->in.offset, tcp_conn->in.copy); + conn->in.offset, conn->in.copy); rc = iscsi_data_recv(conn); if (rc) { - if (rc == -EAGAIN) + if (rc == -EAGAIN) { + rd_desc->count = conn->in.datalen - + conn->in.ctask->data_count; goto again; + } iscsi_conn_failure(conn, rc); return 0; } - tcp_conn->in.copy -= tcp_conn->in.padding; - tcp_conn->in.offset += tcp_conn->in.padding; + conn->in.copy -= conn->in.padding; + conn->in.offset += conn->in.padding; if (conn->datadgst_en) { - if (tcp_conn->in.padding) { - debug_tcp("padding -> %d\n", - tcp_conn->in.padding); - memset(pad, 0, tcp_conn->in.padding); - sg_init_one(&sg, pad, tcp_conn->in.padding); - crypto_digest_update(tcp_conn->data_rx_tfm, - &sg, 1); + if (conn->in.padding) { + debug_tcp("padding -> %d\n", conn->in.padding); + memset(pad, 0, conn->in.padding); + sg_init_one(&sg, pad, conn->in.padding); + crypto_digest_update(conn->data_rx_tfm, &sg, 1); } - crypto_digest_final(tcp_conn->data_rx_tfm, - (u8 *) & tcp_conn->in.datadgst); - debug_tcp("rx digest 0x%x\n", tcp_conn->in.datadgst); - tcp_conn->in_progress = IN_PROGRESS_DDIGEST_RECV; + crypto_digest_final(conn->data_rx_tfm, + (u8 *) & conn->in.datadgst); + debug_tcp("rx digest 0x%x\n", conn->in.datadgst); + conn->in_progress = IN_PROGRESS_DDIGEST_RECV; } else - tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER; + conn->in_progress = IN_PROGRESS_WAIT_HEADER; } debug_tcp("f, processed %d from out of %d padding %d\n", - tcp_conn->in.offset - offset, (int)len, tcp_conn->in.padding); - BUG_ON(tcp_conn->in.offset - offset > len); + conn->in.offset - offset, (int)len, conn->in.padding); + BUG_ON(conn->in.offset - offset > len); - if (tcp_conn->in.offset - offset != len) { + if (conn->in.offset - offset != len) { debug_tcp("continue to process %d bytes\n", - (int)len - (tcp_conn->in.offset - offset)); + (int)len - (conn->in.offset - offset)); goto more; } nomore: - processed = tcp_conn->in.offset - offset; + processed = conn->in.offset - offset; BUG_ON(processed == 0); return processed; again: - processed = tcp_conn->in.offset - offset; + processed = conn->in.offset - offset; debug_tcp("c, processed %d from out of %d rd_desc_cnt %d\n", processed, (int)len, (int)rd_desc->count); BUG_ON(processed == 0); @@ -953,14 +1220,9 @@ iscsi_tcp_data_ready(struct sock *sk, int flag) read_lock(&sk->sk_callback_lock); - /* - * Use rd_desc to pass 'conn' to iscsi_tcp_data_recv. - * We set count to 1 because we want the network layer to - * hand us all the skbs that are available. iscsi_tcp_data_recv - * handled pdus that cross buffers or pdus that still need data. - */ + /* use rd_desc to pass 'conn' to iscsi_tcp_data_recv */ rd_desc.arg.data = conn; - rd_desc.count = 1; + rd_desc.count = 0; tcp_read_sock(sk, &rd_desc, iscsi_tcp_data_recv); read_unlock(&sk->sk_callback_lock); @@ -969,7 +1231,6 @@ iscsi_tcp_data_ready(struct sock *sk, int flag) static void iscsi_tcp_state_change(struct sock *sk) { - struct iscsi_tcp_conn *tcp_conn; struct iscsi_conn *conn; struct iscsi_session *session; void (*old_state_change)(struct sock *); @@ -986,8 +1247,7 @@ iscsi_tcp_state_change(struct sock *sk) iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); } - tcp_conn = conn->dd_data; - old_state_change = tcp_conn->old_state_change; + old_state_change = conn->old_state_change; read_unlock(&sk->sk_callback_lock); @@ -1002,25 +1262,23 @@ static void iscsi_write_space(struct sock *sk) { struct iscsi_conn *conn = (struct iscsi_conn*)sk->sk_user_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - - tcp_conn->old_write_space(sk); + conn->old_write_space(sk); debug_tcp("iscsi_write_space: cid %d\n", conn->id); + clear_bit(SUSPEND_BIT, &conn->suspend_tx); scsi_queue_work(conn->session->host, &conn->xmitwork); } static void iscsi_conn_set_callbacks(struct iscsi_conn *conn) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct sock *sk = tcp_conn->sock->sk; + struct sock *sk = conn->sock->sk; /* assign new callbacks */ write_lock_bh(&sk->sk_callback_lock); sk->sk_user_data = conn; - tcp_conn->old_data_ready = sk->sk_data_ready; - tcp_conn->old_state_change = sk->sk_state_change; - tcp_conn->old_write_space = sk->sk_write_space; + conn->old_data_ready = sk->sk_data_ready; + conn->old_state_change = sk->sk_state_change; + conn->old_write_space = sk->sk_write_space; sk->sk_data_ready = iscsi_tcp_data_ready; sk->sk_state_change = iscsi_tcp_state_change; sk->sk_write_space = iscsi_write_space; @@ -1030,15 +1288,14 @@ iscsi_conn_set_callbacks(struct iscsi_conn *conn) static void iscsi_conn_restore_callbacks(struct iscsi_conn *conn) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct sock *sk = tcp_conn->sock->sk; + struct sock *sk = conn->sock->sk; /* restore socket callbacks, see also: iscsi_conn_set_callbacks() */ write_lock_bh(&sk->sk_callback_lock); sk->sk_user_data = NULL; - sk->sk_data_ready = tcp_conn->old_data_ready; - sk->sk_state_change = tcp_conn->old_state_change; - sk->sk_write_space = tcp_conn->old_write_space; + sk->sk_data_ready = conn->old_data_ready; + sk->sk_state_change = conn->old_state_change; + sk->sk_write_space = conn->old_write_space; sk->sk_no_check = 0; write_unlock_bh(&sk->sk_callback_lock); } @@ -1053,9 +1310,8 @@ iscsi_conn_restore_callbacks(struct iscsi_conn *conn) static inline int iscsi_send(struct iscsi_conn *conn, struct iscsi_buf *buf, int size, int flags) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct socket *sk = tcp_conn->sock; - int offset = buf->sg.offset + buf->sent, res; + struct socket *sk = conn->sock; + int offset = buf->sg.offset + buf->sent; /* * if we got use_sg=0 or are sending something we kmallocd @@ -1066,22 +1322,9 @@ iscsi_send(struct iscsi_conn *conn, struct iscsi_buf *buf, int size, int flags) * slab case. */ if (buf->use_sendmsg) - res = sock_no_sendpage(sk, buf->sg.page, offset, size, flags); - else - res = tcp_conn->sendpage(sk, buf->sg.page, offset, size, flags); - - if (res >= 0) { - conn->txdata_octets += res; - buf->sent += res; - return res; - } - - tcp_conn->sendpage_failures_cnt++; - if (res == -EAGAIN) - res = -ENOBUFS; + return sock_no_sendpage(sk, buf->sg.page, offset, size, flags); else - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - return res; + return conn->sendpage(sk, buf->sg.page, offset, size, flags); } /** @@ -1107,10 +1350,16 @@ iscsi_sendhdr(struct iscsi_conn *conn, struct iscsi_buf *buf, int datalen) res = iscsi_send(conn, buf, size, flags); debug_tcp("sendhdr %d bytes, sent %d res %d\n", size, buf->sent, res); if (res >= 0) { + conn->txdata_octets += res; + buf->sent += res; if (size != res) return -EAGAIN; return 0; - } + } else if (res == -EAGAIN) { + conn->sendpage_failures_cnt++; + set_bit(SUSPEND_BIT, &conn->suspend_tx); + } else if (res == -EPIPE) + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); return res; } @@ -1143,46 +1392,47 @@ iscsi_sendpage(struct iscsi_conn *conn, struct iscsi_buf *buf, debug_tcp("sendpage: %d bytes, sent %d left %d sent %d res %d\n", size, buf->sent, *count, *sent, res); if (res >= 0) { + conn->txdata_octets += res; + buf->sent += res; *count -= res; *sent += res; if (size != res) return -EAGAIN; return 0; - } + } else if (res == -EAGAIN) { + conn->sendpage_failures_cnt++; + set_bit(SUSPEND_BIT, &conn->suspend_tx); + } else if (res == -EPIPE) + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); return res; } static inline void -iscsi_data_digest_init(struct iscsi_tcp_conn *tcp_conn, - struct iscsi_cmd_task *ctask) +iscsi_data_digest_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - - BUG_ON(!tcp_conn->data_tx_tfm); - crypto_digest_init(tcp_conn->data_tx_tfm); - tcp_ctask->digest_count = 4; + BUG_ON(!conn->data_tx_tfm); + crypto_digest_init(conn->data_tx_tfm); + ctask->digest_count = 4; } static int iscsi_digest_final_send(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, struct iscsi_buf *buf, uint32_t *digest, int final) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; int rc = 0; int sent = 0; if (final) - crypto_digest_final(tcp_conn->data_tx_tfm, (u8*)digest); + crypto_digest_final(conn->data_tx_tfm, (u8*)digest); - iscsi_buf_init_iov(buf, (char*)digest, 4); - rc = iscsi_sendpage(conn, buf, &tcp_ctask->digest_count, &sent); + iscsi_buf_init_virt(buf, (char*)digest, 4); + rc = iscsi_sendpage(conn, buf, &ctask->digest_count, &sent); if (rc) { - tcp_ctask->datadigest = *digest; - tcp_ctask->xmstate |= XMSTATE_DATA_DIGEST; + ctask->datadigest = *digest; + ctask->xmstate |= XMSTATE_DATA_DIGEST; } else - tcp_ctask->digest_count = 4; + ctask->digest_count = 4; return rc; } @@ -1203,19 +1453,21 @@ static void iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, struct iscsi_r2t_info *r2t, int left) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; struct iscsi_data *hdr; + struct iscsi_data_task *dtask; struct scsi_cmnd *sc = ctask->sc; int new_offset; - hdr = &r2t->dtask.hdr; + dtask = mempool_alloc(ctask->datapool, GFP_ATOMIC); + BUG_ON(!dtask); + hdr = &dtask->hdr; memset(hdr, 0, sizeof(struct iscsi_data)); hdr->ttt = r2t->ttt; hdr->datasn = cpu_to_be32(r2t->solicit_datasn); r2t->solicit_datasn++; hdr->opcode = ISCSI_OP_SCSI_DATA_OUT; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - hdr->itt = ctask->hdr->itt; + memcpy(hdr->lun, ctask->hdr.lun, sizeof(hdr->lun)); + hdr->itt = ctask->hdr.itt; hdr->exp_statsn = r2t->exp_statsn; new_offset = r2t->data_offset + r2t->sent; hdr->offset = cpu_to_be32(new_offset); @@ -1229,98 +1481,181 @@ iscsi_solicit_data_cont(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, } conn->dataout_pdus_cnt++; - iscsi_buf_init_iov(&r2t->headbuf, (char*)hdr, + iscsi_buf_init_virt(&r2t->headbuf, (char*)hdr, sizeof(struct iscsi_hdr)); + r2t->dtask = dtask; + if (sc->use_sg && !iscsi_buf_left(&r2t->sendbuf)) { - BUG_ON(tcp_ctask->bad_sg == r2t->sg); + BUG_ON(ctask->bad_sg == r2t->sg); iscsi_buf_init_sg(&r2t->sendbuf, r2t->sg); r2t->sg += 1; } else - iscsi_buf_init_iov(&tcp_ctask->sendbuf, + iscsi_buf_init_iov(&ctask->sendbuf, (char*)sc->request_buffer + new_offset, r2t->data_count); + + list_add(&dtask->item, &ctask->dataqueue); } static void iscsi_unsolicit_data_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; + struct iscsi_data *hdr; struct iscsi_data_task *dtask; - dtask = tcp_ctask->dtask = &tcp_ctask->unsol_dtask; - iscsi_prep_unsolicit_data_pdu(ctask, &dtask->hdr, - tcp_ctask->r2t_data_count); - iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)&dtask->hdr, + dtask = mempool_alloc(ctask->datapool, GFP_ATOMIC); + BUG_ON(!dtask); + hdr = &dtask->hdr; + memset(hdr, 0, sizeof(struct iscsi_data)); + hdr->ttt = cpu_to_be32(ISCSI_RESERVED_TAG); + hdr->datasn = cpu_to_be32(ctask->unsol_datasn); + ctask->unsol_datasn++; + hdr->opcode = ISCSI_OP_SCSI_DATA_OUT; + memcpy(hdr->lun, ctask->hdr.lun, sizeof(hdr->lun)); + hdr->itt = ctask->hdr.itt; + hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); + hdr->offset = cpu_to_be32(ctask->total_length - + ctask->r2t_data_count - + ctask->unsol_count); + if (ctask->unsol_count > conn->max_xmit_dlength) { + hton24(hdr->dlength, conn->max_xmit_dlength); + ctask->data_count = conn->max_xmit_dlength; + hdr->flags = 0; + } else { + hton24(hdr->dlength, ctask->unsol_count); + ctask->data_count = ctask->unsol_count; + hdr->flags = ISCSI_FLAG_CMD_FINAL; + } + + iscsi_buf_init_virt(&ctask->headbuf, (char*)hdr, sizeof(struct iscsi_hdr)); + + list_add(&dtask->item, &ctask->dataqueue); + + ctask->dtask = dtask; } /** - * iscsi_tcp_cmd_init - Initialize iSCSI SCSI_READ or SCSI_WRITE commands + * iscsi_cmd_init - Initialize iSCSI SCSI_READ or SCSI_WRITE commands * @conn: iscsi connection * @ctask: scsi command task * @sc: scsi command **/ static void -iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask) +iscsi_cmd_init(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, + struct scsi_cmnd *sc) { - struct scsi_cmnd *sc = ctask->sc; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; + struct iscsi_session *session = conn->session; - BUG_ON(__kfifo_len(tcp_ctask->r2tqueue)); + BUG_ON(__kfifo_len(ctask->r2tqueue)); - tcp_ctask->sent = 0; - tcp_ctask->sg_count = 0; + ctask->sc = sc; + ctask->conn = conn; + ctask->hdr.opcode = ISCSI_OP_SCSI_CMD; + ctask->hdr.flags = ISCSI_ATTR_SIMPLE; + int_to_scsilun(sc->device->lun, (struct scsi_lun *)ctask->hdr.lun); + ctask->hdr.itt = ctask->itt | (conn->id << CID_SHIFT) | + (session->age << AGE_SHIFT); + ctask->hdr.data_length = cpu_to_be32(sc->request_bufflen); + ctask->hdr.cmdsn = cpu_to_be32(session->cmdsn); session->cmdsn++; + ctask->hdr.exp_statsn = cpu_to_be32(conn->exp_statsn); + memcpy(ctask->hdr.cdb, sc->cmnd, sc->cmd_len); + memset(&ctask->hdr.cdb[sc->cmd_len], 0, MAX_COMMAND_SIZE - sc->cmd_len); + + ctask->mtask = NULL; + ctask->sent = 0; + ctask->sg_count = 0; + + ctask->total_length = sc->request_bufflen; if (sc->sc_data_direction == DMA_TO_DEVICE) { - tcp_ctask->xmstate = XMSTATE_W_HDR; - tcp_ctask->exp_r2tsn = 0; + ctask->exp_r2tsn = 0; + ctask->hdr.flags |= ISCSI_FLAG_CMD_WRITE; BUG_ON(ctask->total_length == 0); - if (sc->use_sg) { struct scatterlist *sg = sc->request_buffer; - iscsi_buf_init_sg(&tcp_ctask->sendbuf, - &sg[tcp_ctask->sg_count++]); - tcp_ctask->sg = sg; - tcp_ctask->bad_sg = sg + sc->use_sg; - } else - iscsi_buf_init_iov(&tcp_ctask->sendbuf, - sc->request_buffer, - sc->request_bufflen); - - if (ctask->imm_count) - tcp_ctask->xmstate |= XMSTATE_IMM_DATA; + iscsi_buf_init_sg(&ctask->sendbuf, + &sg[ctask->sg_count++]); + ctask->sg = sg; + ctask->bad_sg = sg + sc->use_sg; + } else { + iscsi_buf_init_iov(&ctask->sendbuf, sc->request_buffer, + sc->request_bufflen); + } - tcp_ctask->pad_count = ctask->total_length & (ISCSI_PAD_LEN-1); - if (tcp_ctask->pad_count) { - tcp_ctask->pad_count = ISCSI_PAD_LEN - - tcp_ctask->pad_count; + /* + * Write counters: + * + * imm_count bytes to be sent right after + * SCSI PDU Header + * + * unsol_count bytes(as Data-Out) to be sent + * without R2T ack right after + * immediate data + * + * r2t_data_count bytes to be sent via R2T ack's + * + * pad_count bytes to be sent as zero-padding + */ + ctask->imm_count = 0; + ctask->unsol_count = 0; + ctask->unsol_datasn = 0; + ctask->xmstate = XMSTATE_W_HDR; + /* calculate write padding */ + ctask->pad_count = ctask->total_length & (ISCSI_PAD_LEN-1); + if (ctask->pad_count) { + ctask->pad_count = ISCSI_PAD_LEN - ctask->pad_count; debug_scsi("write padding %d bytes\n", - tcp_ctask->pad_count); - tcp_ctask->xmstate |= XMSTATE_W_PAD; + ctask->pad_count); + ctask->xmstate |= XMSTATE_W_PAD; } + if (session->imm_data_en) { + if (ctask->total_length >= session->first_burst) + ctask->imm_count = min(session->first_burst, + conn->max_xmit_dlength); + else + ctask->imm_count = min(ctask->total_length, + conn->max_xmit_dlength); + hton24(ctask->hdr.dlength, ctask->imm_count); + ctask->xmstate |= XMSTATE_IMM_DATA; + } else + zero_data(ctask->hdr.dlength); + + if (!session->initial_r2t_en) + ctask->unsol_count = min(session->first_burst, + ctask->total_length) - ctask->imm_count; + if (!ctask->unsol_count) + /* No unsolicit Data-Out's */ + ctask->hdr.flags |= ISCSI_FLAG_CMD_FINAL; + else + ctask->xmstate |= XMSTATE_UNS_HDR | XMSTATE_UNS_INIT; - if (ctask->unsol_count) - tcp_ctask->xmstate |= XMSTATE_UNS_HDR | - XMSTATE_UNS_INIT; - tcp_ctask->r2t_data_count = ctask->total_length - + ctask->r2t_data_count = ctask->total_length - ctask->imm_count - ctask->unsol_count; debug_scsi("cmd [itt %x total %d imm %d imm_data %d " "r2t_data %d]\n", ctask->itt, ctask->total_length, ctask->imm_count, - ctask->unsol_count, tcp_ctask->r2t_data_count); - } else - tcp_ctask->xmstate = XMSTATE_R_HDR; + ctask->unsol_count, ctask->r2t_data_count); + } else { + ctask->hdr.flags |= ISCSI_FLAG_CMD_FINAL; + if (sc->sc_data_direction == DMA_FROM_DEVICE) + ctask->hdr.flags |= ISCSI_FLAG_CMD_READ; + ctask->datasn = 0; + ctask->xmstate = XMSTATE_R_HDR; + zero_data(ctask->hdr.dlength); + } - iscsi_buf_init_iov(&tcp_ctask->headbuf, (char*)ctask->hdr, + iscsi_buf_init_virt(&ctask->headbuf, (char*)&ctask->hdr, sizeof(struct iscsi_hdr)); + conn->scsicmd_pdus_cnt++; } /** - * iscsi_tcp_mtask_xmit - xmit management(immediate) task + * iscsi_mtask_xmit - xmit management(immediate) task * @conn: iscsi connection * @mtask: task management task * @@ -1334,167 +1669,132 @@ iscsi_tcp_cmd_init(struct iscsi_cmd_task *ctask) * IN_PROGRESS_IMM_DATA - PDU Data xmit in progress **/ static int -iscsi_tcp_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask) +iscsi_mtask_xmit(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask) { - struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data; - int rc; debug_scsi("mtask deq [cid %d state %x itt 0x%x]\n", - conn->id, tcp_mtask->xmstate, mtask->itt); + conn->id, mtask->xmstate, mtask->itt); - if (tcp_mtask->xmstate & XMSTATE_IMM_HDR) { - tcp_mtask->xmstate &= ~XMSTATE_IMM_HDR; + if (mtask->xmstate & XMSTATE_IMM_HDR) { + mtask->xmstate &= ~XMSTATE_IMM_HDR; if (mtask->data_count) - tcp_mtask->xmstate |= XMSTATE_IMM_DATA; + mtask->xmstate |= XMSTATE_IMM_DATA; if (conn->c_stage != ISCSI_CONN_INITIAL_STAGE && - conn->stop_stage != STOP_CONN_RECOVER && + conn->stop_stage != STOP_CONN_RECOVER && conn->hdrdgst_en) - iscsi_hdr_digest(conn, &tcp_mtask->headbuf, - (u8*)tcp_mtask->hdrext); - rc = iscsi_sendhdr(conn, &tcp_mtask->headbuf, - mtask->data_count); - if (rc) { - tcp_mtask->xmstate |= XMSTATE_IMM_HDR; + iscsi_hdr_digest(conn, &mtask->headbuf, + (u8*)mtask->hdrext); + if (iscsi_sendhdr(conn, &mtask->headbuf, mtask->data_count)) { + mtask->xmstate |= XMSTATE_IMM_HDR; if (mtask->data_count) - tcp_mtask->xmstate &= ~XMSTATE_IMM_DATA; - return rc; + mtask->xmstate &= ~XMSTATE_IMM_DATA; + return -EAGAIN; } } - if (tcp_mtask->xmstate & XMSTATE_IMM_DATA) { + if (mtask->xmstate & XMSTATE_IMM_DATA) { BUG_ON(!mtask->data_count); - tcp_mtask->xmstate &= ~XMSTATE_IMM_DATA; + mtask->xmstate &= ~XMSTATE_IMM_DATA; /* FIXME: implement. * Virtual buffer could be spreaded across multiple pages... */ do { - int rc; - - rc = iscsi_sendpage(conn, &tcp_mtask->sendbuf, - &mtask->data_count, &tcp_mtask->sent); - if (rc) { - tcp_mtask->xmstate |= XMSTATE_IMM_DATA; - return rc; + if (iscsi_sendpage(conn, &mtask->sendbuf, + &mtask->data_count, &mtask->sent)) { + mtask->xmstate |= XMSTATE_IMM_DATA; + return -EAGAIN; } } while (mtask->data_count); } - BUG_ON(tcp_mtask->xmstate != XMSTATE_IDLE); - if (mtask->hdr->itt == cpu_to_be32(ISCSI_RESERVED_TAG)) { - struct iscsi_session *session = conn->session; - - spin_lock_bh(&session->lock); - list_del(&conn->mtask->running); - __kfifo_put(session->mgmtpool.queue, (void*)&conn->mtask, - sizeof(void*)); - spin_unlock_bh(&session->lock); - } + BUG_ON(mtask->xmstate != XMSTATE_IDLE); return 0; } static inline int -handle_xmstate_r_hdr(struct iscsi_conn *conn, - struct iscsi_tcp_cmd_task *tcp_ctask) +handle_xmstate_r_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - int rc; - - tcp_ctask->xmstate &= ~XMSTATE_R_HDR; + ctask->xmstate &= ~XMSTATE_R_HDR; if (conn->hdrdgst_en) - iscsi_hdr_digest(conn, &tcp_ctask->headbuf, - (u8*)tcp_ctask->hdrext); - rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, 0); - if (!rc) { - BUG_ON(tcp_ctask->xmstate != XMSTATE_IDLE); + iscsi_hdr_digest(conn, &ctask->headbuf, (u8*)ctask->hdrext); + if (!iscsi_sendhdr(conn, &ctask->headbuf, 0)) { + BUG_ON(ctask->xmstate != XMSTATE_IDLE); return 0; /* wait for Data-In */ } - tcp_ctask->xmstate |= XMSTATE_R_HDR; - return rc; + ctask->xmstate |= XMSTATE_R_HDR; + return -EAGAIN; } static inline int -handle_xmstate_w_hdr(struct iscsi_conn *conn, - struct iscsi_cmd_task *ctask) +handle_xmstate_w_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - int rc; - - tcp_ctask->xmstate &= ~XMSTATE_W_HDR; + ctask->xmstate &= ~XMSTATE_W_HDR; if (conn->hdrdgst_en) - iscsi_hdr_digest(conn, &tcp_ctask->headbuf, - (u8*)tcp_ctask->hdrext); - rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->imm_count); - if (rc) - tcp_ctask->xmstate |= XMSTATE_W_HDR; - return rc; + iscsi_hdr_digest(conn, &ctask->headbuf, (u8*)ctask->hdrext); + if (iscsi_sendhdr(conn, &ctask->headbuf, ctask->imm_count)) { + ctask->xmstate |= XMSTATE_W_HDR; + return -EAGAIN; + } + return 0; } static inline int handle_xmstate_data_digest(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - int rc; - - tcp_ctask->xmstate &= ~XMSTATE_DATA_DIGEST; - debug_tcp("resent data digest 0x%x\n", tcp_ctask->datadigest); - rc = iscsi_digest_final_send(conn, ctask, &tcp_ctask->immbuf, - &tcp_ctask->datadigest, 0); - if (rc) { - tcp_ctask->xmstate |= XMSTATE_DATA_DIGEST; + ctask->xmstate &= ~XMSTATE_DATA_DIGEST; + debug_tcp("resent data digest 0x%x\n", ctask->datadigest); + if (iscsi_digest_final_send(conn, ctask, &ctask->immbuf, + &ctask->datadigest, 0)) { + ctask->xmstate |= XMSTATE_DATA_DIGEST; debug_tcp("resent data digest 0x%x fail!\n", - tcp_ctask->datadigest); + ctask->datadigest); + return -EAGAIN; } - - return rc; + return 0; } static inline int handle_xmstate_imm_data(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - int rc; - BUG_ON(!ctask->imm_count); - tcp_ctask->xmstate &= ~XMSTATE_IMM_DATA; + ctask->xmstate &= ~XMSTATE_IMM_DATA; if (conn->datadgst_en) { - iscsi_data_digest_init(tcp_conn, ctask); - tcp_ctask->immdigest = 0; + iscsi_data_digest_init(conn, ctask); + ctask->immdigest = 0; } for (;;) { - rc = iscsi_sendpage(conn, &tcp_ctask->sendbuf, - &ctask->imm_count, &tcp_ctask->sent); - if (rc) { - tcp_ctask->xmstate |= XMSTATE_IMM_DATA; + if (iscsi_sendpage(conn, &ctask->sendbuf, &ctask->imm_count, + &ctask->sent)) { + ctask->xmstate |= XMSTATE_IMM_DATA; if (conn->datadgst_en) { - crypto_digest_final(tcp_conn->data_tx_tfm, - (u8*)&tcp_ctask->immdigest); + crypto_digest_final(conn->data_tx_tfm, + (u8*)&ctask->immdigest); debug_tcp("tx imm sendpage fail 0x%x\n", - tcp_ctask->datadigest); + ctask->datadigest); } - return rc; + return -EAGAIN; } if (conn->datadgst_en) - crypto_digest_update(tcp_conn->data_tx_tfm, - &tcp_ctask->sendbuf.sg, 1); + crypto_digest_update(conn->data_tx_tfm, + &ctask->sendbuf.sg, 1); if (!ctask->imm_count) break; - iscsi_buf_init_sg(&tcp_ctask->sendbuf, - &tcp_ctask->sg[tcp_ctask->sg_count++]); + iscsi_buf_init_sg(&ctask->sendbuf, + &ctask->sg[ctask->sg_count++]); } - if (conn->datadgst_en && !(tcp_ctask->xmstate & XMSTATE_W_PAD)) { - rc = iscsi_digest_final_send(conn, ctask, &tcp_ctask->immbuf, - &tcp_ctask->immdigest, 1); - if (rc) { + if (conn->datadgst_en && !(ctask->xmstate & XMSTATE_W_PAD)) { + if (iscsi_digest_final_send(conn, ctask, &ctask->immbuf, + &ctask->immdigest, 1)) { debug_tcp("sending imm digest 0x%x fail!\n", - tcp_ctask->immdigest); - return rc; + ctask->immdigest); + return -EAGAIN; } - debug_tcp("sending imm digest 0x%x\n", tcp_ctask->immdigest); + debug_tcp("sending imm digest 0x%x\n", ctask->immdigest); } return 0; @@ -1503,81 +1803,74 @@ handle_xmstate_imm_data(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) static inline int handle_xmstate_uns_hdr(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; struct iscsi_data_task *dtask; - int rc; - tcp_ctask->xmstate |= XMSTATE_UNS_DATA; - if (tcp_ctask->xmstate & XMSTATE_UNS_INIT) { + ctask->xmstate |= XMSTATE_UNS_DATA; + if (ctask->xmstate & XMSTATE_UNS_INIT) { iscsi_unsolicit_data_init(conn, ctask); - dtask = tcp_ctask->dtask; + BUG_ON(!ctask->dtask); + dtask = ctask->dtask; if (conn->hdrdgst_en) - iscsi_hdr_digest(conn, &tcp_ctask->headbuf, + iscsi_hdr_digest(conn, &ctask->headbuf, (u8*)dtask->hdrext); - tcp_ctask->xmstate &= ~XMSTATE_UNS_INIT; + ctask->xmstate &= ~XMSTATE_UNS_INIT; } - - rc = iscsi_sendhdr(conn, &tcp_ctask->headbuf, ctask->data_count); - if (rc) { - tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA; - tcp_ctask->xmstate |= XMSTATE_UNS_HDR; - return rc; + if (iscsi_sendhdr(conn, &ctask->headbuf, ctask->data_count)) { + ctask->xmstate &= ~XMSTATE_UNS_DATA; + ctask->xmstate |= XMSTATE_UNS_HDR; + return -EAGAIN; } debug_scsi("uns dout [itt 0x%x dlen %d sent %d]\n", - ctask->itt, ctask->unsol_count, tcp_ctask->sent); + ctask->itt, ctask->unsol_count, ctask->sent); return 0; } static inline int handle_xmstate_uns_data(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_data_task *dtask = tcp_ctask->dtask; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - int rc; + struct iscsi_data_task *dtask = ctask->dtask; BUG_ON(!ctask->data_count); - tcp_ctask->xmstate &= ~XMSTATE_UNS_DATA; + ctask->xmstate &= ~XMSTATE_UNS_DATA; if (conn->datadgst_en) { - iscsi_data_digest_init(tcp_conn, ctask); + iscsi_data_digest_init(conn, ctask); dtask->digest = 0; } for (;;) { - int start = tcp_ctask->sent; + int start = ctask->sent; - rc = iscsi_sendpage(conn, &tcp_ctask->sendbuf, - &ctask->data_count, &tcp_ctask->sent); - if (rc) { - ctask->unsol_count -= tcp_ctask->sent - start; - tcp_ctask->xmstate |= XMSTATE_UNS_DATA; + if (iscsi_sendpage(conn, &ctask->sendbuf, &ctask->data_count, + &ctask->sent)) { + ctask->unsol_count -= ctask->sent - start; + ctask->xmstate |= XMSTATE_UNS_DATA; /* will continue with this ctask later.. */ if (conn->datadgst_en) { - crypto_digest_final(tcp_conn->data_tx_tfm, + crypto_digest_final(conn->data_tx_tfm, (u8 *)&dtask->digest); debug_tcp("tx uns data fail 0x%x\n", dtask->digest); } - return rc; + return -EAGAIN; } - BUG_ON(tcp_ctask->sent > ctask->total_length); - ctask->unsol_count -= tcp_ctask->sent - start; + BUG_ON(ctask->sent > ctask->total_length); + ctask->unsol_count -= ctask->sent - start; /* * XXX:we may run here with un-initial sendbuf. * so pass it */ - if (conn->datadgst_en && tcp_ctask->sent - start > 0) - crypto_digest_update(tcp_conn->data_tx_tfm, - &tcp_ctask->sendbuf.sg, 1); + if (conn->datadgst_en && ctask->sent - start > 0) + crypto_digest_update(conn->data_tx_tfm, + &ctask->sendbuf.sg, 1); if (!ctask->data_count) break; - iscsi_buf_init_sg(&tcp_ctask->sendbuf, - &tcp_ctask->sg[tcp_ctask->sg_count++]); + iscsi_buf_init_sg(&ctask->sendbuf, + &ctask->sg[ctask->sg_count++]); } BUG_ON(ctask->unsol_count < 0); @@ -1587,29 +1880,27 @@ handle_xmstate_uns_data(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) */ if (ctask->unsol_count) { if (conn->datadgst_en) { - rc = iscsi_digest_final_send(conn, ctask, + if (iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, - &dtask->digest, 1); - if (rc) { + &dtask->digest, 1)) { debug_tcp("send uns digest 0x%x fail\n", dtask->digest); - return rc; + return -EAGAIN; } debug_tcp("sending uns digest 0x%x, more uns\n", dtask->digest); } - tcp_ctask->xmstate |= XMSTATE_UNS_INIT; + ctask->xmstate |= XMSTATE_UNS_INIT; return 1; } - if (conn->datadgst_en && !(tcp_ctask->xmstate & XMSTATE_W_PAD)) { - rc = iscsi_digest_final_send(conn, ctask, + if (conn->datadgst_en && !(ctask->xmstate & XMSTATE_W_PAD)) { + if (iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, - &dtask->digest, 1); - if (rc) { + &dtask->digest, 1)) { debug_tcp("send last uns digest 0x%x fail\n", dtask->digest); - return rc; + return -EAGAIN; } debug_tcp("sending uns digest 0x%x\n",dtask->digest); } @@ -1621,17 +1912,15 @@ static inline int handle_xmstate_sol_data(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { struct iscsi_session *session = conn->session; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_r2t_info *r2t = tcp_ctask->r2t; - struct iscsi_data_task *dtask = &r2t->dtask; - int left, rc; + struct iscsi_r2t_info *r2t = ctask->r2t; + struct iscsi_data_task *dtask = r2t->dtask; + int left; - tcp_ctask->xmstate &= ~XMSTATE_SOL_DATA; - tcp_ctask->dtask = dtask; + ctask->xmstate &= ~XMSTATE_SOL_DATA; + ctask->dtask = dtask; if (conn->datadgst_en) { - iscsi_data_digest_init(tcp_conn, ctask); + iscsi_data_digest_init(conn, ctask); dtask->digest = 0; } solicit_again: @@ -1641,27 +1930,25 @@ solicit_again: if (!r2t->data_count) goto data_out_done; - rc = iscsi_sendpage(conn, &r2t->sendbuf, &r2t->data_count, &r2t->sent); - if (rc) { - tcp_ctask->xmstate |= XMSTATE_SOL_DATA; + if (iscsi_sendpage(conn, &r2t->sendbuf, &r2t->data_count, &r2t->sent)) { + ctask->xmstate |= XMSTATE_SOL_DATA; /* will continue with this ctask later.. */ if (conn->datadgst_en) { - crypto_digest_final(tcp_conn->data_tx_tfm, + crypto_digest_final(conn->data_tx_tfm, (u8 *)&dtask->digest); debug_tcp("r2t data send fail 0x%x\n", dtask->digest); } - return rc; + return -EAGAIN; } BUG_ON(r2t->data_count < 0); if (conn->datadgst_en) - crypto_digest_update(tcp_conn->data_tx_tfm, &r2t->sendbuf.sg, - 1); + crypto_digest_update(conn->data_tx_tfm, &r2t->sendbuf.sg, 1); if (r2t->data_count) { BUG_ON(ctask->sc->use_sg == 0); if (!iscsi_buf_left(&r2t->sendbuf)) { - BUG_ON(tcp_ctask->bad_sg == r2t->sg); + BUG_ON(ctask->bad_sg == r2t->sg); iscsi_buf_init_sg(&r2t->sendbuf, r2t->sg); r2t->sg += 1; } @@ -1677,20 +1964,19 @@ data_out_done: left = r2t->data_length - r2t->sent; if (left) { if (conn->datadgst_en) { - rc = iscsi_digest_final_send(conn, ctask, + if (iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, - &dtask->digest, 1); - if (rc) { + &dtask->digest, 1)) { debug_tcp("send r2t data digest 0x%x" "fail\n", dtask->digest); - return rc; + return -EAGAIN; } debug_tcp("r2t data send digest 0x%x\n", dtask->digest); } iscsi_solicit_data_cont(conn, ctask, r2t, left); - tcp_ctask->xmstate |= XMSTATE_SOL_DATA; - tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR; + ctask->xmstate |= XMSTATE_SOL_DATA; + ctask->xmstate &= ~XMSTATE_SOL_HDR; return 1; } @@ -1698,27 +1984,26 @@ data_out_done: * Done with this R2T. Check if there are more * outstanding R2Ts ready to be processed. */ - BUG_ON(tcp_ctask->r2t_data_count - r2t->data_length < 0); + BUG_ON(ctask->r2t_data_count - r2t->data_length < 0); if (conn->datadgst_en) { - rc = iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, - &dtask->digest, 1); - if (rc) { + if (iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, + &dtask->digest, 1)) { debug_tcp("send last r2t data digest 0x%x" "fail\n", dtask->digest); - return rc; + return -EAGAIN; } debug_tcp("r2t done dout digest 0x%x\n", dtask->digest); } - tcp_ctask->r2t_data_count -= r2t->data_length; - tcp_ctask->r2t = NULL; + ctask->r2t_data_count -= r2t->data_length; + ctask->r2t = NULL; spin_lock_bh(&session->lock); - __kfifo_put(tcp_ctask->r2tpool.queue, (void*)&r2t, sizeof(void*)); + __kfifo_put(ctask->r2tpool.queue, (void*)&r2t, sizeof(void*)); spin_unlock_bh(&session->lock); - if (__kfifo_get(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*))) { - tcp_ctask->r2t = r2t; - tcp_ctask->xmstate |= XMSTATE_SOL_DATA; - tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR; + if (__kfifo_get(ctask->r2tqueue, (void*)&r2t, sizeof(void*))) { + ctask->r2t = r2t; + ctask->xmstate |= XMSTATE_SOL_DATA; + ctask->xmstate &= ~XMSTATE_SOL_HDR; return 1; } @@ -1728,44 +2013,36 @@ data_out_done: static inline int handle_xmstate_w_pad(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct iscsi_data_task *dtask = tcp_ctask->dtask; - int sent, rc; - - tcp_ctask->xmstate &= ~XMSTATE_W_PAD; - iscsi_buf_init_iov(&tcp_ctask->sendbuf, (char*)&tcp_ctask->pad, - tcp_ctask->pad_count); - rc = iscsi_sendpage(conn, &tcp_ctask->sendbuf, &tcp_ctask->pad_count, - &sent); - if (rc) { - tcp_ctask->xmstate |= XMSTATE_W_PAD; - return rc; + struct iscsi_data_task *dtask = ctask->dtask; + int sent; + + ctask->xmstate &= ~XMSTATE_W_PAD; + iscsi_buf_init_virt(&ctask->sendbuf, (char*)&ctask->pad, + ctask->pad_count); + if (iscsi_sendpage(conn, &ctask->sendbuf, &ctask->pad_count, &sent)) { + ctask->xmstate |= XMSTATE_W_PAD; + return -EAGAIN; } if (conn->datadgst_en) { - crypto_digest_update(tcp_conn->data_tx_tfm, - &tcp_ctask->sendbuf.sg, 1); + crypto_digest_update(conn->data_tx_tfm, &ctask->sendbuf.sg, 1); /* imm data? */ if (!dtask) { - rc = iscsi_digest_final_send(conn, ctask, - &tcp_ctask->immbuf, - &tcp_ctask->immdigest, 1); - if (rc) { + if (iscsi_digest_final_send(conn, ctask, &ctask->immbuf, + &ctask->immdigest, 1)) { debug_tcp("send padding digest 0x%x" - "fail!\n", tcp_ctask->immdigest); - return rc; + "fail!\n", ctask->immdigest); + return -EAGAIN; } debug_tcp("done with padding, digest 0x%x\n", - tcp_ctask->datadigest); + ctask->datadigest); } else { - rc = iscsi_digest_final_send(conn, ctask, + if (iscsi_digest_final_send(conn, ctask, &dtask->digestbuf, - &dtask->digest, 1); - if (rc) { + &dtask->digest, 1)) { debug_tcp("send padding digest 0x%x" "fail\n", dtask->digest); - return rc; + return -EAGAIN; } debug_tcp("done with padding, digest 0x%x\n", dtask->digest); @@ -1776,13 +2053,12 @@ handle_xmstate_w_pad(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) } static int -iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) +iscsi_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; int rc = 0; debug_scsi("ctask deq [cid %d xmstate %x itt 0x%x]\n", - conn->id, tcp_ctask->xmstate, ctask->itt); + conn->id, ctask->xmstate, ctask->itt); /* * serialize with TMF AbortTask @@ -1790,38 +2066,40 @@ iscsi_tcp_ctask_xmit(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) if (ctask->mtask) return rc; - if (tcp_ctask->xmstate & XMSTATE_R_HDR) - return handle_xmstate_r_hdr(conn, tcp_ctask); + if (ctask->xmstate & XMSTATE_R_HDR) { + rc = handle_xmstate_r_hdr(conn, ctask); + return rc; + } - if (tcp_ctask->xmstate & XMSTATE_W_HDR) { + if (ctask->xmstate & XMSTATE_W_HDR) { rc = handle_xmstate_w_hdr(conn, ctask); if (rc) return rc; } /* XXX: for data digest xmit recover */ - if (tcp_ctask->xmstate & XMSTATE_DATA_DIGEST) { + if (ctask->xmstate & XMSTATE_DATA_DIGEST) { rc = handle_xmstate_data_digest(conn, ctask); if (rc) return rc; } - if (tcp_ctask->xmstate & XMSTATE_IMM_DATA) { + if (ctask->xmstate & XMSTATE_IMM_DATA) { rc = handle_xmstate_imm_data(conn, ctask); if (rc) return rc; } - if (tcp_ctask->xmstate & XMSTATE_UNS_HDR) { + if (ctask->xmstate & XMSTATE_UNS_HDR) { BUG_ON(!ctask->unsol_count); - tcp_ctask->xmstate &= ~XMSTATE_UNS_HDR; + ctask->xmstate &= ~XMSTATE_UNS_HDR; unsolicit_head_again: rc = handle_xmstate_uns_hdr(conn, ctask); if (rc) return rc; } - if (tcp_ctask->xmstate & XMSTATE_UNS_DATA) { + if (ctask->xmstate & XMSTATE_UNS_DATA) { rc = handle_xmstate_uns_data(conn, ctask); if (rc == 1) goto unsolicit_head_again; @@ -1830,24 +2108,23 @@ unsolicit_head_again: goto done; } - if (tcp_ctask->xmstate & XMSTATE_SOL_HDR) { + if (ctask->xmstate & XMSTATE_SOL_HDR) { struct iscsi_r2t_info *r2t; - tcp_ctask->xmstate &= ~XMSTATE_SOL_HDR; - tcp_ctask->xmstate |= XMSTATE_SOL_DATA; - if (!tcp_ctask->r2t) - __kfifo_get(tcp_ctask->r2tqueue, (void*)&tcp_ctask->r2t, + ctask->xmstate &= ~XMSTATE_SOL_HDR; + ctask->xmstate |= XMSTATE_SOL_DATA; + if (!ctask->r2t) + __kfifo_get(ctask->r2tqueue, (void*)&ctask->r2t, sizeof(void*)); solicit_head_again: - r2t = tcp_ctask->r2t; + r2t = ctask->r2t; if (conn->hdrdgst_en) iscsi_hdr_digest(conn, &r2t->headbuf, - (u8*)r2t->dtask.hdrext); - rc = iscsi_sendhdr(conn, &r2t->headbuf, r2t->data_count); - if (rc) { - tcp_ctask->xmstate &= ~XMSTATE_SOL_DATA; - tcp_ctask->xmstate |= XMSTATE_SOL_HDR; - return rc; + (u8*)r2t->dtask->hdrext); + if (iscsi_sendhdr(conn, &r2t->headbuf, r2t->data_count)) { + ctask->xmstate &= ~XMSTATE_SOL_DATA; + ctask->xmstate |= XMSTATE_SOL_HDR; + return -EAGAIN; } debug_scsi("sol dout [dsn %d itt 0x%x dlen %d sent %d]\n", @@ -1855,7 +2132,7 @@ solicit_head_again: r2t->sent); } - if (tcp_ctask->xmstate & XMSTATE_SOL_DATA) { + if (ctask->xmstate & XMSTATE_SOL_DATA) { rc = handle_xmstate_sol_data(conn, ctask); if (rc == 1) goto solicit_head_again; @@ -1868,199 +2145,1022 @@ done: * Last thing to check is whether we need to send write * padding. Note that we check for xmstate equality, not just the bit. */ - if (tcp_ctask->xmstate == XMSTATE_W_PAD) + if (ctask->xmstate == XMSTATE_W_PAD) rc = handle_xmstate_w_pad(conn, ctask); return rc; } +/** + * iscsi_data_xmit - xmit any command into the scheduled connection + * @conn: iscsi connection + * + * Notes: + * The function can return -EAGAIN in which case the caller must + * re-schedule it again later or recover. '0' return code means + * successful xmit. + **/ +static int +iscsi_data_xmit(struct iscsi_conn *conn) +{ + if (unlikely(conn->suspend_tx)) { + debug_tcp("conn %d Tx suspended!\n", conn->id); + return 0; + } + + /* + * Transmit in the following order: + * + * 1) un-finished xmit (ctask or mtask) + * 2) immediate control PDUs + * 3) write data + * 4) SCSI commands + * 5) non-immediate control PDUs + * + * No need to lock around __kfifo_get as long as + * there's one producer and one consumer. + */ + + BUG_ON(conn->ctask && conn->mtask); + + if (conn->ctask) { + if (iscsi_ctask_xmit(conn, conn->ctask)) + goto again; + /* done with this in-progress ctask */ + conn->ctask = NULL; + } + if (conn->mtask) { + if (iscsi_mtask_xmit(conn, conn->mtask)) + goto again; + /* done with this in-progress mtask */ + conn->mtask = NULL; + } + + /* process immediate first */ + if (unlikely(__kfifo_len(conn->immqueue))) { + struct iscsi_session *session = conn->session; + while (__kfifo_get(conn->immqueue, (void*)&conn->mtask, + sizeof(void*))) { + if (iscsi_mtask_xmit(conn, conn->mtask)) + goto again; + + if (conn->mtask->hdr.itt == + cpu_to_be32(ISCSI_RESERVED_TAG)) { + spin_lock_bh(&session->lock); + __kfifo_put(session->mgmtpool.queue, + (void*)&conn->mtask, sizeof(void*)); + spin_unlock_bh(&session->lock); + } + } + /* done with this mtask */ + conn->mtask = NULL; + } + + /* process write queue */ + while (__kfifo_get(conn->writequeue, (void*)&conn->ctask, + sizeof(void*))) { + if (iscsi_ctask_xmit(conn, conn->ctask)) + goto again; + } + + /* process command queue */ + while (__kfifo_get(conn->xmitqueue, (void*)&conn->ctask, + sizeof(void*))) { + if (iscsi_ctask_xmit(conn, conn->ctask)) + goto again; + } + /* done with this ctask */ + conn->ctask = NULL; + + /* process the rest control plane PDUs, if any */ + if (unlikely(__kfifo_len(conn->mgmtqueue))) { + struct iscsi_session *session = conn->session; + + while (__kfifo_get(conn->mgmtqueue, (void*)&conn->mtask, + sizeof(void*))) { + if (iscsi_mtask_xmit(conn, conn->mtask)) + goto again; + + if (conn->mtask->hdr.itt == + cpu_to_be32(ISCSI_RESERVED_TAG)) { + spin_lock_bh(&session->lock); + __kfifo_put(session->mgmtpool.queue, + (void*)&conn->mtask, + sizeof(void*)); + spin_unlock_bh(&session->lock); + } + } + /* done with this mtask */ + conn->mtask = NULL; + } + + return 0; + +again: + if (unlikely(conn->suspend_tx)) + return 0; + + return -EAGAIN; +} + +static void +iscsi_xmitworker(void *data) +{ + struct iscsi_conn *conn = data; + + /* + * serialize Xmit worker on a per-connection basis. + */ + mutex_lock(&conn->xmitmutex); + if (iscsi_data_xmit(conn)) + scsi_queue_work(conn->session->host, &conn->xmitwork); + mutex_unlock(&conn->xmitmutex); +} + +#define FAILURE_BAD_HOST 1 +#define FAILURE_SESSION_FAILED 2 +#define FAILURE_SESSION_FREED 3 +#define FAILURE_WINDOW_CLOSED 4 +#define FAILURE_SESSION_TERMINATE 5 + +static int +iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) +{ + struct Scsi_Host *host; + int reason = 0; + struct iscsi_session *session; + struct iscsi_conn *conn = NULL; + struct iscsi_cmd_task *ctask = NULL; + + sc->scsi_done = done; + sc->result = 0; + + host = sc->device->host; + session = iscsi_hostdata(host->hostdata); + BUG_ON(host != session->host); + + spin_lock(&session->lock); + + if (session->state != ISCSI_STATE_LOGGED_IN) { + if (session->state == ISCSI_STATE_FAILED) { + reason = FAILURE_SESSION_FAILED; + goto reject; + } else if (session->state == ISCSI_STATE_TERMINATE) { + reason = FAILURE_SESSION_TERMINATE; + goto fault; + } + reason = FAILURE_SESSION_FREED; + goto fault; + } + + /* + * Check for iSCSI window and take care of CmdSN wrap-around + */ + if ((int)(session->max_cmdsn - session->cmdsn) < 0) { + reason = FAILURE_WINDOW_CLOSED; + goto reject; + } + + conn = session->leadconn; + + __kfifo_get(session->cmdpool.queue, (void*)&ctask, sizeof(void*)); + BUG_ON(ctask->sc); + + sc->SCp.phase = session->age; + sc->SCp.ptr = (char*)ctask; + iscsi_cmd_init(conn, ctask, sc); + + __kfifo_put(conn->xmitqueue, (void*)&ctask, sizeof(void*)); + debug_scsi( + "ctask enq [%s cid %d sc %lx itt 0x%x len %d cmdsn %d win %d]\n", + sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read", + conn->id, (long)sc, ctask->itt, sc->request_bufflen, + session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1); + spin_unlock(&session->lock); + + scsi_queue_work(host, &conn->xmitwork); + return 0; + +reject: + spin_unlock(&session->lock); + debug_scsi("cmd 0x%x rejected (%d)\n", sc->cmnd[0], reason); + return SCSI_MLQUEUE_HOST_BUSY; + +fault: + spin_unlock(&session->lock); + printk(KERN_ERR "iscsi_tcp: cmd 0x%x is not queued (%d)\n", + sc->cmnd[0], reason); + sc->sense_buffer[0] = 0x70; + sc->sense_buffer[2] = NOT_READY; + sc->sense_buffer[7] = 0x6; + sc->sense_buffer[12] = 0x08; + sc->sense_buffer[13] = 0x00; + sc->result = (DID_NO_CONNECT << 16); + sc->resid = sc->request_bufflen; + sc->scsi_done(sc); + return 0; +} + +static int +iscsi_change_queue_depth(struct scsi_device *sdev, int depth) +{ + if (depth > ISCSI_MAX_CMD_PER_LUN) + depth = ISCSI_MAX_CMD_PER_LUN; + scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); + return sdev->queue_depth; +} + +static int +iscsi_pool_init(struct iscsi_queue *q, int max, void ***items, int item_size) +{ + int i; + + *items = kmalloc(max * sizeof(void*), GFP_KERNEL); + if (*items == NULL) + return -ENOMEM; + + q->max = max; + q->pool = kmalloc(max * sizeof(void*), GFP_KERNEL); + if (q->pool == NULL) { + kfree(*items); + return -ENOMEM; + } + + q->queue = kfifo_init((void*)q->pool, max * sizeof(void*), + GFP_KERNEL, NULL); + if (q->queue == ERR_PTR(-ENOMEM)) { + kfree(q->pool); + kfree(*items); + return -ENOMEM; + } + + for (i = 0; i < max; i++) { + q->pool[i] = kmalloc(item_size, GFP_KERNEL); + if (q->pool[i] == NULL) { + int j; + + for (j = 0; j < i; j++) + kfree(q->pool[j]); + + kfifo_free(q->queue); + kfree(q->pool); + kfree(*items); + return -ENOMEM; + } + memset(q->pool[i], 0, item_size); + (*items)[i] = q->pool[i]; + __kfifo_put(q->queue, (void*)&q->pool[i], sizeof(void*)); + } + return 0; +} + +static void +iscsi_pool_free(struct iscsi_queue *q, void **items) +{ + int i; + + for (i = 0; i < q->max; i++) + kfree(items[i]); + kfree(q->pool); + kfree(items); +} + static struct iscsi_cls_conn * -iscsi_tcp_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx) +iscsi_conn_create(struct iscsi_cls_session *cls_session, uint32_t conn_idx) { + struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); + struct iscsi_session *session = iscsi_hostdata(shost->hostdata); struct iscsi_conn *conn; struct iscsi_cls_conn *cls_conn; - struct iscsi_tcp_conn *tcp_conn; - cls_conn = iscsi_conn_setup(cls_session, conn_idx); + cls_conn = iscsi_create_conn(cls_session, conn_idx); if (!cls_conn) return NULL; conn = cls_conn->dd_data; - /* - * due to strange issues with iser these are not set - * in iscsi_conn_setup - */ - conn->max_recv_dlength = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH; + memset(conn, 0, sizeof(*conn)); - tcp_conn = kzalloc(sizeof(*tcp_conn), GFP_KERNEL); - if (!tcp_conn) - goto tcp_conn_alloc_fail; + conn->cls_conn = cls_conn; + conn->c_stage = ISCSI_CONN_INITIAL_STAGE; + conn->in_progress = IN_PROGRESS_WAIT_HEADER; + conn->id = conn_idx; + conn->exp_statsn = 0; + conn->tmabort_state = TMABORT_INITIAL; - conn->dd_data = tcp_conn; - tcp_conn->iscsi_conn = conn; - tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER; /* initial operational parameters */ - tcp_conn->hdr_size = sizeof(struct iscsi_hdr); - tcp_conn->data_size = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH; + conn->hdr_size = sizeof(struct iscsi_hdr); + conn->data_size = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH; + conn->max_recv_dlength = DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH; + + /* initialize general xmit PDU commands queue */ + conn->xmitqueue = kfifo_alloc(session->cmds_max * sizeof(void*), + GFP_KERNEL, NULL); + if (conn->xmitqueue == ERR_PTR(-ENOMEM)) + goto xmitqueue_alloc_fail; + + /* initialize write response PDU commands queue */ + conn->writequeue = kfifo_alloc(session->cmds_max * sizeof(void*), + GFP_KERNEL, NULL); + if (conn->writequeue == ERR_PTR(-ENOMEM)) + goto writequeue_alloc_fail; + + /* initialize general immediate & non-immediate PDU commands queue */ + conn->immqueue = kfifo_alloc(session->mgmtpool_max * sizeof(void*), + GFP_KERNEL, NULL); + if (conn->immqueue == ERR_PTR(-ENOMEM)) + goto immqueue_alloc_fail; + + conn->mgmtqueue = kfifo_alloc(session->mgmtpool_max * sizeof(void*), + GFP_KERNEL, NULL); + if (conn->mgmtqueue == ERR_PTR(-ENOMEM)) + goto mgmtqueue_alloc_fail; + + INIT_WORK(&conn->xmitwork, iscsi_xmitworker, conn); + + /* allocate login_mtask used for the login/text sequences */ + spin_lock_bh(&session->lock); + if (!__kfifo_get(session->mgmtpool.queue, + (void*)&conn->login_mtask, + sizeof(void*))) { + spin_unlock_bh(&session->lock); + goto login_mtask_alloc_fail; + } + spin_unlock_bh(&session->lock); /* allocate initial PDU receive place holder */ - if (tcp_conn->data_size <= PAGE_SIZE) - tcp_conn->data = kmalloc(tcp_conn->data_size, GFP_KERNEL); + if (conn->data_size <= PAGE_SIZE) + conn->data = kmalloc(conn->data_size, GFP_KERNEL); else - tcp_conn->data = (void*)__get_free_pages(GFP_KERNEL, - get_order(tcp_conn->data_size)); - if (!tcp_conn->data) + conn->data = (void*)__get_free_pages(GFP_KERNEL, + get_order(conn->data_size)); + if (!conn->data) goto max_recv_dlenght_alloc_fail; + init_timer(&conn->tmabort_timer); + mutex_init(&conn->xmitmutex); + init_waitqueue_head(&conn->ehwait); + return cls_conn; max_recv_dlenght_alloc_fail: - kfree(tcp_conn); -tcp_conn_alloc_fail: - iscsi_conn_teardown(cls_conn); + spin_lock_bh(&session->lock); + __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, + sizeof(void*)); + spin_unlock_bh(&session->lock); +login_mtask_alloc_fail: + kfifo_free(conn->mgmtqueue); +mgmtqueue_alloc_fail: + kfifo_free(conn->immqueue); +immqueue_alloc_fail: + kfifo_free(conn->writequeue); +writequeue_alloc_fail: + kfifo_free(conn->xmitqueue); +xmitqueue_alloc_fail: + iscsi_destroy_conn(cls_conn); return NULL; } static void -iscsi_tcp_conn_destroy(struct iscsi_cls_conn *cls_conn) +iscsi_conn_destroy(struct iscsi_cls_conn *cls_conn) { struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - int digest = 0; + struct iscsi_session *session = conn->session; + unsigned long flags; + + mutex_lock(&conn->xmitmutex); + set_bit(SUSPEND_BIT, &conn->suspend_tx); + if (conn->c_stage == ISCSI_CONN_INITIAL_STAGE && conn->sock) { + struct sock *sk = conn->sock->sk; + + /* + * conn_start() has never been called! + * need to cleanup the socket. + */ + write_lock_bh(&sk->sk_callback_lock); + set_bit(SUSPEND_BIT, &conn->suspend_rx); + write_unlock_bh(&sk->sk_callback_lock); + + sock_hold(conn->sock->sk); + iscsi_conn_restore_callbacks(conn); + sock_put(conn->sock->sk); + sock_release(conn->sock); + conn->sock = NULL; + } - if (conn->hdrdgst_en || conn->datadgst_en) - digest = 1; + spin_lock_bh(&session->lock); + conn->c_stage = ISCSI_CONN_CLEANUP_WAIT; + if (session->leadconn == conn) { + /* + * leading connection? then give up on recovery. + */ + session->state = ISCSI_STATE_TERMINATE; + wake_up(&conn->ehwait); + } + spin_unlock_bh(&session->lock); + + mutex_unlock(&conn->xmitmutex); - iscsi_conn_teardown(cls_conn); + /* + * Block until all in-progress commands for this connection + * time out or fail. + */ + for (;;) { + spin_lock_irqsave(session->host->host_lock, flags); + if (!session->host->host_busy) { /* OK for ERL == 0 */ + spin_unlock_irqrestore(session->host->host_lock, flags); + break; + } + spin_unlock_irqrestore(session->host->host_lock, flags); + msleep_interruptible(500); + printk("conn_destroy(): host_busy %d host_failed %d\n", + session->host->host_busy, session->host->host_failed); + /* + * force eh_abort() to unblock + */ + wake_up(&conn->ehwait); + } - /* now free tcp_conn */ - if (digest) { - if (tcp_conn->tx_tfm) - crypto_free_tfm(tcp_conn->tx_tfm); - if (tcp_conn->rx_tfm) - crypto_free_tfm(tcp_conn->rx_tfm); - if (tcp_conn->data_tx_tfm) - crypto_free_tfm(tcp_conn->data_tx_tfm); - if (tcp_conn->data_rx_tfm) - crypto_free_tfm(tcp_conn->data_rx_tfm); + /* now free crypto */ + if (conn->hdrdgst_en || conn->datadgst_en) { + if (conn->tx_tfm) + crypto_free_tfm(conn->tx_tfm); + if (conn->rx_tfm) + crypto_free_tfm(conn->rx_tfm); + if (conn->data_tx_tfm) + crypto_free_tfm(conn->data_tx_tfm); + if (conn->data_rx_tfm) + crypto_free_tfm(conn->data_rx_tfm); } /* free conn->data, size = MaxRecvDataSegmentLength */ - if (tcp_conn->data_size <= PAGE_SIZE) - kfree(tcp_conn->data); + if (conn->data_size <= PAGE_SIZE) + kfree(conn->data); else - free_pages((unsigned long)tcp_conn->data, - get_order(tcp_conn->data_size)); - kfree(tcp_conn); + free_pages((unsigned long)conn->data, + get_order(conn->data_size)); + + spin_lock_bh(&session->lock); + __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, + sizeof(void*)); + list_del(&conn->item); + if (list_empty(&session->connections)) + session->leadconn = NULL; + if (session->leadconn && session->leadconn == conn) + session->leadconn = container_of(session->connections.next, + struct iscsi_conn, item); + + if (session->leadconn == NULL) + /* none connections exits.. reset sequencing */ + session->cmdsn = session->max_cmdsn = session->exp_cmdsn = 1; + spin_unlock_bh(&session->lock); + + kfifo_free(conn->xmitqueue); + kfifo_free(conn->writequeue); + kfifo_free(conn->immqueue); + kfifo_free(conn->mgmtqueue); + + iscsi_destroy_conn(cls_conn); } static int -iscsi_tcp_conn_bind(struct iscsi_cls_session *cls_session, - struct iscsi_cls_conn *cls_conn, uint64_t transport_eph, - int is_leading) +iscsi_conn_bind(struct iscsi_cls_session *cls_session, + struct iscsi_cls_conn *cls_conn, uint32_t transport_fd, + int is_leading) { - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); + struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + struct iscsi_conn *tmp = ERR_PTR(-EEXIST), *conn = cls_conn->dd_data; struct sock *sk; struct socket *sock; int err; /* lookup for existing socket */ - sock = sockfd_lookup((int)transport_eph, &err); + sock = sockfd_lookup(transport_fd, &err); if (!sock) { printk(KERN_ERR "iscsi_tcp: sockfd_lookup failed %d\n", err); return -EEXIST; } - err = iscsi_conn_bind(cls_session, cls_conn, is_leading); - if (err) - return err; + /* lookup for existing connection */ + spin_lock_bh(&session->lock); + list_for_each_entry(tmp, &session->connections, item) { + if (tmp == conn) { + if (conn->c_stage != ISCSI_CONN_STOPPED || + conn->stop_stage == STOP_CONN_TERM) { + printk(KERN_ERR "iscsi_tcp: can't bind " + "non-stopped connection (%d:%d)\n", + conn->c_stage, conn->stop_stage); + spin_unlock_bh(&session->lock); + return -EIO; + } + break; + } + } + if (tmp != conn) { + /* bind new iSCSI connection to session */ + conn->session = session; + + list_add(&conn->item, &session->connections); + } + spin_unlock_bh(&session->lock); - /* bind iSCSI connection and socket */ - tcp_conn->sock = sock; + if (conn->stop_stage != STOP_CONN_SUSPEND) { + /* bind iSCSI connection and socket */ + conn->sock = sock; - /* setup Socket parameters */ - sk = sock->sk; - sk->sk_reuse = 1; - sk->sk_sndtimeo = 15 * HZ; /* FIXME: make it configurable */ - sk->sk_allocation = GFP_ATOMIC; + /* setup Socket parameters */ + sk = sock->sk; + sk->sk_reuse = 1; + sk->sk_sndtimeo = 15 * HZ; /* FIXME: make it configurable */ + sk->sk_allocation = GFP_ATOMIC; - /* FIXME: disable Nagle's algorithm */ + /* FIXME: disable Nagle's algorithm */ + + /* + * Intercept TCP callbacks for sendfile like receive + * processing. + */ + iscsi_conn_set_callbacks(conn); + + conn->sendpage = conn->sock->ops->sendpage; + + /* + * set receive state machine into initial state + */ + conn->in_progress = IN_PROGRESS_WAIT_HEADER; + } + + if (is_leading) + session->leadconn = conn; /* - * Intercept TCP callbacks for sendfile like receive - * processing. - */ - conn->recv_lock = &sk->sk_callback_lock; - iscsi_conn_set_callbacks(conn); - tcp_conn->sendpage = tcp_conn->sock->ops->sendpage; - /* - * set receive state machine into initial state + * Unblock xmitworker(), Login Phase will pass through. */ - tcp_conn->in_progress = IN_PROGRESS_WAIT_HEADER; + clear_bit(SUSPEND_BIT, &conn->suspend_rx); + clear_bit(SUSPEND_BIT, &conn->suspend_tx); return 0; } -static void -iscsi_tcp_cleanup_ctask(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask) +static int +iscsi_conn_start(struct iscsi_cls_conn *cls_conn) { - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - struct iscsi_r2t_info *r2t; + struct iscsi_conn *conn = cls_conn->dd_data; + struct iscsi_session *session = conn->session; + struct sock *sk; - /* flush ctask's r2t queues */ - while (__kfifo_get(tcp_ctask->r2tqueue, (void*)&r2t, sizeof(void*))) - __kfifo_put(tcp_ctask->r2tpool.queue, (void*)&r2t, - sizeof(void*)); + /* FF phase warming up... */ - __iscsi_ctask_cleanup(conn, ctask); + if (session == NULL) { + printk(KERN_ERR "iscsi_tcp: can't start unbound connection\n"); + return -EPERM; + } + + sk = conn->sock->sk; + + write_lock_bh(&sk->sk_callback_lock); + spin_lock_bh(&session->lock); + conn->c_stage = ISCSI_CONN_STARTED; + session->state = ISCSI_STATE_LOGGED_IN; + + switch(conn->stop_stage) { + case STOP_CONN_RECOVER: + /* + * unblock eh_abort() if it is blocked. re-try all + * commands after successful recovery + */ + session->conn_cnt++; + conn->stop_stage = 0; + conn->tmabort_state = TMABORT_INITIAL; + session->age++; + wake_up(&conn->ehwait); + break; + case STOP_CONN_TERM: + session->conn_cnt++; + conn->stop_stage = 0; + break; + case STOP_CONN_SUSPEND: + conn->stop_stage = 0; + clear_bit(SUSPEND_BIT, &conn->suspend_rx); + clear_bit(SUSPEND_BIT, &conn->suspend_tx); + break; + default: + break; + } + spin_unlock_bh(&session->lock); + write_unlock_bh(&sk->sk_callback_lock); + + return 0; } static void -iscsi_tcp_suspend_conn_rx(struct iscsi_conn *conn) +iscsi_conn_stop(struct iscsi_cls_conn *cls_conn, int flag) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + struct iscsi_conn *conn = cls_conn->dd_data; + struct iscsi_session *session = conn->session; struct sock *sk; + unsigned long flags; - if (!tcp_conn->sock) - return; - - sk = tcp_conn->sock->sk; + BUG_ON(!conn->sock); + sk = conn->sock->sk; write_lock_bh(&sk->sk_callback_lock); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); + set_bit(SUSPEND_BIT, &conn->suspend_rx); write_unlock_bh(&sk->sk_callback_lock); + + mutex_lock(&conn->xmitmutex); + + spin_lock_irqsave(session->host->host_lock, flags); + spin_lock(&session->lock); + conn->stop_stage = flag; + conn->c_stage = ISCSI_CONN_STOPPED; + set_bit(SUSPEND_BIT, &conn->suspend_tx); + + if (flag != STOP_CONN_SUSPEND) + session->conn_cnt--; + + if (session->conn_cnt == 0 || session->leadconn == conn) + session->state = ISCSI_STATE_FAILED; + + spin_unlock(&session->lock); + spin_unlock_irqrestore(session->host->host_lock, flags); + + if (flag == STOP_CONN_TERM || flag == STOP_CONN_RECOVER) { + struct iscsi_cmd_task *ctask; + struct iscsi_mgmt_task *mtask; + + /* + * Socket must go now. + */ + sock_hold(conn->sock->sk); + iscsi_conn_restore_callbacks(conn); + sock_put(conn->sock->sk); + + /* + * flush xmit queues. + */ + spin_lock_bh(&session->lock); + while (__kfifo_get(conn->writequeue, (void*)&ctask, + sizeof(void*)) || + __kfifo_get(conn->xmitqueue, (void*)&ctask, + sizeof(void*))) { + struct iscsi_r2t_info *r2t; + + /* + * flush ctask's r2t queues + */ + while (__kfifo_get(ctask->r2tqueue, (void*)&r2t, + sizeof(void*))) + __kfifo_put(ctask->r2tpool.queue, (void*)&r2t, + sizeof(void*)); + + spin_unlock_bh(&session->lock); + local_bh_disable(); + iscsi_ctask_cleanup(conn, ctask); + local_bh_enable(); + spin_lock_bh(&session->lock); + } + conn->ctask = NULL; + while (__kfifo_get(conn->immqueue, (void*)&mtask, + sizeof(void*)) || + __kfifo_get(conn->mgmtqueue, (void*)&mtask, + sizeof(void*))) { + __kfifo_put(session->mgmtpool.queue, + (void*)&mtask, sizeof(void*)); + } + conn->mtask = NULL; + spin_unlock_bh(&session->lock); + + /* + * release socket only after we stopped data_xmit() + * activity and flushed all outstandings + */ + sock_release(conn->sock); + conn->sock = NULL; + + /* + * for connection level recovery we should not calculate + * header digest. conn->hdr_size used for optimization + * in hdr_extract() and will be re-negotiated at + * set_param() time. + */ + if (flag == STOP_CONN_RECOVER) { + conn->hdr_size = sizeof(struct iscsi_hdr); + conn->hdrdgst_en = 0; + conn->datadgst_en = 0; + } + } + mutex_unlock(&conn->xmitmutex); } -static void -iscsi_tcp_terminate_conn(struct iscsi_conn *conn) +static int +iscsi_conn_send_generic(struct iscsi_conn *conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size) { - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + struct iscsi_session *session = conn->session; + struct iscsi_nopout *nop = (struct iscsi_nopout *)hdr; + struct iscsi_mgmt_task *mtask; - if (!tcp_conn->sock) - return; + spin_lock_bh(&session->lock); + if (session->state == ISCSI_STATE_TERMINATE) { + spin_unlock_bh(&session->lock); + return -EPERM; + } + if (hdr->opcode == (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) || + hdr->opcode == (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE)) + /* + * Login and Text are sent serially, in + * request-followed-by-response sequence. + * Same mtask can be used. Same ITT must be used. + * Note that login_mtask is preallocated at conn_create(). + */ + mtask = conn->login_mtask; + else { + BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE); + BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED); + + if (!__kfifo_get(session->mgmtpool.queue, + (void*)&mtask, sizeof(void*))) { + spin_unlock_bh(&session->lock); + return -ENOSPC; + } + } + + /* + * pre-format CmdSN and ExpStatSN for outgoing PDU. + */ + if (hdr->itt != cpu_to_be32(ISCSI_RESERVED_TAG)) { + hdr->itt = mtask->itt | (conn->id << CID_SHIFT) | + (session->age << AGE_SHIFT); + nop->cmdsn = cpu_to_be32(session->cmdsn); + if (conn->c_stage == ISCSI_CONN_STARTED && + !(hdr->opcode & ISCSI_OP_IMMEDIATE)) + session->cmdsn++; + } else + /* do not advance CmdSN */ + nop->cmdsn = cpu_to_be32(session->cmdsn); + + nop->exp_statsn = cpu_to_be32(conn->exp_statsn); + + memcpy(&mtask->hdr, hdr, sizeof(struct iscsi_hdr)); + + iscsi_buf_init_virt(&mtask->headbuf, (char*)&mtask->hdr, + sizeof(struct iscsi_hdr)); + + spin_unlock_bh(&session->lock); + + if (data_size) { + memcpy(mtask->data, data, data_size); + mtask->data_count = data_size; + } else + mtask->data_count = 0; + + mtask->xmstate = XMSTATE_IMM_HDR; + + if (mtask->data_count) { + iscsi_buf_init_iov(&mtask->sendbuf, (char*)mtask->data, + mtask->data_count); + } + + debug_scsi("mgmtpdu [op 0x%x hdr->itt 0x%x datalen %d]\n", + hdr->opcode, hdr->itt, data_size); - sock_hold(tcp_conn->sock->sk); - iscsi_conn_restore_callbacks(conn); - sock_put(tcp_conn->sock->sk); + /* + * since send_pdu() could be called at least from two contexts, + * we need to serialize __kfifo_put, so we don't have to take + * additional lock on fast data-path + */ + if (hdr->opcode & ISCSI_OP_IMMEDIATE) + __kfifo_put(conn->immqueue, (void*)&mtask, sizeof(void*)); + else + __kfifo_put(conn->mgmtqueue, (void*)&mtask, sizeof(void*)); - sock_release(tcp_conn->sock); - tcp_conn->sock = NULL; - conn->recv_lock = NULL; + scsi_queue_work(session->host, &conn->xmitwork); + return 0; +} + +static int +iscsi_eh_host_reset(struct scsi_cmnd *sc) +{ + struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)sc->SCp.ptr; + struct iscsi_conn *conn = ctask->conn; + struct iscsi_session *session = conn->session; + + spin_lock_bh(&session->lock); + if (session->state == ISCSI_STATE_TERMINATE) { + debug_scsi("failing host reset: session terminated " + "[CID %d age %d]", conn->id, session->age); + spin_unlock_bh(&session->lock); + return FAILED; + } + spin_unlock_bh(&session->lock); + + debug_scsi("failing connection CID %d due to SCSI host reset " + "[itt 0x%x age %d]", conn->id, ctask->itt, + session->age); + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + + return SUCCESS; } -/* called with host lock */ static void -iscsi_tcp_mgmt_init(struct iscsi_conn *conn, struct iscsi_mgmt_task *mtask, - char *data, uint32_t data_size) +iscsi_tmabort_timedout(unsigned long data) { - struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data; + struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)data; + struct iscsi_conn *conn = ctask->conn; + struct iscsi_session *session = conn->session; - iscsi_buf_init_iov(&tcp_mtask->headbuf, (char*)mtask->hdr, - sizeof(struct iscsi_hdr)); - tcp_mtask->xmstate = XMSTATE_IMM_HDR; + spin_lock(&session->lock); + if (conn->tmabort_state == TMABORT_INITIAL) { + __kfifo_put(session->mgmtpool.queue, + (void*)&ctask->mtask, sizeof(void*)); + conn->tmabort_state = TMABORT_TIMEDOUT; + debug_scsi("tmabort timedout [sc %lx itt 0x%x]\n", + (long)ctask->sc, ctask->itt); + /* unblock eh_abort() */ + wake_up(&conn->ehwait); + } + spin_unlock(&session->lock); +} - if (mtask->data_count) - iscsi_buf_init_iov(&tcp_mtask->sendbuf, (char*)mtask->data, - mtask->data_count); +static int +iscsi_eh_abort(struct scsi_cmnd *sc) +{ + int rc; + struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)sc->SCp.ptr; + struct iscsi_conn *conn = ctask->conn; + struct iscsi_session *session = conn->session; + + conn->eh_abort_cnt++; + debug_scsi("aborting [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); + + /* + * two cases for ERL=0 here: + * + * 1) connection-level failure; + * 2) recovery due protocol error; + */ + mutex_lock(&conn->xmitmutex); + spin_lock_bh(&session->lock); + if (session->state != ISCSI_STATE_LOGGED_IN) { + if (session->state == ISCSI_STATE_TERMINATE) { + spin_unlock_bh(&session->lock); + mutex_unlock(&conn->xmitmutex); + goto failed; + } + spin_unlock_bh(&session->lock); + } else { + struct iscsi_tm *hdr = &conn->tmhdr; + + /* + * Still LOGGED_IN... + */ + + if (!ctask->sc || sc->SCp.phase != session->age) { + /* + * 1) ctask completed before time out. But session + * is still ok => Happy Retry. + * 2) session was re-open during time out of ctask. + */ + spin_unlock_bh(&session->lock); + mutex_unlock(&conn->xmitmutex); + goto success; + } + conn->tmabort_state = TMABORT_INITIAL; + spin_unlock_bh(&session->lock); + + /* + * ctask timed out but session is OK + * ERL=0 requires task mgmt abort to be issued on each + * failed command. requests must be serialized. + */ + memset(hdr, 0, sizeof(struct iscsi_tm)); + hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE; + hdr->flags = ISCSI_TM_FUNC_ABORT_TASK; + hdr->flags |= ISCSI_FLAG_CMD_FINAL; + memcpy(hdr->lun, ctask->hdr.lun, sizeof(hdr->lun)); + hdr->rtt = ctask->hdr.itt; + hdr->refcmdsn = ctask->hdr.cmdsn; + + rc = iscsi_conn_send_generic(conn, (struct iscsi_hdr *)hdr, + NULL, 0); + if (rc) { + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + debug_scsi("abort sent failure [itt 0x%x]", ctask->itt); + } else { + struct iscsi_r2t_info *r2t; + + /* + * TMF abort vs. TMF response race logic + */ + spin_lock_bh(&session->lock); + ctask->mtask = (struct iscsi_mgmt_task *) + session->mgmt_cmds[(hdr->itt & ITT_MASK) - + ISCSI_MGMT_ITT_OFFSET]; + /* + * have to flush r2tqueue to avoid r2t leaks + */ + while (__kfifo_get(ctask->r2tqueue, (void*)&r2t, + sizeof(void*))) { + __kfifo_put(ctask->r2tpool.queue, (void*)&r2t, + sizeof(void*)); + } + if (conn->tmabort_state == TMABORT_INITIAL) { + conn->tmfcmd_pdus_cnt++; + conn->tmabort_timer.expires = 3*HZ + jiffies; + conn->tmabort_timer.function = + iscsi_tmabort_timedout; + conn->tmabort_timer.data = (unsigned long)ctask; + add_timer(&conn->tmabort_timer); + debug_scsi("abort sent [itt 0x%x]", ctask->itt); + } else { + if (!ctask->sc || + conn->tmabort_state == TMABORT_SUCCESS) { + conn->tmabort_state = TMABORT_INITIAL; + spin_unlock_bh(&session->lock); + mutex_unlock(&conn->xmitmutex); + goto success; + } + conn->tmabort_state = TMABORT_INITIAL; + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + } + spin_unlock_bh(&session->lock); + } + } + mutex_unlock(&conn->xmitmutex); + + + /* + * block eh thread until: + * + * 1) abort response; + * 2) abort timeout; + * 3) session re-opened; + * 4) session terminated; + */ + for (;;) { + int p_state = session->state; + + rc = wait_event_interruptible(conn->ehwait, + (p_state == ISCSI_STATE_LOGGED_IN ? + (session->state == ISCSI_STATE_TERMINATE || + conn->tmabort_state != TMABORT_INITIAL) : + (session->state == ISCSI_STATE_TERMINATE || + session->state == ISCSI_STATE_LOGGED_IN))); + if (rc) { + /* shutdown.. */ + session->state = ISCSI_STATE_TERMINATE; + goto failed; + } + + if (signal_pending(current)) + flush_signals(current); + + if (session->state == ISCSI_STATE_TERMINATE) + goto failed; + + spin_lock_bh(&session->lock); + if (sc->SCp.phase == session->age && + (conn->tmabort_state == TMABORT_TIMEDOUT || + conn->tmabort_state == TMABORT_FAILED)) { + conn->tmabort_state = TMABORT_INITIAL; + if (!ctask->sc) { + /* + * ctask completed before tmf abort response or + * time out. + * But session is still ok => Happy Retry. + */ + spin_unlock_bh(&session->lock); + break; + } + spin_unlock_bh(&session->lock); + iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); + continue; + } + spin_unlock_bh(&session->lock); + break; + } + +success: + debug_scsi("abort success [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); + rc = SUCCESS; + goto exit; + +failed: + debug_scsi("abort failed [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); + rc = FAILED; + +exit: + del_timer_sync(&conn->tmabort_timer); + + mutex_lock(&conn->xmitmutex); + if (conn->sock) { + struct sock *sk = conn->sock->sk; + + write_lock_bh(&sk->sk_callback_lock); + iscsi_ctask_cleanup(conn, ctask); + write_unlock_bh(&sk->sk_callback_lock); + } + mutex_unlock(&conn->xmitmutex); + return rc; } static int @@ -2074,7 +3174,6 @@ iscsi_r2tpool_alloc(struct iscsi_session *session) */ for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) { struct iscsi_cmd_task *ctask = session->cmds[cmd_i]; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; /* * pre-allocated x4 as much r2ts to handle race when @@ -2083,32 +3182,42 @@ iscsi_r2tpool_alloc(struct iscsi_session *session) */ /* R2T pool */ - if (iscsi_pool_init(&tcp_ctask->r2tpool, session->max_r2t * 4, - (void***)&tcp_ctask->r2ts, - sizeof(struct iscsi_r2t_info))) { + if (iscsi_pool_init(&ctask->r2tpool, session->max_r2t * 4, + (void***)&ctask->r2ts, sizeof(struct iscsi_r2t_info))) { goto r2t_alloc_fail; } /* R2T xmit queue */ - tcp_ctask->r2tqueue = kfifo_alloc( + ctask->r2tqueue = kfifo_alloc( session->max_r2t * 4 * sizeof(void*), GFP_KERNEL, NULL); - if (tcp_ctask->r2tqueue == ERR_PTR(-ENOMEM)) { - iscsi_pool_free(&tcp_ctask->r2tpool, - (void**)tcp_ctask->r2ts); + if (ctask->r2tqueue == ERR_PTR(-ENOMEM)) { + iscsi_pool_free(&ctask->r2tpool, (void**)ctask->r2ts); goto r2t_alloc_fail; } + + /* + * number of + * Data-Out PDU's within R2T-sequence can be quite big; + * using mempool + */ + ctask->datapool = mempool_create_slab_pool(ISCSI_DTASK_DEFAULT_MAX, + taskcache); + if (ctask->datapool == NULL) { + kfifo_free(ctask->r2tqueue); + iscsi_pool_free(&ctask->r2tpool, (void**)ctask->r2ts); + goto r2t_alloc_fail; + } + INIT_LIST_HEAD(&ctask->dataqueue); } return 0; r2t_alloc_fail: for (i = 0; i < cmd_i; i++) { - struct iscsi_cmd_task *ctask = session->cmds[i]; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - - kfifo_free(tcp_ctask->r2tqueue); - iscsi_pool_free(&tcp_ctask->r2tpool, - (void**)tcp_ctask->r2ts); + mempool_destroy(session->cmds[i]->datapool); + kfifo_free(session->cmds[i]->r2tqueue); + iscsi_pool_free(&session->cmds[i]->r2tpool, + (void**)session->cmds[i]->r2ts); } return -ENOMEM; } @@ -2119,13 +3228,127 @@ iscsi_r2tpool_free(struct iscsi_session *session) int i; for (i = 0; i < session->cmds_max; i++) { - struct iscsi_cmd_task *ctask = session->cmds[i]; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; + mempool_destroy(session->cmds[i]->datapool); + kfifo_free(session->cmds[i]->r2tqueue); + iscsi_pool_free(&session->cmds[i]->r2tpool, + (void**)session->cmds[i]->r2ts); + } +} + +static struct scsi_host_template iscsi_sht = { + .name = "iSCSI Initiator over TCP/IP, v." + ISCSI_VERSION_STR, + .queuecommand = iscsi_queuecommand, + .change_queue_depth = iscsi_change_queue_depth, + .can_queue = ISCSI_XMIT_CMDS_MAX - 1, + .sg_tablesize = ISCSI_SG_TABLESIZE, + .cmd_per_lun = ISCSI_DEF_CMD_PER_LUN, + .eh_abort_handler = iscsi_eh_abort, + .eh_host_reset_handler = iscsi_eh_host_reset, + .use_clustering = DISABLE_CLUSTERING, + .proc_name = "iscsi_tcp", + .this_id = -1, +}; + +static struct iscsi_transport iscsi_tcp_transport; + +static struct iscsi_cls_session * +iscsi_session_create(struct scsi_transport_template *scsit, + uint32_t initial_cmdsn, uint32_t *sid) +{ + struct Scsi_Host *shost; + struct iscsi_session *session; + int cmd_i; + + shost = iscsi_transport_create_session(scsit, &iscsi_tcp_transport); + if (!shost) + return NULL; + + session = iscsi_hostdata(shost->hostdata); + memset(session, 0, sizeof(struct iscsi_session)); + session->host = shost; + session->state = ISCSI_STATE_FREE; + session->mgmtpool_max = ISCSI_MGMT_CMDS_MAX; + session->cmds_max = ISCSI_XMIT_CMDS_MAX; + session->cmdsn = initial_cmdsn; + session->exp_cmdsn = initial_cmdsn + 1; + session->max_cmdsn = initial_cmdsn + 1; + session->max_r2t = 1; + *sid = shost->host_no; + + /* initialize SCSI PDU commands pool */ + if (iscsi_pool_init(&session->cmdpool, session->cmds_max, + (void***)&session->cmds, sizeof(struct iscsi_cmd_task))) + goto cmdpool_alloc_fail; + + /* pre-format cmds pool with ITT */ + for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) + session->cmds[cmd_i]->itt = cmd_i; + + spin_lock_init(&session->lock); + INIT_LIST_HEAD(&session->connections); + + /* initialize immediate command pool */ + if (iscsi_pool_init(&session->mgmtpool, session->mgmtpool_max, + (void***)&session->mgmt_cmds, sizeof(struct iscsi_mgmt_task))) + goto mgmtpool_alloc_fail; + + + /* pre-format immediate cmds pool with ITT */ + for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) { + session->mgmt_cmds[cmd_i]->itt = ISCSI_MGMT_ITT_OFFSET + cmd_i; + session->mgmt_cmds[cmd_i]->data = kmalloc( + DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH, GFP_KERNEL); + if (!session->mgmt_cmds[cmd_i]->data) { + int j; + + for (j = 0; j < cmd_i; j++) + kfree(session->mgmt_cmds[j]->data); + goto immdata_alloc_fail; + } + } + + if (iscsi_r2tpool_alloc(session)) + goto r2tpool_alloc_fail; - kfifo_free(tcp_ctask->r2tqueue); - iscsi_pool_free(&tcp_ctask->r2tpool, - (void**)tcp_ctask->r2ts); + return hostdata_session(shost->hostdata); + +r2tpool_alloc_fail: + for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) + kfree(session->mgmt_cmds[cmd_i]->data); +immdata_alloc_fail: + iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds); +mgmtpool_alloc_fail: + iscsi_pool_free(&session->cmdpool, (void**)session->cmds); +cmdpool_alloc_fail: + iscsi_transport_destroy_session(shost); + return NULL; +} + +static void +iscsi_session_destroy(struct iscsi_cls_session *cls_session) +{ + struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); + struct iscsi_session *session = iscsi_hostdata(shost->hostdata); + int cmd_i; + struct iscsi_data_task *dtask, *n; + + for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) { + struct iscsi_cmd_task *ctask = session->cmds[cmd_i]; + list_for_each_entry_safe(dtask, n, &ctask->dataqueue, item) { + list_del(&dtask->item); + mempool_free(dtask, ctask->datapool); + } } + + for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) + kfree(session->mgmt_cmds[cmd_i]->data); + + iscsi_r2tpool_free(session); + iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds); + iscsi_pool_free(&session->cmdpool, (void**)session->cmds); + + iscsi_transport_destroy_session(shost); } static int @@ -2134,14 +3357,23 @@ iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, { struct iscsi_conn *conn = cls_conn->dd_data; struct iscsi_session *session = conn->session; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; + + spin_lock_bh(&session->lock); + if (conn->c_stage != ISCSI_CONN_INITIAL_STAGE && + conn->stop_stage != STOP_CONN_RECOVER) { + printk(KERN_ERR "iscsi_tcp: can not change parameter [%d]\n", + param); + spin_unlock_bh(&session->lock); + return 0; + } + spin_unlock_bh(&session->lock); switch(param) { case ISCSI_PARAM_MAX_RECV_DLENGTH: { - char *saveptr = tcp_conn->data; + char *saveptr = conn->data; gfp_t flags = GFP_KERNEL; - if (tcp_conn->data_size >= value) { + if (conn->data_size >= value) { conn->max_recv_dlength = value; break; } @@ -2152,21 +3384,21 @@ iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, spin_unlock_bh(&session->lock); if (value <= PAGE_SIZE) - tcp_conn->data = kmalloc(value, flags); + conn->data = kmalloc(value, flags); else - tcp_conn->data = (void*)__get_free_pages(flags, + conn->data = (void*)__get_free_pages(flags, get_order(value)); - if (tcp_conn->data == NULL) { - tcp_conn->data = saveptr; + if (conn->data == NULL) { + conn->data = saveptr; return -ENOMEM; } - if (tcp_conn->data_size <= PAGE_SIZE) + if (conn->data_size <= PAGE_SIZE) kfree(saveptr); else free_pages((unsigned long)saveptr, - get_order(tcp_conn->data_size)); + get_order(conn->data_size)); conn->max_recv_dlength = value; - tcp_conn->data_size = value; + conn->data_size = value; } break; case ISCSI_PARAM_MAX_XMIT_DLENGTH: @@ -2174,51 +3406,49 @@ iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, break; case ISCSI_PARAM_HDRDGST_EN: conn->hdrdgst_en = value; - tcp_conn->hdr_size = sizeof(struct iscsi_hdr); + conn->hdr_size = sizeof(struct iscsi_hdr); if (conn->hdrdgst_en) { - tcp_conn->hdr_size += sizeof(__u32); - if (!tcp_conn->tx_tfm) - tcp_conn->tx_tfm = crypto_alloc_tfm("crc32c", - 0); - if (!tcp_conn->tx_tfm) + conn->hdr_size += sizeof(__u32); + if (!conn->tx_tfm) + conn->tx_tfm = crypto_alloc_tfm("crc32c", 0); + if (!conn->tx_tfm) return -ENOMEM; - if (!tcp_conn->rx_tfm) - tcp_conn->rx_tfm = crypto_alloc_tfm("crc32c", - 0); - if (!tcp_conn->rx_tfm) { - crypto_free_tfm(tcp_conn->tx_tfm); + if (!conn->rx_tfm) + conn->rx_tfm = crypto_alloc_tfm("crc32c", 0); + if (!conn->rx_tfm) { + crypto_free_tfm(conn->tx_tfm); return -ENOMEM; } } else { - if (tcp_conn->tx_tfm) - crypto_free_tfm(tcp_conn->tx_tfm); - if (tcp_conn->rx_tfm) - crypto_free_tfm(tcp_conn->rx_tfm); + if (conn->tx_tfm) + crypto_free_tfm(conn->tx_tfm); + if (conn->rx_tfm) + crypto_free_tfm(conn->rx_tfm); } break; case ISCSI_PARAM_DATADGST_EN: conn->datadgst_en = value; if (conn->datadgst_en) { - if (!tcp_conn->data_tx_tfm) - tcp_conn->data_tx_tfm = + if (!conn->data_tx_tfm) + conn->data_tx_tfm = crypto_alloc_tfm("crc32c", 0); - if (!tcp_conn->data_tx_tfm) + if (!conn->data_tx_tfm) return -ENOMEM; - if (!tcp_conn->data_rx_tfm) - tcp_conn->data_rx_tfm = + if (!conn->data_rx_tfm) + conn->data_rx_tfm = crypto_alloc_tfm("crc32c", 0); - if (!tcp_conn->data_rx_tfm) { - crypto_free_tfm(tcp_conn->data_tx_tfm); + if (!conn->data_rx_tfm) { + crypto_free_tfm(conn->data_tx_tfm); return -ENOMEM; } } else { - if (tcp_conn->data_tx_tfm) - crypto_free_tfm(tcp_conn->data_tx_tfm); - if (tcp_conn->data_rx_tfm) - crypto_free_tfm(tcp_conn->data_rx_tfm); + if (conn->data_tx_tfm) + crypto_free_tfm(conn->data_tx_tfm); + if (conn->data_rx_tfm) + crypto_free_tfm(conn->data_rx_tfm); } - tcp_conn->sendpage = conn->datadgst_en ? - sock_no_sendpage : tcp_conn->sock->ops->sendpage; + conn->sendpage = conn->datadgst_en ? + sock_no_sendpage : conn->sock->ops->sendpage; break; case ISCSI_PARAM_INITIAL_R2T_EN: session->initial_r2t_en = value; @@ -2259,9 +3489,6 @@ iscsi_conn_set_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, BUG_ON(value); session->ofmarker_en = value; break; - case ISCSI_PARAM_EXP_STATSN: - conn->exp_statsn = value; - break; default: break; } @@ -2308,7 +3535,7 @@ iscsi_session_get_param(struct iscsi_cls_session *cls_session, *value = session->ofmarker_en; break; default: - return -EINVAL; + return ISCSI_ERR_PARAM_NOT_FOUND; } return 0; @@ -2319,8 +3546,6 @@ iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn, enum iscsi_param param, uint32_t *value) { struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct inet_sock *inet; switch(param) { case ISCSI_PARAM_MAX_RECV_DLENGTH: @@ -2335,70 +3560,17 @@ iscsi_conn_get_param(struct iscsi_cls_conn *cls_conn, case ISCSI_PARAM_DATADGST_EN: *value = conn->datadgst_en; break; - case ISCSI_PARAM_CONN_PORT: - mutex_lock(&conn->xmitmutex); - if (!tcp_conn->sock) { - mutex_unlock(&conn->xmitmutex); - return -EINVAL; - } - - inet = inet_sk(tcp_conn->sock->sk); - *value = be16_to_cpu(inet->dport); - mutex_unlock(&conn->xmitmutex); - case ISCSI_PARAM_EXP_STATSN: - *value = conn->exp_statsn; - break; default: - return -EINVAL; + return ISCSI_ERR_PARAM_NOT_FOUND; } return 0; } -static int -iscsi_conn_get_str_param(struct iscsi_cls_conn *cls_conn, - enum iscsi_param param, char *buf) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; - struct sock *sk; - struct inet_sock *inet; - struct ipv6_pinfo *np; - int len = 0; - - switch (param) { - case ISCSI_PARAM_CONN_ADDRESS: - mutex_lock(&conn->xmitmutex); - if (!tcp_conn->sock) { - mutex_unlock(&conn->xmitmutex); - return -EINVAL; - } - - sk = tcp_conn->sock->sk; - if (sk->sk_family == PF_INET) { - inet = inet_sk(sk); - len = sprintf(buf, "%u.%u.%u.%u\n", - NIPQUAD(inet->daddr)); - } else { - np = inet6_sk(sk); - len = sprintf(buf, - "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", - NIP6(np->daddr)); - } - mutex_unlock(&conn->xmitmutex); - break; - default: - return -EINVAL; - } - - return len; -} - static void iscsi_conn_get_stats(struct iscsi_cls_conn *cls_conn, struct iscsi_stats *stats) { struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_tcp_conn *tcp_conn = conn->dd_data; stats->txdata_octets = conn->txdata_octets; stats->rxdata_octets = conn->rxdata_octets; @@ -2411,141 +3583,68 @@ iscsi_conn_get_stats(struct iscsi_cls_conn *cls_conn, struct iscsi_stats *stats) stats->tmfrsp_pdus = conn->tmfrsp_pdus_cnt; stats->custom_length = 3; strcpy(stats->custom[0].desc, "tx_sendpage_failures"); - stats->custom[0].value = tcp_conn->sendpage_failures_cnt; + stats->custom[0].value = conn->sendpage_failures_cnt; strcpy(stats->custom[1].desc, "rx_discontiguous_hdr"); - stats->custom[1].value = tcp_conn->discontiguous_hdr_cnt; + stats->custom[1].value = conn->discontiguous_hdr_cnt; strcpy(stats->custom[2].desc, "eh_abort_cnt"); stats->custom[2].value = conn->eh_abort_cnt; } -static struct iscsi_cls_session * -iscsi_tcp_session_create(struct iscsi_transport *iscsit, - struct scsi_transport_template *scsit, - uint32_t initial_cmdsn, uint32_t *hostno) +static int +iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size) { - struct iscsi_cls_session *cls_session; - struct iscsi_session *session; - uint32_t hn; - int cmd_i; - - cls_session = iscsi_session_setup(iscsit, scsit, - sizeof(struct iscsi_tcp_cmd_task), - sizeof(struct iscsi_tcp_mgmt_task), - initial_cmdsn, &hn); - if (!cls_session) - return NULL; - *hostno = hn; - - session = class_to_transport_session(cls_session); - for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) { - struct iscsi_cmd_task *ctask = session->cmds[cmd_i]; - struct iscsi_tcp_cmd_task *tcp_ctask = ctask->dd_data; - - ctask->hdr = &tcp_ctask->hdr; - } - - for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) { - struct iscsi_mgmt_task *mtask = session->mgmt_cmds[cmd_i]; - struct iscsi_tcp_mgmt_task *tcp_mtask = mtask->dd_data; - - mtask->hdr = &tcp_mtask->hdr; - } - - if (iscsi_r2tpool_alloc(class_to_transport_session(cls_session))) - goto r2tpool_alloc_fail; - - return cls_session; + struct iscsi_conn *conn = cls_conn->dd_data; + int rc; -r2tpool_alloc_fail: - iscsi_session_teardown(cls_session); - return NULL; -} + mutex_lock(&conn->xmitmutex); + rc = iscsi_conn_send_generic(conn, hdr, data, data_size); + mutex_unlock(&conn->xmitmutex); -static void iscsi_tcp_session_destroy(struct iscsi_cls_session *cls_session) -{ - iscsi_r2tpool_free(class_to_transport_session(cls_session)); - iscsi_session_teardown(cls_session); + return rc; } -static struct scsi_host_template iscsi_sht = { - .name = "iSCSI Initiator over TCP/IP, v" - ISCSI_TCP_VERSION, - .queuecommand = iscsi_queuecommand, - .change_queue_depth = iscsi_change_queue_depth, - .can_queue = ISCSI_XMIT_CMDS_MAX - 1, - .sg_tablesize = ISCSI_SG_TABLESIZE, - .cmd_per_lun = ISCSI_DEF_CMD_PER_LUN, - .eh_abort_handler = iscsi_eh_abort, - .eh_host_reset_handler = iscsi_eh_host_reset, - .use_clustering = DISABLE_CLUSTERING, - .proc_name = "iscsi_tcp", - .this_id = -1, -}; - static struct iscsi_transport iscsi_tcp_transport = { .owner = THIS_MODULE, .name = "tcp", .caps = CAP_RECOVERY_L0 | CAP_MULTI_R2T | CAP_HDRDGST | CAP_DATADGST, - .param_mask = ISCSI_MAX_RECV_DLENGTH | - ISCSI_MAX_XMIT_DLENGTH | - ISCSI_HDRDGST_EN | - ISCSI_DATADGST_EN | - ISCSI_INITIAL_R2T_EN | - ISCSI_MAX_R2T | - ISCSI_IMM_DATA_EN | - ISCSI_FIRST_BURST | - ISCSI_MAX_BURST | - ISCSI_PDU_INORDER_EN | - ISCSI_DATASEQ_INORDER_EN | - ISCSI_ERL | - ISCSI_CONN_PORT | - ISCSI_CONN_ADDRESS | - ISCSI_EXP_STATSN, .host_template = &iscsi_sht, + .hostdata_size = sizeof(struct iscsi_session), .conndata_size = sizeof(struct iscsi_conn), .max_conn = 1, .max_cmd_len = ISCSI_TCP_MAX_CMD_LEN, - /* session management */ - .create_session = iscsi_tcp_session_create, - .destroy_session = iscsi_tcp_session_destroy, - /* connection management */ - .create_conn = iscsi_tcp_conn_create, - .bind_conn = iscsi_tcp_conn_bind, - .destroy_conn = iscsi_tcp_conn_destroy, + .create_session = iscsi_session_create, + .destroy_session = iscsi_session_destroy, + .create_conn = iscsi_conn_create, + .bind_conn = iscsi_conn_bind, + .destroy_conn = iscsi_conn_destroy, .set_param = iscsi_conn_set_param, .get_conn_param = iscsi_conn_get_param, - .get_conn_str_param = iscsi_conn_get_str_param, .get_session_param = iscsi_session_get_param, .start_conn = iscsi_conn_start, .stop_conn = iscsi_conn_stop, - /* these are called as part of conn recovery */ - .suspend_conn_recv = iscsi_tcp_suspend_conn_rx, - .terminate_conn = iscsi_tcp_terminate_conn, - /* IO */ .send_pdu = iscsi_conn_send_pdu, .get_stats = iscsi_conn_get_stats, - .init_cmd_task = iscsi_tcp_cmd_init, - .init_mgmt_task = iscsi_tcp_mgmt_init, - .xmit_cmd_task = iscsi_tcp_ctask_xmit, - .xmit_mgmt_task = iscsi_tcp_mtask_xmit, - .cleanup_cmd_task = iscsi_tcp_cleanup_ctask, - /* recovery */ - .session_recovery_timedout = iscsi_session_recovery_timedout, }; static int __init iscsi_tcp_init(void) { if (iscsi_max_lun < 1) { - printk(KERN_ERR "iscsi_tcp: Invalid max_lun value of %u\n", - iscsi_max_lun); + printk(KERN_ERR "Invalid max_lun value of %u\n", iscsi_max_lun); return -EINVAL; } iscsi_tcp_transport.max_lun = iscsi_max_lun; + taskcache = kmem_cache_create("iscsi_taskcache", + sizeof(struct iscsi_data_task), 0, + SLAB_HWCACHE_ALIGN, NULL, NULL); + if (!taskcache) + return -ENOMEM; + if (!iscsi_register_transport(&iscsi_tcp_transport)) - return -ENODEV; + kmem_cache_destroy(taskcache); return 0; } @@ -2554,6 +3653,7 @@ static void __exit iscsi_tcp_exit(void) { iscsi_unregister_transport(&iscsi_tcp_transport); + kmem_cache_destroy(taskcache); } module_init(iscsi_tcp_init); diff --git a/drivers/scsi/iscsi_tcp.h b/drivers/scsi/iscsi_tcp.h index 808302832..ba26741ac 100644 --- a/drivers/scsi/iscsi_tcp.h +++ b/drivers/scsi/iscsi_tcp.h @@ -2,8 +2,7 @@ * iSCSI Initiator TCP Transport * Copyright (C) 2004 Dmitry Yusupov * Copyright (C) 2004 Alex Aizman - * Copyright (C) 2005 - 2006 Mike Christie - * Copyright (C) 2006 Red Hat, Inc. All rights reserved. + * Copyright (C) 2005 Mike Christie * maintained by open-iscsi@googlegroups.com * * This program is free software; you can redistribute it and/or modify @@ -22,7 +21,20 @@ #ifndef ISCSI_TCP_H #define ISCSI_TCP_H -#include +/* Session's states */ +#define ISCSI_STATE_FREE 1 +#define ISCSI_STATE_LOGGED_IN 2 +#define ISCSI_STATE_FAILED 3 +#define ISCSI_STATE_TERMINATE 4 + +/* Connection's states */ +#define ISCSI_CONN_INITIAL_STAGE 0 +#define ISCSI_CONN_STARTED 1 +#define ISCSI_CONN_STOPPED 2 +#define ISCSI_CONN_CLEANUP_WAIT 3 + +/* Connection suspend "bit" */ +#define SUSPEND_BIT 1 /* Socket's Receive state machine */ #define IN_PROGRESS_WAIT_HEADER 0x0 @@ -30,6 +42,12 @@ #define IN_PROGRESS_DATA_RECV 0x2 #define IN_PROGRESS_DDIGEST_RECV 0x3 +/* Task Mgmt states */ +#define TMABORT_INITIAL 0x0 +#define TMABORT_SUCCESS 0x1 +#define TMABORT_FAILED 0x2 +#define TMABORT_TIMEDOUT 0x3 + /* xmit state machine */ #define XMSTATE_IDLE 0x0 #define XMSTATE_R_HDR 0x1 @@ -44,14 +62,34 @@ #define XMSTATE_W_PAD 0x200 #define XMSTATE_DATA_DIGEST 0x400 +#define ISCSI_CONN_MAX 1 #define ISCSI_CONN_RCVBUF_MIN 262144 #define ISCSI_CONN_SNDBUF_MIN 262144 #define ISCSI_PAD_LEN 4 #define ISCSI_R2T_MAX 16 +#define ISCSI_XMIT_CMDS_MAX 128 /* must be power of 2 */ +#define ISCSI_MGMT_CMDS_MAX 32 /* must be power of 2 */ +#define ISCSI_MGMT_ITT_OFFSET 0xa00 #define ISCSI_SG_TABLESIZE SG_ALL +#define ISCSI_DEF_CMD_PER_LUN 32 +#define ISCSI_MAX_CMD_PER_LUN 128 #define ISCSI_TCP_MAX_CMD_LEN 16 -struct socket; +#define ITT_MASK (0xfff) +#define CID_SHIFT 12 +#define CID_MASK (0xffff< -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include -#include - -struct iscsi_session * -class_to_transport_session(struct iscsi_cls_session *cls_session) -{ - struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); - return iscsi_hostdata(shost->hostdata); -} -EXPORT_SYMBOL_GPL(class_to_transport_session); - -#define INVALID_SN_DELTA 0xffff - -int -iscsi_check_assign_cmdsn(struct iscsi_session *session, struct iscsi_nopin *hdr) -{ - uint32_t max_cmdsn = be32_to_cpu(hdr->max_cmdsn); - uint32_t exp_cmdsn = be32_to_cpu(hdr->exp_cmdsn); - - if (max_cmdsn < exp_cmdsn -1 && - max_cmdsn > exp_cmdsn - INVALID_SN_DELTA) - return ISCSI_ERR_MAX_CMDSN; - if (max_cmdsn > session->max_cmdsn || - max_cmdsn < session->max_cmdsn - INVALID_SN_DELTA) - session->max_cmdsn = max_cmdsn; - if (exp_cmdsn > session->exp_cmdsn || - exp_cmdsn < session->exp_cmdsn - INVALID_SN_DELTA) - session->exp_cmdsn = exp_cmdsn; - - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_check_assign_cmdsn); - -void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *ctask, - struct iscsi_data *hdr, - int transport_data_cnt) -{ - struct iscsi_conn *conn = ctask->conn; - - memset(hdr, 0, sizeof(struct iscsi_data)); - hdr->ttt = cpu_to_be32(ISCSI_RESERVED_TAG); - hdr->datasn = cpu_to_be32(ctask->unsol_datasn); - ctask->unsol_datasn++; - hdr->opcode = ISCSI_OP_SCSI_DATA_OUT; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - - hdr->itt = ctask->hdr->itt; - hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); - - hdr->offset = cpu_to_be32(ctask->total_length - - transport_data_cnt - - ctask->unsol_count); - - if (ctask->unsol_count > conn->max_xmit_dlength) { - hton24(hdr->dlength, conn->max_xmit_dlength); - ctask->data_count = conn->max_xmit_dlength; - hdr->flags = 0; - } else { - hton24(hdr->dlength, ctask->unsol_count); - ctask->data_count = ctask->unsol_count; - hdr->flags = ISCSI_FLAG_CMD_FINAL; - } -} -EXPORT_SYMBOL_GPL(iscsi_prep_unsolicit_data_pdu); - -/** - * iscsi_prep_scsi_cmd_pdu - prep iscsi scsi cmd pdu - * @ctask: iscsi cmd task - * - * Prep basic iSCSI PDU fields for a scsi cmd pdu. The LLD should set - * fields like dlength or final based on how much data it sends - */ -static void iscsi_prep_scsi_cmd_pdu(struct iscsi_cmd_task *ctask) -{ - struct iscsi_conn *conn = ctask->conn; - struct iscsi_session *session = conn->session; - struct iscsi_cmd *hdr = ctask->hdr; - struct scsi_cmnd *sc = ctask->sc; - - hdr->opcode = ISCSI_OP_SCSI_CMD; - hdr->flags = ISCSI_ATTR_SIMPLE; - int_to_scsilun(sc->device->lun, (struct scsi_lun *)hdr->lun); - hdr->itt = ctask->itt | (conn->id << ISCSI_CID_SHIFT) | - (session->age << ISCSI_AGE_SHIFT); - hdr->data_length = cpu_to_be32(sc->request_bufflen); - hdr->cmdsn = cpu_to_be32(session->cmdsn); - session->cmdsn++; - hdr->exp_statsn = cpu_to_be32(conn->exp_statsn); - memcpy(hdr->cdb, sc->cmnd, sc->cmd_len); - memset(&hdr->cdb[sc->cmd_len], 0, MAX_COMMAND_SIZE - sc->cmd_len); - - if (sc->sc_data_direction == DMA_TO_DEVICE) { - hdr->flags |= ISCSI_FLAG_CMD_WRITE; - /* - * Write counters: - * - * imm_count bytes to be sent right after - * SCSI PDU Header - * - * unsol_count bytes(as Data-Out) to be sent - * without R2T ack right after - * immediate data - * - * r2t_data_count bytes to be sent via R2T ack's - * - * pad_count bytes to be sent as zero-padding - */ - ctask->imm_count = 0; - ctask->unsol_count = 0; - ctask->unsol_datasn = 0; - - if (session->imm_data_en) { - if (ctask->total_length >= session->first_burst) - ctask->imm_count = min(session->first_burst, - conn->max_xmit_dlength); - else - ctask->imm_count = min(ctask->total_length, - conn->max_xmit_dlength); - hton24(ctask->hdr->dlength, ctask->imm_count); - } else - zero_data(ctask->hdr->dlength); - - if (!session->initial_r2t_en) - ctask->unsol_count = min(session->first_burst, - ctask->total_length) - ctask->imm_count; - if (!ctask->unsol_count) - /* No unsolicit Data-Out's */ - ctask->hdr->flags |= ISCSI_FLAG_CMD_FINAL; - } else { - ctask->datasn = 0; - hdr->flags |= ISCSI_FLAG_CMD_FINAL; - zero_data(hdr->dlength); - - if (sc->sc_data_direction == DMA_FROM_DEVICE) - hdr->flags |= ISCSI_FLAG_CMD_READ; - } - - conn->scsicmd_pdus_cnt++; -} -EXPORT_SYMBOL_GPL(iscsi_prep_scsi_cmd_pdu); - -/** - * iscsi_complete_command - return command back to scsi-ml - * @session: iscsi session - * @ctask: iscsi cmd task - * - * Must be called with session lock. - * This function returns the scsi command to scsi-ml and returns - * the cmd task to the pool of available cmd tasks. - */ -static void iscsi_complete_command(struct iscsi_session *session, - struct iscsi_cmd_task *ctask) -{ - struct scsi_cmnd *sc = ctask->sc; - - ctask->sc = NULL; - list_del_init(&ctask->running); - __kfifo_put(session->cmdpool.queue, (void*)&ctask, sizeof(void*)); - sc->scsi_done(sc); -} - -/** - * iscsi_cmd_rsp - SCSI Command Response processing - * @conn: iscsi connection - * @hdr: iscsi header - * @ctask: scsi command task - * @data: cmd data buffer - * @datalen: len of buffer - * - * iscsi_cmd_rsp sets up the scsi_cmnd fields based on the PDU and - * then completes the command and task. - **/ -static int iscsi_scsi_cmd_rsp(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - struct iscsi_cmd_task *ctask, char *data, - int datalen) -{ - int rc; - struct iscsi_cmd_rsp *rhdr = (struct iscsi_cmd_rsp *)hdr; - struct iscsi_session *session = conn->session; - struct scsi_cmnd *sc = ctask->sc; - - rc = iscsi_check_assign_cmdsn(session, (struct iscsi_nopin*)rhdr); - if (rc) { - sc->result = DID_ERROR << 16; - goto out; - } - - conn->exp_statsn = be32_to_cpu(rhdr->statsn) + 1; - - sc->result = (DID_OK << 16) | rhdr->cmd_status; - - if (rhdr->response != ISCSI_STATUS_CMD_COMPLETED) { - sc->result = DID_ERROR << 16; - goto out; - } - - if (rhdr->cmd_status == SAM_STAT_CHECK_CONDITION) { - int senselen; - - if (datalen < 2) { -invalid_datalen: - printk(KERN_ERR "iscsi: Got CHECK_CONDITION but " - "invalid data buffer size of %d\n", datalen); - sc->result = DID_BAD_TARGET << 16; - goto out; - } - - senselen = (data[0] << 8) | data[1]; - if (datalen < senselen) - goto invalid_datalen; - - memcpy(sc->sense_buffer, data + 2, - min(senselen, SCSI_SENSE_BUFFERSIZE)); - debug_scsi("copied %d bytes of sense\n", - min(senselen, SCSI_SENSE_BUFFERSIZE)); - } - - if (sc->sc_data_direction == DMA_TO_DEVICE) - goto out; - - if (rhdr->flags & ISCSI_FLAG_CMD_UNDERFLOW) { - int res_count = be32_to_cpu(rhdr->residual_count); - - if (res_count > 0 && res_count <= sc->request_bufflen) - sc->resid = res_count; - else - sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; - } else if (rhdr->flags & ISCSI_FLAG_CMD_BIDI_UNDERFLOW) - sc->result = (DID_BAD_TARGET << 16) | rhdr->cmd_status; - else if (rhdr->flags & ISCSI_FLAG_CMD_OVERFLOW) - sc->resid = be32_to_cpu(rhdr->residual_count); - -out: - debug_scsi("done [sc %lx res %d itt 0x%x]\n", - (long)sc, sc->result, ctask->itt); - conn->scsirsp_pdus_cnt++; - - iscsi_complete_command(conn->session, ctask); - return rc; -} - -/** - * __iscsi_complete_pdu - complete pdu - * @conn: iscsi conn - * @hdr: iscsi header - * @data: data buffer - * @datalen: len of data buffer - * - * Completes pdu processing by freeing any resources allocated at - * queuecommand or send generic. session lock must be held and verify - * itt must have been called. - */ -int __iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - char *data, int datalen) -{ - struct iscsi_session *session = conn->session; - int opcode = hdr->opcode & ISCSI_OPCODE_MASK, rc = 0; - struct iscsi_cmd_task *ctask; - struct iscsi_mgmt_task *mtask; - uint32_t itt; - - if (hdr->itt != cpu_to_be32(ISCSI_RESERVED_TAG)) - itt = hdr->itt & ISCSI_ITT_MASK; - else - itt = hdr->itt; - - if (itt < session->cmds_max) { - ctask = session->cmds[itt]; - - debug_scsi("cmdrsp [op 0x%x cid %d itt 0x%x len %d]\n", - opcode, conn->id, ctask->itt, datalen); - - switch(opcode) { - case ISCSI_OP_SCSI_CMD_RSP: - BUG_ON((void*)ctask != ctask->sc->SCp.ptr); - rc = iscsi_scsi_cmd_rsp(conn, hdr, ctask, data, - datalen); - break; - case ISCSI_OP_SCSI_DATA_IN: - BUG_ON((void*)ctask != ctask->sc->SCp.ptr); - if (hdr->flags & ISCSI_FLAG_DATA_STATUS) { - conn->scsirsp_pdus_cnt++; - iscsi_complete_command(session, ctask); - } - break; - case ISCSI_OP_R2T: - /* LLD handles this for now */ - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } - } else if (itt >= ISCSI_MGMT_ITT_OFFSET && - itt < ISCSI_MGMT_ITT_OFFSET + session->mgmtpool_max) { - mtask = session->mgmt_cmds[itt - ISCSI_MGMT_ITT_OFFSET]; - - debug_scsi("immrsp [op 0x%x cid %d itt 0x%x len %d]\n", - opcode, conn->id, mtask->itt, datalen); - - rc = iscsi_check_assign_cmdsn(session, - (struct iscsi_nopin*)hdr); - if (rc) - goto done; - - switch(opcode) { - case ISCSI_OP_LOGOUT_RSP: - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - /* fall through */ - case ISCSI_OP_LOGIN_RSP: - case ISCSI_OP_TEXT_RSP: - /* - * login related PDU's exp_statsn is handled in - * userspace - */ - rc = iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen); - list_del(&mtask->running); - if (conn->login_mtask != mtask) - __kfifo_put(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*)); - break; - case ISCSI_OP_SCSI_TMFUNC_RSP: - if (datalen) { - rc = ISCSI_ERR_PROTO; - break; - } - - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - conn->tmfrsp_pdus_cnt++; - if (conn->tmabort_state == TMABORT_INITIAL) { - conn->tmabort_state = - ((struct iscsi_tm_rsp *)hdr)-> - response == ISCSI_TMF_RSP_COMPLETE ? - TMABORT_SUCCESS:TMABORT_FAILED; - /* unblock eh_abort() */ - wake_up(&conn->ehwait); - } - break; - case ISCSI_OP_NOOP_IN: - if (hdr->ttt != ISCSI_RESERVED_TAG) { - rc = ISCSI_ERR_PROTO; - break; - } - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - - rc = iscsi_recv_pdu(conn->cls_conn, hdr, data, datalen); - list_del(&mtask->running); - if (conn->login_mtask != mtask) - __kfifo_put(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*)); - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } - } else if (itt == ISCSI_RESERVED_TAG) { - switch(opcode) { - case ISCSI_OP_NOOP_IN: - if (!datalen) { - rc = iscsi_check_assign_cmdsn(session, - (struct iscsi_nopin*)hdr); - if (!rc && hdr->ttt != ISCSI_RESERVED_TAG) - rc = iscsi_recv_pdu(conn->cls_conn, - hdr, NULL, 0); - } else - rc = ISCSI_ERR_PROTO; - break; - case ISCSI_OP_REJECT: - /* we need sth like iscsi_reject_rsp()*/ - case ISCSI_OP_ASYNC_EVENT: - conn->exp_statsn = be32_to_cpu(hdr->statsn) + 1; - /* we need sth like iscsi_async_event_rsp() */ - rc = ISCSI_ERR_BAD_OPCODE; - break; - default: - rc = ISCSI_ERR_BAD_OPCODE; - break; - } - } else - rc = ISCSI_ERR_BAD_ITT; - -done: - return rc; -} -EXPORT_SYMBOL_GPL(__iscsi_complete_pdu); - -int iscsi_complete_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - char *data, int datalen) -{ - int rc; - - spin_lock(&conn->session->lock); - rc = __iscsi_complete_pdu(conn, hdr, data, datalen); - spin_unlock(&conn->session->lock); - return rc; -} -EXPORT_SYMBOL_GPL(iscsi_complete_pdu); - -/* verify itt (itt encoding: age+cid+itt) */ -int iscsi_verify_itt(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - uint32_t *ret_itt) -{ - struct iscsi_session *session = conn->session; - struct iscsi_cmd_task *ctask; - uint32_t itt; - - if (hdr->itt != cpu_to_be32(ISCSI_RESERVED_TAG)) { - if ((hdr->itt & ISCSI_AGE_MASK) != - (session->age << ISCSI_AGE_SHIFT)) { - printk(KERN_ERR "iscsi: received itt %x expected " - "session age (%x)\n", hdr->itt, - session->age & ISCSI_AGE_MASK); - return ISCSI_ERR_BAD_ITT; - } - - if ((hdr->itt & ISCSI_CID_MASK) != - (conn->id << ISCSI_CID_SHIFT)) { - printk(KERN_ERR "iscsi: received itt %x, expected " - "CID (%x)\n", hdr->itt, conn->id); - return ISCSI_ERR_BAD_ITT; - } - itt = hdr->itt & ISCSI_ITT_MASK; - } else - itt = hdr->itt; - - if (itt < session->cmds_max) { - ctask = session->cmds[itt]; - - if (!ctask->sc) { - printk(KERN_INFO "iscsi: dropping ctask with " - "itt 0x%x\n", ctask->itt); - /* force drop */ - return ISCSI_ERR_NO_SCSI_CMD; - } - - if (ctask->sc->SCp.phase != session->age) { - printk(KERN_ERR "iscsi: ctask's session age %d, " - "expected %d\n", ctask->sc->SCp.phase, - session->age); - return ISCSI_ERR_SESSION_FAILED; - } - } - - *ret_itt = itt; - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_verify_itt); - -void iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err) -{ - struct iscsi_session *session = conn->session; - unsigned long flags; - - spin_lock_irqsave(&session->lock, flags); - if (session->state == ISCSI_STATE_FAILED) { - spin_unlock_irqrestore(&session->lock, flags); - return; - } - - if (conn->stop_stage == 0) - session->state = ISCSI_STATE_FAILED; - spin_unlock_irqrestore(&session->lock, flags); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); - iscsi_conn_error(conn->cls_conn, err); -} -EXPORT_SYMBOL_GPL(iscsi_conn_failure); - -/** - * iscsi_data_xmit - xmit any command into the scheduled connection - * @conn: iscsi connection - * - * Notes: - * The function can return -EAGAIN in which case the caller must - * re-schedule it again later or recover. '0' return code means - * successful xmit. - **/ -static int iscsi_data_xmit(struct iscsi_conn *conn) -{ - struct iscsi_transport *tt; - int rc = 0; - - if (unlikely(conn->suspend_tx)) { - debug_scsi("conn %d Tx suspended!\n", conn->id); - return -ENODATA; - } - tt = conn->session->tt; - - /* - * Transmit in the following order: - * - * 1) un-finished xmit (ctask or mtask) - * 2) immediate control PDUs - * 3) write data - * 4) SCSI commands - * 5) non-immediate control PDUs - * - * No need to lock around __kfifo_get as long as - * there's one producer and one consumer. - */ - - BUG_ON(conn->ctask && conn->mtask); - - if (conn->ctask) { - rc = tt->xmit_cmd_task(conn, conn->ctask); - if (rc) - goto again; - /* done with this in-progress ctask */ - conn->ctask = NULL; - } - if (conn->mtask) { - rc = tt->xmit_mgmt_task(conn, conn->mtask); - if (rc) - goto again; - /* done with this in-progress mtask */ - conn->mtask = NULL; - } - - /* process immediate first */ - if (unlikely(__kfifo_len(conn->immqueue))) { - while (__kfifo_get(conn->immqueue, (void*)&conn->mtask, - sizeof(void*))) { - spin_lock_bh(&conn->session->lock); - list_add_tail(&conn->mtask->running, - &conn->mgmt_run_list); - spin_unlock_bh(&conn->session->lock); - rc = tt->xmit_mgmt_task(conn, conn->mtask); - if (rc) - goto again; - } - /* done with this mtask */ - conn->mtask = NULL; - } - - /* process command queue */ - while (__kfifo_get(conn->xmitqueue, (void*)&conn->ctask, - sizeof(void*))) { - /* - * iscsi tcp may readd the task to the xmitqueue to send - * write data - */ - spin_lock_bh(&conn->session->lock); - if (list_empty(&conn->ctask->running)) - list_add_tail(&conn->ctask->running, &conn->run_list); - spin_unlock_bh(&conn->session->lock); - rc = tt->xmit_cmd_task(conn, conn->ctask); - if (rc) - goto again; - } - /* done with this ctask */ - conn->ctask = NULL; - - /* process the rest control plane PDUs, if any */ - if (unlikely(__kfifo_len(conn->mgmtqueue))) { - while (__kfifo_get(conn->mgmtqueue, (void*)&conn->mtask, - sizeof(void*))) { - spin_lock_bh(&conn->session->lock); - list_add_tail(&conn->mtask->running, - &conn->mgmt_run_list); - spin_unlock_bh(&conn->session->lock); - rc = tt->xmit_mgmt_task(conn, conn->mtask); - if (rc) - goto again; - } - /* done with this mtask */ - conn->mtask = NULL; - } - - return -ENODATA; - -again: - if (unlikely(conn->suspend_tx)) - return -ENODATA; - - return rc; -} - -static void iscsi_xmitworker(void *data) -{ - struct iscsi_conn *conn = data; - int rc; - /* - * serialize Xmit worker on a per-connection basis. - */ - mutex_lock(&conn->xmitmutex); - do { - rc = iscsi_data_xmit(conn); - } while (rc >= 0 || rc == -EAGAIN); - mutex_unlock(&conn->xmitmutex); -} - -enum { - FAILURE_BAD_HOST = 1, - FAILURE_SESSION_FAILED, - FAILURE_SESSION_FREED, - FAILURE_WINDOW_CLOSED, - FAILURE_SESSION_TERMINATE, - FAILURE_SESSION_IN_RECOVERY, - FAILURE_SESSION_RECOVERY_TIMEOUT, -}; - -int iscsi_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *)) -{ - struct Scsi_Host *host; - int reason = 0; - struct iscsi_session *session; - struct iscsi_conn *conn; - struct iscsi_cmd_task *ctask = NULL; - - sc->scsi_done = done; - sc->result = 0; - - host = sc->device->host; - session = iscsi_hostdata(host->hostdata); - - spin_lock(&session->lock); - - /* - * ISCSI_STATE_FAILED is a temp. state. The recovery - * code will decide what is best to do with command queued - * during this time - */ - if (session->state != ISCSI_STATE_LOGGED_IN && - session->state != ISCSI_STATE_FAILED) { - /* - * to handle the race between when we set the recovery state - * and block the session we requeue here (commands could - * be entering our queuecommand while a block is starting - * up because the block code is not locked) - */ - if (session->state == ISCSI_STATE_IN_RECOVERY) { - reason = FAILURE_SESSION_IN_RECOVERY; - goto reject; - } - - if (session->state == ISCSI_STATE_RECOVERY_FAILED) - reason = FAILURE_SESSION_RECOVERY_TIMEOUT; - else if (session->state == ISCSI_STATE_TERMINATE) - reason = FAILURE_SESSION_TERMINATE; - else - reason = FAILURE_SESSION_FREED; - goto fault; - } - - /* - * Check for iSCSI window and take care of CmdSN wrap-around - */ - if ((int)(session->max_cmdsn - session->cmdsn) < 0) { - reason = FAILURE_WINDOW_CLOSED; - goto reject; - } - - conn = session->leadconn; - - __kfifo_get(session->cmdpool.queue, (void*)&ctask, sizeof(void*)); - sc->SCp.phase = session->age; - sc->SCp.ptr = (char *)ctask; - - ctask->mtask = NULL; - ctask->conn = conn; - ctask->sc = sc; - INIT_LIST_HEAD(&ctask->running); - ctask->total_length = sc->request_bufflen; - iscsi_prep_scsi_cmd_pdu(ctask); - - session->tt->init_cmd_task(ctask); - - __kfifo_put(conn->xmitqueue, (void*)&ctask, sizeof(void*)); - debug_scsi( - "ctask enq [%s cid %d sc %lx itt 0x%x len %d cmdsn %d win %d]\n", - sc->sc_data_direction == DMA_TO_DEVICE ? "write" : "read", - conn->id, (long)sc, ctask->itt, sc->request_bufflen, - session->cmdsn, session->max_cmdsn - session->exp_cmdsn + 1); - spin_unlock(&session->lock); - - scsi_queue_work(host, &conn->xmitwork); - return 0; - -reject: - spin_unlock(&session->lock); - debug_scsi("cmd 0x%x rejected (%d)\n", sc->cmnd[0], reason); - return SCSI_MLQUEUE_HOST_BUSY; - -fault: - spin_unlock(&session->lock); - printk(KERN_ERR "iscsi: cmd 0x%x is not queued (%d)\n", - sc->cmnd[0], reason); - sc->result = (DID_NO_CONNECT << 16); - sc->resid = sc->request_bufflen; - sc->scsi_done(sc); - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_queuecommand); - -int iscsi_change_queue_depth(struct scsi_device *sdev, int depth) -{ - if (depth > ISCSI_MAX_CMD_PER_LUN) - depth = ISCSI_MAX_CMD_PER_LUN; - scsi_adjust_queue_depth(sdev, scsi_get_tag_type(sdev), depth); - return sdev->queue_depth; -} -EXPORT_SYMBOL_GPL(iscsi_change_queue_depth); - -static int -iscsi_conn_send_generic(struct iscsi_conn *conn, struct iscsi_hdr *hdr, - char *data, uint32_t data_size) -{ - struct iscsi_session *session = conn->session; - struct iscsi_nopout *nop = (struct iscsi_nopout *)hdr; - struct iscsi_mgmt_task *mtask; - - spin_lock_bh(&session->lock); - if (session->state == ISCSI_STATE_TERMINATE) { - spin_unlock_bh(&session->lock); - return -EPERM; - } - if (hdr->opcode == (ISCSI_OP_LOGIN | ISCSI_OP_IMMEDIATE) || - hdr->opcode == (ISCSI_OP_TEXT | ISCSI_OP_IMMEDIATE)) - /* - * Login and Text are sent serially, in - * request-followed-by-response sequence. - * Same mtask can be used. Same ITT must be used. - * Note that login_mtask is preallocated at conn_create(). - */ - mtask = conn->login_mtask; - else { - BUG_ON(conn->c_stage == ISCSI_CONN_INITIAL_STAGE); - BUG_ON(conn->c_stage == ISCSI_CONN_STOPPED); - - nop->exp_statsn = cpu_to_be32(conn->exp_statsn); - if (!__kfifo_get(session->mgmtpool.queue, - (void*)&mtask, sizeof(void*))) { - spin_unlock_bh(&session->lock); - return -ENOSPC; - } - } - - /* - * pre-format CmdSN for outgoing PDU. - */ - if (hdr->itt != cpu_to_be32(ISCSI_RESERVED_TAG)) { - hdr->itt = mtask->itt | (conn->id << ISCSI_CID_SHIFT) | - (session->age << ISCSI_AGE_SHIFT); - nop->cmdsn = cpu_to_be32(session->cmdsn); - if (conn->c_stage == ISCSI_CONN_STARTED && - !(hdr->opcode & ISCSI_OP_IMMEDIATE)) - session->cmdsn++; - } else - /* do not advance CmdSN */ - nop->cmdsn = cpu_to_be32(session->cmdsn); - - if (data_size) { - memcpy(mtask->data, data, data_size); - mtask->data_count = data_size; - } else - mtask->data_count = 0; - - INIT_LIST_HEAD(&mtask->running); - memcpy(mtask->hdr, hdr, sizeof(struct iscsi_hdr)); - if (session->tt->init_mgmt_task) - session->tt->init_mgmt_task(conn, mtask, data, data_size); - spin_unlock_bh(&session->lock); - - debug_scsi("mgmtpdu [op 0x%x hdr->itt 0x%x datalen %d]\n", - hdr->opcode, hdr->itt, data_size); - - /* - * since send_pdu() could be called at least from two contexts, - * we need to serialize __kfifo_put, so we don't have to take - * additional lock on fast data-path - */ - if (hdr->opcode & ISCSI_OP_IMMEDIATE) - __kfifo_put(conn->immqueue, (void*)&mtask, sizeof(void*)); - else - __kfifo_put(conn->mgmtqueue, (void*)&mtask, sizeof(void*)); - - scsi_queue_work(session->host, &conn->xmitwork); - return 0; -} - -int iscsi_conn_send_pdu(struct iscsi_cls_conn *cls_conn, struct iscsi_hdr *hdr, - char *data, uint32_t data_size) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - int rc; - - mutex_lock(&conn->xmitmutex); - rc = iscsi_conn_send_generic(conn, hdr, data, data_size); - mutex_unlock(&conn->xmitmutex); - - return rc; -} -EXPORT_SYMBOL_GPL(iscsi_conn_send_pdu); - -void iscsi_session_recovery_timedout(struct iscsi_cls_session *cls_session) -{ - struct iscsi_session *session = class_to_transport_session(cls_session); - struct iscsi_conn *conn = session->leadconn; - - spin_lock_bh(&session->lock); - if (session->state != ISCSI_STATE_LOGGED_IN) { - session->state = ISCSI_STATE_RECOVERY_FAILED; - if (conn) - wake_up(&conn->ehwait); - } - spin_unlock_bh(&session->lock); -} -EXPORT_SYMBOL_GPL(iscsi_session_recovery_timedout); - -int iscsi_eh_host_reset(struct scsi_cmnd *sc) -{ - struct Scsi_Host *host = sc->device->host; - struct iscsi_session *session = iscsi_hostdata(host->hostdata); - struct iscsi_conn *conn = session->leadconn; - int fail_session = 0; - - spin_lock_bh(&session->lock); - if (session->state == ISCSI_STATE_TERMINATE) { -failed: - debug_scsi("failing host reset: session terminated " - "[CID %d age %d]", conn->id, session->age); - spin_unlock_bh(&session->lock); - return FAILED; - } - - if (sc->SCp.phase == session->age) { - debug_scsi("failing connection CID %d due to SCSI host reset", - conn->id); - fail_session = 1; - } - spin_unlock_bh(&session->lock); - - /* - * we drop the lock here but the leadconn cannot be destoyed while - * we are in the scsi eh - */ - if (fail_session) - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - - debug_scsi("iscsi_eh_host_reset wait for relogin\n"); - wait_event_interruptible(conn->ehwait, - session->state == ISCSI_STATE_TERMINATE || - session->state == ISCSI_STATE_LOGGED_IN || - session->state == ISCSI_STATE_RECOVERY_FAILED); - if (signal_pending(current)) - flush_signals(current); - - spin_lock_bh(&session->lock); - if (session->state == ISCSI_STATE_LOGGED_IN) - printk(KERN_INFO "iscsi: host reset succeeded\n"); - else - goto failed; - spin_unlock_bh(&session->lock); - - return SUCCESS; -} -EXPORT_SYMBOL_GPL(iscsi_eh_host_reset); - -static void iscsi_tmabort_timedout(unsigned long data) -{ - struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)data; - struct iscsi_conn *conn = ctask->conn; - struct iscsi_session *session = conn->session; - - spin_lock(&session->lock); - if (conn->tmabort_state == TMABORT_INITIAL) { - conn->tmabort_state = TMABORT_TIMEDOUT; - debug_scsi("tmabort timedout [sc %p itt 0x%x]\n", - ctask->sc, ctask->itt); - /* unblock eh_abort() */ - wake_up(&conn->ehwait); - } - spin_unlock(&session->lock); -} - -/* must be called with the mutex lock */ -static int iscsi_exec_abort_task(struct scsi_cmnd *sc, - struct iscsi_cmd_task *ctask) -{ - struct iscsi_conn *conn = ctask->conn; - struct iscsi_session *session = conn->session; - struct iscsi_tm *hdr = &conn->tmhdr; - int rc; - - /* - * ctask timed out but session is OK requests must be serialized. - */ - memset(hdr, 0, sizeof(struct iscsi_tm)); - hdr->opcode = ISCSI_OP_SCSI_TMFUNC | ISCSI_OP_IMMEDIATE; - hdr->flags = ISCSI_TM_FUNC_ABORT_TASK; - hdr->flags |= ISCSI_FLAG_CMD_FINAL; - memcpy(hdr->lun, ctask->hdr->lun, sizeof(hdr->lun)); - hdr->rtt = ctask->hdr->itt; - hdr->refcmdsn = ctask->hdr->cmdsn; - - rc = iscsi_conn_send_generic(conn, (struct iscsi_hdr *)hdr, - NULL, 0); - if (rc) { - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - debug_scsi("abort sent failure [itt 0x%x] %d", ctask->itt, rc); - return rc; - } - - debug_scsi("abort sent [itt 0x%x]\n", ctask->itt); - - spin_lock_bh(&session->lock); - ctask->mtask = (struct iscsi_mgmt_task *) - session->mgmt_cmds[(hdr->itt & ISCSI_ITT_MASK) - - ISCSI_MGMT_ITT_OFFSET]; - - if (conn->tmabort_state == TMABORT_INITIAL) { - conn->tmfcmd_pdus_cnt++; - conn->tmabort_timer.expires = 10*HZ + jiffies; - conn->tmabort_timer.function = iscsi_tmabort_timedout; - conn->tmabort_timer.data = (unsigned long)ctask; - add_timer(&conn->tmabort_timer); - debug_scsi("abort set timeout [itt 0x%x]", ctask->itt); - } - spin_unlock_bh(&session->lock); - mutex_unlock(&conn->xmitmutex); - - /* - * block eh thread until: - * - * 1) abort response - * 2) abort timeout - * 3) session is terminated or restarted or userspace has - * given up on recovery - */ - wait_event_interruptible(conn->ehwait, - sc->SCp.phase != session->age || - session->state != ISCSI_STATE_LOGGED_IN || - conn->tmabort_state != TMABORT_INITIAL); - if (signal_pending(current)) - flush_signals(current); - del_timer_sync(&conn->tmabort_timer); - - mutex_lock(&conn->xmitmutex); - return 0; -} - -/* - * xmit mutex and session lock must be held - */ -#define iscsi_remove_task(tasktype) \ -static struct iscsi_##tasktype * \ -iscsi_remove_##tasktype(struct kfifo *fifo, uint32_t itt) \ -{ \ - int i, nr_tasks = __kfifo_len(fifo) / sizeof(void*); \ - struct iscsi_##tasktype *task; \ - \ - debug_scsi("searching %d tasks\n", nr_tasks); \ - \ - for (i = 0; i < nr_tasks; i++) { \ - __kfifo_get(fifo, (void*)&task, sizeof(void*)); \ - debug_scsi("check task %u\n", task->itt); \ - \ - if (task->itt == itt) { \ - debug_scsi("matched task\n"); \ - return task; \ - } \ - \ - __kfifo_put(fifo, (void*)&task, sizeof(void*)); \ - } \ - return NULL; \ -} - -iscsi_remove_task(mgmt_task); -iscsi_remove_task(cmd_task); - -static int iscsi_ctask_mtask_cleanup(struct iscsi_cmd_task *ctask) -{ - struct iscsi_conn *conn = ctask->conn; - struct iscsi_session *session = conn->session; - - if (!ctask->mtask) - return -EINVAL; - - if (!iscsi_remove_mgmt_task(conn->immqueue, ctask->mtask->itt)) - list_del(&ctask->mtask->running); - __kfifo_put(session->mgmtpool.queue, (void*)&ctask->mtask, - sizeof(void*)); - ctask->mtask = NULL; - return 0; -} - -/* - * session lock and xmitmutex must be held - */ -static void fail_command(struct iscsi_conn *conn, struct iscsi_cmd_task *ctask, - int err) -{ - struct scsi_cmnd *sc; - - conn->session->tt->cleanup_cmd_task(conn, ctask); - iscsi_ctask_mtask_cleanup(ctask); - - sc = ctask->sc; - if (!sc) - return; - sc->result = err; - sc->resid = sc->request_bufflen; - iscsi_complete_command(conn->session, ctask); -} - -int iscsi_eh_abort(struct scsi_cmnd *sc) -{ - struct iscsi_cmd_task *ctask = (struct iscsi_cmd_task *)sc->SCp.ptr; - struct iscsi_conn *conn = ctask->conn; - struct iscsi_session *session = conn->session; - struct iscsi_cmd_task *pending_ctask; - int rc; - - conn->eh_abort_cnt++; - debug_scsi("aborting [sc %p itt 0x%x]\n", sc, ctask->itt); - - mutex_lock(&conn->xmitmutex); - spin_lock_bh(&session->lock); - - /* - * If we are not logged in or we have started a new session - * then let the host reset code handle this - */ - if (session->state != ISCSI_STATE_LOGGED_IN || - sc->SCp.phase != session->age) - goto failed; - - /* ctask completed before time out */ - if (!ctask->sc) - goto success; - - /* what should we do here ? */ - if (conn->ctask == ctask) { - printk(KERN_INFO "iscsi: sc %p itt 0x%x partially sent. " - "Failing abort\n", sc, ctask->itt); - goto failed; - } - - /* check for the easy pending cmd abort */ - pending_ctask = iscsi_remove_cmd_task(conn->xmitqueue, ctask->itt); - if (pending_ctask) { - /* iscsi_tcp queues write transfers on the xmitqueue */ - if (list_empty(&pending_ctask->running)) { - debug_scsi("found pending task\n"); - goto success; - } else - __kfifo_put(conn->xmitqueue, (void*)&pending_ctask, - sizeof(void*)); - } - - conn->tmabort_state = TMABORT_INITIAL; - - spin_unlock_bh(&session->lock); - rc = iscsi_exec_abort_task(sc, ctask); - spin_lock_bh(&session->lock); - - iscsi_ctask_mtask_cleanup(ctask); - if (rc || sc->SCp.phase != session->age || - session->state != ISCSI_STATE_LOGGED_IN) - goto failed; - - /* ctask completed before tmf abort response */ - if (!ctask->sc) { - debug_scsi("sc completed while abort in progress\n"); - goto success; - } - - if (conn->tmabort_state != TMABORT_SUCCESS) { - spin_unlock_bh(&session->lock); - iscsi_conn_failure(conn, ISCSI_ERR_CONN_FAILED); - spin_lock_bh(&session->lock); - goto failed; - } - -success: - debug_scsi("abort success [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); - spin_unlock_bh(&session->lock); - - /* - * clean up task if aborted. we have the xmitmutex so grab - * the recv lock as a writer - */ - write_lock_bh(conn->recv_lock); - spin_lock(&session->lock); - fail_command(conn, ctask, DID_ABORT << 16); - spin_unlock(&session->lock); - write_unlock_bh(conn->recv_lock); - - mutex_unlock(&conn->xmitmutex); - return SUCCESS; - -failed: - spin_unlock_bh(&session->lock); - mutex_unlock(&conn->xmitmutex); - - debug_scsi("abort failed [sc %lx itt 0x%x]\n", (long)sc, ctask->itt); - return FAILED; -} -EXPORT_SYMBOL_GPL(iscsi_eh_abort); - -int -iscsi_pool_init(struct iscsi_queue *q, int max, void ***items, int item_size) -{ - int i; - - *items = kmalloc(max * sizeof(void*), GFP_KERNEL); - if (*items == NULL) - return -ENOMEM; - - q->max = max; - q->pool = kmalloc(max * sizeof(void*), GFP_KERNEL); - if (q->pool == NULL) { - kfree(*items); - return -ENOMEM; - } - - q->queue = kfifo_init((void*)q->pool, max * sizeof(void*), - GFP_KERNEL, NULL); - if (q->queue == ERR_PTR(-ENOMEM)) { - kfree(q->pool); - kfree(*items); - return -ENOMEM; - } - - for (i = 0; i < max; i++) { - q->pool[i] = kmalloc(item_size, GFP_KERNEL); - if (q->pool[i] == NULL) { - int j; - - for (j = 0; j < i; j++) - kfree(q->pool[j]); - - kfifo_free(q->queue); - kfree(q->pool); - kfree(*items); - return -ENOMEM; - } - memset(q->pool[i], 0, item_size); - (*items)[i] = q->pool[i]; - __kfifo_put(q->queue, (void*)&q->pool[i], sizeof(void*)); - } - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_pool_init); - -void iscsi_pool_free(struct iscsi_queue *q, void **items) -{ - int i; - - for (i = 0; i < q->max; i++) - kfree(items[i]); - kfree(q->pool); - kfree(items); -} -EXPORT_SYMBOL_GPL(iscsi_pool_free); - -/* - * iSCSI Session's hostdata organization: - * - * *------------------* <== hostdata_session(host->hostdata) - * | ptr to class sess| - * |------------------| <== iscsi_hostdata(host->hostdata) - * | iscsi_session | - * *------------------* - */ - -#define hostdata_privsize(_sz) (sizeof(unsigned long) + _sz + \ - _sz % sizeof(unsigned long)) - -#define hostdata_session(_hostdata) (iscsi_ptr(*(unsigned long *)_hostdata)) - -/** - * iscsi_session_setup - create iscsi cls session and host and session - * @scsit: scsi transport template - * @iscsit: iscsi transport template - * @initial_cmdsn: initial CmdSN - * @hostno: host no allocated - * - * This can be used by software iscsi_transports that allocate - * a session per scsi host. - **/ -struct iscsi_cls_session * -iscsi_session_setup(struct iscsi_transport *iscsit, - struct scsi_transport_template *scsit, - int cmd_task_size, int mgmt_task_size, - uint32_t initial_cmdsn, uint32_t *hostno) -{ - struct Scsi_Host *shost; - struct iscsi_session *session; - struct iscsi_cls_session *cls_session; - int cmd_i; - - shost = scsi_host_alloc(iscsit->host_template, - hostdata_privsize(sizeof(*session))); - if (!shost) - return NULL; - - shost->max_id = 1; - shost->max_channel = 0; - shost->max_lun = iscsit->max_lun; - shost->max_cmd_len = iscsit->max_cmd_len; - shost->transportt = scsit; - shost->transportt->create_work_queue = 1; - *hostno = shost->host_no; - - session = iscsi_hostdata(shost->hostdata); - memset(session, 0, sizeof(struct iscsi_session)); - session->host = shost; - session->state = ISCSI_STATE_FREE; - session->mgmtpool_max = ISCSI_MGMT_CMDS_MAX; - session->cmds_max = ISCSI_XMIT_CMDS_MAX; - session->cmdsn = initial_cmdsn; - session->exp_cmdsn = initial_cmdsn + 1; - session->max_cmdsn = initial_cmdsn + 1; - session->max_r2t = 1; - session->tt = iscsit; - - /* initialize SCSI PDU commands pool */ - if (iscsi_pool_init(&session->cmdpool, session->cmds_max, - (void***)&session->cmds, - cmd_task_size + sizeof(struct iscsi_cmd_task))) - goto cmdpool_alloc_fail; - - /* pre-format cmds pool with ITT */ - for (cmd_i = 0; cmd_i < session->cmds_max; cmd_i++) { - struct iscsi_cmd_task *ctask = session->cmds[cmd_i]; - - if (cmd_task_size) - ctask->dd_data = &ctask[1]; - ctask->itt = cmd_i; - } - - spin_lock_init(&session->lock); - INIT_LIST_HEAD(&session->connections); - - /* initialize immediate command pool */ - if (iscsi_pool_init(&session->mgmtpool, session->mgmtpool_max, - (void***)&session->mgmt_cmds, - mgmt_task_size + sizeof(struct iscsi_mgmt_task))) - goto mgmtpool_alloc_fail; - - - /* pre-format immediate cmds pool with ITT */ - for (cmd_i = 0; cmd_i < session->mgmtpool_max; cmd_i++) { - struct iscsi_mgmt_task *mtask = session->mgmt_cmds[cmd_i]; - - if (mgmt_task_size) - mtask->dd_data = &mtask[1]; - mtask->itt = ISCSI_MGMT_ITT_OFFSET + cmd_i; - } - - if (scsi_add_host(shost, NULL)) - goto add_host_fail; - - cls_session = iscsi_create_session(shost, iscsit, 0); - if (!cls_session) - goto cls_session_fail; - *(unsigned long*)shost->hostdata = (unsigned long)cls_session; - - return cls_session; - -cls_session_fail: - scsi_remove_host(shost); -add_host_fail: - iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds); -mgmtpool_alloc_fail: - iscsi_pool_free(&session->cmdpool, (void**)session->cmds); -cmdpool_alloc_fail: - scsi_host_put(shost); - return NULL; -} -EXPORT_SYMBOL_GPL(iscsi_session_setup); - -/** - * iscsi_session_teardown - destroy session, host, and cls_session - * shost: scsi host - * - * This can be used by software iscsi_transports that allocate - * a session per scsi host. - **/ -void iscsi_session_teardown(struct iscsi_cls_session *cls_session) -{ - struct Scsi_Host *shost = iscsi_session_to_shost(cls_session); - struct iscsi_session *session = iscsi_hostdata(shost->hostdata); - - scsi_remove_host(shost); - - iscsi_pool_free(&session->mgmtpool, (void**)session->mgmt_cmds); - iscsi_pool_free(&session->cmdpool, (void**)session->cmds); - - iscsi_destroy_session(cls_session); - scsi_host_put(shost); -} -EXPORT_SYMBOL_GPL(iscsi_session_teardown); - -/** - * iscsi_conn_setup - create iscsi_cls_conn and iscsi_conn - * @cls_session: iscsi_cls_session - * @conn_idx: cid - **/ -struct iscsi_cls_conn * -iscsi_conn_setup(struct iscsi_cls_session *cls_session, uint32_t conn_idx) -{ - struct iscsi_session *session = class_to_transport_session(cls_session); - struct iscsi_conn *conn; - struct iscsi_cls_conn *cls_conn; - char *data; - - cls_conn = iscsi_create_conn(cls_session, conn_idx); - if (!cls_conn) - return NULL; - conn = cls_conn->dd_data; - memset(conn, 0, sizeof(*conn)); - - conn->session = session; - conn->cls_conn = cls_conn; - conn->c_stage = ISCSI_CONN_INITIAL_STAGE; - conn->id = conn_idx; - conn->exp_statsn = 0; - conn->tmabort_state = TMABORT_INITIAL; - INIT_LIST_HEAD(&conn->run_list); - INIT_LIST_HEAD(&conn->mgmt_run_list); - - /* initialize general xmit PDU commands queue */ - conn->xmitqueue = kfifo_alloc(session->cmds_max * sizeof(void*), - GFP_KERNEL, NULL); - if (conn->xmitqueue == ERR_PTR(-ENOMEM)) - goto xmitqueue_alloc_fail; - - /* initialize general immediate & non-immediate PDU commands queue */ - conn->immqueue = kfifo_alloc(session->mgmtpool_max * sizeof(void*), - GFP_KERNEL, NULL); - if (conn->immqueue == ERR_PTR(-ENOMEM)) - goto immqueue_alloc_fail; - - conn->mgmtqueue = kfifo_alloc(session->mgmtpool_max * sizeof(void*), - GFP_KERNEL, NULL); - if (conn->mgmtqueue == ERR_PTR(-ENOMEM)) - goto mgmtqueue_alloc_fail; - - INIT_WORK(&conn->xmitwork, iscsi_xmitworker, conn); - - /* allocate login_mtask used for the login/text sequences */ - spin_lock_bh(&session->lock); - if (!__kfifo_get(session->mgmtpool.queue, - (void*)&conn->login_mtask, - sizeof(void*))) { - spin_unlock_bh(&session->lock); - goto login_mtask_alloc_fail; - } - spin_unlock_bh(&session->lock); - - data = kmalloc(DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH, GFP_KERNEL); - if (!data) - goto login_mtask_data_alloc_fail; - conn->login_mtask->data = data; - - init_timer(&conn->tmabort_timer); - mutex_init(&conn->xmitmutex); - init_waitqueue_head(&conn->ehwait); - - return cls_conn; - -login_mtask_data_alloc_fail: - __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, - sizeof(void*)); -login_mtask_alloc_fail: - kfifo_free(conn->mgmtqueue); -mgmtqueue_alloc_fail: - kfifo_free(conn->immqueue); -immqueue_alloc_fail: - kfifo_free(conn->xmitqueue); -xmitqueue_alloc_fail: - iscsi_destroy_conn(cls_conn); - return NULL; -} -EXPORT_SYMBOL_GPL(iscsi_conn_setup); - -/** - * iscsi_conn_teardown - teardown iscsi connection - * cls_conn: iscsi class connection - * - * TODO: we may need to make this into a two step process - * like scsi-mls remove + put host - */ -void iscsi_conn_teardown(struct iscsi_cls_conn *cls_conn) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_session *session = conn->session; - unsigned long flags; - - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - mutex_lock(&conn->xmitmutex); - if (conn->c_stage == ISCSI_CONN_INITIAL_STAGE) { - if (session->tt->suspend_conn_recv) - session->tt->suspend_conn_recv(conn); - - session->tt->terminate_conn(conn); - } - - spin_lock_bh(&session->lock); - conn->c_stage = ISCSI_CONN_CLEANUP_WAIT; - if (session->leadconn == conn) { - /* - * leading connection? then give up on recovery. - */ - session->state = ISCSI_STATE_TERMINATE; - wake_up(&conn->ehwait); - } - spin_unlock_bh(&session->lock); - - mutex_unlock(&conn->xmitmutex); - - /* - * Block until all in-progress commands for this connection - * time out or fail. - */ - for (;;) { - spin_lock_irqsave(session->host->host_lock, flags); - if (!session->host->host_busy) { /* OK for ERL == 0 */ - spin_unlock_irqrestore(session->host->host_lock, flags); - break; - } - spin_unlock_irqrestore(session->host->host_lock, flags); - msleep_interruptible(500); - printk(KERN_INFO "iscsi: scsi conn_destroy(): host_busy %d " - "host_failed %d\n", session->host->host_busy, - session->host->host_failed); - /* - * force eh_abort() to unblock - */ - wake_up(&conn->ehwait); - } - - spin_lock_bh(&session->lock); - kfree(conn->login_mtask->data); - __kfifo_put(session->mgmtpool.queue, (void*)&conn->login_mtask, - sizeof(void*)); - list_del(&conn->item); - if (list_empty(&session->connections)) - session->leadconn = NULL; - if (session->leadconn && session->leadconn == conn) - session->leadconn = container_of(session->connections.next, - struct iscsi_conn, item); - - if (session->leadconn == NULL) - /* no connections exits.. reset sequencing */ - session->cmdsn = session->max_cmdsn = session->exp_cmdsn = 1; - spin_unlock_bh(&session->lock); - - kfifo_free(conn->xmitqueue); - kfifo_free(conn->immqueue); - kfifo_free(conn->mgmtqueue); - - iscsi_destroy_conn(cls_conn); -} -EXPORT_SYMBOL_GPL(iscsi_conn_teardown); - -int iscsi_conn_start(struct iscsi_cls_conn *cls_conn) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_session *session = conn->session; - - if (session == NULL) { - printk(KERN_ERR "iscsi: can't start unbound connection\n"); - return -EPERM; - } - - spin_lock_bh(&session->lock); - conn->c_stage = ISCSI_CONN_STARTED; - session->state = ISCSI_STATE_LOGGED_IN; - - switch(conn->stop_stage) { - case STOP_CONN_RECOVER: - /* - * unblock eh_abort() if it is blocked. re-try all - * commands after successful recovery - */ - conn->stop_stage = 0; - conn->tmabort_state = TMABORT_INITIAL; - session->age++; - spin_unlock_bh(&session->lock); - - iscsi_unblock_session(session_to_cls(session)); - wake_up(&conn->ehwait); - return 0; - case STOP_CONN_TERM: - conn->stop_stage = 0; - break; - default: - break; - } - spin_unlock_bh(&session->lock); - - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_conn_start); - -static void -flush_control_queues(struct iscsi_session *session, struct iscsi_conn *conn) -{ - struct iscsi_mgmt_task *mtask, *tmp; - - /* handle pending */ - while (__kfifo_get(conn->immqueue, (void*)&mtask, sizeof(void*)) || - __kfifo_get(conn->mgmtqueue, (void*)&mtask, sizeof(void*))) { - if (mtask == conn->login_mtask) - continue; - debug_scsi("flushing pending mgmt task itt 0x%x\n", mtask->itt); - __kfifo_put(session->mgmtpool.queue, (void*)&mtask, - sizeof(void*)); - } - - /* handle running */ - list_for_each_entry_safe(mtask, tmp, &conn->mgmt_run_list, running) { - debug_scsi("flushing running mgmt task itt 0x%x\n", mtask->itt); - list_del(&mtask->running); - - if (mtask == conn->login_mtask) - continue; - __kfifo_put(session->mgmtpool.queue, (void*)&mtask, - sizeof(void*)); - } - - conn->mtask = NULL; -} - -/* Fail commands. Mutex and session lock held and recv side suspended */ -static void fail_all_commands(struct iscsi_conn *conn) -{ - struct iscsi_cmd_task *ctask, *tmp; - - /* flush pending */ - while (__kfifo_get(conn->xmitqueue, (void*)&ctask, sizeof(void*))) { - debug_scsi("failing pending sc %p itt 0x%x\n", ctask->sc, - ctask->itt); - fail_command(conn, ctask, DID_BUS_BUSY << 16); - } - - /* fail all other running */ - list_for_each_entry_safe(ctask, tmp, &conn->run_list, running) { - debug_scsi("failing in progress sc %p itt 0x%x\n", - ctask->sc, ctask->itt); - fail_command(conn, ctask, DID_BUS_BUSY << 16); - } - - conn->ctask = NULL; -} - -static void iscsi_start_session_recovery(struct iscsi_session *session, - struct iscsi_conn *conn, int flag) -{ - int old_stop_stage; - - spin_lock_bh(&session->lock); - if (conn->stop_stage == STOP_CONN_TERM) { - spin_unlock_bh(&session->lock); - return; - } - - /* - * When this is called for the in_login state, we only want to clean - * up the login task and connection. We do not need to block and set - * the recovery state again - */ - if (flag == STOP_CONN_TERM) - session->state = ISCSI_STATE_TERMINATE; - else if (conn->stop_stage != STOP_CONN_RECOVER) - session->state = ISCSI_STATE_IN_RECOVERY; - - old_stop_stage = conn->stop_stage; - conn->stop_stage = flag; - conn->c_stage = ISCSI_CONN_STOPPED; - set_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - spin_unlock_bh(&session->lock); - - if (session->tt->suspend_conn_recv) - session->tt->suspend_conn_recv(conn); - - mutex_lock(&conn->xmitmutex); - /* - * for connection level recovery we should not calculate - * header digest. conn->hdr_size used for optimization - * in hdr_extract() and will be re-negotiated at - * set_param() time. - */ - if (flag == STOP_CONN_RECOVER) { - conn->hdrdgst_en = 0; - conn->datadgst_en = 0; - if (session->state == ISCSI_STATE_IN_RECOVERY && - old_stop_stage != STOP_CONN_RECOVER) { - debug_scsi("blocking session\n"); - iscsi_block_session(session_to_cls(session)); - } - } - - session->tt->terminate_conn(conn); - /* - * flush queues. - */ - spin_lock_bh(&session->lock); - fail_all_commands(conn); - flush_control_queues(session, conn); - spin_unlock_bh(&session->lock); - - mutex_unlock(&conn->xmitmutex); -} - -void iscsi_conn_stop(struct iscsi_cls_conn *cls_conn, int flag) -{ - struct iscsi_conn *conn = cls_conn->dd_data; - struct iscsi_session *session = conn->session; - - switch (flag) { - case STOP_CONN_RECOVER: - case STOP_CONN_TERM: - iscsi_start_session_recovery(session, conn, flag); - break; - default: - printk(KERN_ERR "iscsi: invalid stop flag %d\n", flag); - } -} -EXPORT_SYMBOL_GPL(iscsi_conn_stop); - -int iscsi_conn_bind(struct iscsi_cls_session *cls_session, - struct iscsi_cls_conn *cls_conn, int is_leading) -{ - struct iscsi_session *session = class_to_transport_session(cls_session); - struct iscsi_conn *tmp = ERR_PTR(-EEXIST), *conn = cls_conn->dd_data; - - /* lookup for existing connection */ - spin_lock_bh(&session->lock); - list_for_each_entry(tmp, &session->connections, item) { - if (tmp == conn) { - if (conn->c_stage != ISCSI_CONN_STOPPED || - conn->stop_stage == STOP_CONN_TERM) { - printk(KERN_ERR "iscsi: can't bind " - "non-stopped connection (%d:%d)\n", - conn->c_stage, conn->stop_stage); - spin_unlock_bh(&session->lock); - return -EIO; - } - break; - } - } - if (tmp != conn) { - /* bind new iSCSI connection to session */ - conn->session = session; - list_add(&conn->item, &session->connections); - } - spin_unlock_bh(&session->lock); - - if (is_leading) - session->leadconn = conn; - - /* - * Unblock xmitworker(), Login Phase will pass through. - */ - clear_bit(ISCSI_SUSPEND_BIT, &conn->suspend_rx); - clear_bit(ISCSI_SUSPEND_BIT, &conn->suspend_tx); - return 0; -} -EXPORT_SYMBOL_GPL(iscsi_conn_bind); - -MODULE_AUTHOR("Mike Christie"); -MODULE_DESCRIPTION("iSCSI library functions"); -MODULE_LICENSE("GPL"); diff --git a/drivers/scsi/scsi_ioctl.c b/drivers/scsi/scsi_ioctl.c index a89aff61d..46ababf31 100644 --- a/drivers/scsi/scsi_ioctl.c +++ b/drivers/scsi/scsi_ioctl.c @@ -110,11 +110,8 @@ static int ioctl_internal_command(struct scsi_device *sdev, char *cmd, sshdr.asc, sshdr.ascq); break; case NOT_READY: /* This happens if there is no disc in drive */ - if (sdev->removable && (cmd[0] != TEST_UNIT_READY)) { - printk(KERN_INFO "Device not ready. Make sure" - " there is a disc in the drive.\n"); + if (sdev->removable) break; - } case UNIT_ATTENTION: if (sdev->removable) { sdev->changed = 1; diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 8bb8222ea..2730d507e 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -1,8 +1,10 @@ -/* +/* * iSCSI transport class definitions * * Copyright (C) IBM Corporation, 2004 - * Copyright (C) Mike Christie, 2004 + * Copyright (C) Mike Christie, 2004 - 2005 + * Copyright (C) Dmitry Yusupov, 2004 - 2005 + * Copyright (C) Alex Aizman, 2004 - 2005 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -19,370 +21,1293 @@ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. */ #include +#include +#include +#include #include #include #include #include #include +#include -#define ISCSI_SESSION_ATTRS 20 -#define ISCSI_HOST_ATTRS 2 +#define ISCSI_SESSION_ATTRS 8 +#define ISCSI_CONN_ATTRS 6 struct iscsi_internal { struct scsi_transport_template t; - struct iscsi_function_template *fnt; + struct iscsi_transport *iscsi_transport; + struct list_head list; + /* + * based on transport capabilities, at register time we set these + * bits to tell the transport class it wants attributes displayed + * in sysfs or that it can support different iSCSI Data-Path + * capabilities + */ + uint32_t param_mask; + + struct class_device cdev; /* * We do not have any private or other attrs. */ + struct transport_container conn_cont; + struct class_device_attribute *conn_attrs[ISCSI_CONN_ATTRS + 1]; + struct transport_container session_cont; struct class_device_attribute *session_attrs[ISCSI_SESSION_ATTRS + 1]; - struct class_device_attribute *host_attrs[ISCSI_HOST_ATTRS + 1]; }; -#define to_iscsi_internal(tmpl) container_of(tmpl, struct iscsi_internal, t) +/* + * list of registered transports and lock that must + * be held while accessing list. The iscsi_transport_lock must + * be acquired after the rx_queue_mutex. + */ +static LIST_HEAD(iscsi_transports); +static DEFINE_SPINLOCK(iscsi_transport_lock); + +#define to_iscsi_internal(tmpl) \ + container_of(tmpl, struct iscsi_internal, t) + +#define cdev_to_iscsi_internal(_cdev) \ + container_of(_cdev, struct iscsi_internal, cdev) + +static void iscsi_transport_release(struct class_device *cdev) +{ + struct iscsi_internal *priv = cdev_to_iscsi_internal(cdev); + kfree(priv); +} + +/* + * iscsi_transport_class represents the iscsi_transports that are + * registered. + */ +static struct class iscsi_transport_class = { + .name = "iscsi_transport", + .release = iscsi_transport_release, +}; + +static ssize_t +show_transport_handle(struct class_device *cdev, char *buf) +{ + struct iscsi_internal *priv = cdev_to_iscsi_internal(cdev); + return sprintf(buf, "%llu\n", (unsigned long long)iscsi_handle(priv->iscsi_transport)); +} +static CLASS_DEVICE_ATTR(handle, S_IRUGO, show_transport_handle, NULL); + +#define show_transport_attr(name, format) \ +static ssize_t \ +show_transport_##name(struct class_device *cdev, char *buf) \ +{ \ + struct iscsi_internal *priv = cdev_to_iscsi_internal(cdev); \ + return sprintf(buf, format"\n", priv->iscsi_transport->name); \ +} \ +static CLASS_DEVICE_ATTR(name, S_IRUGO, show_transport_##name, NULL); + +show_transport_attr(caps, "0x%x"); +show_transport_attr(max_lun, "%d"); +show_transport_attr(max_conn, "%d"); +show_transport_attr(max_cmd_len, "%d"); + +static struct attribute *iscsi_transport_attrs[] = { + &class_device_attr_handle.attr, + &class_device_attr_caps.attr, + &class_device_attr_max_lun.attr, + &class_device_attr_max_conn.attr, + &class_device_attr_max_cmd_len.attr, + NULL, +}; + +static struct attribute_group iscsi_transport_group = { + .attrs = iscsi_transport_attrs, +}; -static DECLARE_TRANSPORT_CLASS(iscsi_transport_class, - "iscsi_transport", +static DECLARE_TRANSPORT_CLASS(iscsi_session_class, + "iscsi_session", NULL, NULL, NULL); -static DECLARE_TRANSPORT_CLASS(iscsi_host_class, - "iscsi_host", +static DECLARE_TRANSPORT_CLASS(iscsi_connection_class, + "iscsi_connection", NULL, NULL, NULL); + +static struct sock *nls; +static int daemon_pid; +static DEFINE_MUTEX(rx_queue_mutex); + +struct mempool_zone { + mempool_t *pool; + atomic_t allocated; + int size; + int hiwat; + struct list_head freequeue; + spinlock_t freelock; +}; + +static struct mempool_zone *z_reply; + /* - * iSCSI target and session attrs + * Z_MAX_* - actual mempool size allocated at the mempool_zone_init() time + * Z_HIWAT_* - zone's high watermark when if_error bit will be set to -ENOMEM + * so daemon will notice OOM on NETLINK tranposrt level and will + * be able to predict or change operational behavior */ -#define iscsi_session_show_fn(field, format) \ - \ -static ssize_t \ -show_session_##field(struct class_device *cdev, char *buf) \ -{ \ - struct scsi_target *starget = transport_class_to_starget(cdev); \ - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); \ - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); \ - \ - if (i->fnt->get_##field) \ - i->fnt->get_##field(starget); \ - return snprintf(buf, 20, format"\n", iscsi_##field(starget)); \ +#define Z_MAX_REPLY 8 +#define Z_HIWAT_REPLY 6 +#define Z_MAX_PDU 8 +#define Z_HIWAT_PDU 6 +#define Z_MAX_ERROR 16 +#define Z_HIWAT_ERROR 12 + +static LIST_HEAD(sesslist); +static DEFINE_SPINLOCK(sesslock); +static LIST_HEAD(connlist); +static DEFINE_SPINLOCK(connlock); + +static struct iscsi_cls_session *iscsi_session_lookup(uint64_t handle) +{ + unsigned long flags; + struct iscsi_cls_session *sess; + + spin_lock_irqsave(&sesslock, flags); + list_for_each_entry(sess, &sesslist, sess_list) { + if (sess == iscsi_ptr(handle)) { + spin_unlock_irqrestore(&sesslock, flags); + return sess; + } + } + spin_unlock_irqrestore(&sesslock, flags); + return NULL; } -#define iscsi_session_rd_attr(field, format) \ - iscsi_session_show_fn(field, format) \ -static CLASS_DEVICE_ATTR(field, S_IRUGO, show_session_##field, NULL); +static struct iscsi_cls_conn *iscsi_conn_lookup(uint64_t handle) +{ + unsigned long flags; + struct iscsi_cls_conn *conn; -iscsi_session_rd_attr(tpgt, "%hu"); -iscsi_session_rd_attr(tsih, "%2x"); -iscsi_session_rd_attr(max_recv_data_segment_len, "%u"); -iscsi_session_rd_attr(max_burst_len, "%u"); -iscsi_session_rd_attr(first_burst_len, "%u"); -iscsi_session_rd_attr(def_time2wait, "%hu"); -iscsi_session_rd_attr(def_time2retain, "%hu"); -iscsi_session_rd_attr(max_outstanding_r2t, "%hu"); -iscsi_session_rd_attr(erl, "%d"); + spin_lock_irqsave(&connlock, flags); + list_for_each_entry(conn, &connlist, conn_list) { + if (conn == iscsi_ptr(handle)) { + spin_unlock_irqrestore(&connlock, flags); + return conn; + } + } + spin_unlock_irqrestore(&connlock, flags); + return NULL; +} +/* + * The following functions can be used by LLDs that allocate + * their own scsi_hosts or by software iscsi LLDs + */ +static void iscsi_session_release(struct device *dev) +{ + struct iscsi_cls_session *session = iscsi_dev_to_session(dev); + struct iscsi_transport *transport = session->transport; + struct Scsi_Host *shost; -#define iscsi_session_show_bool_fn(field) \ - \ -static ssize_t \ -show_session_bool_##field(struct class_device *cdev, char *buf) \ -{ \ - struct scsi_target *starget = transport_class_to_starget(cdev); \ - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); \ - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); \ - \ - if (i->fnt->get_##field) \ - i->fnt->get_##field(starget); \ - \ - if (iscsi_##field(starget)) \ - return sprintf(buf, "Yes\n"); \ - return sprintf(buf, "No\n"); \ + shost = iscsi_session_to_shost(session); + scsi_host_put(shost); + kfree(session); + module_put(transport->owner); +} + +static int iscsi_is_session_dev(const struct device *dev) +{ + return dev->release == iscsi_session_release; } -#define iscsi_session_rd_bool_attr(field) \ - iscsi_session_show_bool_fn(field) \ -static CLASS_DEVICE_ATTR(field, S_IRUGO, show_session_bool_##field, NULL); +/** + * iscsi_create_session - create iscsi class session + * @shost: scsi host + * @transport: iscsi transport + * + * This can be called from a LLD or iscsi_transport + **/ +struct iscsi_cls_session * +iscsi_create_session(struct Scsi_Host *shost, struct iscsi_transport *transport) +{ + struct iscsi_cls_session *session; + int err; -iscsi_session_rd_bool_attr(initial_r2t); -iscsi_session_rd_bool_attr(immediate_data); -iscsi_session_rd_bool_attr(data_pdu_in_order); -iscsi_session_rd_bool_attr(data_sequence_in_order); + if (!try_module_get(transport->owner)) + return NULL; -#define iscsi_session_show_digest_fn(field) \ - \ -static ssize_t \ -show_##field(struct class_device *cdev, char *buf) \ -{ \ - struct scsi_target *starget = transport_class_to_starget(cdev); \ - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); \ - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); \ - \ - if (i->fnt->get_##field) \ - i->fnt->get_##field(starget); \ - \ - if (iscsi_##field(starget)) \ - return sprintf(buf, "CRC32C\n"); \ - return sprintf(buf, "None\n"); \ + session = kzalloc(sizeof(*session), GFP_KERNEL); + if (!session) + goto module_put; + session->transport = transport; + + /* this is released in the dev's release function */ + scsi_host_get(shost); + snprintf(session->dev.bus_id, BUS_ID_SIZE, "session%u", shost->host_no); + session->dev.parent = &shost->shost_gendev; + session->dev.release = iscsi_session_release; + err = device_register(&session->dev); + if (err) { + dev_printk(KERN_ERR, &session->dev, "iscsi: could not " + "register session's dev\n"); + goto free_session; + } + transport_register_device(&session->dev); + + return session; + +free_session: + kfree(session); +module_put: + module_put(transport->owner); + return NULL; } -#define iscsi_session_rd_digest_attr(field) \ - iscsi_session_show_digest_fn(field) \ -static CLASS_DEVICE_ATTR(field, S_IRUGO, show_##field, NULL); +EXPORT_SYMBOL_GPL(iscsi_create_session); -iscsi_session_rd_digest_attr(header_digest); -iscsi_session_rd_digest_attr(data_digest); +/** + * iscsi_destroy_session - destroy iscsi session + * @session: iscsi_session + * + * Can be called by a LLD or iscsi_transport. There must not be + * any running connections. + **/ +int iscsi_destroy_session(struct iscsi_cls_session *session) +{ + transport_unregister_device(&session->dev); + device_unregister(&session->dev); + return 0; +} -static ssize_t -show_port(struct class_device *cdev, char *buf) +EXPORT_SYMBOL_GPL(iscsi_destroy_session); + +static void iscsi_conn_release(struct device *dev) { - struct scsi_target *starget = transport_class_to_starget(cdev); - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); + struct iscsi_cls_conn *conn = iscsi_dev_to_conn(dev); + struct device *parent = conn->dev.parent; - if (i->fnt->get_port) - i->fnt->get_port(starget); + kfree(conn); + put_device(parent); +} - return snprintf(buf, 20, "%hu\n", ntohs(iscsi_port(starget))); +static int iscsi_is_conn_dev(const struct device *dev) +{ + return dev->release == iscsi_conn_release; } -static CLASS_DEVICE_ATTR(port, S_IRUGO, show_port, NULL); -static ssize_t -show_ip_address(struct class_device *cdev, char *buf) +/** + * iscsi_create_conn - create iscsi class connection + * @session: iscsi cls session + * @cid: connection id + * + * This can be called from a LLD or iscsi_transport. The connection + * is child of the session so cid must be unique for all connections + * on the session. + **/ +struct iscsi_cls_conn * +iscsi_create_conn(struct iscsi_cls_session *session, uint32_t cid) { - struct scsi_target *starget = transport_class_to_starget(cdev); - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); + struct iscsi_transport *transport = session->transport; + struct Scsi_Host *shost = iscsi_session_to_shost(session); + struct iscsi_cls_conn *conn; + int err; + + conn = kzalloc(sizeof(*conn) + transport->conndata_size, GFP_KERNEL); + if (!conn) + return NULL; + + if (transport->conndata_size) + conn->dd_data = &conn[1]; + + INIT_LIST_HEAD(&conn->conn_list); + conn->transport = transport; - if (i->fnt->get_ip_address) - i->fnt->get_ip_address(starget); + /* this is released in the dev's release function */ + if (!get_device(&session->dev)) + goto free_conn; + snprintf(conn->dev.bus_id, BUS_ID_SIZE, "connection%d:%u", + shost->host_no, cid); + conn->dev.parent = &session->dev; + conn->dev.release = iscsi_conn_release; + err = device_register(&conn->dev); + if (err) { + dev_printk(KERN_ERR, &conn->dev, "iscsi: could not register " + "connection's dev\n"); + goto release_parent_ref; + } + transport_register_device(&conn->dev); + return conn; - if (iscsi_addr_type(starget) == AF_INET) - return sprintf(buf, "%u.%u.%u.%u\n", - NIPQUAD(iscsi_sin_addr(starget))); - else if(iscsi_addr_type(starget) == AF_INET6) - return sprintf(buf, "%04x:%04x:%04x:%04x:%04x:%04x:%04x:%04x\n", - NIP6(iscsi_sin6_addr(starget))); - return -EINVAL; +release_parent_ref: + put_device(&session->dev); +free_conn: + kfree(conn); + return NULL; } -static CLASS_DEVICE_ATTR(ip_address, S_IRUGO, show_ip_address, NULL); -static ssize_t -show_isid(struct class_device *cdev, char *buf) +EXPORT_SYMBOL_GPL(iscsi_create_conn); + +/** + * iscsi_destroy_conn - destroy iscsi class connection + * @session: iscsi cls session + * + * This can be called from a LLD or iscsi_transport. + **/ +int iscsi_destroy_conn(struct iscsi_cls_conn *conn) +{ + transport_unregister_device(&conn->dev); + device_unregister(&conn->dev); + return 0; +} + +EXPORT_SYMBOL_GPL(iscsi_destroy_conn); + +/* + * These functions are used only by software iscsi_transports + * which do not allocate and more their scsi_hosts since this + * is initiated from userspace. + */ + +/* + * iSCSI Session's hostdata organization: + * + * *------------------* <== hostdata_session(host->hostdata) + * | ptr to class sess| + * |------------------| <== iscsi_hostdata(host->hostdata) + * | transport's data | + * *------------------* + */ + +#define hostdata_privsize(_t) (sizeof(unsigned long) + _t->hostdata_size + \ + _t->hostdata_size % sizeof(unsigned long)) + +#define hostdata_session(_hostdata) (iscsi_ptr(*(unsigned long *)_hostdata)) + +/** + * iscsi_transport_create_session - create iscsi cls session and host + * scsit: scsi transport template + * transport: iscsi transport template + * + * This can be used by software iscsi_transports that allocate + * a session per scsi host. + **/ +struct Scsi_Host * +iscsi_transport_create_session(struct scsi_transport_template *scsit, + struct iscsi_transport *transport) { - struct scsi_target *starget = transport_class_to_starget(cdev); - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); + struct iscsi_cls_session *session; + struct Scsi_Host *shost; + unsigned long flags; + + shost = scsi_host_alloc(transport->host_template, + hostdata_privsize(transport)); + if (!shost) { + printk(KERN_ERR "iscsi: can not allocate SCSI host for " + "session\n"); + return NULL; + } + + shost->max_id = 1; + shost->max_channel = 0; + shost->max_lun = transport->max_lun; + shost->max_cmd_len = transport->max_cmd_len; + shost->transportt = scsit; + shost->transportt->create_work_queue = 1; + + if (scsi_add_host(shost, NULL)) + goto free_host; + + session = iscsi_create_session(shost, transport); + if (!session) + goto remove_host; + + *(unsigned long*)shost->hostdata = (unsigned long)session; + spin_lock_irqsave(&sesslock, flags); + list_add(&session->sess_list, &sesslist); + spin_unlock_irqrestore(&sesslock, flags); + return shost; + +remove_host: + scsi_remove_host(shost); +free_host: + scsi_host_put(shost); + return NULL; +} - if (i->fnt->get_isid) - i->fnt->get_isid(starget); +EXPORT_SYMBOL_GPL(iscsi_transport_create_session); - return sprintf(buf, "%02x%02x%02x%02x%02x%02x\n", - iscsi_isid(starget)[0], iscsi_isid(starget)[1], - iscsi_isid(starget)[2], iscsi_isid(starget)[3], - iscsi_isid(starget)[4], iscsi_isid(starget)[5]); +/** + * iscsi_transport_destroy_session - destroy session and scsi host + * shost: scsi host + * + * This can be used by software iscsi_transports that allocate + * a session per scsi host. + **/ +int iscsi_transport_destroy_session(struct Scsi_Host *shost) +{ + struct iscsi_cls_session *session; + unsigned long flags; + + scsi_remove_host(shost); + session = hostdata_session(shost->hostdata); + spin_lock_irqsave(&sesslock, flags); + list_del(&session->sess_list); + spin_unlock_irqrestore(&sesslock, flags); + iscsi_destroy_session(session); + /* ref from host alloc */ + scsi_host_put(shost); + return 0; } -static CLASS_DEVICE_ATTR(isid, S_IRUGO, show_isid, NULL); + +EXPORT_SYMBOL_GPL(iscsi_transport_destroy_session); /* - * This is used for iSCSI names. Normally, we follow - * the transport class convention of having the lld - * set the field, but in these cases the value is - * too large. + * iscsi interface functions */ -#define iscsi_session_show_str_fn(field) \ - \ +static struct iscsi_internal * +iscsi_if_transport_lookup(struct iscsi_transport *tt) +{ + struct iscsi_internal *priv; + unsigned long flags; + + spin_lock_irqsave(&iscsi_transport_lock, flags); + list_for_each_entry(priv, &iscsi_transports, list) { + if (tt == priv->iscsi_transport) { + spin_unlock_irqrestore(&iscsi_transport_lock, flags); + return priv; + } + } + spin_unlock_irqrestore(&iscsi_transport_lock, flags); + return NULL; +} + +static inline struct list_head *skb_to_lh(struct sk_buff *skb) +{ + return (struct list_head *)&skb->cb; +} + +static void* +mempool_zone_alloc_skb(gfp_t gfp_mask, void *pool_data) +{ + struct mempool_zone *zone = pool_data; + + return alloc_skb(zone->size, gfp_mask); +} + +static void +mempool_zone_free_skb(void *element, void *pool_data) +{ + kfree_skb(element); +} + +static void +mempool_zone_complete(struct mempool_zone *zone) +{ + unsigned long flags; + struct list_head *lh, *n; + + spin_lock_irqsave(&zone->freelock, flags); + list_for_each_safe(lh, n, &zone->freequeue) { + struct sk_buff *skb = (struct sk_buff *)((char *)lh - + offsetof(struct sk_buff, cb)); + if (!skb_shared(skb)) { + list_del(skb_to_lh(skb)); + mempool_free(skb, zone->pool); + atomic_dec(&zone->allocated); + } + } + spin_unlock_irqrestore(&zone->freelock, flags); +} + +static struct mempool_zone * +mempool_zone_init(unsigned max, unsigned size, unsigned hiwat) +{ + struct mempool_zone *zp; + + zp = kzalloc(sizeof(*zp), GFP_KERNEL); + if (!zp) + return NULL; + + zp->size = size; + zp->hiwat = hiwat; + INIT_LIST_HEAD(&zp->freequeue); + spin_lock_init(&zp->freelock); + atomic_set(&zp->allocated, 0); + + zp->pool = mempool_create(max, mempool_zone_alloc_skb, + mempool_zone_free_skb, zp); + if (!zp->pool) { + kfree(zp); + return NULL; + } + + return zp; +} + +static void mempool_zone_destroy(struct mempool_zone *zp) +{ + mempool_destroy(zp->pool); + kfree(zp); +} + +static struct sk_buff* +mempool_zone_get_skb(struct mempool_zone *zone) +{ + struct sk_buff *skb; + + skb = mempool_alloc(zone->pool, GFP_ATOMIC); + if (skb) + atomic_inc(&zone->allocated); + return skb; +} + +static int +iscsi_unicast_skb(struct mempool_zone *zone, struct sk_buff *skb) +{ + unsigned long flags; + int rc; + + skb_get(skb); + rc = netlink_unicast(nls, skb, daemon_pid, MSG_DONTWAIT); + if (rc < 0) { + mempool_free(skb, zone->pool); + printk(KERN_ERR "iscsi: can not unicast skb (%d)\n", rc); + return rc; + } + + spin_lock_irqsave(&zone->freelock, flags); + list_add(skb_to_lh(skb), &zone->freequeue); + spin_unlock_irqrestore(&zone->freelock, flags); + + return 0; +} + +int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size) +{ + struct nlmsghdr *nlh; + struct sk_buff *skb; + struct iscsi_uevent *ev; + char *pdu; + int len = NLMSG_SPACE(sizeof(*ev) + sizeof(struct iscsi_hdr) + + data_size); + + mempool_zone_complete(conn->z_pdu); + + skb = mempool_zone_get_skb(conn->z_pdu); + if (!skb) { + iscsi_conn_error(conn, ISCSI_ERR_CONN_FAILED); + dev_printk(KERN_ERR, &conn->dev, "iscsi: can not deliver " + "control PDU: OOM\n"); + return -ENOMEM; + } + + nlh = __nlmsg_put(skb, daemon_pid, 0, 0, (len - sizeof(*nlh)), 0); + ev = NLMSG_DATA(nlh); + memset(ev, 0, sizeof(*ev)); + ev->transport_handle = iscsi_handle(conn->transport); + ev->type = ISCSI_KEVENT_RECV_PDU; + if (atomic_read(&conn->z_pdu->allocated) >= conn->z_pdu->hiwat) + ev->iferror = -ENOMEM; + ev->r.recv_req.conn_handle = iscsi_handle(conn); + pdu = (char*)ev + sizeof(*ev); + memcpy(pdu, hdr, sizeof(struct iscsi_hdr)); + memcpy(pdu + sizeof(struct iscsi_hdr), data, data_size); + + return iscsi_unicast_skb(conn->z_pdu, skb); +} +EXPORT_SYMBOL_GPL(iscsi_recv_pdu); + +void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error) +{ + struct nlmsghdr *nlh; + struct sk_buff *skb; + struct iscsi_uevent *ev; + int len = NLMSG_SPACE(sizeof(*ev)); + + mempool_zone_complete(conn->z_error); + + skb = mempool_zone_get_skb(conn->z_error); + if (!skb) { + dev_printk(KERN_ERR, &conn->dev, "iscsi: gracefully ignored " + "conn error (%d)\n", error); + return; + } + + nlh = __nlmsg_put(skb, daemon_pid, 0, 0, (len - sizeof(*nlh)), 0); + ev = NLMSG_DATA(nlh); + ev->transport_handle = iscsi_handle(conn->transport); + ev->type = ISCSI_KEVENT_CONN_ERROR; + if (atomic_read(&conn->z_error->allocated) >= conn->z_error->hiwat) + ev->iferror = -ENOMEM; + ev->r.connerror.error = error; + ev->r.connerror.conn_handle = iscsi_handle(conn); + + iscsi_unicast_skb(conn->z_error, skb); + + dev_printk(KERN_INFO, &conn->dev, "iscsi: detected conn error (%d)\n", + error); +} +EXPORT_SYMBOL_GPL(iscsi_conn_error); + +static int +iscsi_if_send_reply(int pid, int seq, int type, int done, int multi, + void *payload, int size) +{ + struct sk_buff *skb; + struct nlmsghdr *nlh; + int len = NLMSG_SPACE(size); + int flags = multi ? NLM_F_MULTI : 0; + int t = done ? NLMSG_DONE : type; + + mempool_zone_complete(z_reply); + + skb = mempool_zone_get_skb(z_reply); + /* + * FIXME: + * user is supposed to react on iferror == -ENOMEM; + * see iscsi_if_rx(). + */ + BUG_ON(!skb); + + nlh = __nlmsg_put(skb, pid, seq, t, (len - sizeof(*nlh)), 0); + nlh->nlmsg_flags = flags; + memcpy(NLMSG_DATA(nlh), payload, size); + return iscsi_unicast_skb(z_reply, skb); +} + +static int +iscsi_if_get_stats(struct iscsi_transport *transport, struct nlmsghdr *nlh) +{ + struct iscsi_uevent *ev = NLMSG_DATA(nlh); + struct iscsi_stats *stats; + struct sk_buff *skbstat; + struct iscsi_cls_conn *conn; + struct nlmsghdr *nlhstat; + struct iscsi_uevent *evstat; + int len = NLMSG_SPACE(sizeof(*ev) + + sizeof(struct iscsi_stats) + + sizeof(struct iscsi_stats_custom) * + ISCSI_STATS_CUSTOM_MAX); + int err = 0; + + conn = iscsi_conn_lookup(ev->u.get_stats.conn_handle); + if (!conn) + return -EEXIST; + + do { + int actual_size; + + mempool_zone_complete(conn->z_pdu); + + skbstat = mempool_zone_get_skb(conn->z_pdu); + if (!skbstat) { + dev_printk(KERN_ERR, &conn->dev, "iscsi: can not " + "deliver stats: OOM\n"); + return -ENOMEM; + } + + nlhstat = __nlmsg_put(skbstat, daemon_pid, 0, 0, + (len - sizeof(*nlhstat)), 0); + evstat = NLMSG_DATA(nlhstat); + memset(evstat, 0, sizeof(*evstat)); + evstat->transport_handle = iscsi_handle(conn->transport); + evstat->type = nlh->nlmsg_type; + if (atomic_read(&conn->z_pdu->allocated) >= conn->z_pdu->hiwat) + evstat->iferror = -ENOMEM; + evstat->u.get_stats.conn_handle = + ev->u.get_stats.conn_handle; + stats = (struct iscsi_stats *) + ((char*)evstat + sizeof(*evstat)); + memset(stats, 0, sizeof(*stats)); + + transport->get_stats(conn, stats); + actual_size = NLMSG_SPACE(sizeof(struct iscsi_uevent) + + sizeof(struct iscsi_stats) + + sizeof(struct iscsi_stats_custom) * + stats->custom_length); + actual_size -= sizeof(*nlhstat); + actual_size = NLMSG_LENGTH(actual_size); + skb_trim(skbstat, NLMSG_ALIGN(actual_size)); + nlhstat->nlmsg_len = actual_size; + + err = iscsi_unicast_skb(conn->z_pdu, skbstat); + } while (err < 0 && err != -ECONNREFUSED); + + return err; +} + +static int +iscsi_if_create_session(struct iscsi_internal *priv, struct iscsi_uevent *ev) +{ + struct iscsi_transport *transport = priv->iscsi_transport; + struct iscsi_cls_session *session; + uint32_t sid; + + session = transport->create_session(&priv->t, + ev->u.c_session.initial_cmdsn, + &sid); + if (!session) + return -ENOMEM; + + ev->r.c_session_ret.session_handle = iscsi_handle(session); + ev->r.c_session_ret.sid = sid; + return 0; +} + +static int +iscsi_if_create_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev) +{ + struct iscsi_cls_conn *conn; + struct iscsi_cls_session *session; + unsigned long flags; + + session = iscsi_session_lookup(ev->u.c_conn.session_handle); + if (!session) + return -EINVAL; + + conn = transport->create_conn(session, ev->u.c_conn.cid); + if (!conn) + return -ENOMEM; + + conn->z_pdu = mempool_zone_init(Z_MAX_PDU, + NLMSG_SPACE(sizeof(struct iscsi_uevent) + + sizeof(struct iscsi_hdr) + + DEFAULT_MAX_RECV_DATA_SEGMENT_LENGTH), + Z_HIWAT_PDU); + if (!conn->z_pdu) { + dev_printk(KERN_ERR, &conn->dev, "iscsi: can not allocate " + "pdu zone for new conn\n"); + goto destroy_conn; + } + + conn->z_error = mempool_zone_init(Z_MAX_ERROR, + NLMSG_SPACE(sizeof(struct iscsi_uevent)), + Z_HIWAT_ERROR); + if (!conn->z_error) { + dev_printk(KERN_ERR, &conn->dev, "iscsi: can not allocate " + "error zone for new conn\n"); + goto free_pdu_pool; + } + + ev->r.handle = iscsi_handle(conn); + + spin_lock_irqsave(&connlock, flags); + list_add(&conn->conn_list, &connlist); + conn->active = 1; + spin_unlock_irqrestore(&connlock, flags); + + return 0; + +free_pdu_pool: + mempool_zone_destroy(conn->z_pdu); +destroy_conn: + if (transport->destroy_conn) + transport->destroy_conn(conn->dd_data); + return -ENOMEM; +} + +static int +iscsi_if_destroy_conn(struct iscsi_transport *transport, struct iscsi_uevent *ev) +{ + unsigned long flags; + struct iscsi_cls_conn *conn; + struct mempool_zone *z_error, *z_pdu; + + conn = iscsi_conn_lookup(ev->u.d_conn.conn_handle); + if (!conn) + return -EINVAL; + spin_lock_irqsave(&connlock, flags); + conn->active = 0; + list_del(&conn->conn_list); + spin_unlock_irqrestore(&connlock, flags); + + z_pdu = conn->z_pdu; + z_error = conn->z_error; + + if (transport->destroy_conn) + transport->destroy_conn(conn); + + mempool_zone_destroy(z_pdu); + mempool_zone_destroy(z_error); + + return 0; +} + +static int +iscsi_if_recv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) +{ + int err = 0; + struct iscsi_uevent *ev = NLMSG_DATA(nlh); + struct iscsi_transport *transport = NULL; + struct iscsi_internal *priv; + struct iscsi_cls_session *session; + struct iscsi_cls_conn *conn; + + priv = iscsi_if_transport_lookup(iscsi_ptr(ev->transport_handle)); + if (!priv) + return -EINVAL; + transport = priv->iscsi_transport; + + if (!try_module_get(transport->owner)) + return -EINVAL; + + switch (nlh->nlmsg_type) { + case ISCSI_UEVENT_CREATE_SESSION: + err = iscsi_if_create_session(priv, ev); + break; + case ISCSI_UEVENT_DESTROY_SESSION: + session = iscsi_session_lookup(ev->u.d_session.session_handle); + if (session) + transport->destroy_session(session); + else + err = -EINVAL; + break; + case ISCSI_UEVENT_CREATE_CONN: + err = iscsi_if_create_conn(transport, ev); + break; + case ISCSI_UEVENT_DESTROY_CONN: + err = iscsi_if_destroy_conn(transport, ev); + break; + case ISCSI_UEVENT_BIND_CONN: + session = iscsi_session_lookup(ev->u.b_conn.session_handle); + conn = iscsi_conn_lookup(ev->u.b_conn.conn_handle); + + if (session && conn) + ev->r.retcode = transport->bind_conn(session, conn, + ev->u.b_conn.transport_fd, + ev->u.b_conn.is_leading); + else + err = -EINVAL; + break; + case ISCSI_UEVENT_SET_PARAM: + conn = iscsi_conn_lookup(ev->u.set_param.conn_handle); + if (conn) + ev->r.retcode = transport->set_param(conn, + ev->u.set_param.param, ev->u.set_param.value); + else + err = -EINVAL; + break; + case ISCSI_UEVENT_START_CONN: + conn = iscsi_conn_lookup(ev->u.start_conn.conn_handle); + if (conn) + ev->r.retcode = transport->start_conn(conn); + else + err = -EINVAL; + + break; + case ISCSI_UEVENT_STOP_CONN: + conn = iscsi_conn_lookup(ev->u.stop_conn.conn_handle); + if (conn) + transport->stop_conn(conn, ev->u.stop_conn.flag); + else + err = -EINVAL; + break; + case ISCSI_UEVENT_SEND_PDU: + conn = iscsi_conn_lookup(ev->u.send_pdu.conn_handle); + if (conn) + ev->r.retcode = transport->send_pdu(conn, + (struct iscsi_hdr*)((char*)ev + sizeof(*ev)), + (char*)ev + sizeof(*ev) + ev->u.send_pdu.hdr_size, + ev->u.send_pdu.data_size); + else + err = -EINVAL; + break; + case ISCSI_UEVENT_GET_STATS: + err = iscsi_if_get_stats(transport, nlh); + break; + default: + err = -EINVAL; + break; + } + + module_put(transport->owner); + return err; +} + +/* Get message from skb (based on rtnetlink_rcv_skb). Each message is + * processed by iscsi_if_recv_msg. Malformed skbs with wrong length are + * or invalid creds discarded silently. */ +static void +iscsi_if_rx(struct sock *sk, int len) +{ + struct sk_buff *skb; + + mutex_lock(&rx_queue_mutex); + while ((skb = skb_dequeue(&sk->sk_receive_queue)) != NULL) { + if (NETLINK_CREDS(skb)->uid) { + skb_pull(skb, skb->len); + goto free_skb; + } + daemon_pid = NETLINK_CREDS(skb)->pid; + + while (skb->len >= NLMSG_SPACE(0)) { + int err; + uint32_t rlen; + struct nlmsghdr *nlh; + struct iscsi_uevent *ev; + + nlh = (struct nlmsghdr *)skb->data; + if (nlh->nlmsg_len < sizeof(*nlh) || + skb->len < nlh->nlmsg_len) { + break; + } + + ev = NLMSG_DATA(nlh); + rlen = NLMSG_ALIGN(nlh->nlmsg_len); + if (rlen > skb->len) + rlen = skb->len; + + err = iscsi_if_recv_msg(skb, nlh); + if (err) { + ev->type = ISCSI_KEVENT_IF_ERROR; + ev->iferror = err; + } + do { + /* + * special case for GET_STATS: + * on success - sending reply and stats from + * inside of if_recv_msg(), + * on error - fall through. + */ + if (ev->type == ISCSI_UEVENT_GET_STATS && !err) + break; + err = iscsi_if_send_reply( + NETLINK_CREDS(skb)->pid, nlh->nlmsg_seq, + nlh->nlmsg_type, 0, 0, ev, sizeof(*ev)); + if (atomic_read(&z_reply->allocated) >= + z_reply->hiwat) + ev->iferror = -ENOMEM; + } while (err < 0 && err != -ECONNREFUSED); + skb_pull(skb, rlen); + } +free_skb: + kfree_skb(skb); + } + mutex_unlock(&rx_queue_mutex); +} + +#define iscsi_cdev_to_conn(_cdev) \ + iscsi_dev_to_conn(_cdev->dev) + +/* + * iSCSI connection attrs + */ +#define iscsi_conn_int_attr_show(param, format) \ static ssize_t \ -show_session_str_##field(struct class_device *cdev, char *buf) \ +show_conn_int_param_##param(struct class_device *cdev, char *buf) \ { \ - ssize_t ret = 0; \ - struct scsi_target *starget = transport_class_to_starget(cdev); \ - struct Scsi_Host *shost = dev_to_shost(starget->dev.parent); \ - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); \ + uint32_t value = 0; \ + struct iscsi_cls_conn *conn = iscsi_cdev_to_conn(cdev); \ + struct iscsi_transport *t = conn->transport; \ \ - if (i->fnt->get_##field) \ - ret = i->fnt->get_##field(starget, buf, PAGE_SIZE); \ - return ret; \ + t->get_conn_param(conn, param, &value); \ + return snprintf(buf, 20, format"\n", value); \ } -#define iscsi_session_rd_str_attr(field) \ - iscsi_session_show_str_fn(field) \ -static CLASS_DEVICE_ATTR(field, S_IRUGO, show_session_str_##field, NULL); +#define iscsi_conn_int_attr(field, param, format) \ + iscsi_conn_int_attr_show(param, format) \ +static CLASS_DEVICE_ATTR(field, S_IRUGO, show_conn_int_param_##param, NULL); -iscsi_session_rd_str_attr(target_name); -iscsi_session_rd_str_attr(target_alias); +iscsi_conn_int_attr(max_recv_dlength, ISCSI_PARAM_MAX_RECV_DLENGTH, "%u"); +iscsi_conn_int_attr(max_xmit_dlength, ISCSI_PARAM_MAX_XMIT_DLENGTH, "%u"); +iscsi_conn_int_attr(header_digest, ISCSI_PARAM_HDRDGST_EN, "%d"); +iscsi_conn_int_attr(data_digest, ISCSI_PARAM_DATADGST_EN, "%d"); +iscsi_conn_int_attr(ifmarker, ISCSI_PARAM_IFMARKER_EN, "%d"); +iscsi_conn_int_attr(ofmarker, ISCSI_PARAM_OFMARKER_EN, "%d"); -/* - * iSCSI host attrs - */ +#define iscsi_cdev_to_session(_cdev) \ + iscsi_dev_to_session(_cdev->dev) /* - * Again, this is used for iSCSI names. Normally, we follow - * the transport class convention of having the lld set - * the field, but in these cases the value is too large. + * iSCSI session attrs */ -#define iscsi_host_show_str_fn(field) \ - \ +#define iscsi_session_int_attr_show(param, format) \ static ssize_t \ -show_host_str_##field(struct class_device *cdev, char *buf) \ +show_session_int_param_##param(struct class_device *cdev, char *buf) \ { \ - int ret = 0; \ - struct Scsi_Host *shost = transport_class_to_shost(cdev); \ - struct iscsi_internal *i = to_iscsi_internal(shost->transportt); \ + uint32_t value = 0; \ + struct iscsi_cls_session *session = iscsi_cdev_to_session(cdev); \ + struct iscsi_transport *t = session->transport; \ \ - if (i->fnt->get_##field) \ - ret = i->fnt->get_##field(shost, buf, PAGE_SIZE); \ - return ret; \ + t->get_session_param(session, param, &value); \ + return snprintf(buf, 20, format"\n", value); \ } -#define iscsi_host_rd_str_attr(field) \ - iscsi_host_show_str_fn(field) \ -static CLASS_DEVICE_ATTR(field, S_IRUGO, show_host_str_##field, NULL); +#define iscsi_session_int_attr(field, param, format) \ + iscsi_session_int_attr_show(param, format) \ +static CLASS_DEVICE_ATTR(field, S_IRUGO, show_session_int_param_##param, NULL); -iscsi_host_rd_str_attr(initiator_name); -iscsi_host_rd_str_attr(initiator_alias); +iscsi_session_int_attr(initial_r2t, ISCSI_PARAM_INITIAL_R2T_EN, "%d"); +iscsi_session_int_attr(max_outstanding_r2t, ISCSI_PARAM_MAX_R2T, "%hu"); +iscsi_session_int_attr(immediate_data, ISCSI_PARAM_IMM_DATA_EN, "%d"); +iscsi_session_int_attr(first_burst_len, ISCSI_PARAM_FIRST_BURST, "%u"); +iscsi_session_int_attr(max_burst_len, ISCSI_PARAM_MAX_BURST, "%u"); +iscsi_session_int_attr(data_pdu_in_order, ISCSI_PARAM_PDU_INORDER_EN, "%d"); +iscsi_session_int_attr(data_seq_in_order, ISCSI_PARAM_DATASEQ_INORDER_EN, "%d"); +iscsi_session_int_attr(erl, ISCSI_PARAM_ERL, "%d"); -#define SETUP_SESSION_RD_ATTR(field) \ - if (i->fnt->show_##field) { \ - i->session_attrs[count] = &class_device_attr_##field; \ +#define SETUP_SESSION_RD_ATTR(field, param) \ + if (priv->param_mask & (1 << param)) { \ + priv->session_attrs[count] = &class_device_attr_##field;\ count++; \ } -#define SETUP_HOST_RD_ATTR(field) \ - if (i->fnt->show_##field) { \ - i->host_attrs[count] = &class_device_attr_##field; \ +#define SETUP_CONN_RD_ATTR(field, param) \ + if (priv->param_mask & (1 << param)) { \ + priv->conn_attrs[count] = &class_device_attr_##field; \ count++; \ } -static int iscsi_host_match(struct attribute_container *cont, - struct device *dev) +static int iscsi_session_match(struct attribute_container *cont, + struct device *dev) { + struct iscsi_cls_session *session; struct Scsi_Host *shost; - struct iscsi_internal *i; + struct iscsi_internal *priv; - if (!scsi_is_host_device(dev)) + if (!iscsi_is_session_dev(dev)) return 0; - shost = dev_to_shost(dev); - if (!shost->transportt || shost->transportt->host_attrs.ac.class - != &iscsi_host_class.class) + session = iscsi_dev_to_session(dev); + shost = iscsi_session_to_shost(session); + if (!shost->transportt) return 0; - i = to_iscsi_internal(shost->transportt); - - return &i->t.host_attrs.ac == cont; + priv = to_iscsi_internal(shost->transportt); + if (priv->session_cont.ac.class != &iscsi_session_class.class) + return 0; + + return &priv->session_cont.ac == cont; } -static int iscsi_target_match(struct attribute_container *cont, - struct device *dev) +static int iscsi_conn_match(struct attribute_container *cont, + struct device *dev) { + struct iscsi_cls_session *session; + struct iscsi_cls_conn *conn; struct Scsi_Host *shost; - struct iscsi_internal *i; + struct iscsi_internal *priv; - if (!scsi_is_target_device(dev)) + if (!iscsi_is_conn_dev(dev)) return 0; - shost = dev_to_shost(dev->parent); - if (!shost->transportt || shost->transportt->host_attrs.ac.class - != &iscsi_host_class.class) + conn = iscsi_dev_to_conn(dev); + session = iscsi_dev_to_session(conn->dev.parent); + shost = iscsi_session_to_shost(session); + + if (!shost->transportt) return 0; - i = to_iscsi_internal(shost->transportt); - - return &i->t.target_attrs.ac == cont; + priv = to_iscsi_internal(shost->transportt); + if (priv->conn_cont.ac.class != &iscsi_connection_class.class) + return 0; + + return &priv->conn_cont.ac == cont; } struct scsi_transport_template * -iscsi_attach_transport(struct iscsi_function_template *fnt) +iscsi_register_transport(struct iscsi_transport *tt) { - struct iscsi_internal *i = kmalloc(sizeof(struct iscsi_internal), - GFP_KERNEL); - int count = 0; + struct iscsi_internal *priv; + unsigned long flags; + int count = 0, err; + + BUG_ON(!tt); + + priv = iscsi_if_transport_lookup(tt); + if (priv) + return NULL; - if (unlikely(!i)) + priv = kzalloc(sizeof(*priv), GFP_KERNEL); + if (!priv) return NULL; + INIT_LIST_HEAD(&priv->list); + priv->iscsi_transport = tt; - memset(i, 0, sizeof(struct iscsi_internal)); - i->fnt = fnt; - - i->t.target_attrs.ac.attrs = &i->session_attrs[0]; - i->t.target_attrs.ac.class = &iscsi_transport_class.class; - i->t.target_attrs.ac.match = iscsi_target_match; - transport_container_register(&i->t.target_attrs); - i->t.target_size = sizeof(struct iscsi_class_session); - - SETUP_SESSION_RD_ATTR(tsih); - SETUP_SESSION_RD_ATTR(isid); - SETUP_SESSION_RD_ATTR(header_digest); - SETUP_SESSION_RD_ATTR(data_digest); - SETUP_SESSION_RD_ATTR(target_name); - SETUP_SESSION_RD_ATTR(target_alias); - SETUP_SESSION_RD_ATTR(port); - SETUP_SESSION_RD_ATTR(tpgt); - SETUP_SESSION_RD_ATTR(ip_address); - SETUP_SESSION_RD_ATTR(initial_r2t); - SETUP_SESSION_RD_ATTR(immediate_data); - SETUP_SESSION_RD_ATTR(max_recv_data_segment_len); - SETUP_SESSION_RD_ATTR(max_burst_len); - SETUP_SESSION_RD_ATTR(first_burst_len); - SETUP_SESSION_RD_ATTR(def_time2wait); - SETUP_SESSION_RD_ATTR(def_time2retain); - SETUP_SESSION_RD_ATTR(max_outstanding_r2t); - SETUP_SESSION_RD_ATTR(data_pdu_in_order); - SETUP_SESSION_RD_ATTR(data_sequence_in_order); - SETUP_SESSION_RD_ATTR(erl); + priv->cdev.class = &iscsi_transport_class; + snprintf(priv->cdev.class_id, BUS_ID_SIZE, "%s", tt->name); + err = class_device_register(&priv->cdev); + if (err) + goto free_priv; - BUG_ON(count > ISCSI_SESSION_ATTRS); - i->session_attrs[count] = NULL; + err = sysfs_create_group(&priv->cdev.kobj, &iscsi_transport_group); + if (err) + goto unregister_cdev; - i->t.host_attrs.ac.attrs = &i->host_attrs[0]; - i->t.host_attrs.ac.class = &iscsi_host_class.class; - i->t.host_attrs.ac.match = iscsi_host_match; - transport_container_register(&i->t.host_attrs); - i->t.host_size = 0; + /* setup parameters mask */ + priv->param_mask = 0xFFFFFFFF; + if (!(tt->caps & CAP_MULTI_R2T)) + priv->param_mask &= ~(1 << ISCSI_PARAM_MAX_R2T); + if (!(tt->caps & CAP_HDRDGST)) + priv->param_mask &= ~(1 << ISCSI_PARAM_HDRDGST_EN); + if (!(tt->caps & CAP_DATADGST)) + priv->param_mask &= ~(1 << ISCSI_PARAM_DATADGST_EN); + if (!(tt->caps & CAP_MARKERS)) { + priv->param_mask &= ~(1 << ISCSI_PARAM_IFMARKER_EN); + priv->param_mask &= ~(1 << ISCSI_PARAM_OFMARKER_EN); + } + /* connection parameters */ + priv->conn_cont.ac.attrs = &priv->conn_attrs[0]; + priv->conn_cont.ac.class = &iscsi_connection_class.class; + priv->conn_cont.ac.match = iscsi_conn_match; + transport_container_register(&priv->conn_cont); + + SETUP_CONN_RD_ATTR(max_recv_dlength, ISCSI_PARAM_MAX_RECV_DLENGTH); + SETUP_CONN_RD_ATTR(max_xmit_dlength, ISCSI_PARAM_MAX_XMIT_DLENGTH); + SETUP_CONN_RD_ATTR(header_digest, ISCSI_PARAM_HDRDGST_EN); + SETUP_CONN_RD_ATTR(data_digest, ISCSI_PARAM_DATADGST_EN); + SETUP_CONN_RD_ATTR(ifmarker, ISCSI_PARAM_IFMARKER_EN); + SETUP_CONN_RD_ATTR(ofmarker, ISCSI_PARAM_OFMARKER_EN); + + BUG_ON(count > ISCSI_CONN_ATTRS); + priv->conn_attrs[count] = NULL; count = 0; - SETUP_HOST_RD_ATTR(initiator_name); - SETUP_HOST_RD_ATTR(initiator_alias); - BUG_ON(count > ISCSI_HOST_ATTRS); - i->host_attrs[count] = NULL; + /* session parameters */ + priv->session_cont.ac.attrs = &priv->session_attrs[0]; + priv->session_cont.ac.class = &iscsi_session_class.class; + priv->session_cont.ac.match = iscsi_session_match; + transport_container_register(&priv->session_cont); + + SETUP_SESSION_RD_ATTR(initial_r2t, ISCSI_PARAM_INITIAL_R2T_EN); + SETUP_SESSION_RD_ATTR(max_outstanding_r2t, ISCSI_PARAM_MAX_R2T); + SETUP_SESSION_RD_ATTR(immediate_data, ISCSI_PARAM_IMM_DATA_EN); + SETUP_SESSION_RD_ATTR(first_burst_len, ISCSI_PARAM_FIRST_BURST); + SETUP_SESSION_RD_ATTR(max_burst_len, ISCSI_PARAM_MAX_BURST); + SETUP_SESSION_RD_ATTR(data_pdu_in_order, ISCSI_PARAM_PDU_INORDER_EN); + SETUP_SESSION_RD_ATTR(data_seq_in_order,ISCSI_PARAM_DATASEQ_INORDER_EN) + SETUP_SESSION_RD_ATTR(erl, ISCSI_PARAM_ERL); + + BUG_ON(count > ISCSI_SESSION_ATTRS); + priv->session_attrs[count] = NULL; + + spin_lock_irqsave(&iscsi_transport_lock, flags); + list_add(&priv->list, &iscsi_transports); + spin_unlock_irqrestore(&iscsi_transport_lock, flags); + + printk(KERN_NOTICE "iscsi: registered transport (%s)\n", tt->name); + return &priv->t; - return &i->t; +unregister_cdev: + class_device_unregister(&priv->cdev); +free_priv: + kfree(priv); + return NULL; } +EXPORT_SYMBOL_GPL(iscsi_register_transport); + +int iscsi_unregister_transport(struct iscsi_transport *tt) +{ + struct iscsi_internal *priv; + unsigned long flags; + + BUG_ON(!tt); + + mutex_lock(&rx_queue_mutex); + + priv = iscsi_if_transport_lookup(tt); + BUG_ON (!priv); + + spin_lock_irqsave(&iscsi_transport_lock, flags); + list_del(&priv->list); + spin_unlock_irqrestore(&iscsi_transport_lock, flags); -EXPORT_SYMBOL(iscsi_attach_transport); + transport_container_unregister(&priv->conn_cont); + transport_container_unregister(&priv->session_cont); -void iscsi_release_transport(struct scsi_transport_template *t) + sysfs_remove_group(&priv->cdev.kobj, &iscsi_transport_group); + class_device_unregister(&priv->cdev); + mutex_unlock(&rx_queue_mutex); + + return 0; +} +EXPORT_SYMBOL_GPL(iscsi_unregister_transport); + +static int +iscsi_rcv_nl_event(struct notifier_block *this, unsigned long event, void *ptr) { - struct iscsi_internal *i = to_iscsi_internal(t); + struct netlink_notify *n = ptr; - transport_container_unregister(&i->t.target_attrs); - transport_container_unregister(&i->t.host_attrs); - - kfree(i); + if (event == NETLINK_URELEASE && + n->protocol == NETLINK_ISCSI && n->pid) { + struct iscsi_cls_conn *conn; + unsigned long flags; + + mempool_zone_complete(z_reply); + spin_lock_irqsave(&connlock, flags); + list_for_each_entry(conn, &connlist, conn_list) { + mempool_zone_complete(conn->z_error); + mempool_zone_complete(conn->z_pdu); + } + spin_unlock_irqrestore(&connlock, flags); + } + + return NOTIFY_DONE; } -EXPORT_SYMBOL(iscsi_release_transport); +static struct notifier_block iscsi_nl_notifier = { + .notifier_call = iscsi_rcv_nl_event, +}; static __init int iscsi_transport_init(void) { - int err = transport_class_register(&iscsi_transport_class); + int err; + err = class_register(&iscsi_transport_class); if (err) return err; - return transport_class_register(&iscsi_host_class); + + err = transport_class_register(&iscsi_connection_class); + if (err) + goto unregister_transport_class; + + err = transport_class_register(&iscsi_session_class); + if (err) + goto unregister_conn_class; + + err = netlink_register_notifier(&iscsi_nl_notifier); + if (err) + goto unregister_session_class; + + nls = netlink_kernel_create(NETLINK_ISCSI, 1, iscsi_if_rx, + THIS_MODULE); + if (!nls) { + err = -ENOBUFS; + goto unregister_notifier; + } + + z_reply = mempool_zone_init(Z_MAX_REPLY, + NLMSG_SPACE(sizeof(struct iscsi_uevent)), Z_HIWAT_REPLY); + if (z_reply) + return 0; + + sock_release(nls->sk_socket); +unregister_notifier: + netlink_unregister_notifier(&iscsi_nl_notifier); +unregister_session_class: + transport_class_unregister(&iscsi_session_class); +unregister_conn_class: + transport_class_unregister(&iscsi_connection_class); +unregister_transport_class: + class_unregister(&iscsi_transport_class); + return err; } static void __exit iscsi_transport_exit(void) { - transport_class_unregister(&iscsi_host_class); - transport_class_unregister(&iscsi_transport_class); + mempool_zone_destroy(z_reply); + sock_release(nls->sk_socket); + netlink_unregister_notifier(&iscsi_nl_notifier); + transport_class_unregister(&iscsi_connection_class); + transport_class_unregister(&iscsi_session_class); + class_unregister(&iscsi_transport_class); } module_init(iscsi_transport_init); module_exit(iscsi_transport_exit); -MODULE_AUTHOR("Mike Christie"); -MODULE_DESCRIPTION("iSCSI Transport Attributes"); +MODULE_AUTHOR("Mike Christie , " + "Dmitry Yusupov , " + "Alex Aizman "); +MODULE_DESCRIPTION("iSCSI Transport Interface"); MODULE_LICENSE("GPL"); diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c index fbb53adee..968597781 100644 --- a/drivers/serial/8250.c +++ b/drivers/serial/8250.c @@ -7,6 +7,9 @@ * * Copyright (C) 2001 Russell King. * + * 2005/09/16: Enabled higher baud rates for 16C95x. + * (Mathias Adam ) + * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by * the Free Software Foundation; either version 2 of the License, or @@ -1745,6 +1748,14 @@ static unsigned int serial8250_get_divisor(struct uart_port *port, unsigned int else if ((port->flags & UPF_MAGIC_MULTIPLIER) && baud == (port->uartclk/8)) quot = 0x8002; + /* + * For 16C950s UART_TCR is used in combination with divisor==1 + * to achieve baud rates up to baud_base*4. + */ + else if ((port->type == PORT_16C950) && + baud > (port->uartclk/16)) + quot = 1; + else quot = uart_get_divisor(port, baud); @@ -1758,7 +1769,7 @@ serial8250_set_termios(struct uart_port *port, struct termios *termios, struct uart_8250_port *up = (struct uart_8250_port *)port; unsigned char cval, fcr = 0; unsigned long flags; - unsigned int baud, quot; + unsigned int baud, quot, max_baud; switch (termios->c_cflag & CSIZE) { case CS5: @@ -1790,7 +1801,8 @@ serial8250_set_termios(struct uart_port *port, struct termios *termios, /* * Ask the core to calculate the divisor for us. */ - baud = uart_get_baud_rate(port, termios, old, 0, port->uartclk/16); + max_baud = (up->port.type == PORT_16C950 ? port->uartclk/4 : port->uartclk/16); + baud = uart_get_baud_rate(port, termios, old, 0, max_baud); quot = serial8250_get_divisor(port, baud); /* @@ -1826,6 +1838,19 @@ serial8250_set_termios(struct uart_port *port, struct termios *termios, */ spin_lock_irqsave(&up->port.lock, flags); + /* + * 16C950 supports additional prescaler ratios between 1:16 and 1:4 + * thus increasing max baud rate to uartclk/4. + */ + if (up->port.type == PORT_16C950) { + if (baud == port->uartclk/4) + serial_icr_write(up, UART_TCR, 0x4); + else if (baud == port->uartclk/8) + serial_icr_write(up, UART_TCR, 0x8); + else + serial_icr_write(up, UART_TCR, 0); + } + /* * Update the per-port timeout. */ @@ -2243,10 +2268,14 @@ serial8250_console_write(struct console *co, const char *s, unsigned int count) touch_nmi_watchdog(); - if (oops_in_progress) { - locked = spin_trylock_irqsave(&up->port.lock, flags); + local_irq_save(flags); + if (up->port.sysrq) { + /* serial8250_handle_port() already took the lock */ + locked = 0; + } else if (oops_in_progress) { + locked = spin_trylock(&up->port.lock); } else - spin_lock_irqsave(&up->port.lock, flags); + spin_lock(&up->port.lock); /* * First save the IER then disable the interrupts @@ -2268,7 +2297,8 @@ serial8250_console_write(struct console *co, const char *s, unsigned int count) serial_out(up, UART_IER, ier); if (locked) - spin_unlock_irqrestore(&up->port.lock, flags); + spin_unlock(&up->port.lock); + local_irq_restore(flags); } static int serial8250_console_setup(struct console *co, char *options) diff --git a/drivers/serial/Kconfig b/drivers/serial/Kconfig index 7d22dc047..d2716173d 100644 --- a/drivers/serial/Kconfig +++ b/drivers/serial/Kconfig @@ -11,6 +11,7 @@ menu "Serial drivers" config SERIAL_8250 tristate "8250/16550 and compatible serial support" depends on (BROKEN || !SPARC) + depends on !XEN_DISABLE_SERIAL select SERIAL_CORE ---help--- This selects whether you want to include the driver for the standard @@ -803,6 +804,7 @@ config SERIAL_MPC52xx tristate "Freescale MPC52xx family PSC serial support" depends on PPC_MPC52xx select SERIAL_CORE + select FW_LOADER help This drivers support the MPC52xx PSC serial ports. If you would like to use them, you must answer Y or M to this option. Not that diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c index 7abe532d7..17839e753 100644 --- a/drivers/serial/serial_core.c +++ b/drivers/serial/serial_core.c @@ -1968,16 +1968,16 @@ int uart_resume_port(struct uart_driver *drv, struct uart_port *port) struct termios termios; /* - * Get the termios for this line + * First try to use the console cflag setting. */ - tty_get_termios(drv->tty_driver, port->line, &termios); + memset(&termios, 0, sizeof(struct termios)); + termios.c_cflag = port->cons->cflag; /* - * If the console cflag is still set, subsitute that - * for the termios cflag. + * If that's unset, use the tty termios setting. */ - if (port->cons->cflag) - termios.c_cflag = port->cons->cflag; + if (state->info && state->info->tty && termios.c_cflag == 0) + termios = *state->info->tty->termios; port->ops->set_termios(port, &termios, NULL); console_start(port->cons); diff --git a/drivers/usb/host/uhci-q.c b/drivers/usb/host/uhci-q.c index 2a7c19501..5f44354c4 100644 --- a/drivers/usb/host/uhci-q.c +++ b/drivers/usb/host/uhci-q.c @@ -13,13 +13,9 @@ * (C) Copyright 2000 Yggdrasil Computing, Inc. (port of new PCI interface * support from usb-ohci.c by Adam Richter, adam@yggdrasil.com). * (C) Copyright 1999 Gregory P. Smith (from usb-ohci.c) - * (C) Copyright 2004 Alan Stern, stern@rowland.harvard.edu + * (C) Copyright 2004-2005 Alan Stern, stern@rowland.harvard.edu */ -static int uhci_urb_dequeue(struct usb_hcd *hcd, struct urb *urb); -static void uhci_unlink_generic(struct uhci_hcd *uhci, struct urb *urb); -static void uhci_remove_pending_urbps(struct uhci_hcd *uhci); -static void uhci_free_pending_qhs(struct uhci_hcd *uhci); static void uhci_free_pending_tds(struct uhci_hcd *uhci); /* @@ -30,8 +26,10 @@ static void uhci_free_pending_tds(struct uhci_hcd *uhci); * games with the FSBR code to make sure we get the correct order in all * the cases. I don't think it's worth the effort */ -static inline void uhci_set_next_interrupt(struct uhci_hcd *uhci) +static void uhci_set_next_interrupt(struct uhci_hcd *uhci) { + if (uhci->is_stopped) + mod_timer(&uhci_to_hcd(uhci)->rh_timer, jiffies); uhci->term_td->status |= cpu_to_le32(TD_CTRL_IOC); } @@ -40,13 +38,7 @@ static inline void uhci_clear_next_interrupt(struct uhci_hcd *uhci) uhci->term_td->status &= ~cpu_to_le32(TD_CTRL_IOC); } -static inline void uhci_moveto_complete(struct uhci_hcd *uhci, - struct urb_priv *urbp) -{ - list_move_tail(&urbp->urb_list, &uhci->complete_list); -} - -static struct uhci_td *uhci_alloc_td(struct uhci_hcd *uhci, struct usb_device *dev) +static struct uhci_td *uhci_alloc_td(struct uhci_hcd *uhci) { dma_addr_t dma_handle; struct uhci_td *td; @@ -56,22 +48,27 @@ static struct uhci_td *uhci_alloc_td(struct uhci_hcd *uhci, struct usb_device *d return NULL; td->dma_handle = dma_handle; - - td->link = UHCI_PTR_TERM; - td->buffer = 0; - td->frame = -1; - td->dev = dev; INIT_LIST_HEAD(&td->list); INIT_LIST_HEAD(&td->remove_list); INIT_LIST_HEAD(&td->fl_list); - usb_get_dev(dev); - return td; } +static void uhci_free_td(struct uhci_hcd *uhci, struct uhci_td *td) +{ + if (!list_empty(&td->list)) + dev_warn(uhci_dev(uhci), "td %p still in list!\n", td); + if (!list_empty(&td->remove_list)) + dev_warn(uhci_dev(uhci), "td %p still in remove_list!\n", td); + if (!list_empty(&td->fl_list)) + dev_warn(uhci_dev(uhci), "td %p still in fl_list!\n", td); + + dma_pool_free(uhci->td_pool, td, td->dma_handle); +} + static inline void uhci_fill_td(struct uhci_td *td, u32 status, u32 token, u32 buffer) { @@ -81,19 +78,20 @@ static inline void uhci_fill_td(struct uhci_td *td, u32 status, } /* - * We insert Isochronous URB's directly into the frame list at the beginning + * We insert Isochronous URBs directly into the frame list at the beginning */ -static void uhci_insert_td_frame_list(struct uhci_hcd *uhci, struct uhci_td *td, unsigned framenum) +static inline void uhci_insert_td_in_frame_list(struct uhci_hcd *uhci, + struct uhci_td *td, unsigned framenum) { framenum &= (UHCI_NUMFRAMES - 1); td->frame = framenum; /* Is there a TD already mapped there? */ - if (uhci->fl->frame_cpu[framenum]) { + if (uhci->frame_cpu[framenum]) { struct uhci_td *ftd, *ltd; - ftd = uhci->fl->frame_cpu[framenum]; + ftd = uhci->frame_cpu[framenum]; ltd = list_entry(ftd->fl_list.prev, struct uhci_td, fl_list); list_add_tail(&td->fl_list, &ftd->fl_list); @@ -102,29 +100,32 @@ static void uhci_insert_td_frame_list(struct uhci_hcd *uhci, struct uhci_td *td, wmb(); ltd->link = cpu_to_le32(td->dma_handle); } else { - td->link = uhci->fl->frame[framenum]; + td->link = uhci->frame[framenum]; wmb(); - uhci->fl->frame[framenum] = cpu_to_le32(td->dma_handle); - uhci->fl->frame_cpu[framenum] = td; + uhci->frame[framenum] = cpu_to_le32(td->dma_handle); + uhci->frame_cpu[framenum] = td; } } -static void uhci_remove_td(struct uhci_hcd *uhci, struct uhci_td *td) +static inline void uhci_remove_td_from_frame_list(struct uhci_hcd *uhci, + struct uhci_td *td) { /* If it's not inserted, don't remove it */ - if (td->frame == -1 && list_empty(&td->fl_list)) + if (td->frame == -1) { + WARN_ON(!list_empty(&td->fl_list)); return; + } - if (td->frame != -1 && uhci->fl->frame_cpu[td->frame] == td) { + if (uhci->frame_cpu[td->frame] == td) { if (list_empty(&td->fl_list)) { - uhci->fl->frame[td->frame] = td->link; - uhci->fl->frame_cpu[td->frame] = NULL; + uhci->frame[td->frame] = td->link; + uhci->frame_cpu[td->frame] = NULL; } else { struct uhci_td *ntd; ntd = list_entry(td->fl_list.next, struct uhci_td, fl_list); - uhci->fl->frame[td->frame] = cpu_to_le32(ntd->dma_handle); - uhci->fl->frame_cpu[td->frame] = ntd; + uhci->frame[td->frame] = cpu_to_le32(ntd->dma_handle); + uhci->frame_cpu[td->frame] = ntd; } } else { struct uhci_td *ptd; @@ -133,48 +134,25 @@ static void uhci_remove_td(struct uhci_hcd *uhci, struct uhci_td *td) ptd->link = td->link; } - wmb(); - td->link = UHCI_PTR_TERM; - list_del_init(&td->fl_list); td->frame = -1; } /* - * Inserts a td list into qh. + * Remove all the TDs for an Isochronous URB from the frame list */ -static void uhci_insert_tds_in_qh(struct uhci_qh *qh, struct urb *urb, __le32 breadth) +static void uhci_unlink_isochronous_tds(struct uhci_hcd *uhci, struct urb *urb) { - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; + struct urb_priv *urbp = (struct urb_priv *) urb->hcpriv; struct uhci_td *td; - __le32 *plink; - /* Ordering isn't important here yet since the QH hasn't been */ - /* inserted into the schedule yet */ - plink = &qh->element; - list_for_each_entry(td, &urbp->td_list, list) { - *plink = cpu_to_le32(td->dma_handle) | breadth; - plink = &td->link; - } - *plink = UHCI_PTR_TERM; -} - -static void uhci_free_td(struct uhci_hcd *uhci, struct uhci_td *td) -{ - if (!list_empty(&td->list)) - dev_warn(uhci_dev(uhci), "td %p still in list!\n", td); - if (!list_empty(&td->remove_list)) - dev_warn(uhci_dev(uhci), "td %p still in remove_list!\n", td); - if (!list_empty(&td->fl_list)) - dev_warn(uhci_dev(uhci), "td %p still in fl_list!\n", td); - - if (td->dev) - usb_put_dev(td->dev); - - dma_pool_free(uhci->td_pool, td, td->dma_handle); + list_for_each_entry(td, &urbp->td_list, list) + uhci_remove_td_from_frame_list(uhci, td); + wmb(); } -static struct uhci_qh *uhci_alloc_qh(struct uhci_hcd *uhci, struct usb_device *dev) +static struct uhci_qh *uhci_alloc_qh(struct uhci_hcd *uhci, + struct usb_device *udev, struct usb_host_endpoint *hep) { dma_addr_t dma_handle; struct uhci_qh *qh; @@ -188,262 +166,217 @@ static struct uhci_qh *uhci_alloc_qh(struct uhci_hcd *uhci, struct usb_device *d qh->element = UHCI_PTR_TERM; qh->link = UHCI_PTR_TERM; - qh->dev = dev; - qh->urbp = NULL; - - INIT_LIST_HEAD(&qh->list); - INIT_LIST_HEAD(&qh->remove_list); - - usb_get_dev(dev); + INIT_LIST_HEAD(&qh->queue); + INIT_LIST_HEAD(&qh->node); + if (udev) { /* Normal QH */ + qh->dummy_td = uhci_alloc_td(uhci); + if (!qh->dummy_td) { + dma_pool_free(uhci->qh_pool, qh, dma_handle); + return NULL; + } + qh->state = QH_STATE_IDLE; + qh->hep = hep; + qh->udev = udev; + hep->hcpriv = qh; + + } else { /* Skeleton QH */ + qh->state = QH_STATE_ACTIVE; + qh->udev = NULL; + } return qh; } static void uhci_free_qh(struct uhci_hcd *uhci, struct uhci_qh *qh) { - if (!list_empty(&qh->list)) + WARN_ON(qh->state != QH_STATE_IDLE && qh->udev); + if (!list_empty(&qh->queue)) dev_warn(uhci_dev(uhci), "qh %p list not empty!\n", qh); - if (!list_empty(&qh->remove_list)) - dev_warn(uhci_dev(uhci), "qh %p still in remove_list!\n", qh); - - if (qh->dev) - usb_put_dev(qh->dev); + list_del(&qh->node); + if (qh->udev) { + qh->hep->hcpriv = NULL; + uhci_free_td(uhci, qh->dummy_td); + } dma_pool_free(uhci->qh_pool, qh, qh->dma_handle); } /* - * Append this urb's qh after the last qh in skelqh->list - * - * Note that urb_priv.queue_list doesn't have a separate queue head; - * it's a ring with every element "live". + * When the currently executing URB is dequeued, save its current toggle value */ -static void uhci_insert_qh(struct uhci_hcd *uhci, struct uhci_qh *skelqh, struct urb *urb) +static void uhci_save_toggle(struct uhci_qh *qh, struct urb *urb) { - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; - struct urb_priv *turbp; - struct uhci_qh *lqh; + struct urb_priv *urbp = (struct urb_priv *) urb->hcpriv; + struct uhci_td *td; - /* Grab the last QH */ - lqh = list_entry(skelqh->list.prev, struct uhci_qh, list); + /* If the QH element pointer is UHCI_PTR_TERM then then currently + * executing URB has already been unlinked, so this one isn't it. */ + if (qh_element(qh) == UHCI_PTR_TERM || + qh->queue.next != &urbp->node) + return; + qh->element = UHCI_PTR_TERM; - /* Point to the next skelqh */ - urbp->qh->link = lqh->link; - wmb(); /* Ordering is important */ + /* Only bulk and interrupt pipes have to worry about toggles */ + if (!(usb_pipetype(urb->pipe) == PIPE_BULK || + usb_pipetype(urb->pipe) == PIPE_INTERRUPT)) + return; - /* - * Patch QHs for previous endpoint's queued URBs? HC goes - * here next, not to the next skelqh it now points to. - * - * lqh --> td ... --> qh ... --> td --> qh ... --> td - * | | | - * v v v - * +<----------------+-----------------+ - * v - * newqh --> td ... --> td - * | - * v - * ... - * - * The HC could see (and use!) any of these as we write them. - */ - lqh->link = cpu_to_le32(urbp->qh->dma_handle) | UHCI_PTR_QH; - if (lqh->urbp) { - list_for_each_entry(turbp, &lqh->urbp->queue_list, queue_list) - turbp->qh->link = lqh->link; + /* Find the first active TD; that's the device's toggle state */ + list_for_each_entry(td, &urbp->td_list, list) { + if (td_status(td) & TD_CTRL_ACTIVE) { + qh->needs_fixup = 1; + qh->initial_toggle = uhci_toggle(td_token(td)); + return; + } } - list_add_tail(&urbp->qh->list, &skelqh->list); + WARN_ON(1); } /* - * Start removal of QH from schedule; it finishes next frame. - * TDs should be unlinked before this is called. + * Fix up the data toggles for URBs in a queue, when one of them + * terminates early (short transfer, error, or dequeued). */ -static void uhci_remove_qh(struct uhci_hcd *uhci, struct uhci_qh *qh) +static void uhci_fixup_toggles(struct uhci_qh *qh, int skip_first) { - struct uhci_qh *pqh; - __le32 newlink; - - if (!qh) - return; - - /* - * Only go through the hoops if it's actually linked in - */ - if (!list_empty(&qh->list)) { - - /* If our queue is nonempty, make the next URB the head */ - if (!list_empty(&qh->urbp->queue_list)) { - struct urb_priv *nurbp; - - nurbp = list_entry(qh->urbp->queue_list.next, - struct urb_priv, queue_list); - nurbp->queued = 0; - list_add(&nurbp->qh->list, &qh->list); - newlink = cpu_to_le32(nurbp->qh->dma_handle) | UHCI_PTR_QH; - } else - newlink = qh->link; - - /* Fix up the previous QH's queue to link to either - * the new head of this queue or the start of the - * next endpoint's queue. */ - pqh = list_entry(qh->list.prev, struct uhci_qh, list); - pqh->link = newlink; - if (pqh->urbp) { - struct urb_priv *turbp; - - list_for_each_entry(turbp, &pqh->urbp->queue_list, - queue_list) - turbp->qh->link = newlink; + struct urb_priv *urbp = NULL; + struct uhci_td *td; + unsigned int toggle = qh->initial_toggle; + unsigned int pipe; + + /* Fixups for a short transfer start with the second URB in the + * queue (the short URB is the first). */ + if (skip_first) + urbp = list_entry(qh->queue.next, struct urb_priv, node); + + /* When starting with the first URB, if the QH element pointer is + * still valid then we know the URB's toggles are okay. */ + else if (qh_element(qh) != UHCI_PTR_TERM) + toggle = 2; + + /* Fix up the toggle for the URBs in the queue. Normally this + * loop won't run more than once: When an error or short transfer + * occurs, the queue usually gets emptied. */ + urbp = list_prepare_entry(urbp, &qh->queue, node); + list_for_each_entry_continue(urbp, &qh->queue, node) { + + /* If the first TD has the right toggle value, we don't + * need to change any toggles in this URB */ + td = list_entry(urbp->td_list.next, struct uhci_td, list); + if (toggle > 1 || uhci_toggle(td_token(td)) == toggle) { + td = list_entry(urbp->td_list.prev, struct uhci_td, + list); + toggle = uhci_toggle(td_token(td)) ^ 1; + + /* Otherwise all the toggles in the URB have to be switched */ + } else { + list_for_each_entry(td, &urbp->td_list, list) { + td->token ^= __constant_cpu_to_le32( + TD_TOKEN_TOGGLE); + toggle ^= 1; + } } - wmb(); - - /* Leave qh->link in case the HC is on the QH now, it will */ - /* continue the rest of the schedule */ - qh->element = UHCI_PTR_TERM; - - list_del_init(&qh->list); - } - - list_del_init(&qh->urbp->queue_list); - qh->urbp = NULL; - - uhci_get_current_frame_number(uhci); - if (uhci->frame_number + uhci->is_stopped != uhci->qh_remove_age) { - uhci_free_pending_qhs(uhci); - uhci->qh_remove_age = uhci->frame_number; } - /* Check to see if the remove list is empty. Set the IOC bit */ - /* to force an interrupt so we can remove the QH */ - if (list_empty(&uhci->qh_remove_list)) - uhci_set_next_interrupt(uhci); - - list_add(&qh->remove_list, &uhci->qh_remove_list); + wmb(); + pipe = list_entry(qh->queue.next, struct urb_priv, node)->urb->pipe; + usb_settoggle(qh->udev, usb_pipeendpoint(pipe), + usb_pipeout(pipe), toggle); + qh->needs_fixup = 0; } -static int uhci_fixup_toggle(struct urb *urb, unsigned int toggle) +/* + * Put a QH on the schedule in both hardware and software + */ +static void uhci_activate_qh(struct uhci_hcd *uhci, struct uhci_qh *qh) { - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; - struct uhci_td *td; - - list_for_each_entry(td, &urbp->td_list, list) { - if (toggle) - td->token |= cpu_to_le32(TD_TOKEN_TOGGLE); - else - td->token &= ~cpu_to_le32(TD_TOKEN_TOGGLE); - - toggle ^= 1; - } - - return toggle; -} + struct uhci_qh *pqh; -/* This function will append one URB's QH to another URB's QH. This is for */ -/* queuing interrupt, control or bulk transfers */ -static void uhci_append_queued_urb(struct uhci_hcd *uhci, struct urb *eurb, struct urb *urb) -{ - struct urb_priv *eurbp, *urbp, *furbp, *lurbp; - struct uhci_td *lltd; + WARN_ON(list_empty(&qh->queue)); - eurbp = eurb->hcpriv; - urbp = urb->hcpriv; + /* Set the element pointer if it isn't set already. + * This isn't needed for Isochronous queues, but it doesn't hurt. */ + if (qh_element(qh) == UHCI_PTR_TERM) { + struct urb_priv *urbp = list_entry(qh->queue.next, + struct urb_priv, node); + struct uhci_td *td = list_entry(urbp->td_list.next, + struct uhci_td, list); - /* Find the first URB in the queue */ - furbp = eurbp; - if (eurbp->queued) { - list_for_each_entry(furbp, &eurbp->queue_list, queue_list) - if (!furbp->queued) - break; + qh->element = cpu_to_le32(td->dma_handle); } - lurbp = list_entry(furbp->queue_list.prev, struct urb_priv, queue_list); - - lltd = list_entry(lurbp->td_list.prev, struct uhci_td, list); - - /* Control transfers always start with toggle 0 */ - if (!usb_pipecontrol(urb->pipe)) - usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe), - uhci_fixup_toggle(urb, - uhci_toggle(td_token(lltd)) ^ 1)); - - /* All qh's in the queue need to link to the next queue */ - urbp->qh->link = eurbp->qh->link; - - wmb(); /* Make sure we flush everything */ - - lltd->link = cpu_to_le32(urbp->qh->dma_handle) | UHCI_PTR_QH; - - list_add_tail(&urbp->queue_list, &furbp->queue_list); - - urbp->queued = 1; + if (qh->state == QH_STATE_ACTIVE) + return; + qh->state = QH_STATE_ACTIVE; + + /* Move the QH from its old list to the end of the appropriate + * skeleton's list */ + if (qh == uhci->next_qh) + uhci->next_qh = list_entry(qh->node.next, struct uhci_qh, + node); + list_move_tail(&qh->node, &qh->skel->node); + + /* Link it into the schedule */ + pqh = list_entry(qh->node.prev, struct uhci_qh, node); + qh->link = pqh->link; + wmb(); + pqh->link = UHCI_PTR_QH | cpu_to_le32(qh->dma_handle); } -static void uhci_delete_queued_urb(struct uhci_hcd *uhci, struct urb *urb) +/* + * Take a QH off the hardware schedule + */ +static void uhci_unlink_qh(struct uhci_hcd *uhci, struct uhci_qh *qh) { - struct urb_priv *urbp, *nurbp, *purbp, *turbp; - struct uhci_td *pltd; - unsigned int toggle; - - urbp = urb->hcpriv; + struct uhci_qh *pqh; - if (list_empty(&urbp->queue_list)) + if (qh->state == QH_STATE_UNLINKING) return; + WARN_ON(qh->state != QH_STATE_ACTIVE || !qh->udev); + qh->state = QH_STATE_UNLINKING; - nurbp = list_entry(urbp->queue_list.next, struct urb_priv, queue_list); + /* Unlink the QH from the schedule and record when we did it */ + pqh = list_entry(qh->node.prev, struct uhci_qh, node); + pqh->link = qh->link; + mb(); - /* - * Fix up the toggle for the following URBs in the queue. - * Only needed for bulk and interrupt: control and isochronous - * endpoints don't propagate toggles between messages. - */ - if (usb_pipebulk(urb->pipe) || usb_pipeint(urb->pipe)) { - if (!urbp->queued) - /* We just set the toggle in uhci_unlink_generic */ - toggle = usb_gettoggle(urb->dev, - usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe)); - else { - /* If we're in the middle of the queue, grab the */ - /* toggle from the TD previous to us */ - purbp = list_entry(urbp->queue_list.prev, - struct urb_priv, queue_list); - pltd = list_entry(purbp->td_list.prev, - struct uhci_td, list); - toggle = uhci_toggle(td_token(pltd)) ^ 1; - } + uhci_get_current_frame_number(uhci); + qh->unlink_frame = uhci->frame_number; - list_for_each_entry(turbp, &urbp->queue_list, queue_list) { - if (!turbp->queued) - break; - toggle = uhci_fixup_toggle(turbp->urb, toggle); - } + /* Force an interrupt so we know when the QH is fully unlinked */ + if (list_empty(&uhci->skel_unlink_qh->node)) + uhci_set_next_interrupt(uhci); - usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe), toggle); - } + /* Move the QH from its old list to the end of the unlinking list */ + if (qh == uhci->next_qh) + uhci->next_qh = list_entry(qh->node.next, struct uhci_qh, + node); + list_move_tail(&qh->node, &uhci->skel_unlink_qh->node); +} - if (urbp->queued) { - /* We're somewhere in the middle (or end). The case where - * we're at the head is handled in uhci_remove_qh(). */ - purbp = list_entry(urbp->queue_list.prev, struct urb_priv, - queue_list); +/* + * When we and the controller are through with a QH, it becomes IDLE. + * This happens when a QH has been off the schedule (on the unlinking + * list) for more than one frame, or when an error occurs while adding + * the first URB onto a new QH. + */ +static void uhci_make_qh_idle(struct uhci_hcd *uhci, struct uhci_qh *qh) +{ + WARN_ON(qh->state == QH_STATE_ACTIVE); - pltd = list_entry(purbp->td_list.prev, struct uhci_td, list); - if (nurbp->queued) - pltd->link = cpu_to_le32(nurbp->qh->dma_handle) | UHCI_PTR_QH; - else - /* The next URB happens to be the beginning, so */ - /* we're the last, end the chain */ - pltd->link = UHCI_PTR_TERM; - } + if (qh == uhci->next_qh) + uhci->next_qh = list_entry(qh->node.next, struct uhci_qh, + node); + list_move(&qh->node, &uhci->idle_qh_list); + qh->state = QH_STATE_IDLE; - /* urbp->queue_list is handled in uhci_remove_qh() */ + /* If anyone is waiting for a QH to become idle, wake them up */ + if (uhci->num_waiting) + wake_up_all(&uhci->waitqh); } -static struct urb_priv *uhci_alloc_urb_priv(struct uhci_hcd *uhci, struct urb *urb) +static inline struct urb_priv *uhci_alloc_urb_priv(struct uhci_hcd *uhci, + struct urb *urb) { struct urb_priv *urbp; @@ -453,17 +386,11 @@ static struct urb_priv *uhci_alloc_urb_priv(struct uhci_hcd *uhci, struct urb *u memset((void *)urbp, 0, sizeof(*urbp)); - urbp->inserttime = jiffies; - urbp->fsbrtime = jiffies; urbp->urb = urb; + urb->hcpriv = urbp; + INIT_LIST_HEAD(&urbp->node); INIT_LIST_HEAD(&urbp->td_list); - INIT_LIST_HEAD(&urbp->queue_list); - INIT_LIST_HEAD(&urbp->urb_list); - - list_add_tail(&urbp->urb_list, &uhci->urb_list); - - urb->hcpriv = urbp; return urbp; } @@ -472,8 +399,6 @@ static void uhci_add_td_to_urb(struct urb *urb, struct uhci_td *td) { struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; - td->urb = urb; - list_add_tail(&td->list, &urbp->td_list); } @@ -483,22 +408,16 @@ static void uhci_remove_td_from_urb(struct uhci_td *td) return; list_del_init(&td->list); - - td->urb = NULL; } -static void uhci_destroy_urb_priv(struct uhci_hcd *uhci, struct urb *urb) +static void uhci_free_urb_priv(struct uhci_hcd *uhci, + struct urb_priv *urbp) { struct uhci_td *td, *tmp; - struct urb_priv *urbp; - - urbp = (struct urb_priv *)urb->hcpriv; - if (!urbp) - return; - if (!list_empty(&urbp->urb_list)) - dev_warn(uhci_dev(uhci), "urb %p still on uhci->urb_list " - "or uhci->remove_list!\n", urb); + if (!list_empty(&urbp->node)) + dev_warn(uhci_dev(uhci), "urb %p still on QH's list!\n", + urbp->urb); uhci_get_current_frame_number(uhci); if (uhci->frame_number + uhci->is_stopped != uhci->td_remove_age) { @@ -507,17 +426,16 @@ static void uhci_destroy_urb_priv(struct uhci_hcd *uhci, struct urb *urb) } /* Check to see if the remove list is empty. Set the IOC bit */ - /* to force an interrupt so we can remove the TD's*/ + /* to force an interrupt so we can remove the TDs. */ if (list_empty(&uhci->td_remove_list)) uhci_set_next_interrupt(uhci); list_for_each_entry_safe(td, tmp, &urbp->td_list, list) { uhci_remove_td_from_urb(td); - uhci_remove_td(uhci, td); list_add(&td->remove_list, &uhci->td_remove_list); } - urb->hcpriv = NULL; + urbp->urb->hcpriv = NULL; kmem_cache_free(uhci_up_cachep, urbp); } @@ -576,34 +494,33 @@ static int uhci_map_status(int status, int dir_out) /* * Control transfers */ -static int uhci_submit_control(struct uhci_hcd *uhci, struct urb *urb, struct urb *eurb) +static int uhci_submit_control(struct uhci_hcd *uhci, struct urb *urb, + struct uhci_qh *qh) { - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; struct uhci_td *td; - struct uhci_qh *qh, *skelqh; unsigned long destination, status; - int maxsze = usb_maxpacket(urb->dev, urb->pipe, usb_pipeout(urb->pipe)); + int maxsze = le16_to_cpu(qh->hep->desc.wMaxPacketSize); int len = urb->transfer_buffer_length; dma_addr_t data = urb->transfer_dma; + __le32 *plink; /* The "pipe" thing contains the destination in bits 8--18 */ destination = (urb->pipe & PIPE_DEVEP_MASK) | USB_PID_SETUP; - /* 3 errors */ - status = TD_CTRL_ACTIVE | uhci_maxerr(3); + /* 3 errors, dummy TD remains inactive */ + status = uhci_maxerr(3); if (urb->dev->speed == USB_SPEED_LOW) status |= TD_CTRL_LS; /* * Build the TD for the control request setup packet */ - td = uhci_alloc_td(uhci, urb->dev); - if (!td) - return -ENOMEM; - + td = qh->dummy_td; uhci_add_td_to_urb(urb, td); - uhci_fill_td(td, status, destination | uhci_explen(7), - urb->setup_dma); + uhci_fill_td(td, status, destination | uhci_explen(8), + urb->setup_dma); + plink = &td->link; + status |= TD_CTRL_ACTIVE; /* * If direction is "send", change the packet ID from SETUP (0x2D) @@ -618,24 +535,23 @@ static int uhci_submit_control(struct uhci_hcd *uhci, struct urb *urb, struct ur } /* - * Build the DATA TD's + * Build the DATA TDs */ while (len > 0) { - int pktsze = len; - - if (pktsze > maxsze) - pktsze = maxsze; + int pktsze = min(len, maxsze); - td = uhci_alloc_td(uhci, urb->dev); + td = uhci_alloc_td(uhci); if (!td) - return -ENOMEM; + goto nomem; + *plink = cpu_to_le32(td->dma_handle); /* Alternate Data0/1 (start with Data1) */ destination ^= TD_TOKEN_TOGGLE; uhci_add_td_to_urb(urb, td); - uhci_fill_td(td, status, destination | uhci_explen(pktsze - 1), - data); + uhci_fill_td(td, status, destination | uhci_explen(pktsze), + data); + plink = &td->link; data += pktsze; len -= pktsze; @@ -644,9 +560,10 @@ static int uhci_submit_control(struct uhci_hcd *uhci, struct urb *urb, struct ur /* * Build the final TD for control status */ - td = uhci_alloc_td(uhci, urb->dev); + td = uhci_alloc_td(uhci); if (!td) - return -ENOMEM; + goto nomem; + *plink = cpu_to_le32(td->dma_handle); /* * It's IN if the pipe is an output pipe or we're not expecting @@ -664,35 +581,39 @@ static int uhci_submit_control(struct uhci_hcd *uhci, struct urb *urb, struct ur uhci_add_td_to_urb(urb, td); uhci_fill_td(td, status | TD_CTRL_IOC, - destination | uhci_explen(UHCI_NULL_DATA_SIZE), 0); - - qh = uhci_alloc_qh(uhci, urb->dev); - if (!qh) - return -ENOMEM; + destination | uhci_explen(0), 0); + plink = &td->link; - urbp->qh = qh; - qh->urbp = urbp; + /* + * Build the new dummy TD and activate the old one + */ + td = uhci_alloc_td(uhci); + if (!td) + goto nomem; + *plink = cpu_to_le32(td->dma_handle); - uhci_insert_tds_in_qh(qh, urb, UHCI_PTR_BREADTH); + uhci_fill_td(td, 0, USB_PID_OUT | uhci_explen(0), 0); + wmb(); + qh->dummy_td->status |= __constant_cpu_to_le32(TD_CTRL_ACTIVE); + qh->dummy_td = td; /* Low-speed transfers get a different queue, and won't hog the bus. * Also, some devices enumerate better without FSBR; the easiest way * to do that is to put URBs on the low-speed queue while the device - * is in the DEFAULT state. */ + * isn't in the CONFIGURED state. */ if (urb->dev->speed == USB_SPEED_LOW || - urb->dev->state == USB_STATE_DEFAULT) - skelqh = uhci->skel_ls_control_qh; + urb->dev->state != USB_STATE_CONFIGURED) + qh->skel = uhci->skel_ls_control_qh; else { - skelqh = uhci->skel_fs_control_qh; + qh->skel = uhci->skel_fs_control_qh; uhci_inc_fsbr(uhci, urb); } + return 0; - if (eurb) - uhci_append_queued_urb(uhci, eurb, urb); - else - uhci_insert_qh(uhci, skelqh, urb); - - return -EINPROGRESS; +nomem: + /* Remove the dummy TD from the td_list so it doesn't get freed */ + uhci_remove_td_from_urb(qh->dummy_td); + return -ENOMEM; } /* @@ -709,7 +630,7 @@ static int usb_control_retrigger_status(struct uhci_hcd *uhci, struct urb *urb) struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; struct uhci_td *td; - urbp->short_control_packet = 1; + urbp->short_transfer = 1; td = list_entry(urbp->td_list.prev, struct uhci_td, list); urbp->qh->element = cpu_to_le32(td->dma_handle); @@ -726,16 +647,14 @@ static int uhci_result_control(struct uhci_hcd *uhci, struct urb *urb) unsigned int status; int ret = 0; - if (list_empty(&urbp->td_list)) - return -EINVAL; - head = &urbp->td_list; - - if (urbp->short_control_packet) { + if (urbp->short_transfer) { tmp = head->prev; goto status_stage; } + urb->actual_length = 0; + tmp = head->next; td = list_entry(tmp, struct uhci_td, list); @@ -748,9 +667,7 @@ static int uhci_result_control(struct uhci_hcd *uhci, struct urb *urb) if (status) goto td_error; - urb->actual_length = 0; - - /* The rest of the TD's (but the last) are data */ + /* The rest of the TDs (but the last) are data */ tmp = tmp->next; while (tmp != head && tmp->next != head) { unsigned int ctrlstat; @@ -776,10 +693,7 @@ static int uhci_result_control(struct uhci_hcd *uhci, struct urb *urb) goto err; } - if (uhci_packetid(td_token(td)) == USB_PID_IN) - return usb_control_retrigger_status(uhci, urb); - else - return 0; + return usb_control_retrigger_status(uhci, urb); } } @@ -820,66 +734,76 @@ err: if (errbuf) { /* Print the chain for debugging purposes */ uhci_show_qh(urbp->qh, errbuf, ERRBUF_LEN, 0); - lprintk(errbuf); } } + /* Note that the queue has stopped */ + urbp->qh->element = UHCI_PTR_TERM; + urbp->qh->is_stopped = 1; return ret; } /* * Common submit for bulk and interrupt */ -static int uhci_submit_common(struct uhci_hcd *uhci, struct urb *urb, struct urb *eurb, struct uhci_qh *skelqh) +static int uhci_submit_common(struct uhci_hcd *uhci, struct urb *urb, + struct uhci_qh *qh) { struct uhci_td *td; - struct uhci_qh *qh; unsigned long destination, status; - int maxsze = usb_maxpacket(urb->dev, urb->pipe, usb_pipeout(urb->pipe)); + int maxsze = le16_to_cpu(qh->hep->desc.wMaxPacketSize); int len = urb->transfer_buffer_length; - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; dma_addr_t data = urb->transfer_dma; + __le32 *plink; + unsigned int toggle; if (len < 0) return -EINVAL; /* The "pipe" thing contains the destination in bits 8--18 */ destination = (urb->pipe & PIPE_DEVEP_MASK) | usb_packetid(urb->pipe); + toggle = usb_gettoggle(urb->dev, usb_pipeendpoint(urb->pipe), + usb_pipeout(urb->pipe)); - status = uhci_maxerr(3) | TD_CTRL_ACTIVE; + /* 3 errors, dummy TD remains inactive */ + status = uhci_maxerr(3); if (urb->dev->speed == USB_SPEED_LOW) status |= TD_CTRL_LS; if (usb_pipein(urb->pipe)) status |= TD_CTRL_SPD; /* - * Build the DATA TD's + * Build the DATA TDs */ + plink = NULL; + td = qh->dummy_td; do { /* Allow zero length packets */ int pktsze = maxsze; - if (pktsze >= len) { + if (len <= pktsze) { /* The last packet */ pktsze = len; if (!(urb->transfer_flags & URB_SHORT_NOT_OK)) status &= ~TD_CTRL_SPD; } - td = uhci_alloc_td(uhci, urb->dev); - if (!td) - return -ENOMEM; - + if (plink) { + td = uhci_alloc_td(uhci); + if (!td) + goto nomem; + *plink = cpu_to_le32(td->dma_handle); + } uhci_add_td_to_urb(urb, td); - uhci_fill_td(td, status, destination | uhci_explen(pktsze - 1) | - (usb_gettoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe)) << TD_TOKEN_TOGGLE_SHIFT), - data); + uhci_fill_td(td, status, + destination | uhci_explen(pktsze) | + (toggle << TD_TOKEN_TOGGLE_SHIFT), + data); + plink = &td->link; + status |= TD_CTRL_ACTIVE; data += pktsze; len -= maxsze; - - usb_dotoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe)); + toggle ^= 1; } while (len > 0); /* @@ -889,20 +813,22 @@ static int uhci_submit_common(struct uhci_hcd *uhci, struct urb *urb, struct urb * however, if transfer_length == 0, the zero packet was already * prepared above. */ - if (usb_pipeout(urb->pipe) && (urb->transfer_flags & URB_ZERO_PACKET) && - !len && urb->transfer_buffer_length) { - td = uhci_alloc_td(uhci, urb->dev); + if ((urb->transfer_flags & URB_ZERO_PACKET) && + usb_pipeout(urb->pipe) && len == 0 && + urb->transfer_buffer_length > 0) { + td = uhci_alloc_td(uhci); if (!td) - return -ENOMEM; + goto nomem; + *plink = cpu_to_le32(td->dma_handle); uhci_add_td_to_urb(urb, td); - uhci_fill_td(td, status, destination | uhci_explen(UHCI_NULL_DATA_SIZE) | - (usb_gettoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe)) << TD_TOKEN_TOGGLE_SHIFT), - data); + uhci_fill_td(td, status, + destination | uhci_explen(0) | + (toggle << TD_TOKEN_TOGGLE_SHIFT), + data); + plink = &td->link; - usb_dotoggle(urb->dev, usb_pipeendpoint(urb->pipe), - usb_pipeout(urb->pipe)); + toggle ^= 1; } /* Set the interrupt-on-completion flag on the last packet. @@ -911,24 +837,29 @@ static int uhci_submit_common(struct uhci_hcd *uhci, struct urb *urb, struct urb * fast side but not enough to justify delaying an interrupt * more than 2 or 3 URBs, so we will ignore the URB_NO_INTERRUPT * flag setting. */ - td->status |= cpu_to_le32(TD_CTRL_IOC); - - qh = uhci_alloc_qh(uhci, urb->dev); - if (!qh) - return -ENOMEM; + td->status |= __constant_cpu_to_le32(TD_CTRL_IOC); - urbp->qh = qh; - qh->urbp = urbp; + /* + * Build the new dummy TD and activate the old one + */ + td = uhci_alloc_td(uhci); + if (!td) + goto nomem; + *plink = cpu_to_le32(td->dma_handle); - /* Always breadth first */ - uhci_insert_tds_in_qh(qh, urb, UHCI_PTR_BREADTH); + uhci_fill_td(td, 0, USB_PID_OUT | uhci_explen(0), 0); + wmb(); + qh->dummy_td->status |= __constant_cpu_to_le32(TD_CTRL_ACTIVE); + qh->dummy_td = td; - if (eurb) - uhci_append_queued_urb(uhci, eurb, urb); - else - uhci_insert_qh(uhci, skelqh, urb); + usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), + usb_pipeout(urb->pipe), toggle); + return 0; - return -EINPROGRESS; +nomem: + /* Remove the dummy TD from the td_list so it doesn't get freed */ + uhci_remove_td_from_urb(qh->dummy_td); + return -ENOMEM; } /* @@ -960,8 +891,29 @@ static int uhci_result_common(struct uhci_hcd *uhci, struct urb *urb) if (urb->transfer_flags & URB_SHORT_NOT_OK) { ret = -EREMOTEIO; goto err; - } else - return 0; + } + + /* + * This URB stopped short of its end. We have to + * fix up the toggles of the following URBs on the + * queue and restart the queue. But only if this + * TD isn't the last one in the URB. + * + * Do this only the first time we encounter the + * short URB. + */ + if (!urbp->short_transfer && + &td->list != urbp->td_list.prev) { + urbp->short_transfer = 1; + urbp->qh->initial_toggle = + uhci_toggle(td_token(td)) ^ 1; + uhci_fixup_toggles(urbp->qh, 1); + + td = list_entry(urbp->td_list.prev, + struct uhci_td, list); + urbp->qh->element = td->link; + } + break; } } @@ -970,31 +922,30 @@ static int uhci_result_common(struct uhci_hcd *uhci, struct urb *urb) td_error: ret = uhci_map_status(status, uhci_packetout(td_token(td))); -err: - /* - * Enable this chunk of code if you want to see some more debugging. - * But be careful, it has the tendancy to starve out khubd and prevent - * disconnects from happening successfully if you have a slow debug - * log interface (like a serial console. - */ -#if 0 if ((debug == 1 && ret != -EPIPE) || debug > 1) { /* Some debugging code */ dev_dbg(uhci_dev(uhci), "%s: failed with status %x\n", __FUNCTION__, status); - if (errbuf) { + if (debug > 1 && errbuf) { /* Print the chain for debugging purposes */ uhci_show_qh(urbp->qh, errbuf, ERRBUF_LEN, 0); - lprintk(errbuf); } } -#endif +err: + + /* Note that the queue has stopped and save the next toggle value */ + urbp->qh->element = UHCI_PTR_TERM; + urbp->qh->is_stopped = 1; + urbp->qh->needs_fixup = 1; + urbp->qh->initial_toggle = uhci_toggle(td_token(td)) ^ + (ret == -EREMOTEIO); return ret; } -static inline int uhci_submit_bulk(struct uhci_hcd *uhci, struct urb *urb, struct urb *eurb) +static inline int uhci_submit_bulk(struct uhci_hcd *uhci, struct urb *urb, + struct uhci_qh *qh) { int ret; @@ -1002,115 +953,86 @@ static inline int uhci_submit_bulk(struct uhci_hcd *uhci, struct urb *urb, struc if (urb->dev->speed == USB_SPEED_LOW) return -EINVAL; - ret = uhci_submit_common(uhci, urb, eurb, uhci->skel_bulk_qh); - if (ret == -EINPROGRESS) + qh->skel = uhci->skel_bulk_qh; + ret = uhci_submit_common(uhci, urb, qh); + if (ret == 0) uhci_inc_fsbr(uhci, urb); - return ret; } -static inline int uhci_submit_interrupt(struct uhci_hcd *uhci, struct urb *urb, struct urb *eurb) +static inline int uhci_submit_interrupt(struct uhci_hcd *uhci, struct urb *urb, + struct uhci_qh *qh) { - /* USB 1.1 interrupt transfers only involve one packet per interval; - * that's the uhci_submit_common() "breadth first" policy. Drivers - * can submit urbs of any length, but longer ones might need many - * intervals to complete. + /* USB 1.1 interrupt transfers only involve one packet per interval. + * Drivers can submit URBs of any length, but longer ones will need + * multiple intervals to complete. */ - return uhci_submit_common(uhci, urb, eurb, uhci->skelqh[__interval_to_skel(urb->interval)]); + qh->skel = uhci->skelqh[__interval_to_skel(urb->interval)]; + return uhci_submit_common(uhci, urb, qh); } /* * Isochronous transfers */ -static int isochronous_find_limits(struct uhci_hcd *uhci, struct urb *urb, unsigned int *start, unsigned int *end) +static int uhci_submit_isochronous(struct uhci_hcd *uhci, struct urb *urb, + struct uhci_qh *qh) { - struct urb *last_urb = NULL; - struct urb_priv *up; - int ret = 0; - - list_for_each_entry(up, &uhci->urb_list, urb_list) { - struct urb *u = up->urb; - - /* look for pending URB's with identical pipe handle */ - if ((urb->pipe == u->pipe) && (urb->dev == u->dev) && - (u->status == -EINPROGRESS) && (u != urb)) { - if (!last_urb) - *start = u->start_frame; - last_urb = u; - } - } - - if (last_urb) { - *end = (last_urb->start_frame + last_urb->number_of_packets * - last_urb->interval) & (UHCI_NUMFRAMES-1); - ret = 0; - } else - ret = -1; /* no previous urb found */ - - return ret; -} - -static int isochronous_find_start(struct uhci_hcd *uhci, struct urb *urb) -{ - int limits; - unsigned int start = 0, end = 0; + struct uhci_td *td = NULL; /* Since urb->number_of_packets > 0 */ + int i, frame; + unsigned long destination, status; + struct urb_priv *urbp = (struct urb_priv *) urb->hcpriv; if (urb->number_of_packets > 900) /* 900? Why? */ return -EFBIG; - limits = isochronous_find_limits(uhci, urb, &start, &end); + status = TD_CTRL_ACTIVE | TD_CTRL_IOS; + destination = (urb->pipe & PIPE_DEVEP_MASK) | usb_packetid(urb->pipe); + /* Figure out the starting frame number */ if (urb->transfer_flags & URB_ISO_ASAP) { - if (limits) { + if (list_empty(&qh->queue)) { uhci_get_current_frame_number(uhci); - urb->start_frame = (uhci->frame_number + 10) - & (UHCI_NUMFRAMES - 1); - } else - urb->start_frame = end; + urb->start_frame = (uhci->frame_number + 10); + + } else { /* Go right after the last one */ + struct urb *last_urb; + + last_urb = list_entry(qh->queue.prev, + struct urb_priv, node)->urb; + urb->start_frame = (last_urb->start_frame + + last_urb->number_of_packets * + last_urb->interval); + } } else { - urb->start_frame &= (UHCI_NUMFRAMES - 1); /* FIXME: Sanity check */ } + urb->start_frame &= (UHCI_NUMFRAMES - 1); - return 0; -} - -/* - * Isochronous transfers - */ -static int uhci_submit_isochronous(struct uhci_hcd *uhci, struct urb *urb) -{ - struct uhci_td *td; - int i, ret, frame; - int status, destination; - - status = TD_CTRL_ACTIVE | TD_CTRL_IOS; - destination = (urb->pipe & PIPE_DEVEP_MASK) | usb_packetid(urb->pipe); - - ret = isochronous_find_start(uhci, urb); - if (ret) - return ret; - - frame = urb->start_frame; - for (i = 0; i < urb->number_of_packets; i++, frame += urb->interval) { - if (!urb->iso_frame_desc[i].length) - continue; - - td = uhci_alloc_td(uhci, urb->dev); + for (i = 0; i < urb->number_of_packets; i++) { + td = uhci_alloc_td(uhci); if (!td) return -ENOMEM; uhci_add_td_to_urb(urb, td); - uhci_fill_td(td, status, destination | uhci_explen(urb->iso_frame_desc[i].length - 1), - urb->transfer_dma + urb->iso_frame_desc[i].offset); + uhci_fill_td(td, status, destination | + uhci_explen(urb->iso_frame_desc[i].length), + urb->transfer_dma + + urb->iso_frame_desc[i].offset); + } + + /* Set the interrupt-on-completion flag on the last packet. */ + td->status |= __constant_cpu_to_le32(TD_CTRL_IOC); - if (i + 1 >= urb->number_of_packets) - td->status |= cpu_to_le32(TD_CTRL_IOC); + qh->skel = uhci->skel_iso_qh; - uhci_insert_td_frame_list(uhci, td, frame); + /* Add the TDs to the frame list */ + frame = urb->start_frame; + list_for_each_entry(td, &urbp->td_list, list) { + uhci_insert_td_in_frame_list(uhci, td, frame); + frame += urb->interval; } - return -EINPROGRESS; + return 0; } static int uhci_result_isochronous(struct uhci_hcd *uhci, struct urb *urb) @@ -1120,7 +1042,7 @@ static int uhci_result_isochronous(struct uhci_hcd *uhci, struct urb *urb) int status; int i, ret = 0; - urb->actual_length = 0; + urb->actual_length = urb->error_count = 0; i = 0; list_for_each_entry(td, &urbp->td_list, list) { @@ -1148,75 +1070,63 @@ static int uhci_result_isochronous(struct uhci_hcd *uhci, struct urb *urb) return ret; } -static struct urb *uhci_find_urb_ep(struct uhci_hcd *uhci, struct urb *urb) -{ - struct urb_priv *up; - - /* We don't match Isoc transfers since they are special */ - if (usb_pipeisoc(urb->pipe)) - return NULL; - - list_for_each_entry(up, &uhci->urb_list, urb_list) { - struct urb *u = up->urb; - - if (u->dev == urb->dev && u->status == -EINPROGRESS) { - /* For control, ignore the direction */ - if (usb_pipecontrol(urb->pipe) && - (u->pipe & ~USB_DIR_IN) == (urb->pipe & ~USB_DIR_IN)) - return u; - else if (u->pipe == urb->pipe) - return u; - } - } - - return NULL; -} - static int uhci_urb_enqueue(struct usb_hcd *hcd, - struct usb_host_endpoint *ep, - struct urb *urb, int mem_flags) + struct usb_host_endpoint *hep, + struct urb *urb, gfp_t mem_flags) { int ret; struct uhci_hcd *uhci = hcd_to_uhci(hcd); unsigned long flags; - struct urb *eurb; + struct urb_priv *urbp; + struct uhci_qh *qh; int bustime; spin_lock_irqsave(&uhci->lock, flags); ret = urb->status; if (ret != -EINPROGRESS) /* URB already unlinked! */ - goto out; + goto done; - eurb = uhci_find_urb_ep(uhci, urb); + ret = -ENOMEM; + urbp = uhci_alloc_urb_priv(uhci, urb); + if (!urbp) + goto done; - if (!uhci_alloc_urb_priv(uhci, urb)) { - ret = -ENOMEM; - goto out; + if (hep->hcpriv) + qh = (struct uhci_qh *) hep->hcpriv; + else { + qh = uhci_alloc_qh(uhci, urb->dev, hep); + if (!qh) + goto err_no_qh; } + urbp->qh = qh; switch (usb_pipetype(urb->pipe)) { case PIPE_CONTROL: - ret = uhci_submit_control(uhci, urb, eurb); + ret = uhci_submit_control(uhci, urb, qh); + break; + case PIPE_BULK: + ret = uhci_submit_bulk(uhci, urb, qh); break; case PIPE_INTERRUPT: - if (!eurb) { + if (list_empty(&qh->queue)) { bustime = usb_check_bandwidth(urb->dev, urb); if (bustime < 0) ret = bustime; else { - ret = uhci_submit_interrupt(uhci, urb, eurb); - if (ret == -EINPROGRESS) + ret = uhci_submit_interrupt(uhci, urb, qh); + if (ret == 0) usb_claim_bandwidth(urb->dev, urb, bustime, 0); } } else { /* inherit from parent */ - urb->bandwidth = eurb->bandwidth; - ret = uhci_submit_interrupt(uhci, urb, eurb); + struct urb_priv *eurbp; + + eurbp = list_entry(qh->queue.prev, struct urb_priv, + node); + urb->bandwidth = eurbp->urb->bandwidth; + ret = uhci_submit_interrupt(uhci, urb, qh); } break; - case PIPE_BULK: - ret = uhci_submit_bulk(uhci, urb, eurb); - break; case PIPE_ISOCHRONOUS: bustime = usb_check_bandwidth(urb->dev, urb); if (bustime < 0) { @@ -1224,219 +1134,208 @@ static int uhci_urb_enqueue(struct usb_hcd *hcd, break; } - ret = uhci_submit_isochronous(uhci, urb); - if (ret == -EINPROGRESS) + ret = uhci_submit_isochronous(uhci, urb, qh); + if (ret == 0) usb_claim_bandwidth(urb->dev, urb, bustime, 1); break; } + if (ret != 0) + goto err_submit_failed; - if (ret != -EINPROGRESS) { - /* Submit failed, so delete it from the urb_list */ - struct urb_priv *urbp = urb->hcpriv; + /* Add this URB to the QH */ + urbp->qh = qh; + list_add_tail(&urbp->node, &qh->queue); + + /* If the new URB is the first and only one on this QH then either + * the QH is new and idle or else it's unlinked and waiting to + * become idle, so we can activate it right away. */ + if (qh->queue.next == &urbp->node) + uhci_activate_qh(uhci, qh); + goto done; + +err_submit_failed: + if (qh->state == QH_STATE_IDLE) + uhci_make_qh_idle(uhci, qh); /* Reclaim unused QH */ - list_del_init(&urbp->urb_list); - uhci_destroy_urb_priv(uhci, urb); - } else - ret = 0; +err_no_qh: + uhci_free_urb_priv(uhci, urbp); -out: +done: spin_unlock_irqrestore(&uhci->lock, flags); return ret; } -/* - * Return the result of a transfer - */ -static void uhci_transfer_result(struct uhci_hcd *uhci, struct urb *urb) +static int uhci_urb_dequeue(struct usb_hcd *hcd, struct urb *urb) { - int ret = -EINPROGRESS; + struct uhci_hcd *uhci = hcd_to_uhci(hcd); + unsigned long flags; struct urb_priv *urbp; - spin_lock(&urb->lock); + spin_lock_irqsave(&uhci->lock, flags); + urbp = urb->hcpriv; + if (!urbp) /* URB was never linked! */ + goto done; + + /* Remove Isochronous TDs from the frame list ASAP */ + if (usb_pipetype(urb->pipe) == PIPE_ISOCHRONOUS) + uhci_unlink_isochronous_tds(uhci, urb); + uhci_unlink_qh(uhci, urbp->qh); - urbp = (struct urb_priv *)urb->hcpriv; +done: + spin_unlock_irqrestore(&uhci->lock, flags); + return 0; +} - if (urb->status != -EINPROGRESS) /* URB already dequeued */ - goto out; +/* + * Finish unlinking an URB and give it back + */ +static void uhci_giveback_urb(struct uhci_hcd *uhci, struct uhci_qh *qh, + struct urb *urb, struct pt_regs *regs) +__releases(uhci->lock) +__acquires(uhci->lock) +{ + struct urb_priv *urbp = (struct urb_priv *) urb->hcpriv; + + /* Isochronous TDs get unlinked directly from the frame list */ + if (usb_pipetype(urb->pipe) == PIPE_ISOCHRONOUS) + uhci_unlink_isochronous_tds(uhci, urb); + + /* If the URB isn't first on its queue, adjust the link pointer + * of the last TD in the previous URB. */ + else if (qh->queue.next != &urbp->node) { + struct urb_priv *purbp; + struct uhci_td *ptd, *ltd; + + purbp = list_entry(urbp->node.prev, struct urb_priv, node); + ptd = list_entry(purbp->td_list.prev, struct uhci_td, + list); + ltd = list_entry(urbp->td_list.prev, struct uhci_td, + list); + ptd->link = ltd->link; + } - switch (usb_pipetype(urb->pipe)) { - case PIPE_CONTROL: - ret = uhci_result_control(uhci, urb); - break; - case PIPE_BULK: - case PIPE_INTERRUPT: - ret = uhci_result_common(uhci, urb); - break; - case PIPE_ISOCHRONOUS: - ret = uhci_result_isochronous(uhci, urb); - break; + /* Take the URB off the QH's queue. If the queue is now empty, + * this is a perfect time for a toggle fixup. */ + list_del_init(&urbp->node); + if (list_empty(&qh->queue) && qh->needs_fixup) { + usb_settoggle(urb->dev, usb_pipeendpoint(urb->pipe), + usb_pipeout(urb->pipe), qh->initial_toggle); + qh->needs_fixup = 0; } - if (ret == -EINPROGRESS) - goto out; - urb->status = ret; + uhci_dec_fsbr(uhci, urb); /* Safe since it checks */ + uhci_free_urb_priv(uhci, urbp); switch (usb_pipetype(urb->pipe)) { - case PIPE_CONTROL: - case PIPE_BULK: case PIPE_ISOCHRONOUS: /* Release bandwidth for Interrupt or Isoc. transfers */ if (urb->bandwidth) usb_release_bandwidth(urb->dev, urb, 1); - uhci_unlink_generic(uhci, urb); break; case PIPE_INTERRUPT: /* Release bandwidth for Interrupt or Isoc. transfers */ /* Make sure we don't release if we have a queued URB */ - if (list_empty(&urbp->queue_list) && urb->bandwidth) + if (list_empty(&qh->queue) && urb->bandwidth) usb_release_bandwidth(urb->dev, urb, 0); else /* bandwidth was passed on to queued URB, */ /* so don't let usb_unlink_urb() release it */ urb->bandwidth = 0; - uhci_unlink_generic(uhci, urb); break; - default: - dev_info(uhci_dev(uhci), "%s: unknown pipe type %d " - "for urb %p\n", - __FUNCTION__, usb_pipetype(urb->pipe), urb); } - /* Move it from uhci->urb_list to uhci->complete_list */ - uhci_moveto_complete(uhci, urbp); - -out: - spin_unlock(&urb->lock); -} - -static void uhci_unlink_generic(struct uhci_hcd *uhci, struct urb *urb) -{ - struct list_head *head; - struct uhci_td *td; - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; - int prevactive = 0; + spin_unlock(&uhci->lock); + usb_hcd_giveback_urb(uhci_to_hcd(uhci), urb, regs); + spin_lock(&uhci->lock); - uhci_dec_fsbr(uhci, urb); /* Safe since it checks */ + /* If the queue is now empty, we can unlink the QH and give up its + * reserved bandwidth. */ + if (list_empty(&qh->queue)) { + uhci_unlink_qh(uhci, qh); - /* - * Now we need to find out what the last successful toggle was - * so we can update the local data toggle for the next transfer - * - * There are 2 ways the last successful completed TD is found: - * - * 1) The TD is NOT active and the actual length < expected length - * 2) The TD is NOT active and it's the last TD in the chain - * - * and a third way the first uncompleted TD is found: - * - * 3) The TD is active and the previous TD is NOT active - * - * Control and Isochronous ignore the toggle, so this is safe - * for all types - * - * FIXME: The toggle fixups won't be 100% reliable until we - * change over to using a single queue for each endpoint and - * stop the queue before unlinking. - */ - head = &urbp->td_list; - list_for_each_entry(td, head, list) { - unsigned int ctrlstat = td_status(td); - - if (!(ctrlstat & TD_CTRL_ACTIVE) && - (uhci_actual_length(ctrlstat) < - uhci_expected_length(td_token(td)) || - td->list.next == head)) - usb_settoggle(urb->dev, uhci_endpoint(td_token(td)), - uhci_packetout(td_token(td)), - uhci_toggle(td_token(td)) ^ 1); - else if ((ctrlstat & TD_CTRL_ACTIVE) && !prevactive) - usb_settoggle(urb->dev, uhci_endpoint(td_token(td)), - uhci_packetout(td_token(td)), - uhci_toggle(td_token(td))); - - prevactive = ctrlstat & TD_CTRL_ACTIVE; + /* Bandwidth stuff not yet implemented */ } - - uhci_delete_queued_urb(uhci, urb); - - /* The interrupt loop will reclaim the QH's */ - uhci_remove_qh(uhci, urbp->qh); - urbp->qh = NULL; } -static int uhci_urb_dequeue(struct usb_hcd *hcd, struct urb *urb) +/* + * Scan the URBs in a QH's queue + */ +#define QH_FINISHED_UNLINKING(qh) \ + (qh->state == QH_STATE_UNLINKING && \ + uhci->frame_number + uhci->is_stopped != qh->unlink_frame) + +static void uhci_scan_qh(struct uhci_hcd *uhci, struct uhci_qh *qh, + struct pt_regs *regs) { - struct uhci_hcd *uhci = hcd_to_uhci(hcd); - unsigned long flags; struct urb_priv *urbp; + struct urb *urb; + int status; - spin_lock_irqsave(&uhci->lock, flags); - urbp = urb->hcpriv; - if (!urbp) /* URB was never linked! */ - goto done; - list_del_init(&urbp->urb_list); - - uhci_unlink_generic(uhci, urb); - - uhci_get_current_frame_number(uhci); - if (uhci->frame_number + uhci->is_stopped != uhci->urb_remove_age) { - uhci_remove_pending_urbps(uhci); - uhci->urb_remove_age = uhci->frame_number; - } - - /* If we're the first, set the next interrupt bit */ - if (list_empty(&uhci->urb_remove_list)) - uhci_set_next_interrupt(uhci); - list_add_tail(&urbp->urb_list, &uhci->urb_remove_list); - -done: - spin_unlock_irqrestore(&uhci->lock, flags); - return 0; -} + while (!list_empty(&qh->queue)) { + urbp = list_entry(qh->queue.next, struct urb_priv, node); + urb = urbp->urb; -static int uhci_fsbr_timeout(struct uhci_hcd *uhci, struct urb *urb) -{ - struct urb_priv *urbp = (struct urb_priv *)urb->hcpriv; - struct list_head *head; - struct uhci_td *td; - int count = 0; + switch (usb_pipetype(urb->pipe)) { + case PIPE_CONTROL: + status = uhci_result_control(uhci, urb); + break; + case PIPE_ISOCHRONOUS: + status = uhci_result_isochronous(uhci, urb); + break; + default: /* PIPE_BULK or PIPE_INTERRUPT */ + status = uhci_result_common(uhci, urb); + break; + } + if (status == -EINPROGRESS) + break; - uhci_dec_fsbr(uhci, urb); + spin_lock(&urb->lock); + if (urb->status == -EINPROGRESS) /* Not dequeued */ + urb->status = status; + else + status = -ECONNRESET; + spin_unlock(&urb->lock); - urbp->fsbr_timeout = 1; + /* Dequeued but completed URBs can't be given back unless + * the QH is stopped or has finished unlinking. */ + if (status == -ECONNRESET && + !(qh->is_stopped || QH_FINISHED_UNLINKING(qh))) + return; - /* - * Ideally we would want to fix qh->element as well, but it's - * read/write by the HC, so that can introduce a race. It's not - * really worth the hassle - */ - - head = &urbp->td_list; - list_for_each_entry(td, head, list) { - /* - * Make sure we don't do the last one (since it'll have the - * TERM bit set) as well as we skip every so many TD's to - * make sure it doesn't hog the bandwidth - */ - if (td->list.next != head && (count % DEPTH_INTERVAL) == - (DEPTH_INTERVAL - 1)) - td->link |= UHCI_PTR_DEPTH; - - count++; + uhci_giveback_urb(uhci, qh, urb, regs); + if (qh->is_stopped) + break; } - return 0; -} - -static void uhci_free_pending_qhs(struct uhci_hcd *uhci) -{ - struct uhci_qh *qh, *tmp; - - list_for_each_entry_safe(qh, tmp, &uhci->qh_remove_list, remove_list) { - list_del_init(&qh->remove_list); + /* If the QH is neither stopped nor finished unlinking (normal case), + * our work here is done. */ + restart: + if (!(qh->is_stopped || QH_FINISHED_UNLINKING(qh))) + return; - uhci_free_qh(uhci, qh); + /* Otherwise give back each of the dequeued URBs */ + list_for_each_entry(urbp, &qh->queue, node) { + urb = urbp->urb; + if (urb->status != -EINPROGRESS) { + uhci_save_toggle(qh, urb); + uhci_giveback_urb(uhci, qh, urb, regs); + goto restart; + } + } + qh->is_stopped = 0; + + /* There are no more dequeued URBs. If there are still URBs on the + * queue, the QH can now be re-activated. */ + if (!list_empty(&qh->queue)) { + if (qh->needs_fixup) + uhci_fixup_toggles(qh, 0); + uhci_activate_qh(uhci, qh); } + + /* The queue is empty. The QH can become idle if it is fully + * unlinked. */ + else if (QH_FINISHED_UNLINKING(qh)) + uhci_make_qh_idle(uhci, qh); } static void uhci_free_pending_tds(struct uhci_hcd *uhci) @@ -1450,43 +1349,13 @@ static void uhci_free_pending_tds(struct uhci_hcd *uhci) } } -static void -uhci_finish_urb(struct usb_hcd *hcd, struct urb *urb, struct pt_regs *regs) -__releases(uhci->lock) -__acquires(uhci->lock) -{ - struct uhci_hcd *uhci = hcd_to_uhci(hcd); - - uhci_destroy_urb_priv(uhci, urb); - - spin_unlock(&uhci->lock); - usb_hcd_giveback_urb(hcd, urb, regs); - spin_lock(&uhci->lock); -} - -static void uhci_finish_completion(struct uhci_hcd *uhci, struct pt_regs *regs) -{ - struct urb_priv *urbp, *tmp; - - list_for_each_entry_safe(urbp, tmp, &uhci->complete_list, urb_list) { - struct urb *urb = urbp->urb; - - list_del_init(&urbp->urb_list); - uhci_finish_urb(uhci_to_hcd(uhci), urb, regs); - } -} - -static void uhci_remove_pending_urbps(struct uhci_hcd *uhci) -{ - - /* Splice the urb_remove_list onto the end of the complete_list */ - list_splice_init(&uhci->urb_remove_list, uhci->complete_list.prev); -} - -/* Process events in the schedule, but only in one thread at a time */ +/* + * Process events in the schedule, but only in one thread at a time + */ static void uhci_scan_schedule(struct uhci_hcd *uhci, struct pt_regs *regs) { - struct urb_priv *urbp, *tmp; + int i; + struct uhci_qh *qh; /* Don't allow re-entrant calls */ if (uhci->scan_in_progress) { @@ -1497,43 +1366,46 @@ static void uhci_scan_schedule(struct uhci_hcd *uhci, struct pt_regs *regs) rescan: uhci->need_rescan = 0; + uhci_clear_next_interrupt(uhci); uhci_get_current_frame_number(uhci); - if (uhci->frame_number + uhci->is_stopped != uhci->qh_remove_age) - uhci_free_pending_qhs(uhci); if (uhci->frame_number + uhci->is_stopped != uhci->td_remove_age) uhci_free_pending_tds(uhci); - if (uhci->frame_number + uhci->is_stopped != uhci->urb_remove_age) - uhci_remove_pending_urbps(uhci); - - /* Walk the list of pending URBs to see which ones completed - * (must be _safe because uhci_transfer_result() dequeues URBs) */ - list_for_each_entry_safe(urbp, tmp, &uhci->urb_list, urb_list) { - struct urb *urb = urbp->urb; - /* Checks the status and does all of the magic necessary */ - uhci_transfer_result(uhci, urb); - } - uhci_finish_completion(uhci, regs); - - /* If the controller is stopped, we can finish these off right now */ - if (uhci->is_stopped) { - uhci_free_pending_qhs(uhci); - uhci_free_pending_tds(uhci); - uhci_remove_pending_urbps(uhci); + /* Go through all the QH queues and process the URBs in each one */ + for (i = 0; i < UHCI_NUM_SKELQH - 1; ++i) { + uhci->next_qh = list_entry(uhci->skelqh[i]->node.next, + struct uhci_qh, node); + while ((qh = uhci->next_qh) != uhci->skelqh[i]) { + uhci->next_qh = list_entry(qh->node.next, + struct uhci_qh, node); + uhci_scan_qh(uhci, qh, regs); + } } if (uhci->need_rescan) goto rescan; uhci->scan_in_progress = 0; - if (list_empty(&uhci->urb_remove_list) && - list_empty(&uhci->td_remove_list) && - list_empty(&uhci->qh_remove_list)) + /* If the controller is stopped, we can finish these off right now */ + if (uhci->is_stopped) + uhci_free_pending_tds(uhci); + + if (list_empty(&uhci->td_remove_list) && + list_empty(&uhci->skel_unlink_qh->node)) uhci_clear_next_interrupt(uhci); else uhci_set_next_interrupt(uhci); +} - /* Wake up anyone waiting for an URB to complete */ - wake_up_all(&uhci->waitqh); +static void check_fsbr(struct uhci_hcd *uhci) +{ + /* For now, don't scan URBs for FSBR timeouts. + * Add it back in later... */ + + /* Really disable FSBR */ + if (!uhci->fsbr && uhci->fsbrtimeout && time_after_eq(jiffies, uhci->fsbrtimeout)) { + uhci->fsbrtimeout = 0; + uhci->skel_term_qh->link = UHCI_PTR_TERM; + } } diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c index 986d76222..ad9ddecf1 100644 --- a/drivers/usb/serial/ftdi_sio.c +++ b/drivers/usb/serial/ftdi_sio.c @@ -553,6 +553,10 @@ struct ftdi_private { int force_baud; /* if non-zero, force the baud rate to this value */ int force_rtscts; /* if non-zero, force RTS-CTS to always be enabled */ + + spinlock_t tx_lock; /* spinlock for transmit state */ + unsigned long tx_outstanding_bytes; + unsigned long tx_outstanding_urbs; }; /* Used for TIOCMIWAIT */ @@ -626,6 +630,9 @@ static struct usb_serial_driver ftdi_sio_device = { #define HIGH 1 #define LOW 0 +/* number of outstanding urbs to prevent userspace DoS from happening */ +#define URB_UPPER_LIMIT 42 + /* * *************************************************************************** * Utlity functions @@ -1156,6 +1163,7 @@ static int ftdi_sio_attach (struct usb_serial *serial) } spin_lock_init(&priv->rx_lock); + spin_lock_init(&priv->tx_lock); init_waitqueue_head(&priv->delta_msr_wait); /* This will push the characters through immediately rather than queue a task to deliver them */ @@ -1372,6 +1380,7 @@ static int ftdi_write (struct usb_serial_port *port, int data_offset ; /* will be 1 for the SIO and 0 otherwise */ int status; int transfer_size; + unsigned long flags; dbg("%s port %d, %d bytes", __FUNCTION__, port->number, count); @@ -1379,6 +1388,13 @@ static int ftdi_write (struct usb_serial_port *port, dbg("write request of 0 bytes"); return 0; } + spin_lock_irqsave(&priv->tx_lock, flags); + if (priv->tx_outstanding_urbs > URB_UPPER_LIMIT) { + spin_unlock_irqrestore(&priv->tx_lock, flags); + dbg("%s - write limit hit\n", __FUNCTION__); + return 0; + } + spin_unlock_irqrestore(&priv->tx_lock, flags); data_offset = priv->write_offset; dbg("data_offset set to %d",data_offset); @@ -1445,6 +1461,11 @@ static int ftdi_write (struct usb_serial_port *port, err("%s - failed submitting write urb, error %d", __FUNCTION__, status); count = status; kfree (buffer); + } else { + spin_lock_irqsave(&priv->tx_lock, flags); + ++priv->tx_outstanding_urbs; + priv->tx_outstanding_bytes += count; + spin_unlock_irqrestore(&priv->tx_lock, flags); } /* we are done with this urb, so let the host driver @@ -1460,7 +1481,11 @@ static int ftdi_write (struct usb_serial_port *port, static void ftdi_write_bulk_callback (struct urb *urb, struct pt_regs *regs) { + unsigned long flags; struct usb_serial_port *port = (struct usb_serial_port *)urb->context; + struct ftdi_private *priv; + int data_offset; /* will be 1 for the SIO and 0 otherwise */ + unsigned long countback; /* free up the transfer buffer, as usb_free_urb() does not do this */ kfree (urb->transfer_buffer); @@ -1472,34 +1497,67 @@ static void ftdi_write_bulk_callback (struct urb *urb, struct pt_regs *regs) return; } + priv = usb_get_serial_port_data(port); + if (!priv) { + dbg("%s - bad port private data pointer - exiting", __FUNCTION__); + return; + } + /* account for transferred data */ + countback = urb->actual_length; + data_offset = priv->write_offset; + if (data_offset > 0) { + /* Subtract the control bytes */ + countback -= (data_offset * ((countback + (PKTSZ - 1)) / PKTSZ)); + } + spin_lock_irqsave(&priv->tx_lock, flags); + --priv->tx_outstanding_urbs; + priv->tx_outstanding_bytes -= countback; + spin_unlock_irqrestore(&priv->tx_lock, flags); + schedule_work(&port->work); } /* ftdi_write_bulk_callback */ static int ftdi_write_room( struct usb_serial_port *port ) { + struct ftdi_private *priv = usb_get_serial_port_data(port); + int room; + unsigned long flags; + dbg("%s - port %d", __FUNCTION__, port->number); - /* - * We really can take anything the user throws at us - * but let's pick a nice big number to tell the tty - * layer that we have lots of free space - */ - return 2048; + spin_lock_irqsave(&priv->tx_lock, flags); + if (priv->tx_outstanding_urbs < URB_UPPER_LIMIT) { + /* + * We really can take anything the user throws at us + * but let's pick a nice big number to tell the tty + * layer that we have lots of free space + */ + room = 2048; + } else { + room = 0; + } + spin_unlock_irqrestore(&priv->tx_lock, flags); + return room; } /* ftdi_write_room */ static int ftdi_chars_in_buffer (struct usb_serial_port *port) { /* ftdi_chars_in_buffer */ + struct ftdi_private *priv = usb_get_serial_port_data(port); + int buffered; + unsigned long flags; + dbg("%s - port %d", __FUNCTION__, port->number); - /* - * We can't really account for how much data we - * have sent out, but hasn't made it through to the - * device, so just tell the tty layer that everything - * is flushed. - */ - return 0; + spin_lock_irqsave(&priv->tx_lock, flags); + buffered = (int)priv->tx_outstanding_bytes; + spin_unlock_irqrestore(&priv->tx_lock, flags); + if (buffered < 0) { + err("%s outstanding tx bytes is negative!", __FUNCTION__); + buffered = 0; + } + return buffered; } /* ftdi_chars_in_buffer */ diff --git a/drivers/usb/storage/unusual_devs.h b/drivers/usb/storage/unusual_devs.h index 2de6528aa..a1a83029c 100644 --- a/drivers/usb/storage/unusual_devs.h +++ b/drivers/usb/storage/unusual_devs.h @@ -113,6 +113,12 @@ UNUSUAL_DEV( 0x0411, 0x001c, 0x0113, 0x0113, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_FIX_INQUIRY ), +/* Reported by Orgad Shaneh */ +UNUSUAL_DEV( 0x0419, 0xaace, 0x0100, 0x0100, + "Samsung", "MP3 Player", + US_SC_DEVICE, US_PR_DEVICE, NULL, + US_FL_IGNORE_RESIDUE), + /* Reported by Christian Leber */ UNUSUAL_DEV( 0x0419, 0xaaf5, 0x0100, 0x0100, "TrekStor", @@ -133,12 +139,6 @@ UNUSUAL_DEV( 0x0420, 0x0001, 0x0100, 0x0100, US_SC_DEVICE, US_PR_DEVICE, NULL, US_FL_IGNORE_RESIDUE ), -/* Reported by Pete Zaitcev , bz#176584 */ -UNUSUAL_DEV( 0x0420, 0x0001, 0x0100, 0x0100, - "GENERIC", "MP3 PLAYER", /* MyMusix PD-205 on the outside. */ - US_SC_DEVICE, US_PR_DEVICE, NULL, - US_FL_IGNORE_RESIDUE ), - /* Reported by Olaf Hering from novell bug #105878 */ UNUSUAL_DEV( 0x0424, 0x0fdc, 0x0210, 0x0210, "SMSC", diff --git a/drivers/usb/storage/usb.c b/drivers/usb/storage/usb.c index dd1086343..0142fe82f 100644 --- a/drivers/usb/storage/usb.c +++ b/drivers/usb/storage/usb.c @@ -593,6 +593,15 @@ static int get_transport(struct us_data *us) break; #endif +#ifdef CONFIG_USB_STORAGE_ALAUDA + case US_PR_ALAUDA: + us->transport_name = "Alauda Control/Bulk"; + us->transport = alauda_transport; + us->transport_reset = usb_stor_Bulk_reset; + us->max_lun = 1; + break; +#endif + default: return -EIO; } @@ -648,15 +657,6 @@ static int get_protocol(struct us_data *us) break; #endif -#ifdef CONFIG_USB_STORAGE_ALAUDA - case US_PR_ALAUDA: - us->transport_name = "Alauda Control/Bulk"; - us->transport = alauda_transport; - us->transport_reset = usb_stor_Bulk_reset; - us->max_lun = 1; - break; -#endif - default: return -EIO; } diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig index 4587087d7..2fd9ab7f4 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -513,7 +513,7 @@ config FB_HGA_ACCEL config VIDEO_SELECT bool - depends on (FB = y) && X86 + depends on (FB = y) && X86 && !XEN default y config FB_SGIVW diff --git a/drivers/video/console/softcursor.c b/drivers/video/console/softcursor.c index 0a2ed2fd3..2ade75291 100644 --- a/drivers/video/console/softcursor.c +++ b/drivers/video/console/softcursor.c @@ -27,7 +27,7 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor) struct fb_image *image; u8 *dst; static u8 *src=NULL; - static int allocsize=0; + static int allocsize = 0; if (info->state != FBINFO_STATE_RUNNING) return 0; @@ -41,8 +41,10 @@ int soft_cursor(struct fb_info *info, struct fb_cursor *cursor) allocsize = dsize + sizeof(struct fb_image); src = kmalloc(allocsize, GFP_ATOMIC); - if (!src) + if (!src) { + allocsize = 0; return -ENOMEM; + } } image = (struct fb_image *) (src + dsize); diff --git a/drivers/xen/Makefile b/drivers/xen/Makefile new file mode 100644 index 000000000..79eddeec1 --- /dev/null +++ b/drivers/xen/Makefile @@ -0,0 +1,17 @@ +obj-y += core/ +obj-y += console/ +obj-y += evtchn/ +obj-y += privcmd/ +obj-y += xenbus/ + +obj-$(CONFIG_XEN_UTIL) += util.o +obj-$(CONFIG_XEN_BALLOON) += balloon/ +obj-$(CONFIG_XEN_DEVMEM) += char/ +obj-$(CONFIG_XEN_BLKDEV_BACKEND) += blkback/ +obj-$(CONFIG_XEN_BLKDEV_TAP) += blktap/ +obj-$(CONFIG_XEN_NETDEV_BACKEND) += netback/ +obj-$(CONFIG_XEN_TPMDEV_BACKEND) += tpmback/ +obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += blkfront/ +obj-$(CONFIG_XEN_NETDEV_FRONTEND) += netfront/ +obj-$(CONFIG_XEN_PCIDEV_BACKEND) += pciback/ +obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront/ diff --git a/drivers/xen/balloon/Makefile b/drivers/xen/balloon/Makefile new file mode 100644 index 000000000..0e3a3485c --- /dev/null +++ b/drivers/xen/balloon/Makefile @@ -0,0 +1,2 @@ + +obj-y += balloon.o diff --git a/drivers/xen/balloon/balloon.c b/drivers/xen/balloon/balloon.c new file mode 100644 index 000000000..322675334 --- /dev/null +++ b/drivers/xen/balloon/balloon.c @@ -0,0 +1,617 @@ +/****************************************************************************** + * balloon.c + * + * Xen balloon driver - enables returning/claiming memory to/from Xen. + * + * Copyright (c) 2003, B Dragovic + * Copyright (c) 2003-2004, M Williamson, K Fraser + * Copyright (c) 2005 Dan M. Smith, IBM Corporation + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10)) + +#ifdef CONFIG_PROC_FS +static struct proc_dir_entry *balloon_pde; +#endif + +static DECLARE_MUTEX(balloon_mutex); + +/* + * Protects atomic reservation decrease/increase against concurrent increases. + * Also protects non-atomic updates of current_pages and driver_pages, and + * balloon lists. + */ +DEFINE_SPINLOCK(balloon_lock); + +/* We aim for 'current allocation' == 'target allocation'. */ +static unsigned long current_pages; +static unsigned long target_pages; + +/* We increase/decrease in batches which fit in a page */ +static unsigned long frame_list[PAGE_SIZE / sizeof(unsigned long)]; + +/* VM /proc information for memory */ +extern unsigned long totalram_pages; + +/* We may hit the hard limit in Xen. If we do then we remember it. */ +static unsigned long hard_limit; + +/* + * Drivers may alter the memory reservation independently, but they must + * inform the balloon driver so that we can avoid hitting the hard limit. + */ +static unsigned long driver_pages; + +/* List of ballooned pages, threaded through the mem_map array. */ +static LIST_HEAD(ballooned_pages); +static unsigned long balloon_low, balloon_high; + +/* Main work function, always executed in process context. */ +static void balloon_process(void *unused); +static DECLARE_WORK(balloon_worker, balloon_process, NULL); +static struct timer_list balloon_timer; + +/* When ballooning out (allocating memory to return to Xen) we don't really + want the kernel to try too hard since that can trigger the oom killer. */ +#define GFP_BALLOON \ + (GFP_HIGHUSER | __GFP_NOWARN | __GFP_NORETRY | __GFP_NOMEMALLOC) + +#define PAGE_TO_LIST(p) (&(p)->lru) +#define LIST_TO_PAGE(l) list_entry((l), struct page, lru) +#define UNLIST_PAGE(p) \ + do { \ + list_del(PAGE_TO_LIST(p)); \ + PAGE_TO_LIST(p)->next = NULL; \ + PAGE_TO_LIST(p)->prev = NULL; \ + } while(0) + +#define IPRINTK(fmt, args...) \ + printk(KERN_INFO "xen_mem: " fmt, ##args) +#define WPRINTK(fmt, args...) \ + printk(KERN_WARNING "xen_mem: " fmt, ##args) + +/* balloon_append: add the given page to the balloon. */ +static void balloon_append(struct page *page) +{ + /* Lowmem is re-populated first, so highmem pages go at list tail. */ + if (PageHighMem(page)) { + list_add_tail(PAGE_TO_LIST(page), &ballooned_pages); + balloon_high++; + } else { + list_add(PAGE_TO_LIST(page), &ballooned_pages); + balloon_low++; + } +} + +/* balloon_retrieve: rescue a page from the balloon, if it is not empty. */ +static struct page *balloon_retrieve(void) +{ + struct page *page; + + if (list_empty(&ballooned_pages)) + return NULL; + + page = LIST_TO_PAGE(ballooned_pages.next); + UNLIST_PAGE(page); + + if (PageHighMem(page)) + balloon_high--; + else + balloon_low--; + + return page; +} + +static struct page *balloon_first_page(void) +{ + if (list_empty(&ballooned_pages)) + return NULL; + return LIST_TO_PAGE(ballooned_pages.next); +} + +static struct page *balloon_next_page(struct page *page) +{ + struct list_head *next = PAGE_TO_LIST(page)->next; + if (next == &ballooned_pages) + return NULL; + return LIST_TO_PAGE(next); +} + +static void balloon_alarm(unsigned long unused) +{ + schedule_work(&balloon_worker); +} + +static unsigned long current_target(void) +{ + unsigned long target = min(target_pages, hard_limit); + if (target > (current_pages + balloon_low + balloon_high)) + target = current_pages + balloon_low + balloon_high; + return target; +} + +static int increase_reservation(unsigned long nr_pages) +{ + unsigned long pfn, i, flags; + struct page *page; + long rc; + struct xen_memory_reservation reservation = { + .address_bits = 0, + .extent_order = 0, + .domid = DOMID_SELF + }; + + if (nr_pages > ARRAY_SIZE(frame_list)) + nr_pages = ARRAY_SIZE(frame_list); + + balloon_lock(flags); + + page = balloon_first_page(); + for (i = 0; i < nr_pages; i++) { + BUG_ON(page == NULL); + frame_list[i] = page_to_pfn(page);; + page = balloon_next_page(page); + } + + set_xen_guest_handle(reservation.extent_start, frame_list); + reservation.nr_extents = nr_pages; + rc = HYPERVISOR_memory_op( + XENMEM_populate_physmap, &reservation); + if (rc < nr_pages) { + if (rc > 0) { + int ret; + + /* We hit the Xen hard limit: reprobe. */ + reservation.nr_extents = rc; + ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, + &reservation); + BUG_ON(ret != rc); + } + if (rc >= 0) + hard_limit = current_pages + rc - driver_pages; + goto out; + } + + for (i = 0; i < nr_pages; i++) { + page = balloon_retrieve(); + BUG_ON(page == NULL); + + pfn = page_to_pfn(page); + BUG_ON(!xen_feature(XENFEAT_auto_translated_physmap) && + phys_to_machine_mapping_valid(pfn)); + + /* Update P->M and M->P tables. */ + set_phys_to_machine(pfn, frame_list[i]); + xen_machphys_update(frame_list[i], pfn); + + /* Link back into the page tables if not highmem. */ + if (pfn < max_low_pfn) { + int ret; + ret = HYPERVISOR_update_va_mapping( + (unsigned long)__va(pfn << PAGE_SHIFT), + pfn_pte_ma(frame_list[i], PAGE_KERNEL), + 0); + BUG_ON(ret); + } + + /* Relinquish the page back to the allocator. */ + ClearPageReserved(page); + init_page_count(page); + __free_page(page); + } + + current_pages += nr_pages; + totalram_pages = current_pages; + + out: + balloon_unlock(flags); + + return 0; +} + +static int decrease_reservation(unsigned long nr_pages) +{ + unsigned long pfn, i, flags; + struct page *page; + void *v; + int need_sleep = 0; + int ret; + struct xen_memory_reservation reservation = { + .address_bits = 0, + .extent_order = 0, + .domid = DOMID_SELF + }; + + if (nr_pages > ARRAY_SIZE(frame_list)) + nr_pages = ARRAY_SIZE(frame_list); + + for (i = 0; i < nr_pages; i++) { + if ((page = alloc_page(GFP_BALLOON)) == NULL) { + nr_pages = i; + need_sleep = 1; + break; + } + + pfn = page_to_pfn(page); + frame_list[i] = pfn_to_mfn(pfn); + + if (!PageHighMem(page)) { + v = phys_to_virt(pfn << PAGE_SHIFT); + scrub_pages(v, 1); + ret = HYPERVISOR_update_va_mapping( + (unsigned long)v, __pte_ma(0), 0); + BUG_ON(ret); + } +#ifdef CONFIG_XEN_SCRUB_PAGES + else { + v = kmap(page); + scrub_pages(v, 1); + kunmap(page); + } +#endif + } + + /* Ensure that ballooned highmem pages don't have kmaps. */ + kmap_flush_unused(); + flush_tlb_all(); + + balloon_lock(flags); + + /* No more mappings: invalidate P2M and add to balloon. */ + for (i = 0; i < nr_pages; i++) { + pfn = mfn_to_pfn(frame_list[i]); + set_phys_to_machine(pfn, INVALID_P2M_ENTRY); + balloon_append(pfn_to_page(pfn)); + } + + set_xen_guest_handle(reservation.extent_start, frame_list); + reservation.nr_extents = nr_pages; + ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); + BUG_ON(ret != nr_pages); + + current_pages -= nr_pages; + totalram_pages = current_pages; + + balloon_unlock(flags); + + return need_sleep; +} + +/* + * We avoid multiple worker processes conflicting via the balloon mutex. + * We may of course race updates of the target counts (which are protected + * by the balloon lock), or with changes to the Xen hard limit, but we will + * recover from these in time. + */ +static void balloon_process(void *unused) +{ + int need_sleep = 0; + long credit; + + down(&balloon_mutex); + + do { + credit = current_target() - current_pages; + if (credit > 0) + need_sleep = (increase_reservation(credit) != 0); + if (credit < 0) + need_sleep = (decrease_reservation(-credit) != 0); + +#ifndef CONFIG_PREEMPT + if (need_resched()) + schedule(); +#endif + } while ((credit != 0) && !need_sleep); + + /* Schedule more work if there is some still to be done. */ + if (current_target() != current_pages) + mod_timer(&balloon_timer, jiffies + HZ); + + up(&balloon_mutex); +} + +/* Resets the Xen limit, sets new target, and kicks off processing. */ +static void set_new_target(unsigned long target) +{ + /* No need for lock. Not read-modify-write updates. */ + hard_limit = ~0UL; + target_pages = target; + schedule_work(&balloon_worker); +} + +static struct xenbus_watch target_watch = +{ + .node = "memory/target" +}; + +/* React to a change in the target key */ +static void watch_target(struct xenbus_watch *watch, + const char **vec, unsigned int len) +{ + unsigned long long new_target; + int err; + + err = xenbus_scanf(XBT_NIL, "memory", "target", "%llu", &new_target); + if (err != 1) { + /* This is ok (for domain0 at least) - so just return */ + return; + } + + /* The given memory/target value is in KiB, so it needs converting to + * pages. PAGE_SHIFT converts bytes to pages, hence PAGE_SHIFT - 10. + */ + set_new_target(new_target >> (PAGE_SHIFT - 10)); +} + +static int balloon_init_watcher(struct notifier_block *notifier, + unsigned long event, + void *data) +{ + int err; + + err = register_xenbus_watch(&target_watch); + if (err) + printk(KERN_ERR "Failed to set balloon watcher\n"); + + return NOTIFY_DONE; +} + +#ifdef CONFIG_PROC_FS +static int balloon_write(struct file *file, const char __user *buffer, + unsigned long count, void *data) +{ + char memstring[64], *endchar; + unsigned long long target_bytes; + + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + if (count <= 1) + return -EBADMSG; /* runt */ + if (count > sizeof(memstring)) + return -EFBIG; /* too long */ + + if (copy_from_user(memstring, buffer, count)) + return -EFAULT; + memstring[sizeof(memstring)-1] = '\0'; + + target_bytes = memparse(memstring, &endchar); + set_new_target(target_bytes >> PAGE_SHIFT); + + return count; +} + +static int balloon_read(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int len; + + len = sprintf( + page, + "Current allocation: %8lu kB\n" + "Requested target: %8lu kB\n" + "Low-mem balloon: %8lu kB\n" + "High-mem balloon: %8lu kB\n" + "Xen hard limit: ", + PAGES2KB(current_pages), PAGES2KB(target_pages), + PAGES2KB(balloon_low), PAGES2KB(balloon_high)); + + if (hard_limit != ~0UL) { + len += sprintf( + page + len, + "%8lu kB (inc. %8lu kB driver headroom)\n", + PAGES2KB(hard_limit), PAGES2KB(driver_pages)); + } else { + len += sprintf( + page + len, + " ??? kB\n"); + } + + *eof = 1; + return len; +} +#endif + +static struct notifier_block xenstore_notifier; + +static int __init balloon_init(void) +{ + unsigned long pfn; + struct page *page; + + if (!is_running_on_xen()) + return -ENODEV; + + IPRINTK("Initialising balloon driver.\n"); + + current_pages = min(xen_start_info->nr_pages, max_pfn); + totalram_pages = current_pages; + target_pages = current_pages; + balloon_low = 0; + balloon_high = 0; + driver_pages = 0UL; + hard_limit = ~0UL; + + init_timer(&balloon_timer); + balloon_timer.data = 0; + balloon_timer.function = balloon_alarm; + +#ifdef CONFIG_PROC_FS + if ((balloon_pde = create_xen_proc_entry("balloon", 0644)) == NULL) { + WPRINTK("Unable to create /proc/xen/balloon.\n"); + return -1; + } + + balloon_pde->read_proc = balloon_read; + balloon_pde->write_proc = balloon_write; +#endif + + /* Initialise the balloon with excess memory space. */ + for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) { + page = pfn_to_page(pfn); + if (!PageReserved(page)) + balloon_append(page); + } + + target_watch.callback = watch_target; + xenstore_notifier.notifier_call = balloon_init_watcher; + + register_xenstore_notifier(&xenstore_notifier); + + return 0; +} + +subsys_initcall(balloon_init); + +void balloon_update_driver_allowance(long delta) +{ + unsigned long flags; + + balloon_lock(flags); + driver_pages += delta; + balloon_unlock(flags); +} + +static int dealloc_pte_fn( + pte_t *pte, struct page *pmd_page, unsigned long addr, void *data) +{ + unsigned long mfn = pte_mfn(*pte); + int ret; + struct xen_memory_reservation reservation = { + .nr_extents = 1, + .extent_order = 0, + .domid = DOMID_SELF + }; + set_xen_guest_handle(reservation.extent_start, &mfn); + set_pte_at(&init_mm, addr, pte, __pte_ma(0)); + set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY); + ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); + BUG_ON(ret != 1); + return 0; +} + +struct page *balloon_alloc_empty_page_range(unsigned long nr_pages) +{ + unsigned long vstart, flags; + unsigned int order = get_order(nr_pages * PAGE_SIZE); + int ret; + unsigned long i; + struct page *page; + + vstart = __get_free_pages(GFP_KERNEL, order); + if (vstart == 0) + return NULL; + + scrub_pages(vstart, 1 << order); + + balloon_lock(flags); + if (xen_feature(XENFEAT_auto_translated_physmap)) { + unsigned long gmfn = __pa(vstart) >> PAGE_SHIFT; + struct xen_memory_reservation reservation = { + .nr_extents = 1, + .extent_order = order, + .domid = DOMID_SELF + }; + set_xen_guest_handle(reservation.extent_start, &gmfn); + ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, + &reservation); + if (ret == -ENOSYS) + goto err; + BUG_ON(ret != 1); + } else { + ret = apply_to_page_range(&init_mm, vstart, PAGE_SIZE << order, + dealloc_pte_fn, NULL); + if (ret == -ENOSYS) + goto err; + BUG_ON(ret); + } + current_pages -= 1UL << order; + totalram_pages = current_pages; + balloon_unlock(flags); + + schedule_work(&balloon_worker); + + flush_tlb_all(); + + page = virt_to_page(vstart); + + for (i = 0; i < (1UL << order); i++) + init_page_count(page + i); + + return page; + + err: + free_pages(vstart, order); + balloon_unlock(flags); + return NULL; +} + +void balloon_dealloc_empty_page_range( + struct page *page, unsigned long nr_pages) +{ + unsigned long i, flags; + unsigned int order = get_order(nr_pages * PAGE_SIZE); + + balloon_lock(flags); + for (i = 0; i < (1UL << order); i++) { + BUG_ON(page_count(page + i) != 1); + balloon_append(page + i); + } + balloon_unlock(flags); + + schedule_work(&balloon_worker); +} + +EXPORT_SYMBOL_GPL(balloon_update_driver_allowance); +EXPORT_SYMBOL_GPL(balloon_alloc_empty_page_range); +EXPORT_SYMBOL_GPL(balloon_dealloc_empty_page_range); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/blkback/Makefile b/drivers/xen/blkback/Makefile new file mode 100644 index 000000000..8bab63da3 --- /dev/null +++ b/drivers/xen/blkback/Makefile @@ -0,0 +1,3 @@ +obj-$(CONFIG_XEN_BLKDEV_BACKEND) := blkbk.o + +blkbk-y := blkback.o xenbus.o interface.o vbd.o diff --git a/drivers/xen/blkback/blkback.c b/drivers/xen/blkback/blkback.c new file mode 100644 index 000000000..c5b3eee0b --- /dev/null +++ b/drivers/xen/blkback/blkback.c @@ -0,0 +1,566 @@ +/****************************************************************************** + * arch/xen/drivers/blkif/backend/main.c + * + * Back-end of the driver for virtual block devices. This portion of the + * driver exports a 'unified' block-device interface that can be accessed + * by any operating system that implements a compatible front end. A + * reference front-end implementation can be found in: + * arch/xen/drivers/blkif/frontend + * + * Copyright (c) 2003-2004, Keir Fraser & Steve Hand + * Copyright (c) 2005, Christopher Clark + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include "common.h" + +/* + * These are rather arbitrary. They are fairly large because adjacent requests + * pulled from a communication ring are quite likely to end up being part of + * the same scatter/gather request at the disc. + * + * ** TRY INCREASING 'blkif_reqs' IF WRITE SPEEDS SEEM TOO LOW ** + * + * This will increase the chances of being able to write whole tracks. + * 64 should be enough to keep us competitive with Linux. + */ +static int blkif_reqs = 64; +module_param_named(reqs, blkif_reqs, int, 0); +MODULE_PARM_DESC(reqs, "Number of blkback requests to allocate"); + +static int mmap_pages; + +/* Run-time switchable: /sys/module/blkback/parameters/ */ +static unsigned int log_stats = 0; +static unsigned int debug_lvl = 0; +module_param(log_stats, int, 0644); +module_param(debug_lvl, int, 0644); + +/* + * Each outstanding request that we've passed to the lower device layers has a + * 'pending_req' allocated to it. Each buffer_head that completes decrements + * the pendcnt towards zero. When it hits zero, the specified domain has a + * response queued for it, with the saved 'id' passed back. + */ +typedef struct { + blkif_t *blkif; + unsigned long id; + int nr_pages; + atomic_t pendcnt; + unsigned short operation; + int status; + struct list_head free_list; +} pending_req_t; + +static pending_req_t *pending_reqs; +static struct list_head pending_free; +static DEFINE_SPINLOCK(pending_free_lock); +static DECLARE_WAIT_QUEUE_HEAD(pending_free_wq); + +#define BLKBACK_INVALID_HANDLE (~0) + +static unsigned long mmap_vstart; +static unsigned long *pending_vaddrs; +static grant_handle_t *pending_grant_handles; + +static inline int vaddr_pagenr(pending_req_t *req, int seg) +{ + return (req - pending_reqs) * BLKIF_MAX_SEGMENTS_PER_REQUEST + seg; +} + +static inline unsigned long vaddr(pending_req_t *req, int seg) +{ + return pending_vaddrs[vaddr_pagenr(req, seg)]; +} + +#define pending_handle(_req, _seg) \ + (pending_grant_handles[vaddr_pagenr(_req, _seg)]) + + +static int do_block_io_op(blkif_t *blkif); +static void dispatch_rw_block_io(blkif_t *blkif, + blkif_request_t *req, + pending_req_t *pending_req); +static void make_response(blkif_t *blkif, unsigned long id, + unsigned short op, int st); + +/****************************************************************** + * misc small helpers + */ +static pending_req_t* alloc_req(void) +{ + pending_req_t *req = NULL; + unsigned long flags; + + spin_lock_irqsave(&pending_free_lock, flags); + if (!list_empty(&pending_free)) { + req = list_entry(pending_free.next, pending_req_t, free_list); + list_del(&req->free_list); + } + spin_unlock_irqrestore(&pending_free_lock, flags); + return req; +} + +static void free_req(pending_req_t *req) +{ + unsigned long flags; + int was_empty; + + spin_lock_irqsave(&pending_free_lock, flags); + was_empty = list_empty(&pending_free); + list_add(&req->free_list, &pending_free); + spin_unlock_irqrestore(&pending_free_lock, flags); + if (was_empty) + wake_up(&pending_free_wq); +} + +static void unplug_queue(blkif_t *blkif) +{ + if (blkif->plug == NULL) + return; + if (blkif->plug->unplug_fn) + blkif->plug->unplug_fn(blkif->plug); + blk_put_queue(blkif->plug); + blkif->plug = NULL; +} + +static void plug_queue(blkif_t *blkif, struct bio *bio) +{ + request_queue_t *q = bdev_get_queue(bio->bi_bdev); + + if (q == blkif->plug) + return; + unplug_queue(blkif); + blk_get_queue(q); + blkif->plug = q; +} + +static void fast_flush_area(pending_req_t *req) +{ + struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + unsigned int i, invcount = 0; + grant_handle_t handle; + int ret; + + for (i = 0; i < req->nr_pages; i++) { + handle = pending_handle(req, i); + if (handle == BLKBACK_INVALID_HANDLE) + continue; + gnttab_set_unmap_op(&unmap[i], vaddr(req, i), GNTMAP_host_map, + handle); + pending_handle(req, i) = BLKBACK_INVALID_HANDLE; + invcount++; + } + + ret = HYPERVISOR_grant_table_op( + GNTTABOP_unmap_grant_ref, unmap, invcount); + BUG_ON(ret); +} + +/****************************************************************** + * SCHEDULER FUNCTIONS + */ + +static void print_stats(blkif_t *blkif) +{ + printk(KERN_DEBUG "%s: oo %3d | rd %4d | wr %4d\n", + current->comm, blkif->st_oo_req, + blkif->st_rd_req, blkif->st_wr_req); + blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000); + blkif->st_rd_req = 0; + blkif->st_wr_req = 0; + blkif->st_oo_req = 0; +} + +int blkif_schedule(void *arg) +{ + blkif_t *blkif = arg; + + blkif_get(blkif); + + if (debug_lvl) + printk(KERN_DEBUG "%s: started\n", current->comm); + + while (!kthread_should_stop()) { + wait_event_interruptible( + blkif->wq, + blkif->waiting_reqs || kthread_should_stop()); + wait_event_interruptible( + pending_free_wq, + !list_empty(&pending_free) || kthread_should_stop()); + + blkif->waiting_reqs = 0; + smp_mb(); /* clear flag *before* checking for work */ + + if (do_block_io_op(blkif)) + blkif->waiting_reqs = 1; + unplug_queue(blkif); + + if (log_stats && time_after(jiffies, blkif->st_print)) + print_stats(blkif); + } + + if (log_stats) + print_stats(blkif); + if (debug_lvl) + printk(KERN_DEBUG "%s: exiting\n", current->comm); + + blkif->xenblkd = NULL; + blkif_put(blkif); + + return 0; +} + +/****************************************************************** + * COMPLETION CALLBACK -- Called as bh->b_end_io() + */ + +static void __end_block_io_op(pending_req_t *pending_req, int uptodate) +{ + /* An error fails the entire request. */ + if (!uptodate) { + DPRINTK("Buffer not up-to-date at end of operation\n"); + pending_req->status = BLKIF_RSP_ERROR; + } + + if (atomic_dec_and_test(&pending_req->pendcnt)) { + fast_flush_area(pending_req); + make_response(pending_req->blkif, pending_req->id, + pending_req->operation, pending_req->status); + blkif_put(pending_req->blkif); + free_req(pending_req); + } +} + +static int end_block_io_op(struct bio *bio, unsigned int done, int error) +{ + if (bio->bi_size != 0) + return 1; + __end_block_io_op(bio->bi_private, !error); + bio_put(bio); + return error; +} + + +/****************************************************************************** + * NOTIFICATION FROM GUEST OS. + */ + +static void blkif_notify_work(blkif_t *blkif) +{ + blkif->waiting_reqs = 1; + wake_up(&blkif->wq); +} + +irqreturn_t blkif_be_int(int irq, void *dev_id, struct pt_regs *regs) +{ + blkif_notify_work(dev_id); + return IRQ_HANDLED; +} + + + +/****************************************************************** + * DOWNWARD CALLS -- These interface with the block-device layer proper. + */ + +static int do_block_io_op(blkif_t *blkif) +{ + blkif_back_ring_t *blk_ring = &blkif->blk_ring; + blkif_request_t *req; + pending_req_t *pending_req; + RING_IDX rc, rp; + int more_to_do = 0; + + rc = blk_ring->req_cons; + rp = blk_ring->sring->req_prod; + rmb(); /* Ensure we see queued requests up to 'rp'. */ + + while ((rc != rp) && !RING_REQUEST_CONS_OVERFLOW(blk_ring, rc)) { + + pending_req = alloc_req(); + if (NULL == pending_req) { + blkif->st_oo_req++; + more_to_do = 1; + break; + } + + req = RING_GET_REQUEST(blk_ring, rc); + blk_ring->req_cons = ++rc; /* before make_response() */ + + switch (req->operation) { + case BLKIF_OP_READ: + blkif->st_rd_req++; + dispatch_rw_block_io(blkif, req, pending_req); + break; + case BLKIF_OP_WRITE: + blkif->st_wr_req++; + dispatch_rw_block_io(blkif, req, pending_req); + break; + default: + DPRINTK("error: unknown block io operation [%d]\n", + req->operation); + make_response(blkif, req->id, req->operation, + BLKIF_RSP_ERROR); + free_req(pending_req); + break; + } + } + return more_to_do; +} + +static void dispatch_rw_block_io(blkif_t *blkif, + blkif_request_t *req, + pending_req_t *pending_req) +{ + extern void ll_rw_block(int rw, int nr, struct buffer_head * bhs[]); + int operation = (req->operation == BLKIF_OP_WRITE) ? WRITE : READ; + struct gnttab_map_grant_ref map[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + struct phys_req preq; + struct { + unsigned long buf; unsigned int nsec; + } seg[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + unsigned int nseg; + struct bio *bio = NULL, *biolist[BLKIF_MAX_SEGMENTS_PER_REQUEST]; + int ret, i, nbio = 0; + + /* Check that number of segments is sane. */ + nseg = req->nr_segments; + if (unlikely(nseg == 0) || + unlikely(nseg > BLKIF_MAX_SEGMENTS_PER_REQUEST)) { + DPRINTK("Bad number of segments in request (%d)\n", nseg); + goto fail_response; + } + + preq.dev = req->handle; + preq.sector_number = req->sector_number; + preq.nr_sects = 0; + + pending_req->blkif = blkif; + pending_req->id = req->id; + pending_req->operation = operation; + pending_req->status = BLKIF_RSP_OKAY; + pending_req->nr_pages = nseg; + + for (i = 0; i < nseg; i++) { + uint32_t flags; + + seg[i].nsec = req->seg[i].last_sect - + req->seg[i].first_sect + 1; + + if ((req->seg[i].last_sect >= (PAGE_SIZE >> 9)) || + (seg[i].nsec <= 0)) + goto fail_response; + preq.nr_sects += seg[i].nsec; + + flags = GNTMAP_host_map; + if ( operation == WRITE ) + flags |= GNTMAP_readonly; + gnttab_set_map_op(&map[i], vaddr(pending_req, i), flags, + req->seg[i].gref, blkif->domid); + } + + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map, nseg); + BUG_ON(ret); + + for (i = 0; i < nseg; i++) { + if (unlikely(map[i].status != 0)) { + DPRINTK("invalid buffer -- could not remap it\n"); + goto fail_flush; + } + + pending_handle(pending_req, i) = map[i].handle; + set_phys_to_machine(__pa(vaddr( + pending_req, i)) >> PAGE_SHIFT, + FOREIGN_FRAME(map[i].dev_bus_addr >> PAGE_SHIFT)); + seg[i].buf = map[i].dev_bus_addr | + (req->seg[i].first_sect << 9); + } + + if (vbd_translate(&preq, blkif, operation) != 0) { + DPRINTK("access denied: %s of [%llu,%llu] on dev=%04x\n", + operation == READ ? "read" : "write", + preq.sector_number, + preq.sector_number + preq.nr_sects, preq.dev); + goto fail_flush; + } + + for (i = 0; i < nseg; i++) { + if (((int)preq.sector_number|(int)seg[i].nsec) & + ((bdev_hardsect_size(preq.bdev) >> 9) - 1)) { + DPRINTK("Misaligned I/O request from domain %d", + blkif->domid); + goto fail_put_bio; + } + + while ((bio == NULL) || + (bio_add_page(bio, + virt_to_page(vaddr(pending_req, i)), + seg[i].nsec << 9, + seg[i].buf & ~PAGE_MASK) == 0)) { + bio = biolist[nbio++] = bio_alloc(GFP_KERNEL, nseg-i); + if (unlikely(bio == NULL)) + goto fail_put_bio; + + bio->bi_bdev = preq.bdev; + bio->bi_private = pending_req; + bio->bi_end_io = end_block_io_op; + bio->bi_sector = preq.sector_number; + } + + preq.sector_number += seg[i].nsec; + } + + plug_queue(blkif, bio); + atomic_set(&pending_req->pendcnt, nbio); + blkif_get(blkif); + + for (i = 0; i < nbio; i++) + submit_bio(operation, biolist[i]); + + return; + + fail_put_bio: + for (i = 0; i < (nbio-1); i++) + bio_put(biolist[i]); + fail_flush: + fast_flush_area(pending_req); + fail_response: + make_response(blkif, req->id, req->operation, BLKIF_RSP_ERROR); + free_req(pending_req); +} + + + +/****************************************************************** + * MISCELLANEOUS SETUP / TEARDOWN / DEBUGGING + */ + + +static void make_response(blkif_t *blkif, unsigned long id, + unsigned short op, int st) +{ + blkif_response_t *resp; + unsigned long flags; + blkif_back_ring_t *blk_ring = &blkif->blk_ring; + int more_to_do = 0; + int notify; + + spin_lock_irqsave(&blkif->blk_ring_lock, flags); + + /* Place on the response ring for the relevant domain. */ + resp = RING_GET_RESPONSE(blk_ring, blk_ring->rsp_prod_pvt); + resp->id = id; + resp->operation = op; + resp->status = st; + blk_ring->rsp_prod_pvt++; + RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(blk_ring, notify); + + if (blk_ring->rsp_prod_pvt == blk_ring->req_cons) { + /* + * Tail check for pending requests. Allows frontend to avoid + * notifications if requests are already in flight (lower + * overheads and promotes batching). + */ + RING_FINAL_CHECK_FOR_REQUESTS(blk_ring, more_to_do); + + } else if (RING_HAS_UNCONSUMED_REQUESTS(blk_ring)) { + more_to_do = 1; + + } + spin_unlock_irqrestore(&blkif->blk_ring_lock, flags); + + if (more_to_do) + blkif_notify_work(blkif); + if (notify) + notify_remote_via_irq(blkif->irq); +} + +static int __init blkif_init(void) +{ + struct page *page; + int i; + + if (!is_running_on_xen()) + return -ENODEV; + + mmap_pages = blkif_reqs * BLKIF_MAX_SEGMENTS_PER_REQUEST; + +#ifdef CONFIG_XEN_IA64_DOM0_NON_VP + extern unsigned long alloc_empty_foreign_map_page_range( + unsigned long pages); + mmap_vstart = (unsigned long) + alloc_empty_foreign_map_page_range(mmap_pages); +#else /* ! ia64 */ + page = balloon_alloc_empty_page_range(mmap_pages); + if (page == NULL) + return -ENOMEM; + mmap_vstart = (unsigned long)pfn_to_kaddr(page_to_pfn(page)); +#endif + + pending_reqs = kmalloc(sizeof(pending_reqs[0]) * + blkif_reqs, GFP_KERNEL); + pending_grant_handles = kmalloc(sizeof(pending_grant_handles[0]) * + mmap_pages, GFP_KERNEL); + pending_vaddrs = kmalloc(sizeof(pending_vaddrs[0]) * + mmap_pages, GFP_KERNEL); + if (!pending_reqs || !pending_grant_handles || !pending_vaddrs) { + kfree(pending_reqs); + kfree(pending_grant_handles); + kfree(pending_vaddrs); + printk("%s: out of memory\n", __FUNCTION__); + return -ENOMEM; + } + + blkif_interface_init(); + + printk("%s: reqs=%d, pages=%d, mmap_vstart=0x%lx\n", + __FUNCTION__, blkif_reqs, mmap_pages, mmap_vstart); + BUG_ON(mmap_vstart == 0); + for (i = 0; i < mmap_pages; i++) { + pending_vaddrs[i] = mmap_vstart + (i << PAGE_SHIFT); + pending_grant_handles[i] = BLKBACK_INVALID_HANDLE; + } + + memset(pending_reqs, 0, sizeof(pending_reqs)); + INIT_LIST_HEAD(&pending_free); + + for (i = 0; i < blkif_reqs; i++) + list_add_tail(&pending_reqs[i].free_list, &pending_free); + + blkif_xenbus_init(); + + return 0; +} + +module_init(blkif_init); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/blkback/common.h b/drivers/xen/blkback/common.h new file mode 100644 index 000000000..d057f7924 --- /dev/null +++ b/drivers/xen/blkback/common.h @@ -0,0 +1,134 @@ +/* + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef __BLKIF__BACKEND__COMMON_H__ +#define __BLKIF__BACKEND__COMMON_H__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DPRINTK(_f, _a...) \ + pr_debug("(file=%s, line=%d) " _f, \ + __FILE__ , __LINE__ , ## _a ) + +struct vbd { + blkif_vdev_t handle; /* what the domain refers to this vbd as */ + unsigned char readonly; /* Non-zero -> read-only */ + unsigned char type; /* VDISK_xxx */ + u32 pdevice; /* phys device that this vbd maps to */ + struct block_device *bdev; +}; + +struct backend_info; + +typedef struct blkif_st { + /* Unique identifier for this interface. */ + domid_t domid; + unsigned int handle; + /* Physical parameters of the comms window. */ + unsigned int evtchn; + unsigned int irq; + /* Comms information. */ + blkif_back_ring_t blk_ring; + struct vm_struct *blk_ring_area; + /* The VBD attached to this interface. */ + struct vbd vbd; + /* Back pointer to the backend_info. */ + struct backend_info *be; + /* Private fields. */ + spinlock_t blk_ring_lock; + atomic_t refcnt; + + wait_queue_head_t wq; + struct task_struct *xenblkd; + unsigned int waiting_reqs; + request_queue_t *plug; + + /* statistics */ + unsigned long st_print; + int st_rd_req; + int st_wr_req; + int st_oo_req; + + wait_queue_head_t waiting_to_free; + + grant_handle_t shmem_handle; + grant_ref_t shmem_ref; +} blkif_t; + +blkif_t *blkif_alloc(domid_t domid); +void blkif_disconnect(blkif_t *blkif); +void blkif_free(blkif_t *blkif); +int blkif_map(blkif_t *blkif, unsigned long shared_page, unsigned int evtchn); + +#define blkif_get(_b) (atomic_inc(&(_b)->refcnt)) +#define blkif_put(_b) \ + do { \ + if (atomic_dec_and_test(&(_b)->refcnt)) \ + wake_up(&(_b)->waiting_to_free);\ + } while (0) + +/* Create a vbd. */ +int vbd_create(blkif_t *blkif, blkif_vdev_t vdevice, unsigned major, + unsigned minor, int readonly); +void vbd_free(struct vbd *vbd); + +unsigned long vbd_size(struct vbd *vbd); +unsigned int vbd_info(struct vbd *vbd); +unsigned long vbd_secsize(struct vbd *vbd); + +struct phys_req { + unsigned short dev; + unsigned short nr_sects; + struct block_device *bdev; + blkif_sector_t sector_number; +}; + +int vbd_translate(struct phys_req *req, blkif_t *blkif, int operation); + +void blkif_interface_init(void); + +void blkif_xenbus_init(void); + +irqreturn_t blkif_be_int(int irq, void *dev_id, struct pt_regs *regs); +int blkif_schedule(void *arg); + +#endif /* __BLKIF__BACKEND__COMMON_H__ */ diff --git a/drivers/xen/blkback/interface.c b/drivers/xen/blkback/interface.c new file mode 100644 index 000000000..53b4764c4 --- /dev/null +++ b/drivers/xen/blkback/interface.c @@ -0,0 +1,171 @@ +/****************************************************************************** + * arch/xen/drivers/blkif/backend/interface.c + * + * Block-device interface management. + * + * Copyright (c) 2004, Keir Fraser + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "common.h" +#include +#include + +static kmem_cache_t *blkif_cachep; + +blkif_t *blkif_alloc(domid_t domid) +{ + blkif_t *blkif; + + blkif = kmem_cache_alloc(blkif_cachep, GFP_KERNEL); + if (!blkif) + return ERR_PTR(-ENOMEM); + + memset(blkif, 0, sizeof(*blkif)); + blkif->domid = domid; + spin_lock_init(&blkif->blk_ring_lock); + atomic_set(&blkif->refcnt, 1); + init_waitqueue_head(&blkif->wq); + blkif->st_print = jiffies; + init_waitqueue_head(&blkif->waiting_to_free); + + return blkif; +} + +static int map_frontend_page(blkif_t *blkif, unsigned long shared_page) +{ + struct gnttab_map_grant_ref op; + int ret; + + gnttab_set_map_op(&op, (unsigned long)blkif->blk_ring_area->addr, + GNTMAP_host_map, shared_page, blkif->domid); + + lock_vm_area(blkif->blk_ring_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1); + unlock_vm_area(blkif->blk_ring_area); + BUG_ON(ret); + + if (op.status) { + DPRINTK(" Grant table operation failure !\n"); + return op.status; + } + + blkif->shmem_ref = shared_page; + blkif->shmem_handle = op.handle; + + return 0; +} + +static void unmap_frontend_page(blkif_t *blkif) +{ + struct gnttab_unmap_grant_ref op; + int ret; + + gnttab_set_unmap_op(&op, (unsigned long)blkif->blk_ring_area->addr, + GNTMAP_host_map, blkif->shmem_handle); + + lock_vm_area(blkif->blk_ring_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1); + unlock_vm_area(blkif->blk_ring_area); + BUG_ON(ret); +} + +int blkif_map(blkif_t *blkif, unsigned long shared_page, unsigned int evtchn) +{ + blkif_sring_t *sring; + int err; + struct evtchn_bind_interdomain bind_interdomain; + + /* Already connected through? */ + if (blkif->irq) + return 0; + + if ( (blkif->blk_ring_area = alloc_vm_area(PAGE_SIZE)) == NULL ) + return -ENOMEM; + + err = map_frontend_page(blkif, shared_page); + if (err) { + free_vm_area(blkif->blk_ring_area); + return err; + } + + bind_interdomain.remote_dom = blkif->domid; + bind_interdomain.remote_port = evtchn; + + err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain, + &bind_interdomain); + if (err) { + unmap_frontend_page(blkif); + free_vm_area(blkif->blk_ring_area); + return err; + } + + blkif->evtchn = bind_interdomain.local_port; + + sring = (blkif_sring_t *)blkif->blk_ring_area->addr; + BACK_RING_INIT(&blkif->blk_ring, sring, PAGE_SIZE); + + blkif->irq = bind_evtchn_to_irqhandler( + blkif->evtchn, blkif_be_int, 0, "blkif-backend", blkif); + + return 0; +} + +void blkif_disconnect(blkif_t *blkif) +{ + if (blkif->xenblkd) { + kthread_stop(blkif->xenblkd); + blkif->xenblkd = NULL; + } + + atomic_dec(&blkif->refcnt); + wait_event(blkif->waiting_to_free, atomic_read(&blkif->refcnt) == 0); + atomic_inc(&blkif->refcnt); + + if (blkif->irq) { + unbind_from_irqhandler(blkif->irq, blkif); + blkif->irq = 0; + } + + if (blkif->blk_ring.sring) { + unmap_frontend_page(blkif); + free_vm_area(blkif->blk_ring_area); + blkif->blk_ring.sring = NULL; + } +} + +void blkif_free(blkif_t *blkif) +{ + if (!atomic_dec_and_test(&blkif->refcnt)) + BUG(); + kmem_cache_free(blkif_cachep, blkif); +} + +void __init blkif_interface_init(void) +{ + blkif_cachep = kmem_cache_create("blkif_cache", sizeof(blkif_t), + 0, 0, NULL, NULL); +} diff --git a/drivers/xen/blkback/vbd.c b/drivers/xen/blkback/vbd.c new file mode 100644 index 000000000..a809b04cd --- /dev/null +++ b/drivers/xen/blkback/vbd.c @@ -0,0 +1,119 @@ +/****************************************************************************** + * blkback/vbd.c + * + * Routines for managing virtual block devices (VBDs). + * + * Copyright (c) 2003-2005, Keir Fraser & Steve Hand + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "common.h" +#include + +#define vbd_sz(_v) ((_v)->bdev->bd_part ? \ + (_v)->bdev->bd_part->nr_sects : (_v)->bdev->bd_disk->capacity) + +unsigned long vbd_size(struct vbd *vbd) +{ + return vbd_sz(vbd); +} + +unsigned int vbd_info(struct vbd *vbd) +{ + return vbd->type | (vbd->readonly?VDISK_READONLY:0); +} + +unsigned long vbd_secsize(struct vbd *vbd) +{ + return bdev_hardsect_size(vbd->bdev); +} + +int vbd_create(blkif_t *blkif, blkif_vdev_t handle, unsigned major, + unsigned minor, int readonly) +{ + struct vbd *vbd; + struct block_device *bdev; + + vbd = &blkif->vbd; + vbd->handle = handle; + vbd->readonly = readonly; + vbd->type = 0; + + vbd->pdevice = MKDEV(major, minor); + + bdev = open_by_devnum(vbd->pdevice, + vbd->readonly ? FMODE_READ : FMODE_WRITE); + + if (IS_ERR(bdev)) { + DPRINTK("vbd_creat: device %08x could not be opened.\n", + vbd->pdevice); + return -ENOENT; + } + + vbd->bdev = bdev; + + if (vbd->bdev->bd_disk == NULL) { + DPRINTK("vbd_creat: device %08x doesn't exist.\n", + vbd->pdevice); + vbd_free(vbd); + return -ENOENT; + } + + if (vbd->bdev->bd_disk->flags & GENHD_FL_CD) + vbd->type |= VDISK_CDROM; + if (vbd->bdev->bd_disk->flags & GENHD_FL_REMOVABLE) + vbd->type |= VDISK_REMOVABLE; + + DPRINTK("Successful creation of handle=%04x (dom=%u)\n", + handle, blkif->domid); + return 0; +} + +void vbd_free(struct vbd *vbd) +{ + if (vbd->bdev) + blkdev_put(vbd->bdev); + vbd->bdev = NULL; +} + +int vbd_translate(struct phys_req *req, blkif_t *blkif, int operation) +{ + struct vbd *vbd = &blkif->vbd; + int rc = -EACCES; + + if ((operation == WRITE) && vbd->readonly) + goto out; + + if (unlikely((req->sector_number + req->nr_sects) > vbd_sz(vbd))) + goto out; + + req->dev = vbd->pdevice; + req->bdev = vbd->bdev; + rc = 0; + + out: + return rc; +} diff --git a/drivers/xen/blkfront/Kconfig b/drivers/xen/blkfront/Kconfig new file mode 100644 index 000000000..edde837fa --- /dev/null +++ b/drivers/xen/blkfront/Kconfig @@ -0,0 +1,6 @@ + +config XENBLOCK + tristate "Block device driver" + depends on ARCH_XEN + help + Block device driver for Xen diff --git a/drivers/xen/blkfront/Makefile b/drivers/xen/blkfront/Makefile new file mode 100644 index 000000000..182ef6531 --- /dev/null +++ b/drivers/xen/blkfront/Makefile @@ -0,0 +1,5 @@ + +obj-$(CONFIG_XEN_BLKDEV_FRONTEND) := xenblk.o + +xenblk-objs := blkfront.o vbd.o + diff --git a/drivers/xen/blkfront/blkfront.c b/drivers/xen/blkfront/blkfront.c new file mode 100644 index 000000000..a911c766a --- /dev/null +++ b/drivers/xen/blkfront/blkfront.c @@ -0,0 +1,841 @@ +/****************************************************************************** + * blkfront.c + * + * XenLinux virtual block-device driver. + * + * Copyright (c) 2003-2004, Keir Fraser & Steve Hand + * Modifications by Mark A. Williamson are (c) Intel Research Cambridge + * Copyright (c) 2004, Christian Limpach + * Copyright (c) 2004, Andrew Warfield + * Copyright (c) 2005, Christopher Clark + * Copyright (c) 2005, XenSource Ltd + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include "block.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define BLKIF_STATE_DISCONNECTED 0 +#define BLKIF_STATE_CONNECTED 1 +#define BLKIF_STATE_SUSPENDED 2 + +#define MAXIMUM_OUTSTANDING_BLOCK_REQS \ + (BLKIF_MAX_SEGMENTS_PER_REQUEST * BLK_RING_SIZE) +#define GRANT_INVALID_REF 0 + +static void connect(struct blkfront_info *); +static void blkfront_closing(struct xenbus_device *); +static int blkfront_remove(struct xenbus_device *); +static int talk_to_backend(struct xenbus_device *, struct blkfront_info *); +static int setup_blkring(struct xenbus_device *, struct blkfront_info *); + +static void kick_pending_request_queues(struct blkfront_info *); + +static irqreturn_t blkif_int(int irq, void *dev_id, struct pt_regs *ptregs); +static void blkif_restart_queue(void *arg); +static void blkif_recover(struct blkfront_info *); +static void blkif_completion(struct blk_shadow *); +static void blkif_free(struct blkfront_info *, int); + + +/** + * Entry point to this code when a new device is created. Allocate the basic + * structures and the ring buffer for communication with the backend, and + * inform the backend of the appropriate details for those. Switch to + * Initialised state. + */ +static int blkfront_probe(struct xenbus_device *dev, + const struct xenbus_device_id *id) +{ + int err, vdevice, i; + struct blkfront_info *info; + + /* FIXME: Use dynamic device id if this is not set. */ + err = xenbus_scanf(XBT_NIL, dev->nodename, + "virtual-device", "%i", &vdevice); + if (err != 1) { + xenbus_dev_fatal(dev, err, "reading virtual-device"); + return err; + } + + info = kzalloc(sizeof(*info), GFP_KERNEL); + if (!info) { + xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure"); + return -ENOMEM; + } + + info->xbdev = dev; + info->vdevice = vdevice; + info->connected = BLKIF_STATE_DISCONNECTED; + INIT_WORK(&info->work, blkif_restart_queue, (void *)info); + + for (i = 0; i < BLK_RING_SIZE; i++) + info->shadow[i].req.id = i+1; + info->shadow[BLK_RING_SIZE-1].req.id = 0x0fffffff; + + /* Front end dir is a number, which is used as the id. */ + info->handle = simple_strtoul(strrchr(dev->nodename,'/')+1, NULL, 0); + dev->dev.driver_data = info; + + err = talk_to_backend(dev, info); + if (err) { + kfree(info); + dev->dev.driver_data = NULL; + return err; + } + + return 0; +} + + +/** + * We are reconnecting to the backend, due to a suspend/resume, or a backend + * driver restart. We tear down our blkif structure and recreate it, but + * leave the device-layer structures intact so that this is transparent to the + * rest of the kernel. + */ +static int blkfront_resume(struct xenbus_device *dev) +{ + struct blkfront_info *info = dev->dev.driver_data; + int err; + + DPRINTK("blkfront_resume: %s\n", dev->nodename); + + blkif_free(info, 1); + + err = talk_to_backend(dev, info); + if (!err) + blkif_recover(info); + + return err; +} + + +/* Common code used when first setting up, and when resuming. */ +static int talk_to_backend(struct xenbus_device *dev, + struct blkfront_info *info) +{ + const char *message = NULL; + struct xenbus_transaction xbt; + int err; + + /* Create shared ring, alloc event channel. */ + err = setup_blkring(dev, info); + if (err) + goto out; + +again: + err = xenbus_transaction_start(&xbt); + if (err) { + xenbus_dev_fatal(dev, err, "starting transaction"); + goto destroy_blkring; + } + + err = xenbus_printf(xbt, dev->nodename, + "ring-ref","%u", info->ring_ref); + if (err) { + message = "writing ring-ref"; + goto abort_transaction; + } + err = xenbus_printf(xbt, dev->nodename, + "event-channel", "%u", info->evtchn); + if (err) { + message = "writing event-channel"; + goto abort_transaction; + } + + err = xenbus_transaction_end(xbt, 0); + if (err) { + if (err == -EAGAIN) + goto again; + xenbus_dev_fatal(dev, err, "completing transaction"); + goto destroy_blkring; + } + + xenbus_switch_state(dev, XenbusStateInitialised); + + return 0; + + abort_transaction: + xenbus_transaction_end(xbt, 1); + if (message) + xenbus_dev_fatal(dev, err, "%s", message); + destroy_blkring: + blkif_free(info, 0); + out: + return err; +} + + +static int setup_blkring(struct xenbus_device *dev, + struct blkfront_info *info) +{ + blkif_sring_t *sring; + int err; + + info->ring_ref = GRANT_INVALID_REF; + + sring = (blkif_sring_t *)__get_free_page(GFP_KERNEL); + if (!sring) { + xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring"); + return -ENOMEM; + } + SHARED_RING_INIT(sring); + FRONT_RING_INIT(&info->ring, sring, PAGE_SIZE); + + err = xenbus_grant_ring(dev, virt_to_mfn(info->ring.sring)); + if (err < 0) { + free_page((unsigned long)sring); + info->ring.sring = NULL; + goto fail; + } + info->ring_ref = err; + + err = xenbus_alloc_evtchn(dev, &info->evtchn); + if (err) + goto fail; + + err = bind_evtchn_to_irqhandler( + info->evtchn, blkif_int, SA_SAMPLE_RANDOM, "blkif", info); + if (err <= 0) { + xenbus_dev_fatal(dev, err, + "bind_evtchn_to_irqhandler failed"); + goto fail; + } + info->irq = err; + + return 0; +fail: + blkif_free(info, 0); + return err; +} + + +/** + * Callback received when the backend's state changes. + */ +static void backend_changed(struct xenbus_device *dev, + enum xenbus_state backend_state) +{ + struct blkfront_info *info = dev->dev.driver_data; + struct block_device *bd; + + DPRINTK("blkfront:backend_changed.\n"); + + switch (backend_state) { + case XenbusStateUnknown: + case XenbusStateInitialising: + case XenbusStateInitWait: + case XenbusStateInitialised: + case XenbusStateClosed: + break; + + case XenbusStateConnected: + connect(info); + break; + + case XenbusStateClosing: + bd = bdget(info->dev); + if (bd == NULL) + xenbus_dev_fatal(dev, -ENODEV, "bdget failed"); + + mutex_lock(&bd->bd_mutex); + if (info->users > 0) + xenbus_dev_error(dev, -EBUSY, + "Device in use; refusing to close"); + else + blkfront_closing(dev); + mutex_unlock(&bd->bd_mutex); + bdput(bd); + break; + } +} + + +/* ** Connection ** */ + + +/* + * Invoked when the backend is finally 'ready' (and has told produced + * the details about the physical device - #sectors, size, etc). + */ +static void connect(struct blkfront_info *info) +{ + unsigned long sectors, sector_size; + unsigned int binfo; + int err; + + if ((info->connected == BLKIF_STATE_CONNECTED) || + (info->connected == BLKIF_STATE_SUSPENDED) ) + return; + + DPRINTK("blkfront.c:connect:%s.\n", info->xbdev->otherend); + + err = xenbus_gather(XBT_NIL, info->xbdev->otherend, + "sectors", "%lu", §ors, + "info", "%u", &binfo, + "sector-size", "%lu", §or_size, + NULL); + if (err) { + xenbus_dev_fatal(info->xbdev, err, + "reading backend fields at %s", + info->xbdev->otherend); + return; + } + + err = xlvbd_add(sectors, info->vdevice, binfo, sector_size, info); + if (err) { + xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s", + info->xbdev->otherend); + return; + } + + (void)xenbus_switch_state(info->xbdev, XenbusStateConnected); + + /* Kick pending requests. */ + spin_lock_irq(&blkif_io_lock); + info->connected = BLKIF_STATE_CONNECTED; + kick_pending_request_queues(info); + spin_unlock_irq(&blkif_io_lock); + + add_disk(info->gd); +} + +/** + * Handle the change of state of the backend to Closing. We must delete our + * device-layer structures now, to ensure that writes are flushed through to + * the backend. Once is this done, we can switch to Closed in + * acknowledgement. + */ +static void blkfront_closing(struct xenbus_device *dev) +{ + struct blkfront_info *info = dev->dev.driver_data; + unsigned long flags; + + DPRINTK("blkfront_closing: %s removed\n", dev->nodename); + + if (info->rq == NULL) + return; + + spin_lock_irqsave(&blkif_io_lock, flags); + /* No more blkif_request(). */ + blk_stop_queue(info->rq); + /* No more gnttab callback work. */ + gnttab_cancel_free_callback(&info->callback); + flush_scheduled_work(); + spin_unlock_irqrestore(&blkif_io_lock, flags); + + xlvbd_del(info); + + xenbus_switch_state(dev, XenbusStateClosed); +} + + +static int blkfront_remove(struct xenbus_device *dev) +{ + struct blkfront_info *info = dev->dev.driver_data; + + DPRINTK("blkfront_remove: %s removed\n", dev->nodename); + + blkif_free(info, 0); + + kfree(info); + + return 0; +} + + +static inline int GET_ID_FROM_FREELIST( + struct blkfront_info *info) +{ + unsigned long free = info->shadow_free; + BUG_ON(free > BLK_RING_SIZE); + info->shadow_free = info->shadow[free].req.id; + info->shadow[free].req.id = 0x0fffffee; /* debug */ + return free; +} + +static inline void ADD_ID_TO_FREELIST( + struct blkfront_info *info, unsigned long id) +{ + info->shadow[id].req.id = info->shadow_free; + info->shadow[id].request = 0; + info->shadow_free = id; +} + +static inline void flush_requests(struct blkfront_info *info) +{ + int notify; + + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&info->ring, notify); + + if (notify) + notify_remote_via_irq(info->irq); +} + +static void kick_pending_request_queues(struct blkfront_info *info) +{ + if (!RING_FULL(&info->ring)) { + /* Re-enable calldowns. */ + blk_start_queue(info->rq); + /* Kick things off immediately. */ + do_blkif_request(info->rq); + } +} + +static void blkif_restart_queue(void *arg) +{ + struct blkfront_info *info = (struct blkfront_info *)arg; + spin_lock_irq(&blkif_io_lock); + if (info->connected == BLKIF_STATE_CONNECTED) + kick_pending_request_queues(info); + spin_unlock_irq(&blkif_io_lock); +} + +static void blkif_restart_queue_callback(void *arg) +{ + struct blkfront_info *info = (struct blkfront_info *)arg; + schedule_work(&info->work); +} + +int blkif_open(struct inode *inode, struct file *filep) +{ + struct blkfront_info *info = inode->i_bdev->bd_disk->private_data; + info->users++; + return 0; +} + + +int blkif_release(struct inode *inode, struct file *filep) +{ + struct blkfront_info *info = inode->i_bdev->bd_disk->private_data; + info->users--; + if (info->users == 0) { + /* Check whether we have been instructed to close. We will + have ignored this request initially, as the device was + still mounted. */ + struct xenbus_device * dev = info->xbdev; + enum xenbus_state state = xenbus_read_driver_state(dev->otherend); + + if (state == XenbusStateClosing) + blkfront_closing(dev); + } + return 0; +} + + +int blkif_ioctl(struct inode *inode, struct file *filep, + unsigned command, unsigned long argument) +{ + int i; + + DPRINTK_IOCTL("command: 0x%x, argument: 0x%lx, dev: 0x%04x\n", + command, (long)argument, inode->i_rdev); + + switch (command) { + case CDROMMULTISESSION: + DPRINTK("FIXME: support multisession CDs later\n"); + for (i = 0; i < sizeof(struct cdrom_multisession); i++) + if (put_user(0, (char __user *)(argument + i))) + return -EFAULT; + return 0; + + default: + /*printk(KERN_ALERT "ioctl %08x not supported by Xen blkdev\n", + command);*/ + return -EINVAL; /* same return as native Linux */ + } + + return 0; +} + + +int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg) +{ + /* We don't have real geometry info, but let's at least return + values consistent with the size of the device */ + sector_t nsect = get_capacity(bd->bd_disk); + sector_t cylinders = nsect; + + hg->heads = 0xff; + hg->sectors = 0x3f; + sector_div(cylinders, hg->heads * hg->sectors); + hg->cylinders = cylinders; + if ((sector_t)(hg->cylinders + 1) * hg->heads * hg->sectors < nsect) + hg->cylinders = 0xffff; + return 0; +} + + +/* + * blkif_queue_request + * + * request block io + * + * id: for guest use only. + * operation: BLKIF_OP_{READ,WRITE,PROBE} + * buffer: buffer to read/write into. this should be a + * virtual address in the guest os. + */ +static int blkif_queue_request(struct request *req) +{ + struct blkfront_info *info = req->rq_disk->private_data; + unsigned long buffer_mfn; + blkif_request_t *ring_req; + struct bio *bio; + struct bio_vec *bvec; + int idx; + unsigned long id; + unsigned int fsect, lsect; + int ref; + grant_ref_t gref_head; + + if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) + return 1; + + if (gnttab_alloc_grant_references( + BLKIF_MAX_SEGMENTS_PER_REQUEST, &gref_head) < 0) { + gnttab_request_free_callback( + &info->callback, + blkif_restart_queue_callback, + info, + BLKIF_MAX_SEGMENTS_PER_REQUEST); + return 1; + } + + /* Fill out a communications ring structure. */ + ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt); + id = GET_ID_FROM_FREELIST(info); + info->shadow[id].request = (unsigned long)req; + + ring_req->id = id; + ring_req->operation = rq_data_dir(req) ? + BLKIF_OP_WRITE : BLKIF_OP_READ; + ring_req->sector_number = (blkif_sector_t)req->sector; + ring_req->handle = info->handle; + + ring_req->nr_segments = 0; + rq_for_each_bio (bio, req) { + bio_for_each_segment (bvec, bio, idx) { + BUG_ON(ring_req->nr_segments + == BLKIF_MAX_SEGMENTS_PER_REQUEST); + buffer_mfn = page_to_phys(bvec->bv_page) >> PAGE_SHIFT; + fsect = bvec->bv_offset >> 9; + lsect = fsect + (bvec->bv_len >> 9) - 1; + /* install a grant reference. */ + ref = gnttab_claim_grant_reference(&gref_head); + BUG_ON(ref == -ENOSPC); + + gnttab_grant_foreign_access_ref( + ref, + info->xbdev->otherend_id, + buffer_mfn, + rq_data_dir(req) ); + + info->shadow[id].frame[ring_req->nr_segments] = + mfn_to_pfn(buffer_mfn); + + ring_req->seg[ring_req->nr_segments] = + (struct blkif_request_segment) { + .gref = ref, + .first_sect = fsect, + .last_sect = lsect }; + + ring_req->nr_segments++; + } + } + + info->ring.req_prod_pvt++; + + /* Keep a private copy so we can reissue requests when recovering. */ + info->shadow[id].req = *ring_req; + + gnttab_free_grant_references(gref_head); + + return 0; +} + +/* + * do_blkif_request + * read a block; request is in a request queue + */ +void do_blkif_request(request_queue_t *rq) +{ + struct blkfront_info *info = NULL; + struct request *req; + int queued; + + DPRINTK("Entered do_blkif_request\n"); + + queued = 0; + + while ((req = elv_next_request(rq)) != NULL) { + info = req->rq_disk->private_data; + if (!blk_fs_request(req)) { + end_request(req, 0); + continue; + } + + if (RING_FULL(&info->ring)) + goto wait; + + DPRINTK("do_blk_req %p: cmd %p, sec %lx, " + "(%u/%li) buffer:%p [%s]\n", + req, req->cmd, req->sector, req->current_nr_sectors, + req->nr_sectors, req->buffer, + rq_data_dir(req) ? "write" : "read"); + + + blkdev_dequeue_request(req); + if (blkif_queue_request(req)) { + blk_requeue_request(rq, req); + wait: + /* Avoid pointless unplugs. */ + blk_stop_queue(rq); + break; + } + + queued++; + } + + if (queued != 0) + flush_requests(info); +} + + +static irqreturn_t blkif_int(int irq, void *dev_id, struct pt_regs *ptregs) +{ + struct request *req; + blkif_response_t *bret; + RING_IDX i, rp; + unsigned long flags; + struct blkfront_info *info = (struct blkfront_info *)dev_id; + + spin_lock_irqsave(&blkif_io_lock, flags); + + if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) { + spin_unlock_irqrestore(&blkif_io_lock, flags); + return IRQ_HANDLED; + } + + again: + rp = info->ring.sring->rsp_prod; + rmb(); /* Ensure we see queued responses up to 'rp'. */ + + for (i = info->ring.rsp_cons; i != rp; i++) { + unsigned long id; + int ret; + + bret = RING_GET_RESPONSE(&info->ring, i); + id = bret->id; + req = (struct request *)info->shadow[id].request; + + blkif_completion(&info->shadow[id]); + + ADD_ID_TO_FREELIST(info, id); + + switch (bret->operation) { + case BLKIF_OP_READ: + case BLKIF_OP_WRITE: + if (unlikely(bret->status != BLKIF_RSP_OKAY)) + DPRINTK("Bad return from blkdev data " + "request: %x\n", bret->status); + + ret = end_that_request_first( + req, (bret->status == BLKIF_RSP_OKAY), + req->hard_nr_sectors); + BUG_ON(ret); + end_that_request_last( + req, (bret->status == BLKIF_RSP_OKAY)); + break; + default: + BUG(); + } + } + + info->ring.rsp_cons = i; + + if (i != info->ring.req_prod_pvt) { + int more_to_do; + RING_FINAL_CHECK_FOR_RESPONSES(&info->ring, more_to_do); + if (more_to_do) + goto again; + } else + info->ring.sring->rsp_event = i + 1; + + kick_pending_request_queues(info); + + spin_unlock_irqrestore(&blkif_io_lock, flags); + + return IRQ_HANDLED; +} + +static void blkif_free(struct blkfront_info *info, int suspend) +{ + /* Prevent new requests being issued until we fix things up. */ + spin_lock_irq(&blkif_io_lock); + info->connected = suspend ? + BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED; + /* No more blkif_request(). */ + if (info->rq) + blk_stop_queue(info->rq); + /* No more gnttab callback work. */ + gnttab_cancel_free_callback(&info->callback); + flush_scheduled_work(); + spin_unlock_irq(&blkif_io_lock); + + /* Free resources associated with old device channel. */ + if (info->ring_ref != GRANT_INVALID_REF) { + gnttab_end_foreign_access(info->ring_ref, 0, + (unsigned long)info->ring.sring); + info->ring_ref = GRANT_INVALID_REF; + info->ring.sring = NULL; + } + if (info->irq) + unbind_from_irqhandler(info->irq, info); + info->evtchn = info->irq = 0; + +} + +static void blkif_completion(struct blk_shadow *s) +{ + int i; + for (i = 0; i < s->req.nr_segments; i++) + gnttab_end_foreign_access(s->req.seg[i].gref, 0, 0UL); +} + +static void blkif_recover(struct blkfront_info *info) +{ + int i; + blkif_request_t *req; + struct blk_shadow *copy; + int j; + + /* Stage 1: Make a safe copy of the shadow state. */ + copy = kmalloc(sizeof(info->shadow), GFP_KERNEL | __GFP_NOFAIL); + memcpy(copy, info->shadow, sizeof(info->shadow)); + + /* Stage 2: Set up free list. */ + memset(&info->shadow, 0, sizeof(info->shadow)); + for (i = 0; i < BLK_RING_SIZE; i++) + info->shadow[i].req.id = i+1; + info->shadow_free = info->ring.req_prod_pvt; + info->shadow[BLK_RING_SIZE-1].req.id = 0x0fffffff; + + /* Stage 3: Find pending requests and requeue them. */ + for (i = 0; i < BLK_RING_SIZE; i++) { + /* Not in use? */ + if (copy[i].request == 0) + continue; + + /* Grab a request slot and copy shadow state into it. */ + req = RING_GET_REQUEST( + &info->ring, info->ring.req_prod_pvt); + *req = copy[i].req; + + /* We get a new request id, and must reset the shadow state. */ + req->id = GET_ID_FROM_FREELIST(info); + memcpy(&info->shadow[req->id], ©[i], sizeof(copy[i])); + + /* Rewrite any grant references invalidated by susp/resume. */ + for (j = 0; j < req->nr_segments; j++) + gnttab_grant_foreign_access_ref( + req->seg[j].gref, + info->xbdev->otherend_id, + pfn_to_mfn(info->shadow[req->id].frame[j]), + rq_data_dir( + (struct request *) + info->shadow[req->id].request)); + info->shadow[req->id].req = *req; + + info->ring.req_prod_pvt++; + } + + kfree(copy); + + (void)xenbus_switch_state(info->xbdev, XenbusStateConnected); + + spin_lock_irq(&blkif_io_lock); + + /* Now safe for us to use the shared ring */ + info->connected = BLKIF_STATE_CONNECTED; + + /* Send off requeued requests */ + flush_requests(info); + + /* Kick any other new requests queued since we resumed */ + kick_pending_request_queues(info); + + spin_unlock_irq(&blkif_io_lock); +} + + +/* ** Driver Registration ** */ + + +static struct xenbus_device_id blkfront_ids[] = { + { "vbd" }, + { "" } +}; + + +static struct xenbus_driver blkfront = { + .name = "vbd", + .owner = THIS_MODULE, + .ids = blkfront_ids, + .probe = blkfront_probe, + .remove = blkfront_remove, + .resume = blkfront_resume, + .otherend_changed = backend_changed, +}; + + +static int __init xlblk_init(void) +{ + if (!is_running_on_xen()) + return -ENODEV; + + return xenbus_register_frontend(&blkfront); +} +module_init(xlblk_init); + + +static void xlblk_exit(void) +{ + return xenbus_unregister_driver(&blkfront); +} +module_exit(xlblk_exit); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/blkfront/block.h b/drivers/xen/blkfront/block.h new file mode 100644 index 000000000..5ba3d1ebc --- /dev/null +++ b/drivers/xen/blkfront/block.h @@ -0,0 +1,156 @@ +/****************************************************************************** + * block.h + * + * Shared definitions between all levels of XenLinux Virtual block devices. + * + * Copyright (c) 2003-2004, Keir Fraser & Steve Hand + * Modifications by Mark A. Williamson are (c) Intel Research Cambridge + * Copyright (c) 2004-2005, Christian Limpach + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef __XEN_DRIVERS_BLOCK_H__ +#define __XEN_DRIVERS_BLOCK_H__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#if 1 +#define IPRINTK(fmt, args...) \ + printk(KERN_INFO "xen_blk: " fmt, ##args) +#else +#define IPRINTK(fmt, args...) ((void)0) +#endif + +#if 1 +#define WPRINTK(fmt, args...) \ + printk(KERN_WARNING "xen_blk: " fmt, ##args) +#else +#define WPRINTK(fmt, args...) ((void)0) +#endif + +#define DPRINTK(_f, _a...) pr_debug(_f, ## _a) + +#if 0 +#define DPRINTK_IOCTL(_f, _a...) printk(KERN_ALERT _f, ## _a) +#else +#define DPRINTK_IOCTL(_f, _a...) ((void)0) +#endif + +struct xlbd_type_info +{ + int partn_shift; + int disks_per_major; + char *devname; + char *diskname; +}; + +struct xlbd_major_info +{ + int major; + int index; + int usage; + struct xlbd_type_info *type; +}; + +struct blk_shadow { + blkif_request_t req; + unsigned long request; + unsigned long frame[BLKIF_MAX_SEGMENTS_PER_REQUEST]; +}; + +#define BLK_RING_SIZE __RING_SIZE((blkif_sring_t *)0, PAGE_SIZE) + +/* + * We have one of these per vbd, whether ide, scsi or 'other'. They + * hang in private_data off the gendisk structure. We may end up + * putting all kinds of interesting stuff here :-) + */ +struct blkfront_info +{ + struct xenbus_device *xbdev; + dev_t dev; + struct gendisk *gd; + int vdevice; + blkif_vdev_t handle; + int connected; + int ring_ref; + blkif_front_ring_t ring; + unsigned int evtchn, irq; + struct xlbd_major_info *mi; + request_queue_t *rq; + struct work_struct work; + struct gnttab_free_callback callback; + struct blk_shadow shadow[BLK_RING_SIZE]; + unsigned long shadow_free; + + /** + * The number of people holding this device open. We won't allow a + * hot-unplug unless this is 0. + */ + int users; +}; + +extern spinlock_t blkif_io_lock; + +extern int blkif_open(struct inode *inode, struct file *filep); +extern int blkif_release(struct inode *inode, struct file *filep); +extern int blkif_ioctl(struct inode *inode, struct file *filep, + unsigned command, unsigned long argument); +extern int blkif_getgeo(struct block_device *, struct hd_geometry *); +extern int blkif_check(dev_t dev); +extern int blkif_revalidate(dev_t dev); +extern void do_blkif_request (request_queue_t *rq); + +/* Virtual block-device subsystem. */ +/* Note that xlvbd_add doesn't call add_disk for you: you're expected + to call add_disk on info->gd once the disk is properly connected + up. */ +int xlvbd_add(blkif_sector_t capacity, int device, + u16 vdisk_info, u16 sector_size, struct blkfront_info *info); +void xlvbd_del(struct blkfront_info *info); + +#endif /* __XEN_DRIVERS_BLOCK_H__ */ diff --git a/drivers/xen/blkfront/vbd.c b/drivers/xen/blkfront/vbd.c new file mode 100644 index 000000000..8aa453d3a --- /dev/null +++ b/drivers/xen/blkfront/vbd.c @@ -0,0 +1,318 @@ +/****************************************************************************** + * vbd.c + * + * XenLinux virtual block-device driver (xvd). + * + * Copyright (c) 2003-2004, Keir Fraser & Steve Hand + * Modifications by Mark A. Williamson are (c) Intel Research Cambridge + * Copyright (c) 2004-2005, Christian Limpach + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "block.h" +#include +#include + +#define BLKIF_MAJOR(dev) ((dev)>>8) +#define BLKIF_MINOR(dev) ((dev) & 0xff) + +/* + * For convenience we distinguish between ide, scsi and 'other' (i.e., + * potentially combinations of the two) in the naming scheme and in a few other + * places. + */ + +#define NUM_IDE_MAJORS 10 +#define NUM_SCSI_MAJORS 9 +#define NUM_VBD_MAJORS 1 + +static struct xlbd_type_info xlbd_ide_type = { + .partn_shift = 6, + .disks_per_major = 2, + .devname = "ide", + .diskname = "hd", +}; + +static struct xlbd_type_info xlbd_scsi_type = { + .partn_shift = 4, + .disks_per_major = 16, + .devname = "sd", + .diskname = "sd", +}; + +static struct xlbd_type_info xlbd_vbd_type = { + .partn_shift = 4, + .disks_per_major = 16, + .devname = "xvd", + .diskname = "xvd", +}; + +static struct xlbd_major_info *major_info[NUM_IDE_MAJORS + NUM_SCSI_MAJORS + + NUM_VBD_MAJORS]; + +#define XLBD_MAJOR_IDE_START 0 +#define XLBD_MAJOR_SCSI_START (NUM_IDE_MAJORS) +#define XLBD_MAJOR_VBD_START (NUM_IDE_MAJORS + NUM_SCSI_MAJORS) + +#define XLBD_MAJOR_IDE_RANGE XLBD_MAJOR_IDE_START ... XLBD_MAJOR_SCSI_START - 1 +#define XLBD_MAJOR_SCSI_RANGE XLBD_MAJOR_SCSI_START ... XLBD_MAJOR_VBD_START - 1 +#define XLBD_MAJOR_VBD_RANGE XLBD_MAJOR_VBD_START ... XLBD_MAJOR_VBD_START + NUM_VBD_MAJORS - 1 + +/* Information about our VBDs. */ +#define MAX_VBDS 64 +static LIST_HEAD(vbds_list); + +static struct block_device_operations xlvbd_block_fops = +{ + .owner = THIS_MODULE, + .open = blkif_open, + .release = blkif_release, + .ioctl = blkif_ioctl, + .getgeo = blkif_getgeo +}; + +DEFINE_SPINLOCK(blkif_io_lock); + +static struct xlbd_major_info * +xlbd_alloc_major_info(int major, int minor, int index) +{ + struct xlbd_major_info *ptr; + + ptr = kzalloc(sizeof(struct xlbd_major_info), GFP_KERNEL); + if (ptr == NULL) + return NULL; + + ptr->major = major; + + switch (index) { + case XLBD_MAJOR_IDE_RANGE: + ptr->type = &xlbd_ide_type; + ptr->index = index - XLBD_MAJOR_IDE_START; + break; + case XLBD_MAJOR_SCSI_RANGE: + ptr->type = &xlbd_scsi_type; + ptr->index = index - XLBD_MAJOR_SCSI_START; + break; + case XLBD_MAJOR_VBD_RANGE: + ptr->type = &xlbd_vbd_type; + ptr->index = index - XLBD_MAJOR_VBD_START; + break; + } + + printk("Registering block device major %i\n", ptr->major); + if (register_blkdev(ptr->major, ptr->type->devname)) { + WPRINTK("can't get major %d with name %s\n", + ptr->major, ptr->type->devname); + kfree(ptr); + return NULL; + } + + devfs_mk_dir(ptr->type->devname); + major_info[index] = ptr; + return ptr; +} + +static struct xlbd_major_info * +xlbd_get_major_info(int vdevice) +{ + struct xlbd_major_info *mi; + int major, minor, index; + + major = BLKIF_MAJOR(vdevice); + minor = BLKIF_MINOR(vdevice); + + switch (major) { + case IDE0_MAJOR: index = 0; break; + case IDE1_MAJOR: index = 1; break; + case IDE2_MAJOR: index = 2; break; + case IDE3_MAJOR: index = 3; break; + case IDE4_MAJOR: index = 4; break; + case IDE5_MAJOR: index = 5; break; + case IDE6_MAJOR: index = 6; break; + case IDE7_MAJOR: index = 7; break; + case IDE8_MAJOR: index = 8; break; + case IDE9_MAJOR: index = 9; break; + case SCSI_DISK0_MAJOR: index = 10; break; + case SCSI_DISK1_MAJOR ... SCSI_DISK7_MAJOR: + index = 11 + major - SCSI_DISK1_MAJOR; + break; + case SCSI_CDROM_MAJOR: index = 18; break; + default: index = 19; break; + } + + mi = ((major_info[index] != NULL) ? major_info[index] : + xlbd_alloc_major_info(major, minor, index)); + if (mi) + mi->usage++; + return mi; +} + +static void +xlbd_put_major_info(struct xlbd_major_info *mi) +{ + mi->usage--; + /* XXX: release major if 0 */ +} + +static int +xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size) +{ + request_queue_t *rq; + + rq = blk_init_queue(do_blkif_request, &blkif_io_lock); + if (rq == NULL) + return -1; + + elevator_init(rq, "noop"); + + /* Hard sector size and max sectors impersonate the equiv. hardware. */ + blk_queue_hardsect_size(rq, sector_size); + blk_queue_max_sectors(rq, 512); + + /* Each segment in a request is up to an aligned page in size. */ + blk_queue_segment_boundary(rq, PAGE_SIZE - 1); + blk_queue_max_segment_size(rq, PAGE_SIZE); + + /* Ensure a merged request will fit in a single I/O ring slot. */ + blk_queue_max_phys_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST); + blk_queue_max_hw_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST); + + /* Make sure buffer addresses are sector-aligned. */ + blk_queue_dma_alignment(rq, 511); + + gd->queue = rq; + + return 0; +} + +static int +xlvbd_alloc_gendisk(int minor, blkif_sector_t capacity, int vdevice, + u16 vdisk_info, u16 sector_size, + struct blkfront_info *info) +{ + struct gendisk *gd; + struct xlbd_major_info *mi; + int nr_minors = 1; + int err = -ENODEV; + + BUG_ON(info->gd != NULL); + BUG_ON(info->mi != NULL); + BUG_ON(info->rq != NULL); + + mi = xlbd_get_major_info(vdevice); + if (mi == NULL) + goto out; + info->mi = mi; + + if ((minor & ((1 << mi->type->partn_shift) - 1)) == 0) + nr_minors = 1 << mi->type->partn_shift; + + gd = alloc_disk(nr_minors); + if (gd == NULL) + goto out; + + if (nr_minors > 1) + sprintf(gd->disk_name, "%s%c", mi->type->diskname, + 'a' + mi->index * mi->type->disks_per_major + + (minor >> mi->type->partn_shift)); + else + sprintf(gd->disk_name, "%s%c%d", mi->type->diskname, + 'a' + mi->index * mi->type->disks_per_major + + (minor >> mi->type->partn_shift), + minor & ((1 << mi->type->partn_shift) - 1)); + + gd->major = mi->major; + gd->first_minor = minor; + gd->fops = &xlvbd_block_fops; + gd->private_data = info; + gd->driverfs_dev = &(info->xbdev->dev); + set_capacity(gd, capacity); + + if (xlvbd_init_blk_queue(gd, sector_size)) { + del_gendisk(gd); + goto out; + } + + info->rq = gd->queue; + + if (vdisk_info & VDISK_READONLY) + set_disk_ro(gd, 1); + + if (vdisk_info & VDISK_REMOVABLE) + gd->flags |= GENHD_FL_REMOVABLE; + + if (vdisk_info & VDISK_CDROM) + gd->flags |= GENHD_FL_CD; + + info->gd = gd; + + return 0; + + out: + if (mi) + xlbd_put_major_info(mi); + info->mi = NULL; + return err; +} + +int +xlvbd_add(blkif_sector_t capacity, int vdevice, u16 vdisk_info, + u16 sector_size, struct blkfront_info *info) +{ + struct block_device *bd; + int err = 0; + + info->dev = MKDEV(BLKIF_MAJOR(vdevice), BLKIF_MINOR(vdevice)); + + bd = bdget(info->dev); + if (bd == NULL) + return -ENODEV; + + err = xlvbd_alloc_gendisk(BLKIF_MINOR(vdevice), capacity, vdevice, + vdisk_info, sector_size, info); + + bdput(bd); + return err; +} + +void +xlvbd_del(struct blkfront_info *info) +{ + if (info->mi == NULL) + return; + + BUG_ON(info->gd == NULL); + del_gendisk(info->gd); + put_disk(info->gd); + info->gd = NULL; + + xlbd_put_major_info(info->mi); + info->mi = NULL; + + BUG_ON(info->rq == NULL); + blk_cleanup_queue(info->rq); + info->rq = NULL; +} diff --git a/drivers/xen/blktap/Makefile b/drivers/xen/blktap/Makefile new file mode 100644 index 000000000..409b07896 --- /dev/null +++ b/drivers/xen/blktap/Makefile @@ -0,0 +1,3 @@ +LINUXINCLUDE += -I../xen/include/public/io +obj-y := xenbus.o interface.o blktap.o + diff --git a/drivers/xen/blktap/blktap.c b/drivers/xen/blktap/blktap.c new file mode 100644 index 000000000..8c5cf68d7 --- /dev/null +++ b/drivers/xen/blktap/blktap.c @@ -0,0 +1,1450 @@ +/****************************************************************************** + * drivers/xen/blktap/blktap.c + * + * Back-end driver for user level virtual block devices. This portion of the + * driver exports a 'unified' block-device interface that can be accessed + * by any operating system that implements a compatible front end. Requests + * are remapped to a user-space memory region. + * + * Based on the blkback driver code. + * + * Copyright (c) 2004-2005, Andrew Warfield and Julian Chesterfield + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include "common.h" +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define MAX_TAP_DEV 100 /*the maximum number of tapdisk ring devices */ +#define MAX_DEV_NAME 100 /*the max tapdisk ring device name e.g. blktap0 */ + +/* + * The maximum number of requests that can be outstanding at any time + * is determined by + * + * [mmap_alloc * MAX_PENDING_REQS * BLKIF_MAX_SEGMENTS_PER_REQUEST] + * + * where mmap_alloc < MAX_DYNAMIC_MEM. + * + * TODO: + * mmap_alloc is initialised to 2 and should be adjustable on the fly via + * sysfs. + */ +#define MAX_DYNAMIC_MEM 64 +#define MAX_PENDING_REQS 64 +#define MMAP_PAGES (MAX_PENDING_REQS * BLKIF_MAX_SEGMENTS_PER_REQUEST) +#define MMAP_VADDR(_start, _req,_seg) \ + (_start + \ + ((_req) * BLKIF_MAX_SEGMENTS_PER_REQUEST * PAGE_SIZE) + \ + ((_seg) * PAGE_SIZE)) +static int blkif_reqs = MAX_PENDING_REQS; +static int mmap_pages = MMAP_PAGES; + +#define RING_PAGES 1 /* BLKTAP - immediately before the mmap area, we + * have a bunch of pages reserved for shared + * memory rings. + */ + +/*Data struct associated with each of the tapdisk devices*/ +typedef struct tap_blkif { + struct vm_area_struct *vma; /*Shared memory area */ + unsigned long rings_vstart; /*Kernel memory mapping */ + unsigned long user_vstart; /*User memory mapping */ + unsigned long dev_inuse; /*One process opens device at a time. */ + unsigned long dev_pending; /*In process of being opened */ + unsigned long ring_ok; /*make this ring->state */ + blkif_front_ring_t ufe_ring; /*Rings up to user space. */ + wait_queue_head_t wait; /*for poll */ + unsigned long mode; /*current switching mode */ + int minor; /*Minor number for tapdisk device */ + pid_t pid; /*tapdisk process id */ + enum { RUNNING, CLEANSHUTDOWN } status; /*Detect a clean userspace + shutdown */ + unsigned long *idx_map; /*Record the user ring id to kern + [req id, idx] tuple */ + blkif_t *blkif; /*Associate blkif with tapdev */ +} tap_blkif_t; + +/*Private data struct associated with the inode*/ +typedef struct private_info { + int idx; +} private_info_t; + +/*Data struct handed back to userspace for tapdisk device to VBD mapping*/ +typedef struct domid_translate { + unsigned short domid; + unsigned short busid; +} domid_translate_t ; + + +domid_translate_t translate_domid[MAX_TAP_DEV]; +tap_blkif_t *tapfds[MAX_TAP_DEV]; + +static int __init set_blkif_reqs(char *str) +{ + get_option(&str, &blkif_reqs); + return 1; +} +__setup("blkif_reqs=", set_blkif_reqs); + +/* Run-time switchable: /sys/module/blktap/parameters/ */ +static unsigned int log_stats = 0; +static unsigned int debug_lvl = 0; +module_param(log_stats, int, 0644); +module_param(debug_lvl, int, 0644); + +/* + * Each outstanding request that we've passed to the lower device layers has a + * 'pending_req' allocated to it. Each buffer_head that completes decrements + * the pendcnt towards zero. When it hits zero, the specified domain has a + * response queued for it, with the saved 'id' passed back. + */ +typedef struct { + blkif_t *blkif; + unsigned long id; + unsigned short mem_idx; + int nr_pages; + atomic_t pendcnt; + unsigned short operation; + int status; + struct list_head free_list; + int inuse; +} pending_req_t; + +static pending_req_t *pending_reqs[MAX_PENDING_REQS]; +static struct list_head pending_free; +static DEFINE_SPINLOCK(pending_free_lock); +static DECLARE_WAIT_QUEUE_HEAD (pending_free_wq); +static int alloc_pending_reqs; + +typedef unsigned int PEND_RING_IDX; + +static inline int MASK_PEND_IDX(int i) { + return (i & (MAX_PENDING_REQS-1)); +} + +static inline unsigned int RTN_PEND_IDX(pending_req_t *req, int idx) { + return (req - pending_reqs[idx]); +} + +#define NR_PENDING_REQS (MAX_PENDING_REQS - pending_prod + pending_cons) + +#define BLKBACK_INVALID_HANDLE (~0) + +typedef struct mmap_page { + unsigned long start; + struct page *mpage; +} mmap_page_t; + +static mmap_page_t mmap_start[MAX_DYNAMIC_MEM]; +static unsigned short mmap_alloc = 0; +static unsigned short mmap_lock = 0; +static unsigned short mmap_inuse = 0; +static unsigned long *pending_addrs[MAX_DYNAMIC_MEM]; + +/****************************************************************** + * GRANT HANDLES + */ + +/* When using grant tables to map a frame for device access then the + * handle returned must be used to unmap the frame. This is needed to + * drop the ref count on the frame. + */ +struct grant_handle_pair +{ + grant_handle_t kernel; + grant_handle_t user; +}; + +static struct grant_handle_pair + pending_grant_handles[MAX_DYNAMIC_MEM][MMAP_PAGES]; +#define pending_handle(_id, _idx, _i) \ + (pending_grant_handles[_id][((_idx) * BLKIF_MAX_SEGMENTS_PER_REQUEST) \ + + (_i)]) + + +static int blktap_read_ufe_ring(int idx); /*local prototypes*/ + +#define BLKTAP_MINOR 0 /*/dev/xen/blktap resides at device number + major=254, minor numbers begin at 0 */ +#define BLKTAP_DEV_MAJOR 254 /* TODO: Make major number dynamic * + * and create devices in the kernel * + */ +#define BLKTAP_DEV_DIR "/dev/xen" + +/* blktap IOCTLs: */ +#define BLKTAP_IOCTL_KICK_FE 1 +#define BLKTAP_IOCTL_KICK_BE 2 /* currently unused */ +#define BLKTAP_IOCTL_SETMODE 3 +#define BLKTAP_IOCTL_SENDPID 4 +#define BLKTAP_IOCTL_NEWINTF 5 +#define BLKTAP_IOCTL_MINOR 6 +#define BLKTAP_IOCTL_MAJOR 7 +#define BLKTAP_QUERY_ALLOC_REQS 8 +#define BLKTAP_IOCTL_FREEINTF 9 +#define BLKTAP_IOCTL_PRINT_IDXS 100 + +/* blktap switching modes: (Set with BLKTAP_IOCTL_SETMODE) */ +#define BLKTAP_MODE_PASSTHROUGH 0x00000000 /* default */ +#define BLKTAP_MODE_INTERCEPT_FE 0x00000001 +#define BLKTAP_MODE_INTERCEPT_BE 0x00000002 /* unimp. */ + +#define BLKTAP_MODE_INTERPOSE \ + (BLKTAP_MODE_INTERCEPT_FE | BLKTAP_MODE_INTERCEPT_BE) + + +static inline int BLKTAP_MODE_VALID(unsigned long arg) +{ + return ((arg == BLKTAP_MODE_PASSTHROUGH ) || + (arg == BLKTAP_MODE_INTERCEPT_FE) || + (arg == BLKTAP_MODE_INTERPOSE )); +} + +/* Requests passing through the tap to userspace are re-assigned an ID. + * We must record a mapping between the BE [IDX,ID] tuple and the userspace + * ring ID. + */ + +static inline unsigned long MAKE_ID(domid_t fe_dom, PEND_RING_IDX idx) +{ + return ((fe_dom << 16) | MASK_PEND_IDX(idx)); +} + +extern inline PEND_RING_IDX ID_TO_IDX(unsigned long id) +{ + return (PEND_RING_IDX)(id & 0x0000ffff); +} + +extern inline int ID_TO_MIDX(unsigned long id) +{ + return (int)(id >> 16); +} + +#define INVALID_REQ 0xdead0000 + +/*TODO: Convert to a free list*/ +static inline int GET_NEXT_REQ(unsigned long *idx_map) +{ + int i; + for (i = 0; i < MAX_PENDING_REQS; i++) + if (idx_map[i] == INVALID_REQ) return i; + + return INVALID_REQ; +} + + +#define BLKTAP_INVALID_HANDLE(_g) \ + (((_g->kernel) == 0xFFFF) && ((_g->user) == 0xFFFF)) + +#define BLKTAP_INVALIDATE_HANDLE(_g) do { \ + (_g)->kernel = 0xFFFF; (_g)->user = 0xFFFF; \ + } while(0) + + +/****************************************************************** + * BLKTAP VM OPS + */ + +static struct page *blktap_nopage(struct vm_area_struct *vma, + unsigned long address, + int *type) +{ + /* + * if the page has not been mapped in by the driver then return + * NOPAGE_SIGBUS to the domain. + */ + + return NOPAGE_SIGBUS; +} + +struct vm_operations_struct blktap_vm_ops = { + nopage: blktap_nopage, +}; + +/****************************************************************** + * BLKTAP FILE OPS + */ + +/*Function Declarations*/ +static int get_next_free_dev(void); +static int blktap_open(struct inode *inode, struct file *filp); +static int blktap_release(struct inode *inode, struct file *filp); +static int blktap_mmap(struct file *filp, struct vm_area_struct *vma); +static int blktap_ioctl(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg); +static unsigned int blktap_poll(struct file *file, poll_table *wait); + +struct miscdevice *set_misc(int minor, char *name, int dev); + +static struct file_operations blktap_fops = { + .owner = THIS_MODULE, + .poll = blktap_poll, + .ioctl = blktap_ioctl, + .open = blktap_open, + .release = blktap_release, + .mmap = blktap_mmap, +}; + + +static int get_next_free_dev(void) +{ + tap_blkif_t *info; + int i = 0, ret = -1; + unsigned long flags; + + spin_lock_irqsave(&pending_free_lock, flags); + + while (i < MAX_TAP_DEV) { + info = tapfds[i]; + if ( (tapfds[i] != NULL) && (info->dev_inuse == 0) + && (info->dev_pending == 0) ) { + info->dev_pending = 1; + ret = i; + goto done; + } + i++; + } + +done: + spin_unlock_irqrestore(&pending_free_lock, flags); + return ret; +} + +int dom_to_devid(domid_t domid, int xenbus_id, blkif_t *blkif) +{ + int i; + + for (i = 0; i < MAX_TAP_DEV; i++) + if ( (translate_domid[i].domid == domid) + && (translate_domid[i].busid == xenbus_id) ) { + tapfds[i]->blkif = blkif; + tapfds[i]->status = RUNNING; + return i; + } + return -1; +} + +void signal_tapdisk(int idx) +{ + tap_blkif_t *info; + struct task_struct *ptask; + + info = tapfds[idx]; + if ( (idx > 0) && (idx < MAX_TAP_DEV) && (info->pid > 0) ) { + ptask = find_task_by_pid(info->pid); + if (ptask) { + info->status = CLEANSHUTDOWN; + } + } + info->blkif = NULL; + return; +} + +static int blktap_open(struct inode *inode, struct file *filp) +{ + blkif_sring_t *sring; + int idx = iminor(inode) - BLKTAP_MINOR; + tap_blkif_t *info; + private_info_t *prv; + int i; + + if (tapfds[idx] == NULL) { + WPRINTK("Unable to open device /dev/xen/blktap%d\n", + idx); + return -ENOMEM; + } + DPRINTK("Opening device /dev/xen/blktap%d\n",idx); + + info = tapfds[idx]; + + /*Only one process can access device at a time*/ + if (test_and_set_bit(0, &info->dev_inuse)) + return -EBUSY; + + info->dev_pending = 0; + + /* Allocate the fe ring. */ + sring = (blkif_sring_t *)get_zeroed_page(GFP_KERNEL); + if (sring == NULL) + goto fail_nomem; + + SetPageReserved(virt_to_page(sring)); + + SHARED_RING_INIT(sring); + FRONT_RING_INIT(&info->ufe_ring, sring, PAGE_SIZE); + + prv = kzalloc(sizeof(private_info_t),GFP_KERNEL); + prv->idx = idx; + filp->private_data = prv; + info->vma = NULL; + + info->idx_map = kmalloc(sizeof(unsigned long) * MAX_PENDING_REQS, + GFP_KERNEL); + + if (idx > 0) { + init_waitqueue_head(&info->wait); + for (i = 0; i < MAX_PENDING_REQS; i++) + info->idx_map[i] = INVALID_REQ; + } + + DPRINTK("Tap open: device /dev/xen/blktap%d\n",idx); + return 0; + + fail_nomem: + return -ENOMEM; +} + +static int blktap_release(struct inode *inode, struct file *filp) +{ + int idx = iminor(inode) - BLKTAP_MINOR; + tap_blkif_t *info; + + if (tapfds[idx] == NULL) { + WPRINTK("Trying to free device that doesn't exist " + "[/dev/xen/blktap%d]\n",idx); + return -1; + } + info = tapfds[idx]; + info->dev_inuse = 0; + DPRINTK("Freeing device [/dev/xen/blktap%d]\n",idx); + + /* Free the ring page. */ + ClearPageReserved(virt_to_page(info->ufe_ring.sring)); + free_page((unsigned long) info->ufe_ring.sring); + + /* Clear any active mappings and free foreign map table */ + if (info->vma) { + zap_page_range( + info->vma, info->vma->vm_start, + info->vma->vm_end - info->vma->vm_start, NULL); + info->vma = NULL; + } + + if (filp->private_data) kfree(filp->private_data); + + if ( (info->status != CLEANSHUTDOWN) && (info->blkif != NULL) ) { + kthread_stop(info->blkif->xenblkd); + info->blkif->xenblkd = NULL; + info->status = CLEANSHUTDOWN; + } + return 0; +} + + +/* Note on mmap: + * We need to map pages to user space in a way that will allow the block + * subsystem set up direct IO to them. This couldn't be done before, because + * there isn't really a sane way to translate a user virtual address down to a + * physical address when the page belongs to another domain. + * + * My first approach was to map the page in to kernel memory, add an entry + * for it in the physical frame list (using alloc_lomem_region as in blkback) + * and then attempt to map that page up to user space. This is disallowed + * by xen though, which realizes that we don't really own the machine frame + * underlying the physical page. + * + * The new approach is to provide explicit support for this in xen linux. + * The VMA now has a flag, VM_FOREIGN, to indicate that it contains pages + * mapped from other vms. vma->vm_private_data is set up as a mapping + * from pages to actual page structs. There is a new clause in get_user_pages + * that does the right thing for this sort of mapping. + */ +static int blktap_mmap(struct file *filp, struct vm_area_struct *vma) +{ + int size; + struct page **map; + int i; + private_info_t *prv; + tap_blkif_t *info; + + /*Retrieve the dev info*/ + prv = (private_info_t *)filp->private_data; + if (prv == NULL) { + WPRINTK("blktap: mmap, retrieving idx failed\n"); + return -ENOMEM; + } + info = tapfds[prv->idx]; + + vma->vm_flags |= VM_RESERVED; + vma->vm_ops = &blktap_vm_ops; + + size = vma->vm_end - vma->vm_start; + if (size != ((mmap_pages + RING_PAGES) << PAGE_SHIFT)) { + WPRINTK("you _must_ map exactly %d pages!\n", + mmap_pages + RING_PAGES); + return -EAGAIN; + } + + size >>= PAGE_SHIFT; + info->rings_vstart = vma->vm_start; + info->user_vstart = info->rings_vstart + (RING_PAGES << PAGE_SHIFT); + + /* Map the ring pages to the start of the region and reserve it. */ + vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot); + + if (remap_pfn_range(vma, vma->vm_start, + __pa(info->ufe_ring.sring) >> PAGE_SHIFT, + PAGE_SIZE, vma->vm_page_prot)) { + WPRINTK("Mapping user ring failed!\n"); + goto fail; + } + + /* Mark this VM as containing foreign pages, and set up mappings. */ + map = kzalloc(((vma->vm_end - vma->vm_start) >> PAGE_SHIFT) + * sizeof(struct page_struct*), + GFP_KERNEL); + if (map == NULL) { + WPRINTK("Couldn't alloc VM_FOREIGN map.\n"); + goto fail; + } + + for (i = 0; i < ((vma->vm_end - vma->vm_start) >> PAGE_SHIFT); i++) + map[i] = NULL; + + vma->vm_private_data = map; + vma->vm_flags |= VM_FOREIGN; + + info->vma = vma; + info->ring_ok = 1; + return 0; + fail: + /* Clear any active mappings. */ + zap_page_range(vma, vma->vm_start, + vma->vm_end - vma->vm_start, NULL); + + return -ENOMEM; +} + + +static int blktap_ioctl(struct inode *inode, struct file *filp, + unsigned int cmd, unsigned long arg) +{ + int idx = iminor(inode) - BLKTAP_MINOR; + switch(cmd) { + case BLKTAP_IOCTL_KICK_FE: + { + /* There are fe messages to process. */ + return blktap_read_ufe_ring(idx); + } + case BLKTAP_IOCTL_SETMODE: + { + tap_blkif_t *info = tapfds[idx]; + + if ( (idx > 0) && (idx < MAX_TAP_DEV) + && (tapfds[idx] != NULL) ) + { + if (BLKTAP_MODE_VALID(arg)) { + info->mode = arg; + /* XXX: may need to flush rings here. */ + DPRINTK("blktap: set mode to %lx\n", + arg); + return 0; + } + } + return 0; + } + case BLKTAP_IOCTL_PRINT_IDXS: + { + tap_blkif_t *info = tapfds[idx]; + + if ( (idx > 0) && (idx < MAX_TAP_DEV) + && (tapfds[idx] != NULL) ) + { + printk("User Rings: \n-----------\n"); + printk("UF: rsp_cons: %2d, req_prod_prv: %2d " + "| req_prod: %2d, rsp_prod: %2d\n", + info->ufe_ring.rsp_cons, + info->ufe_ring.req_prod_pvt, + info->ufe_ring.sring->req_prod, + info->ufe_ring.sring->rsp_prod); + } + return 0; + } + case BLKTAP_IOCTL_SENDPID: + { + tap_blkif_t *info = tapfds[idx]; + + if ( (idx > 0) && (idx < MAX_TAP_DEV) + && (tapfds[idx] != NULL) ) + { + info->pid = (pid_t)arg; + DPRINTK("blktap: pid received %d\n", + info->pid); + } + return 0; + } + case BLKTAP_IOCTL_NEWINTF: + { + uint64_t val = (uint64_t)arg; + domid_translate_t *tr = (domid_translate_t *)&val; + int newdev; + + DPRINTK("NEWINTF Req for domid %d and bus id %d\n", + tr->domid, tr->busid); + newdev = get_next_free_dev(); + if (newdev < 1) { + WPRINTK("Error initialising /dev/xen/blktap - " + "No more devices\n"); + return -1; + } + translate_domid[newdev].domid = tr->domid; + translate_domid[newdev].busid = tr->busid; + return newdev; + } + case BLKTAP_IOCTL_FREEINTF: + { + unsigned long dev = arg; + tap_blkif_t *info = NULL; + + if ( (dev > 0) && (dev < MAX_TAP_DEV) ) info = tapfds[dev]; + + if ( (info != NULL) && (info->dev_pending) ) + info->dev_pending = 0; + return 0; + } + case BLKTAP_IOCTL_MINOR: + { + unsigned long dev = arg; + tap_blkif_t *info = NULL; + + if ( (dev > 0) && (dev < MAX_TAP_DEV) ) info = tapfds[dev]; + + if (info != NULL) return info->minor; + else return -1; + } + case BLKTAP_IOCTL_MAJOR: + return BLKTAP_DEV_MAJOR; + + case BLKTAP_QUERY_ALLOC_REQS: + { + WPRINTK("BLKTAP_QUERY_ALLOC_REQS ioctl: %d/%d\n", + alloc_pending_reqs, blkif_reqs); + return (alloc_pending_reqs/blkif_reqs) * 100; + } + } + return -ENOIOCTLCMD; +} + +static unsigned int blktap_poll(struct file *file, poll_table *wait) +{ + private_info_t *prv; + tap_blkif_t *info; + + /*Retrieve the dev info*/ + prv = (private_info_t *)file->private_data; + if (prv == NULL) { + WPRINTK(" poll, retrieving idx failed\n"); + return 0; + } + + if (prv->idx == 0) return 0; + + info = tapfds[prv->idx]; + + poll_wait(file, &info->wait, wait); + if (info->ufe_ring.req_prod_pvt != info->ufe_ring.sring->req_prod) { + flush_tlb_all(); + RING_PUSH_REQUESTS(&info->ufe_ring); + return POLLIN | POLLRDNORM; + } + return 0; +} + +void blktap_kick_user(int idx) +{ + tap_blkif_t *info; + + if (idx == 0) return; + + info = tapfds[idx]; + + if (info != NULL) wake_up_interruptible(&info->wait); + return; +} + +static int do_block_io_op(blkif_t *blkif); +static void dispatch_rw_block_io(blkif_t *blkif, + blkif_request_t *req, + pending_req_t *pending_req); +static void make_response(blkif_t *blkif, unsigned long id, + unsigned short op, int st); + +/****************************************************************** + * misc small helpers + */ +static int req_increase(void) +{ + int i, j; + struct page *page; + unsigned long flags; + int ret; + + spin_lock_irqsave(&pending_free_lock, flags); + + ret = -EINVAL; + if (mmap_alloc >= MAX_PENDING_REQS || mmap_lock) + goto done; + +#ifdef __ia64__ + extern unsigned long alloc_empty_foreign_map_page_range( + unsigned long pages); + mmap_start[mmap_alloc].start = (unsigned long) + alloc_empty_foreign_map_page_range(mmap_pages); +#else /* ! ia64 */ + page = balloon_alloc_empty_page_range(mmap_pages); + ret = -ENOMEM; + if (page == NULL) { + printk("%s balloon_alloc_empty_page_range gave NULL\n", __FUNCTION__); + goto done; + } + + /* Pin all of the pages. */ + for (i=0; iinuse == 0) { + list_del(&req->free_list); + mmap_inuse--; + } + } + if (mmap_inuse == 0) mmap_req_del(mmap_alloc-1); + done: + spin_unlock_irqrestore(&pending_free_lock, flags); + return; +} + +static pending_req_t* alloc_req(void) +{ + pending_req_t *req = NULL; + unsigned long flags; + + spin_lock_irqsave(&pending_free_lock, flags); + + if (!list_empty(&pending_free)) { + req = list_entry(pending_free.next, pending_req_t, free_list); + list_del(&req->free_list); + } + + if (req) { + req->inuse = 1; + alloc_pending_reqs++; + } + spin_unlock_irqrestore(&pending_free_lock, flags); + + return req; +} + +static void free_req(pending_req_t *req) +{ + unsigned long flags; + int was_empty; + + spin_lock_irqsave(&pending_free_lock, flags); + + alloc_pending_reqs--; + req->inuse = 0; + if (mmap_lock && (req->mem_idx == mmap_alloc-1)) { + mmap_inuse--; + if (mmap_inuse == 0) mmap_req_del(mmap_alloc-1); + spin_unlock_irqrestore(&pending_free_lock, flags); + return; + } + was_empty = list_empty(&pending_free); + list_add(&req->free_list, &pending_free); + + spin_unlock_irqrestore(&pending_free_lock, flags); + + if (was_empty) + wake_up(&pending_free_wq); +} + +static void fast_flush_area(pending_req_t *req, int k_idx, int u_idx, int + tapidx) +{ + struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST*2]; + unsigned int i, invcount = 0; + struct grant_handle_pair *khandle; + uint64_t ptep; + int ret, mmap_idx; + unsigned long kvaddr, uvaddr; + + tap_blkif_t *info = tapfds[tapidx]; + + if (info == NULL) { + WPRINTK("fast_flush: Couldn't get info!\n"); + return; + } + mmap_idx = req->mem_idx; + + for (i = 0; i < req->nr_pages; i++) { + kvaddr = MMAP_VADDR(mmap_start[mmap_idx].start, k_idx, i); + uvaddr = MMAP_VADDR(info->user_vstart, u_idx, i); + + khandle = &pending_handle(mmap_idx, k_idx, i); + if (BLKTAP_INVALID_HANDLE(khandle)) { + WPRINTK("BLKTAP_INVALID_HANDLE\n"); + continue; + } + gnttab_set_unmap_op(&unmap[invcount], + MMAP_VADDR(mmap_start[mmap_idx].start, k_idx, i), + GNTMAP_host_map, khandle->kernel); + invcount++; + + if (create_lookup_pte_addr( + info->vma->vm_mm, + MMAP_VADDR(info->user_vstart, u_idx, i), + &ptep) !=0) { + WPRINTK("Couldn't get a pte addr!\n"); + return; + } + + gnttab_set_unmap_op(&unmap[invcount], + ptep, GNTMAP_host_map, + khandle->user); + invcount++; + + BLKTAP_INVALIDATE_HANDLE(khandle); + } + ret = HYPERVISOR_grant_table_op( + GNTTABOP_unmap_grant_ref, unmap, invcount); + BUG_ON(ret); + + if (info->vma != NULL) + zap_page_range(info->vma, + MMAP_VADDR(info->user_vstart, u_idx, 0), + req->nr_pages << PAGE_SHIFT, NULL); +} + +/****************************************************************** + * SCHEDULER FUNCTIONS + */ + +static void print_stats(blkif_t *blkif) +{ + printk(KERN_DEBUG "%s: oo %3d | rd %4d | wr %4d\n", + current->comm, blkif->st_oo_req, + blkif->st_rd_req, blkif->st_wr_req); + blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000); + blkif->st_rd_req = 0; + blkif->st_wr_req = 0; + blkif->st_oo_req = 0; +} + +int tap_blkif_schedule(void *arg) +{ + blkif_t *blkif = arg; + + blkif_get(blkif); + + if (debug_lvl) + printk(KERN_DEBUG "%s: started\n", current->comm); + + while (!kthread_should_stop()) { + wait_event_interruptible( + blkif->wq, + blkif->waiting_reqs || kthread_should_stop()); + wait_event_interruptible( + pending_free_wq, + !list_empty(&pending_free) || kthread_should_stop()); + + blkif->waiting_reqs = 0; + smp_mb(); /* clear flag *before* checking for work */ + + if (do_block_io_op(blkif)) + blkif->waiting_reqs = 1; + + if (log_stats && time_after(jiffies, blkif->st_print)) + print_stats(blkif); + } + + if (log_stats) + print_stats(blkif); + if (debug_lvl) + printk(KERN_DEBUG "%s: exiting\n", current->comm); + + blkif->xenblkd = NULL; + blkif_put(blkif); + + return 0; +} + +/****************************************************************** + * COMPLETION CALLBACK -- Called by user level ioctl() + */ + +static int blktap_read_ufe_ring(int idx) +{ + /* This is called to read responses from the UFE ring. */ + RING_IDX i, j, rp; + blkif_response_t *resp; + blkif_t *blkif=NULL; + int pending_idx, usr_idx, mmap_idx; + pending_req_t *pending_req; + tap_blkif_t *info; + + info = tapfds[idx]; + if (info == NULL) { + return 0; + } + + /* We currently only forward packets in INTERCEPT_FE mode. */ + if (!(info->mode & BLKTAP_MODE_INTERCEPT_FE)) + return 0; + + /* for each outstanding message on the UFEring */ + rp = info->ufe_ring.sring->rsp_prod; + rmb(); + + for (i = info->ufe_ring.rsp_cons; i != rp; i++) { + resp = RING_GET_RESPONSE(&info->ufe_ring, i); + ++info->ufe_ring.rsp_cons; + + /*retrieve [usr_idx] to [mmap_idx,pending_idx] mapping*/ + usr_idx = (int)resp->id; + pending_idx = MASK_PEND_IDX(ID_TO_IDX(info->idx_map[usr_idx])); + mmap_idx = ID_TO_MIDX(info->idx_map[usr_idx]); + + if ( (mmap_idx >= mmap_alloc) || + (ID_TO_IDX(info->idx_map[usr_idx]) >= MAX_PENDING_REQS) ) + WPRINTK("Incorrect req map" + "[%d], internal map [%d,%d (%d)]\n", + usr_idx, mmap_idx, + ID_TO_IDX(info->idx_map[usr_idx]), + MASK_PEND_IDX( + ID_TO_IDX(info->idx_map[usr_idx]))); + + pending_req = &pending_reqs[mmap_idx][pending_idx]; + blkif = pending_req->blkif; + + for (j = 0; j < pending_req->nr_pages; j++) { + + unsigned long kvaddr, uvaddr; + struct page **map = info->vma->vm_private_data; + struct page *pg; + int offset; + + uvaddr = MMAP_VADDR(info->user_vstart, usr_idx, j); + kvaddr = MMAP_VADDR(mmap_start[mmap_idx].start, + pending_idx, j); + + pg = pfn_to_page(__pa(kvaddr) >> PAGE_SHIFT); + ClearPageReserved(pg); + offset = (uvaddr - info->vma->vm_start) + >> PAGE_SHIFT; + map[offset] = NULL; + } + fast_flush_area(pending_req, pending_idx, usr_idx, idx); + make_response(blkif, pending_req->id, resp->operation, + resp->status); + info->idx_map[usr_idx] = INVALID_REQ; + blkif_put(pending_req->blkif); + free_req(pending_req); + } + + return 0; +} + + +/****************************************************************************** + * NOTIFICATION FROM GUEST OS. + */ + +static void blkif_notify_work(blkif_t *blkif) +{ + blkif->waiting_reqs = 1; + wake_up(&blkif->wq); +} + +irqreturn_t tap_blkif_be_int(int irq, void *dev_id, struct pt_regs *regs) +{ + blkif_notify_work(dev_id); + return IRQ_HANDLED; +} + + + +/****************************************************************** + * DOWNWARD CALLS -- These interface with the block-device layer proper. + */ +static int print_dbug = 1; +static int do_block_io_op(blkif_t *blkif) +{ + blkif_back_ring_t *blk_ring = &blkif->blk_ring; + blkif_request_t *req; + pending_req_t *pending_req; + RING_IDX rc, rp; + int more_to_do = 0; + tap_blkif_t *info; + + rc = blk_ring->req_cons; + rp = blk_ring->sring->req_prod; + rmb(); /* Ensure we see queued requests up to 'rp'. */ + + /*Check blkif has corresponding UE ring*/ + if (blkif->dev_num == -1) { + /*oops*/ + if (print_dbug) { + WPRINTK("Corresponding UE " + "ring does not exist!\n"); + print_dbug = 0; /*We only print this message once*/ + } + return 1; + } + + info = tapfds[blkif->dev_num]; + if (info == NULL || !info->dev_inuse) { + if (print_dbug) { + WPRINTK("Can't get UE info!\n"); + print_dbug = 0; + } + return 1; + } + + while (rc != rp) { + + if (RING_FULL(&info->ufe_ring)) { + WPRINTK("RING_FULL! More to do\n"); + more_to_do = 1; + break; + } + + if (RING_REQUEST_CONS_OVERFLOW(blk_ring, rc)) { + WPRINTK("RING_REQUEST_CONS_OVERFLOW!" + " More to do\n"); + more_to_do = 1; + break; + } + + pending_req = alloc_req(); + if (NULL == pending_req) { + blkif->st_oo_req++; + more_to_do = 1; + break; + } + + req = RING_GET_REQUEST(blk_ring, rc); + blk_ring->req_cons = ++rc; /* before make_response() */ + + switch (req->operation) { + case BLKIF_OP_READ: + blkif->st_rd_req++; + dispatch_rw_block_io(blkif, req, pending_req); + break; + + case BLKIF_OP_WRITE: + blkif->st_wr_req++; + dispatch_rw_block_io(blkif, req, pending_req); + break; + + default: + WPRINTK("unknown operation [%d]\n", + req->operation); + make_response(blkif, req->id, req->operation, + BLKIF_RSP_ERROR); + free_req(pending_req); + break; + } + } + + blktap_kick_user(blkif->dev_num); + + return more_to_do; +} + +static void dispatch_rw_block_io(blkif_t *blkif, + blkif_request_t *req, + pending_req_t *pending_req) +{ + extern void ll_rw_block(int rw, int nr, struct buffer_head * bhs[]); + int op, operation = (req->operation == BLKIF_OP_WRITE) ? WRITE : READ; + struct gnttab_map_grant_ref map[BLKIF_MAX_SEGMENTS_PER_REQUEST*2]; + unsigned int nseg; + int ret, i; + tap_blkif_t *info = tapfds[blkif->dev_num]; + uint64_t sector; + + blkif_request_t *target; + int pending_idx = RTN_PEND_IDX(pending_req,pending_req->mem_idx); + int usr_idx = GET_NEXT_REQ(info->idx_map); + uint16_t mmap_idx = pending_req->mem_idx; + + /*Check we have space on user ring - should never fail*/ + if(usr_idx == INVALID_REQ) goto fail_flush; + + /* Check that number of segments is sane. */ + nseg = req->nr_segments; + if ( unlikely(nseg == 0) || + unlikely(nseg > BLKIF_MAX_SEGMENTS_PER_REQUEST) ) { + WPRINTK("Bad number of segments in request (%d)\n", nseg); + goto fail_response; + } + + /* Make sure userspace is ready. */ + if (!info->ring_ok) { + WPRINTK("blktap: ring not ready for requests!\n"); + goto fail_response; + } + + if (RING_FULL(&info->ufe_ring)) { + WPRINTK("blktap: fe_ring is full, can't add " + "IO Request will be dropped. %d %d\n", + RING_SIZE(&info->ufe_ring), + RING_SIZE(&blkif->blk_ring)); + goto fail_response; + } + + pending_req->blkif = blkif; + pending_req->id = req->id; + pending_req->operation = operation; + pending_req->status = BLKIF_RSP_OKAY; + pending_req->nr_pages = nseg; + op = 0; + for (i = 0; i < nseg; i++) { + unsigned long uvaddr; + unsigned long kvaddr; + uint64_t ptep; + struct page *page; + uint32_t flags; + + uvaddr = MMAP_VADDR(info->user_vstart, usr_idx, i); + kvaddr = MMAP_VADDR(mmap_start[mmap_idx].start, + pending_idx, i); + page = virt_to_page(kvaddr); + + sector = req->sector_number + (8*i); + if( (blkif->sectors > 0) && (sector >= blkif->sectors) ) { + WPRINTK("BLKTAP: Sector request greater" + "than size\n"); + WPRINTK("BLKTAP: %s request sector" + "[%llu,%llu], Total [%llu]\n", + (req->operation == + BLKIF_OP_WRITE ? "WRITE" : "READ"), + (long long unsigned) sector, + (long long unsigned) sector>>9, + blkif->sectors); + } + + flags = GNTMAP_host_map; + if (operation == WRITE) + flags |= GNTMAP_readonly; + gnttab_set_map_op(&map[op], kvaddr, flags, + req->seg[i].gref, blkif->domid); + op++; + + /* Now map it to user. */ + ret = create_lookup_pte_addr(info->vma->vm_mm, + uvaddr, &ptep); + if (ret) { + WPRINTK("Couldn't get a pte addr!\n"); + fast_flush_area(pending_req, pending_idx, usr_idx, + blkif->dev_num); + goto fail_flush; + } + + flags = GNTMAP_host_map | GNTMAP_application_map + | GNTMAP_contains_pte; + if (operation == WRITE) + flags |= GNTMAP_readonly; + gnttab_set_map_op(&map[op], ptep, flags, + req->seg[i].gref, blkif->domid); + op++; + } + + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map, op); + BUG_ON(ret); + + for (i = 0; i < (nseg*2); i+=2) { + unsigned long uvaddr; + unsigned long kvaddr; + unsigned long offset; + struct page *pg; + + uvaddr = MMAP_VADDR(info->user_vstart, usr_idx, i/2); + kvaddr = MMAP_VADDR(mmap_start[mmap_idx].start, + pending_idx, i/2); + + if (unlikely(map[i].status != 0)) { + WPRINTK("invalid kernel buffer -- " + "could not remap it\n"); + goto fail_flush; + } + + if (unlikely(map[i+1].status != 0)) { + WPRINTK("invalid user buffer -- " + "could not remap it\n"); + goto fail_flush; + } + + pending_handle(mmap_idx, pending_idx, i/2).kernel + = map[i].handle; + pending_handle(mmap_idx, pending_idx, i/2).user + = map[i+1].handle; + set_phys_to_machine(__pa(kvaddr) >> PAGE_SHIFT, + FOREIGN_FRAME(map[i].dev_bus_addr >> PAGE_SHIFT)); + offset = (uvaddr - info->vma->vm_start) >> PAGE_SHIFT; + pg = pfn_to_page(__pa(kvaddr) >> PAGE_SHIFT); + ((struct page **)info->vma->vm_private_data)[offset] = + pg; + } + /* Mark mapped pages as reserved: */ + for (i = 0; i < req->nr_segments; i++) { + unsigned long kvaddr; + struct page *pg; + + kvaddr = MMAP_VADDR(mmap_start[mmap_idx].start, + pending_idx, i); + pg = pfn_to_page(__pa(kvaddr) >> PAGE_SHIFT); + SetPageReserved(pg); + } + + /*record [mmap_idx,pending_idx] to [usr_idx] mapping*/ + info->idx_map[usr_idx] = MAKE_ID(mmap_idx, pending_idx); + + blkif_get(blkif); + /* Finally, write the request message to the user ring. */ + target = RING_GET_REQUEST(&info->ufe_ring, + info->ufe_ring.req_prod_pvt); + memcpy(target, req, sizeof(*req)); + target->id = usr_idx; + info->ufe_ring.req_prod_pvt++; + return; + + fail_flush: + WPRINTK("Reached Fail_flush\n"); + fast_flush_area(pending_req, pending_idx, usr_idx, blkif->dev_num); + fail_response: + make_response(blkif, req->id, req->operation, BLKIF_RSP_ERROR); + free_req(pending_req); +} + + + +/****************************************************************** + * MISCELLANEOUS SETUP / TEARDOWN / DEBUGGING + */ + + +static void make_response(blkif_t *blkif, unsigned long id, + unsigned short op, int st) +{ + blkif_response_t *resp; + unsigned long flags; + blkif_back_ring_t *blk_ring = &blkif->blk_ring; + int more_to_do = 0; + int notify; + + spin_lock_irqsave(&blkif->blk_ring_lock, flags); + /* Place on the response ring for the relevant domain. */ + resp = RING_GET_RESPONSE(blk_ring, blk_ring->rsp_prod_pvt); + resp->id = id; + resp->operation = op; + resp->status = st; + blk_ring->rsp_prod_pvt++; + RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(blk_ring, notify); + + if (blk_ring->rsp_prod_pvt == blk_ring->req_cons) { + /* + * Tail check for pending requests. Allows frontend to avoid + * notifications if requests are already in flight (lower + * overheads and promotes batching). + */ + RING_FINAL_CHECK_FOR_REQUESTS(blk_ring, more_to_do); + } else if (RING_HAS_UNCONSUMED_REQUESTS(blk_ring)) { + more_to_do = 1; + + } + spin_unlock_irqrestore(&blkif->blk_ring_lock, flags); + if (more_to_do) + blkif_notify_work(blkif); + if (notify) + notify_remote_via_irq(blkif->irq); +} + +static int __init blkif_init(void) +{ + int i,ret,blktap_dir; + tap_blkif_t *info; + + if (!is_running_on_xen()) + return -ENODEV; + + INIT_LIST_HEAD(&pending_free); + for(i = 0; i < 2; i++) { + ret = req_increase(); + if (ret) + break; + } + if (i == 0) + return ret; + + tap_blkif_interface_init(); + + alloc_pending_reqs = 0; + + tap_blkif_xenbus_init(); + + /*Create the blktap devices, but do not map memory or waitqueue*/ + for(i = 0; i < MAX_TAP_DEV; i++) translate_domid[i].domid = 0xFFFF; + + ret = register_chrdev(BLKTAP_DEV_MAJOR,"blktap",&blktap_fops); + blktap_dir = devfs_mk_dir(NULL, "xen", 0, NULL); + + if ( (ret < 0)||(blktap_dir < 0) ) { + WPRINTK("Couldn't register /dev/xen/blktap\n"); + return -ENOMEM; + } + + for(i = 0; i < MAX_TAP_DEV; i++ ) { + info = tapfds[i] = kzalloc(sizeof(tap_blkif_t),GFP_KERNEL); + if(tapfds[i] == NULL) return -ENOMEM; + info->minor = i; + info->pid = 0; + info->blkif = NULL; + + ret = devfs_mk_cdev(MKDEV(BLKTAP_DEV_MAJOR, i), + S_IFCHR|S_IRUGO|S_IWUSR, "xen/blktap%d", i); + + if(ret != 0) return -ENOMEM; + info->dev_pending = info->dev_inuse = 0; + + DPRINTK("Created misc_dev [/dev/xen/blktap%d]\n",i); + } + + DPRINTK("Blktap device successfully created\n"); + + return 0; +} + +module_init(blkif_init); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/console/Makefile b/drivers/xen/console/Makefile new file mode 100644 index 000000000..35de3e933 --- /dev/null +++ b/drivers/xen/console/Makefile @@ -0,0 +1,2 @@ + +obj-y := console.o xencons_ring.o diff --git a/drivers/xen/console/console.c b/drivers/xen/console/console.c new file mode 100644 index 000000000..ec7dd9e58 --- /dev/null +++ b/drivers/xen/console/console.c @@ -0,0 +1,644 @@ +/****************************************************************************** + * console.c + * + * Virtual console driver. + * + * Copyright (c) 2002-2004, K A Fraser. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +/* + * Modes: + * 'xencons=off' [XC_OFF]: Console is disabled. + * 'xencons=tty' [XC_TTY]: Console attached to '/dev/tty[0-9]+'. + * 'xencons=ttyS' [XC_SERIAL]: Console attached to '/dev/ttyS[0-9]+'. + * [XC_DEFAULT]: DOM0 -> XC_SERIAL ; all others -> XC_TTY. + * + * NB. In mode XC_TTY, we create dummy consoles for tty2-63. This suppresses + * warnings from standard distro startup scripts. + */ +static enum { XC_OFF, XC_DEFAULT, XC_TTY, XC_SERIAL } xc_mode = XC_DEFAULT; +static int xc_num = -1; + +#ifdef CONFIG_MAGIC_SYSRQ +static unsigned long sysrq_requested; +extern int sysrq_enabled; +#endif + +static int __init xencons_setup(char *str) +{ + char *q; + int n; + + if (!strncmp(str, "ttyS", 4)) + xc_mode = XC_SERIAL; + else if (!strncmp(str, "tty", 3)) + xc_mode = XC_TTY; + else if (!strncmp(str, "off", 3)) + xc_mode = XC_OFF; + + switch (xc_mode) { + case XC_SERIAL: + n = simple_strtol(str+4, &q, 10); + if (q > (str + 4)) + xc_num = n; + break; + case XC_TTY: + n = simple_strtol(str+3, &q, 10); + if (q > (str + 3)) + xc_num = n; + break; + default: + break; + } + + return 1; +} +__setup("xencons=", xencons_setup); + +/* The kernel and user-land drivers share a common transmit buffer. */ +static unsigned int wbuf_size = 4096; +#define WBUF_MASK(_i) ((_i)&(wbuf_size-1)) +static char *wbuf; +static unsigned int wc, wp; /* write_cons, write_prod */ + +static int __init xencons_bufsz_setup(char *str) +{ + unsigned int goal; + goal = simple_strtoul(str, NULL, 0); + if (goal) { + goal = roundup_pow_of_two(goal); + if (wbuf_size < goal) + wbuf_size = goal; + } + return 1; +} +__setup("xencons_bufsz=", xencons_bufsz_setup); + +/* This lock protects accesses to the common transmit buffer. */ +static DEFINE_SPINLOCK(xencons_lock); + +/* Common transmit-kick routine. */ +static void __xencons_tx_flush(void); + +static struct tty_driver *xencons_driver; + +/******************** Kernel console driver ********************************/ + +static void kcons_write(struct console *c, const char *s, unsigned int count) +{ + int i = 0; + unsigned long flags; + + spin_lock_irqsave(&xencons_lock, flags); + + while (i < count) { + for (; i < count; i++) { + if ((wp - wc) >= (wbuf_size - 1)) + break; + if ((wbuf[WBUF_MASK(wp++)] = s[i]) == '\n') + wbuf[WBUF_MASK(wp++)] = '\r'; + } + + __xencons_tx_flush(); + } + + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static void kcons_write_dom0(struct console *c, const char *s, unsigned int count) +{ + + while (count > 0) { + int rc; + rc = HYPERVISOR_console_io( CONSOLEIO_write, count, (char *)s); + if (rc <= 0) + break; + count -= rc; + s += rc; + } +} + +static struct tty_driver *kcons_device(struct console *c, int *index) +{ + *index = 0; + return xencons_driver; +} + +static struct console kcons_info = { + .device = kcons_device, + .flags = CON_PRINTBUFFER | CON_ENABLED, + .index = -1, +}; + +#define __RETCODE 0 +static int __init xen_console_init(void) +{ + if (!is_running_on_xen()) + return __RETCODE; + + if (xen_start_info->flags & SIF_INITDOMAIN) { + if (xc_mode == XC_DEFAULT) + xc_mode = XC_SERIAL; + kcons_info.write = kcons_write_dom0; + } else { + if (xc_mode == XC_DEFAULT) + xc_mode = XC_TTY; + kcons_info.write = kcons_write; + } + + switch (xc_mode) { + case XC_SERIAL: + strcpy(kcons_info.name, "ttyS"); + if (xc_num == -1) + xc_num = 0; + break; + + case XC_TTY: + strcpy(kcons_info.name, "tty"); + if (xc_num == -1) + xc_num = 1; + break; + + default: + return __RETCODE; + } + + wbuf = alloc_bootmem(wbuf_size); + + register_console(&kcons_info); + + return __RETCODE; +} +console_initcall(xen_console_init); + +/*** Useful function for console debugging -- goes straight to Xen. ***/ +asmlinkage int xprintk(const char *fmt, ...) +{ + va_list args; + int printk_len; + static char printk_buf[1024]; + + /* Emit the output into the temporary buffer */ + va_start(args, fmt); + printk_len = vsnprintf(printk_buf, sizeof(printk_buf), fmt, args); + va_end(args); + + /* Send the processed output directly to Xen. */ + kcons_write_dom0(NULL, printk_buf, printk_len); + + return 0; +} + +/*** Forcibly flush console data before dying. ***/ +void xencons_force_flush(void) +{ + int sz; + + /* Emergency console is synchronous, so there's nothing to flush. */ + if (xen_start_info->flags & SIF_INITDOMAIN) + return; + + /* Spin until console data is flushed through to the daemon. */ + while (wc != wp) { + int sent = 0; + if ((sz = wp - wc) == 0) + continue; + sent = xencons_ring_send(&wbuf[WBUF_MASK(wc)], sz); + if (sent > 0) + wc += sent; + } +} + + +/******************** User-space console driver (/dev/console) ************/ + +#define DRV(_d) (_d) +#define DUMMY_TTY(_tty) ((xc_mode != XC_SERIAL) && \ + ((_tty)->index != (xc_num - 1))) + +static struct termios *xencons_termios[MAX_NR_CONSOLES]; +static struct termios *xencons_termios_locked[MAX_NR_CONSOLES]; +static struct tty_struct *xencons_tty; +static int xencons_priv_irq; +static char x_char; + +void xencons_rx(char *buf, unsigned len, struct pt_regs *regs) +{ + int i; + unsigned long flags; + + spin_lock_irqsave(&xencons_lock, flags); + if (xencons_tty == NULL) + goto out; + + for (i = 0; i < len; i++) { +#ifdef CONFIG_MAGIC_SYSRQ + if (sysrq_enabled) { + if (buf[i] == '\x0f') { /* ^O */ + sysrq_requested = jiffies; + continue; /* don't print the sysrq key */ + } else if (sysrq_requested) { + unsigned long sysrq_timeout = + sysrq_requested + HZ*2; + sysrq_requested = 0; + if (time_before(jiffies, sysrq_timeout)) { + spin_unlock_irqrestore( + &xencons_lock, flags); + handle_sysrq( + buf[i], regs, xencons_tty); + spin_lock_irqsave( + &xencons_lock, flags); + continue; + } + } + } +#endif + tty_insert_flip_char(xencons_tty, buf[i], 0); + } + tty_flip_buffer_push(xencons_tty); + + out: + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static void __xencons_tx_flush(void) +{ + int sent, sz, work_done = 0; + + if (x_char) { + if (xen_start_info->flags & SIF_INITDOMAIN) + kcons_write_dom0(NULL, &x_char, 1); + else + while (x_char) + if (xencons_ring_send(&x_char, 1) == 1) + break; + x_char = 0; + work_done = 1; + } + + while (wc != wp) { + sz = wp - wc; + if (sz > (wbuf_size - WBUF_MASK(wc))) + sz = wbuf_size - WBUF_MASK(wc); + if (xen_start_info->flags & SIF_INITDOMAIN) { + kcons_write_dom0(NULL, &wbuf[WBUF_MASK(wc)], sz); + wc += sz; + } else { + sent = xencons_ring_send(&wbuf[WBUF_MASK(wc)], sz); + if (sent == 0) + break; + wc += sent; + } + work_done = 1; + } + + if (work_done && (xencons_tty != NULL)) { + wake_up_interruptible(&xencons_tty->write_wait); + if ((xencons_tty->flags & (1 << TTY_DO_WRITE_WAKEUP)) && + (xencons_tty->ldisc.write_wakeup != NULL)) + (xencons_tty->ldisc.write_wakeup)(xencons_tty); + } +} + +void xencons_tx(void) +{ + unsigned long flags; + + spin_lock_irqsave(&xencons_lock, flags); + __xencons_tx_flush(); + spin_unlock_irqrestore(&xencons_lock, flags); +} + +/* Privileged receive callback and transmit kicker. */ +static irqreturn_t xencons_priv_interrupt(int irq, void *dev_id, + struct pt_regs *regs) +{ + static char rbuf[16]; + int l; + + while ((l = HYPERVISOR_console_io(CONSOLEIO_read, 16, rbuf)) > 0) + xencons_rx(rbuf, l, regs); + + xencons_tx(); + + return IRQ_HANDLED; +} + +static int xencons_write_room(struct tty_struct *tty) +{ + return wbuf_size - (wp - wc); +} + +static int xencons_chars_in_buffer(struct tty_struct *tty) +{ + return wp - wc; +} + +static void xencons_send_xchar(struct tty_struct *tty, char ch) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return; + + spin_lock_irqsave(&xencons_lock, flags); + x_char = ch; + __xencons_tx_flush(); + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static void xencons_throttle(struct tty_struct *tty) +{ + if (DUMMY_TTY(tty)) + return; + + if (I_IXOFF(tty)) + xencons_send_xchar(tty, STOP_CHAR(tty)); +} + +static void xencons_unthrottle(struct tty_struct *tty) +{ + if (DUMMY_TTY(tty)) + return; + + if (I_IXOFF(tty)) { + if (x_char != 0) + x_char = 0; + else + xencons_send_xchar(tty, START_CHAR(tty)); + } +} + +static void xencons_flush_buffer(struct tty_struct *tty) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return; + + spin_lock_irqsave(&xencons_lock, flags); + wc = wp = 0; + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static inline int __xencons_put_char(int ch) +{ + char _ch = (char)ch; + if ((wp - wc) == wbuf_size) + return 0; + wbuf[WBUF_MASK(wp++)] = _ch; + return 1; +} + +static int xencons_write( + struct tty_struct *tty, + const unsigned char *buf, + int count) +{ + int i; + unsigned long flags; + + if (DUMMY_TTY(tty)) + return count; + + spin_lock_irqsave(&xencons_lock, flags); + + for (i = 0; i < count; i++) + if (!__xencons_put_char(buf[i])) + break; + + if (i != 0) + __xencons_tx_flush(); + + spin_unlock_irqrestore(&xencons_lock, flags); + + return i; +} + +static void xencons_put_char(struct tty_struct *tty, u_char ch) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return; + + spin_lock_irqsave(&xencons_lock, flags); + (void)__xencons_put_char(ch); + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static void xencons_flush_chars(struct tty_struct *tty) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return; + + spin_lock_irqsave(&xencons_lock, flags); + __xencons_tx_flush(); + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static void xencons_wait_until_sent(struct tty_struct *tty, int timeout) +{ + unsigned long orig_jiffies = jiffies; + + if (DUMMY_TTY(tty)) + return; + + while (DRV(tty->driver)->chars_in_buffer(tty)) { + set_current_state(TASK_INTERRUPTIBLE); + schedule_timeout(1); + if (signal_pending(current)) + break; + if (timeout && time_after(jiffies, orig_jiffies + timeout)) + break; + } + + set_current_state(TASK_RUNNING); +} + +static int xencons_open(struct tty_struct *tty, struct file *filp) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return 0; + + spin_lock_irqsave(&xencons_lock, flags); + tty->driver_data = NULL; + if (xencons_tty == NULL) + xencons_tty = tty; + __xencons_tx_flush(); + spin_unlock_irqrestore(&xencons_lock, flags); + + return 0; +} + +static void xencons_close(struct tty_struct *tty, struct file *filp) +{ + unsigned long flags; + + if (DUMMY_TTY(tty)) + return; + + mutex_lock(&tty_mutex); + + if (tty->count != 1) { + mutex_unlock(&tty_mutex); + return; + } + + /* Prevent other threads from re-opening this tty. */ + set_bit(TTY_CLOSING, &tty->flags); + mutex_unlock(&tty_mutex); + + tty->closing = 1; + tty_wait_until_sent(tty, 0); + if (DRV(tty->driver)->flush_buffer != NULL) + DRV(tty->driver)->flush_buffer(tty); + if (tty->ldisc.flush_buffer != NULL) + tty->ldisc.flush_buffer(tty); + tty->closing = 0; + spin_lock_irqsave(&xencons_lock, flags); + xencons_tty = NULL; + spin_unlock_irqrestore(&xencons_lock, flags); +} + +static struct tty_operations xencons_ops = { + .open = xencons_open, + .close = xencons_close, + .write = xencons_write, + .write_room = xencons_write_room, + .put_char = xencons_put_char, + .flush_chars = xencons_flush_chars, + .chars_in_buffer = xencons_chars_in_buffer, + .send_xchar = xencons_send_xchar, + .flush_buffer = xencons_flush_buffer, + .throttle = xencons_throttle, + .unthrottle = xencons_unthrottle, + .wait_until_sent = xencons_wait_until_sent, +}; + +static int __init xencons_init(void) +{ + int rc; + + if (!is_running_on_xen()) + return -ENODEV; + + if (xc_mode == XC_OFF) + return 0; + + xencons_ring_init(); + + xencons_driver = alloc_tty_driver((xc_mode == XC_SERIAL) ? + 1 : MAX_NR_CONSOLES); + if (xencons_driver == NULL) + return -ENOMEM; + + DRV(xencons_driver)->name = "xencons"; + DRV(xencons_driver)->major = TTY_MAJOR; + DRV(xencons_driver)->type = TTY_DRIVER_TYPE_SERIAL; + DRV(xencons_driver)->subtype = SERIAL_TYPE_NORMAL; + DRV(xencons_driver)->init_termios = tty_std_termios; + DRV(xencons_driver)->flags = + TTY_DRIVER_REAL_RAW | + TTY_DRIVER_RESET_TERMIOS; + DRV(xencons_driver)->termios = xencons_termios; + DRV(xencons_driver)->termios_locked = xencons_termios_locked; + + if (xc_mode == XC_SERIAL) { + DRV(xencons_driver)->name = "ttyS"; + DRV(xencons_driver)->minor_start = 64 + xc_num; + DRV(xencons_driver)->name_base = 0 + xc_num; + } else { + DRV(xencons_driver)->name = "tty"; + DRV(xencons_driver)->minor_start = 1; + DRV(xencons_driver)->name_base = 1; + } + + tty_set_operations(xencons_driver, &xencons_ops); + + if ((rc = tty_register_driver(DRV(xencons_driver))) != 0) { + printk("WARNING: Failed to register Xen virtual " + "console driver as '%s%d'\n", + DRV(xencons_driver)->name, + DRV(xencons_driver)->name_base); + put_tty_driver(xencons_driver); + xencons_driver = NULL; + return rc; + } + + if (xen_start_info->flags & SIF_INITDOMAIN) { + xencons_priv_irq = bind_virq_to_irqhandler( + VIRQ_CONSOLE, + 0, + xencons_priv_interrupt, + 0, + "console", + NULL); + BUG_ON(xencons_priv_irq < 0); + } + + printk("Xen virtual console successfully installed as %s%d\n", + DRV(xencons_driver)->name, xc_num); + + return 0; +} + +module_init(xencons_init); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/evtchn/Makefile b/drivers/xen/evtchn/Makefile new file mode 100644 index 000000000..7b082a081 --- /dev/null +++ b/drivers/xen/evtchn/Makefile @@ -0,0 +1,2 @@ + +obj-y := evtchn.o diff --git a/drivers/xen/evtchn/evtchn.c b/drivers/xen/evtchn/evtchn.c new file mode 100644 index 000000000..76bfab82e --- /dev/null +++ b/drivers/xen/evtchn/evtchn.c @@ -0,0 +1,458 @@ +/****************************************************************************** + * evtchn.c + * + * Driver for receiving and demuxing event-channel signals. + * + * Copyright (c) 2004-2005, K A Fraser + * Multi-process extensions Copyright (c) 2004, Steven Smith + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +struct per_user_data { + /* Notification ring, accessed via /dev/xen/evtchn. */ +#define EVTCHN_RING_SIZE (PAGE_SIZE / sizeof(evtchn_port_t)) +#define EVTCHN_RING_MASK(_i) ((_i)&(EVTCHN_RING_SIZE-1)) + evtchn_port_t *ring; + unsigned int ring_cons, ring_prod, ring_overflow; + + /* Processes wait on this queue when ring is empty. */ + wait_queue_head_t evtchn_wait; + struct fasync_struct *evtchn_async_queue; +}; + +/* Who's bound to each port? */ +static struct per_user_data *port_user[NR_EVENT_CHANNELS]; +static spinlock_t port_user_lock; + +void evtchn_device_upcall(int port) +{ + struct per_user_data *u; + + spin_lock(&port_user_lock); + + mask_evtchn(port); + clear_evtchn(port); + + if ((u = port_user[port]) != NULL) { + if ((u->ring_prod - u->ring_cons) < EVTCHN_RING_SIZE) { + u->ring[EVTCHN_RING_MASK(u->ring_prod)] = port; + if (u->ring_cons == u->ring_prod++) { + wake_up_interruptible(&u->evtchn_wait); + kill_fasync(&u->evtchn_async_queue, + SIGIO, POLL_IN); + } + } else { + u->ring_overflow = 1; + } + } + + spin_unlock(&port_user_lock); +} + +static ssize_t evtchn_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + int rc; + unsigned int c, p, bytes1 = 0, bytes2 = 0; + struct per_user_data *u = file->private_data; + + /* Whole number of ports. */ + count &= ~(sizeof(evtchn_port_t)-1); + + if (count == 0) + return 0; + + if (count > PAGE_SIZE) + count = PAGE_SIZE; + + for (;;) { + if (u->ring_overflow) + return -EFBIG; + + if ((c = u->ring_cons) != (p = u->ring_prod)) + break; + + if (file->f_flags & O_NONBLOCK) + return -EAGAIN; + + rc = wait_event_interruptible( + u->evtchn_wait, u->ring_cons != u->ring_prod); + if (rc) + return rc; + } + + /* Byte lengths of two chunks. Chunk split (if any) is at ring wrap. */ + if (((c ^ p) & EVTCHN_RING_SIZE) != 0) { + bytes1 = (EVTCHN_RING_SIZE - EVTCHN_RING_MASK(c)) * + sizeof(evtchn_port_t); + bytes2 = EVTCHN_RING_MASK(p) * sizeof(evtchn_port_t); + } else { + bytes1 = (p - c) * sizeof(evtchn_port_t); + bytes2 = 0; + } + + /* Truncate chunks according to caller's maximum byte count. */ + if (bytes1 > count) { + bytes1 = count; + bytes2 = 0; + } else if ((bytes1 + bytes2) > count) { + bytes2 = count - bytes1; + } + + if (copy_to_user(buf, &u->ring[EVTCHN_RING_MASK(c)], bytes1) || + ((bytes2 != 0) && + copy_to_user(&buf[bytes1], &u->ring[0], bytes2))) + return -EFAULT; + + u->ring_cons += (bytes1 + bytes2) / sizeof(evtchn_port_t); + + return bytes1 + bytes2; +} + +static ssize_t evtchn_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + int rc, i; + evtchn_port_t *kbuf = (evtchn_port_t *)__get_free_page(GFP_KERNEL); + struct per_user_data *u = file->private_data; + + if (kbuf == NULL) + return -ENOMEM; + + /* Whole number of ports. */ + count &= ~(sizeof(evtchn_port_t)-1); + + if (count == 0) { + rc = 0; + goto out; + } + + if (count > PAGE_SIZE) + count = PAGE_SIZE; + + if (copy_from_user(kbuf, buf, count) != 0) { + rc = -EFAULT; + goto out; + } + + spin_lock_irq(&port_user_lock); + for (i = 0; i < (count/sizeof(evtchn_port_t)); i++) + if ((kbuf[i] < NR_EVENT_CHANNELS) && (port_user[kbuf[i]] == u)) + unmask_evtchn(kbuf[i]); + spin_unlock_irq(&port_user_lock); + + rc = count; + + out: + free_page((unsigned long)kbuf); + return rc; +} + +static void evtchn_bind_to_user(struct per_user_data *u, int port) +{ + spin_lock_irq(&port_user_lock); + BUG_ON(port_user[port] != NULL); + port_user[port] = u; + unmask_evtchn(port); + spin_unlock_irq(&port_user_lock); +} + +static int evtchn_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long arg) +{ + int rc; + struct per_user_data *u = file->private_data; + void __user *uarg = (void __user *) arg; + + switch (cmd) { + case IOCTL_EVTCHN_BIND_VIRQ: { + struct ioctl_evtchn_bind_virq bind; + struct evtchn_bind_virq bind_virq; + + rc = -EFAULT; + if (copy_from_user(&bind, uarg, sizeof(bind))) + break; + + bind_virq.virq = bind.virq; + bind_virq.vcpu = 0; + rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq, + &bind_virq); + if (rc != 0) + break; + + rc = bind_virq.port; + evtchn_bind_to_user(u, rc); + break; + } + + case IOCTL_EVTCHN_BIND_INTERDOMAIN: { + struct ioctl_evtchn_bind_interdomain bind; + struct evtchn_bind_interdomain bind_interdomain; + + rc = -EFAULT; + if (copy_from_user(&bind, uarg, sizeof(bind))) + break; + + bind_interdomain.remote_dom = bind.remote_domain; + bind_interdomain.remote_port = bind.remote_port; + rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain, + &bind_interdomain); + if (rc != 0) + break; + + rc = bind_interdomain.local_port; + evtchn_bind_to_user(u, rc); + break; + } + + case IOCTL_EVTCHN_BIND_UNBOUND_PORT: { + struct ioctl_evtchn_bind_unbound_port bind; + struct evtchn_alloc_unbound alloc_unbound; + + rc = -EFAULT; + if (copy_from_user(&bind, uarg, sizeof(bind))) + break; + + alloc_unbound.dom = DOMID_SELF; + alloc_unbound.remote_dom = bind.remote_domain; + rc = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound, + &alloc_unbound); + if (rc != 0) + break; + + rc = alloc_unbound.port; + evtchn_bind_to_user(u, rc); + break; + } + + case IOCTL_EVTCHN_UNBIND: { + struct ioctl_evtchn_unbind unbind; + struct evtchn_close close; + int ret; + + rc = -EFAULT; + if (copy_from_user(&unbind, uarg, sizeof(unbind))) + break; + + rc = -EINVAL; + if (unbind.port >= NR_EVENT_CHANNELS) + break; + + spin_lock_irq(&port_user_lock); + + rc = -ENOTCONN; + if (port_user[unbind.port] != u) { + spin_unlock_irq(&port_user_lock); + break; + } + + port_user[unbind.port] = NULL; + mask_evtchn(unbind.port); + + spin_unlock_irq(&port_user_lock); + + close.port = unbind.port; + ret = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close); + BUG_ON(ret); + + rc = 0; + break; + } + + case IOCTL_EVTCHN_NOTIFY: { + struct ioctl_evtchn_notify notify; + + rc = -EFAULT; + if (copy_from_user(¬ify, uarg, sizeof(notify))) + break; + + if (notify.port >= NR_EVENT_CHANNELS) { + rc = -EINVAL; + } else if (port_user[notify.port] != u) { + rc = -ENOTCONN; + } else { + notify_remote_via_evtchn(notify.port); + rc = 0; + } + break; + } + + case IOCTL_EVTCHN_RESET: { + /* Initialise the ring to empty. Clear errors. */ + spin_lock_irq(&port_user_lock); + u->ring_cons = u->ring_prod = u->ring_overflow = 0; + spin_unlock_irq(&port_user_lock); + rc = 0; + break; + } + + default: + rc = -ENOSYS; + break; + } + + return rc; +} + +static unsigned int evtchn_poll(struct file *file, poll_table *wait) +{ + unsigned int mask = POLLOUT | POLLWRNORM; + struct per_user_data *u = file->private_data; + + poll_wait(file, &u->evtchn_wait, wait); + if (u->ring_cons != u->ring_prod) + mask |= POLLIN | POLLRDNORM; + if (u->ring_overflow) + mask = POLLERR; + return mask; +} + +static int evtchn_fasync(int fd, struct file *filp, int on) +{ + struct per_user_data *u = filp->private_data; + return fasync_helper(fd, filp, on, &u->evtchn_async_queue); +} + +static int evtchn_open(struct inode *inode, struct file *filp) +{ + struct per_user_data *u; + + if ((u = kmalloc(sizeof(*u), GFP_KERNEL)) == NULL) + return -ENOMEM; + + memset(u, 0, sizeof(*u)); + init_waitqueue_head(&u->evtchn_wait); + + u->ring = (evtchn_port_t *)__get_free_page(GFP_KERNEL); + if (u->ring == NULL) { + kfree(u); + return -ENOMEM; + } + + filp->private_data = u; + + return 0; +} + +static int evtchn_release(struct inode *inode, struct file *filp) +{ + int i; + struct per_user_data *u = filp->private_data; + struct evtchn_close close; + + spin_lock_irq(&port_user_lock); + + free_page((unsigned long)u->ring); + + for (i = 0; i < NR_EVENT_CHANNELS; i++) { + int ret; + if (port_user[i] != u) + continue; + + port_user[i] = NULL; + mask_evtchn(i); + + close.port = i; + ret = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close); + BUG_ON(ret); + } + + spin_unlock_irq(&port_user_lock); + + kfree(u); + + return 0; +} + +static struct file_operations evtchn_fops = { + .owner = THIS_MODULE, + .read = evtchn_read, + .write = evtchn_write, + .ioctl = evtchn_ioctl, + .poll = evtchn_poll, + .fasync = evtchn_fasync, + .open = evtchn_open, + .release = evtchn_release, +}; + +static struct miscdevice evtchn_miscdev = { + .minor = EVTCHN_MINOR, + .name = "evtchn", + .fops = &evtchn_fops, + .devfs_name = "misc/evtchn", +}; + +static int __init evtchn_init(void) +{ + int err; + + if (!is_running_on_xen()) + return -ENODEV; + + spin_lock_init(&port_user_lock); + memset(port_user, 0, sizeof(port_user)); + + /* Create '/dev/misc/evtchn'. */ + err = misc_register(&evtchn_miscdev); + if (err != 0) { + printk(KERN_ALERT "Could not register /dev/misc/evtchn\n"); + return err; + } + + printk("Event-channel device installed.\n"); + + return 0; +} + +static void evtchn_cleanup(void) +{ + misc_deregister(&evtchn_miscdev); +} + +module_init(evtchn_init); +module_exit(evtchn_cleanup); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/netback/Makefile b/drivers/xen/netback/Makefile new file mode 100644 index 000000000..d5d232837 --- /dev/null +++ b/drivers/xen/netback/Makefile @@ -0,0 +1,5 @@ +obj-$(CONFIG_XEN_NETDEV_BACKEND) := netbk.o +obj-$(CONFIG_XEN_NETDEV_LOOPBACK) += netloop.o + +netbk-y := netback.o xenbus.o interface.o +netloop-y := loopback.o diff --git a/drivers/xen/netback/common.h b/drivers/xen/netback/common.h new file mode 100644 index 000000000..c5ec30071 --- /dev/null +++ b/drivers/xen/netback/common.h @@ -0,0 +1,138 @@ +/****************************************************************************** + * arch/xen/drivers/netif/backend/common.h + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#ifndef __NETIF__BACKEND__COMMON_H__ +#define __NETIF__BACKEND__COMMON_H__ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define DPRINTK(_f, _a...) \ + pr_debug("(file=%s, line=%d) " _f, \ + __FILE__ , __LINE__ , ## _a ) +#define IPRINTK(fmt, args...) \ + printk(KERN_INFO "xen_net: " fmt, ##args) +#define WPRINTK(fmt, args...) \ + printk(KERN_WARNING "xen_net: " fmt, ##args) + +typedef struct netif_st { + /* Unique identifier for this interface. */ + domid_t domid; + unsigned int handle; + + u8 fe_dev_addr[6]; + + /* Physical parameters of the comms window. */ + grant_handle_t tx_shmem_handle; + grant_ref_t tx_shmem_ref; + grant_handle_t rx_shmem_handle; + grant_ref_t rx_shmem_ref; + unsigned int evtchn; + unsigned int irq; + + /* The shared rings and indexes. */ + netif_tx_back_ring_t tx; + netif_rx_back_ring_t rx; + struct vm_struct *tx_comms_area; + struct vm_struct *rx_comms_area; + + /* Set of features that can be turned on in dev->features. */ + int features; + int can_queue; + + /* Allow netif_be_start_xmit() to peek ahead in the rx request ring. */ + RING_IDX rx_req_cons_peek; + + /* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */ + unsigned long credit_bytes; + unsigned long credit_usec; + unsigned long remaining_credit; + struct timer_list credit_timeout; + + /* Miscellaneous private stuff. */ + struct list_head list; /* scheduling list */ + atomic_t refcnt; + struct net_device *dev; + struct net_device_stats stats; + + wait_queue_head_t waiting_to_free; +} netif_t; + +#define NET_TX_RING_SIZE __RING_SIZE((netif_tx_sring_t *)0, PAGE_SIZE) +#define NET_RX_RING_SIZE __RING_SIZE((netif_rx_sring_t *)0, PAGE_SIZE) + +void netif_disconnect(netif_t *netif); + +netif_t *netif_alloc(domid_t domid, unsigned int handle, u8 be_mac[ETH_ALEN]); +int netif_map(netif_t *netif, unsigned long tx_ring_ref, + unsigned long rx_ring_ref, unsigned int evtchn); + +#define netif_get(_b) (atomic_inc(&(_b)->refcnt)) +#define netif_put(_b) \ + do { \ + if ( atomic_dec_and_test(&(_b)->refcnt) ) \ + wake_up(&(_b)->waiting_to_free); \ + } while (0) + +void netif_xenbus_init(void); + +void netif_schedule_work(netif_t *netif); +void netif_deschedule_work(netif_t *netif); + +int netif_be_start_xmit(struct sk_buff *skb, struct net_device *dev); +struct net_device_stats *netif_be_get_stats(struct net_device *dev); +irqreturn_t netif_be_int(int irq, void *dev_id, struct pt_regs *regs); + +static inline int netbk_can_queue(struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + return netif->can_queue; +} + +static inline int netbk_can_sg(struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + return netif->features & NETIF_F_SG; +} + +#endif /* __NETIF__BACKEND__COMMON_H__ */ diff --git a/drivers/xen/netback/interface.c b/drivers/xen/netback/interface.c new file mode 100644 index 000000000..d60b23b0f --- /dev/null +++ b/drivers/xen/netback/interface.c @@ -0,0 +1,339 @@ +/****************************************************************************** + * arch/xen/drivers/netif/backend/interface.c + * + * Network-device interface management. + * + * Copyright (c) 2004-2005, Keir Fraser + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "common.h" +#include +#include + +static void __netif_up(netif_t *netif) +{ + enable_irq(netif->irq); + netif_schedule_work(netif); +} + +static void __netif_down(netif_t *netif) +{ + disable_irq(netif->irq); + netif_deschedule_work(netif); +} + +static int net_open(struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + if (netif_carrier_ok(dev)) + __netif_up(netif); + return 0; +} + +static int net_close(struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + if (netif_carrier_ok(dev)) + __netif_down(netif); + return 0; +} + +static int netbk_change_mtu(struct net_device *dev, int mtu) +{ + int max = netbk_can_sg(dev) ? 65535 - ETH_HLEN : ETH_DATA_LEN; + + if (mtu > max) + return -EINVAL; + dev->mtu = mtu; + return 0; +} + +static int netbk_set_sg(struct net_device *dev, u32 data) +{ + if (data) { + netif_t *netif = netdev_priv(dev); + + if (!(netif->features & NETIF_F_SG)) + return -ENOSYS; + } + + return ethtool_op_set_sg(dev, data); +} + +static int netbk_set_tso(struct net_device *dev, u32 data) +{ + if (data) { + netif_t *netif = netdev_priv(dev); + + if (!(netif->features & NETIF_F_TSO)) + return -ENOSYS; + } + + return ethtool_op_set_tso(dev, data); +} + +static struct ethtool_ops network_ethtool_ops = +{ + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = ethtool_op_set_tx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = netbk_set_sg, + .get_tso = ethtool_op_get_tso, + .set_tso = netbk_set_tso, + .get_link = ethtool_op_get_link, +}; + +netif_t *netif_alloc(domid_t domid, unsigned int handle, u8 be_mac[ETH_ALEN]) +{ + int err = 0, i; + struct net_device *dev; + netif_t *netif; + char name[IFNAMSIZ] = {}; + + snprintf(name, IFNAMSIZ - 1, "vif%u.%u", domid, handle); + dev = alloc_netdev(sizeof(netif_t), name, ether_setup); + if (dev == NULL) { + DPRINTK("Could not create netif: out of memory\n"); + return ERR_PTR(-ENOMEM); + } + + netif_carrier_off(dev); + + netif = netdev_priv(dev); + memset(netif, 0, sizeof(*netif)); + netif->domid = domid; + netif->handle = handle; + atomic_set(&netif->refcnt, 1); + init_waitqueue_head(&netif->waiting_to_free); + netif->dev = dev; + + netif->credit_bytes = netif->remaining_credit = ~0UL; + netif->credit_usec = 0UL; + init_timer(&netif->credit_timeout); + + dev->hard_start_xmit = netif_be_start_xmit; + dev->get_stats = netif_be_get_stats; + dev->open = net_open; + dev->stop = net_close; + dev->change_mtu = netbk_change_mtu; + dev->features = NETIF_F_IP_CSUM; + + SET_ETHTOOL_OPS(dev, &network_ethtool_ops); + + /* + * Reduce default TX queuelen so that each guest interface only + * allows it to eat around 6.4MB of host memory. + */ + dev->tx_queue_len = 100; + + for (i = 0; i < ETH_ALEN; i++) + if (be_mac[i] != 0) + break; + if (i == ETH_ALEN) { + /* + * Initialise a dummy MAC address. We choose the numerically + * largest non-broadcast address to prevent the address getting + * stolen by an Ethernet bridge for STP purposes. + * (FE:FF:FF:FF:FF:FF) + */ + memset(dev->dev_addr, 0xFF, ETH_ALEN); + dev->dev_addr[0] &= ~0x01; + } else + memcpy(dev->dev_addr, be_mac, ETH_ALEN); + + rtnl_lock(); + err = register_netdevice(dev); + rtnl_unlock(); + if (err) { + DPRINTK("Could not register new net device %s: err=%d\n", + dev->name, err); + free_netdev(dev); + return ERR_PTR(err); + } + + DPRINTK("Successfully created netif\n"); + return netif; +} + +static int map_frontend_pages( + netif_t *netif, grant_ref_t tx_ring_ref, grant_ref_t rx_ring_ref) +{ + struct gnttab_map_grant_ref op; + int ret; + + gnttab_set_map_op(&op, (unsigned long)netif->tx_comms_area->addr, + GNTMAP_host_map, tx_ring_ref, netif->domid); + + lock_vm_area(netif->tx_comms_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1); + unlock_vm_area(netif->tx_comms_area); + BUG_ON(ret); + + if (op.status) { + DPRINTK(" Gnttab failure mapping tx_ring_ref!\n"); + return op.status; + } + + netif->tx_shmem_ref = tx_ring_ref; + netif->tx_shmem_handle = op.handle; + + gnttab_set_map_op(&op, (unsigned long)netif->rx_comms_area->addr, + GNTMAP_host_map, rx_ring_ref, netif->domid); + + lock_vm_area(netif->rx_comms_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1); + unlock_vm_area(netif->rx_comms_area); + BUG_ON(ret); + + if (op.status) { + DPRINTK(" Gnttab failure mapping rx_ring_ref!\n"); + return op.status; + } + + netif->rx_shmem_ref = rx_ring_ref; + netif->rx_shmem_handle = op.handle; + + return 0; +} + +static void unmap_frontend_pages(netif_t *netif) +{ + struct gnttab_unmap_grant_ref op; + int ret; + + gnttab_set_unmap_op(&op, (unsigned long)netif->tx_comms_area->addr, + GNTMAP_host_map, netif->tx_shmem_handle); + + lock_vm_area(netif->tx_comms_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1); + unlock_vm_area(netif->tx_comms_area); + BUG_ON(ret); + + gnttab_set_unmap_op(&op, (unsigned long)netif->rx_comms_area->addr, + GNTMAP_host_map, netif->rx_shmem_handle); + + lock_vm_area(netif->rx_comms_area); + ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1); + unlock_vm_area(netif->rx_comms_area); + BUG_ON(ret); +} + +int netif_map(netif_t *netif, unsigned long tx_ring_ref, + unsigned long rx_ring_ref, unsigned int evtchn) +{ + int err = -ENOMEM; + netif_tx_sring_t *txs; + netif_rx_sring_t *rxs; + struct evtchn_bind_interdomain bind_interdomain; + + /* Already connected through? */ + if (netif->irq) + return 0; + + netif->tx_comms_area = alloc_vm_area(PAGE_SIZE); + if (netif->tx_comms_area == NULL) + return -ENOMEM; + netif->rx_comms_area = alloc_vm_area(PAGE_SIZE); + if (netif->rx_comms_area == NULL) + goto err_rx; + + err = map_frontend_pages(netif, tx_ring_ref, rx_ring_ref); + if (err) + goto err_map; + + bind_interdomain.remote_dom = netif->domid; + bind_interdomain.remote_port = evtchn; + + err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain, + &bind_interdomain); + if (err) + goto err_hypervisor; + + netif->evtchn = bind_interdomain.local_port; + + netif->irq = bind_evtchn_to_irqhandler( + netif->evtchn, netif_be_int, 0, netif->dev->name, netif); + disable_irq(netif->irq); + + txs = (netif_tx_sring_t *)netif->tx_comms_area->addr; + BACK_RING_INIT(&netif->tx, txs, PAGE_SIZE); + + rxs = (netif_rx_sring_t *) + ((char *)netif->rx_comms_area->addr); + BACK_RING_INIT(&netif->rx, rxs, PAGE_SIZE); + + netif->rx_req_cons_peek = 0; + + netif_get(netif); + + rtnl_lock(); + netif_carrier_on(netif->dev); + if (netif_running(netif->dev)) + __netif_up(netif); + rtnl_unlock(); + + return 0; +err_hypervisor: + unmap_frontend_pages(netif); +err_map: + free_vm_area(netif->rx_comms_area); +err_rx: + free_vm_area(netif->tx_comms_area); + return err; +} + +static void netif_free(netif_t *netif) +{ + atomic_dec(&netif->refcnt); + wait_event(netif->waiting_to_free, atomic_read(&netif->refcnt) == 0); + + if (netif->irq) + unbind_from_irqhandler(netif->irq, netif); + + unregister_netdev(netif->dev); + + if (netif->tx.sring) { + unmap_frontend_pages(netif); + free_vm_area(netif->tx_comms_area); + free_vm_area(netif->rx_comms_area); + } + + free_netdev(netif->dev); +} + +void netif_disconnect(netif_t *netif) +{ + if (netif_carrier_ok(netif->dev)) { + rtnl_lock(); + netif_carrier_off(netif->dev); + if (netif_running(netif->dev)) + __netif_down(netif); + rtnl_unlock(); + netif_put(netif); + } + netif_free(netif); +} diff --git a/drivers/xen/netback/netback.c b/drivers/xen/netback/netback.c new file mode 100644 index 000000000..0316f3b05 --- /dev/null +++ b/drivers/xen/netback/netback.c @@ -0,0 +1,1345 @@ +/****************************************************************************** + * drivers/xen/netback/netback.c + * + * Back-end of the driver for virtual network devices. This portion of the + * driver exports a 'unified' network-device interface that can be accessed + * by any operating system that implements a compatible front end. A + * reference front-end implementation can be found in: + * drivers/xen/netfront/netfront.c + * + * Copyright (c) 2002-2005, K A Fraser + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include "common.h" +#include +#include + +/*#define NETBE_DEBUG_INTERRUPT*/ + +struct netbk_rx_meta { + skb_frag_t frag; + int id; +}; + +static void netif_idx_release(u16 pending_idx); +static void netif_page_release(struct page *page); +static void make_tx_response(netif_t *netif, + netif_tx_request_t *txp, + s8 st); +static netif_rx_response_t *make_rx_response(netif_t *netif, + u16 id, + s8 st, + u16 offset, + u16 size, + u16 flags); + +static void net_tx_action(unsigned long unused); +static DECLARE_TASKLET(net_tx_tasklet, net_tx_action, 0); + +static void net_rx_action(unsigned long unused); +static DECLARE_TASKLET(net_rx_tasklet, net_rx_action, 0); + +static struct timer_list net_timer; + +#define MAX_PENDING_REQS 256 + +static struct sk_buff_head rx_queue; +static multicall_entry_t rx_mcl[NET_RX_RING_SIZE+1]; +static mmu_update_t rx_mmu[NET_RX_RING_SIZE]; +static gnttab_transfer_t grant_rx_op[NET_RX_RING_SIZE]; +static unsigned char rx_notify[NR_IRQS]; + +static unsigned long mmap_vstart; +#define MMAP_VADDR(_req) (mmap_vstart + ((_req) * PAGE_SIZE)) + +#define PKT_PROT_LEN 64 + +static struct { + netif_tx_request_t req; + netif_t *netif; +} pending_tx_info[MAX_PENDING_REQS]; +static u16 pending_ring[MAX_PENDING_REQS]; +typedef unsigned int PEND_RING_IDX; +#define MASK_PEND_IDX(_i) ((_i)&(MAX_PENDING_REQS-1)) +static PEND_RING_IDX pending_prod, pending_cons; +#define NR_PENDING_REQS (MAX_PENDING_REQS - pending_prod + pending_cons) + +/* Freed TX SKBs get batched on this ring before return to pending_ring. */ +static u16 dealloc_ring[MAX_PENDING_REQS]; +static PEND_RING_IDX dealloc_prod, dealloc_cons; + +static struct sk_buff_head tx_queue; + +static grant_handle_t grant_tx_handle[MAX_PENDING_REQS]; +static gnttab_unmap_grant_ref_t tx_unmap_ops[MAX_PENDING_REQS]; +static gnttab_map_grant_ref_t tx_map_ops[MAX_PENDING_REQS]; + +static struct list_head net_schedule_list; +static spinlock_t net_schedule_list_lock; + +#define MAX_MFN_ALLOC 64 +static unsigned long mfn_list[MAX_MFN_ALLOC]; +static unsigned int alloc_index = 0; + +static inline unsigned long alloc_mfn(void) +{ + return mfn_list[--alloc_index]; +} + +static int check_mfn(int nr) +{ + struct xen_memory_reservation reservation = { + .extent_order = 0, + .domid = DOMID_SELF + }; + + if (likely(alloc_index >= nr)) + return 0; + + set_xen_guest_handle(reservation.extent_start, mfn_list + alloc_index); + reservation.nr_extents = MAX_MFN_ALLOC - alloc_index; + alloc_index += HYPERVISOR_memory_op(XENMEM_increase_reservation, + &reservation); + + return alloc_index >= nr ? 0 : -ENOMEM; +} + +static inline void maybe_schedule_tx_action(void) +{ + smp_mb(); + if ((NR_PENDING_REQS < (MAX_PENDING_REQS/2)) && + !list_empty(&net_schedule_list)) + tasklet_schedule(&net_tx_tasklet); +} + +/* + * A gross way of confirming the origin of an skb data page. The slab + * allocator abuses a field in the page struct to cache the kmem_cache_t ptr. + */ +static inline int is_xen_skb(struct sk_buff *skb) +{ + extern kmem_cache_t *skbuff_cachep; + kmem_cache_t *cp = (kmem_cache_t *)virt_to_page(skb->head)->lru.next; + return (cp == skbuff_cachep); +} + +static struct sk_buff *netbk_copy_skb(struct sk_buff *skb) +{ + struct skb_shared_info *ninfo; + struct sk_buff *nskb; + unsigned long offset; + int ret; + int len; + int headlen; + + nskb = alloc_skb(SKB_MAX_HEAD(0), GFP_ATOMIC); + if (unlikely(!nskb)) + goto err; + + skb_reserve(nskb, 16); + headlen = nskb->end - nskb->data; + if (headlen > skb_headlen(skb)) + headlen = skb_headlen(skb); + ret = skb_copy_bits(skb, 0, __skb_put(nskb, headlen), headlen); + BUG_ON(ret); + + ninfo = skb_shinfo(nskb); + ninfo->gso_size = skb_shinfo(skb)->gso_size; + ninfo->gso_type = skb_shinfo(skb)->gso_type; + + offset = headlen; + len = skb->len - headlen; + + nskb->len = skb->len; + nskb->data_len = len; + nskb->truesize += len; + + while (len) { + struct page *page; + int copy; + int zero; + + if (unlikely(ninfo->nr_frags >= MAX_SKB_FRAGS)) { + dump_stack(); + goto err_free; + } + + copy = len >= PAGE_SIZE ? PAGE_SIZE : len; + zero = len >= PAGE_SIZE ? 0 : __GFP_ZERO; + + page = alloc_page(GFP_ATOMIC | zero); + if (unlikely(!page)) + goto err_free; + + ret = skb_copy_bits(skb, offset, page_address(page), copy); + BUG_ON(ret); + + ninfo->frags[ninfo->nr_frags].page = page; + ninfo->frags[ninfo->nr_frags].page_offset = 0; + ninfo->frags[ninfo->nr_frags].size = copy; + ninfo->nr_frags++; + + offset += copy; + len -= copy; + } + + offset = nskb->data - skb->data; + + nskb->h.raw = skb->h.raw + offset; + nskb->nh.raw = skb->nh.raw + offset; + nskb->mac.raw = skb->mac.raw + offset; + + return nskb; + + err_free: + kfree_skb(nskb); + err: + return NULL; +} + +static inline int netbk_max_required_rx_slots(netif_t *netif) +{ + if (netif->features & (NETIF_F_SG|NETIF_F_TSO)) + return MAX_SKB_FRAGS + 2; /* header + extra_info + frags */ + return 1; /* all in one */ +} + +static inline int netbk_queue_full(netif_t *netif) +{ + RING_IDX peek = netif->rx_req_cons_peek; + RING_IDX needed = netbk_max_required_rx_slots(netif); + + return ((netif->rx.sring->req_prod - peek) < needed) || + ((netif->rx.rsp_prod_pvt + NET_RX_RING_SIZE - peek) < needed); +} + +int netif_be_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + + BUG_ON(skb->dev != dev); + + /* Drop the packet if the target domain has no receive buffers. */ + if (unlikely(!netif_running(dev) || !netif_carrier_ok(dev))) + goto drop; + + if (unlikely(netbk_queue_full(netif))) { + /* Not a BUG_ON() -- misbehaving netfront can trigger this. */ + if (netbk_can_queue(dev)) + DPRINTK("Queue full but not stopped!\n"); + goto drop; + } + + /* + * We do not copy the packet unless: + * 1. The data is shared; or + * 2. The data is not allocated from our special cache. + * 3. The data is fragmented. + */ + if (skb_cloned(skb) || skb_is_nonlinear(skb) || !is_xen_skb(skb)) { + struct sk_buff *nskb = netbk_copy_skb(skb); + if ( unlikely(nskb == NULL) ) + goto drop; + /* Copy only the header fields we use in this driver. */ + nskb->dev = skb->dev; + nskb->ip_summed = skb->ip_summed; + nskb->proto_data_valid = skb->proto_data_valid; + dev_kfree_skb(skb); + skb = nskb; + } + + netif->rx_req_cons_peek += skb_shinfo(skb)->nr_frags + 1 + + !!skb_shinfo(skb)->gso_size; + netif_get(netif); + + if (netbk_can_queue(dev) && netbk_queue_full(netif)) { + netif->rx.sring->req_event = netif->rx_req_cons_peek + + netbk_max_required_rx_slots(netif); + mb(); /* request notification /then/ check & stop the queue */ + if (netbk_queue_full(netif)) + netif_stop_queue(dev); + } + + skb_queue_tail(&rx_queue, skb); + tasklet_schedule(&net_rx_tasklet); + + return 0; + + drop: + netif->stats.tx_dropped++; + dev_kfree_skb(skb); + return 0; +} + +#if 0 +static void xen_network_done_notify(void) +{ + static struct net_device *eth0_dev = NULL; + if (unlikely(eth0_dev == NULL)) + eth0_dev = __dev_get_by_name("eth0"); + netif_rx_schedule(eth0_dev); +} +/* + * Add following to poll() function in NAPI driver (Tigon3 is example): + * if ( xen_network_done() ) + * tg3_enable_ints(tp); + */ +int xen_network_done(void) +{ + return skb_queue_empty(&rx_queue); +} +#endif + +static u16 netbk_gop_frag(netif_t *netif, struct page *page, int count, int i) +{ + multicall_entry_t *mcl = rx_mcl + count; + mmu_update_t *mmu = rx_mmu + count; + gnttab_transfer_t *gop = grant_rx_op + count; + netif_rx_request_t *req; + unsigned long old_mfn, new_mfn; + + old_mfn = virt_to_mfn(page_address(page)); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + new_mfn = alloc_mfn(); + + /* + * Set the new P2M table entry before reassigning + * the old data page. Heed the comment in + * pgtable-2level.h:pte_page(). :-) + */ + set_phys_to_machine(page_to_pfn(page), new_mfn); + + MULTI_update_va_mapping(mcl, (unsigned long)page_address(page), + pfn_pte_ma(new_mfn, PAGE_KERNEL), 0); + + mmu->ptr = ((maddr_t)new_mfn << PAGE_SHIFT) | + MMU_MACHPHYS_UPDATE; + mmu->val = page_to_pfn(page); + } + + req = RING_GET_REQUEST(&netif->rx, netif->rx.req_cons + i); + gop->mfn = old_mfn; + gop->domid = netif->domid; + gop->ref = req->gref; + return req->id; +} + +static void netbk_gop_skb(struct sk_buff *skb, struct netbk_rx_meta *meta, + int count) +{ + netif_t *netif = netdev_priv(skb->dev); + int nr_frags = skb_shinfo(skb)->nr_frags; + int i; + int extra; + + meta[count].frag.page_offset = skb_shinfo(skb)->gso_type; + meta[count].frag.size = skb_shinfo(skb)->gso_size; + extra = !!meta[count].frag.size + 1; + + for (i = 0; i < nr_frags; i++) { + meta[++count].frag = skb_shinfo(skb)->frags[i]; + meta[count].id = netbk_gop_frag(netif, meta[count].frag.page, + count, i + extra); + } + + /* + * This must occur at the end to ensure that we don't trash + * skb_shinfo until we're done. + */ + meta[count - nr_frags].id = netbk_gop_frag(netif, + virt_to_page(skb->data), + count - nr_frags, 0); + netif->rx.req_cons += nr_frags + extra; +} + +static inline void netbk_free_pages(int nr_frags, struct netbk_rx_meta *meta) +{ + int i; + + for (i = 0; i < nr_frags; i++) + put_page(meta[i].frag.page); +} + +static int netbk_check_gop(int nr_frags, domid_t domid, int count) +{ + multicall_entry_t *mcl = rx_mcl + count; + gnttab_transfer_t *gop = grant_rx_op + count; + int status = NETIF_RSP_OKAY; + int i; + + for (i = 0; i <= nr_frags; i++) { + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* The update_va_mapping() must not fail. */ + BUG_ON(mcl->result != 0); + mcl++; + } + + /* Check the reassignment error code. */ + if (gop->status != 0) { + DPRINTK("Bad status %d from grant transfer to DOM%u\n", + gop->status, domid); + /* + * Page no longer belongs to us unless GNTST_bad_page, + * but that should be a fatal error anyway. + */ + BUG_ON(gop->status == GNTST_bad_page); + status = NETIF_RSP_ERROR; + } + gop++; + } + + return status; +} + +static void netbk_add_frag_responses(netif_t *netif, int status, + struct netbk_rx_meta *meta, int nr_frags) +{ + int i; + + for (i = 0; i < nr_frags; i++) { + int id = meta[i].id; + int flags = (i == nr_frags - 1) ? 0 : NETRXF_more_data; + + make_rx_response(netif, id, status, meta[i].frag.page_offset, + meta[i].frag.size, flags); + } +} + +static void net_rx_action(unsigned long unused) +{ + netif_t *netif = NULL; + s8 status; + u16 id, irq, flags; + netif_rx_response_t *resp; + struct netif_extra_info *extra; + multicall_entry_t *mcl; + struct sk_buff_head rxq; + struct sk_buff *skb; + int notify_nr = 0; + int ret; + int nr_frags; + int count; + + /* + * Putting hundreds of bytes on the stack is considered rude. + * Static works because a tasklet can only be on one CPU at any time. + */ + static u16 notify_list[NET_RX_RING_SIZE]; + static struct netbk_rx_meta meta[NET_RX_RING_SIZE]; + + skb_queue_head_init(&rxq); + + count = 0; + + while ((skb = skb_dequeue(&rx_queue)) != NULL) { + nr_frags = skb_shinfo(skb)->nr_frags; + *(int *)skb->cb = nr_frags; + + if (!xen_feature(XENFEAT_auto_translated_physmap) && + check_mfn(nr_frags + 1)) { + /* Memory squeeze? Back off for an arbitrary while. */ + if ( net_ratelimit() ) + WPRINTK("Memory squeeze in netback " + "driver.\n"); + mod_timer(&net_timer, jiffies + HZ); + skb_queue_head(&rx_queue, skb); + break; + } + + netbk_gop_skb(skb, meta, count); + + count += nr_frags + 1; + + __skb_queue_tail(&rxq, skb); + + /* Filled the batch queue? */ + if (count + MAX_SKB_FRAGS >= NET_RX_RING_SIZE) + break; + } + + if (!count) + return; + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + mcl = rx_mcl + count; + + mcl[-1].args[MULTI_UVMFLAGS_INDEX] = UVMF_TLB_FLUSH|UVMF_ALL; + + mcl->op = __HYPERVISOR_mmu_update; + mcl->args[0] = (unsigned long)rx_mmu; + mcl->args[1] = count; + mcl->args[2] = 0; + mcl->args[3] = DOMID_SELF; + + ret = HYPERVISOR_multicall(rx_mcl, count + 1); + BUG_ON(ret != 0); + } + + ret = HYPERVISOR_grant_table_op(GNTTABOP_transfer, grant_rx_op, count); + BUG_ON(ret != 0); + + count = 0; + while ((skb = __skb_dequeue(&rxq)) != NULL) { + nr_frags = *(int *)skb->cb; + + atomic_set(&(skb_shinfo(skb)->dataref), 1); + skb_shinfo(skb)->nr_frags = 0; + skb_shinfo(skb)->frag_list = NULL; + + netif = netdev_priv(skb->dev); + netif->stats.tx_bytes += skb->len; + netif->stats.tx_packets++; + + netbk_free_pages(nr_frags, meta + count + 1); + status = netbk_check_gop(nr_frags, netif->domid, count); + + id = meta[count].id; + flags = nr_frags ? NETRXF_more_data : 0; + + if (skb->ip_summed == CHECKSUM_HW) /* local packet? */ + flags |= NETRXF_csum_blank | NETRXF_data_validated; + else if (skb->proto_data_valid) /* remote but checksummed? */ + flags |= NETRXF_data_validated; + + resp = make_rx_response(netif, id, status, + offset_in_page(skb->data), + skb_headlen(skb), flags); + + extra = NULL; + + if (meta[count].frag.size) { + struct netif_extra_info *gso = + (struct netif_extra_info *) + RING_GET_RESPONSE(&netif->rx, + netif->rx.rsp_prod_pvt++); + + if (extra) + extra->flags |= XEN_NETIF_EXTRA_FLAG_MORE; + else + resp->flags |= NETRXF_extra_info; + + gso->u.gso.size = meta[count].frag.size; + gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4; + gso->u.gso.pad = 0; + gso->u.gso.features = 0; + + gso->type = XEN_NETIF_EXTRA_TYPE_GSO; + gso->flags = 0; + extra = gso; + } + + netbk_add_frag_responses(netif, status, meta + count + 1, + nr_frags); + + RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&netif->rx, ret); + irq = netif->irq; + if (ret && !rx_notify[irq]) { + rx_notify[irq] = 1; + notify_list[notify_nr++] = irq; + } + + if (netif_queue_stopped(netif->dev) && + !netbk_queue_full(netif)) + netif_wake_queue(netif->dev); + + netif_put(netif); + dev_kfree_skb(skb); + count += nr_frags + 1; + } + + while (notify_nr != 0) { + irq = notify_list[--notify_nr]; + rx_notify[irq] = 0; + notify_remote_via_irq(irq); + } + + /* More work to do? */ + if (!skb_queue_empty(&rx_queue) && !timer_pending(&net_timer)) + tasklet_schedule(&net_rx_tasklet); +#if 0 + else + xen_network_done_notify(); +#endif +} + +static void net_alarm(unsigned long unused) +{ + tasklet_schedule(&net_rx_tasklet); +} + +struct net_device_stats *netif_be_get_stats(struct net_device *dev) +{ + netif_t *netif = netdev_priv(dev); + return &netif->stats; +} + +static int __on_net_schedule_list(netif_t *netif) +{ + return netif->list.next != NULL; +} + +static void remove_from_net_schedule_list(netif_t *netif) +{ + spin_lock_irq(&net_schedule_list_lock); + if (likely(__on_net_schedule_list(netif))) { + list_del(&netif->list); + netif->list.next = NULL; + netif_put(netif); + } + spin_unlock_irq(&net_schedule_list_lock); +} + +static void add_to_net_schedule_list_tail(netif_t *netif) +{ + if (__on_net_schedule_list(netif)) + return; + + spin_lock_irq(&net_schedule_list_lock); + if (!__on_net_schedule_list(netif) && + likely(netif_running(netif->dev) && + netif_carrier_ok(netif->dev))) { + list_add_tail(&netif->list, &net_schedule_list); + netif_get(netif); + } + spin_unlock_irq(&net_schedule_list_lock); +} + +/* + * Note on CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER: + * If this driver is pipelining transmit requests then we can be very + * aggressive in avoiding new-packet notifications -- frontend only needs to + * send a notification if there are no outstanding unreceived responses. + * If we may be buffer transmit buffers for any reason then we must be rather + * more conservative and treat this as the final check for pending work. + */ +void netif_schedule_work(netif_t *netif) +{ + int more_to_do; + +#ifdef CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER + more_to_do = RING_HAS_UNCONSUMED_REQUESTS(&netif->tx); +#else + RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, more_to_do); +#endif + + if (more_to_do) { + add_to_net_schedule_list_tail(netif); + maybe_schedule_tx_action(); + } +} + +void netif_deschedule_work(netif_t *netif) +{ + remove_from_net_schedule_list(netif); +} + + +static void tx_credit_callback(unsigned long data) +{ + netif_t *netif = (netif_t *)data; + netif->remaining_credit = netif->credit_bytes; + netif_schedule_work(netif); +} + +inline static void net_tx_action_dealloc(void) +{ + gnttab_unmap_grant_ref_t *gop; + u16 pending_idx; + PEND_RING_IDX dc, dp; + netif_t *netif; + int ret; + + dc = dealloc_cons; + dp = dealloc_prod; + + /* Ensure we see all indexes enqueued by netif_idx_release(). */ + smp_rmb(); + + /* + * Free up any grants we have finished using + */ + gop = tx_unmap_ops; + while (dc != dp) { + pending_idx = dealloc_ring[MASK_PEND_IDX(dc++)]; + gnttab_set_unmap_op(gop, MMAP_VADDR(pending_idx), + GNTMAP_host_map, + grant_tx_handle[pending_idx]); + gop++; + } + ret = HYPERVISOR_grant_table_op( + GNTTABOP_unmap_grant_ref, tx_unmap_ops, gop - tx_unmap_ops); + BUG_ON(ret); + + while (dealloc_cons != dp) { + pending_idx = dealloc_ring[MASK_PEND_IDX(dealloc_cons++)]; + + netif = pending_tx_info[pending_idx].netif; + + make_tx_response(netif, &pending_tx_info[pending_idx].req, + NETIF_RSP_OKAY); + + pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx; + + netif_put(netif); + } +} + +static void netbk_tx_err(netif_t *netif, netif_tx_request_t *txp, RING_IDX end) +{ + RING_IDX cons = netif->tx.req_cons; + + do { + make_tx_response(netif, txp, NETIF_RSP_ERROR); + if (cons >= end) + break; + txp = RING_GET_REQUEST(&netif->tx, cons++); + } while (1); + netif->tx.req_cons = cons; + netif_schedule_work(netif); + netif_put(netif); +} + +static int netbk_count_requests(netif_t *netif, netif_tx_request_t *txp, + int work_to_do) +{ + netif_tx_request_t *first = txp; + RING_IDX cons = netif->tx.req_cons; + int frags = 0; + + while (txp->flags & NETTXF_more_data) { + if (frags >= work_to_do) { + DPRINTK("Need more frags\n"); + return -frags; + } + + txp = RING_GET_REQUEST(&netif->tx, cons + frags); + if (txp->size > first->size) { + DPRINTK("Frags galore\n"); + return -frags; + } + + first->size -= txp->size; + frags++; + + if (unlikely((txp->offset + txp->size) > PAGE_SIZE)) { + DPRINTK("txp->offset: %x, size: %u\n", + txp->offset, txp->size); + return -frags; + } + } + + return frags; +} + +static gnttab_map_grant_ref_t *netbk_get_requests(netif_t *netif, + struct sk_buff *skb, + gnttab_map_grant_ref_t *mop) +{ + struct skb_shared_info *shinfo = skb_shinfo(skb); + skb_frag_t *frags = shinfo->frags; + netif_tx_request_t *txp; + unsigned long pending_idx = *((u16 *)skb->data); + RING_IDX cons = netif->tx.req_cons; + int i, start; + + /* Skip first skb fragment if it is on same page as header fragment. */ + start = ((unsigned long)shinfo->frags[0].page == pending_idx); + + for (i = start; i < shinfo->nr_frags; i++) { + txp = RING_GET_REQUEST(&netif->tx, cons++); + pending_idx = pending_ring[MASK_PEND_IDX(pending_cons++)]; + + gnttab_set_map_op(mop++, MMAP_VADDR(pending_idx), + GNTMAP_host_map | GNTMAP_readonly, + txp->gref, netif->domid); + + memcpy(&pending_tx_info[pending_idx].req, txp, sizeof(*txp)); + netif_get(netif); + pending_tx_info[pending_idx].netif = netif; + frags[i].page = (void *)pending_idx; + } + + return mop; +} + +static int netbk_tx_check_mop(struct sk_buff *skb, + gnttab_map_grant_ref_t **mopp) +{ + gnttab_map_grant_ref_t *mop = *mopp; + int pending_idx = *((u16 *)skb->data); + netif_t *netif = pending_tx_info[pending_idx].netif; + netif_tx_request_t *txp; + struct skb_shared_info *shinfo = skb_shinfo(skb); + int nr_frags = shinfo->nr_frags; + int i, err, start; + + /* Check status of header. */ + err = mop->status; + if (unlikely(err)) { + txp = &pending_tx_info[pending_idx].req; + make_tx_response(netif, txp, NETIF_RSP_ERROR); + pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx; + netif_put(netif); + } else { + set_phys_to_machine( + __pa(MMAP_VADDR(pending_idx)) >> PAGE_SHIFT, + FOREIGN_FRAME(mop->dev_bus_addr >> PAGE_SHIFT)); + grant_tx_handle[pending_idx] = mop->handle; + } + + /* Skip first skb fragment if it is on same page as header fragment. */ + start = ((unsigned long)shinfo->frags[0].page == pending_idx); + + for (i = start; i < nr_frags; i++) { + int j, newerr; + + pending_idx = (unsigned long)shinfo->frags[i].page; + + /* Check error status: if okay then remember grant handle. */ + newerr = (++mop)->status; + if (likely(!newerr)) { + set_phys_to_machine( + __pa(MMAP_VADDR(pending_idx))>>PAGE_SHIFT, + FOREIGN_FRAME(mop->dev_bus_addr>>PAGE_SHIFT)); + grant_tx_handle[pending_idx] = mop->handle; + /* Had a previous error? Invalidate this fragment. */ + if (unlikely(err)) + netif_idx_release(pending_idx); + continue; + } + + /* Error on this fragment: respond to client with an error. */ + txp = &pending_tx_info[pending_idx].req; + make_tx_response(netif, txp, NETIF_RSP_ERROR); + pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx; + netif_put(netif); + + /* Not the first error? Preceding frags already invalidated. */ + if (err) + continue; + + /* First error: invalidate header and preceding fragments. */ + pending_idx = *((u16 *)skb->data); + netif_idx_release(pending_idx); + for (j = start; j < i; j++) { + pending_idx = (unsigned long)shinfo->frags[i].page; + netif_idx_release(pending_idx); + } + + /* Remember the error: invalidate all subsequent fragments. */ + err = newerr; + } + + *mopp = mop + 1; + return err; +} + +static void netbk_fill_frags(struct sk_buff *skb) +{ + struct skb_shared_info *shinfo = skb_shinfo(skb); + int nr_frags = shinfo->nr_frags; + int i; + + for (i = 0; i < nr_frags; i++) { + skb_frag_t *frag = shinfo->frags + i; + netif_tx_request_t *txp; + unsigned long pending_idx; + + pending_idx = (unsigned long)frag->page; + txp = &pending_tx_info[pending_idx].req; + frag->page = virt_to_page(MMAP_VADDR(pending_idx)); + frag->size = txp->size; + frag->page_offset = txp->offset; + + skb->len += txp->size; + skb->data_len += txp->size; + skb->truesize += txp->size; + } +} + +int netbk_get_extras(netif_t *netif, struct netif_extra_info *extras, + int work_to_do) +{ + struct netif_extra_info *extra; + RING_IDX cons = netif->tx.req_cons; + + do { + if (unlikely(work_to_do-- <= 0)) { + DPRINTK("Missing extra info\n"); + return -EBADR; + } + + extra = (struct netif_extra_info *) + RING_GET_REQUEST(&netif->tx, cons); + if (unlikely(!extra->type || + extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) { + netif->tx.req_cons = ++cons; + DPRINTK("Invalid extra type: %d\n", extra->type); + return -EINVAL; + } + + memcpy(&extras[extra->type - 1], extra, sizeof(*extra)); + netif->tx.req_cons = ++cons; + } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE); + + return work_to_do; +} + +static int netbk_set_skb_gso(struct sk_buff *skb, struct netif_extra_info *gso) +{ + if (!gso->u.gso.size) { + DPRINTK("GSO size must not be zero.\n"); + return -EINVAL; + } + + /* Currently only TCPv4 S.O. is supported. */ + if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4) { + DPRINTK("Bad GSO type %d.\n", gso->u.gso.type); + return -EINVAL; + } + + skb_shinfo(skb)->gso_size = gso->u.gso.size; + skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; + + /* Header must be checked, and gso_segs computed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + + return 0; +} + +/* Called after netfront has transmitted */ +static void net_tx_action(unsigned long unused) +{ + struct list_head *ent; + struct sk_buff *skb; + netif_t *netif; + netif_tx_request_t txreq; + struct netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1]; + u16 pending_idx; + RING_IDX i; + gnttab_map_grant_ref_t *mop; + unsigned int data_len; + int ret, work_to_do; + + if (dealloc_cons != dealloc_prod) + net_tx_action_dealloc(); + + mop = tx_map_ops; + while (((NR_PENDING_REQS + MAX_SKB_FRAGS) < MAX_PENDING_REQS) && + !list_empty(&net_schedule_list)) { + /* Get a netif from the list with work to do. */ + ent = net_schedule_list.next; + netif = list_entry(ent, netif_t, list); + netif_get(netif); + remove_from_net_schedule_list(netif); + + RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, work_to_do); + if (!work_to_do) { + netif_put(netif); + continue; + } + + i = netif->tx.req_cons; + rmb(); /* Ensure that we see the request before we copy it. */ + memcpy(&txreq, RING_GET_REQUEST(&netif->tx, i), sizeof(txreq)); + /* Credit-based scheduling. */ + if (txreq.size > netif->remaining_credit) { + unsigned long now = jiffies; + unsigned long next_credit = + netif->credit_timeout.expires + + msecs_to_jiffies(netif->credit_usec / 1000); + + /* Timer could already be pending in rare cases. */ + if (timer_pending(&netif->credit_timeout)) + break; + + /* Passed the point where we can replenish credit? */ + if (time_after_eq(now, next_credit)) { + netif->credit_timeout.expires = now; + netif->remaining_credit = netif->credit_bytes; + } + + /* Still too big to send right now? Set a callback. */ + if (txreq.size > netif->remaining_credit) { + netif->remaining_credit = 0; + netif->credit_timeout.data = + (unsigned long)netif; + netif->credit_timeout.function = + tx_credit_callback; + __mod_timer(&netif->credit_timeout, + next_credit); + break; + } + } + netif->remaining_credit -= txreq.size; + + work_to_do--; + netif->tx.req_cons = ++i; + + memset(extras, 0, sizeof(extras)); + if (txreq.flags & NETTXF_extra_info) { + work_to_do = netbk_get_extras(netif, extras, + work_to_do); + i = netif->tx.req_cons; + if (unlikely(work_to_do < 0)) { + netbk_tx_err(netif, &txreq, i); + continue; + } + } + + ret = netbk_count_requests(netif, &txreq, work_to_do); + if (unlikely(ret < 0)) { + netbk_tx_err(netif, &txreq, i - ret); + continue; + } + i += ret; + + if (unlikely(ret > MAX_SKB_FRAGS)) { + DPRINTK("Too many frags\n"); + netbk_tx_err(netif, &txreq, i); + continue; + } + + if (unlikely(txreq.size < ETH_HLEN)) { + DPRINTK("Bad packet size: %d\n", txreq.size); + netbk_tx_err(netif, &txreq, i); + continue; + } + + /* No crossing a page as the payload mustn't fragment. */ + if (unlikely((txreq.offset + txreq.size) > PAGE_SIZE)) { + DPRINTK("txreq.offset: %x, size: %u, end: %lu\n", + txreq.offset, txreq.size, + (txreq.offset &~PAGE_MASK) + txreq.size); + netbk_tx_err(netif, &txreq, i); + continue; + } + + pending_idx = pending_ring[MASK_PEND_IDX(pending_cons)]; + + data_len = (txreq.size > PKT_PROT_LEN && + ret < MAX_SKB_FRAGS) ? + PKT_PROT_LEN : txreq.size; + + skb = alloc_skb(data_len+16, GFP_ATOMIC); + if (unlikely(skb == NULL)) { + DPRINTK("Can't allocate a skb in start_xmit.\n"); + netbk_tx_err(netif, &txreq, i); + break; + } + + /* Packets passed to netif_rx() must have some headroom. */ + skb_reserve(skb, 16); + + if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) { + struct netif_extra_info *gso; + gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1]; + + if (netbk_set_skb_gso(skb, gso)) { + kfree_skb(skb); + netbk_tx_err(netif, &txreq, i); + continue; + } + } + + gnttab_set_map_op(mop, MMAP_VADDR(pending_idx), + GNTMAP_host_map | GNTMAP_readonly, + txreq.gref, netif->domid); + mop++; + + memcpy(&pending_tx_info[pending_idx].req, + &txreq, sizeof(txreq)); + pending_tx_info[pending_idx].netif = netif; + *((u16 *)skb->data) = pending_idx; + + __skb_put(skb, data_len); + + skb_shinfo(skb)->nr_frags = ret; + if (data_len < txreq.size) { + skb_shinfo(skb)->nr_frags++; + skb_shinfo(skb)->frags[0].page = + (void *)(unsigned long)pending_idx; + } + + __skb_queue_tail(&tx_queue, skb); + + pending_cons++; + + mop = netbk_get_requests(netif, skb, mop); + + netif->tx.req_cons = i; + netif_schedule_work(netif); + + if ((mop - tx_map_ops) >= ARRAY_SIZE(tx_map_ops)) + break; + } + + if (mop == tx_map_ops) + return; + + ret = HYPERVISOR_grant_table_op( + GNTTABOP_map_grant_ref, tx_map_ops, mop - tx_map_ops); + BUG_ON(ret); + + mop = tx_map_ops; + while ((skb = __skb_dequeue(&tx_queue)) != NULL) { + netif_tx_request_t *txp; + + pending_idx = *((u16 *)skb->data); + netif = pending_tx_info[pending_idx].netif; + txp = &pending_tx_info[pending_idx].req; + + /* Check the remap error code. */ + if (unlikely(netbk_tx_check_mop(skb, &mop))) { + printk(KERN_ALERT "#### netback grant fails\n"); + skb_shinfo(skb)->nr_frags = 0; + kfree_skb(skb); + continue; + } + + data_len = skb->len; + memcpy(skb->data, + (void *)(MMAP_VADDR(pending_idx)|txp->offset), + data_len); + if (data_len < txp->size) { + /* Append the packet payload as a fragment. */ + txp->offset += data_len; + txp->size -= data_len; + } else { + /* Schedule a response immediately. */ + netif_idx_release(pending_idx); + } + + /* + * Old frontends do not assert data_validated but we + * can infer it from csum_blank so test both flags. + */ + if (txp->flags & (NETTXF_data_validated|NETTXF_csum_blank)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + skb->proto_data_valid = 1; + } else { + skb->ip_summed = CHECKSUM_NONE; + skb->proto_data_valid = 0; + } + skb->proto_csum_blank = !!(txp->flags & NETTXF_csum_blank); + + netbk_fill_frags(skb); + + skb->dev = netif->dev; + skb->protocol = eth_type_trans(skb, skb->dev); + + netif->stats.rx_bytes += skb->len; + netif->stats.rx_packets++; + + netif_rx(skb); + netif->dev->last_rx = jiffies; + } +} + +static void netif_idx_release(u16 pending_idx) +{ + static DEFINE_SPINLOCK(_lock); + unsigned long flags; + + spin_lock_irqsave(&_lock, flags); + dealloc_ring[MASK_PEND_IDX(dealloc_prod)] = pending_idx; + /* Sync with net_tx_action_dealloc: insert idx /then/ incr producer. */ + smp_wmb(); + dealloc_prod++; + spin_unlock_irqrestore(&_lock, flags); + + tasklet_schedule(&net_tx_tasklet); +} + +static void netif_page_release(struct page *page) +{ + u16 pending_idx = page - virt_to_page(mmap_vstart); + + /* Ready for next use. */ + init_page_count(page); + + netif_idx_release(pending_idx); +} + +irqreturn_t netif_be_int(int irq, void *dev_id, struct pt_regs *regs) +{ + netif_t *netif = dev_id; + + add_to_net_schedule_list_tail(netif); + maybe_schedule_tx_action(); + + if (netif_queue_stopped(netif->dev) && !netbk_queue_full(netif)) + netif_wake_queue(netif->dev); + + return IRQ_HANDLED; +} + +static void make_tx_response(netif_t *netif, + netif_tx_request_t *txp, + s8 st) +{ + RING_IDX i = netif->tx.rsp_prod_pvt; + netif_tx_response_t *resp; + int notify; + + resp = RING_GET_RESPONSE(&netif->tx, i); + resp->id = txp->id; + resp->status = st; + + if (txp->flags & NETTXF_extra_info) + RING_GET_RESPONSE(&netif->tx, ++i)->status = NETIF_RSP_NULL; + + netif->tx.rsp_prod_pvt = ++i; + RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&netif->tx, notify); + if (notify) + notify_remote_via_irq(netif->irq); + +#ifdef CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER + if (i == netif->tx.req_cons) { + int more_to_do; + RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, more_to_do); + if (more_to_do) + add_to_net_schedule_list_tail(netif); + } +#endif +} + +static netif_rx_response_t *make_rx_response(netif_t *netif, + u16 id, + s8 st, + u16 offset, + u16 size, + u16 flags) +{ + RING_IDX i = netif->rx.rsp_prod_pvt; + netif_rx_response_t *resp; + + resp = RING_GET_RESPONSE(&netif->rx, i); + resp->offset = offset; + resp->flags = flags; + resp->id = id; + resp->status = (s16)size; + if (st < 0) + resp->status = (s16)st; + + netif->rx.rsp_prod_pvt = ++i; + + return resp; +} + +#ifdef NETBE_DEBUG_INTERRUPT +static irqreturn_t netif_be_dbg(int irq, void *dev_id, struct pt_regs *regs) +{ + struct list_head *ent; + netif_t *netif; + int i = 0; + + printk(KERN_ALERT "netif_schedule_list:\n"); + spin_lock_irq(&net_schedule_list_lock); + + list_for_each (ent, &net_schedule_list) { + netif = list_entry(ent, netif_t, list); + printk(KERN_ALERT " %d: private(rx_req_cons=%08x " + "rx_resp_prod=%08x\n", + i, netif->rx.req_cons, netif->rx.rsp_prod_pvt); + printk(KERN_ALERT " tx_req_cons=%08x tx_resp_prod=%08x)\n", + netif->tx.req_cons, netif->tx.rsp_prod_pvt); + printk(KERN_ALERT " shared(rx_req_prod=%08x " + "rx_resp_prod=%08x\n", + netif->rx.sring->req_prod, netif->rx.sring->rsp_prod); + printk(KERN_ALERT " rx_event=%08x tx_req_prod=%08x\n", + netif->rx.sring->rsp_event, netif->tx.sring->req_prod); + printk(KERN_ALERT " tx_resp_prod=%08x, tx_event=%08x)\n", + netif->tx.sring->rsp_prod, netif->tx.sring->rsp_event); + i++; + } + + spin_unlock_irq(&net_schedule_list_lock); + printk(KERN_ALERT " ** End of netif_schedule_list **\n"); + + return IRQ_HANDLED; +} +#endif + +static int __init netback_init(void) +{ + int i; + struct page *page; + + if (!is_running_on_xen()) + return -ENODEV; + + /* We can increase reservation by this much in net_rx_action(). */ + balloon_update_driver_allowance(NET_RX_RING_SIZE); + + skb_queue_head_init(&rx_queue); + skb_queue_head_init(&tx_queue); + + init_timer(&net_timer); + net_timer.data = 0; + net_timer.function = net_alarm; + + page = balloon_alloc_empty_page_range(MAX_PENDING_REQS); + if (page == NULL) + return -ENOMEM; + + mmap_vstart = (unsigned long)pfn_to_kaddr(page_to_pfn(page)); + + for (i = 0; i < MAX_PENDING_REQS; i++) { + page = virt_to_page(MMAP_VADDR(i)); + init_page_count(page); + SetPageForeign(page, netif_page_release); + } + + pending_cons = 0; + pending_prod = MAX_PENDING_REQS; + for (i = 0; i < MAX_PENDING_REQS; i++) + pending_ring[i] = i; + + spin_lock_init(&net_schedule_list_lock); + INIT_LIST_HEAD(&net_schedule_list); + + netif_xenbus_init(); + +#ifdef NETBE_DEBUG_INTERRUPT + (void)bind_virq_to_irqhandler( + VIRQ_DEBUG, + 0, + netif_be_dbg, + SA_SHIRQ, + "net-be-dbg", + &netif_be_dbg); +#endif + + return 0; +} + +module_init(netback_init); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/netfront/Kconfig b/drivers/xen/netfront/Kconfig new file mode 100644 index 000000000..334e6c386 --- /dev/null +++ b/drivers/xen/netfront/Kconfig @@ -0,0 +1,6 @@ + +config XENNET + tristate "Xen network driver" + depends on NETDEVICES && ARCH_XEN + help + Network driver for Xen diff --git a/drivers/xen/netfront/Makefile b/drivers/xen/netfront/Makefile new file mode 100644 index 000000000..dc22829f5 --- /dev/null +++ b/drivers/xen/netfront/Makefile @@ -0,0 +1,4 @@ + +obj-$(CONFIG_XEN_NETDEV_FRONTEND) := xennet.o + +xennet-objs := netfront.o diff --git a/drivers/xen/netfront/netfront.c b/drivers/xen/netfront/netfront.c new file mode 100644 index 000000000..ef6e134a7 --- /dev/null +++ b/drivers/xen/netfront/netfront.c @@ -0,0 +1,1841 @@ +/****************************************************************************** + * Virtual network driver for conversing with remote driver backends. + * + * Copyright (c) 2002-2005, K A Fraser + * Copyright (c) 2005, XenSource Ltd + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License version 2 + * as published by the Free Software Foundation; or, when distributed + * separately from the Linux kernel or incorporated into other + * software packages, subject to the following license: + * + * Permission is hereby granted, free of charge, to any person obtaining a copy + * of this source file (the "Software"), to deal in the Software without + * restriction, including without limitation the rights to use, copy, modify, + * merge, publish, distribute, sublicense, and/or sell copies of the Software, + * and to permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#define RX_COPY_THRESHOLD 256 + +#define GRANT_INVALID_REF 0 + +#define NET_TX_RING_SIZE __RING_SIZE((struct netif_tx_sring *)0, PAGE_SIZE) +#define NET_RX_RING_SIZE __RING_SIZE((struct netif_rx_sring *)0, PAGE_SIZE) + +struct netfront_info { + struct list_head list; + struct net_device *netdev; + + struct net_device_stats stats; + + struct netif_tx_front_ring tx; + struct netif_rx_front_ring rx; + + spinlock_t tx_lock; + spinlock_t rx_lock; + + unsigned int handle; + unsigned int evtchn, irq; + + /* Receive-ring batched refills. */ +#define RX_MIN_TARGET 8 +#define RX_DFL_MIN_TARGET 64 +#define RX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256) + unsigned rx_min_target, rx_max_target, rx_target; + struct sk_buff_head rx_batch; + + struct timer_list rx_refill_timer; + + /* + * {tx,rx}_skbs store outstanding skbuffs. The first entry in tx_skbs + * is an index into a chain of free entries. + */ + struct sk_buff *tx_skbs[NET_TX_RING_SIZE+1]; + struct sk_buff *rx_skbs[NET_RX_RING_SIZE]; + +#define TX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256) + grant_ref_t gref_tx_head; + grant_ref_t grant_tx_ref[NET_TX_RING_SIZE + 1]; + grant_ref_t gref_rx_head; + grant_ref_t grant_rx_ref[NET_TX_RING_SIZE]; + + struct xenbus_device *xbdev; + int tx_ring_ref; + int rx_ring_ref; + u8 mac[ETH_ALEN]; + + unsigned long rx_pfn_array[NET_RX_RING_SIZE]; + struct multicall_entry rx_mcl[NET_RX_RING_SIZE+1]; + struct mmu_update rx_mmu[NET_RX_RING_SIZE]; +}; + +struct netfront_rx_info { + struct netif_rx_response rx; + struct netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1]; +}; + +/* + * Access macros for acquiring freeing slots in tx_skbs[]. + */ + +static inline void add_id_to_freelist(struct sk_buff **list, unsigned short id) +{ + list[id] = list[0]; + list[0] = (void *)(unsigned long)id; +} + +static inline unsigned short get_id_from_freelist(struct sk_buff **list) +{ + unsigned int id = (unsigned int)(unsigned long)list[0]; + list[0] = list[id]; + return id; +} + +static inline int xennet_rxidx(RING_IDX idx) +{ + return idx & (NET_RX_RING_SIZE - 1); +} + +static inline struct sk_buff *xennet_get_rx_skb(struct netfront_info *np, + RING_IDX ri) +{ + int i = xennet_rxidx(ri); + struct sk_buff *skb = np->rx_skbs[i]; + np->rx_skbs[i] = NULL; + return skb; +} + +static inline grant_ref_t xennet_get_rx_ref(struct netfront_info *np, + RING_IDX ri) +{ + int i = xennet_rxidx(ri); + grant_ref_t ref = np->grant_rx_ref[i]; + np->grant_rx_ref[i] = GRANT_INVALID_REF; + return ref; +} + +#define DPRINTK(fmt, args...) \ + pr_debug("netfront (%s:%d) " fmt, \ + __FUNCTION__, __LINE__, ##args) +#define IPRINTK(fmt, args...) \ + printk(KERN_INFO "netfront: " fmt, ##args) +#define WPRINTK(fmt, args...) \ + printk(KERN_WARNING "netfront: " fmt, ##args) + +static int talk_to_backend(struct xenbus_device *, struct netfront_info *); +static int setup_device(struct xenbus_device *, struct netfront_info *); +static struct net_device *create_netdev(int, struct xenbus_device *); + +static void netfront_closing(struct xenbus_device *); + +static void end_access(int, void *); +static void netif_disconnect_backend(struct netfront_info *); +static void close_netdev(struct netfront_info *); +static void netif_free(struct netfront_info *); + +static void network_connect(struct net_device *); +static void network_tx_buf_gc(struct net_device *); +static void network_alloc_rx_buffers(struct net_device *); +static int send_fake_arp(struct net_device *); + +static irqreturn_t netif_int(int irq, void *dev_id, struct pt_regs *ptregs); + +#ifdef CONFIG_SYSFS +static int xennet_sysfs_addif(struct net_device *netdev); +static void xennet_sysfs_delif(struct net_device *netdev); +#else /* !CONFIG_SYSFS */ +#define xennet_sysfs_addif(dev) (0) +#define xennet_sysfs_delif(dev) do { } while(0) +#endif + +static inline int xennet_can_sg(struct net_device *dev) +{ + return dev->features & NETIF_F_SG; +} + +/** + * Entry point to this code when a new device is created. Allocate the basic + * structures and the ring buffers for communication with the backend, and + * inform the backend of the appropriate details for those. Switch to + * Connected state. + */ +static int __devinit netfront_probe(struct xenbus_device *dev, + const struct xenbus_device_id *id) +{ + int err; + struct net_device *netdev; + struct netfront_info *info; + unsigned int handle; + + err = xenbus_scanf(XBT_NIL, dev->nodename, "handle", "%u", &handle); + if (err != 1) { + xenbus_dev_fatal(dev, err, "reading handle"); + return err; + } + + netdev = create_netdev(handle, dev); + if (IS_ERR(netdev)) { + err = PTR_ERR(netdev); + xenbus_dev_fatal(dev, err, "creating netdev"); + return err; + } + + info = netdev_priv(netdev); + dev->dev.driver_data = info; + + err = talk_to_backend(dev, info); + if (err) { + xennet_sysfs_delif(info->netdev); + unregister_netdev(netdev); + free_netdev(netdev); + dev->dev.driver_data = NULL; + return err; + } + + return 0; +} + + +/** + * We are reconnecting to the backend, due to a suspend/resume, or a backend + * driver restart. We tear down our netif structure and recreate it, but + * leave the device-layer structures intact so that this is transparent to the + * rest of the kernel. + */ +static int netfront_resume(struct xenbus_device *dev) +{ + struct netfront_info *info = dev->dev.driver_data; + + DPRINTK("%s\n", dev->nodename); + + netif_disconnect_backend(info); + return talk_to_backend(dev, info); +} + +static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[]) +{ + char *s, *e, *macstr; + int i; + + macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL); + if (IS_ERR(macstr)) + return PTR_ERR(macstr); + + for (i = 0; i < ETH_ALEN; i++) { + mac[i] = simple_strtoul(s, &e, 16); + if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) { + kfree(macstr); + return -ENOENT; + } + s = e+1; + } + + kfree(macstr); + return 0; +} + +/* Common code used when first setting up, and when resuming. */ +static int talk_to_backend(struct xenbus_device *dev, + struct netfront_info *info) +{ + const char *message; + struct xenbus_transaction xbt; + int err; + + err = xen_net_read_mac(dev, info->mac); + if (err) { + xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename); + goto out; + } + + /* Create shared ring, alloc event channel. */ + err = setup_device(dev, info); + if (err) + goto out; + +again: + err = xenbus_transaction_start(&xbt); + if (err) { + xenbus_dev_fatal(dev, err, "starting transaction"); + goto destroy_ring; + } + + err = xenbus_printf(xbt, dev->nodename, "tx-ring-ref","%u", + info->tx_ring_ref); + if (err) { + message = "writing tx ring-ref"; + goto abort_transaction; + } + err = xenbus_printf(xbt, dev->nodename, "rx-ring-ref","%u", + info->rx_ring_ref); + if (err) { + message = "writing rx ring-ref"; + goto abort_transaction; + } + err = xenbus_printf(xbt, dev->nodename, + "event-channel", "%u", info->evtchn); + if (err) { + message = "writing event-channel"; + goto abort_transaction; + } + + err = xenbus_printf(xbt, dev->nodename, "feature-rx-notify", "%d", 1); + if (err) { + message = "writing feature-rx-notify"; + goto abort_transaction; + } + + err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", 1); + if (err) { + message = "writing feature-sg"; + goto abort_transaction; + } + + err = xenbus_printf(xbt, dev->nodename, "feature-gso-tcpv4", "%d", 1); + if (err) { + message = "writing feature-gso-tcpv4"; + goto abort_transaction; + } + + err = xenbus_transaction_end(xbt, 0); + if (err) { + if (err == -EAGAIN) + goto again; + xenbus_dev_fatal(dev, err, "completing transaction"); + goto destroy_ring; + } + + return 0; + + abort_transaction: + xenbus_transaction_end(xbt, 1); + xenbus_dev_fatal(dev, err, "%s", message); + destroy_ring: + netif_free(info); + out: + return err; +} + + +static int setup_device(struct xenbus_device *dev, struct netfront_info *info) +{ + struct netif_tx_sring *txs; + struct netif_rx_sring *rxs; + int err; + struct net_device *netdev = info->netdev; + + info->tx_ring_ref = GRANT_INVALID_REF; + info->rx_ring_ref = GRANT_INVALID_REF; + info->rx.sring = NULL; + info->tx.sring = NULL; + info->irq = 0; + + txs = (struct netif_tx_sring *)get_zeroed_page(GFP_KERNEL); + if (!txs) { + err = -ENOMEM; + xenbus_dev_fatal(dev, err, "allocating tx ring page"); + goto fail; + } + SHARED_RING_INIT(txs); + FRONT_RING_INIT(&info->tx, txs, PAGE_SIZE); + + err = xenbus_grant_ring(dev, virt_to_mfn(txs)); + if (err < 0) { + free_page((unsigned long)txs); + goto fail; + } + info->tx_ring_ref = err; + + rxs = (struct netif_rx_sring *)get_zeroed_page(GFP_KERNEL); + if (!rxs) { + err = -ENOMEM; + xenbus_dev_fatal(dev, err, "allocating rx ring page"); + goto fail; + } + SHARED_RING_INIT(rxs); + FRONT_RING_INIT(&info->rx, rxs, PAGE_SIZE); + + err = xenbus_grant_ring(dev, virt_to_mfn(rxs)); + if (err < 0) { + free_page((unsigned long)rxs); + goto fail; + } + info->rx_ring_ref = err; + + err = xenbus_alloc_evtchn(dev, &info->evtchn); + if (err) + goto fail; + + memcpy(netdev->dev_addr, info->mac, ETH_ALEN); + err = bind_evtchn_to_irqhandler(info->evtchn, netif_int, + SA_SAMPLE_RANDOM, netdev->name, netdev); + if (err < 0) + goto fail; + info->irq = err; + return 0; + + fail: + netif_free(info); + return err; +} + + +/** + * Callback received when the backend's state changes. + */ +static void backend_changed(struct xenbus_device *dev, + enum xenbus_state backend_state) +{ + struct netfront_info *np = dev->dev.driver_data; + struct net_device *netdev = np->netdev; + + DPRINTK("\n"); + + switch (backend_state) { + case XenbusStateInitialising: + case XenbusStateInitialised: + case XenbusStateConnected: + case XenbusStateUnknown: + case XenbusStateClosed: + break; + + case XenbusStateInitWait: + network_connect(netdev); + xenbus_switch_state(dev, XenbusStateConnected); + (void)send_fake_arp(netdev); + break; + + case XenbusStateClosing: + netfront_closing(dev); + break; + } +} + + +/** Send a packet on a net device to encourage switches to learn the + * MAC. We send a fake ARP request. + * + * @param dev device + * @return 0 on success, error code otherwise + */ +static int send_fake_arp(struct net_device *dev) +{ + struct sk_buff *skb; + u32 src_ip, dst_ip; + + dst_ip = INADDR_BROADCAST; + src_ip = inet_select_addr(dev, dst_ip, RT_SCOPE_LINK); + + /* No IP? Then nothing to do. */ + if (src_ip == 0) + return 0; + + skb = arp_create(ARPOP_REPLY, ETH_P_ARP, + dst_ip, dev, src_ip, + /*dst_hw*/ NULL, /*src_hw*/ NULL, + /*target_hw*/ dev->dev_addr); + if (skb == NULL) + return -ENOMEM; + + return dev_queue_xmit(skb); +} + + +static int network_open(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + + memset(&np->stats, 0, sizeof(np->stats)); + + network_alloc_rx_buffers(dev); + np->rx.sring->rsp_event = np->rx.rsp_cons + 1; + + if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx)) + netif_rx_schedule(dev); + + netif_start_queue(dev); + + return 0; +} + +static inline int netfront_tx_slot_available(struct netfront_info *np) +{ + return RING_FREE_REQUESTS(&np->tx) >= MAX_SKB_FRAGS + 2; +} + +static inline void network_maybe_wake_tx(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + + if (unlikely(netif_queue_stopped(dev)) && + netfront_tx_slot_available(np) && + likely(netif_running(dev))) + netif_wake_queue(dev); +} + +static void network_tx_buf_gc(struct net_device *dev) +{ + RING_IDX cons, prod; + unsigned short id; + struct netfront_info *np = netdev_priv(dev); + struct sk_buff *skb; + + if (unlikely(!netif_carrier_ok(dev))) + return; + + do { + prod = np->tx.sring->rsp_prod; + rmb(); /* Ensure we see responses up to 'rp'. */ + + for (cons = np->tx.rsp_cons; cons != prod; cons++) { + struct netif_tx_response *txrsp; + + txrsp = RING_GET_RESPONSE(&np->tx, cons); + if (txrsp->status == NETIF_RSP_NULL) + continue; + + id = txrsp->id; + skb = np->tx_skbs[id]; + if (unlikely(gnttab_query_foreign_access( + np->grant_tx_ref[id]) != 0)) { + printk(KERN_ALERT "network_tx_buf_gc: warning " + "-- grant still in use by backend " + "domain.\n"); + BUG(); + } + gnttab_end_foreign_access_ref( + np->grant_tx_ref[id], GNTMAP_readonly); + gnttab_release_grant_reference( + &np->gref_tx_head, np->grant_tx_ref[id]); + np->grant_tx_ref[id] = GRANT_INVALID_REF; + add_id_to_freelist(np->tx_skbs, id); + dev_kfree_skb_irq(skb); + } + + np->tx.rsp_cons = prod; + + /* + * Set a new event, then check for race with update of tx_cons. + * Note that it is essential to schedule a callback, no matter + * how few buffers are pending. Even if there is space in the + * transmit ring, higher layers may be blocked because too much + * data is outstanding: in such cases notification from Xen is + * likely to be the only kick that we'll get. + */ + np->tx.sring->rsp_event = + prod + ((np->tx.sring->req_prod - prod) >> 1) + 1; + mb(); + } while ((cons == prod) && (prod != np->tx.sring->rsp_prod)); + + network_maybe_wake_tx(dev); +} + + +static void rx_refill_timeout(unsigned long data) +{ + struct net_device *dev = (struct net_device *)data; + netif_rx_schedule(dev); +} + + +static void network_alloc_rx_buffers(struct net_device *dev) +{ + unsigned short id; + struct netfront_info *np = netdev_priv(dev); + struct sk_buff *skb; + struct page *page; + int i, batch_target, notify; + RING_IDX req_prod = np->rx.req_prod_pvt; + struct xen_memory_reservation reservation; + grant_ref_t ref; + unsigned long pfn; + void *vaddr; + + if (unlikely(!netif_carrier_ok(dev))) + return; + + /* + * Allocate skbuffs greedily, even though we batch updates to the + * receive ring. This creates a less bursty demand on the memory + * allocator, so should reduce the chance of failed allocation requests + * both for ourself and for other kernel subsystems. + */ + batch_target = np->rx_target - (req_prod - np->rx.rsp_cons); + for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) { + /* + * Allocate an skb and a page. Do not use __dev_alloc_skb as + * that will allocate page-sized buffers which is not + * necessary here. + * 16 bytes added as necessary headroom for netif_receive_skb. + */ + skb = alloc_skb(RX_COPY_THRESHOLD + 16, + GFP_ATOMIC | __GFP_NOWARN); + if (unlikely(!skb)) + goto no_skb; + + page = alloc_page(GFP_ATOMIC | __GFP_NOWARN); + if (!page) { + kfree_skb(skb); +no_skb: + /* Any skbuffs queued for refill? Force them out. */ + if (i != 0) + goto refill; + /* Could not allocate any skbuffs. Try again later. */ + mod_timer(&np->rx_refill_timer, + jiffies + (HZ/10)); + break; + } + + skb_reserve(skb, 16); /* mimic dev_alloc_skb() */ + skb_shinfo(skb)->frags[0].page = page; + skb_shinfo(skb)->nr_frags = 1; + __skb_queue_tail(&np->rx_batch, skb); + } + + /* Is the batch large enough to be worthwhile? */ + if (i < (np->rx_target/2)) { + if (req_prod > np->rx.sring->req_prod) + goto push; + return; + } + + /* Adjust our fill target if we risked running out of buffers. */ + if (((req_prod - np->rx.sring->rsp_prod) < (np->rx_target / 4)) && + ((np->rx_target *= 2) > np->rx_max_target)) + np->rx_target = np->rx_max_target; + + refill: + for (i = 0; ; i++) { + if ((skb = __skb_dequeue(&np->rx_batch)) == NULL) + break; + + skb->dev = dev; + + id = xennet_rxidx(req_prod + i); + + BUG_ON(np->rx_skbs[id]); + np->rx_skbs[id] = skb; + + RING_GET_REQUEST(&np->rx, req_prod + i)->id = id; + ref = gnttab_claim_grant_reference(&np->gref_rx_head); + BUG_ON((signed short)ref < 0); + np->grant_rx_ref[id] = ref; + + pfn = page_to_pfn(skb_shinfo(skb)->frags[0].page); + vaddr = page_address(skb_shinfo(skb)->frags[0].page); + + gnttab_grant_foreign_transfer_ref(ref, + np->xbdev->otherend_id, pfn); + RING_GET_REQUEST(&np->rx, req_prod + i)->gref = ref; + np->rx_pfn_array[i] = pfn_to_mfn(pfn); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* Remove this page before passing back to Xen. */ + set_phys_to_machine(pfn, INVALID_P2M_ENTRY); + MULTI_update_va_mapping(np->rx_mcl+i, + (unsigned long)vaddr, + __pte(0), 0); + } + } + + /* Tell the ballon driver what is going on. */ + balloon_update_driver_allowance(i); + + set_xen_guest_handle(reservation.extent_start, np->rx_pfn_array); + reservation.nr_extents = i; + reservation.extent_order = 0; + reservation.address_bits = 0; + reservation.domid = DOMID_SELF; + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* After all PTEs have been zapped, flush the TLB. */ + np->rx_mcl[i-1].args[MULTI_UVMFLAGS_INDEX] = + UVMF_TLB_FLUSH|UVMF_ALL; + + /* Give away a batch of pages. */ + np->rx_mcl[i].op = __HYPERVISOR_memory_op; + np->rx_mcl[i].args[0] = XENMEM_decrease_reservation; + np->rx_mcl[i].args[1] = (unsigned long)&reservation; + + /* Zap PTEs and give away pages in one big multicall. */ + (void)HYPERVISOR_multicall(np->rx_mcl, i+1); + + /* Check return status of HYPERVISOR_memory_op(). */ + if (unlikely(np->rx_mcl[i].result != i)) + panic("Unable to reduce memory reservation\n"); + } else + if (HYPERVISOR_memory_op(XENMEM_decrease_reservation, + &reservation) != i) + panic("Unable to reduce memory reservation\n"); + + /* Above is a suitable barrier to ensure backend will see requests. */ + np->rx.req_prod_pvt = req_prod + i; + push: + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->rx, notify); + if (notify) + notify_remote_via_irq(np->irq); +} + +static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev, + struct netif_tx_request *tx) +{ + struct netfront_info *np = netdev_priv(dev); + char *data = skb->data; + unsigned long mfn; + RING_IDX prod = np->tx.req_prod_pvt; + int frags = skb_shinfo(skb)->nr_frags; + unsigned int offset = offset_in_page(data); + unsigned int len = skb_headlen(skb); + unsigned int id; + grant_ref_t ref; + int i; + + while (len > PAGE_SIZE - offset) { + tx->size = PAGE_SIZE - offset; + tx->flags |= NETTXF_more_data; + len -= tx->size; + data += tx->size; + offset = 0; + + id = get_id_from_freelist(np->tx_skbs); + np->tx_skbs[id] = skb_get(skb); + tx = RING_GET_REQUEST(&np->tx, prod++); + tx->id = id; + ref = gnttab_claim_grant_reference(&np->gref_tx_head); + BUG_ON((signed short)ref < 0); + + mfn = virt_to_mfn(data); + gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id, + mfn, GNTMAP_readonly); + + tx->gref = np->grant_tx_ref[id] = ref; + tx->offset = offset; + tx->size = len; + tx->flags = 0; + } + + for (i = 0; i < frags; i++) { + skb_frag_t *frag = skb_shinfo(skb)->frags + i; + + tx->flags |= NETTXF_more_data; + + id = get_id_from_freelist(np->tx_skbs); + np->tx_skbs[id] = skb_get(skb); + tx = RING_GET_REQUEST(&np->tx, prod++); + tx->id = id; + ref = gnttab_claim_grant_reference(&np->gref_tx_head); + BUG_ON((signed short)ref < 0); + + mfn = pfn_to_mfn(page_to_pfn(frag->page)); + gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id, + mfn, GNTMAP_readonly); + + tx->gref = np->grant_tx_ref[id] = ref; + tx->offset = frag->page_offset; + tx->size = frag->size; + tx->flags = 0; + } + + np->tx.req_prod_pvt = prod; +} + +static int network_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + unsigned short id; + struct netfront_info *np = netdev_priv(dev); + struct netif_tx_request *tx; + struct netif_extra_info *extra; + char *data = skb->data; + RING_IDX i; + grant_ref_t ref; + unsigned long mfn; + int notify; + int frags = skb_shinfo(skb)->nr_frags; + unsigned int offset = offset_in_page(data); + unsigned int len = skb_headlen(skb); + + frags += (offset + len + PAGE_SIZE - 1) / PAGE_SIZE; + if (unlikely(frags > MAX_SKB_FRAGS + 1)) { + printk(KERN_ALERT "xennet: skb rides the rocket: %d frags\n", + frags); + dump_stack(); + goto drop; + } + + spin_lock_irq(&np->tx_lock); + + if (unlikely(!netif_carrier_ok(dev) || + (frags > 1 && !xennet_can_sg(dev)) || + netif_needs_gso(dev, skb))) { + spin_unlock_irq(&np->tx_lock); + goto drop; + } + + i = np->tx.req_prod_pvt; + + id = get_id_from_freelist(np->tx_skbs); + np->tx_skbs[id] = skb; + + tx = RING_GET_REQUEST(&np->tx, i); + + tx->id = id; + ref = gnttab_claim_grant_reference(&np->gref_tx_head); + BUG_ON((signed short)ref < 0); + mfn = virt_to_mfn(data); + gnttab_grant_foreign_access_ref( + ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly); + tx->gref = np->grant_tx_ref[id] = ref; + tx->offset = offset; + tx->size = len; + + tx->flags = 0; + extra = NULL; + + if (skb->ip_summed == CHECKSUM_HW) /* local packet? */ + tx->flags |= NETTXF_csum_blank | NETTXF_data_validated; + if (skb->proto_data_valid) /* remote but checksummed? */ + tx->flags |= NETTXF_data_validated; + + if (skb_shinfo(skb)->gso_size) { + struct netif_extra_info *gso = (struct netif_extra_info *) + RING_GET_REQUEST(&np->tx, ++i); + + if (extra) + extra->flags |= XEN_NETIF_EXTRA_FLAG_MORE; + else + tx->flags |= NETTXF_extra_info; + + gso->u.gso.size = skb_shinfo(skb)->gso_size; + gso->u.gso.type = XEN_NETIF_GSO_TYPE_TCPV4; + gso->u.gso.pad = 0; + gso->u.gso.features = 0; + + gso->type = XEN_NETIF_EXTRA_TYPE_GSO; + gso->flags = 0; + extra = gso; + } + + np->tx.req_prod_pvt = i + 1; + + xennet_make_frags(skb, dev, tx); + tx->size = skb->len; + + RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->tx, notify); + if (notify) + notify_remote_via_irq(np->irq); + + network_tx_buf_gc(dev); + + if (!netfront_tx_slot_available(np)) + netif_stop_queue(dev); + + spin_unlock_irq(&np->tx_lock); + + np->stats.tx_bytes += skb->len; + np->stats.tx_packets++; + + return 0; + + drop: + np->stats.tx_dropped++; + dev_kfree_skb(skb); + return 0; +} + +static irqreturn_t netif_int(int irq, void *dev_id, struct pt_regs *ptregs) +{ + struct net_device *dev = dev_id; + struct netfront_info *np = netdev_priv(dev); + unsigned long flags; + + spin_lock_irqsave(&np->tx_lock, flags); + network_tx_buf_gc(dev); + spin_unlock_irqrestore(&np->tx_lock, flags); + + if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx) && + likely(netif_running(dev))) + netif_rx_schedule(dev); + + return IRQ_HANDLED; +} + +static void xennet_move_rx_slot(struct netfront_info *np, struct sk_buff *skb, + grant_ref_t ref) +{ + int new = xennet_rxidx(np->rx.req_prod_pvt); + + BUG_ON(np->rx_skbs[new]); + np->rx_skbs[new] = skb; + np->grant_rx_ref[new] = ref; + RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->id = new; + RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->gref = ref; + np->rx.req_prod_pvt++; +} + +int xennet_get_extras(struct netfront_info *np, + struct netif_extra_info *extras, RING_IDX rp) + +{ + struct netif_extra_info *extra; + RING_IDX cons = np->rx.rsp_cons; + int err = 0; + + do { + struct sk_buff *skb; + grant_ref_t ref; + + if (unlikely(cons + 1 == rp)) { + if (net_ratelimit()) + WPRINTK("Missing extra info\n"); + err = -EBADR; + break; + } + + extra = (struct netif_extra_info *) + RING_GET_RESPONSE(&np->rx, ++cons); + + if (unlikely(!extra->type || + extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) { + if (net_ratelimit()) + WPRINTK("Invalid extra type: %d\n", + extra->type); + err = -EINVAL; + } else + memcpy(&extras[extra->type - 1], extra, sizeof(*extra)); + + skb = xennet_get_rx_skb(np, cons); + ref = xennet_get_rx_ref(np, cons); + xennet_move_rx_slot(np, skb, ref); + } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE); + + np->rx.rsp_cons = cons; + return err; +} + +static int xennet_get_responses(struct netfront_info *np, + struct netfront_rx_info *rinfo, RING_IDX rp, + struct sk_buff_head *list, int count) +{ + struct mmu_update *mmu = np->rx_mmu + count; + struct multicall_entry *mcl = np->rx_mcl + count; + struct netif_rx_response *rx = &rinfo->rx; + struct netif_extra_info *extras = rinfo->extras; + RING_IDX cons = np->rx.rsp_cons; + struct sk_buff *skb = xennet_get_rx_skb(np, cons); + grant_ref_t ref = xennet_get_rx_ref(np, cons); + int max = MAX_SKB_FRAGS + (rx->status <= RX_COPY_THRESHOLD); + int frags = 1; + int err = 0; + + if (rx->flags & NETRXF_extra_info) { + err = xennet_get_extras(np, extras, rp); + cons = np->rx.rsp_cons; + } + + for (;;) { + unsigned long mfn; + + if (unlikely(rx->status < 0 || + rx->offset + rx->status > PAGE_SIZE)) { + if (net_ratelimit()) + WPRINTK("rx->offset: %x, size: %u\n", + rx->offset, rx->status); + err = -EINVAL; + } + + /* + * This definitely indicates a bug, either in this driver or in + * the backend driver. In future this should flag the bad + * situation to the system controller to reboot the backed. + */ + if (ref == GRANT_INVALID_REF) { + WPRINTK("Bad rx response id %d.\n", rx->id); + err = -EINVAL; + goto next; + } + + /* Memory pressure, insufficient buffer headroom, ... */ + if ((mfn = gnttab_end_foreign_transfer_ref(ref)) == 0) { + if (net_ratelimit()) + WPRINTK("Unfulfilled rx req (id=%d, st=%d).\n", + rx->id, rx->status); + xennet_move_rx_slot(np, skb, ref); + err = -ENOMEM; + goto next; + } + + gnttab_release_grant_reference(&np->gref_rx_head, ref); + + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + /* Remap the page. */ + struct page *page = skb_shinfo(skb)->frags[0].page; + unsigned long pfn = page_to_pfn(page); + void *vaddr = page_address(page); + + MULTI_update_va_mapping(mcl, (unsigned long)vaddr, + pfn_pte_ma(mfn, PAGE_KERNEL), + 0); + mcl++; + mmu->ptr = ((maddr_t)mfn << PAGE_SHIFT) + | MMU_MACHPHYS_UPDATE; + mmu->val = pfn; + mmu++; + + set_phys_to_machine(pfn, mfn); + } + + __skb_queue_tail(list, skb); + +next: + if (!(rx->flags & NETRXF_more_data)) + break; + + if (cons + frags == rp) { + if (net_ratelimit()) + WPRINTK("Need more frags\n"); + err = -ENOENT; + break; + } + + rx = RING_GET_RESPONSE(&np->rx, cons + frags); + skb = xennet_get_rx_skb(np, cons + frags); + ref = xennet_get_rx_ref(np, cons + frags); + frags++; + } + + if (unlikely(frags > max)) { + if (net_ratelimit()) + WPRINTK("Too many frags\n"); + err = -E2BIG; + } + + return err; +} + +static RING_IDX xennet_fill_frags(struct netfront_info *np, + struct sk_buff *skb, + struct sk_buff_head *list) +{ + struct skb_shared_info *shinfo = skb_shinfo(skb); + int nr_frags = shinfo->nr_frags; + RING_IDX cons = np->rx.rsp_cons; + skb_frag_t *frag = shinfo->frags + nr_frags; + struct sk_buff *nskb; + + while ((nskb = __skb_dequeue(list))) { + struct netif_rx_response *rx = + RING_GET_RESPONSE(&np->rx, ++cons); + + frag->page = skb_shinfo(nskb)->frags[0].page; + frag->page_offset = rx->offset; + frag->size = rx->status; + + skb->data_len += rx->status; + + skb_shinfo(nskb)->nr_frags = 0; + kfree_skb(nskb); + + frag++; + nr_frags++; + } + + shinfo->nr_frags = nr_frags; + return cons; +} + +static int xennet_set_skb_gso(struct sk_buff *skb, struct netif_extra_info *gso) +{ + if (!gso->u.gso.size) { + if (net_ratelimit()) + WPRINTK("GSO size must not be zero.\n"); + return -EINVAL; + } + + /* Currently only TCPv4 S.O. is supported. */ + if (gso->u.gso.type != XEN_NETIF_GSO_TYPE_TCPV4) { + if (net_ratelimit()) + WPRINTK("Bad GSO type %d.\n", gso->u.gso.type); + return -EINVAL; + } + + skb_shinfo(skb)->gso_size = gso->u.gso.size; + skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; + + /* Header must be checked, and gso_segs computed. */ + skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY; + skb_shinfo(skb)->gso_segs = 0; + + return 0; +} + +static int netif_poll(struct net_device *dev, int *pbudget) +{ + struct netfront_info *np = netdev_priv(dev); + struct sk_buff *skb; + struct netfront_rx_info rinfo; + struct netif_rx_response *rx = &rinfo.rx; + struct netif_extra_info *extras = rinfo.extras; + RING_IDX i, rp; + struct multicall_entry *mcl; + int work_done, budget, more_to_do = 1; + struct sk_buff_head rxq; + struct sk_buff_head errq; + struct sk_buff_head tmpq; + unsigned long flags; + unsigned int len; + int pages_done; + int err; + + spin_lock(&np->rx_lock); + + if (unlikely(!netif_carrier_ok(dev))) { + spin_unlock(&np->rx_lock); + return 0; + } + + skb_queue_head_init(&rxq); + skb_queue_head_init(&errq); + skb_queue_head_init(&tmpq); + + if ((budget = *pbudget) > dev->quota) + budget = dev->quota; + rp = np->rx.sring->rsp_prod; + rmb(); /* Ensure we see queued responses up to 'rp'. */ + + for (i = np->rx.rsp_cons, work_done = 0, pages_done = 0; + (i != rp) && (work_done < budget); + np->rx.rsp_cons = ++i, work_done++) { + memcpy(rx, RING_GET_RESPONSE(&np->rx, i), sizeof(*rx)); + memset(extras, 0, sizeof(extras)); + + err = xennet_get_responses(np, &rinfo, rp, &tmpq, pages_done); + pages_done += skb_queue_len(&tmpq); + + if (unlikely(err)) { +err: + i = np->rx.rsp_cons + skb_queue_len(&tmpq) - 1; + work_done--; + while ((skb = __skb_dequeue(&tmpq))) + __skb_queue_tail(&errq, skb); + np->stats.rx_errors++; + continue; + } + + skb = __skb_dequeue(&tmpq); + + if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) { + struct netif_extra_info *gso; + gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1]; + + if (unlikely(xennet_set_skb_gso(skb, gso))) { + __skb_queue_head(&tmpq, skb); + goto err; + } + } + + skb->nh.raw = (void *)skb_shinfo(skb)->frags[0].page; + skb->h.raw = skb->nh.raw + rx->offset; + + len = rx->status; + if (len > RX_COPY_THRESHOLD) + len = RX_COPY_THRESHOLD; + skb_put(skb, len); + + if (rx->status > len) { + skb_shinfo(skb)->frags[0].page_offset = + rx->offset + len; + skb_shinfo(skb)->frags[0].size = rx->status - len; + skb->data_len = rx->status - len; + } else { + skb_shinfo(skb)->frags[0].page = NULL; + skb_shinfo(skb)->nr_frags = 0; + } + + i = xennet_fill_frags(np, skb, &tmpq); + skb->truesize += skb->data_len; + skb->len += skb->data_len; + + /* + * Old backends do not assert data_validated but we + * can infer it from csum_blank so test both flags. + */ + if (rx->flags & (NETRXF_data_validated|NETRXF_csum_blank)) { + skb->ip_summed = CHECKSUM_UNNECESSARY; + skb->proto_data_valid = 1; + } else { + skb->ip_summed = CHECKSUM_NONE; + skb->proto_data_valid = 0; + } + skb->proto_csum_blank = !!(rx->flags & NETRXF_csum_blank); + + np->stats.rx_packets++; + np->stats.rx_bytes += skb->len; + + __skb_queue_tail(&rxq, skb); + } + + /* Some pages are no longer absent... */ + balloon_update_driver_allowance(-pages_done); + + /* Do all the remapping work, and M2P updates, in one big hypercall. */ + if (likely(pages_done)) { + mcl = np->rx_mcl + pages_done; + mcl->op = __HYPERVISOR_mmu_update; + mcl->args[0] = (unsigned long)np->rx_mmu; + mcl->args[1] = pages_done; + mcl->args[2] = 0; + mcl->args[3] = DOMID_SELF; + (void)HYPERVISOR_multicall(np->rx_mcl, pages_done + 1); + } + + while ((skb = __skb_dequeue(&errq))) + kfree_skb(skb); + + while ((skb = __skb_dequeue(&rxq)) != NULL) { + struct page *page = (struct page *)skb->nh.raw; + void *vaddr = page_address(page); + + memcpy(skb->data, vaddr + (skb->h.raw - skb->nh.raw), + skb_headlen(skb)); + + if (page != skb_shinfo(skb)->frags[0].page) + __free_page(page); + + /* Ethernet work: Delayed to here as it peeks the header. */ + skb->protocol = eth_type_trans(skb, dev); + + /* Pass it up. */ + netif_receive_skb(skb); + dev->last_rx = jiffies; + } + + /* If we get a callback with very few responses, reduce fill target. */ + /* NB. Note exponential increase, linear decrease. */ + if (((np->rx.req_prod_pvt - np->rx.sring->rsp_prod) > + ((3*np->rx_target) / 4)) && + (--np->rx_target < np->rx_min_target)) + np->rx_target = np->rx_min_target; + + network_alloc_rx_buffers(dev); + + *pbudget -= work_done; + dev->quota -= work_done; + + if (work_done < budget) { + local_irq_save(flags); + + RING_FINAL_CHECK_FOR_RESPONSES(&np->rx, more_to_do); + if (!more_to_do) + __netif_rx_complete(dev); + + local_irq_restore(flags); + } + + spin_unlock(&np->rx_lock); + + return more_to_do; +} + + +static int network_close(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + netif_stop_queue(np->netdev); + return 0; +} + + +static struct net_device_stats *network_get_stats(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + return &np->stats; +} + +static int xennet_change_mtu(struct net_device *dev, int mtu) +{ + int max = xennet_can_sg(dev) ? 65535 - ETH_HLEN : ETH_DATA_LEN; + + if (mtu > max) + return -EINVAL; + dev->mtu = mtu; + return 0; +} + +static int xennet_set_sg(struct net_device *dev, u32 data) +{ + if (data) { + struct netfront_info *np = netdev_priv(dev); + int val; + + if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-sg", + "%d", &val) < 0) + val = 0; + if (!val) + return -ENOSYS; + } else if (dev->mtu > ETH_DATA_LEN) + dev->mtu = ETH_DATA_LEN; + + return ethtool_op_set_sg(dev, data); +} + +static int xennet_set_tso(struct net_device *dev, u32 data) +{ + if (data) { + struct netfront_info *np = netdev_priv(dev); + int val; + + if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, + "feature-gso-tcpv4", "%d", &val) < 0) + val = 0; + if (!val) + return -ENOSYS; + } + + return ethtool_op_set_tso(dev, data); +} + +static void xennet_set_features(struct net_device *dev) +{ + /* Turn off all GSO bits except ROBUST. */ + dev->features &= (1 << NETIF_F_GSO_SHIFT) - 1; + dev->features |= NETIF_F_GSO_ROBUST; + xennet_set_sg(dev, 0); + + if (!xennet_set_sg(dev, 1)) + xennet_set_tso(dev, 1); +} + +static void network_connect(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + int i, requeue_idx; + struct sk_buff *skb; + grant_ref_t ref; + + xennet_set_features(dev); + + spin_lock_irq(&np->tx_lock); + spin_lock(&np->rx_lock); + + /* + * Recovery procedure: + * NB. Freelist index entries are always going to be less than + * PAGE_OFFSET, whereas pointers to skbs will always be equal or + * greater than PAGE_OFFSET: we use this property to distinguish + * them. + */ + + /* Step 1: Discard all pending TX packet fragments. */ + for (requeue_idx = 0, i = 1; i <= NET_TX_RING_SIZE; i++) { + if ((unsigned long)np->tx_skbs[i] < PAGE_OFFSET) + continue; + + skb = np->tx_skbs[i]; + gnttab_end_foreign_access_ref( + np->grant_tx_ref[i], GNTMAP_readonly); + gnttab_release_grant_reference( + &np->gref_tx_head, np->grant_tx_ref[i]); + np->grant_tx_ref[i] = GRANT_INVALID_REF; + add_id_to_freelist(np->tx_skbs, i); + dev_kfree_skb_irq(skb); + } + + /* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */ + for (requeue_idx = 0, i = 0; i < NET_RX_RING_SIZE; i++) { + if (!np->rx_skbs[i]) + continue; + + skb = np->rx_skbs[requeue_idx] = xennet_get_rx_skb(np, i); + ref = np->grant_rx_ref[requeue_idx] = xennet_get_rx_ref(np, i); + + gnttab_grant_foreign_transfer_ref( + ref, np->xbdev->otherend_id, + page_to_pfn(skb_shinfo(skb)->frags->page)); + + RING_GET_REQUEST(&np->rx, requeue_idx)->gref = ref; + RING_GET_REQUEST(&np->rx, requeue_idx)->id = requeue_idx; + + requeue_idx++; + } + + np->rx.req_prod_pvt = requeue_idx; + + /* + * Step 3: All public and private state should now be sane. Get + * ready to start sending and receiving packets and give the driver + * domain a kick because we've probably just requeued some + * packets. + */ + netif_carrier_on(dev); + notify_remote_via_irq(np->irq); + network_tx_buf_gc(dev); + network_alloc_rx_buffers(dev); + + spin_unlock(&np->rx_lock); + spin_unlock_irq(&np->tx_lock); +} + +static void netif_uninit(struct net_device *dev) +{ + struct netfront_info *np = netdev_priv(dev); + gnttab_free_grant_references(np->gref_tx_head); + gnttab_free_grant_references(np->gref_rx_head); +} + +static struct ethtool_ops network_ethtool_ops = +{ + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = ethtool_op_set_tx_csum, + .get_sg = ethtool_op_get_sg, + .set_sg = xennet_set_sg, + .get_tso = ethtool_op_get_tso, + .set_tso = xennet_set_tso, + .get_link = ethtool_op_get_link, +}; + +#ifdef CONFIG_SYSFS +static ssize_t show_rxbuf_min(struct class_device *cd, char *buf) +{ + struct net_device *netdev = container_of(cd, struct net_device, + class_dev); + struct netfront_info *info = netdev_priv(netdev); + + return sprintf(buf, "%u\n", info->rx_min_target); +} + +static ssize_t store_rxbuf_min(struct class_device *cd, + const char *buf, size_t len) +{ + struct net_device *netdev = container_of(cd, struct net_device, + class_dev); + struct netfront_info *np = netdev_priv(netdev); + char *endp; + unsigned long target; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + target = simple_strtoul(buf, &endp, 0); + if (endp == buf) + return -EBADMSG; + + if (target < RX_MIN_TARGET) + target = RX_MIN_TARGET; + if (target > RX_MAX_TARGET) + target = RX_MAX_TARGET; + + spin_lock(&np->rx_lock); + if (target > np->rx_max_target) + np->rx_max_target = target; + np->rx_min_target = target; + if (target > np->rx_target) + np->rx_target = target; + + network_alloc_rx_buffers(netdev); + + spin_unlock(&np->rx_lock); + return len; +} + +static ssize_t show_rxbuf_max(struct class_device *cd, char *buf) +{ + struct net_device *netdev = container_of(cd, struct net_device, + class_dev); + struct netfront_info *info = netdev_priv(netdev); + + return sprintf(buf, "%u\n", info->rx_max_target); +} + +static ssize_t store_rxbuf_max(struct class_device *cd, + const char *buf, size_t len) +{ + struct net_device *netdev = container_of(cd, struct net_device, + class_dev); + struct netfront_info *np = netdev_priv(netdev); + char *endp; + unsigned long target; + + if (!capable(CAP_NET_ADMIN)) + return -EPERM; + + target = simple_strtoul(buf, &endp, 0); + if (endp == buf) + return -EBADMSG; + + if (target < RX_MIN_TARGET) + target = RX_MIN_TARGET; + if (target > RX_MAX_TARGET) + target = RX_MAX_TARGET; + + spin_lock(&np->rx_lock); + if (target < np->rx_min_target) + np->rx_min_target = target; + np->rx_max_target = target; + if (target < np->rx_target) + np->rx_target = target; + + network_alloc_rx_buffers(netdev); + + spin_unlock(&np->rx_lock); + return len; +} + +static ssize_t show_rxbuf_cur(struct class_device *cd, char *buf) +{ + struct net_device *netdev = container_of(cd, struct net_device, + class_dev); + struct netfront_info *info = netdev_priv(netdev); + + return sprintf(buf, "%u\n", info->rx_target); +} + +static const struct class_device_attribute xennet_attrs[] = { + __ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf_min, store_rxbuf_min), + __ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf_max, store_rxbuf_max), + __ATTR(rxbuf_cur, S_IRUGO, show_rxbuf_cur, NULL), +}; + +static int xennet_sysfs_addif(struct net_device *netdev) +{ + int i; + int error = 0; + + for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) { + error = class_device_create_file(&netdev->class_dev, + &xennet_attrs[i]); + if (error) + goto fail; + } + return 0; + + fail: + while (--i >= 0) + class_device_remove_file(&netdev->class_dev, + &xennet_attrs[i]); + return error; +} + +static void xennet_sysfs_delif(struct net_device *netdev) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) { + class_device_remove_file(&netdev->class_dev, + &xennet_attrs[i]); + } +} + +#endif /* CONFIG_SYSFS */ + + +/* + * Nothing to do here. Virtual interface is point-to-point and the + * physical interface is probably promiscuous anyway. + */ +static void network_set_multicast_list(struct net_device *dev) +{ +} + +/** Create a network device. + * @param handle device handle + * @param val return parameter for created device + * @return 0 on success, error code otherwise + */ +static struct net_device * __devinit create_netdev(int handle, + struct xenbus_device *dev) +{ + int i, err = 0; + struct net_device *netdev = NULL; + struct netfront_info *np = NULL; + + netdev = alloc_etherdev(sizeof(struct netfront_info)); + if (!netdev) { + printk(KERN_WARNING "%s> alloc_etherdev failed.\n", + __FUNCTION__); + return ERR_PTR(-ENOMEM); + } + + np = netdev_priv(netdev); + np->handle = handle; + np->xbdev = dev; + + netif_carrier_off(netdev); + + spin_lock_init(&np->tx_lock); + spin_lock_init(&np->rx_lock); + + skb_queue_head_init(&np->rx_batch); + np->rx_target = RX_DFL_MIN_TARGET; + np->rx_min_target = RX_DFL_MIN_TARGET; + np->rx_max_target = RX_MAX_TARGET; + + init_timer(&np->rx_refill_timer); + np->rx_refill_timer.data = (unsigned long)netdev; + np->rx_refill_timer.function = rx_refill_timeout; + + /* Initialise {tx,rx}_skbs as a free chain containing every entry. */ + for (i = 0; i <= NET_TX_RING_SIZE; i++) { + np->tx_skbs[i] = (void *)((unsigned long) i+1); + np->grant_tx_ref[i] = GRANT_INVALID_REF; + } + + for (i = 0; i < NET_RX_RING_SIZE; i++) { + np->rx_skbs[i] = NULL; + np->grant_rx_ref[i] = GRANT_INVALID_REF; + } + + /* A grant for every tx ring slot */ + if (gnttab_alloc_grant_references(TX_MAX_TARGET, + &np->gref_tx_head) < 0) { + printk(KERN_ALERT "#### netfront can't alloc tx grant refs\n"); + err = -ENOMEM; + goto exit; + } + /* A grant for every rx ring slot */ + if (gnttab_alloc_grant_references(RX_MAX_TARGET, + &np->gref_rx_head) < 0) { + printk(KERN_ALERT "#### netfront can't alloc rx grant refs\n"); + err = -ENOMEM; + goto exit_free_tx; + } + + netdev->open = network_open; + netdev->hard_start_xmit = network_start_xmit; + netdev->stop = network_close; + netdev->get_stats = network_get_stats; + netdev->poll = netif_poll; + netdev->set_multicast_list = network_set_multicast_list; + netdev->uninit = netif_uninit; + netdev->change_mtu = xennet_change_mtu; + netdev->weight = 64; + netdev->features = NETIF_F_IP_CSUM; + + SET_ETHTOOL_OPS(netdev, &network_ethtool_ops); + SET_MODULE_OWNER(netdev); + SET_NETDEV_DEV(netdev, &dev->dev); + + err = register_netdev(netdev); + if (err) { + printk(KERN_WARNING "%s> register_netdev err=%d\n", + __FUNCTION__, err); + goto exit_free_rx; + } + + err = xennet_sysfs_addif(netdev); + if (err) { + /* This can be non-fatal: it only means no tuning parameters */ + printk(KERN_WARNING "%s> add sysfs failed err=%d\n", + __FUNCTION__, err); + } + + np->netdev = netdev; + + return netdev; + + + exit_free_rx: + gnttab_free_grant_references(np->gref_rx_head); + exit_free_tx: + gnttab_free_grant_references(np->gref_tx_head); + exit: + free_netdev(netdev); + return ERR_PTR(err); +} + +/* + * We use this notifier to send out a fake ARP reply to reset switches and + * router ARP caches when an IP interface is brought up on a VIF. + */ +static int +inetdev_notify(struct notifier_block *this, unsigned long event, void *ptr) +{ + struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; + struct net_device *dev = ifa->ifa_dev->dev; + + /* UP event and is it one of our devices? */ + if (event == NETDEV_UP && dev->open == network_open) + (void)send_fake_arp(dev); + + return NOTIFY_DONE; +} + + +/* ** Close down ** */ + + +/** + * Handle the change of state of the backend to Closing. We must delete our + * device-layer structures now, to ensure that writes are flushed through to + * the backend. Once is this done, we can switch to Closed in + * acknowledgement. + */ +static void netfront_closing(struct xenbus_device *dev) +{ + struct netfront_info *info = dev->dev.driver_data; + + DPRINTK("netfront_closing: %s removed\n", dev->nodename); + + close_netdev(info); + + xenbus_switch_state(dev, XenbusStateClosed); +} + + +static int __devexit netfront_remove(struct xenbus_device *dev) +{ + struct netfront_info *info = dev->dev.driver_data; + + DPRINTK("%s\n", dev->nodename); + + netif_disconnect_backend(info); + free_netdev(info->netdev); + + return 0; +} + + +static void close_netdev(struct netfront_info *info) +{ + del_timer_sync(&info->rx_refill_timer); + + xennet_sysfs_delif(info->netdev); + unregister_netdev(info->netdev); +} + + +static void netif_disconnect_backend(struct netfront_info *info) +{ + /* Stop old i/f to prevent errors whilst we rebuild the state. */ + spin_lock_irq(&info->tx_lock); + spin_lock(&info->rx_lock); + netif_carrier_off(info->netdev); + spin_unlock(&info->rx_lock); + spin_unlock_irq(&info->tx_lock); + + if (info->irq) + unbind_from_irqhandler(info->irq, info->netdev); + info->evtchn = info->irq = 0; + + end_access(info->tx_ring_ref, info->tx.sring); + end_access(info->rx_ring_ref, info->rx.sring); + info->tx_ring_ref = GRANT_INVALID_REF; + info->rx_ring_ref = GRANT_INVALID_REF; + info->tx.sring = NULL; + info->rx.sring = NULL; +} + + +static void netif_free(struct netfront_info *info) +{ + close_netdev(info); + netif_disconnect_backend(info); + free_netdev(info->netdev); +} + + +static void end_access(int ref, void *page) +{ + if (ref != GRANT_INVALID_REF) + gnttab_end_foreign_access(ref, 0, (unsigned long)page); +} + + +/* ** Driver registration ** */ + + +static struct xenbus_device_id netfront_ids[] = { + { "vif" }, + { "" } +}; + + +static struct xenbus_driver netfront = { + .name = "vif", + .owner = THIS_MODULE, + .ids = netfront_ids, + .probe = netfront_probe, + .remove = __devexit_p(netfront_remove), + .resume = netfront_resume, + .otherend_changed = backend_changed, +}; + + +static struct notifier_block notifier_inetdev = { + .notifier_call = inetdev_notify, + .next = NULL, + .priority = 0 +}; + +static int __init netif_init(void) +{ + if (!is_running_on_xen()) + return -ENODEV; + + if (xen_start_info->flags & SIF_INITDOMAIN) + return 0; + + IPRINTK("Initialising virtual ethernet driver.\n"); + + (void)register_inetaddr_notifier(¬ifier_inetdev); + + return xenbus_register_frontend(&netfront); +} +module_init(netif_init); + + +static void __exit netif_exit(void) +{ + unregister_inetaddr_notifier(¬ifier_inetdev); + + return xenbus_unregister_driver(&netfront); +} +module_exit(netif_exit); + +MODULE_LICENSE("Dual BSD/GPL"); diff --git a/drivers/xen/privcmd/Makefile b/drivers/xen/privcmd/Makefile new file mode 100644 index 000000000..de090ba6b --- /dev/null +++ b/drivers/xen/privcmd/Makefile @@ -0,0 +1,2 @@ + +obj-$(CONFIG_XEN_PRIVCMD) := privcmd.o diff --git a/drivers/xen/privcmd/privcmd.c b/drivers/xen/privcmd/privcmd.c new file mode 100644 index 000000000..3c9bb9f0d --- /dev/null +++ b/drivers/xen/privcmd/privcmd.c @@ -0,0 +1,286 @@ +/****************************************************************************** + * privcmd.c + * + * Interface to privileged domain-0 commands. + * + * Copyright (c) 2002-2004, K A Fraser, B Dragovic + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#include + +static struct proc_dir_entry *privcmd_intf; +static struct proc_dir_entry *capabilities_intf; + +#define NR_HYPERCALLS 64 +static DECLARE_BITMAP(hypercall_permission_map, NR_HYPERCALLS); + +static int privcmd_ioctl(struct inode *inode, struct file *file, + unsigned int cmd, unsigned long data) +{ + int ret = -ENOSYS; + void __user *udata = (void __user *) data; + + switch (cmd) { + case IOCTL_PRIVCMD_HYPERCALL: { + privcmd_hypercall_t hypercall; + + if (copy_from_user(&hypercall, udata, sizeof(hypercall))) + return -EFAULT; + + /* Check hypercall number for validity. */ + if (hypercall.op >= NR_HYPERCALLS) + return -EINVAL; + if (!test_bit(hypercall.op, hypercall_permission_map)) + return -EINVAL; + +#if defined(__i386__) + __asm__ __volatile__ ( + "pushl %%ebx; pushl %%ecx; pushl %%edx; " + "pushl %%esi; pushl %%edi; " + "movl 8(%%eax),%%ebx ;" + "movl 16(%%eax),%%ecx ;" + "movl 24(%%eax),%%edx ;" + "movl 32(%%eax),%%esi ;" + "movl 40(%%eax),%%edi ;" + "movl (%%eax),%%eax ;" + "shll $5,%%eax ;" + "addl $hypercall_page,%%eax ;" + "call *%%eax ;" + "popl %%edi; popl %%esi; popl %%edx; " + "popl %%ecx; popl %%ebx" + : "=a" (ret) : "0" (&hypercall) : "memory" ); +#elif defined (__x86_64__) + { + long ign1, ign2, ign3; + __asm__ __volatile__ ( + "movq %8,%%r10; movq %9,%%r8;" + "shlq $5,%%rax ;" + "addq $hypercall_page,%%rax ;" + "call *%%rax" + : "=a" (ret), "=D" (ign1), + "=S" (ign2), "=d" (ign3) + : "0" ((unsigned long)hypercall.op), + "1" ((unsigned long)hypercall.arg[0]), + "2" ((unsigned long)hypercall.arg[1]), + "3" ((unsigned long)hypercall.arg[2]), + "g" ((unsigned long)hypercall.arg[3]), + "g" ((unsigned long)hypercall.arg[4]) + : "r8", "r10", "memory" ); + } +#elif defined (__ia64__) + __asm__ __volatile__ ( + ";; mov r14=%2; mov r15=%3; " + "mov r16=%4; mov r17=%5; mov r18=%6;" + "mov r2=%1; break 0x1000;; mov %0=r8 ;;" + : "=r" (ret) + : "r" (hypercall.op), + "r" (hypercall.arg[0]), + "r" (hypercall.arg[1]), + "r" (hypercall.arg[2]), + "r" (hypercall.arg[3]), + "r" (hypercall.arg[4]) + : "r14","r15","r16","r17","r18","r2","r8","memory"); +#endif + } + break; + +#if defined(CONFIG_XEN_PRIVILEGED_GUEST) + case IOCTL_PRIVCMD_MMAP: { +#define PRIVCMD_MMAP_SZ 32 + privcmd_mmap_t mmapcmd; + privcmd_mmap_entry_t msg[PRIVCMD_MMAP_SZ]; + privcmd_mmap_entry_t __user *p; + int i, rc; + + if (copy_from_user(&mmapcmd, udata, sizeof(mmapcmd))) + return -EFAULT; + + p = mmapcmd.entry; + + for (i = 0; i < mmapcmd.num; + i += PRIVCMD_MMAP_SZ, p += PRIVCMD_MMAP_SZ) { + int j, n = ((mmapcmd.num-i)>PRIVCMD_MMAP_SZ)? + PRIVCMD_MMAP_SZ:(mmapcmd.num-i); + + if (copy_from_user(&msg, p, + n*sizeof(privcmd_mmap_entry_t))) + return -EFAULT; + + for (j = 0; j < n; j++) { + struct vm_area_struct *vma = + find_vma( current->mm, msg[j].va ); + + if (!vma) + return -EINVAL; + + if (msg[j].va > PAGE_OFFSET) + return -EINVAL; + + if ((msg[j].va + (msg[j].npages << PAGE_SHIFT)) + > vma->vm_end ) + return -EINVAL; + + if ((rc = direct_remap_pfn_range( + vma, + msg[j].va&PAGE_MASK, + msg[j].mfn, + msg[j].npages<vm_page_prot, + mmapcmd.dom)) < 0) + return rc; + } + } + ret = 0; + } + break; + + case IOCTL_PRIVCMD_MMAPBATCH: { + privcmd_mmapbatch_t m; + struct vm_area_struct *vma = NULL; + xen_pfn_t __user *p; + unsigned long addr, mfn; + int i; + + if (copy_from_user(&m, udata, sizeof(m))) { + ret = -EFAULT; + goto batch_err; + } + + if (m.dom == DOMID_SELF) { + ret = -EINVAL; + goto batch_err; + } + + vma = find_vma(current->mm, m.addr); + if (!vma) { + ret = -EINVAL; + goto batch_err; + } + + if (m.addr > PAGE_OFFSET) { + ret = -EFAULT; + goto batch_err; + } + + if ((m.addr + (m.num< vma->vm_end) { + ret = -EFAULT; + goto batch_err; + } + + p = m.arr; + addr = m.addr; + for (i = 0; i < m.num; i++, addr += PAGE_SIZE, p++) { + if (get_user(mfn, p)) + return -EFAULT; + + ret = direct_remap_pfn_range(vma, addr & PAGE_MASK, + mfn, PAGE_SIZE, + vma->vm_page_prot, m.dom); + if (ret < 0) + put_user(0xF0000000 | mfn, p); + } + + ret = 0; + break; + + batch_err: + printk("batch_err ret=%d vma=%p addr=%lx " + "num=%d arr=%p %lx-%lx\n", + ret, vma, (unsigned long)m.addr, m.num, m.arr, + vma ? vma->vm_start : 0, vma ? vma->vm_end : 0); + break; + } + break; +#endif + + default: + ret = -EINVAL; + break; + } + + return ret; +} + +#ifndef HAVE_ARCH_PRIVCMD_MMAP +static int privcmd_mmap(struct file * file, struct vm_area_struct * vma) +{ + /* DONTCOPY is essential for Xen as copy_page_range is broken. */ + vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY; + + return 0; +} +#endif + +static struct file_operations privcmd_file_ops = { + .ioctl = privcmd_ioctl, + .mmap = privcmd_mmap, +}; + +static int capabilities_read(char *page, char **start, off_t off, + int count, int *eof, void *data) +{ + int len = 0; + *page = 0; + + if (xen_start_info->flags & SIF_INITDOMAIN) + len = sprintf( page, "control_d\n" ); + + *eof = 1; + return len; +} + +static int __init privcmd_init(void) +{ + if (!is_running_on_xen()) + return -ENODEV; + + /* Set of hypercalls that privileged applications may execute. */ + set_bit(__HYPERVISOR_acm_op, hypercall_permission_map); + set_bit(__HYPERVISOR_dom0_op, hypercall_permission_map); + set_bit(__HYPERVISOR_event_channel_op, hypercall_permission_map); + set_bit(__HYPERVISOR_memory_op, hypercall_permission_map); + set_bit(__HYPERVISOR_mmu_update, hypercall_permission_map); + set_bit(__HYPERVISOR_mmuext_op, hypercall_permission_map); + set_bit(__HYPERVISOR_xen_version, hypercall_permission_map); + set_bit(__HYPERVISOR_sched_op, hypercall_permission_map); + set_bit(__HYPERVISOR_sched_op_compat, hypercall_permission_map); + set_bit(__HYPERVISOR_event_channel_op_compat, + hypercall_permission_map); + set_bit(__HYPERVISOR_hvm_op, hypercall_permission_map); + + privcmd_intf = create_xen_proc_entry("privcmd", 0400); + if (privcmd_intf != NULL) + privcmd_intf->proc_fops = &privcmd_file_ops; + + capabilities_intf = create_xen_proc_entry("capabilities", 0400 ); + if (capabilities_intf != NULL) + capabilities_intf->read_proc = capabilities_read; + + return 0; +} + +__initcall(privcmd_init); diff --git a/fs/Kconfig b/fs/Kconfig index 0167990d8..1d0cf4054 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -842,6 +842,7 @@ config TMPFS config HUGETLBFS bool "HugeTLB file system support" depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || BROKEN + depends !XEN help hugetlbfs is a filesystem backing for HugeTLB pages, based on ramfs. For architectures that support it, say Y here and read diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 68ebd10f3..aa50dc692 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -512,7 +512,11 @@ befs_utf2nls(struct super_block *sb, const char *in, wchar_t uni; int unilen, utflen; char *result; - int maxlen = in_len; /* The utf8->nls conversion can't make more chars */ + /* The utf8->nls conversion won't make the final nls string bigger + * than the utf one, but if the string is pure ascii they'll have the + * same width and an extra char is needed to save the additional \0 + */ + int maxlen = in_len + 1; befs_debug(sb, "---> utf2nls()"); @@ -588,7 +592,10 @@ befs_nls2utf(struct super_block *sb, const char *in, wchar_t uni; int unilen, utflen; char *result; - int maxlen = 3 * in_len; + /* There're nls characters that will translate to 3-chars-wide UTF-8 + * characters, a additional byte is needed to save the final \0 + * in special cases */ + int maxlen = (3 * in_len) + 1; befs_debug(sb, "---> nls2utf()\n"); diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 093645b22..479ee5fde 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -88,7 +88,7 @@ static struct linux_binfmt elf_format = { .min_coredump = ELF_EXEC_PAGESIZE }; -#define BAD_ADDR(x) ((unsigned long)(x) > PAGE_MASK) +#define BAD_ADDR(x) ((unsigned long)(x) >= PAGE_MASK) static int set_brk(unsigned long start, unsigned long end) { @@ -441,7 +441,7 @@ static unsigned long load_elf_interp(struct elfhdr * interp_elf_ex, * <= p_memsize so it is only necessary to check p_memsz. */ k = load_addr + eppnt->p_vaddr; - if (k > TASK_SIZE || eppnt->p_filesz > eppnt->p_memsz || + if (BAD_ADDR(k) || eppnt->p_filesz > eppnt->p_memsz || eppnt->p_memsz > TASK_SIZE || TASK_SIZE - eppnt->p_memsz < k) { error = -ENOMEM; goto out_close; @@ -944,7 +944,7 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) * allowed task size. Note that p_filesz must always be * <= p_memsz so it is only necessary to check p_memsz. */ - if (k > TASK_SIZE || elf_ppnt->p_filesz > elf_ppnt->p_memsz || + if (BAD_ADDR(k) || elf_ppnt->p_filesz > elf_ppnt->p_memsz || elf_ppnt->p_memsz > TASK_SIZE || TASK_SIZE - elf_ppnt->p_memsz < k) { /* set_brk can never work. Avoid overflows. */ @@ -1005,10 +1005,9 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) } } if (BAD_ADDR(elf_entry)) { - printk(KERN_ERR "Unable to load interpreter %.128s\n", - elf_interpreter); force_sig(SIGSEGV, current); - retval = -ENOEXEC; /* Nobody gets to see this, but.. */ + retval = IS_ERR((void *)elf_entry) ? + (int)elf_entry : -EINVAL; goto out_free_dentry; } reloc_func_desc = interp_load_addr; @@ -1019,8 +1018,8 @@ static int load_elf_binary(struct linux_binprm * bprm, struct pt_regs * regs) } else { elf_entry = loc->elf_ex.e_entry; if (BAD_ADDR(elf_entry)) { - send_sig(SIGSEGV, current, 0); - retval = -ENOEXEC; /* Nobody gets to see this, but.. */ + force_sig(SIGSEGV, current); + retval = -EINVAL; goto out_free_dentry; } } diff --git a/fs/buffer.c b/fs/buffer.c index 23f1f3a68..7f6d65925 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -473,13 +473,18 @@ out: pass does the actual I/O. */ void invalidate_bdev(struct block_device *bdev, int destroy_dirty_buffers) { + struct address_space *mapping = bdev->bd_inode->i_mapping; + + if (mapping->nrpages == 0) + return; + invalidate_bh_lrus(); /* * FIXME: what about destroy_dirty_buffers? * We really want to use invalidate_inode_pages2() for * that, but not until that's cleaned up. */ - invalidate_inode_pages(bdev->bd_inode->i_mapping); + invalidate_inode_pages(mapping); } /* diff --git a/fs/ext2/super.c b/fs/ext2/super.c index f05b2cff4..22993252e 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -252,6 +252,46 @@ static struct super_operations ext2_sops = { #endif }; +static struct dentry *ext2_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino != EXT2_ROOT_INO && ino < EXT2_FIRST_INO(sb)) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT2_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * ext2_read_inode currently does appropriate checks, but + * it might be "neater" to call ext2_get_inode first and check + * if the inode is valid..... + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) + || (generation && inode->i_generation != generation) + ) { + /* we didn't find the right inode.. */ + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + + /* Yes, most of these are left as NULL!! * A NULL value implies the default, which works with ext2-like file * systems, but can be improved upon. @@ -259,6 +299,7 @@ static struct super_operations ext2_sops = { */ static struct export_operations ext2_export_ops = { .get_parent = ext2_get_parent, + .get_dentry = ext2_get_dentry, }; static unsigned long get_sb_block(void **data) diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index bdd65f45d..4ce217b6d 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1160,7 +1160,7 @@ retry: ret = PTR_ERR(handle); goto out; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_prepare_write(page, from, to, ext3_get_block); else ret = block_prepare_write(page, from, to, ext3_get_block); @@ -1246,7 +1246,7 @@ static int ext3_writeback_commit_write(struct file *file, struct page *page, if (new_i_size > EXT3_I(inode)->i_disksize) EXT3_I(inode)->i_disksize = new_i_size; - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_commit_write(file, page, from, to); else ret = generic_commit_write(file, page, from, to); @@ -1496,7 +1496,7 @@ static int ext3_writeback_writepage(struct page *page, goto out_fail; } - if (test_opt(inode->i_sb, NOBH)) + if (test_opt(inode->i_sb, NOBH) && ext3_should_writeback_data(inode)) ret = nobh_writepage(page, ext3_get_block, wbc); else ret = block_write_full_page(page, ext3_get_block, wbc); @@ -2403,14 +2403,15 @@ static unsigned long ext3_get_inode_block(struct super_block *sb, struct buffer_head *bh; struct ext3_group_desc * gdp; - - if ((ino != EXT3_ROOT_INO && ino != EXT3_JOURNAL_INO && - ino != EXT3_RESIZE_INO && ino < EXT3_FIRST_INO(sb)) || - ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) { - ext3_error(sb, "ext3_get_inode_block", - "bad inode number: %lu", ino); + if (!ext3_valid_inum(sb, ino)) { + /* + * This error is already checked for in namei.c unless we are + * looking at an NFS filehandle, in which case no error + * report is needed + */ return 0; } + block_group = (ino - 1) / EXT3_INODES_PER_GROUP(sb); if (block_group >= EXT3_SB(sb)->s_groups_count) { ext3_error(sb,"ext3_get_inode_block","group >= groups count"); diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 329eae367..045f58812 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1001,7 +1001,12 @@ static struct dentry *ext3_lookup(struct inode * dir, struct dentry *dentry, str if (bh) { unsigned long ino = le32_to_cpu(de->inode); brelse (bh); - inode = iget(dir->i_sb, ino); + if (!ext3_valid_inum(dir->i_sb, ino)) { + ext3_error(dir->i_sb, "ext3_lookup", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(dir->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); @@ -1030,7 +1035,13 @@ struct dentry *ext3_get_parent(struct dentry *child) return ERR_PTR(-ENOENT); ino = le32_to_cpu(de->inode); brelse(bh); - inode = iget(child->d_inode->i_sb, ino); + + if (!ext3_valid_inum(child->d_inode->i_sb, ino)) { + ext3_error(child->d_inode->i_sb, "ext3_get_parent", + "bad inode number: %lu", ino); + inode = NULL; + } else + inode = iget(child->d_inode->i_sb, ino); if (!inode) return ERR_PTR(-EACCES); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 2a0cb4ca8..ed22106c9 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -620,8 +620,48 @@ static struct super_operations ext3_sops = { #endif }; +static struct dentry *ext3_get_dentry(struct super_block *sb, void *vobjp) +{ + __u32 *objp = vobjp; + unsigned long ino = objp[0]; + __u32 generation = objp[1]; + struct inode *inode; + struct dentry *result; + + if (ino != EXT3_ROOT_INO && ino < EXT3_FIRST_INO(sb)) + return ERR_PTR(-ESTALE); + if (ino > le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)) + return ERR_PTR(-ESTALE); + + /* iget isn't really right if the inode is currently unallocated!! + * ext3_read_inode currently does appropriate checks, but + * it might be "neater" to call ext3_get_inode first and check + * if the inode is valid..... + */ + inode = iget(sb, ino); + if (inode == NULL) + return ERR_PTR(-ENOMEM); + if (is_bad_inode(inode) + || (generation && inode->i_generation != generation) + ) { + /* we didn't find the right inode.. */ + iput(inode); + return ERR_PTR(-ESTALE); + } + /* now to find a dentry. + * If possible, get a well-connected one + */ + result = d_alloc_anon(inode); + if (!result) { + iput(inode); + return ERR_PTR(-ENOMEM); + } + return result; +} + static struct export_operations ext3_export_ops = { .get_parent = ext3_get_parent, + .get_dentry = ext3_get_dentry, }; enum { diff --git a/fs/file.c b/fs/file.c index 55f4e7022..4c2cacca4 100644 --- a/fs/file.c +++ b/fs/file.c @@ -277,11 +277,13 @@ static struct fdtable *alloc_fdtable(int nr) } while (nfds <= nr); new_fds = alloc_fd_array(nfds); if (!new_fds) - goto out; + goto out2; fdt->fd = new_fds; fdt->max_fds = nfds; fdt->free_files = NULL; return fdt; +out2: + nfds = fdt->max_fdset; out: if (new_openset) free_fdset(new_openset, nfds); diff --git a/fs/locks.c b/fs/locks.c index 7ec121cb3..718983409 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1409,8 +1409,9 @@ static int __setlease(struct file *filp, long arg, struct file_lock **flp) if (!leases_enable) goto out; - error = lease_alloc(filp, arg, &fl); - if (error) + error = -ENOMEM; + fl = locks_alloc_lock(); + if (fl == NULL) goto out; locks_copy_lock(fl, lease); @@ -1418,6 +1419,7 @@ static int __setlease(struct file *filp, long arg, struct file_lock **flp) locks_insert_lock(before, fl); *flp = fl; + error = 0; out: return error; } diff --git a/fs/namei.c b/fs/namei.c index 4a36030c8..75649bf3e 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1791,8 +1791,14 @@ do_link: if (error) goto exit_dput; error = __do_follow_link(&path, nd); - if (error) + if (error) { + /* Does someone understand code flow here? Or it is only + * me so stupid? Anathema to whoever designed this non-sense + * with "intent.open". + */ + release_open_intent(nd); return error; + } nd->flags &= ~LOOKUP_PARENT; if (nd->last_type == LAST_BIND) goto ok; diff --git a/fs/proc/base.c b/fs/proc/base.c index 8c880128d..f6e0c6274 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -631,6 +631,27 @@ static int proc_permission(struct inode *inode, int mask, struct nameidata *nd) return proc_check_root(inode); } +static int proc_setattr(struct dentry *dentry, struct iattr *attr) +{ + int error; + struct inode *inode = dentry->d_inode; + + if (attr->ia_valid & ATTR_MODE) + return -EPERM; + + error = inode_change_ok(inode, attr); + if (!error) { + error = security_inode_setattr(dentry, attr); + if (!error) + error = inode_setattr(inode, attr); + } + return error; +} + +static struct inode_operations proc_def_inode_operations = { + .setattr = proc_setattr, +}; + static int proc_task_permission(struct inode *inode, int mask, struct nameidata *nd) { struct dentry *root; @@ -1022,6 +1043,7 @@ static struct file_operations proc_oom_adjust_operations = { static struct inode_operations proc_mem_inode_operations = { .permission = proc_permission, + .setattr = proc_setattr, }; #ifdef CONFIG_AUDITSYSCALL @@ -1219,7 +1241,8 @@ out: static struct inode_operations proc_pid_link_inode_operations = { .readlink = proc_pid_readlink, - .follow_link = proc_pid_follow_link + .follow_link = proc_pid_follow_link, + .setattr = proc_setattr, }; #define NUMBUF 10 @@ -1400,6 +1423,7 @@ static struct inode *proc_pid_make_inode(struct super_block * sb, struct task_st ei->task = NULL; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_ino = fake_ino(task->pid, ino); + inode->i_op = &proc_def_inode_operations; if (!pid_alive(task)) goto out_unlock; @@ -1455,6 +1479,7 @@ static int pid_revalidate(struct dentry *dentry, struct nameidata *nd) inode->i_uid = 0; inode->i_gid = 0; } + inode->i_mode &= ~(S_ISUID | S_ISGID); security_task_to_inode(task, inode); return 1; } @@ -1483,6 +1508,7 @@ static int tid_fd_revalidate(struct dentry *dentry, struct nameidata *nd) inode->i_uid = 0; inode->i_gid = 0; } + inode->i_mode &= ~(S_ISUID | S_ISGID); security_task_to_inode(task, inode); return 1; } @@ -1628,11 +1654,13 @@ static struct file_operations proc_task_operations = { static struct inode_operations proc_fd_inode_operations = { .lookup = proc_lookupfd, .permission = proc_permission, + .setattr = proc_setattr, }; static struct inode_operations proc_task_inode_operations = { .lookup = proc_task_lookup, .permission = proc_task_permission, + .setattr = proc_setattr, }; #ifdef CONFIG_SECURITY @@ -1943,10 +1971,12 @@ static struct file_operations proc_tid_base_operations = { static struct inode_operations proc_tgid_base_inode_operations = { .lookup = proc_tgid_base_lookup, + .setattr = proc_setattr, }; static struct inode_operations proc_tid_base_inode_operations = { .lookup = proc_tid_base_lookup, + .setattr = proc_setattr, }; #ifdef CONFIG_SECURITY @@ -1988,10 +2018,12 @@ static struct dentry *proc_tid_attr_lookup(struct inode *dir, static struct inode_operations proc_tgid_attr_inode_operations = { .lookup = proc_tgid_attr_lookup, + .setattr = proc_setattr, }; static struct inode_operations proc_tid_attr_inode_operations = { .lookup = proc_tid_attr_lookup, + .setattr = proc_setattr, }; #endif @@ -2016,6 +2048,7 @@ static void *proc_self_follow_link(struct dentry *dentry, struct nameidata *nd) static struct inode_operations proc_self_inode_operations = { .readlink = proc_self_readlink, .follow_link = proc_self_follow_link, + .setattr = proc_setattr, }; /** diff --git a/fs/splice.c b/fs/splice.c index a285fd746..8fef6672b 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1294,6 +1294,85 @@ asmlinkage long sys_splice(int fd_in, loff_t __user *off_in, return error; } +/* + * Make sure there's data to read. Wait for input if we can, otherwise + * return an appropriate error. + */ +static int link_ipipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs) + return 0; + + ret = 0; + mutex_lock(&pipe->inode->i_mutex); + + while (!pipe->nrbufs) { + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + if (!pipe->writers) + break; + if (!pipe->waiting_writers) { + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + } + pipe_wait(pipe); + } + + mutex_unlock(&pipe->inode->i_mutex); + return ret; +} + +/* + * Make sure there's writeable room. Wait for room if we can, otherwise + * return an appropriate error. + */ +static int link_opipe_prep(struct pipe_inode_info *pipe, unsigned int flags) +{ + int ret; + + /* + * Check ->nrbufs without the inode lock first. This function + * is speculative anyways, so missing one is ok. + */ + if (pipe->nrbufs < PIPE_BUFFERS) + return 0; + + ret = 0; + mutex_lock(&pipe->inode->i_mutex); + + while (pipe->nrbufs >= PIPE_BUFFERS) { + if (!pipe->readers) { + send_sig(SIGPIPE, current, 0); + ret = -EPIPE; + break; + } + if (flags & SPLICE_F_NONBLOCK) { + ret = -EAGAIN; + break; + } + if (signal_pending(current)) { + ret = -ERESTARTSYS; + break; + } + pipe->waiting_writers++; + pipe_wait(pipe); + pipe->waiting_writers--; + } + + mutex_unlock(&pipe->inode->i_mutex); + return ret; +} + /* * Link contents of ipipe to opipe. */ @@ -1302,9 +1381,7 @@ static int link_pipe(struct pipe_inode_info *ipipe, size_t len, unsigned int flags) { struct pipe_buffer *ibuf, *obuf; - int ret, do_wakeup, i, ipipe_first; - - ret = do_wakeup = ipipe_first = 0; + int ret = 0, i = 0, nbuf; /* * Potential ABBA deadlock, work around it by ordering lock @@ -1312,7 +1389,6 @@ static int link_pipe(struct pipe_inode_info *ipipe, * could deadlock (one doing tee from A -> B, the other from B -> A). */ if (ipipe->inode < opipe->inode) { - ipipe_first = 1; mutex_lock(&ipipe->inode->i_mutex); mutex_lock(&opipe->inode->i_mutex); } else { @@ -1320,118 +1396,55 @@ static int link_pipe(struct pipe_inode_info *ipipe, mutex_lock(&ipipe->inode->i_mutex); } - for (i = 0;; i++) { + do { if (!opipe->readers) { send_sig(SIGPIPE, current, 0); if (!ret) ret = -EPIPE; break; } - if (ipipe->nrbufs - i) { - ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1)); - /* - * If we have room, fill this buffer - */ - if (opipe->nrbufs < PIPE_BUFFERS) { - int nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1); - - /* - * Get a reference to this pipe buffer, - * so we can copy the contents over. - */ - ibuf->ops->get(ipipe, ibuf); - - obuf = opipe->bufs + nbuf; - *obuf = *ibuf; - - /* - * Don't inherit the gift flag, we need to - * prevent multiple steals of this page. - */ - obuf->flags &= ~PIPE_BUF_FLAG_GIFT; - - if (obuf->len > len) - obuf->len = len; - - opipe->nrbufs++; - do_wakeup = 1; - ret += obuf->len; - len -= obuf->len; - - if (!len) - break; - if (opipe->nrbufs < PIPE_BUFFERS) - continue; - } - - /* - * We have input available, but no output room. - * If we already copied data, return that. If we - * need to drop the opipe lock, it must be ordered - * last to avoid deadlocks. - */ - if ((flags & SPLICE_F_NONBLOCK) || !ipipe_first) { - if (!ret) - ret = -EAGAIN; - break; - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - break; - } - if (do_wakeup) { - smp_mb(); - if (waitqueue_active(&opipe->wait)) - wake_up_interruptible(&opipe->wait); - kill_fasync(&opipe->fasync_readers, SIGIO, POLL_IN); - do_wakeup = 0; - } + /* + * If we have iterated all input buffers or ran out of + * output room, break. + */ + if (i >= ipipe->nrbufs || opipe->nrbufs >= PIPE_BUFFERS) + break; - opipe->waiting_writers++; - pipe_wait(opipe); - opipe->waiting_writers--; - continue; - } + ibuf = ipipe->bufs + ((ipipe->curbuf + i) & (PIPE_BUFFERS - 1)); + nbuf = (opipe->curbuf + opipe->nrbufs) & (PIPE_BUFFERS - 1); /* - * No input buffers, do the usual checks for available - * writers and blocking and wait if necessary + * Get a reference to this pipe buffer, + * so we can copy the contents over. */ - if (!ipipe->writers) - break; - if (!ipipe->waiting_writers) { - if (ret) - break; - } + ibuf->ops->get(ipipe, ibuf); + + obuf = opipe->bufs + nbuf; + *obuf = *ibuf; + /* - * pipe_wait() drops the ipipe mutex. To avoid deadlocks - * with another process, we can only safely do that if - * the ipipe lock is ordered last. + * Don't inherit the gift flag, we need to + * prevent multiple steals of this page. */ - if ((flags & SPLICE_F_NONBLOCK) || ipipe_first) { - if (!ret) - ret = -EAGAIN; - break; - } - if (signal_pending(current)) { - if (!ret) - ret = -ERESTARTSYS; - break; - } + obuf->flags &= ~PIPE_BUF_FLAG_GIFT; - if (waitqueue_active(&ipipe->wait)) - wake_up_interruptible_sync(&ipipe->wait); - kill_fasync(&ipipe->fasync_writers, SIGIO, POLL_OUT); + if (obuf->len > len) + obuf->len = len; - pipe_wait(ipipe); - } + opipe->nrbufs++; + ret += obuf->len; + len -= obuf->len; + i++; + } while (len); mutex_unlock(&ipipe->inode->i_mutex); mutex_unlock(&opipe->inode->i_mutex); - if (do_wakeup) { + /* + * If we put data in the output pipe, wakeup any potential readers. + */ + if (ret > 0) { smp_mb(); if (waitqueue_active(&opipe->wait)) wake_up_interruptible(&opipe->wait); @@ -1452,14 +1465,29 @@ static long do_tee(struct file *in, struct file *out, size_t len, { struct pipe_inode_info *ipipe = in->f_dentry->d_inode->i_pipe; struct pipe_inode_info *opipe = out->f_dentry->d_inode->i_pipe; + int ret = -EINVAL; /* - * Link ipipe to the two output pipes, consuming as we go along. + * Duplicate the contents of ipipe to opipe without actually + * copying the data. */ - if (ipipe && opipe) - return link_pipe(ipipe, opipe, len, flags); + if (ipipe && opipe && ipipe != opipe) { + /* + * Keep going, unless we encounter an error. The ipipe/opipe + * ordering doesn't really matter. + */ + ret = link_ipipe_prep(ipipe, flags); + if (!ret) { + ret = link_opipe_prep(opipe, flags); + if (!ret) { + ret = link_pipe(ipipe, opipe, len, flags); + if (!ret && (flags & SPLICE_F_NONBLOCK)) + ret = -EAGAIN; + } + } + } - return -EINVAL; + return ret; } asmlinkage long sys_tee(int fdin, int fdout, size_t len, unsigned int flags) diff --git a/fs/udf/super.c b/fs/udf/super.c index e45789fe3..73fc0d8ae 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -1653,7 +1653,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent) iput(inode); goto error_out; } - sb->s_maxbytes = MAX_LFS_FILESIZE; + sb->s_maxbytes = 1<<30; return 0; error_out: diff --git a/fs/udf/truncate.c b/fs/udf/truncate.c index e1b0e8cfe..0abd66ce3 100644 --- a/fs/udf/truncate.c +++ b/fs/udf/truncate.c @@ -239,37 +239,51 @@ void udf_truncate_extents(struct inode * inode) { if (offset) { - extoffset -= adsize; - etype = udf_next_aext(inode, &bloc, &extoffset, &eloc, &elen, &bh, 1); - if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) - { - extoffset -= adsize; - elen = EXT_NOT_RECORDED_NOT_ALLOCATED | (elen + offset); - udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 0); + /* + * OK, there is not extent covering inode->i_size and + * no extent above inode->i_size => truncate is + * extending the file by 'offset'. + */ + if ((!bh && extoffset == udf_file_entry_alloc_offset(inode)) || + (bh && extoffset == sizeof(struct allocExtDesc))) { + /* File has no extents at all! */ + memset(&eloc, 0x00, sizeof(kernel_lb_addr)); + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; + udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } - else if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) - { - kernel_lb_addr neloc = { 0, 0 }; + else { extoffset -= adsize; - nelen = EXT_NOT_RECORDED_NOT_ALLOCATED | - ((elen + offset + inode->i_sb->s_blocksize - 1) & - ~(inode->i_sb->s_blocksize - 1)); - udf_write_aext(inode, bloc, &extoffset, neloc, nelen, bh, 1); - udf_add_aext(inode, &bloc, &extoffset, eloc, (etype << 30) | elen, &bh, 1); - } - else - { - if (elen & (inode->i_sb->s_blocksize - 1)) + etype = udf_next_aext(inode, &bloc, &extoffset, &eloc, &elen, &bh, 1); + if (etype == (EXT_NOT_RECORDED_NOT_ALLOCATED >> 30)) + { + extoffset -= adsize; + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | (elen + offset); + udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 0); + } + else if (etype == (EXT_NOT_RECORDED_ALLOCATED >> 30)) { + kernel_lb_addr neloc = { 0, 0 }; extoffset -= adsize; - elen = EXT_RECORDED_ALLOCATED | - ((elen + inode->i_sb->s_blocksize - 1) & + nelen = EXT_NOT_RECORDED_NOT_ALLOCATED | + ((elen + offset + inode->i_sb->s_blocksize - 1) & ~(inode->i_sb->s_blocksize - 1)); - udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 1); + udf_write_aext(inode, bloc, &extoffset, neloc, nelen, bh, 1); + udf_add_aext(inode, &bloc, &extoffset, eloc, (etype << 30) | elen, &bh, 1); + } + else + { + if (elen & (inode->i_sb->s_blocksize - 1)) + { + extoffset -= adsize; + elen = EXT_RECORDED_ALLOCATED | + ((elen + inode->i_sb->s_blocksize - 1) & + ~(inode->i_sb->s_blocksize - 1)); + udf_write_aext(inode, bloc, &extoffset, eloc, elen, bh, 1); + } + memset(&eloc, 0x00, sizeof(kernel_lb_addr)); + elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; + udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } - memset(&eloc, 0x00, sizeof(kernel_lb_addr)); - elen = EXT_NOT_RECORDED_NOT_ALLOCATED | offset; - udf_add_aext(inode, &bloc, &extoffset, eloc, elen, &bh, 1); } } } diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c index ac511ab9c..0655cc3cb 100644 --- a/fs/xfs/xfs_dir2_node.c +++ b/fs/xfs/xfs_dir2_node.c @@ -970,7 +970,7 @@ xfs_dir2_leafn_remove( /* * One less used entry in the free table. */ - free->hdr.nused = cpu_to_be32(-1); + be32_add(&free->hdr.nused, -1); xfs_dir2_free_log_header(tp, fbp); /* * If this was the last entry in the table, we can diff --git a/include/asm-generic/mman.h b/include/asm-generic/mman.h index 3b41d2bb7..010ced7cf 100644 --- a/include/asm-generic/mman.h +++ b/include/asm-generic/mman.h @@ -39,4 +39,10 @@ #define MAP_ANON MAP_ANONYMOUS #define MAP_FILE 0 +#ifdef __KERNEL__ +#ifndef arch_mmap_check +#define arch_mmap_check(addr, len, flags) (0) +#endif +#endif + #endif diff --git a/include/asm-i386/apic.h b/include/asm-i386/apic.h index 288233fd7..87476de38 100644 --- a/include/asm-i386/apic.h +++ b/include/asm-i386/apic.h @@ -132,10 +132,12 @@ extern unsigned int nmi_watchdog; extern int disable_timer_pin_1; +#ifndef CONFIG_XEN void smp_send_timer_broadcast_ipi(struct pt_regs *regs); void switch_APIC_timer_to_ipi(void *cpumask); void switch_ipi_to_APIC_timer(void *cpumask); #define ARCH_APICTIMER_STOPS_ON_C3 1 +#endif extern int timer_over_8254; diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h index aa371537c..fdecfa0d1 100644 --- a/include/asm-i386/desc.h +++ b/include/asm-i386/desc.h @@ -170,7 +170,7 @@ static inline void set_user_cs(struct desc_struct *desc, unsigned long limit) } #define load_user_cs_desc(cpu, mm) \ - get_cpu_gdt_table(cpu)[GDT_ENTRY_DEFAULT_USER_CS] = (mm)->context.user_cs + get_cpu_gdt_table(cpu)[GDT_ENTRY_DEFAULT_USER_CS] = (mm)->context.user_cs extern void arch_add_exec_range(struct mm_struct *mm, unsigned long limit); extern void arch_remove_exec_range(struct mm_struct *mm, unsigned long limit); diff --git a/include/asm-i386/fixmap.h b/include/asm-i386/fixmap.h index cfb1c61d3..f11513aa9 100644 --- a/include/asm-i386/fixmap.h +++ b/include/asm-i386/fixmap.h @@ -20,7 +20,7 @@ * Leave one empty page between vmalloc'ed areas and * the start of the fixmap. */ -#define __FIXADDR_TOP 0xfffff000 +extern unsigned long __FIXADDR_TOP; #ifndef __ASSEMBLY__ #include @@ -95,6 +95,8 @@ enum fixed_addresses { extern void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags); +extern void set_fixaddr_top(unsigned long top); + #define set_fixmap(idx, phys) \ __set_fixmap(idx, phys, PAGE_KERNEL) /* diff --git a/include/asm-i386/mach-default/mach_traps.h b/include/asm-i386/mach-default/mach_traps.h index 625438b8a..4e34a91be 100644 --- a/include/asm-i386/mach-default/mach_traps.h +++ b/include/asm-i386/mach-default/mach_traps.h @@ -15,6 +15,18 @@ static inline void clear_mem_error(unsigned char reason) outb(reason, 0x61); } +static inline void clear_io_check_error(unsigned char reason) +{ + unsigned long i; + + reason = (reason & 0xf) | 8; + outb(reason, 0x61); + i = 2000; + while (--i) udelay(1000); + reason &= ~8; + outb(reason, 0x61); +} + static inline unsigned char get_nmi_reason(void) { return inb(0x61); diff --git a/include/asm-i386/page.h b/include/asm-i386/page.h index 104653b50..2b2683ea4 100644 --- a/include/asm-i386/page.h +++ b/include/asm-i386/page.h @@ -128,7 +128,7 @@ extern int devmem_is_allowed(unsigned long pagenr); #define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET) #define VMALLOC_RESERVE ((unsigned long)__VMALLOC_RESERVE) -#define MAXMEM (-__PAGE_OFFSET-__VMALLOC_RESERVE) +#define MAXMEM (__FIXADDR_TOP-__PAGE_OFFSET-__VMALLOC_RESERVE) #define __pa(x) ((unsigned long)(x)-PAGE_OFFSET) #define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET)) #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT) diff --git a/include/asm-i386/pgtable-2level-defs.h b/include/asm-i386/pgtable-2level-defs.h index 02518079f..bd6346f41 100644 --- a/include/asm-i386/pgtable-2level-defs.h +++ b/include/asm-i386/pgtable-2level-defs.h @@ -1,6 +1,8 @@ #ifndef _I386_PGTABLE_2LEVEL_DEFS_H #define _I386_PGTABLE_2LEVEL_DEFS_H +#define HAVE_SHARED_KERNEL_PMD 0 + /* * traditional i386 two-level paging structure: */ diff --git a/include/asm-i386/pgtable-3level-defs.h b/include/asm-i386/pgtable-3level-defs.h index eb3a1ea88..0c7599201 100644 --- a/include/asm-i386/pgtable-3level-defs.h +++ b/include/asm-i386/pgtable-3level-defs.h @@ -1,6 +1,8 @@ #ifndef _I386_PGTABLE_3LEVEL_DEFS_H #define _I386_PGTABLE_3LEVEL_DEFS_H +#define HAVE_SHARED_KERNEL_PMD 1 + /* * PGDIR_SHIFT determines what a top-level page table entry can map */ diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h index d76b7693c..6d794a373 100644 --- a/include/asm-i386/spinlock.h +++ b/include/asm-i386/spinlock.h @@ -36,7 +36,7 @@ "\n1:\t" \ "lock ; decb %0\n\t" \ "jns 5f\n" \ - "2:\t" \ + "2:\t" \ "testl $0x200, %1\n\t" \ "jz 4f\n\t" \ "sti\n" \ diff --git a/include/asm-ia64/mman.h b/include/asm-ia64/mman.h index 6ba179f12..b242f9522 100644 --- a/include/asm-ia64/mman.h +++ b/include/asm-ia64/mman.h @@ -8,6 +8,14 @@ * David Mosberger-Tang , Hewlett-Packard Co */ +#ifdef __KERNEL__ +#ifndef __ASSEMBLY__ +#define arch_mmap_check ia64_map_check_rgn +int ia64_map_check_rgn(unsigned long addr, unsigned long len, + unsigned long flags); +#endif +#endif + #include #define MAP_GROWSDOWN 0x00100 /* stack-like segment */ diff --git a/include/asm-ia64/sn/xp.h b/include/asm-ia64/sn/xp.h index 9902185c0..6f807e019 100644 --- a/include/asm-ia64/sn/xp.h +++ b/include/asm-ia64/sn/xp.h @@ -16,9 +16,9 @@ #define _ASM_IA64_SN_XP_H -#include #include #include +#include #include #include @@ -50,7 +50,7 @@ * C-brick nasids, thus the need for bitmaps which don't account for * odd-numbered (non C-brick) nasids. */ -#define XP_MAX_PHYSNODE_ID (MAX_PHYSNODE_ID / 2) +#define XP_MAX_PHYSNODE_ID (MAX_NUMALINK_NODES / 2) #define XP_NASID_MASK_BYTES ((XP_MAX_PHYSNODE_ID + 7) / 8) #define XP_NASID_MASK_WORDS ((XP_MAX_PHYSNODE_ID + 63) / 64) @@ -60,23 +60,37 @@ * the bte_copy() once in the hope that the failure was due to a temporary * aberration (i.e., the link going down temporarily). * - * See bte_copy for definition of the input parameters. + * src - physical address of the source of the transfer. + * vdst - virtual address of the destination of the transfer. + * len - number of bytes to transfer from source to destination. + * mode - see bte_copy() for definition. + * notification - see bte_copy() for definition. * * Note: xp_bte_copy() should never be called while holding a spinlock. */ static inline bte_result_t -xp_bte_copy(u64 src, u64 dest, u64 len, u64 mode, void *notification) +xp_bte_copy(u64 src, u64 vdst, u64 len, u64 mode, void *notification) { bte_result_t ret; + u64 pdst = ia64_tpa(vdst); - ret = bte_copy(src, dest, len, mode, notification); + /* + * Ensure that the physically mapped memory is contiguous. + * + * We do this by ensuring that the memory is from region 7 only. + * If the need should arise to use memory from one of the other + * regions, then modify the BUG_ON() statement to ensure that the + * memory from that region is always physically contiguous. + */ + BUG_ON(REGION_NUMBER(vdst) != RGN_KERNEL); + ret = bte_copy(src, pdst, len, mode, notification); if (ret != BTE_SUCCESS) { if (!in_interrupt()) { cond_resched(); } - ret = bte_copy(src, dest, len, mode, notification); + ret = bte_copy(src, pdst, len, mode, notification); } return ret; @@ -218,7 +232,19 @@ enum xpc_retval { xpcInvalidPartid, /* 42: invalid partition ID */ xpcLocalPartid, /* 43: local partition ID */ - xpcUnknownReason /* 44: unknown reason -- must be last in list */ + xpcOtherGoingDown, /* 44: other side going down, reason unknown */ + xpcSystemGoingDown, /* 45: system is going down, reason unknown */ + xpcSystemHalt, /* 46: system is being halted */ + xpcSystemReboot, /* 47: system is being rebooted */ + xpcSystemPoweroff, /* 48: system is being powered off */ + + xpcDisconnecting, /* 49: channel disconnecting (closing) */ + + xpcOpenCloseError, /* 50: channel open/close protocol error */ + + xpcDisconnected, /* 51: channel disconnected (closed) */ + + xpcUnknownReason /* 52: unknown reason -- must be last in list */ }; @@ -343,12 +369,12 @@ typedef void (*xpc_notify_func)(enum xpc_retval reason, partid_t partid, * * The 'func' field points to the function to call when aynchronous * notification is required for such events as: a connection established/lost, - * or an incomming message received, or an error condition encountered. A + * or an incoming message received, or an error condition encountered. A * non-NULL 'func' field indicates that there is an active registration for * the channel. */ struct xpc_registration { - struct semaphore sema; + struct mutex mutex; xpc_channel_func func; /* function to call */ void *key; /* pointer to user's key */ u16 nentries; /* #of msg entries in local msg queue */ diff --git a/include/asm-ia64/sn/xpc.h b/include/asm-ia64/sn/xpc.h index aa3b8ace9..b454ad4e2 100644 --- a/include/asm-ia64/sn/xpc.h +++ b/include/asm-ia64/sn/xpc.h @@ -684,7 +684,9 @@ extern struct xpc_vars *xpc_vars; extern struct xpc_rsvd_page *xpc_rsvd_page; extern struct xpc_vars_part *xpc_vars_part; extern struct xpc_partition xpc_partitions[XP_MAX_PARTITIONS + 1]; -extern char xpc_remote_copy_buffer[]; +extern char *xpc_remote_copy_buffer; +extern void *xpc_remote_copy_buffer_base; +extern void *xpc_kmalloc_cacheline_aligned(size_t, gfp_t, void **); extern struct xpc_rsvd_page *xpc_rsvd_page_init(void); extern void xpc_allow_IPI_ops(void); extern void xpc_restrict_IPI_ops(void); diff --git a/include/asm-s390/futex.h b/include/asm-s390/futex.h index 180277556..ffedf14f8 100644 --- a/include/asm-s390/futex.h +++ b/include/asm-s390/futex.h @@ -98,9 +98,10 @@ futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, int newval) if (! access_ok (VERIFY_WRITE, uaddr, sizeof(int))) return -EFAULT; - asm volatile(" cs %1,%4,0(%5)\n" + asm volatile(" sacf 256\n" + " cs %1,%4,0(%5)\n" "0: lr %0,%1\n" - "1:\n" + "1: sacf 0\n" #ifndef __s390x__ ".section __ex_table,\"a\"\n" " .align 4\n" diff --git a/include/asm-sparc/mman.h b/include/asm-sparc/mman.h index 88d1886ab..95ecab588 100644 --- a/include/asm-sparc/mman.h +++ b/include/asm-sparc/mman.h @@ -2,6 +2,12 @@ #ifndef __SPARC_MMAN_H__ #define __SPARC_MMAN_H__ +#ifdef __KERNEL__ +#define arch_mmap_check sparc_mmap_check +int sparc_mmap_check(unsigned long addr, unsigned long len, + unsigned long flags); +#endif + #include /* SunOS'ified... */ diff --git a/include/asm-sparc64/mman.h b/include/asm-sparc64/mman.h index 6fd878e61..b3002763c 100644 --- a/include/asm-sparc64/mman.h +++ b/include/asm-sparc64/mman.h @@ -2,6 +2,12 @@ #ifndef __SPARC64_MMAN_H__ #define __SPARC64_MMAN_H__ +#ifdef __KERNEL__ +#define arch_mmap_check sparc64_mmap_check +int sparc64_mmap_check(unsigned long addr, unsigned long len, + unsigned long flags); +#endif + #include /* SunOS'ified... */ diff --git a/include/asm-sparc64/sfp-machine.h b/include/asm-sparc64/sfp-machine.h index 5015bb8d6..89d42431e 100644 --- a/include/asm-sparc64/sfp-machine.h +++ b/include/asm-sparc64/sfp-machine.h @@ -34,7 +34,7 @@ #define _FP_MUL_MEAT_D(R,X,Y) \ _FP_MUL_MEAT_1_wide(_FP_WFRACBITS_D,R,X,Y,umul_ppmm) #define _FP_MUL_MEAT_Q(R,X,Y) \ - _FP_MUL_MEAT_2_wide_3mul(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm) + _FP_MUL_MEAT_2_wide(_FP_WFRACBITS_Q,R,X,Y,umul_ppmm) #define _FP_DIV_MEAT_S(R,X,Y) _FP_DIV_MEAT_1_imm(S,R,X,Y,_FP_DIV_HELP_imm) #define _FP_DIV_MEAT_D(R,X,Y) _FP_DIV_MEAT_1_udiv_norm(D,R,X,Y) diff --git a/include/asm-um/page.h b/include/asm-um/page.h index 0f7229ce7..459311435 100644 --- a/include/asm-um/page.h +++ b/include/asm-um/page.h @@ -116,7 +116,7 @@ extern struct page *arch_validate(struct page *page, gfp_t mask, int order); #define HAVE_ARCH_VALIDATE #define devmem_is_allowed(x) 1 -extern void arch_free_page(struct page *page, int order); +extern int arch_free_page(struct page *page, int order); #define HAVE_ARCH_FREE_PAGE #include diff --git a/include/asm-x86_64/apic.h b/include/asm-x86_64/apic.h index bdbd89356..79db75f17 100644 --- a/include/asm-x86_64/apic.h +++ b/include/asm-x86_64/apic.h @@ -105,11 +105,13 @@ extern int disable_timer_pin_1; extern void setup_threshold_lvt(unsigned long lvt_off); +#ifndef CONFIG_XEN void smp_send_timer_broadcast_ipi(void); void switch_APIC_timer_to_ipi(void *cpumask); void switch_ipi_to_APIC_timer(void *cpumask); #define ARCH_APICTIMER_STOPS_ON_C3 1 +#endif #endif /* CONFIG_X86_LOCAL_APIC */ diff --git a/include/asm-x86_64/hw_irq.h b/include/asm-x86_64/hw_irq.h index 0df1715de..91f751ba0 100644 --- a/include/asm-x86_64/hw_irq.h +++ b/include/asm-x86_64/hw_irq.h @@ -127,7 +127,7 @@ asmlinkage void IRQ_NAME(nr); \ __asm__( \ "\n.p2align\n" \ "IRQ" #nr "_interrupt:\n\t" \ - "push $" #nr "-256 ; " \ + "push $~(" #nr ") ; " \ "jmp common_interrupt"); #if defined(CONFIG_X86_IO_APIC) diff --git a/include/asm-x86_64/ipi.h b/include/asm-x86_64/ipi.h index 2a5c162b7..be4bc7638 100644 --- a/include/asm-x86_64/ipi.h +++ b/include/asm-x86_64/ipi.h @@ -49,6 +49,7 @@ static inline int __prepare_ICR2 (unsigned int mask) return SET_APIC_DEST_FIELD(mask); } +#ifndef CONFIG_XEN_UNPRIVILEGED_GUEST static inline void __send_IPI_shortcut(unsigned int shortcut, int vector, unsigned int dest) { /* @@ -113,5 +114,6 @@ static inline void send_IPI_mask_sequence(cpumask_t mask, int vector) } local_irq_restore(flags); } +#endif /* CONFIG_XEN_UNPRIVILEGED_GUEST */ #endif /* __ASM_IPI_H */ diff --git a/include/asm-x86_64/pgtable.h b/include/asm-x86_64/pgtable.h index 31e83c3bd..195952b2c 100644 --- a/include/asm-x86_64/pgtable.h +++ b/include/asm-x86_64/pgtable.h @@ -21,7 +21,7 @@ extern unsigned long __supported_pte_mask; #define swapper_pg_dir init_level4_pgt -extern int nonx_setup(char *str); +extern void nonx_setup(const char *str); extern void paging_init(void); extern void clear_kernel_mapping(unsigned long addr, unsigned long size); diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h index 93535f093..9269df738 100644 --- a/include/linux/ethtool.h +++ b/include/linux/ethtool.h @@ -408,6 +408,8 @@ struct ethtool_ops { #define ETHTOOL_GPERMADDR 0x00000020 /* Get permanent hardware address */ #define ETHTOOL_GUFO 0x00000021 /* Get UFO enable (ethtool_value) */ #define ETHTOOL_SUFO 0x00000022 /* Set UFO enable (ethtool_value) */ +#define ETHTOOL_GGSO 0x00000023 /* Get GSO enable (ethtool_value) */ +#define ETHTOOL_SGSO 0x00000024 /* Set GSO enable (ethtool_value) */ /* compatibility with older code */ #define SPARC_ETH_GSET ETHTOOL_GSET diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h index 75c1dc82c..33bc72c36 100644 --- a/include/linux/ext3_fs.h +++ b/include/linux/ext3_fs.h @@ -508,6 +508,15 @@ static inline struct ext3_inode_info *EXT3_I(struct inode *inode) { return container_of(inode, struct ext3_inode_info, vfs_inode); } + +static inline int ext3_valid_inum(struct super_block *sb, unsigned long ino) +{ + return ino == EXT3_ROOT_INO || + ino == EXT3_JOURNAL_INO || + ino == EXT3_RESIZE_INO || + (ino >= EXT3_FIRST_INO(sb) && + ino <= le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count)); +} #else /* Assume that user mode programs are passing in an ext3fs superblock, not * a kernel struct super_block. This will allow us to call the feature-test diff --git a/include/linux/gfp.h b/include/linux/gfp.h index a308228b7..83ece62bc 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -100,7 +100,11 @@ static inline int gfp_zone(gfp_t gfp) */ #ifndef HAVE_ARCH_FREE_PAGE -static inline void arch_free_page(struct page *page, int order) { } +/* + * If arch_free_page returns non-zero then the generic free_page code can + * immediately bail: the arch-specific function has done all the work. + */ +static inline int arch_free_page(struct page *page, int order) { return 0; } #endif extern struct page * diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 892c4ea1b..8f2bcfb85 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -25,10 +25,16 @@ static inline void flush_kernel_dcache_page(struct page *page) /* declarations for linux/mm/highmem.c */ unsigned int nr_free_highpages(void); +#ifdef CONFIG_XEN +void kmap_flush_unused(void); +#endif #else /* CONFIG_HIGHMEM */ static inline unsigned int nr_free_highpages(void) { return 0; } +#ifdef CONFIG_XEN +static inline void kmap_flush_unused(void) { } +#endif static inline void *kmap(struct page *page) { diff --git a/include/linux/idr.h b/include/linux/idr.h index d37c8d808..f559a719d 100644 --- a/include/linux/idr.h +++ b/include/linux/idr.h @@ -78,6 +78,7 @@ void *idr_find(struct idr *idp, int id); int idr_pre_get(struct idr *idp, gfp_t gfp_mask); int idr_get_new(struct idr *idp, void *ptr, int *id); int idr_get_new_above(struct idr *idp, void *ptr, int starting_id, int *id); +void *idr_replace(struct idr *idp, void *ptr, int id); void idr_remove(struct idr *idp, int id); void idr_destroy(struct idr *idp); void idr_init(struct idr *idp); diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h index 2c08fdc2b..d55ff51ff 100644 --- a/include/linux/interrupt.h +++ b/include/linux/interrupt.h @@ -58,6 +58,12 @@ extern void disable_irq(unsigned int irq); extern void enable_irq(unsigned int irq); #endif +#ifdef CONFIG_HAVE_IRQ_IGNORE_UNHANDLED +int irq_ignore_unhandled(unsigned int irq); +#else +#define irq_ignore_unhandled(irq) 0 +#endif + #ifndef __ARCH_SET_SOFTIRQ_PENDING #define set_softirq_pending(x) (local_softirq_pending() = (x)) #define or_softirq_pending(x) (local_softirq_pending() |= (x)) diff --git a/include/linux/mm.h b/include/linux/mm.h index e4183fa67..51eea5b38 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -166,6 +166,9 @@ extern unsigned int kobjsize(const void *objp); #define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */ #define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */ #define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */ +#ifdef CONFIG_XEN +#define VM_FOREIGN 0x04000000 /* Has pages belonging to another VM */ +#endif #ifndef VM_STACK_DEFAULT_FLAGS /* arch can override this */ #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS @@ -1026,6 +1029,13 @@ struct page *follow_page(struct vm_area_struct *, unsigned long address, #define FOLL_GET 0x04 /* do get_page on page */ #define FOLL_ANON 0x08 /* give ZERO_PAGE if no pgtable */ +#ifdef CONFIG_XEN +typedef int (*pte_fn_t)(pte_t *pte, struct page *pmd_page, unsigned long addr, + void *data); +extern int apply_to_page_range(struct mm_struct *mm, unsigned long address, + unsigned long size, pte_fn_t fn, void *data); +#endif + #ifdef CONFIG_PROC_FS void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long); #else diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h index f4169bbb6..c5b97cf59 100644 --- a/include/linux/netdevice.h +++ b/include/linux/netdevice.h @@ -232,6 +232,7 @@ enum netdev_state_t __LINK_STATE_RX_SCHED, __LINK_STATE_LINKWATCH_PENDING, __LINK_STATE_DORMANT, + __LINK_STATE_QDISC_RUNNING, }; @@ -307,9 +308,17 @@ struct net_device #define NETIF_F_HW_VLAN_RX 256 /* Receive VLAN hw acceleration */ #define NETIF_F_HW_VLAN_FILTER 512 /* Receive filtering on VLAN */ #define NETIF_F_VLAN_CHALLENGED 1024 /* Device cannot handle VLAN packets */ -#define NETIF_F_TSO 2048 /* Can offload TCP/IP segmentation */ +#define NETIF_F_GSO 2048 /* Enable software GSO. */ #define NETIF_F_LLTX 4096 /* LockLess TX */ -#define NETIF_F_UFO 8192 /* Can offload UDP Large Send*/ + + /* Segmentation offload features */ +#define NETIF_F_GSO_SHIFT 16 +#define NETIF_F_TSO (SKB_GSO_TCPV4 << NETIF_F_GSO_SHIFT) +#define NETIF_F_UFO (SKB_GSO_UDPV4 << NETIF_F_GSO_SHIFT) +#define NETIF_F_GSO_ROBUST (SKB_GSO_DODGY << NETIF_F_GSO_SHIFT) + +#define NETIF_F_GEN_CSUM (NETIF_F_NO_CSUM | NETIF_F_HW_CSUM) +#define NETIF_F_ALL_CSUM (NETIF_F_IP_CSUM | NETIF_F_GEN_CSUM) struct net_device *next_sched; @@ -398,6 +407,9 @@ struct net_device struct list_head qdisc_list; unsigned long tx_queue_len; /* Max frames per queue allowed */ + /* Partially transmitted GSO packet. */ + struct sk_buff *gso_skb; + /* ingress path synchronizer */ spinlock_t ingress_lock; struct Qdisc *qdisc_ingress; @@ -406,7 +418,7 @@ struct net_device * One part is mostly used on xmit path (device) */ /* hard_start_xmit synchronizer */ - spinlock_t xmit_lock ____cacheline_aligned_in_smp; + spinlock_t _xmit_lock ____cacheline_aligned_in_smp; /* cpu id of processor entered to hard_start_xmit or -1, if nobody entered there. */ @@ -532,6 +544,9 @@ struct packet_type { struct net_device *, struct packet_type *, struct net_device *); + struct sk_buff *(*gso_segment)(struct sk_buff *skb, + int features); + int (*gso_send_check)(struct sk_buff *skb); void *af_packet_priv; struct list_head list; }; @@ -679,7 +694,8 @@ extern int dev_change_name(struct net_device *, char *); extern int dev_set_mtu(struct net_device *, int); extern int dev_set_mac_address(struct net_device *, struct sockaddr *); -extern void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev); +extern int dev_hard_start_xmit(struct sk_buff *skb, + struct net_device *dev); extern void dev_init(void); @@ -889,11 +905,43 @@ static inline void __netif_rx_complete(struct net_device *dev) clear_bit(__LINK_STATE_RX_SCHED, &dev->state); } +static inline void netif_tx_lock(struct net_device *dev) +{ + spin_lock(&dev->_xmit_lock); + dev->xmit_lock_owner = smp_processor_id(); +} + +static inline void netif_tx_lock_bh(struct net_device *dev) +{ + spin_lock_bh(&dev->_xmit_lock); + dev->xmit_lock_owner = smp_processor_id(); +} + +static inline int netif_tx_trylock(struct net_device *dev) +{ + int ok = spin_trylock(&dev->_xmit_lock); + if (likely(ok)) + dev->xmit_lock_owner = smp_processor_id(); + return ok; +} + +static inline void netif_tx_unlock(struct net_device *dev) +{ + dev->xmit_lock_owner = -1; + spin_unlock(&dev->_xmit_lock); +} + +static inline void netif_tx_unlock_bh(struct net_device *dev) +{ + dev->xmit_lock_owner = -1; + spin_unlock_bh(&dev->_xmit_lock); +} + static inline void netif_tx_disable(struct net_device *dev) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); netif_stop_queue(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } /* These functions live elsewhere (drivers/net/net_init.c, but related) */ @@ -921,6 +969,7 @@ extern int netdev_max_backlog; extern int weight_p; extern int netdev_set_master(struct net_device *dev, struct net_device *master); extern int skb_checksum_help(struct sk_buff *skb, int inward); +extern struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features); #ifdef CONFIG_BUG extern void netdev_rx_csum_fault(struct net_device *dev); #else @@ -940,6 +989,19 @@ extern void dev_seq_stop(struct seq_file *seq, void *v); extern void linkwatch_run_queue(void); +static inline int skb_gso_ok(struct sk_buff *skb, int features) +{ + int feature = skb_shinfo(skb)->gso_type << NETIF_F_GSO_SHIFT; + return (features & feature) == feature; +} + +static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb) +{ + return skb_is_gso(skb) && + (!skb_gso_ok(skb, dev->features) || + unlikely(skb->ip_summed != CHECKSUM_HW)); +} + #endif /* __KERNEL__ */ #endif /* _LINUX_DEV_H */ diff --git a/include/linux/netfilter_bridge.h b/include/linux/netfilter_bridge.h index a75b84bb9..178a97e1a 100644 --- a/include/linux/netfilter_bridge.h +++ b/include/linux/netfilter_bridge.h @@ -47,18 +47,26 @@ enum nf_br_hook_priorities { #define BRNF_BRIDGED 0x08 #define BRNF_NF_BRIDGE_PREROUTING 0x10 - /* Only used in br_forward.c */ -static inline -void nf_bridge_maybe_copy_header(struct sk_buff *skb) +static inline int nf_bridge_maybe_copy_header(struct sk_buff *skb) { + int err; + if (skb->nf_bridge) { if (skb->protocol == __constant_htons(ETH_P_8021Q)) { + err = skb_cow(skb, 18); + if (err) + return err; memcpy(skb->data - 18, skb->nf_bridge->data, 18); skb_push(skb, 4); - } else + } else { + err = skb_cow(skb, 16); + if (err) + return err; memcpy(skb->data - 16, skb->nf_bridge->data, 16); + } } + return 0; } /* This is called by the IP fragmenting code and it ensures there is diff --git a/include/linux/oprofile.h b/include/linux/oprofile.h index 0d514b252..6c33ee68d 100644 --- a/include/linux/oprofile.h +++ b/include/linux/oprofile.h @@ -16,6 +16,10 @@ #include #include #include + +#ifdef CONFIG_XEN +#include +#endif struct super_block; struct dentry; @@ -27,6 +31,11 @@ struct oprofile_operations { /* create any necessary configuration files in the oprofile fs. * Optional. */ int (*create_files)(struct super_block * sb, struct dentry * root); + /* setup active domains with Xen */ + int (*set_active)(int *active_domains, unsigned int adomains); + /* setup passive domains with Xen */ + int (*set_passive)(int *passive_domains, unsigned int pdomains); + /* Do any necessary interrupt setup. Optional. */ int (*setup)(void); /* Do any necessary interrupt shutdown. Optional. */ @@ -78,6 +87,8 @@ void oprofile_add_pc(unsigned long pc, int is_kernel, unsigned long event); /* add a backtrace entry, to be called from the ->backtrace callback */ void oprofile_add_trace(unsigned long eip); +/* add a domain switch entry */ +int oprofile_add_domain_switch(int32_t domain_id); /** * Create a file of the given name as a child of the given root, with diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 4550ab035..2bd4cbafc 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -1242,6 +1242,7 @@ #define PCI_DEVICE_ID_VIA_PX8X0_0 0x0259 #define PCI_DEVICE_ID_VIA_3269_0 0x0269 #define PCI_DEVICE_ID_VIA_K8T800PRO_0 0x0282 +#define PCI_DEVICE_ID_VIA_3296_0 0x0296 #define PCI_DEVICE_ID_VIA_8363_0 0x0305 #define PCI_DEVICE_ID_VIA_P4M800CE 0x0314 #define PCI_DEVICE_ID_VIA_8371_0 0x0391 @@ -1249,6 +1250,7 @@ #define PCI_DEVICE_ID_VIA_82C561 0x0561 #define PCI_DEVICE_ID_VIA_82C586_1 0x0571 #define PCI_DEVICE_ID_VIA_82C576 0x0576 +#define PCI_DEVICE_ID_VIA_SATA_EIDE 0x0581 #define PCI_DEVICE_ID_VIA_82C586_0 0x0586 #define PCI_DEVICE_ID_VIA_82C596 0x0596 #define PCI_DEVICE_ID_VIA_82C597_0 0x0597 @@ -1289,10 +1291,11 @@ #define PCI_DEVICE_ID_VIA_8783_0 0x3208 #define PCI_DEVICE_ID_VIA_8237 0x3227 #define PCI_DEVICE_ID_VIA_8251 0x3287 -#define PCI_DEVICE_ID_VIA_3296_0 0x0296 +#define PCI_DEVICE_ID_VIA_8237A 0x3337 #define PCI_DEVICE_ID_VIA_8231 0x8231 #define PCI_DEVICE_ID_VIA_8231_4 0x8235 #define PCI_DEVICE_ID_VIA_8365_1 0x8305 +#define PCI_DEVICE_ID_VIA_CX700 0x8324 #define PCI_DEVICE_ID_VIA_8371_1 0x8391 #define PCI_DEVICE_ID_VIA_82C598_1 0x8598 #define PCI_DEVICE_ID_VIA_838X_1 0xB188 diff --git a/include/linux/pfn.h b/include/linux/pfn.h index bb01f8b92..ba2a86677 100644 --- a/include/linux/pfn.h +++ b/include/linux/pfn.h @@ -4,6 +4,6 @@ #define PFN_ALIGN(x) (((unsigned long)(x) + (PAGE_SIZE - 1)) & PAGE_MASK) #define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT) #define PFN_DOWN(x) ((x) >> PAGE_SHIFT) -#define PFN_PHYS(x) ((x) << PAGE_SHIFT) +#define PFN_PHYS(x) ((unsigned long long)(x) << PAGE_SHIFT) #endif diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h index c7e0f55bd..68ed67092 100644 --- a/include/linux/skbuff.h +++ b/include/linux/skbuff.h @@ -134,9 +134,10 @@ struct skb_frag_struct { struct skb_shared_info { atomic_t dataref; unsigned short nr_frags; - unsigned short tso_size; - unsigned short tso_segs; - unsigned short ufo_size; + unsigned short gso_size; + /* Warning: this field is not always filled in (UFO)! */ + unsigned short gso_segs; + unsigned short gso_type; unsigned int ip6_frag_id; struct sk_buff *frag_list; skb_frag_t frags[MAX_SKB_FRAGS]; @@ -168,6 +169,14 @@ enum { SKB_FCLONE_CLONE, }; +enum { + SKB_GSO_TCPV4 = 1 << 0, + SKB_GSO_UDPV4 = 1 << 1, + + /* This indicates the skb is from an untrusted source. */ + SKB_GSO_DODGY = 1 << 2, +}; + /** * struct sk_buff - socket buffer * @next: Next buffer in list @@ -189,6 +198,8 @@ enum { * @local_df: allow local fragmentation * @cloned: Head may be cloned (check refcnt to be sure) * @nohdr: Payload reference only, must not modify header + * @proto_data_valid: Protocol data validated since arriving at localhost + * @proto_csum_blank: Protocol csum must be added before leaving localhost * @pkt_type: Packet class * @fclone: skbuff clone status * @ip_summed: Driver fed us an IP checksum @@ -265,7 +276,13 @@ struct sk_buff { nfctinfo:3; __u8 pkt_type:3, fclone:2, +#ifndef CONFIG_XEN ipvs_property:1; +#else + ipvs_property:1, + proto_data_valid:1, + proto_csum_blank:1; +#endif __be16 protocol; void (*destructor)(struct sk_buff *skb); @@ -324,7 +341,8 @@ static inline struct sk_buff *alloc_skb_fclone(unsigned int size, extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, unsigned int size, - gfp_t priority); + gfp_t priority, + int fclone); extern void kfree_skbmem(struct sk_buff *skb); extern struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t priority); @@ -969,15 +987,16 @@ static inline void skb_reserve(struct sk_buff *skb, int len) #define NET_SKB_PAD 16 #endif -extern int ___pskb_trim(struct sk_buff *skb, unsigned int len, int realloc); +extern int ___pskb_trim(struct sk_buff *skb, unsigned int len); static inline void __skb_trim(struct sk_buff *skb, unsigned int len) { - if (!skb->data_len) { - skb->len = len; - skb->tail = skb->data + len; - } else - ___pskb_trim(skb, len, 0); + if (unlikely(skb->data_len)) { + WARN_ON(1); + return; + } + skb->len = len; + skb->tail = skb->data + len; } /** @@ -987,6 +1006,7 @@ static inline void __skb_trim(struct sk_buff *skb, unsigned int len) * * Cut the length of a buffer down by removing data from the tail. If * the buffer is already under the length specified it is not modified. + * The skb must be linear. */ static inline void skb_trim(struct sk_buff *skb, unsigned int len) { @@ -997,12 +1017,10 @@ static inline void skb_trim(struct sk_buff *skb, unsigned int len) static inline int __pskb_trim(struct sk_buff *skb, unsigned int len) { - if (!skb->data_len) { - skb->len = len; - skb->tail = skb->data+len; - return 0; - } - return ___pskb_trim(skb, len, 1); + if (skb->data_len) + return ___pskb_trim(skb, len); + __skb_trim(skb, len); + return 0; } static inline int pskb_trim(struct sk_buff *skb, unsigned int len) @@ -1010,6 +1028,21 @@ static inline int pskb_trim(struct sk_buff *skb, unsigned int len) return (len < skb->len) ? __pskb_trim(skb, len) : 0; } +/** + * pskb_trim_unique - remove end from a paged unique (not cloned) buffer + * @skb: buffer to alter + * @len: new length + * + * This is identical to pskb_trim except that the caller knows that + * the skb is not cloned so we should never get an error due to out- + * of-memory. + */ +static inline void pskb_trim_unique(struct sk_buff *skb, unsigned int len) +{ + int err = pskb_trim(skb, len); + BUG_ON(err); +} + /** * skb_orphan - orphan a buffer * @skb: buffer to orphan @@ -1064,7 +1097,7 @@ static inline struct sk_buff *__dev_alloc_skb(unsigned int length, return skb; } #else -extern struct sk_buff *__dev_alloc_skb(unsigned int length, int gfp_mask); +extern struct sk_buff *__dev_alloc_skb(unsigned int length, gfp_t gfp_mask); #endif /** @@ -1163,18 +1196,34 @@ static inline int skb_can_coalesce(struct sk_buff *skb, int i, return 0; } +static inline int __skb_linearize(struct sk_buff *skb) +{ + return __pskb_pull_tail(skb, skb->data_len) ? 0 : -ENOMEM; +} + /** * skb_linearize - convert paged skb to linear one * @skb: buffer to linarize - * @gfp: allocation mode * * If there is no free memory -ENOMEM is returned, otherwise zero * is returned and the old skb data released. */ -extern int __skb_linearize(struct sk_buff *skb, gfp_t gfp); -static inline int skb_linearize(struct sk_buff *skb, gfp_t gfp) +static inline int skb_linearize(struct sk_buff *skb) +{ + return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0; +} + +/** + * skb_linearize_cow - make sure skb is linear and writable + * @skb: buffer to process + * + * If there is no free memory -ENOMEM is returned, otherwise zero + * is returned and the old skb data released. + */ +static inline int skb_linearize_cow(struct sk_buff *skb) { - return __skb_linearize(skb, gfp); + return skb_is_nonlinear(skb) || skb_cloned(skb) ? + __skb_linearize(skb) : 0; } /** @@ -1271,6 +1320,7 @@ extern void skb_split(struct sk_buff *skb, struct sk_buff *skb1, const u32 len); extern void skb_release_data(struct sk_buff *skb); +extern struct sk_buff *skb_segment(struct sk_buff *skb, int features); static inline void *skb_header_pointer(const struct sk_buff *skb, int offset, int len, void *buffer) @@ -1400,5 +1450,10 @@ static inline void nf_reset(struct sk_buff *skb) static inline void nf_reset(struct sk_buff *skb) {} #endif /* CONFIG_NETFILTER */ +static inline int skb_is_gso(const struct sk_buff *skb) +{ + return skb_shinfo(skb)->gso_size; +} + #endif /* __KERNEL__ */ #endif /* _LINUX_SKBUFF_H */ diff --git a/include/linux/tty.h b/include/linux/tty.h index 87931555b..f13f49afe 100644 --- a/include/linux/tty.h +++ b/include/linux/tty.h @@ -297,8 +297,6 @@ extern int tty_read_raw_data(struct tty_struct *tty, unsigned char *bufp, int buflen); extern void tty_write_message(struct tty_struct *tty, char *msg); -extern void tty_get_termios(struct tty_driver *drv, int idx, struct termios *tio); - extern int is_orphaned_pgrp(int pgrp); extern int is_ignored(int sig); extern int tty_signal(int sig, struct tty_struct *tty); diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h index b94d1ad92..75b5b9333 100644 --- a/include/net/pkt_sched.h +++ b/include/net/pkt_sched.h @@ -218,12 +218,13 @@ extern struct qdisc_rate_table *qdisc_get_rtab(struct tc_ratespec *r, struct rtattr *tab); extern void qdisc_put_rtab(struct qdisc_rate_table *tab); -extern int qdisc_restart(struct net_device *dev); +extern void __qdisc_run(struct net_device *dev); static inline void qdisc_run(struct net_device *dev) { - while (!netif_queue_stopped(dev) && qdisc_restart(dev) < 0) - /* NOTHING */; + if (!netif_queue_stopped(dev) && + !test_and_set_bit(__LINK_STATE_QDISC_RUNNING, &dev->state)) + __qdisc_run(dev); } extern int tc_classify(struct sk_buff *skb, struct tcf_proto *tp, diff --git a/include/net/protocol.h b/include/net/protocol.h index 6dc597061..d516c58ce 100644 --- a/include/net/protocol.h +++ b/include/net/protocol.h @@ -37,6 +37,9 @@ struct net_protocol { int (*handler)(struct sk_buff *skb); void (*err_handler)(struct sk_buff *skb, u32 info); + int (*gso_send_check)(struct sk_buff *skb); + struct sk_buff *(*gso_segment)(struct sk_buff *skb, + int features); int no_policy; }; diff --git a/include/net/sctp/sctp.h b/include/net/sctp/sctp.h index aa6033ca7..225dcea94 100644 --- a/include/net/sctp/sctp.h +++ b/include/net/sctp/sctp.h @@ -405,19 +405,6 @@ static inline int sctp_list_single_entry(struct list_head *head) return ((head->next != head) && (head->next == head->prev)); } -/* Calculate the size (in bytes) occupied by the data of an iovec. */ -static inline size_t get_user_iov_size(struct iovec *iov, int iovlen) -{ - size_t retval = 0; - - for (; iovlen > 0; --iovlen) { - retval += iov->iov_len; - iov++; - } - - return retval; -} - /* Generate a random jitter in the range of -50% ~ +50% of input RTO. */ static inline __s32 sctp_jitter(__u32 rto) { diff --git a/include/net/sctp/sm.h b/include/net/sctp/sm.h index 1eac3d0eb..de313de4f 100644 --- a/include/net/sctp/sm.h +++ b/include/net/sctp/sm.h @@ -221,8 +221,7 @@ struct sctp_chunk *sctp_make_abort_no_data(const struct sctp_association *, const struct sctp_chunk *, __u32 tsn); struct sctp_chunk *sctp_make_abort_user(const struct sctp_association *, - const struct sctp_chunk *, - const struct msghdr *); + const struct msghdr *, size_t msg_len); struct sctp_chunk *sctp_make_abort_violation(const struct sctp_association *, const struct sctp_chunk *, const __u8 *, diff --git a/include/net/sock.h b/include/net/sock.h index 21950e757..f937108a4 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -1042,9 +1042,13 @@ static inline void sk_setup_caps(struct sock *sk, struct dst_entry *dst) { __sk_dst_set(sk, dst); sk->sk_route_caps = dst->dev->features; + if (sk->sk_route_caps & NETIF_F_GSO) + sk->sk_route_caps |= NETIF_F_TSO; if (sk->sk_route_caps & NETIF_F_TSO) { if (sock_flag(sk, SOCK_NO_LARGESEND) || dst->header_len) sk->sk_route_caps &= ~NETIF_F_TSO; + else + sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM; } } diff --git a/include/net/tcp.h b/include/net/tcp.h index c0f884027..8fa246247 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -568,13 +568,13 @@ struct tcp_skb_cb { */ static inline int tcp_skb_pcount(const struct sk_buff *skb) { - return skb_shinfo(skb)->tso_segs; + return skb_shinfo(skb)->gso_segs; } /* This is valid iff tcp_skb_pcount() > 1. */ static inline int tcp_skb_mss(const struct sk_buff *skb) { - return skb_shinfo(skb)->tso_size; + return skb_shinfo(skb)->gso_size; } static inline void tcp_dec_pcount_approx(__u32 *count, @@ -1079,6 +1079,9 @@ extern struct request_sock_ops tcp_request_sock_ops; extern int tcp_v4_destroy_sock(struct sock *sk); +extern int tcp_v4_gso_send_check(struct sk_buff *skb); +extern struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features); + #ifdef CONFIG_PROC_FS extern int tcp4_proc_init(void); extern void tcp4_proc_exit(void); diff --git a/include/scsi/iscsi_if.h b/include/scsi/iscsi_if.h index 0196291f6..e5618b909 100644 --- a/include/scsi/iscsi_if.h +++ b/include/scsi/iscsi_if.h @@ -43,10 +43,6 @@ enum iscsi_uevent_e { ISCSI_UEVENT_GET_STATS = UEVENT_BASE + 10, ISCSI_UEVENT_GET_PARAM = UEVENT_BASE + 11, - ISCSI_UEVENT_TRANSPORT_EP_CONNECT = UEVENT_BASE + 12, - ISCSI_UEVENT_TRANSPORT_EP_POLL = UEVENT_BASE + 13, - ISCSI_UEVENT_TRANSPORT_EP_DISCONNECT = UEVENT_BASE + 14, - /* up events */ ISCSI_KEVENT_RECV_PDU = KEVENT_BASE + 1, ISCSI_KEVENT_CONN_ERROR = KEVENT_BASE + 2, @@ -64,83 +60,61 @@ struct iscsi_uevent { uint32_t initial_cmdsn; } c_session; struct msg_destroy_session { + uint64_t session_handle; uint32_t sid; } d_session; struct msg_create_conn { - uint32_t sid; + uint64_t session_handle; uint32_t cid; + uint32_t sid; } c_conn; struct msg_bind_conn { - uint32_t sid; - uint32_t cid; - uint64_t transport_eph; + uint64_t session_handle; + uint64_t conn_handle; + uint32_t transport_fd; uint32_t is_leading; } b_conn; struct msg_destroy_conn { - uint32_t sid; + uint64_t conn_handle; uint32_t cid; } d_conn; struct msg_send_pdu { - uint32_t sid; - uint32_t cid; uint32_t hdr_size; uint32_t data_size; + uint64_t conn_handle; } send_pdu; struct msg_set_param { - uint32_t sid; - uint32_t cid; + uint64_t conn_handle; uint32_t param; /* enum iscsi_param */ - uint32_t len; + uint32_t value; } set_param; struct msg_start_conn { - uint32_t sid; - uint32_t cid; + uint64_t conn_handle; } start_conn; struct msg_stop_conn { - uint32_t sid; - uint32_t cid; uint64_t conn_handle; uint32_t flag; } stop_conn; struct msg_get_stats { - uint32_t sid; - uint32_t cid; + uint64_t conn_handle; } get_stats; - struct msg_transport_connect { - uint32_t non_blocking; - } ep_connect; - struct msg_transport_poll { - uint64_t ep_handle; - uint32_t timeout_ms; - } ep_poll; - struct msg_transport_disconnect { - uint64_t ep_handle; - } ep_disconnect; } u; union { /* messages k -> u */ + uint64_t handle; int retcode; struct msg_create_session_ret { + uint64_t session_handle; uint32_t sid; - uint32_t host_no; } c_session_ret; - struct msg_create_conn_ret { - uint32_t sid; - uint32_t cid; - } c_conn_ret; struct msg_recv_req { - uint32_t sid; - uint32_t cid; uint64_t recv_handle; + uint64_t conn_handle; } recv_req; struct msg_conn_error { - uint32_t sid; - uint32_t cid; + uint64_t conn_handle; uint32_t error; /* enum iscsi_err */ } connerror; - struct msg_transport_connect_ret { - uint64_t handle; - } ep_connect_ret; } r; } __attribute__ ((aligned (sizeof(uint64_t)))); @@ -165,66 +139,29 @@ enum iscsi_err { ISCSI_ERR_SESSION_FAILED = ISCSI_ERR_BASE + 13, ISCSI_ERR_HDR_DGST = ISCSI_ERR_BASE + 14, ISCSI_ERR_DATA_DGST = ISCSI_ERR_BASE + 15, - ISCSI_ERR_PARAM_NOT_FOUND = ISCSI_ERR_BASE + 16, - ISCSI_ERR_NO_SCSI_CMD = ISCSI_ERR_BASE + 17, + ISCSI_ERR_PARAM_NOT_FOUND = ISCSI_ERR_BASE + 16 }; /* * iSCSI Parameters (RFC3720) */ enum iscsi_param { - /* passed in using netlink set param */ - ISCSI_PARAM_MAX_RECV_DLENGTH, - ISCSI_PARAM_MAX_XMIT_DLENGTH, - ISCSI_PARAM_HDRDGST_EN, - ISCSI_PARAM_DATADGST_EN, - ISCSI_PARAM_INITIAL_R2T_EN, - ISCSI_PARAM_MAX_R2T, - ISCSI_PARAM_IMM_DATA_EN, - ISCSI_PARAM_FIRST_BURST, - ISCSI_PARAM_MAX_BURST, - ISCSI_PARAM_PDU_INORDER_EN, - ISCSI_PARAM_DATASEQ_INORDER_EN, - ISCSI_PARAM_ERL, - ISCSI_PARAM_IFMARKER_EN, - ISCSI_PARAM_OFMARKER_EN, - ISCSI_PARAM_EXP_STATSN, - ISCSI_PARAM_TARGET_NAME, - ISCSI_PARAM_TPGT, - ISCSI_PARAM_PERSISTENT_ADDRESS, - ISCSI_PARAM_PERSISTENT_PORT, - ISCSI_PARAM_SESS_RECOVERY_TMO, - - /* pased in through bind conn using transport_fd */ - ISCSI_PARAM_CONN_PORT, - ISCSI_PARAM_CONN_ADDRESS, - - /* must always be last */ - ISCSI_PARAM_MAX, + ISCSI_PARAM_MAX_RECV_DLENGTH = 0, + ISCSI_PARAM_MAX_XMIT_DLENGTH = 1, + ISCSI_PARAM_HDRDGST_EN = 2, + ISCSI_PARAM_DATADGST_EN = 3, + ISCSI_PARAM_INITIAL_R2T_EN = 4, + ISCSI_PARAM_MAX_R2T = 5, + ISCSI_PARAM_IMM_DATA_EN = 6, + ISCSI_PARAM_FIRST_BURST = 7, + ISCSI_PARAM_MAX_BURST = 8, + ISCSI_PARAM_PDU_INORDER_EN = 9, + ISCSI_PARAM_DATASEQ_INORDER_EN = 10, + ISCSI_PARAM_ERL = 11, + ISCSI_PARAM_IFMARKER_EN = 12, + ISCSI_PARAM_OFMARKER_EN = 13, }; - -#define ISCSI_MAX_RECV_DLENGTH (1 << ISCSI_PARAM_MAX_RECV_DLENGTH) -#define ISCSI_MAX_XMIT_DLENGTH (1 << ISCSI_PARAM_MAX_XMIT_DLENGTH) -#define ISCSI_HDRDGST_EN (1 << ISCSI_PARAM_HDRDGST_EN) -#define ISCSI_DATADGST_EN (1 << ISCSI_PARAM_DATADGST_EN) -#define ISCSI_INITIAL_R2T_EN (1 << ISCSI_PARAM_INITIAL_R2T_EN) -#define ISCSI_MAX_R2T (1 << ISCSI_PARAM_MAX_R2T) -#define ISCSI_IMM_DATA_EN (1 << ISCSI_PARAM_IMM_DATA_EN) -#define ISCSI_FIRST_BURST (1 << ISCSI_PARAM_FIRST_BURST) -#define ISCSI_MAX_BURST (1 << ISCSI_PARAM_MAX_BURST) -#define ISCSI_PDU_INORDER_EN (1 << ISCSI_PARAM_PDU_INORDER_EN) -#define ISCSI_DATASEQ_INORDER_EN (1 << ISCSI_PARAM_DATASEQ_INORDER_EN) -#define ISCSI_ERL (1 << ISCSI_PARAM_ERL) -#define ISCSI_IFMARKER_EN (1 << ISCSI_PARAM_IFMARKER_EN) -#define ISCSI_OFMARKER_EN (1 << ISCSI_PARAM_OFMARKER_EN) -#define ISCSI_EXP_STATSN (1 << ISCSI_PARAM_EXP_STATSN) -#define ISCSI_TARGET_NAME (1 << ISCSI_PARAM_TARGET_NAME) -#define ISCSI_TPGT (1 << ISCSI_PARAM_TPGT) -#define ISCSI_PERSISTENT_ADDRESS (1 << ISCSI_PARAM_PERSISTENT_ADDRESS) -#define ISCSI_PERSISTENT_PORT (1 << ISCSI_PARAM_PERSISTENT_PORT) -#define ISCSI_SESS_RECOVERY_TMO (1 << ISCSI_PARAM_SESS_RECOVERY_TMO) -#define ISCSI_CONN_PORT (1 << ISCSI_PARAM_CONN_PORT) -#define ISCSI_CONN_ADDRESS (1 << ISCSI_PARAM_CONN_ADDRESS) +#define ISCSI_PARAM_MAX 14 #define iscsi_ptr(_handle) ((void*)(unsigned long)_handle) #define iscsi_handle(_ptr) ((uint64_t)(unsigned long)_ptr) @@ -253,6 +190,7 @@ enum iscsi_param { * These flags describes reason of stop_conn() call */ #define STOP_CONN_TERM 0x1 +#define STOP_CONN_SUSPEND 0x2 #define STOP_CONN_RECOVER 0x3 #define ISCSI_STATS_CUSTOM_MAX 32 diff --git a/include/scsi/iscsi_proto.h b/include/scsi/iscsi_proto.h index 02f6e4b9e..4feda05fd 100644 --- a/include/scsi/iscsi_proto.h +++ b/include/scsi/iscsi_proto.h @@ -21,6 +21,8 @@ #ifndef ISCSI_PROTO_H #define ISCSI_PROTO_H +#define ISCSI_VERSION_STR "0.3" +#define ISCSI_DATE_STR "22-Apr-2005" #define ISCSI_DRAFT20_VERSION 0x00 /* default iSCSI listen port for incoming connections */ diff --git a/include/scsi/libiscsi.h b/include/scsi/libiscsi.h deleted file mode 100644 index cbf7e58bd..000000000 --- a/include/scsi/libiscsi.h +++ /dev/null @@ -1,282 +0,0 @@ -/* - * iSCSI lib definitions - * - * Copyright (C) 2006 Red Hat, Inc. All rights reserved. - * Copyright (C) 2004 - 2006 Mike Christie - * Copyright (C) 2004 - 2005 Dmitry Yusupov - * Copyright (C) 2004 - 2005 Alex Aizman - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA. - */ -#ifndef LIBISCSI_H -#define LIBISCSI_H - -#include -#include -#include -#include - -struct scsi_transport_template; -struct scsi_device; -struct Scsi_Host; -struct scsi_cmnd; -struct socket; -struct iscsi_transport; -struct iscsi_cls_session; -struct iscsi_cls_conn; -struct iscsi_session; -struct iscsi_nopin; - -/* #define DEBUG_SCSI */ -#ifdef DEBUG_SCSI -#define debug_scsi(fmt...) printk(KERN_INFO "iscsi: " fmt) -#else -#define debug_scsi(fmt...) -#endif - -#define ISCSI_XMIT_CMDS_MAX 128 /* must be power of 2 */ -#define ISCSI_MGMT_CMDS_MAX 32 /* must be power of 2 */ -#define ISCSI_CONN_MAX 1 - -#define ISCSI_MGMT_ITT_OFFSET 0xa00 - -#define ISCSI_DEF_CMD_PER_LUN 32 -#define ISCSI_MAX_CMD_PER_LUN 128 - -/* Task Mgmt states */ -#define TMABORT_INITIAL 0x0 -#define TMABORT_SUCCESS 0x1 -#define TMABORT_FAILED 0x2 -#define TMABORT_TIMEDOUT 0x3 - -/* Connection suspend "bit" */ -#define ISCSI_SUSPEND_BIT 1 - -#define ISCSI_ITT_MASK (0xfff) -#define ISCSI_CID_SHIFT 12 -#define ISCSI_CID_MASK (0xffff << ISCSI_CID_SHIFT) -#define ISCSI_AGE_SHIFT 28 -#define ISCSI_AGE_MASK (0xf << ISCSI_AGE_SHIFT) - -struct iscsi_mgmt_task { - /* - * Becuae LLDs allocate their hdr differently, this is a pointer to - * that storage. It must be setup at session creation time. - */ - struct iscsi_hdr *hdr; - char *data; /* mgmt payload */ - int data_count; /* counts data to be sent */ - uint32_t itt; /* this ITT */ - void *dd_data; /* driver/transport data */ - struct list_head running; -}; - -struct iscsi_cmd_task { - /* - * Becuae LLDs allocate their hdr differently, this is a pointer to - * that storage. It must be setup at session creation time. - */ - struct iscsi_cmd *hdr; - int itt; /* this ITT */ - int datasn; /* DataSN */ - - uint32_t unsol_datasn; - int imm_count; /* imm-data (bytes) */ - int unsol_count; /* unsolicited (bytes)*/ - int data_count; /* remaining Data-Out */ - struct scsi_cmnd *sc; /* associated SCSI cmd*/ - int total_length; - struct iscsi_conn *conn; /* used connection */ - struct iscsi_mgmt_task *mtask; /* tmf mtask in progr */ - - struct list_head running; /* running cmd list */ - void *dd_data; /* driver/transport data */ -}; - -struct iscsi_conn { - struct iscsi_cls_conn *cls_conn; /* ptr to class connection */ - void *dd_data; /* iscsi_transport data */ - struct iscsi_session *session; /* parent session */ - /* - * LLDs should set this lock. It protects the transport recv - * code - */ - rwlock_t *recv_lock; - /* - * conn_stop() flag: stop to recover, stop to terminate - */ - int stop_stage; - - /* iSCSI connection-wide sequencing */ - uint32_t exp_statsn; - - /* control data */ - int id; /* CID */ - struct list_head item; /* maintains list of conns */ - int c_stage; /* connection state */ - struct iscsi_mgmt_task *login_mtask; /* mtask used for login/text */ - struct iscsi_mgmt_task *mtask; /* xmit mtask in progress */ - struct iscsi_cmd_task *ctask; /* xmit ctask in progress */ - - /* xmit */ - struct kfifo *immqueue; /* immediate xmit queue */ - struct kfifo *mgmtqueue; /* mgmt (control) xmit queue */ - struct list_head mgmt_run_list; /* list of control tasks */ - struct kfifo *xmitqueue; /* data-path cmd queue */ - struct list_head run_list; /* list of cmds in progress */ - struct work_struct xmitwork; /* per-conn. xmit workqueue */ - /* - * serializes connection xmit, access to kfifos: - * xmitqueue, immqueue, mgmtqueue - */ - struct mutex xmitmutex; - - unsigned long suspend_tx; /* suspend Tx */ - unsigned long suspend_rx; /* suspend Rx */ - - /* abort */ - wait_queue_head_t ehwait; /* used in eh_abort() */ - struct iscsi_tm tmhdr; - struct timer_list tmabort_timer; - int tmabort_state; /* see TMABORT_INITIAL, etc.*/ - - /* negotiated params */ - int max_recv_dlength; /* initiator_max_recv_dsl*/ - int max_xmit_dlength; /* target_max_recv_dsl */ - int hdrdgst_en; - int datadgst_en; - - /* MIB-statistics */ - uint64_t txdata_octets; - uint64_t rxdata_octets; - uint32_t scsicmd_pdus_cnt; - uint32_t dataout_pdus_cnt; - uint32_t scsirsp_pdus_cnt; - uint32_t datain_pdus_cnt; - uint32_t r2t_pdus_cnt; - uint32_t tmfcmd_pdus_cnt; - int32_t tmfrsp_pdus_cnt; - - /* custom statistics */ - uint32_t eh_abort_cnt; -}; - -struct iscsi_queue { - struct kfifo *queue; /* FIFO Queue */ - void **pool; /* Pool of elements */ - int max; /* Max number of elements */ -}; - -struct iscsi_session { - /* iSCSI session-wide sequencing */ - uint32_t cmdsn; - uint32_t exp_cmdsn; - uint32_t max_cmdsn; - - /* configuration */ - int initial_r2t_en; - int max_r2t; - int imm_data_en; - int first_burst; - int max_burst; - int time2wait; - int time2retain; - int pdu_inorder_en; - int dataseq_inorder_en; - int erl; - int ifmarker_en; - int ofmarker_en; - - /* control data */ - struct iscsi_transport *tt; - struct Scsi_Host *host; - struct iscsi_conn *leadconn; /* leading connection */ - spinlock_t lock; /* protects session state, * - * sequence numbers, * - * session resources: * - * - cmdpool, * - * - mgmtpool, * - * - r2tpool */ - int state; /* session state */ - struct list_head item; - int age; /* counts session re-opens */ - - struct list_head connections; /* list of connections */ - int cmds_max; /* size of cmds array */ - struct iscsi_cmd_task **cmds; /* Original Cmds arr */ - struct iscsi_queue cmdpool; /* PDU's pool */ - int mgmtpool_max; /* size of mgmt array */ - struct iscsi_mgmt_task **mgmt_cmds; /* Original mgmt arr */ - struct iscsi_queue mgmtpool; /* Mgmt PDU's pool */ -}; - -/* - * scsi host template - */ -extern int iscsi_change_queue_depth(struct scsi_device *sdev, int depth); -extern int iscsi_eh_abort(struct scsi_cmnd *sc); -extern int iscsi_eh_host_reset(struct scsi_cmnd *sc); -extern int iscsi_queuecommand(struct scsi_cmnd *sc, - void (*done)(struct scsi_cmnd *)); - -/* - * session management - */ -extern struct iscsi_cls_session * -iscsi_session_setup(struct iscsi_transport *, struct scsi_transport_template *, - int, int, uint32_t, uint32_t *); -extern void iscsi_session_teardown(struct iscsi_cls_session *); -extern struct iscsi_session *class_to_transport_session(struct iscsi_cls_session *); -extern void iscsi_session_recovery_timedout(struct iscsi_cls_session *); - -#define session_to_cls(_sess) \ - hostdata_session(_sess->host->hostdata) - -/* - * connection management - */ -extern struct iscsi_cls_conn *iscsi_conn_setup(struct iscsi_cls_session *, - uint32_t); -extern void iscsi_conn_teardown(struct iscsi_cls_conn *); -extern int iscsi_conn_start(struct iscsi_cls_conn *); -extern void iscsi_conn_stop(struct iscsi_cls_conn *, int); -extern int iscsi_conn_bind(struct iscsi_cls_session *, struct iscsi_cls_conn *, - int); -extern void iscsi_conn_failure(struct iscsi_conn *conn, enum iscsi_err err); - -/* - * pdu and task processing - */ -extern int iscsi_check_assign_cmdsn(struct iscsi_session *, - struct iscsi_nopin *); -extern void iscsi_prep_unsolicit_data_pdu(struct iscsi_cmd_task *, - struct iscsi_data *hdr, - int transport_data_cnt); -extern int iscsi_conn_send_pdu(struct iscsi_cls_conn *, struct iscsi_hdr *, - char *, uint32_t); -extern int iscsi_complete_pdu(struct iscsi_conn *, struct iscsi_hdr *, - char *, int); -extern int __iscsi_complete_pdu(struct iscsi_conn *, struct iscsi_hdr *, - char *, int); -extern int iscsi_verify_itt(struct iscsi_conn *, struct iscsi_hdr *, - uint32_t *); - -/* - * generic helpers - */ -extern void iscsi_pool_free(struct iscsi_queue *, void **); -extern int iscsi_pool_init(struct iscsi_queue *, int, void ***, int); - -#endif diff --git a/include/scsi/scsi_transport_iscsi.h b/include/scsi/scsi_transport_iscsi.h index 1b26a6c0a..b41cf077e 100644 --- a/include/scsi/scsi_transport_iscsi.h +++ b/include/scsi/scsi_transport_iscsi.h @@ -1,8 +1,10 @@ -/* +/* * iSCSI transport class definitions * * Copyright (C) IBM Corporation, 2004 - * Copyright (C) Mike Christie, 2004 + * Copyright (C) Mike Christie, 2004 - 2005 + * Copyright (C) Dmitry Yusupov, 2004 - 2005 + * Copyright (C) Alex Aizman, 2004 - 2005 * * This program is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -21,158 +23,125 @@ #ifndef SCSI_TRANSPORT_ISCSI_H #define SCSI_TRANSPORT_ISCSI_H -#include -#include -#include +#include +#include struct scsi_transport_template; +struct Scsi_Host; +struct mempool_zone; +struct iscsi_cls_conn; -struct iscsi_class_session { - uint8_t isid[6]; - uint16_t tsih; - int header_digest; /* 1 CRC32, 0 None */ - int data_digest; /* 1 CRC32, 0 None */ - uint16_t tpgt; - union { - struct in6_addr sin6_addr; - struct in_addr sin_addr; - } u; - sa_family_t addr_type; /* must be AF_INET or AF_INET6 */ - uint16_t port; /* must be in network byte order */ - int initial_r2t; /* 1 Yes, 0 No */ - int immediate_data; /* 1 Yes, 0 No */ - uint32_t max_recv_data_segment_len; - uint32_t max_burst_len; - uint32_t first_burst_len; - uint16_t def_time2wait; - uint16_t def_time2retain; - uint16_t max_outstanding_r2t; - int data_pdu_in_order; /* 1 Yes, 0 No */ - int data_sequence_in_order; /* 1 Yes, 0 No */ - int erl; +/** + * struct iscsi_transport - iSCSI Transport template + * + * @name: transport name + * @caps: iSCSI Data-Path capabilities + * @create_session: create new iSCSI session object + * @destroy_session: destroy existing iSCSI session object + * @create_conn: create new iSCSI connection + * @bind_conn: associate this connection with existing iSCSI session + * and specified transport descriptor + * @destroy_conn: destroy inactive iSCSI connection + * @set_param: set iSCSI Data-Path operational parameter + * @start_conn: set connection to be operational + * @stop_conn: suspend/recover/terminate connection + * @send_pdu: send iSCSI PDU, Login, Logout, NOP-Out, Reject, Text. + * + * Template API provided by iSCSI Transport + */ +struct iscsi_transport { + struct module *owner; + char *name; + unsigned int caps; + struct scsi_host_template *host_template; + /* LLD session/scsi_host data size */ + int hostdata_size; + /* LLD iscsi_host data size */ + int ihostdata_size; + /* LLD connection data size */ + int conndata_size; + int max_lun; + unsigned int max_conn; + unsigned int max_cmd_len; + struct iscsi_cls_session *(*create_session) + (struct scsi_transport_template *t, uint32_t sn, uint32_t *sid); + void (*destroy_session) (struct iscsi_cls_session *session); + struct iscsi_cls_conn *(*create_conn) (struct iscsi_cls_session *sess, + uint32_t cid); + int (*bind_conn) (struct iscsi_cls_session *session, + struct iscsi_cls_conn *cls_conn, + uint32_t transport_fd, int is_leading); + int (*start_conn) (struct iscsi_cls_conn *conn); + void (*stop_conn) (struct iscsi_cls_conn *conn, int flag); + void (*destroy_conn) (struct iscsi_cls_conn *conn); + int (*set_param) (struct iscsi_cls_conn *conn, enum iscsi_param param, + uint32_t value); + int (*get_conn_param) (struct iscsi_cls_conn *conn, + enum iscsi_param param, + uint32_t *value); + int (*get_session_param) (struct iscsi_cls_session *session, + enum iscsi_param param, uint32_t *value); + int (*send_pdu) (struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size); + void (*get_stats) (struct iscsi_cls_conn *conn, + struct iscsi_stats *stats); }; /* - * accessor macros + * transport registration upcalls */ -#define iscsi_isid(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->isid) -#define iscsi_tsih(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->tsih) -#define iscsi_header_digest(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->header_digest) -#define iscsi_data_digest(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->data_digest) -#define iscsi_port(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->port) -#define iscsi_addr_type(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->addr_type) -#define iscsi_sin_addr(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->u.sin_addr) -#define iscsi_sin6_addr(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->u.sin6_addr) -#define iscsi_tpgt(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->tpgt) -#define iscsi_initial_r2t(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->initial_r2t) -#define iscsi_immediate_data(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->immediate_data) -#define iscsi_max_recv_data_segment_len(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->max_recv_data_segment_len) -#define iscsi_max_burst_len(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->max_burst_len) -#define iscsi_first_burst_len(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->first_burst_len) -#define iscsi_def_time2wait(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->def_time2wait) -#define iscsi_def_time2retain(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->def_time2retain) -#define iscsi_max_outstanding_r2t(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->max_outstanding_r2t) -#define iscsi_data_pdu_in_order(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->data_pdu_in_order) -#define iscsi_data_sequence_in_order(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->data_sequence_in_order) -#define iscsi_erl(x) \ - (((struct iscsi_class_session *)&(x)->starget_data)->erl) +extern struct scsi_transport_template *iscsi_register_transport(struct iscsi_transport *tt); +extern int iscsi_unregister_transport(struct iscsi_transport *tt); /* - * The functions by which the transport class and the driver communicate + * control plane upcalls */ -struct iscsi_function_template { - /* - * target attrs - */ - void (*get_isid)(struct scsi_target *); - void (*get_tsih)(struct scsi_target *); - void (*get_header_digest)(struct scsi_target *); - void (*get_data_digest)(struct scsi_target *); - void (*get_port)(struct scsi_target *); - void (*get_tpgt)(struct scsi_target *); - /* - * In get_ip_address the lld must set the address and - * the address type - */ - void (*get_ip_address)(struct scsi_target *); - /* - * The lld should snprintf the name or alias to the buffer - */ - ssize_t (*get_target_name)(struct scsi_target *, char *, ssize_t); - ssize_t (*get_target_alias)(struct scsi_target *, char *, ssize_t); - void (*get_initial_r2t)(struct scsi_target *); - void (*get_immediate_data)(struct scsi_target *); - void (*get_max_recv_data_segment_len)(struct scsi_target *); - void (*get_max_burst_len)(struct scsi_target *); - void (*get_first_burst_len)(struct scsi_target *); - void (*get_def_time2wait)(struct scsi_target *); - void (*get_def_time2retain)(struct scsi_target *); - void (*get_max_outstanding_r2t)(struct scsi_target *); - void (*get_data_pdu_in_order)(struct scsi_target *); - void (*get_data_sequence_in_order)(struct scsi_target *); - void (*get_erl)(struct scsi_target *); +extern void iscsi_conn_error(struct iscsi_cls_conn *conn, enum iscsi_err error); +extern int iscsi_recv_pdu(struct iscsi_cls_conn *conn, struct iscsi_hdr *hdr, + char *data, uint32_t data_size); - /* - * host atts - */ +struct iscsi_cls_conn { + struct list_head conn_list; /* item in connlist */ + void *dd_data; /* LLD private data */ + struct iscsi_transport *transport; + int active; /* must be accessed with the connlock */ + struct device dev; /* sysfs transport/container device */ + struct mempool_zone *z_error; + struct mempool_zone *z_pdu; + struct list_head freequeue; +}; + +#define iscsi_dev_to_conn(_dev) \ + container_of(_dev, struct iscsi_cls_conn, dev) - /* - * The lld should snprintf the name or alias to the buffer - */ - ssize_t (*get_initiator_alias)(struct Scsi_Host *, char *, ssize_t); - ssize_t (*get_initiator_name)(struct Scsi_Host *, char *, ssize_t); - /* - * The driver sets these to tell the transport class it - * wants the attributes displayed in sysfs. If the show_ flag - * is not set, the attribute will be private to the transport - * class. We could probably just test if a get_ fn was set - * since we only use the values for sysfs but this is how - * fc does it too. - */ - unsigned long show_isid:1; - unsigned long show_tsih:1; - unsigned long show_header_digest:1; - unsigned long show_data_digest:1; - unsigned long show_port:1; - unsigned long show_tpgt:1; - unsigned long show_ip_address:1; - unsigned long show_target_name:1; - unsigned long show_target_alias:1; - unsigned long show_initial_r2t:1; - unsigned long show_immediate_data:1; - unsigned long show_max_recv_data_segment_len:1; - unsigned long show_max_burst_len:1; - unsigned long show_first_burst_len:1; - unsigned long show_def_time2wait:1; - unsigned long show_def_time2retain:1; - unsigned long show_max_outstanding_r2t:1; - unsigned long show_data_pdu_in_order:1; - unsigned long show_data_sequence_in_order:1; - unsigned long show_erl:1; - unsigned long show_initiator_name:1; - unsigned long show_initiator_alias:1; +struct iscsi_cls_session { + struct list_head sess_list; /* item in session_list */ + struct iscsi_transport *transport; + struct device dev; /* sysfs transport/container device */ }; -struct scsi_transport_template *iscsi_attach_transport(struct iscsi_function_template *); -void iscsi_release_transport(struct scsi_transport_template *); +#define iscsi_dev_to_session(_dev) \ + container_of(_dev, struct iscsi_cls_session, dev) + +#define iscsi_session_to_shost(_session) \ + dev_to_shost(_session->dev.parent) + +/* + * session and connection functions that can be used by HW iSCSI LLDs + */ +extern struct iscsi_cls_session *iscsi_create_session(struct Scsi_Host *shost, + struct iscsi_transport *t); +extern int iscsi_destroy_session(struct iscsi_cls_session *session); +extern struct iscsi_cls_conn *iscsi_create_conn(struct iscsi_cls_session *sess, + uint32_t cid); +extern int iscsi_destroy_conn(struct iscsi_cls_conn *conn); + +/* + * session functions used by software iscsi + */ +extern struct Scsi_Host * +iscsi_transport_create_session(struct scsi_transport_template *scsit, + struct iscsi_transport *transport); +extern int iscsi_transport_destroy_session(struct Scsi_Host *shost); #endif diff --git a/include/sound/initval.h b/include/sound/initval.h index d29e3d31d..d45170b9e 100644 --- a/include/sound/initval.h +++ b/include/sound/initval.h @@ -62,7 +62,8 @@ static int snd_legacy_find_free_irq(int *irq_table) { while (*irq_table != -1) { if (!request_irq(*irq_table, snd_legacy_empty_irq_handler, - SA_INTERRUPT, "ALSA Test IRQ", (void *) irq_table)) { + SA_INTERRUPT | SA_PROBEIRQ, "ALSA Test IRQ", + (void *) irq_table)) { free_irq(*irq_table, (void *) irq_table); return *irq_table; } diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt index 0b46a5dff..17ab32232 100644 --- a/kernel/Kconfig.preempt +++ b/kernel/Kconfig.preempt @@ -35,6 +35,7 @@ config PREEMPT_VOLUNTARY config PREEMPT bool "Preemptible Kernel (Low-Latency Desktop)" + depends on !XEN help This option reduces the latency of the kernel by making all kernel code (that is not executing in a critical section) diff --git a/kernel/fork.c b/kernel/fork.c index fcaaa9978..471d19f71 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -280,6 +280,9 @@ static inline int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) if (retval) goto out; } +#ifdef arch_dup_mmap + arch_dup_mmap(mm, oldmm); +#endif retval = 0; out: up_write(&mm->mmap_sem); diff --git a/kernel/futex.c b/kernel/futex.c index 9ddf5010a..c27bb560b 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -594,6 +594,7 @@ static int unqueue_me(struct futex_q *q) /* In the common case we don't take the spinlock, which is nice. */ retry: lock_ptr = q->lock_ptr; + barrier(); if (lock_ptr != 0) { spin_lock(lock_ptr); /* diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 1279e3499..97d555999 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -204,14 +204,10 @@ int setup_irq(unsigned int irq, struct irqaction * new) p = &desc->action; if ((old = *p) != NULL) { /* Can't share interrupts unless both agree to */ - if (!(old->flags & new->flags & SA_SHIRQ)) - goto mismatch; - -#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ) - /* All handlers must agree on per-cpuness */ - if ((old->flags & IRQ_PER_CPU) != (new->flags & IRQ_PER_CPU)) - goto mismatch; -#endif + if (!(old->flags & new->flags & SA_SHIRQ)) { + spin_unlock_irqrestore(&desc->lock,flags); + return -EBUSY; + } /* add new interrupt at end of irq queue */ do { @@ -222,10 +218,7 @@ int setup_irq(unsigned int irq, struct irqaction * new) } *p = new; -#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ) - if (new->flags & SA_PERCPU_IRQ) - desc->status |= IRQ_PER_CPU; -#endif + if (!shared) { desc->depth = 0; desc->status &= ~(IRQ_DISABLED | IRQ_AUTODETECT | @@ -243,14 +236,6 @@ int setup_irq(unsigned int irq, struct irqaction * new) register_handler_proc(irq, new); return 0; - -mismatch: - spin_unlock_irqrestore(&desc->lock, flags); - if (!(new->flags & SA_PROBEIRQ)) { - printk(KERN_ERR "%s: irq handler mismatch\n", __FUNCTION__); - dump_stack(); - } - return -EBUSY; } /** @@ -273,7 +258,6 @@ void free_irq(unsigned int irq, void *dev_id) struct irqaction **p; unsigned long flags; - WARN_ON(in_interrupt()); if (irq >= NR_IRQS) return; diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c index 7df9abd5e..f9353e960 100644 --- a/kernel/irq/spurious.c +++ b/kernel/irq/spurious.c @@ -137,7 +137,8 @@ void note_interrupt(unsigned int irq, irq_desc_t *desc, irqreturn_t action_ret, struct pt_regs *regs) { if (action_ret != IRQ_HANDLED) { - desc->irqs_unhandled++; + if (!irq_ignore_unhandled(irq)) + desc->irqs_unhandled++; if (action_ret != IRQ_NONE) report_bad_irq(irq, desc, action_ret); } diff --git a/kernel/sched.c b/kernel/sched.c index 4d2010694..eeaa85844 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -4238,17 +4238,22 @@ asmlinkage long sys_sched_yield(void) return 0; } -static inline void __cond_resched(void) +static inline int __resched_legal(int expected_preempt_count) +{ + if (unlikely(preempt_count() != expected_preempt_count)) + return 0; + if (unlikely(system_state != SYSTEM_RUNNING)) + return 0; + return 1; +} + +static void __cond_resched(void) { /* * The BKS might be reacquired before we have dropped * PREEMPT_ACTIVE, which could trigger a second * cond_resched() call. */ - if (unlikely(preempt_count())) - return; - if (unlikely(system_state != SYSTEM_RUNNING)) - return; do { add_preempt_count(PREEMPT_ACTIVE); schedule(); @@ -4258,13 +4263,12 @@ static inline void __cond_resched(void) int __sched cond_resched(void) { - if (need_resched()) { + if (need_resched() && __resched_legal(0)) { __cond_resched(); return 1; } return 0; } - EXPORT_SYMBOL(cond_resched); /* @@ -4285,7 +4289,7 @@ int cond_resched_lock(spinlock_t *lock) ret = 1; spin_lock(lock); } - if (need_resched()) { + if (need_resched() && __resched_legal(1)) { _raw_spin_unlock(lock); preempt_enable_no_resched(); __cond_resched(); @@ -4294,14 +4298,13 @@ int cond_resched_lock(spinlock_t *lock) } return ret; } - EXPORT_SYMBOL(cond_resched_lock); int __sched cond_resched_softirq(void) { BUG_ON(!in_softirq()); - if (need_resched()) { + if (need_resched() && __resched_legal(0)) { __local_bh_enable(); __cond_resched(); local_bh_disable(); @@ -4309,10 +4312,8 @@ int __sched cond_resched_softirq(void) } return 0; } - EXPORT_SYMBOL(cond_resched_softirq); - /** * yield - yield the current processor to other threads. * diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index dcfb5d731..51cacd111 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -111,7 +111,6 @@ static int stop_machine(void) /* If some failed, kill them all. */ if (ret < 0) { stopmachine_set_state(STOPMACHINE_EXIT); - up(&stopmachine_mutex); return ret; } diff --git a/kernel/timer.c b/kernel/timer.c index 2f8cdbe00..20a4bca5e 100644 --- a/kernel/timer.c +++ b/kernel/timer.c @@ -543,6 +543,22 @@ found: } spin_unlock(&base->lock); + /* + * It can happen that other CPUs service timer IRQs and increment + * jiffies, but we have not yet got a local timer tick to process + * the timer wheels. In that case, the expiry time can be before + * jiffies, but since the high-resolution timer here is relative to + * jiffies, the default expression when high-resolution timers are + * not active, + * + * time_before(MAX_JIFFY_OFFSET + jiffies, expires) + * + * would falsely evaluate to true. If that is the case, just + * return jiffies so that we can immediately fire the local timer + */ + if (time_before(expires, jiffies)) + return jiffies; + /* * It can happen that other CPUs service timer IRQs and increment * jiffies, but we have not yet got a local timer tick to process @@ -971,46 +987,18 @@ asmlinkage long sys_getpid(void) } /* - * Accessing ->group_leader->real_parent is not SMP-safe, it could - * change from under us. However, rather than getting any lock - * we can use an optimistic algorithm: get the parent - * pid, and go back and check that the parent is still - * the same. If it has changed (which is extremely unlikely - * indeed), we just try again.. - * - * NOTE! This depends on the fact that even if we _do_ - * get an old value of "parent", we can happily dereference - * the pointer (it was and remains a dereferencable kernel pointer - * no matter what): we just can't necessarily trust the result - * until we know that the parent pointer is valid. - * - * NOTE2: ->group_leader never changes from under us. + * Accessing ->real_parent is not SMP-safe, it could + * change from under us. However, we can use a stale + * value of ->real_parent under rcu_read_lock(), see + * release_task()->call_rcu(delayed_put_task_struct). */ asmlinkage long sys_getppid(void) { int pid; - struct task_struct *me = current; - struct task_struct *parent; - parent = me->group_leader->real_parent; - for (;;) { - pid = parent->tgid; -#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT) -{ - struct task_struct *old = parent; - - /* - * Make sure we read the pid before re-reading the - * parent pointer: - */ - smp_rmb(); - parent = me->group_leader->real_parent; - if (old != parent) - continue; -} -#endif - break; - } + rcu_read_lock(); + pid = rcu_dereference(current->real_parent)->tgid; + rcu_read_unlock(); return vx_map_pid(pid); } @@ -1025,7 +1013,7 @@ asmlinkage long do_getxpid(long *ppid) *ppid = sys_getppid(); return sys_getpid(); } - + #else /* _alpha_ */ asmlinkage long sys_getuid(void) diff --git a/lib/Makefile b/lib/Makefile index b830c9a15..033ff7c67 100644 --- a/lib/Makefile +++ b/lib/Makefile @@ -48,6 +48,9 @@ obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o obj-$(CONFIG_SWIOTLB) += swiotlb.o +ifneq ($(CONFIG_XEN_IA64_DOM0_NON_VP),y) +swiotlb-$(CONFIG_XEN) := ../arch/i386/kernel/swiotlb.o +endif hostprogs-y := gen_crc32table clean-files := crc32table.h diff --git a/lib/idr.c b/lib/idr.c index de19030a9..4d0968195 100644 --- a/lib/idr.c +++ b/lib/idr.c @@ -29,6 +29,7 @@ #include #include #endif +#include #include #include @@ -398,6 +399,48 @@ void *idr_find(struct idr *idp, int id) } EXPORT_SYMBOL(idr_find); +/** + * idr_replace - replace pointer for given id + * @idp: idr handle + * @ptr: pointer you want associated with the id + * @id: lookup key + * + * Replace the pointer registered with an id and return the old value. + * A -ENOENT return indicates that @id was not found. + * A -EINVAL return indicates that @id was not within valid constraints. + * + * The caller must serialize vs idr_find(), idr_get_new(), and idr_remove(). + */ +void *idr_replace(struct idr *idp, void *ptr, int id) +{ + int n; + struct idr_layer *p, *old_p; + + n = idp->layers * IDR_BITS; + p = idp->top; + + id &= MAX_ID_MASK; + + if (id >= (1 << n)) + return ERR_PTR(-EINVAL); + + n -= IDR_BITS; + while ((n > 0) && p) { + p = p->ary[(id >> n) & IDR_MASK]; + n -= IDR_BITS; + } + + n = id & IDR_MASK; + if (unlikely(p == NULL || !test_bit(n, &p->bitmap))) + return ERR_PTR(-ENOENT); + + old_p = p->ary[n]; + p->ary[n] = ptr; + + return old_p; +} +EXPORT_SYMBOL(idr_replace); + static void idr_cache_ctor(void * idr_layer, kmem_cache_t *idr_layer_cache, unsigned long flags) { diff --git a/lib/spinlock_debug.c b/lib/spinlock_debug.c index 34022dd9a..8ebd8ea08 100644 --- a/lib/spinlock_debug.c +++ b/lib/spinlock_debug.c @@ -150,6 +150,7 @@ static void rwlock_bug(rwlock_t *lock, const char *msg) #define RWLOCK_BUG_ON(cond, lock, msg) if (unlikely(cond)) rwlock_bug(lock, msg) +#if 0 /* __write_lock_debug() can lock up - maybe this can too? */ static void __read_lock_debug(rwlock_t *lock) { int print_once = 1; @@ -171,12 +172,12 @@ static void __read_lock_debug(rwlock_t *lock) } } } +#endif void _raw_read_lock(rwlock_t *lock) { RWLOCK_BUG_ON(lock->magic != RWLOCK_MAGIC, lock, "bad magic"); - if (unlikely(!__raw_read_trylock(&lock->raw_lock))) - __read_lock_debug(lock); + __raw_read_lock(&lock->raw_lock); } int _raw_read_trylock(rwlock_t *lock) @@ -222,6 +223,7 @@ static inline void debug_write_unlock(rwlock_t *lock) lock->owner_cpu = -1; } +#if 0 /* This can cause lockups */ static void __write_lock_debug(rwlock_t *lock) { int print_once = 1; @@ -243,12 +245,12 @@ static void __write_lock_debug(rwlock_t *lock) } } } +#endif void _raw_write_lock(rwlock_t *lock) { debug_write_lock_before(lock); - if (unlikely(!__raw_write_trylock(&lock->raw_lock))) - __write_lock_debug(lock); + __raw_write_lock(&lock->raw_lock); debug_write_lock_after(lock); } diff --git a/lib/ts_bm.c b/lib/ts_bm.c index c4c1ac5fb..791726579 100644 --- a/lib/ts_bm.c +++ b/lib/ts_bm.c @@ -112,15 +112,14 @@ static int subpattern(u8 *pattern, int i, int j, int g) return ret; } -static void compute_prefix_tbl(struct ts_bm *bm, const u8 *pattern, - unsigned int len) +static void compute_prefix_tbl(struct ts_bm *bm) { int i, j, g; for (i = 0; i < ASIZE; i++) - bm->bad_shift[i] = len; - for (i = 0; i < len - 1; i++) - bm->bad_shift[pattern[i]] = len - 1 - i; + bm->bad_shift[i] = bm->patlen; + for (i = 0; i < bm->patlen - 1; i++) + bm->bad_shift[bm->pattern[i]] = bm->patlen - 1 - i; /* Compute the good shift array, used to match reocurrences * of a subpattern */ @@ -151,8 +150,8 @@ static struct ts_config *bm_init(const void *pattern, unsigned int len, bm = ts_config_priv(conf); bm->patlen = len; bm->pattern = (u8 *) bm->good_shift + prefix_tbl_len; - compute_prefix_tbl(bm, pattern, len); memcpy(bm->pattern, pattern, len); + compute_prefix_tbl(bm); return conf; } diff --git a/mm/Kconfig b/mm/Kconfig index 67b3f74ca..ab1007556 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -115,7 +115,7 @@ config SPARSEMEM_EXTREME # eventually, we can have this option just 'select SPARSEMEM' config MEMORY_HOTPLUG bool "Allow for memory hot-add" - depends on SPARSEMEM && HOTPLUG && !SOFTWARE_SUSPEND + depends on SPARSEMEM && HOTPLUG && !SOFTWARE_SUSPEND && ARCH_ENABLE_MEMORY_HOTPLUG comment "Memory hotplug is currently incompatible with Software Suspend" depends on SPARSEMEM && HOTPLUG && SOFTWARE_SUSPEND @@ -126,12 +126,16 @@ comment "Memory hotplug is currently incompatible with Software Suspend" # Default to 4 for wider testing, though 8 might be more appropriate. # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. +# XEN on x86 architecture uses the mapping field on pagetable pages to store a +# pointer to the destructor. This conflicts with pte_lock_deinit(). # config SPLIT_PTLOCK_CPUS int default "4096" if ARM && !CPU_CACHE_VIPT default "4096" if PARISC && !PA20 - default "4096" + default "4096" if X86_XEN || X86_64_XEN + default "4096" if !64BIT + default "4" # # support for page migration diff --git a/mm/filemap.c b/mm/filemap.c index 4b0449581..96f354e18 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -518,18 +518,11 @@ EXPORT_SYMBOL(unlock_page); */ void end_page_writeback(struct page *page) { - struct zone *zone = page_zone(page); if (!TestClearPageReclaim(page) || rotate_reclaimable_page(page)) { if (!test_clear_page_writeback(page)) BUG(); } smp_mb__after_clear_bit(); - if (zone->all_unreclaimable) { - spin_lock(&zone->lock); - zone->all_unreclaimable = 0; - zone->pages_scanned = 0; - spin_unlock(&zone->lock); - } wake_up_page(page, PG_writeback); } EXPORT_SYMBOL(end_page_writeback); @@ -2022,14 +2015,21 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, do { unsigned long index; unsigned long offset; - unsigned long maxlen; size_t copied; offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */ index = pos >> PAGE_CACHE_SHIFT; bytes = PAGE_CACHE_SIZE - offset; - if (bytes > count) - bytes = count; + + /* Limit the size of the copy to the caller's write size */ + bytes = min(bytes, count); + + /* + * Limit the size of the copy to that of the current segment, + * because fault_in_pages_readable() doesn't know how to walk + * segments. + */ + bytes = min(bytes, cur_iov->iov_len - iov_base); /* * Bring in the user page that we will copy from _first_. @@ -2037,10 +2037,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, * same page as we're writing to, without it being marked * up-to-date. */ - maxlen = cur_iov->iov_len - iov_base; - if (maxlen > bytes) - maxlen = bytes; - fault_in_pages_readable(buf, maxlen); + fault_in_pages_readable(buf, bytes); page = __grab_cache_page(mapping,index,&cached_page,&lru_pvec); if (!page) { @@ -2048,6 +2045,12 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, break; } + if (unlikely(bytes == 0)) { + status = 0; + copied = 0; + goto zero_length_segment; + } + status = a_ops->prepare_write(file, page, offset, offset+bytes); if (unlikely(status)) { loff_t isize = i_size_read(inode); @@ -2077,7 +2080,8 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov, page_cache_release(page); continue; } - if (likely(copied > 0)) { +zero_length_segment: + if (likely(copied >= 0)) { if (!status) status = copied; diff --git a/mm/filemap.h b/mm/filemap.h index 13793ba0c..efd01426b 100644 --- a/mm/filemap.h +++ b/mm/filemap.h @@ -78,7 +78,7 @@ filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes) const struct iovec *iov = *iovp; size_t base = *basep; - while (bytes) { + do { int copy = min(bytes, iov->iov_len - base); bytes -= copy; @@ -87,7 +87,7 @@ filemap_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes) iov++; base = 0; } - } + } while (bytes); *iovp = iov; *basep = base; } diff --git a/mm/highmem.c b/mm/highmem.c index 9b274fdf9..a02c6dec9 100644 --- a/mm/highmem.c +++ b/mm/highmem.c @@ -142,6 +142,17 @@ start: return vaddr; } +#ifdef CONFIG_XEN +void kmap_flush_unused(void) +{ + spin_lock(&kmap_lock); + flush_all_zero_pkmaps(); + spin_unlock(&kmap_lock); +} + +EXPORT_SYMBOL(kmap_flush_unused); +#endif + void fastcall *kmap_high(struct page *page) { unsigned long vaddr; diff --git a/mm/memory.c b/mm/memory.c index 1bd040470..9f75bb2ca 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -402,7 +402,8 @@ struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr, pte_ * and that the resulting page looks ok. */ if (unlikely(!pfn_valid(pfn))) { - print_bad_pte(vma, pte, addr); + if (!(vma->vm_flags & VM_RESERVED)) + print_bad_pte(vma, pte, addr); return NULL; } @@ -1017,6 +1018,26 @@ int get_user_pages(struct task_struct *tsk, struct mm_struct *mm, continue; } +#ifdef CONFIG_XEN + if (vma && (vma->vm_flags & VM_FOREIGN)) { + struct page **map = vma->vm_private_data; + int offset = (start - vma->vm_start) >> PAGE_SHIFT; + if (map[offset] != NULL) { + if (pages) { + struct page *page = map[offset]; + + pages[i] = page; + get_page(page); + } + if (vmas) + vmas[i] = vma; + i++; + start += PAGE_SIZE; + len--; + continue; + } + } +#endif if (!vma || (vma->vm_flags & (VM_IO | VM_PFNMAP)) || !(vm_flags & vma->vm_flags)) return i ? : -EFAULT; @@ -1356,6 +1377,102 @@ int remap_pfn_range(struct vm_area_struct *vma, unsigned long addr, } EXPORT_SYMBOL(remap_pfn_range); +#ifdef CONFIG_XEN +static inline int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pte_t *pte; + int err; + struct page *pmd_page; + spinlock_t *ptl; + + pte = (mm == &init_mm) ? + pte_alloc_kernel(pmd, addr) : + pte_alloc_map_lock(mm, pmd, addr, &ptl); + if (!pte) + return -ENOMEM; + + BUG_ON(pmd_huge(*pmd)); + + pmd_page = pmd_page(*pmd); + + do { + err = fn(pte, pmd_page, addr, data); + if (err) + break; + } while (pte++, addr += PAGE_SIZE, addr != end); + + if (mm != &init_mm) + pte_unmap_unlock(pte-1, ptl); + return err; +} + +static inline int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pmd_t *pmd; + unsigned long next; + int err; + + pmd = pmd_alloc(mm, pud, addr); + if (!pmd) + return -ENOMEM; + do { + next = pmd_addr_end(addr, end); + err = apply_to_pte_range(mm, pmd, addr, next, fn, data); + if (err) + break; + } while (pmd++, addr = next, addr != end); + return err; +} + +static inline int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd, + unsigned long addr, unsigned long end, + pte_fn_t fn, void *data) +{ + pud_t *pud; + unsigned long next; + int err; + + pud = pud_alloc(mm, pgd, addr); + if (!pud) + return -ENOMEM; + do { + next = pud_addr_end(addr, end); + err = apply_to_pmd_range(mm, pud, addr, next, fn, data); + if (err) + break; + } while (pud++, addr = next, addr != end); + return err; +} + +/* + * Scan a region of virtual memory, filling in page tables as necessary + * and calling a provided function on each leaf page table. + */ +int apply_to_page_range(struct mm_struct *mm, unsigned long addr, + unsigned long size, pte_fn_t fn, void *data) +{ + pgd_t *pgd; + unsigned long next; + unsigned long end = addr + size; + int err; + + BUG_ON(addr >= end); + pgd = pgd_offset(mm, addr); + do { + next = pgd_addr_end(addr, end); + err = apply_to_pud_range(mm, pgd, addr, next, fn, data); + if (err) + break; + } while (pgd++, addr = next, addr != end); + return err; +} +EXPORT_SYMBOL_GPL(apply_to_page_range); +#endif + /* * handle_pte_fault chooses page fault handler according to an entry * which was read non-atomically. Before making any commitment, on diff --git a/mm/mmap.c b/mm/mmap.c index c34750cf9..f6940d1b3 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -920,6 +920,10 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr, if (!len) return -EINVAL; + error = arch_mmap_check(addr, len, flags); + if (error) + return error; + /* Careful about overflows.. */ len = PAGE_ALIGN(len); if (!len || len > TASK_SIZE) @@ -1940,6 +1944,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len) unsigned long flags; struct rb_node ** rb_link, * rb_parent; pgoff_t pgoff = addr >> PAGE_SHIFT; + int error; len = PAGE_ALIGN(len); if (!len) @@ -1948,6 +1953,12 @@ unsigned long do_brk(unsigned long addr, unsigned long len) if ((addr + len) > TASK_SIZE || (addr + len) < addr) return -EINVAL; + flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; + + error = arch_mmap_check(addr, len, flags); + if (error) + return error; + /* * mlock MCL_FUTURE? */ @@ -1991,8 +2002,6 @@ unsigned long do_brk(unsigned long addr, unsigned long len) !vx_vmpages_avail(mm, len >> PAGE_SHIFT)) return -ENOMEM; - flags = VM_DATA_DEFAULT_FLAGS | VM_ACCOUNT | mm->def_flags; - /* Can we just expand an old private anonymous mapping? */ if (vma_merge(mm, prev, addr, addr + len, flags, NULL, NULL, pgoff, NULL)) @@ -2033,6 +2042,10 @@ void exit_mmap(struct mm_struct *mm) unsigned long nr_accounted = 0; unsigned long end; +#ifdef arch_exit_mmap + arch_exit_mmap(mm); +#endif + lru_add_drain(); flush_cache_mm(mm); tlb = tlb_gather_mmu(mm, 1); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index b747cbf8b..5581dd1a4 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -439,7 +439,8 @@ static void __free_pages_ok(struct page *page, unsigned int order) int i; int reserved = 0; - arch_free_page(page, order); + if (arch_free_page(page, order)) + return; if (!PageHighMem(page)) mutex_debug_check_no_locks_freed(page_address(page), PAGE_SIZE<mapping = NULL; @@ -751,11 +753,6 @@ static void fastcall free_hot_cold_page(struct page *page, int cold) if (pcp->count >= pcp->high) { free_pages_bulk(zone, pcp->batch, &pcp->list, 0); pcp->count -= pcp->batch; - } else if (zone->all_unreclaimable) { - spin_lock(&zone->lock); - zone->all_unreclaimable = 0; - zone->pages_scanned = 0; - spin_unlock(&zone->lock); } local_irq_restore(flags); put_cpu(); diff --git a/mm/pdflush.c b/mm/pdflush.c index c4b6d0afd..4842716d1 100644 --- a/mm/pdflush.c +++ b/mm/pdflush.c @@ -104,21 +104,20 @@ static int __pdflush(struct pdflush_work *my_work) list_move(&my_work->list, &pdflush_list); my_work->when_i_went_to_sleep = jiffies; spin_unlock_irq(&pdflush_lock); - schedule(); - if (try_to_freeze()) { - spin_lock_irq(&pdflush_lock); - continue; - } - + try_to_freeze(); spin_lock_irq(&pdflush_lock); if (!list_empty(&my_work->list)) { - printk("pdflush: bogus wakeup!\n"); + /* + * Someone woke us up, but without removing our control + * structure from the global list. swsusp will do this + * in try_to_freeze()->refrigerator(). Handle it. + */ my_work->fn = NULL; continue; } if (my_work->fn == NULL) { - printk("pdflush: NULL work function\n"); + printk("pdflush: bogus wakeup\n"); continue; } spin_unlock_irq(&pdflush_lock); diff --git a/mm/swapfile.c b/mm/swapfile.c index 234b2f173..a3bed3ff3 100644 --- a/mm/swapfile.c +++ b/mm/swapfile.c @@ -441,11 +441,12 @@ int swap_type_of(dev_t device) if (!(swap_info[i].flags & SWP_WRITEOK)) continue; + if (!device) { spin_unlock(&swap_lock); return i; } - inode = swap_info->swap_file->f_dentry->d_inode; + inode = swap_info[i].swap_file->f_dentry->d_inode; if (S_ISBLK(inode->i_mode) && device == MKDEV(imajor(inode), iminor(inode))) { spin_unlock(&swap_lock); diff --git a/net/8021q/vlan.c b/net/8021q/vlan.c index 3948949a6..729abc439 100644 --- a/net/8021q/vlan.c +++ b/net/8021q/vlan.c @@ -67,10 +67,6 @@ static struct packet_type vlan_packet_type = { .func = vlan_skb_recv, /* VLAN receive method */ }; -/* Bits of netdev state that are propagated from real device to virtual */ -#define VLAN_LINK_STATE_MASK \ - ((1<<__LINK_STATE_PRESENT)|(1<<__LINK_STATE_NOCARRIER)|(1<<__LINK_STATE_DORMANT)) - /* End of global variables definitions. */ /* @@ -470,7 +466,9 @@ static struct net_device *register_vlan_device(const char *eth_IF_name, new_dev->flags = real_dev->flags; new_dev->flags &= ~IFF_UP; - new_dev->state = real_dev->state & ~(1<<__LINK_STATE_START); + new_dev->state = (real_dev->state & ((1<<__LINK_STATE_NOCARRIER) | + (1<<__LINK_STATE_DORMANT))) | + (1<<__LINK_STATE_PRESENT); /* need 4 bytes for extra VLAN header info, * hope the underlying device can handle it. diff --git a/net/atm/clip.c b/net/atm/clip.c index 72d852982..4f44e98f1 100644 --- a/net/atm/clip.c +++ b/net/atm/clip.c @@ -98,7 +98,7 @@ static void unlink_clip_vcc(struct clip_vcc *clip_vcc) printk(KERN_CRIT "!clip_vcc->entry (clip_vcc %p)\n", clip_vcc); return; } - spin_lock_bh(&entry->neigh->dev->xmit_lock); /* block clip_start_xmit() */ + netif_tx_lock_bh(entry->neigh->dev); /* block clip_start_xmit() */ entry->neigh->used = jiffies; for (walk = &entry->vccs; *walk; walk = &(*walk)->next) if (*walk == clip_vcc) { @@ -120,9 +120,9 @@ static void unlink_clip_vcc(struct clip_vcc *clip_vcc) goto out; } printk(KERN_CRIT "ATMARP: unlink_clip_vcc failed (entry %p, vcc " - "0x%p)\n", entry, clip_vcc); - out: - spin_unlock_bh(&entry->neigh->dev->xmit_lock); + "0x%p)\n",entry,clip_vcc); +out: + netif_tx_unlock_bh(entry->neigh->dev); } /* The neighbour entry n->lock is held. */ diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 0c88a2ac3..f8dbcee80 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -145,9 +145,9 @@ static int br_set_tx_csum(struct net_device *dev, u32 data) struct net_bridge *br = netdev_priv(dev); if (data) - br->feature_mask |= NETIF_F_IP_CSUM; + br->feature_mask |= NETIF_F_NO_CSUM; else - br->feature_mask &= ~NETIF_F_IP_CSUM; + br->feature_mask &= ~NETIF_F_ALL_CSUM; br_features_recompute(br); return 0; @@ -184,6 +184,6 @@ void br_dev_setup(struct net_device *dev) dev->set_mac_address = br_set_mac_address; dev->priv_flags = IFF_EBRIDGE; - dev->features = NETIF_F_SG | NETIF_F_FRAGLIST - | NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_IP_CSUM; + dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA | + NETIF_F_TSO | NETIF_F_NO_CSUM | NETIF_F_GSO_ROBUST; } diff --git a/net/bridge/br_forward.c b/net/bridge/br_forward.c index 56f3aa47e..056df7ceb 100644 --- a/net/bridge/br_forward.c +++ b/net/bridge/br_forward.c @@ -38,16 +38,20 @@ static inline unsigned packet_length(const struct sk_buff *skb) int br_dev_queue_push_xmit(struct sk_buff *skb) { /* drop mtu oversized packets except tso */ - if (packet_length(skb) > skb->dev->mtu && !skb_shinfo(skb)->tso_size) + if (skb->len > skb->dev->mtu && !skb_is_gso(skb)) kfree_skb(skb); else { #ifdef CONFIG_BRIDGE_NETFILTER /* ip_refrag calls ip_fragment, doesn't copy the MAC header. */ - nf_bridge_maybe_copy_header(skb); + if (nf_bridge_maybe_copy_header(skb)) + kfree_skb(skb); + else #endif - skb_push(skb, ETH_HLEN); + { + skb_push(skb, ETH_HLEN); - dev_queue_xmit(skb); + dev_queue_xmit(skb); + } } return 0; diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c index f5d47bf4f..f55ef682e 100644 --- a/net/bridge/br_if.c +++ b/net/bridge/br_if.c @@ -372,17 +372,28 @@ void br_features_recompute(struct net_bridge *br) struct net_bridge_port *p; unsigned long features, checksum; - features = br->feature_mask &~ NETIF_F_IP_CSUM; - checksum = br->feature_mask & NETIF_F_IP_CSUM; + checksum = br->feature_mask & NETIF_F_ALL_CSUM ? NETIF_F_NO_CSUM : 0; + features = br->feature_mask & ~NETIF_F_ALL_CSUM; list_for_each_entry(p, &br->port_list, list) { - if (!(p->dev->features - & (NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM))) + unsigned long feature = p->dev->features; + + if (checksum & NETIF_F_NO_CSUM && !(feature & NETIF_F_NO_CSUM)) + checksum ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM; + if (checksum & NETIF_F_HW_CSUM && !(feature & NETIF_F_HW_CSUM)) + checksum ^= NETIF_F_HW_CSUM | NETIF_F_IP_CSUM; + if (!(feature & NETIF_F_IP_CSUM)) checksum = 0; - features &= p->dev->features; + + if (feature & NETIF_F_GSO) + feature |= NETIF_F_TSO; + feature |= NETIF_F_GSO; + + features &= feature; } - br->dev->features = features | checksum | NETIF_F_LLTX; + br->dev->features = features | checksum | NETIF_F_LLTX | + NETIF_F_GSO_ROBUST; } /* called with RTNL */ diff --git a/net/bridge/br_netfilter.c b/net/bridge/br_netfilter.c index 3da926444..a64272783 100644 --- a/net/bridge/br_netfilter.c +++ b/net/bridge/br_netfilter.c @@ -769,7 +769,7 @@ static int br_nf_dev_queue_xmit(struct sk_buff *skb) { if (skb->protocol == htons(ETH_P_IP) && skb->len > skb->dev->mtu && - !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size)) + !skb_is_gso(skb)) return ip_fragment(skb, br_dev_queue_push_xmit); else return br_dev_queue_push_xmit(skb); @@ -877,8 +877,9 @@ static unsigned int ip_sabotage_out(unsigned int hook, struct sk_buff **pskb, struct sk_buff *skb = *pskb; if ((out->hard_start_xmit == br_dev_xmit && - okfn != br_nf_forward_finish && - okfn != br_nf_local_out_finish && okfn != br_nf_dev_queue_xmit) + okfn != br_nf_forward_finish && + okfn != br_nf_local_out_finish && + okfn != br_nf_dev_queue_xmit) #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE) || ((out->priv_flags & IFF_802_1Q_VLAN) && VLAN_DEV_INFO(out)->real_dev->hard_start_xmit == br_dev_xmit) diff --git a/net/bridge/netfilter/ebt_ulog.c b/net/bridge/netfilter/ebt_ulog.c index d051e2c09..1b37fa7ac 100644 --- a/net/bridge/netfilter/ebt_ulog.c +++ b/net/bridge/netfilter/ebt_ulog.c @@ -75,6 +75,9 @@ static void ulog_send(unsigned int nlgroup) if (timer_pending(&ub->timer)) del_timer(&ub->timer); + if (!ub->skb) + return; + /* last nlmsg needs NLMSG_DONE */ if (ub->qlen > 1) ub->lastnlh->nlmsg_type = NLMSG_DONE; diff --git a/net/core/dev.c b/net/core/dev.c index 279539330..cbc26ac62 100644 --- a/net/core/dev.c +++ b/net/core/dev.c @@ -115,8 +115,15 @@ #include #include #include +#include #include +#ifdef CONFIG_XEN +#include +#include +#include +#endif + /* * The list of packet types we will receive (as opposed to discard) * and the routines to invoke. @@ -1042,7 +1049,7 @@ static inline void net_timestamp(struct sk_buff *skb) * taps currently in use. */ -void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) +static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev) { struct packet_type *ptype; @@ -1180,6 +1187,45 @@ out: return ret; } +/** + * skb_gso_segment - Perform segmentation on skb. + * @skb: buffer to segment + * @features: features for the output path (see dev->features) + * + * This function segments the given skb and returns a list of segments. + * + * It may return NULL if the skb requires no segmentation. This is + * only possible when GSO is used for verifying header integrity. + */ +struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = ERR_PTR(-EPROTONOSUPPORT); + struct packet_type *ptype; + int type = skb->protocol; + + BUG_ON(skb_shinfo(skb)->frag_list); + BUG_ON(skb->ip_summed != CHECKSUM_HW); + + skb->mac.raw = skb->data; + skb->mac_len = skb->nh.raw - skb->data; + __skb_pull(skb, skb->mac_len); + + rcu_read_lock(); + list_for_each_entry_rcu(ptype, &ptype_base[ntohs(type) & 15], list) { + if (ptype->type == type && !ptype->dev && ptype->gso_segment) { + segs = ptype->gso_segment(skb, features); + break; + } + } + rcu_read_unlock(); + + __skb_push(skb, skb->data - skb->mac.raw); + + return segs; +} + +EXPORT_SYMBOL(skb_gso_segment); + /* Take action when hardware reception checksum errors are detected. */ #ifdef CONFIG_BUG void netdev_rx_csum_fault(struct net_device *dev) @@ -1216,78 +1262,148 @@ static inline int illegal_highdma(struct net_device *dev, struct sk_buff *skb) #define illegal_highdma(dev, skb) (0) #endif -/* Keep head the same: replace data */ -int __skb_linearize(struct sk_buff *skb, gfp_t gfp_mask) -{ - unsigned int size; - u8 *data; - long offset; - struct skb_shared_info *ninfo; - int headerlen = skb->data - skb->head; - int expand = (skb->tail + skb->data_len) - skb->end; - - if (skb_shared(skb)) - BUG(); - - if (expand <= 0) - expand = 0; - - size = skb->end - skb->head + expand; - size = SKB_DATA_ALIGN(size); - data = kmalloc(size + sizeof(struct skb_shared_info), gfp_mask); - if (!data) - return -ENOMEM; - - /* Copy entire thing */ - if (skb_copy_bits(skb, -headerlen, data, headerlen + skb->len)) - BUG(); - - /* Set up shinfo */ - ninfo = (struct skb_shared_info*)(data + size); - atomic_set(&ninfo->dataref, 1); - ninfo->tso_size = skb_shinfo(skb)->tso_size; - ninfo->tso_segs = skb_shinfo(skb)->tso_segs; - ninfo->nr_frags = 0; - ninfo->frag_list = NULL; - - /* Offset between the two in bytes */ - offset = data - skb->head; - - /* Free old data. */ - skb_release_data(skb); - - skb->head = data; - skb->end = data + size; - - /* Set up new pointers */ - skb->h.raw += offset; - skb->nh.raw += offset; - skb->mac.raw += offset; - skb->tail += offset; - skb->data += offset; - - /* We are no longer a clone, even if we were. */ - skb->cloned = 0; - - skb->tail += skb->data_len; - skb->data_len = 0; +struct dev_gso_cb { + void (*destructor)(struct sk_buff *skb); +}; + +#define DEV_GSO_CB(skb) ((struct dev_gso_cb *)(skb)->cb) + +static void dev_gso_skb_destructor(struct sk_buff *skb) +{ + struct dev_gso_cb *cb; + + do { + struct sk_buff *nskb = skb->next; + + skb->next = nskb->next; + nskb->next = NULL; + kfree_skb(nskb); + } while (skb->next); + + cb = DEV_GSO_CB(skb); + if (cb->destructor) + cb->destructor(skb); +} + +/** + * dev_gso_segment - Perform emulated hardware segmentation on skb. + * @skb: buffer to segment + * + * This function segments the given skb and stores the list of segments + * in skb->next. + */ +static int dev_gso_segment(struct sk_buff *skb) +{ + struct net_device *dev = skb->dev; + struct sk_buff *segs; + int features = dev->features & ~(illegal_highdma(dev, skb) ? + NETIF_F_SG : 0); + + segs = skb_gso_segment(skb, features); + + /* Verifying header integrity only. */ + if (!segs) + return 0; + + if (unlikely(IS_ERR(segs))) + return PTR_ERR(segs); + + skb->next = segs; + DEV_GSO_CB(skb)->destructor = skb->destructor; + skb->destructor = dev_gso_skb_destructor; + + return 0; +} + +int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev) +{ + if (likely(!skb->next)) { + if (netdev_nit) + dev_queue_xmit_nit(skb, dev); + + if (netif_needs_gso(dev, skb)) { + if (unlikely(dev_gso_segment(skb))) + goto out_kfree_skb; + if (skb->next) + goto gso; + } + + return dev->hard_start_xmit(skb, dev); + } + +gso: + do { + struct sk_buff *nskb = skb->next; + int rc; + + skb->next = nskb->next; + nskb->next = NULL; + rc = dev->hard_start_xmit(nskb, dev); + if (unlikely(rc)) { + nskb->next = skb->next; + skb->next = nskb; + return rc; + } + if (unlikely(netif_queue_stopped(dev) && skb->next)) + return NETDEV_TX_BUSY; + } while (skb->next); + + skb->destructor = DEV_GSO_CB(skb)->destructor; + +out_kfree_skb: + kfree_skb(skb); return 0; } #define HARD_TX_LOCK(dev, cpu) { \ if ((dev->features & NETIF_F_LLTX) == 0) { \ - spin_lock(&dev->xmit_lock); \ - dev->xmit_lock_owner = cpu; \ + netif_tx_lock(dev); \ } \ } #define HARD_TX_UNLOCK(dev) { \ if ((dev->features & NETIF_F_LLTX) == 0) { \ - dev->xmit_lock_owner = -1; \ - spin_unlock(&dev->xmit_lock); \ + netif_tx_unlock(dev); \ } \ } +#ifdef CONFIG_XEN +inline int skb_checksum_setup(struct sk_buff *skb) +{ + if (skb->proto_csum_blank) { + if (skb->protocol != htons(ETH_P_IP)) + goto out; + skb->h.raw = (unsigned char *)skb->nh.iph + 4*skb->nh.iph->ihl; + if (skb->h.raw >= skb->tail) + goto out; + switch (skb->nh.iph->protocol) { + case IPPROTO_TCP: + skb->csum = offsetof(struct tcphdr, check); + break; + case IPPROTO_UDP: + skb->csum = offsetof(struct udphdr, check); + break; + default: + if (net_ratelimit()) + printk(KERN_ERR "Attempting to checksum a non-" + "TCP/UDP packet, dropping a protocol" + " %d packet", skb->nh.iph->protocol); + goto out; + } + if ((skb->h.raw + skb->csum + 2) > skb->tail) + goto out; + skb->ip_summed = CHECKSUM_HW; + skb->proto_csum_blank = 0; + } + return 0; +out: + return -EPROTO; +} +#else +inline int skb_checksum_setup(struct sk_buff *skb) { return 0; } +#endif + + /** * dev_queue_xmit - transmit a buffer * @skb: buffer to transmit @@ -1320,9 +1436,19 @@ int dev_queue_xmit(struct sk_buff *skb) struct Qdisc *q; int rc = -ENOMEM; + /* If a checksum-deferred packet is forwarded to a device that needs a + * checksum, correct the pointers and force checksumming. + */ + if (skb_checksum_setup(skb)) + goto out_kfree_skb; + + /* GSO will handle the following emulations directly. */ + if (netif_needs_gso(dev, skb)) + goto gso; + if (skb_shinfo(skb)->frag_list && !(dev->features & NETIF_F_FRAGLIST) && - __skb_linearize(skb, GFP_ATOMIC)) + __skb_linearize(skb)) goto out_kfree_skb; /* Fragmented skb is linearized if device does not support SG, @@ -1331,25 +1457,26 @@ int dev_queue_xmit(struct sk_buff *skb) */ if (skb_shinfo(skb)->nr_frags && (!(dev->features & NETIF_F_SG) || illegal_highdma(dev, skb)) && - __skb_linearize(skb, GFP_ATOMIC)) + __skb_linearize(skb)) goto out_kfree_skb; /* If packet is not checksummed and device does not support * checksumming for this protocol, complete checksumming here. */ if (skb->ip_summed == CHECKSUM_HW && - (!(dev->features & (NETIF_F_HW_CSUM | NETIF_F_NO_CSUM)) && + (!(dev->features & NETIF_F_GEN_CSUM) && (!(dev->features & NETIF_F_IP_CSUM) || skb->protocol != htons(ETH_P_IP)))) if (skb_checksum_help(skb, 0)) goto out_kfree_skb; +gso: spin_lock_prefetch(&dev->queue_lock); /* Disable soft irqs for various locks below. Also * stops preemption for RCU. */ - local_bh_disable(); + rcu_read_lock_bh(); /* Updates of qdisc are serialized by queue_lock. * The struct Qdisc which is pointed to by qdisc is now a @@ -1383,8 +1510,8 @@ int dev_queue_xmit(struct sk_buff *skb) /* The device has no queue. Common case for software devices: loopback, all the sorts of tunnels... - Really, it is unlikely that xmit_lock protection is necessary here. - (f.e. loopback and IP tunnels are clean ignoring statistics + Really, it is unlikely that netif_tx_lock protection is necessary + here. (f.e. loopback and IP tunnels are clean ignoring statistics counters.) However, it is possible, that they rely on protection made by us here. @@ -1400,11 +1527,8 @@ int dev_queue_xmit(struct sk_buff *skb) HARD_TX_LOCK(dev, cpu); if (!netif_queue_stopped(dev)) { - if (netdev_nit) - dev_queue_xmit_nit(skb, dev); - rc = 0; - if (!dev->hard_start_xmit(skb, dev)) { + if (!dev_hard_start_xmit(skb, dev)) { HARD_TX_UNLOCK(dev); goto out; } @@ -1423,13 +1547,13 @@ int dev_queue_xmit(struct sk_buff *skb) } rc = -ENETDOWN; - local_bh_enable(); + rcu_read_unlock_bh(); out_kfree_skb: kfree_skb(skb); return rc; out: - local_bh_enable(); + rcu_read_unlock_bh(); return rc; } @@ -1707,6 +1831,19 @@ int netif_receive_skb(struct sk_buff *skb) } #endif +#ifdef CONFIG_XEN + switch (skb->ip_summed) { + case CHECKSUM_UNNECESSARY: + skb->proto_data_valid = 1; + break; + case CHECKSUM_HW: + /* XXX Implement me. */ + default: + skb->proto_data_valid = 0; + break; + } +#endif + list_for_each_entry_rcu(ptype, &ptype_all, list) { if (!ptype->dev || ptype->dev == skb->dev) { if (pt_prev) @@ -2793,7 +2930,7 @@ int register_netdevice(struct net_device *dev) BUG_ON(dev->reg_state != NETREG_UNINITIALIZED); spin_lock_init(&dev->queue_lock); - spin_lock_init(&dev->xmit_lock); + spin_lock_init(&dev->_xmit_lock); dev->xmit_lock_owner = -1; #ifdef CONFIG_NET_CLS_ACT spin_lock_init(&dev->ingress_lock); @@ -2837,9 +2974,7 @@ int register_netdevice(struct net_device *dev) /* Fix illegal SG+CSUM combinations. */ if ((dev->features & NETIF_F_SG) && - !(dev->features & (NETIF_F_IP_CSUM | - NETIF_F_NO_CSUM | - NETIF_F_HW_CSUM))) { + !(dev->features & NETIF_F_ALL_CSUM)) { printk("%s: Dropping NETIF_F_SG since no checksum feature.\n", dev->name); dev->features &= ~NETIF_F_SG; @@ -3379,7 +3514,6 @@ subsys_initcall(net_dev_init); EXPORT_SYMBOL(__dev_get_by_index); EXPORT_SYMBOL(__dev_get_by_name); EXPORT_SYMBOL(__dev_remove_pack); -EXPORT_SYMBOL(__skb_linearize); EXPORT_SYMBOL(dev_valid_name); EXPORT_SYMBOL(dev_add_pack); EXPORT_SYMBOL(dev_alloc_name); @@ -3414,6 +3548,7 @@ EXPORT_SYMBOL(unregister_netdevice_notifier); EXPORT_SYMBOL(net_enable_timestamp); EXPORT_SYMBOL(net_disable_timestamp); EXPORT_SYMBOL(dev_get_flags); +EXPORT_SYMBOL(skb_checksum_setup); #if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE) EXPORT_SYMBOL(br_handle_frame_hook); diff --git a/net/core/dev_mcast.c b/net/core/dev_mcast.c index 05d608508..c57d887da 100644 --- a/net/core/dev_mcast.c +++ b/net/core/dev_mcast.c @@ -62,7 +62,7 @@ * Device mc lists are changed by bh at least if IPv6 is enabled, * so that it must be bh protected. * - * We block accesses to device mc filters with dev->xmit_lock. + * We block accesses to device mc filters with netif_tx_lock. */ /* @@ -93,9 +93,9 @@ static void __dev_mc_upload(struct net_device *dev) void dev_mc_upload(struct net_device *dev) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); __dev_mc_upload(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } /* @@ -107,7 +107,7 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl) int err = 0; struct dev_mc_list *dmi, **dmip; - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); for (dmip = &dev->mc_list; (dmi = *dmip) != NULL; dmip = &dmi->next) { /* @@ -139,13 +139,13 @@ int dev_mc_delete(struct net_device *dev, void *addr, int alen, int glbl) */ __dev_mc_upload(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return 0; } } err = -ENOENT; done: - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return err; } @@ -160,7 +160,7 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl) dmi1 = kmalloc(sizeof(*dmi), GFP_ATOMIC); - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); for (dmi = dev->mc_list; dmi != NULL; dmi = dmi->next) { if (memcmp(dmi->dmi_addr, addr, dmi->dmi_addrlen) == 0 && dmi->dmi_addrlen == alen) { @@ -176,7 +176,7 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl) } if ((dmi = dmi1) == NULL) { - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return -ENOMEM; } memcpy(dmi->dmi_addr, addr, alen); @@ -189,11 +189,11 @@ int dev_mc_add(struct net_device *dev, void *addr, int alen, int glbl) __dev_mc_upload(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return 0; done: - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); kfree(dmi1); return err; } @@ -204,7 +204,7 @@ done: void dev_mc_discard(struct net_device *dev) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); while (dev->mc_list != NULL) { struct dev_mc_list *tmp = dev->mc_list; @@ -215,7 +215,7 @@ void dev_mc_discard(struct net_device *dev) } dev->mc_count = 0; - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } #ifdef CONFIG_PROC_FS @@ -250,7 +250,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v) struct dev_mc_list *m; struct net_device *dev = v; - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); for (m = dev->mc_list; m; m = m->next) { int i; @@ -262,7 +262,7 @@ static int dev_mc_seq_show(struct seq_file *seq, void *v) seq_putc(seq, '\n'); } - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); return 0; } diff --git a/net/core/dst.c b/net/core/dst.c index 470c05bc4..1a5e49da0 100644 --- a/net/core/dst.c +++ b/net/core/dst.c @@ -95,12 +95,11 @@ static void dst_run_gc(unsigned long dummy) dst_gc_timer_inc = DST_GC_INC; dst_gc_timer_expires = DST_GC_MIN; } - dst_gc_timer.expires = jiffies + dst_gc_timer_expires; #if RT_CACHE_DEBUG >= 2 printk("dst_total: %d/%d %ld\n", atomic_read(&dst_total), delayed, dst_gc_timer_expires); #endif - add_timer(&dst_gc_timer); + mod_timer(&dst_gc_timer, jiffies + dst_gc_timer_expires); out: spin_unlock(&dst_lock); diff --git a/net/core/ethtool.c b/net/core/ethtool.c index c680b7e04..987cccaa0 100644 --- a/net/core/ethtool.c +++ b/net/core/ethtool.c @@ -30,7 +30,7 @@ u32 ethtool_op_get_link(struct net_device *dev) u32 ethtool_op_get_tx_csum(struct net_device *dev) { - return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0; + return (dev->features & NETIF_F_ALL_CSUM) != 0; } int ethtool_op_set_tx_csum(struct net_device *dev, u32 data) @@ -437,7 +437,7 @@ static int ethtool_set_pauseparam(struct net_device *dev, void __user *useraddr) { struct ethtool_pauseparam pauseparam; - if (!dev->ethtool_ops->get_pauseparam) + if (!dev->ethtool_ops->set_pauseparam) return -EOPNOTSUPP; if (copy_from_user(&pauseparam, useraddr, sizeof(pauseparam))) @@ -551,9 +551,7 @@ static int ethtool_set_sg(struct net_device *dev, char __user *useraddr) return -EFAULT; if (edata.data && - !(dev->features & (NETIF_F_IP_CSUM | - NETIF_F_NO_CSUM | - NETIF_F_HW_CSUM))) + !(dev->features & NETIF_F_ALL_CSUM)) return -EINVAL; return __ethtool_set_sg(dev, edata.data); @@ -561,7 +559,7 @@ static int ethtool_set_sg(struct net_device *dev, char __user *useraddr) static int ethtool_get_tso(struct net_device *dev, char __user *useraddr) { - struct ethtool_value edata = { ETHTOOL_GTSO }; + struct ethtool_value edata = { ETHTOOL_GUFO }; if (!dev->ethtool_ops->get_tso) return -EOPNOTSUPP; @@ -616,6 +614,29 @@ static int ethtool_set_ufo(struct net_device *dev, char __user *useraddr) return dev->ethtool_ops->set_ufo(dev, edata.data); } +static int ethtool_get_gso(struct net_device *dev, char __user *useraddr) +{ + struct ethtool_value edata = { ETHTOOL_GGSO }; + + edata.data = dev->features & NETIF_F_GSO; + if (copy_to_user(useraddr, &edata, sizeof(edata))) + return -EFAULT; + return 0; +} + +static int ethtool_set_gso(struct net_device *dev, char __user *useraddr) +{ + struct ethtool_value edata; + + if (copy_from_user(&edata, useraddr, sizeof(edata))) + return -EFAULT; + if (edata.data) + dev->features |= NETIF_F_GSO; + else + dev->features &= ~NETIF_F_GSO; + return 0; +} + static int ethtool_self_test(struct net_device *dev, char __user *useraddr) { struct ethtool_test test; @@ -907,6 +928,12 @@ int dev_ethtool(struct ifreq *ifr) case ETHTOOL_SUFO: rc = ethtool_set_ufo(dev, useraddr); break; + case ETHTOOL_GGSO: + rc = ethtool_get_gso(dev, useraddr); + break; + case ETHTOOL_SGSO: + rc = ethtool_set_gso(dev, useraddr); + break; default: rc = -EOPNOTSUPP; } diff --git a/net/core/netpoll.c b/net/core/netpoll.c index e8e05cebd..9cb781830 100644 --- a/net/core/netpoll.c +++ b/net/core/netpoll.c @@ -273,24 +273,21 @@ static void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb) do { npinfo->tries--; - spin_lock(&np->dev->xmit_lock); - np->dev->xmit_lock_owner = smp_processor_id(); + netif_tx_lock(np->dev); /* * network drivers do not expect to be called if the queue is * stopped. */ if (netif_queue_stopped(np->dev)) { - np->dev->xmit_lock_owner = -1; - spin_unlock(&np->dev->xmit_lock); + netif_tx_unlock(np->dev); netpoll_poll(np); udelay(50); continue; } status = np->dev->hard_start_xmit(skb, np->dev); - np->dev->xmit_lock_owner = -1; - spin_unlock(&np->dev->xmit_lock); + netif_tx_unlock(np->dev); /* success */ if(!status) { diff --git a/net/core/pktgen.c b/net/core/pktgen.c index c23e9c06e..4e7437a56 100644 --- a/net/core/pktgen.c +++ b/net/core/pktgen.c @@ -2149,6 +2149,8 @@ static struct sk_buff *fill_packet_ipv4(struct net_device *odev, skb->mac.raw = ((u8 *) iph) - 14 - pkt_dev->nr_labels*sizeof(u32); skb->dev = odev; skb->pkt_type = PACKET_HOST; + skb->nh.iph = iph; + skb->h.uh = udph; if (pkt_dev->nfrags <= 0) pgh = (struct pktgen_hdr *)skb_put(skb, datalen); @@ -2460,6 +2462,8 @@ static struct sk_buff *fill_packet_ipv6(struct net_device *odev, skb->protocol = protocol; skb->dev = odev; skb->pkt_type = PACKET_HOST; + skb->nh.ipv6h = iph; + skb->h.uh = udph; if (pkt_dev->nfrags <= 0) pgh = (struct pktgen_hdr *)skb_put(skb, datalen); @@ -2896,8 +2900,8 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev) pkt_dev->clone_count = 0; /* reset counter */ } } - - spin_lock_bh(&odev->xmit_lock); + + netif_tx_lock_bh(odev); if (!netif_queue_stopped(odev)) { atomic_inc(&(pkt_dev->skb->users)); @@ -2942,8 +2946,8 @@ static __inline__ void pktgen_xmit(struct pktgen_dev *pkt_dev) pkt_dev->next_tx_ns = 0; } - spin_unlock_bh(&odev->xmit_lock); - + netif_tx_unlock_bh(odev); + /* If pkt_dev->count is zero, then run forever */ if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) { if (atomic_read(&(pkt_dev->skb->users)) != 1) { diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c index cfd2d1ef8..375dd9061 100644 --- a/net/core/rtnetlink.c +++ b/net/core/rtnetlink.c @@ -398,6 +398,9 @@ static int do_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) } if (ida[IFLA_ADDRESS - 1]) { + struct sockaddr *sa; + int len; + if (!dev->set_mac_address) { err = -EOPNOTSUPP; goto out; @@ -409,7 +412,17 @@ static int do_setlink(struct sk_buff *skb, struct nlmsghdr *nlh, void *arg) if (ida[IFLA_ADDRESS - 1]->rta_len != RTA_LENGTH(dev->addr_len)) goto out; - err = dev->set_mac_address(dev, RTA_DATA(ida[IFLA_ADDRESS - 1])); + len = sizeof(sa_family_t) + dev->addr_len; + sa = kmalloc(len, GFP_KERNEL); + if (!sa) { + err = -ENOMEM; + goto out; + } + sa->sa_family = dev->type; + memcpy(sa->sa_data, RTA_DATA(ida[IFLA_ADDRESS - 1]), + dev->addr_len); + err = dev->set_mac_address(dev, sa); + kfree(sa); if (err) goto out; send_addr_notify = 1; diff --git a/net/core/skbuff.c b/net/core/skbuff.c index a304412a7..e5b26c279 100644 --- a/net/core/skbuff.c +++ b/net/core/skbuff.c @@ -140,6 +140,7 @@ EXPORT_SYMBOL(skb_truesize_bug); * Buffers may only be allocated from interrupts using a @gfp_mask of * %GFP_ATOMIC. */ +#ifndef CONFIG_HAVE_ARCH_ALLOC_SKB struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, int fclone) { @@ -172,9 +173,9 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask, shinfo = skb_shinfo(skb); atomic_set(&shinfo->dataref, 1); shinfo->nr_frags = 0; - shinfo->tso_size = 0; - shinfo->tso_segs = 0; - shinfo->ufo_size = 0; + shinfo->gso_size = 0; + shinfo->gso_segs = 0; + shinfo->gso_type = 0; shinfo->ip6_frag_id = 0; shinfo->frag_list = NULL; @@ -194,6 +195,7 @@ nodata: skb = NULL; goto out; } +#endif /* !CONFIG_HAVE_ARCH_ALLOC_SKB */ /** * alloc_skb_from_cache - allocate a network buffer @@ -211,14 +213,18 @@ nodata: */ struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, unsigned int size, - gfp_t gfp_mask) + gfp_t gfp_mask, + int fclone) { + kmem_cache_t *cache; + struct skb_shared_info *shinfo; struct sk_buff *skb; u8 *data; + cache = fclone ? skbuff_fclone_cache : skbuff_head_cache; + /* Get the HEAD */ - skb = kmem_cache_alloc(skbuff_head_cache, - gfp_mask & ~__GFP_DMA); + skb = kmem_cache_alloc(cache, gfp_mask & ~__GFP_DMA); if (!skb) goto out; @@ -235,26 +241,39 @@ struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp, skb->data = data; skb->tail = data; skb->end = data + size; + /* make sure we initialize shinfo sequentially */ + shinfo = skb_shinfo(skb); + atomic_set(&shinfo->dataref, 1); + shinfo->nr_frags = 0; + shinfo->gso_size = 0; + shinfo->gso_segs = 0; + shinfo->gso_type = 0; + shinfo->ip6_frag_id = 0; + shinfo->frag_list = NULL; - atomic_set(&(skb_shinfo(skb)->dataref), 1); - skb_shinfo(skb)->nr_frags = 0; - skb_shinfo(skb)->tso_size = 0; - skb_shinfo(skb)->tso_segs = 0; - skb_shinfo(skb)->frag_list = NULL; + if (fclone) { + struct sk_buff *child = skb + 1; + atomic_t *fclone_ref = (atomic_t *) (child + 1); + + skb->fclone = SKB_FCLONE_ORIG; + atomic_set(fclone_ref, 1); + + child->fclone = SKB_FCLONE_UNAVAILABLE; + } out: return skb; nodata: - kmem_cache_free(skbuff_head_cache, skb); + kmem_cache_free(cache, skb); skb = NULL; goto out; } -static void skb_drop_fraglist(struct sk_buff *skb) +static void skb_drop_list(struct sk_buff **listp) { - struct sk_buff *list = skb_shinfo(skb)->frag_list; + struct sk_buff *list = *listp; - skb_shinfo(skb)->frag_list = NULL; + *listp = NULL; do { struct sk_buff *this = list; @@ -263,6 +282,11 @@ static void skb_drop_fraglist(struct sk_buff *skb) } while (list); } +static inline void skb_drop_fraglist(struct sk_buff *skb) +{ + skb_drop_list(&skb_shinfo(skb)->frag_list); +} + static void skb_clone_fraglist(struct sk_buff *skb) { struct sk_buff *list; @@ -434,6 +458,10 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask) C(local_df); n->cloned = 1; n->nohdr = 0; +#ifdef CONFIG_XEN + C(proto_data_valid); + C(proto_csum_blank); +#endif C(pkt_type); C(ip_summed); C(priority); @@ -533,8 +561,9 @@ static void copy_skb_header(struct sk_buff *new, const struct sk_buff *old) new->xid = old->xid; #endif atomic_set(&new->users, 1); - skb_shinfo(new)->tso_size = skb_shinfo(old)->tso_size; - skb_shinfo(new)->tso_segs = skb_shinfo(old)->tso_segs; + skb_shinfo(new)->gso_size = skb_shinfo(old)->gso_size; + skb_shinfo(new)->gso_segs = skb_shinfo(old)->gso_segs; + skb_shinfo(new)->gso_type = skb_shinfo(old)->gso_type; } /** @@ -806,49 +835,86 @@ struct sk_buff *skb_pad(struct sk_buff *skb, int pad) return nskb; } -/* Trims skb to length len. It can change skb pointers, if "realloc" is 1. - * If realloc==0 and trimming is impossible without change of data, - * it is BUG(). +/* Trims skb to length len. It can change skb pointers. */ -int ___pskb_trim(struct sk_buff *skb, unsigned int len, int realloc) +int ___pskb_trim(struct sk_buff *skb, unsigned int len) { + struct sk_buff **fragp; + struct sk_buff *frag; int offset = skb_headlen(skb); int nfrags = skb_shinfo(skb)->nr_frags; int i; + int err; - for (i = 0; i < nfrags; i++) { + if (skb_cloned(skb) && + unlikely((err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC)))) + return err; + + i = 0; + if (offset >= len) + goto drop_pages; + + for (; i < nfrags; i++) { int end = offset + skb_shinfo(skb)->frags[i].size; - if (end > len) { - if (skb_cloned(skb)) { - BUG_ON(!realloc); - if (pskb_expand_head(skb, 0, 0, GFP_ATOMIC)) - return -ENOMEM; - } - if (len <= offset) { - put_page(skb_shinfo(skb)->frags[i].page); - skb_shinfo(skb)->nr_frags--; - } else { - skb_shinfo(skb)->frags[i].size = len - offset; - } + + if (end < len) { + offset = end; + continue; } - offset = end; + + skb_shinfo(skb)->frags[i++].size = len - offset; + +drop_pages: + skb_shinfo(skb)->nr_frags = i; + + for (; i < nfrags; i++) + put_page(skb_shinfo(skb)->frags[i].page); + + if (skb_shinfo(skb)->frag_list) + skb_drop_fraglist(skb); + goto done; } - if (offset < len) { + for (fragp = &skb_shinfo(skb)->frag_list; (frag = *fragp); + fragp = &frag->next) { + int end = offset + frag->len; + + if (skb_shared(frag)) { + struct sk_buff *nfrag; + + nfrag = skb_clone(frag, GFP_ATOMIC); + if (unlikely(!nfrag)) + return -ENOMEM; + + nfrag->next = frag->next; + kfree_skb(frag); + frag = nfrag; + *fragp = frag; + } + + if (end < len) { + offset = end; + continue; + } + + if (end > len && + unlikely((err = pskb_trim(frag, len - offset)))) + return err; + + if (frag->next) + skb_drop_list(&frag->next); + break; + } + +done: + if (len > skb_headlen(skb)) { skb->data_len -= skb->len - len; skb->len = len; } else { - if (len <= skb_headlen(skb)) { - skb->len = len; - skb->data_len = 0; - skb->tail = skb->data + len; - if (skb_shinfo(skb)->frag_list && !skb_cloned(skb)) - skb_drop_fraglist(skb); - } else { - skb->data_len -= skb->len - len; - skb->len = len; - } + skb->len = len; + skb->data_len = 0; + skb->tail = skb->data + len; } return 0; @@ -1832,6 +1898,133 @@ unsigned char *skb_pull_rcsum(struct sk_buff *skb, unsigned int len) EXPORT_SYMBOL_GPL(skb_pull_rcsum); +/** + * skb_segment - Perform protocol segmentation on skb. + * @skb: buffer to segment + * @features: features for the output path (see dev->features) + * + * This function performs segmentation on the given skb. It returns + * the segment at the given position. It returns NULL if there are + * no more segments to generate, or when an error is encountered. + */ +struct sk_buff *skb_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = NULL; + struct sk_buff *tail = NULL; + unsigned int mss = skb_shinfo(skb)->gso_size; + unsigned int doffset = skb->data - skb->mac.raw; + unsigned int offset = doffset; + unsigned int headroom; + unsigned int len; + int sg = features & NETIF_F_SG; + int nfrags = skb_shinfo(skb)->nr_frags; + int err = -ENOMEM; + int i = 0; + int pos; + + __skb_push(skb, doffset); + headroom = skb_headroom(skb); + pos = skb_headlen(skb); + + do { + struct sk_buff *nskb; + skb_frag_t *frag; + int hsize, nsize; + int k; + int size; + + len = skb->len - offset; + if (len > mss) + len = mss; + + hsize = skb_headlen(skb) - offset; + if (hsize < 0) + hsize = 0; + nsize = hsize + doffset; + if (nsize > len + doffset || !sg) + nsize = len + doffset; + + nskb = alloc_skb(nsize + headroom, GFP_ATOMIC); + if (unlikely(!nskb)) + goto err; + + if (segs) + tail->next = nskb; + else + segs = nskb; + tail = nskb; + + nskb->dev = skb->dev; + nskb->priority = skb->priority; + nskb->protocol = skb->protocol; + nskb->dst = dst_clone(skb->dst); + memcpy(nskb->cb, skb->cb, sizeof(skb->cb)); + nskb->pkt_type = skb->pkt_type; + nskb->mac_len = skb->mac_len; + + skb_reserve(nskb, headroom); + nskb->mac.raw = nskb->data; + nskb->nh.raw = nskb->data + skb->mac_len; + nskb->h.raw = nskb->nh.raw + (skb->h.raw - skb->nh.raw); + memcpy(skb_put(nskb, doffset), skb->data, doffset); + + if (!sg) { + nskb->csum = skb_copy_and_csum_bits(skb, offset, + skb_put(nskb, len), + len, 0); + continue; + } + + frag = skb_shinfo(nskb)->frags; + k = 0; + + nskb->ip_summed = CHECKSUM_HW; + nskb->csum = skb->csum; + memcpy(skb_put(nskb, hsize), skb->data + offset, hsize); + + while (pos < offset + len) { + BUG_ON(i >= nfrags); + + *frag = skb_shinfo(skb)->frags[i]; + get_page(frag->page); + size = frag->size; + + if (pos < offset) { + frag->page_offset += offset - pos; + frag->size -= offset - pos; + } + + k++; + + if (pos + size <= offset + len) { + i++; + pos += size; + } else { + frag->size -= pos + size - (offset + len); + break; + } + + frag++; + } + + skb_shinfo(nskb)->nr_frags = k; + nskb->data_len = len - hsize; + nskb->len += nskb->data_len; + nskb->truesize += nskb->data_len; + } while ((offset += len) < skb->len); + + return segs; + +err: + while ((skb = segs)) { + segs = skb->next; + kfree(skb); + } + return ERR_PTR(err); +} + +EXPORT_SYMBOL_GPL(skb_segment); + void __init skb_init(void) { skbuff_head_cache = kmem_cache_create("skbuff_head_cache", diff --git a/net/decnet/dn_nsp_in.c b/net/decnet/dn_nsp_in.c index 547523b41..a2ba9db1c 100644 --- a/net/decnet/dn_nsp_in.c +++ b/net/decnet/dn_nsp_in.c @@ -801,8 +801,7 @@ got_it: * We linearize everything except data segments here. */ if (cb->nsp_flags & ~0x60) { - if (unlikely(skb_is_nonlinear(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) + if (unlikely(skb_linearize(skb))) goto free_out; } diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c index e172cf98d..5abf7057a 100644 --- a/net/decnet/dn_route.c +++ b/net/decnet/dn_route.c @@ -629,8 +629,7 @@ int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type padlen); if (flags & DN_RT_PKT_CNTL) { - if (unlikely(skb_is_nonlinear(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) + if (unlikely(skb_linearize(skb))) goto dump_it; switch(flags & DN_RT_CNTL_MSK) { diff --git a/net/decnet/dn_rules.c b/net/decnet/dn_rules.c index 446faafe2..2440d0504 100644 --- a/net/decnet/dn_rules.c +++ b/net/decnet/dn_rules.c @@ -400,9 +400,10 @@ int dn_fib_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) rcu_read_lock(); hlist_for_each_entry(r, node, &dn_fib_rules, r_hlist) { if (idx < s_idx) - continue; + goto next; if (dn_fib_fill_rule(skb, r, cb, NLM_F_MULTI) < 0) break; +next: idx++; } rcu_read_unlock(); diff --git a/net/ieee80211/Kconfig b/net/ieee80211/Kconfig index dbb08528d..f7e84e9d1 100644 --- a/net/ieee80211/Kconfig +++ b/net/ieee80211/Kconfig @@ -58,6 +58,7 @@ config IEEE80211_CRYPT_TKIP depends on IEEE80211 && NET_RADIO select CRYPTO select CRYPTO_MICHAEL_MIC + select CRC32 ---help--- Include software based cipher suites in support of IEEE 802.11i (aka TGi, WPA, WPA2, WPA-PSK, etc.) for use with TKIP enabled diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c index c8c6a86d3..7351fa14d 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -68,6 +68,7 @@ */ #include +#include #include #include #include @@ -1157,6 +1158,88 @@ int inet_sk_rebuild_header(struct sock *sk) EXPORT_SYMBOL(inet_sk_rebuild_header); +static int inet_gso_send_check(struct sk_buff *skb) +{ + struct iphdr *iph; + struct net_protocol *ops; + int proto; + int ihl; + int err = -EINVAL; + + if (unlikely(!pskb_may_pull(skb, sizeof(*iph)))) + goto out; + + iph = skb->nh.iph; + ihl = iph->ihl * 4; + if (ihl < sizeof(*iph)) + goto out; + + if (unlikely(!pskb_may_pull(skb, ihl))) + goto out; + + skb->h.raw = __skb_pull(skb, ihl); + iph = skb->nh.iph; + proto = iph->protocol & (MAX_INET_PROTOS - 1); + err = -EPROTONOSUPPORT; + + rcu_read_lock(); + ops = rcu_dereference(inet_protos[proto]); + if (likely(ops && ops->gso_send_check)) + err = ops->gso_send_check(skb); + rcu_read_unlock(); + +out: + return err; +} + +static struct sk_buff *inet_gso_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct iphdr *iph; + struct net_protocol *ops; + int proto; + int ihl; + int id; + + if (!pskb_may_pull(skb, sizeof(*iph))) + goto out; + + iph = skb->nh.iph; + ihl = iph->ihl * 4; + if (ihl < sizeof(*iph)) + goto out; + + if (!pskb_may_pull(skb, ihl)) + goto out; + + skb->h.raw = __skb_pull(skb, ihl); + iph = skb->nh.iph; + id = ntohs(iph->id); + proto = iph->protocol & (MAX_INET_PROTOS - 1); + segs = ERR_PTR(-EPROTONOSUPPORT); + + rcu_read_lock(); + ops = rcu_dereference(inet_protos[proto]); + if (ops && ops->gso_segment) + segs = ops->gso_segment(skb, features); + rcu_read_unlock(); + + if (!segs || unlikely(IS_ERR(segs))) + goto out; + + skb = segs; + do { + iph = skb->nh.iph; + iph->id = htons(id++); + iph->tot_len = htons(skb->len - skb->mac_len); + iph->check = 0; + iph->check = ip_fast_csum(skb->nh.raw, iph->ihl); + } while ((skb = skb->next)); + +out: + return segs; +} + #ifdef CONFIG_IP_MULTICAST static struct net_protocol igmp_protocol = { .handler = igmp_rcv, @@ -1166,6 +1249,8 @@ static struct net_protocol igmp_protocol = { static struct net_protocol tcp_protocol = { .handler = tcp_v4_rcv, .err_handler = tcp_v4_err, + .gso_send_check = tcp_v4_gso_send_check, + .gso_segment = tcp_tso_segment, .no_policy = 1, }; @@ -1211,6 +1296,8 @@ static int ipv4_proc_init(void); static struct packet_type ip_packet_type = { .type = __constant_htons(ETH_P_IP), .func = ip_rcv, + .gso_send_check = inet_gso_send_check, + .gso_segment = inet_gso_segment, }; static int __init inet_init(void) diff --git a/net/ipv4/fib_rules.c b/net/ipv4/fib_rules.c index ec566f3e6..a66c96a25 100644 --- a/net/ipv4/fib_rules.c +++ b/net/ipv4/fib_rules.c @@ -458,13 +458,13 @@ int inet_dump_rules(struct sk_buff *skb, struct netlink_callback *cb) rcu_read_lock(); hlist_for_each_entry(r, node, &fib_rules, hlist) { - if (idx < s_idx) - continue; + goto next; if (inet_fill_rule(skb, r, NETLINK_CB(cb->skb).pid, cb->nlh->nlmsg_seq, RTM_NEWRULE, NLM_F_MULTI) < 0) break; +next: idx++; } rcu_read_unlock(); diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index 0f4145bab..1de08c186 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -160,7 +160,7 @@ void free_fib_info(struct fib_info *fi) void fib_release_info(struct fib_info *fi) { - write_lock(&fib_info_lock); + write_lock_bh(&fib_info_lock); if (fi && --fi->fib_treeref == 0) { hlist_del(&fi->fib_hash); if (fi->fib_prefsrc) @@ -173,7 +173,7 @@ void fib_release_info(struct fib_info *fi) fi->fib_dead = 1; fib_info_put(fi); } - write_unlock(&fib_info_lock); + write_unlock_bh(&fib_info_lock); } static __inline__ int nh_comp(const struct fib_info *fi, const struct fib_info *ofi) @@ -599,7 +599,7 @@ static void fib_hash_move(struct hlist_head *new_info_hash, unsigned int old_size = fib_hash_size; unsigned int i, bytes; - write_lock(&fib_info_lock); + write_lock_bh(&fib_info_lock); old_info_hash = fib_info_hash; old_laddrhash = fib_info_laddrhash; fib_hash_size = new_size; @@ -640,7 +640,7 @@ static void fib_hash_move(struct hlist_head *new_info_hash, } fib_info_laddrhash = new_laddrhash; - write_unlock(&fib_info_lock); + write_unlock_bh(&fib_info_lock); bytes = old_size * sizeof(struct hlist_head *); fib_hash_free(old_info_hash, bytes); @@ -822,7 +822,7 @@ link_it: fi->fib_treeref++; atomic_inc(&fi->fib_clntref); - write_lock(&fib_info_lock); + write_lock_bh(&fib_info_lock); hlist_add_head(&fi->fib_hash, &fib_info_hash[fib_info_hashfn(fi)]); if (fi->fib_prefsrc) { @@ -841,7 +841,7 @@ link_it: head = &fib_info_devhash[hash]; hlist_add_head(&nh->nh_hash, head); } endfor_nexthops(fi) - write_unlock(&fib_info_lock); + write_unlock_bh(&fib_info_lock); return fi; err_inval: diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index cff9c3a72..20b17047c 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -210,8 +210,7 @@ static inline int ip_finish_output(struct sk_buff *skb) return dst_output(skb); } #endif - if (skb->len > dst_mtu(skb->dst) && - !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size)) + if (skb->len > dst_mtu(skb->dst) && !skb_is_gso(skb)) return ip_fragment(skb, ip_finish_output2); else return ip_finish_output2(skb); @@ -362,7 +361,7 @@ packet_routed: } ip_select_ident_more(iph, &rt->u.dst, sk, - (skb_shinfo(skb)->tso_segs ?: 1) - 1); + (skb_shinfo(skb)->gso_segs ?: 1) - 1); /* Add an IP checksum. */ ip_send_check(iph); @@ -743,7 +742,8 @@ static inline int ip_ufo_append_data(struct sock *sk, (length - transhdrlen)); if (!err) { /* specify the length of each IP datagram fragment*/ - skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen); + skb_shinfo(skb)->gso_size = mtu - fragheaderlen; + skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; __skb_queue_tail(&sk->sk_write_queue, skb); return 0; @@ -839,7 +839,7 @@ int ip_append_data(struct sock *sk, */ if (transhdrlen && length + fragheaderlen <= mtu && - rt->u.dst.dev->features&(NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM) && + rt->u.dst.dev->features & NETIF_F_ALL_CSUM && !exthdrlen) csummode = CHECKSUM_HW; @@ -946,7 +946,7 @@ alloc_new_skb: skb_prev->csum = csum_sub(skb_prev->csum, skb->csum); data += fraggap; - skb_trim(skb_prev, maxfraglen); + pskb_trim_unique(skb_prev, maxfraglen); } copy = datalen - transhdrlen - fraggap; @@ -1086,14 +1086,16 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, inet->cork.length += size; if ((sk->sk_protocol == IPPROTO_UDP) && - (rt->u.dst.dev->features & NETIF_F_UFO)) - skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen); + (rt->u.dst.dev->features & NETIF_F_UFO)) { + skb_shinfo(skb)->gso_size = mtu - fragheaderlen; + skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; + } while (size > 0) { int i; - if (skb_shinfo(skb)->ufo_size) + if (skb_is_gso(skb)) len = size; else { @@ -1139,7 +1141,7 @@ ssize_t ip_append_page(struct sock *sk, struct page *page, data, fraggap, 0); skb_prev->csum = csum_sub(skb_prev->csum, skb->csum); - skb_trim(skb_prev, maxfraglen); + pskb_trim_unique(skb_prev, maxfraglen); } /* diff --git a/net/ipv4/ipcomp.c b/net/ipv4/ipcomp.c index 95278b22b..098d103ec 100644 --- a/net/ipv4/ipcomp.c +++ b/net/ipv4/ipcomp.c @@ -84,7 +84,7 @@ out: static int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb) { u8 nexthdr; - int err = 0; + int err = -ENOMEM; struct iphdr *iph; union { struct iphdr iph; @@ -92,11 +92,8 @@ static int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb) } tmp_iph; - if ((skb_is_nonlinear(skb) || skb_cloned(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) { - err = -ENOMEM; + if (skb_linearize_cow(skb)) goto out; - } skb->ip_summed = CHECKSUM_NONE; @@ -171,10 +168,8 @@ static int ipcomp_output(struct xfrm_state *x, struct sk_buff *skb) goto out_ok; } - if ((skb_is_nonlinear(skb) || skb_cloned(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) { + if (skb_linearize_cow(skb)) goto out_ok; - } err = ipcomp_compress(x, skb); iph = skb->nh.iph; diff --git a/net/ipv4/netfilter/arp_tables.c b/net/ipv4/netfilter/arp_tables.c index d0d191920..92adfebec 100644 --- a/net/ipv4/netfilter/arp_tables.c +++ b/net/ipv4/netfilter/arp_tables.c @@ -237,7 +237,7 @@ unsigned int arpt_do_table(struct sk_buff **pskb, struct arpt_entry *e, *back; const char *indev, *outdev; void *table_base; - struct xt_table_info *private = table->private; + struct xt_table_info *private; /* ARP header, plus 2 device addresses, plus 2 IP addresses. */ if (!pskb_may_pull((*pskb), (sizeof(struct arphdr) + @@ -249,6 +249,7 @@ unsigned int arpt_do_table(struct sk_buff **pskb, outdev = out ? out->name : nulldevname; read_lock_bh(&table->lock); + private = table->private; table_base = (void *)private->entries[smp_processor_id()]; e = get_entry(table_base, private->hook_entry[hook]); back = get_entry(table_base, private->underflow[hook]); diff --git a/net/ipv4/netfilter/ip_conntrack_helper_h323.c b/net/ipv4/netfilter/ip_conntrack_helper_h323.c index 518f581d3..853a3d512 100644 --- a/net/ipv4/netfilter/ip_conntrack_helper_h323.c +++ b/net/ipv4/netfilter/ip_conntrack_helper_h323.c @@ -1092,7 +1092,7 @@ static struct ip_conntrack_expect *find_expect(struct ip_conntrack *ct, tuple.dst.protonum = IPPROTO_TCP; exp = __ip_conntrack_expect_find(&tuple); - if (exp->master == ct) + if (exp && exp->master == ct) return exp; return NULL; } diff --git a/net/ipv4/netfilter/ip_nat_proto_tcp.c b/net/ipv4/netfilter/ip_nat_proto_tcp.c index a3d14079e..35d1d90c9 100644 --- a/net/ipv4/netfilter/ip_nat_proto_tcp.c +++ b/net/ipv4/netfilter/ip_nat_proto_tcp.c @@ -129,7 +129,12 @@ tcp_manip_pkt(struct sk_buff **pskb, if (hdrsize < sizeof(*hdr)) return 1; - hdr->check = ip_nat_cheat_check(~oldip, newip, +#ifdef CONFIG_XEN + if ((*pskb)->proto_csum_blank) + hdr->check = ip_nat_cheat_check(oldip, ~newip, hdr->check); + else +#endif + hdr->check = ip_nat_cheat_check(~oldip, newip, ip_nat_cheat_check(oldport ^ 0xFFFF, newport, hdr->check)); diff --git a/net/ipv4/netfilter/ip_nat_proto_udp.c b/net/ipv4/netfilter/ip_nat_proto_udp.c index ec6053fdc..e8eaf6a37 100644 --- a/net/ipv4/netfilter/ip_nat_proto_udp.c +++ b/net/ipv4/netfilter/ip_nat_proto_udp.c @@ -113,11 +113,17 @@ udp_manip_pkt(struct sk_buff **pskb, newport = tuple->dst.u.udp.port; portptr = &hdr->dest; } - if (hdr->check) /* 0 is a special case meaning no checksum */ - hdr->check = ip_nat_cheat_check(~oldip, newip, + if (hdr->check) { /* 0 is a special case meaning no checksum */ +#ifdef CONFIG_XEN + if ((*pskb)->proto_csum_blank) + hdr->check = ip_nat_cheat_check(oldip, ~newip, hdr->check); + else +#endif + hdr->check = ip_nat_cheat_check(~oldip, newip, ip_nat_cheat_check(*portptr ^ 0xFFFF, newport, hdr->check)); + } *portptr = newport; return 1; } diff --git a/net/ipv4/netfilter/ip_tables.c b/net/ipv4/netfilter/ip_tables.c index cee3397ec..71871c4f9 100644 --- a/net/ipv4/netfilter/ip_tables.c +++ b/net/ipv4/netfilter/ip_tables.c @@ -231,7 +231,7 @@ ipt_do_table(struct sk_buff **pskb, const char *indev, *outdev; void *table_base; struct ipt_entry *e, *back; - struct xt_table_info *private = table->private; + struct xt_table_info *private; /* Initialization */ ip = (*pskb)->nh.iph; @@ -248,6 +248,7 @@ ipt_do_table(struct sk_buff **pskb, read_lock_bh(&table->lock); IP_NF_ASSERT(table->valid_hooks & (1 << hook)); + private = table->private; table_base = (void *)private->entries[smp_processor_id()]; e = get_entry(table_base, private->hook_entry[hook]); diff --git a/net/ipv4/netfilter/ipt_ULOG.c b/net/ipv4/netfilter/ipt_ULOG.c index f2b2171d3..2a79b6137 100644 --- a/net/ipv4/netfilter/ipt_ULOG.c +++ b/net/ipv4/netfilter/ipt_ULOG.c @@ -116,6 +116,11 @@ static void ulog_send(unsigned int nlgroupnum) del_timer(&ub->timer); } + if (!ub->skb) { + DEBUGP("ipt_ULOG: ulog_send: nothing to send\n"); + return; + } + /* last nlmsg needs NLMSG_DONE */ if (ub->qlen > 1) ub->lastnlh->nlmsg_type = NLMSG_DONE; diff --git a/net/ipv4/route.c b/net/ipv4/route.c index cc9423de7..5fe2fcf99 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -3144,7 +3144,7 @@ int __init ip_rt_init(void) rhash_entries, (num_physpages >= 128 * 1024) ? 15 : 17, - HASH_HIGHMEM, + 0, &rt_hash_log, &rt_hash_mask, 0); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 6a228319f..e369c0b9d 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -258,6 +258,7 @@ #include #include #include +#include #include #include @@ -572,7 +573,7 @@ new_segment: skb->ip_summed = CHECKSUM_HW; tp->write_seq += copy; TCP_SKB_CB(skb)->end_seq += copy; - skb_shinfo(skb)->tso_segs = 0; + skb_shinfo(skb)->gso_segs = 0; if (!copied) TCP_SKB_CB(skb)->flags &= ~TCPCB_FLAG_PSH; @@ -623,14 +624,10 @@ ssize_t tcp_sendpage(struct socket *sock, struct page *page, int offset, ssize_t res; struct sock *sk = sock->sk; -#define TCP_ZC_CSUM_FLAGS (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM | NETIF_F_HW_CSUM) - if (!(sk->sk_route_caps & NETIF_F_SG) || - !(sk->sk_route_caps & TCP_ZC_CSUM_FLAGS)) + !(sk->sk_route_caps & NETIF_F_ALL_CSUM)) return sock_no_sendpage(sock, page, offset, size, flags); -#undef TCP_ZC_CSUM_FLAGS - lock_sock(sk); TCP_CHECK_TIMER(sk); res = do_tcp_sendpages(sk, &page, offset, size, flags); @@ -727,9 +724,7 @@ new_segment: /* * Check whether we can use HW checksum. */ - if (sk->sk_route_caps & - (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM | - NETIF_F_HW_CSUM)) + if (sk->sk_route_caps & NETIF_F_ALL_CSUM) skb->ip_summed = CHECKSUM_HW; skb_entail(sk, tp, skb); @@ -825,7 +820,7 @@ new_segment: tp->write_seq += copy; TCP_SKB_CB(skb)->end_seq += copy; - skb_shinfo(skb)->tso_segs = 0; + skb_shinfo(skb)->gso_segs = 0; from += copy; copied += copy; @@ -2072,6 +2067,77 @@ int compat_tcp_getsockopt(struct sock *sk, int level, int optname, EXPORT_SYMBOL(compat_tcp_getsockopt); #endif +struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features) +{ + struct sk_buff *segs = ERR_PTR(-EINVAL); + struct tcphdr *th; + unsigned thlen; + unsigned int seq; + unsigned int delta; + unsigned int oldlen; + unsigned int len; + + if (!pskb_may_pull(skb, sizeof(*th))) + goto out; + + th = skb->h.th; + thlen = th->doff * 4; + if (thlen < sizeof(*th)) + goto out; + + if (!pskb_may_pull(skb, thlen)) + goto out; + + oldlen = (u16)~skb->len; + __skb_pull(skb, thlen); + + if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) { + /* Packet is from an untrusted source, reset gso_segs. */ + int mss = skb_shinfo(skb)->gso_size; + + skb_shinfo(skb)->gso_segs = (skb->len + mss - 1) / mss; + + segs = NULL; + goto out; + } + + segs = skb_segment(skb, features); + if (IS_ERR(segs)) + goto out; + + len = skb_shinfo(skb)->gso_size; + delta = htonl(oldlen + (thlen + len)); + + skb = segs; + th = skb->h.th; + seq = ntohl(th->seq); + + do { + th->fin = th->psh = 0; + + th->check = ~csum_fold(th->check + delta); + if (skb->ip_summed != CHECKSUM_HW) + th->check = csum_fold(csum_partial(skb->h.raw, thlen, + skb->csum)); + + seq += len; + skb = skb->next; + th = skb->h.th; + + th->seq = htonl(seq); + th->cwr = 0; + } while (skb->next); + + delta = htonl(oldlen + (skb->tail - skb->h.raw) + skb->data_len); + th->check = ~csum_fold(th->check + delta); + if (skb->ip_summed != CHECKSUM_HW) + th->check = csum_fold(csum_partial(skb->h.raw, thlen, + skb->csum)); + +out: + return segs; +} + extern void __skb_cb_too_small_for_tcp(int, int); extern struct tcp_congestion_ops tcp_reno; diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 662bd247b..0eadd3a27 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -1072,7 +1072,7 @@ tcp_sacktag_write_queue(struct sock *sk, struct sk_buff *ack_skb, u32 prior_snd_ else pkt_len = (end_seq - TCP_SKB_CB(skb)->seq); - if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->tso_size)) + if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->gso_size)) break; pcount = tcp_skb_pcount(skb); } diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 14cffdea6..17f5857c5 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -496,6 +496,24 @@ void tcp_v4_send_check(struct sock *sk, int len, struct sk_buff *skb) } } +int tcp_v4_gso_send_check(struct sk_buff *skb) +{ + struct iphdr *iph; + struct tcphdr *th; + + if (!pskb_may_pull(skb, sizeof(*th))) + return -EINVAL; + + iph = skb->nh.iph; + th = skb->h.th; + + th->check = 0; + th->check = ~tcp_v4_check(th, skb->len, iph->saddr, iph->daddr, 0); + skb->csum = offsetof(struct tcphdr, check); + skb->ip_summed = CHECKSUM_HW; + return 0; +} + /* * This routine will send an RST to the other tcp. * diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 1e848adbc..ba50b52f8 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -197,6 +197,7 @@ void tcp_select_initial_window(int __space, __u32 mss, * See RFC1323 for an explanation of the limit to 14 */ space = max_t(u32, sysctl_tcp_rmem[2], sysctl_rmem_max); + space = min_t(u32, space, *window_clamp); while (space > 65535 && (*rcv_wscale) < 14) { space >>= 1; (*rcv_wscale)++; @@ -511,15 +512,17 @@ static void tcp_set_skb_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned /* Avoid the costly divide in the normal * non-TSO case. */ - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; } else { unsigned int factor; factor = skb->len + (mss_now - 1); factor /= mss_now; - skb_shinfo(skb)->tso_segs = factor; - skb_shinfo(skb)->tso_size = mss_now; + skb_shinfo(skb)->gso_segs = factor; + skb_shinfo(skb)->gso_size = mss_now; + skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4; } } @@ -912,7 +915,7 @@ static int tcp_init_tso_segs(struct sock *sk, struct sk_buff *skb, unsigned int if (!tso_segs || (tso_segs > 1 && - skb_shinfo(skb)->tso_size != mss_now)) { + tcp_skb_mss(skb) != mss_now)) { tcp_set_skb_tso_segs(sk, skb, mss_now); tso_segs = tcp_skb_pcount(skb); } @@ -1723,8 +1726,9 @@ int tcp_retransmit_skb(struct sock *sk, struct sk_buff *skb) tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) { if (!pskb_trim(skb, 0)) { TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1; - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; skb->ip_summed = CHECKSUM_NONE; skb->csum = 0; } @@ -1929,8 +1933,9 @@ void tcp_send_fin(struct sock *sk) skb->csum = 0; TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_FIN); TCP_SKB_CB(skb)->sacked = 0; - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */ TCP_SKB_CB(skb)->seq = tp->write_seq; @@ -1962,8 +1967,9 @@ void tcp_send_active_reset(struct sock *sk, gfp_t priority) skb->csum = 0; TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_RST); TCP_SKB_CB(skb)->sacked = 0; - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; /* Send it off. */ TCP_SKB_CB(skb)->seq = tcp_acceptable_seq(sk, tp); @@ -2046,8 +2052,9 @@ struct sk_buff * tcp_make_synack(struct sock *sk, struct dst_entry *dst, TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn; TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1; TCP_SKB_CB(skb)->sacked = 0; - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; th->seq = htonl(TCP_SKB_CB(skb)->seq); th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1); if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */ @@ -2151,8 +2158,9 @@ int tcp_connect(struct sock *sk) TCP_SKB_CB(buff)->flags = TCPCB_FLAG_SYN; TCP_ECN_send_syn(sk, tp, buff); TCP_SKB_CB(buff)->sacked = 0; - skb_shinfo(buff)->tso_segs = 1; - skb_shinfo(buff)->tso_size = 0; + skb_shinfo(buff)->gso_segs = 1; + skb_shinfo(buff)->gso_size = 0; + skb_shinfo(buff)->gso_type = 0; buff->csum = 0; TCP_SKB_CB(buff)->seq = tp->write_seq++; TCP_SKB_CB(buff)->end_seq = tp->write_seq; @@ -2256,8 +2264,9 @@ void tcp_send_ack(struct sock *sk) buff->csum = 0; TCP_SKB_CB(buff)->flags = TCPCB_FLAG_ACK; TCP_SKB_CB(buff)->sacked = 0; - skb_shinfo(buff)->tso_segs = 1; - skb_shinfo(buff)->tso_size = 0; + skb_shinfo(buff)->gso_segs = 1; + skb_shinfo(buff)->gso_size = 0; + skb_shinfo(buff)->gso_type = 0; /* Send it off, this clears delayed acks for us. */ TCP_SKB_CB(buff)->seq = TCP_SKB_CB(buff)->end_seq = tcp_acceptable_seq(sk, tp); @@ -2292,8 +2301,9 @@ static int tcp_xmit_probe_skb(struct sock *sk, int urgent) skb->csum = 0; TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK; TCP_SKB_CB(skb)->sacked = urgent; - skb_shinfo(skb)->tso_segs = 1; - skb_shinfo(skb)->tso_size = 0; + skb_shinfo(skb)->gso_segs = 1; + skb_shinfo(skb)->gso_size = 0; + skb_shinfo(skb)->gso_type = 0; /* Use a previous sequence. This should cause the other * end to send an ack. Don't queue or clone SKB, just diff --git a/net/ipv4/xfrm4_output.c b/net/ipv4/xfrm4_output.c index af2392ae5..ecc4119df 100644 --- a/net/ipv4/xfrm4_output.c +++ b/net/ipv4/xfrm4_output.c @@ -8,13 +8,19 @@ * 2 of the License, or (at your option) any later version. */ +#include +#include +#include #include #include +#include #include #include #include #include +extern int skb_checksum_setup(struct sk_buff *skb); + /* Add encapsulation header. * * In transport mode, the IP header will be moved forward to make space @@ -33,6 +39,7 @@ static void xfrm4_encap(struct sk_buff *skb) struct dst_entry *dst = skb->dst; struct xfrm_state *x = dst->xfrm; struct iphdr *iph, *top_iph; + int flags; iph = skb->nh.iph; skb->h.ipiph = iph; @@ -51,12 +58,15 @@ static void xfrm4_encap(struct sk_buff *skb) /* DS disclosed */ top_iph->tos = INET_ECN_encapsulate(iph->tos, iph->tos); - if (x->props.flags & XFRM_STATE_NOECN) + + flags = x->props.flags; + if (flags & XFRM_STATE_NOECN) IP_ECN_clear(top_iph); - top_iph->frag_off = iph->frag_off & htons(IP_DF); + top_iph->frag_off = (flags & XFRM_STATE_NOPMTUDISC) ? + 0 : (iph->frag_off & htons(IP_DF)); if (!top_iph->frag_off) - __ip_select_ident(top_iph, dst, 0); + __ip_select_ident(top_iph, dst->child, 0); top_iph->ttl = dst_metric(dst->child, RTAX_HOPLIMIT); @@ -91,12 +101,16 @@ out: return ret; } -int xfrm4_output(struct sk_buff *skb) +static int xfrm4_output_one(struct sk_buff *skb) { struct dst_entry *dst = skb->dst; struct xfrm_state *x = dst->xfrm; int err; + err = skb_checksum_setup(skb); + if (err) + goto error_nolock; + if (skb->ip_summed == CHECKSUM_HW) { err = skb_checksum_help(skb, 0); if (err) @@ -109,27 +123,33 @@ int xfrm4_output(struct sk_buff *skb) goto error_nolock; } - spin_lock_bh(&x->lock); - err = xfrm_state_check(x, skb); - if (err) - goto error; + do { + spin_lock_bh(&x->lock); + err = xfrm_state_check(x, skb); + if (err) + goto error; - xfrm4_encap(skb); + xfrm4_encap(skb); - err = x->type->output(x, skb); - if (err) - goto error; + err = x->type->output(x, skb); + if (err) + goto error; - x->curlft.bytes += skb->len; - x->curlft.packets++; + x->curlft.bytes += skb->len; + x->curlft.packets++; - spin_unlock_bh(&x->lock); + spin_unlock_bh(&x->lock); - if (!(skb->dst = dst_pop(dst))) { - err = -EHOSTUNREACH; - goto error_nolock; - } - err = NET_XMIT_BYPASS; + if (!(skb->dst = dst_pop(dst))) { + err = -EHOSTUNREACH; + goto error_nolock; + } + dst = skb->dst; + x = dst->xfrm; + } while (x && !x->props.mode); + + IPCB(skb)->flags |= IPSKB_XFRM_TRANSFORMED; + err = 0; out_exit: return err; @@ -139,3 +159,76 @@ error_nolock: kfree_skb(skb); goto out_exit; } + +static int xfrm4_output_finish2(struct sk_buff *skb) +{ + int err; + + while (likely((err = xfrm4_output_one(skb)) == 0)) { + nf_reset(skb); + + err = nf_hook(PF_INET, NF_IP_LOCAL_OUT, &skb, NULL, + skb->dst->dev, dst_output); + if (unlikely(err != 1)) + break; + + if (!skb->dst->xfrm) + return dst_output(skb); + + err = nf_hook(PF_INET, NF_IP_POST_ROUTING, &skb, NULL, + skb->dst->dev, xfrm4_output_finish2); + if (unlikely(err != 1)) + break; + } + + return err; +} + +static int xfrm4_output_finish(struct sk_buff *skb) +{ + struct sk_buff *segs; + +#ifdef CONFIG_NETFILTER + if (!skb->dst->xfrm) { + IPCB(skb)->flags |= IPSKB_REROUTED; + return dst_output(skb); + } +#endif + + if (!skb_is_gso(skb)) + return xfrm4_output_finish2(skb); + + skb->protocol = htons(ETH_P_IP); + segs = skb_gso_segment(skb, 0); + kfree_skb(skb); + if (unlikely(IS_ERR(segs))) + return PTR_ERR(segs); + + do { + struct sk_buff *nskb = segs->next; + int err; + + segs->next = NULL; + err = xfrm4_output_finish2(segs); + + if (unlikely(err)) { + while ((segs = nskb)) { + nskb = segs->next; + segs->next = NULL; + kfree_skb(segs); + } + return err; + } + + segs = nskb; + } while (segs); + + return 0; +} + +int xfrm4_output(struct sk_buff *skb) +{ + return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev, + xfrm4_output_finish, + !(IPCB(skb)->flags & IPSKB_REROUTED)); +} diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index 1f5e0c63a..386ca5c41 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -2470,6 +2470,7 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp, u32 flags) spin_lock_bh(&ifp->lock); if (dev->flags&(IFF_NOARP|IFF_LOOPBACK) || + !(dev->flags&IFF_MULTICAST) || !(ifp->flags&IFA_F_TENTATIVE)) { ifp->flags &= ~IFA_F_TENTATIVE; spin_unlock_bh(&ifp->lock); @@ -2554,6 +2555,7 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp) if (ifp->idev->cnf.forwarding == 0 && ifp->idev->cnf.rtr_solicits > 0 && (dev->flags&IFF_LOOPBACK) == 0 && + (dev->flags & IFF_MULTICAST) && (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)) { struct in6_addr all_routers; diff --git a/net/ipv6/exthdrs.c b/net/ipv6/exthdrs.c index a18d42563..9ca783dcc 100644 --- a/net/ipv6/exthdrs.c +++ b/net/ipv6/exthdrs.c @@ -635,14 +635,17 @@ ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt, struct ipv6_txoptions *opt2; int err; - if (newtype != IPV6_HOPOPTS && opt->hopopt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt)); - if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt)); - if (newtype != IPV6_RTHDR && opt->srcrt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt)); - if (newtype != IPV6_DSTOPTS && opt->dst1opt) - tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt)); + if (opt) { + if (newtype != IPV6_HOPOPTS && opt->hopopt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->hopopt)); + if (newtype != IPV6_RTHDRDSTOPTS && opt->dst0opt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst0opt)); + if (newtype != IPV6_RTHDR && opt->srcrt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->srcrt)); + if (newtype != IPV6_DSTOPTS && opt->dst1opt) + tot_len += CMSG_ALIGN(ipv6_optlen(opt->dst1opt)); + } + if (newopt && newoptlen) tot_len += CMSG_ALIGN(newoptlen); @@ -659,25 +662,25 @@ ipv6_renew_options(struct sock *sk, struct ipv6_txoptions *opt, opt2->tot_len = tot_len; p = (char *)(opt2 + 1); - err = ipv6_renew_option(opt->hopopt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->hopopt : NULL, newopt, newoptlen, newtype != IPV6_HOPOPTS, &opt2->hopopt, &p); if (err) goto out; - err = ipv6_renew_option(opt->dst0opt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->dst0opt : NULL, newopt, newoptlen, newtype != IPV6_RTHDRDSTOPTS, &opt2->dst0opt, &p); if (err) goto out; - err = ipv6_renew_option(opt->srcrt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->srcrt : NULL, newopt, newoptlen, newtype != IPV6_RTHDR, - (struct ipv6_opt_hdr **)opt2->srcrt, &p); + (struct ipv6_opt_hdr **)&opt2->srcrt, &p); if (err) goto out; - err = ipv6_renew_option(opt->dst1opt, newopt, newoptlen, + err = ipv6_renew_option(opt ? opt->dst1opt : NULL, newopt, newoptlen, newtype != IPV6_DSTOPTS, &opt2->dst1opt, &p); if (err) diff --git a/net/ipv6/ip6_output.c b/net/ipv6/ip6_output.c index e46048974..1d5dfeb26 100644 --- a/net/ipv6/ip6_output.c +++ b/net/ipv6/ip6_output.c @@ -147,7 +147,7 @@ static int ip6_output2(struct sk_buff *skb) int ip6_output(struct sk_buff *skb) { - if ((skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->ufo_size) || + if ((skb->len > dst_mtu(skb->dst) && !skb_is_gso(skb)) || dst_allfrag(skb->dst)) return ip6_fragment(skb, ip6_output2); else @@ -830,8 +830,9 @@ static inline int ip6_ufo_append_data(struct sock *sk, struct frag_hdr fhdr; /* specify the length of each IP datagram fragment*/ - skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen) - - sizeof(struct frag_hdr); + skb_shinfo(skb)->gso_size = mtu - fragheaderlen - + sizeof(struct frag_hdr); + skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4; ipv6_select_ident(skb, &fhdr); skb_shinfo(skb)->ip6_frag_id = fhdr.identification; __skb_queue_tail(&sk->sk_write_queue, skb); @@ -1047,7 +1048,7 @@ alloc_new_skb: skb_prev->csum = csum_sub(skb_prev->csum, skb->csum); data += fraggap; - skb_trim(skb_prev, maxfraglen); + pskb_trim_unique(skb_prev, maxfraglen); } copy = datalen - transhdrlen - fraggap; if (copy < 0) { diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 486364360..96cac9a94 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -65,7 +65,7 @@ static LIST_HEAD(ipcomp6_tfms_list); static int ipcomp6_input(struct xfrm_state *x, struct sk_buff *skb) { - int err = 0; + int err = -ENOMEM; u8 nexthdr = 0; int hdr_len = skb->h.raw - skb->nh.raw; unsigned char *tmp_hdr = NULL; @@ -76,11 +76,8 @@ static int ipcomp6_input(struct xfrm_state *x, struct sk_buff *skb) struct crypto_tfm *tfm; int cpu; - if ((skb_is_nonlinear(skb) || skb_cloned(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) { - err = -ENOMEM; + if (skb_linearize_cow(skb)) goto out; - } skb->ip_summed = CHECKSUM_NONE; @@ -159,10 +156,8 @@ static int ipcomp6_output(struct xfrm_state *x, struct sk_buff *skb) goto out_ok; } - if ((skb_is_nonlinear(skb) || skb_cloned(skb)) && - skb_linearize(skb, GFP_ATOMIC) != 0) { + if (skb_linearize_cow(skb)) goto out_ok; - } /* compression */ plen = skb->len - hdr_len; diff --git a/net/ipv6/xfrm6_output.c b/net/ipv6/xfrm6_output.c index 6b9867717..e9ea33892 100644 --- a/net/ipv6/xfrm6_output.c +++ b/net/ipv6/xfrm6_output.c @@ -9,9 +9,11 @@ * 2 of the License, or (at your option) any later version. */ +#include #include #include #include +#include #include #include #include @@ -92,7 +94,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb) return ret; } -int xfrm6_output(struct sk_buff *skb) +static int xfrm6_output_one(struct sk_buff *skb) { struct dst_entry *dst = skb->dst; struct xfrm_state *x = dst->xfrm; @@ -110,29 +112,35 @@ int xfrm6_output(struct sk_buff *skb) goto error_nolock; } - spin_lock_bh(&x->lock); - err = xfrm_state_check(x, skb); - if (err) - goto error; + do { + spin_lock_bh(&x->lock); + err = xfrm_state_check(x, skb); + if (err) + goto error; - xfrm6_encap(skb); + xfrm6_encap(skb); - err = x->type->output(x, skb); - if (err) - goto error; + err = x->type->output(x, skb); + if (err) + goto error; - x->curlft.bytes += skb->len; - x->curlft.packets++; + x->curlft.bytes += skb->len; + x->curlft.packets++; - spin_unlock_bh(&x->lock); + spin_unlock_bh(&x->lock); - skb->nh.raw = skb->data; - - if (!(skb->dst = dst_pop(dst))) { - err = -EHOSTUNREACH; - goto error_nolock; - } - err = NET_XMIT_BYPASS; + skb->nh.raw = skb->data; + + if (!(skb->dst = dst_pop(dst))) { + err = -EHOSTUNREACH; + goto error_nolock; + } + dst = skb->dst; + x = dst->xfrm; + } while (x && !x->props.mode); + + IP6CB(skb)->flags |= IP6SKB_XFRM_TRANSFORMED; + err = 0; out_exit: return err; @@ -142,3 +150,68 @@ error_nolock: kfree_skb(skb); goto out_exit; } + +static int xfrm6_output_finish2(struct sk_buff *skb) +{ + int err; + + while (likely((err = xfrm6_output_one(skb)) == 0)) { + nf_reset(skb); + + err = nf_hook(PF_INET6, NF_IP6_LOCAL_OUT, &skb, NULL, + skb->dst->dev, dst_output); + if (unlikely(err != 1)) + break; + + if (!skb->dst->xfrm) + return dst_output(skb); + + err = nf_hook(PF_INET6, NF_IP6_POST_ROUTING, &skb, NULL, + skb->dst->dev, xfrm6_output_finish2); + if (unlikely(err != 1)) + break; + } + + return err; +} + +static int xfrm6_output_finish(struct sk_buff *skb) +{ + struct sk_buff *segs; + + if (!skb_is_gso(skb)) + return xfrm6_output_finish2(skb); + + skb->protocol = htons(ETH_P_IP); + segs = skb_gso_segment(skb, 0); + kfree_skb(skb); + if (unlikely(IS_ERR(segs))) + return PTR_ERR(segs); + + do { + struct sk_buff *nskb = segs->next; + int err; + + segs->next = NULL; + err = xfrm6_output_finish2(segs); + + if (unlikely(err)) { + while ((segs = nskb)) { + nskb = segs->next; + segs->next = NULL; + kfree_skb(segs); + } + return err; + } + + segs = nskb; + } while (segs); + + return 0; +} + +int xfrm6_output(struct sk_buff *skb) +{ + return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL, skb->dst->dev, + xfrm6_output_finish); +} diff --git a/net/ipx/af_ipx.c b/net/ipx/af_ipx.c index 811d99872..e6a50e826 100644 --- a/net/ipx/af_ipx.c +++ b/net/ipx/af_ipx.c @@ -1647,7 +1647,8 @@ static int ipx_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_ty ipx_pktsize = ntohs(ipx->ipx_pktsize); /* Too small or invalid header? */ - if (ipx_pktsize < sizeof(struct ipxhdr) || ipx_pktsize > skb->len) + if (ipx_pktsize < sizeof(struct ipxhdr) + || !pskb_may_pull(skb, ipx_pktsize)) goto drop; if (ipx->ipx_checksum != IPX_NO_CHECKSUM && diff --git a/net/netfilter/nfnetlink_log.c b/net/netfilter/nfnetlink_log.c index 9cc07e68b..23af261e0 100644 --- a/net/netfilter/nfnetlink_log.c +++ b/net/netfilter/nfnetlink_log.c @@ -366,6 +366,9 @@ __nfulnl_send(struct nfulnl_instance *inst) if (timer_pending(&inst->timer)) del_timer(&inst->timer); + if (!inst->skb) + return 0; + if (inst->qlen > 1) inst->lastnlh->nlmsg_type = NLMSG_DONE; diff --git a/net/sched/act_api.c b/net/sched/act_api.c index 914c85ff8..eb7dc2947 100644 --- a/net/sched/act_api.c +++ b/net/sched/act_api.c @@ -34,7 +34,7 @@ #include #include -#if 1 /* control */ +#if 0 /* control */ #define DPRINTK(format, args...) printk(KERN_DEBUG format, ##args) #else #define DPRINTK(format, args...) @@ -165,7 +165,7 @@ int tcf_action_exec(struct sk_buff *skb, struct tc_action *act, while ((a = act) != NULL) { repeat: if (a->ops && a->ops->act) { - ret = a->ops->act(&skb, a); + ret = a->ops->act(skb, a, res); if (TC_MUNGED & skb->tc_verd) { /* copied already, allow trampling */ skb->tc_verd = SET_TC_OK2MUNGE(skb->tc_verd); @@ -179,11 +179,6 @@ repeat: act = a->next; } exec_done: - if (skb->tc_classid > 0) { - res->classid = skb->tc_classid; - res->class = 0; - skb->tc_classid = 0; - } return ret; } @@ -256,15 +251,17 @@ tcf_action_dump(struct sk_buff *skb, struct tc_action *act, int bind, int ref) RTA_PUT(skb, a->order, 0, NULL); err = tcf_action_dump_1(skb, a, bind, ref); if (err < 0) - goto rtattr_failure; + goto errout; r->rta_len = skb->tail - (u8*)r; } return 0; rtattr_failure: + err = -EINVAL; +errout: skb_trim(skb, b - skb->data); - return -err; + return err; } struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est, @@ -295,7 +292,7 @@ struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est, if (a_o == NULL) { #ifdef CONFIG_KMOD rtnl_unlock(); - request_module(act_name); + request_module("act_%s", act_name); rtnl_lock(); a_o = tc_lookup_action_n(act_name); @@ -311,6 +308,7 @@ struct tc_action *tcf_action_init_1(struct rtattr *rta, struct rtattr *est, goto err_mod; } #endif + *err = -ENOENT; goto err_out; } @@ -428,17 +426,19 @@ errout: static int tca_get_fill(struct sk_buff *skb, struct tc_action *a, u32 pid, u32 seq, - unsigned flags, int event, int bind, int ref) + u16 flags, int event, int bind, int ref) { struct tcamsg *t; struct nlmsghdr *nlh; unsigned char *b = skb->tail; struct rtattr *x; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*t)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); + t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; x = (struct rtattr*) skb->tail; RTA_PUT(skb, TCA_ACT_TAB, 0, NULL); @@ -580,6 +580,8 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid) nlh = NLMSG_PUT(skb, pid, n->nlmsg_seq, RTM_DELACTION, sizeof(*t)); t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; x = (struct rtattr *) skb->tail; RTA_PUT(skb, TCA_ACT_TAB, 0, NULL); @@ -594,7 +596,7 @@ static int tca_action_flush(struct rtattr *rta, struct nlmsghdr *n, u32 pid) nlh->nlmsg_flags |= NLM_F_ROOT; module_put(a->ops->owner); kfree(a); - err = rtnetlink_send(skb, pid, RTMGRP_TC, n->nlmsg_flags&NLM_F_ECHO); + err = rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO); if (err > 0) return 0; @@ -657,7 +659,7 @@ tca_action_gd(struct rtattr *rta, struct nlmsghdr *n, u32 pid, int event) /* now do the delete */ tcf_action_destroy(head, 0); - ret = rtnetlink_send(skb, pid, RTMGRP_TC, + ret = rtnetlink_send(skb, pid, RTNLGRP_TC, n->nlmsg_flags&NLM_F_ECHO); if (ret > 0) return 0; @@ -669,7 +671,7 @@ err: } static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event, - unsigned flags) + u16 flags) { struct tcamsg *t; struct nlmsghdr *nlh; @@ -684,11 +686,12 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event, b = (unsigned char *)skb->tail; - nlh = NLMSG_PUT(skb, pid, seq, event, sizeof(*t)); - nlh->nlmsg_flags = flags; + nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*t), flags); t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; - + t->tca__pad1 = 0; + t->tca__pad2 = 0; + x = (struct rtattr*) skb->tail; RTA_PUT(skb, TCA_ACT_TAB, 0, NULL); @@ -698,16 +701,16 @@ static int tcf_add_notify(struct tc_action *a, u32 pid, u32 seq, int event, x->rta_len = skb->tail - (u8*)x; nlh->nlmsg_len = skb->tail - b; - NETLINK_CB(skb).dst_groups = RTMGRP_TC; + NETLINK_CB(skb).dst_group = RTNLGRP_TC; - err = rtnetlink_send(skb, pid, RTMGRP_TC, flags&NLM_F_ECHO); + err = rtnetlink_send(skb, pid, RTNLGRP_TC, flags&NLM_F_ECHO); if (err > 0) err = 0; return err; rtattr_failure: nlmsg_failure: - skb_trim(skb, b - skb->data); + kfree_skb(skb); return -1; } @@ -777,7 +780,7 @@ replay: return ret; } -static char * +static struct rtattr * find_dump_kind(struct nlmsghdr *n) { struct rtattr *tb1, *tb2[TCA_ACT_MAX+1]; @@ -805,7 +808,7 @@ find_dump_kind(struct nlmsghdr *n) return NULL; kind = tb2[TCA_ACT_KIND-1]; - return (char *) RTA_DATA(kind); + return kind; } static int @@ -818,16 +821,15 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) struct tc_action a; int ret = 0; struct tcamsg *t = (struct tcamsg *) NLMSG_DATA(cb->nlh); - char *kind = find_dump_kind(cb->nlh); + struct rtattr *kind = find_dump_kind(cb->nlh); if (kind == NULL) { printk("tc_dump_action: action bad kind\n"); return 0; } - a_o = tc_lookup_action_n(kind); + a_o = tc_lookup_action(kind); if (a_o == NULL) { - printk("failed to find %s\n", kind); return 0; } @@ -835,7 +837,7 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) a.ops = a_o; if (a_o->walk == NULL) { - printk("tc_dump_action: %s !capable of dumping table\n", kind); + printk("tc_dump_action: %s !capable of dumping table\n", a_o->kind); goto rtattr_failure; } @@ -843,6 +845,8 @@ tc_dump_action(struct sk_buff *skb, struct netlink_callback *cb) cb->nlh->nlmsg_type, sizeof(*t)); t = NLMSG_DATA(nlh); t->tca_family = AF_UNSPEC; + t->tca__pad1 = 0; + t->tca__pad2 = 0; x = (struct rtattr *) skb->tail; RTA_PUT(skb, TCA_ACT_TAB, 0, NULL); diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c index 138ea92ed..74d4a1dce 100644 --- a/net/sched/sch_generic.c +++ b/net/sched/sch_generic.c @@ -72,9 +72,9 @@ void qdisc_unlock_tree(struct net_device *dev) dev->queue_lock serializes queue accesses for this device AND dev->qdisc pointer itself. - dev->xmit_lock serializes accesses to device driver. + netif_tx_lock serializes accesses to device driver. - dev->queue_lock and dev->xmit_lock are mutually exclusive, + dev->queue_lock and netif_tx_lock are mutually exclusive, if one is grabbed, another must be free. */ @@ -90,14 +90,17 @@ void qdisc_unlock_tree(struct net_device *dev) NOTE: Called under dev->queue_lock with locally disabled BH. */ -int qdisc_restart(struct net_device *dev) +static inline int qdisc_restart(struct net_device *dev) { struct Qdisc *q = dev->qdisc; struct sk_buff *skb; /* Dequeue packet */ - if ((skb = q->dequeue(q)) != NULL) { + if (((skb = dev->gso_skb)) || ((skb = q->dequeue(q)))) { unsigned nolock = (dev->features & NETIF_F_LLTX); + + dev->gso_skb = NULL; + /* * When the driver has LLTX set it does its own locking * in start_xmit. No need to add additional overhead by @@ -108,7 +111,7 @@ int qdisc_restart(struct net_device *dev) * will be requeued. */ if (!nolock) { - if (!spin_trylock(&dev->xmit_lock)) { + if (!netif_tx_trylock(dev)) { collision: /* So, someone grabbed the driver. */ @@ -126,8 +129,6 @@ int qdisc_restart(struct net_device *dev) __get_cpu_var(netdev_rx_stat).cpu_collision++; goto requeue; } - /* Remember that the driver is grabbed by us. */ - dev->xmit_lock_owner = smp_processor_id(); } { @@ -136,14 +137,11 @@ int qdisc_restart(struct net_device *dev) if (!netif_queue_stopped(dev)) { int ret; - if (netdev_nit) - dev_queue_xmit_nit(skb, dev); - ret = dev->hard_start_xmit(skb, dev); + ret = dev_hard_start_xmit(skb, dev); if (ret == NETDEV_TX_OK) { if (!nolock) { - dev->xmit_lock_owner = -1; - spin_unlock(&dev->xmit_lock); + netif_tx_unlock(dev); } spin_lock(&dev->queue_lock); return -1; @@ -157,8 +155,7 @@ int qdisc_restart(struct net_device *dev) /* NETDEV_TX_BUSY - we need to requeue */ /* Release the driver */ if (!nolock) { - dev->xmit_lock_owner = -1; - spin_unlock(&dev->xmit_lock); + netif_tx_unlock(dev); } spin_lock(&dev->queue_lock); q = dev->qdisc; @@ -175,7 +172,10 @@ int qdisc_restart(struct net_device *dev) */ requeue: - q->ops->requeue(skb, q); + if (skb->next) + dev->gso_skb = skb; + else + q->ops->requeue(skb, q); netif_schedule(dev); return 1; } @@ -183,11 +183,23 @@ requeue: return q->q.qlen; } +void __qdisc_run(struct net_device *dev) +{ + if (unlikely(dev->qdisc == &noop_qdisc)) + goto out; + + while (qdisc_restart(dev) < 0 && !netif_queue_stopped(dev)) + /* NOTHING */; + +out: + clear_bit(__LINK_STATE_QDISC_RUNNING, &dev->state); +} + static void dev_watchdog(unsigned long arg) { struct net_device *dev = (struct net_device *)arg; - spin_lock(&dev->xmit_lock); + netif_tx_lock(dev); if (dev->qdisc != &noop_qdisc) { if (netif_device_present(dev) && netif_running(dev) && @@ -203,7 +215,7 @@ static void dev_watchdog(unsigned long arg) dev_hold(dev); } } - spin_unlock(&dev->xmit_lock); + netif_tx_unlock(dev); dev_put(dev); } @@ -227,17 +239,17 @@ void __netdev_watchdog_up(struct net_device *dev) static void dev_watchdog_up(struct net_device *dev) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); __netdev_watchdog_up(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } static void dev_watchdog_down(struct net_device *dev) { - spin_lock_bh(&dev->xmit_lock); + netif_tx_lock_bh(dev); if (del_timer(&dev->watchdog_timer)) dev_put(dev); - spin_unlock_bh(&dev->xmit_lock); + netif_tx_unlock_bh(dev); } void netif_carrier_on(struct net_device *dev) @@ -579,10 +591,17 @@ void dev_deactivate(struct net_device *dev) dev_watchdog_down(dev); - while (test_bit(__LINK_STATE_SCHED, &dev->state)) + /* Wait for outstanding dev_queue_xmit calls. */ + synchronize_rcu(); + + /* Wait for outstanding qdisc_run calls. */ + while (test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state)) yield(); - spin_unlock_wait(&dev->xmit_lock); + if (dev->gso_skb) { + kfree_skb(dev->gso_skb); + dev->gso_skb = NULL; + } } void dev_init_scheduler(struct net_device *dev) @@ -624,6 +643,5 @@ EXPORT_SYMBOL(qdisc_create_dflt); EXPORT_SYMBOL(qdisc_alloc); EXPORT_SYMBOL(qdisc_destroy); EXPORT_SYMBOL(qdisc_reset); -EXPORT_SYMBOL(qdisc_restart); EXPORT_SYMBOL(qdisc_lock_tree); EXPORT_SYMBOL(qdisc_unlock_tree); diff --git a/net/sched/sch_teql.c b/net/sched/sch_teql.c index 79b8ef34c..4c16ad57a 100644 --- a/net/sched/sch_teql.c +++ b/net/sched/sch_teql.c @@ -302,20 +302,17 @@ restart: switch (teql_resolve(skb, skb_res, slave)) { case 0: - if (spin_trylock(&slave->xmit_lock)) { - slave->xmit_lock_owner = smp_processor_id(); + if (netif_tx_trylock(slave)) { if (!netif_queue_stopped(slave) && slave->hard_start_xmit(skb, slave) == 0) { - slave->xmit_lock_owner = -1; - spin_unlock(&slave->xmit_lock); + netif_tx_unlock(slave); master->slaves = NEXT_SLAVE(q); netif_wake_queue(dev); master->stats.tx_packets++; master->stats.tx_bytes += len; return 0; } - slave->xmit_lock_owner = -1; - spin_unlock(&slave->xmit_lock); + netif_tx_unlock(slave); } if (netif_queue_stopped(dev)) busy = 1; diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c index 5e0de3c0e..b9b907023 100644 --- a/net/sctp/sm_make_chunk.c +++ b/net/sctp/sm_make_chunk.c @@ -806,38 +806,26 @@ no_mem: /* Helper to create ABORT with a SCTP_ERROR_USER_ABORT error. */ struct sctp_chunk *sctp_make_abort_user(const struct sctp_association *asoc, - const struct sctp_chunk *chunk, - const struct msghdr *msg) + const struct msghdr *msg, + size_t paylen) { struct sctp_chunk *retval; - void *payload = NULL, *payoff; - size_t paylen = 0; - struct iovec *iov = NULL; - int iovlen = 0; - - if (msg) { - iov = msg->msg_iov; - iovlen = msg->msg_iovlen; - paylen = get_user_iov_size(iov, iovlen); - } + void *payload = NULL; + int err; - retval = sctp_make_abort(asoc, chunk, sizeof(sctp_errhdr_t) + paylen); + retval = sctp_make_abort(asoc, NULL, sizeof(sctp_errhdr_t) + paylen); if (!retval) goto err_chunk; if (paylen) { /* Put the msg_iov together into payload. */ - payload = kmalloc(paylen, GFP_ATOMIC); + payload = kmalloc(paylen, GFP_KERNEL); if (!payload) goto err_payload; - payoff = payload; - for (; iovlen > 0; --iovlen) { - if (copy_from_user(payoff, iov->iov_base,iov->iov_len)) - goto err_copy; - payoff += iov->iov_len; - iov++; - } + err = memcpy_fromiovec(payload, msg->msg_iov, paylen); + if (err < 0) + goto err_copy; } sctp_init_cause(retval, SCTP_ERROR_USER_ABORT, payload, paylen); diff --git a/net/sctp/sm_statefuns.c b/net/sctp/sm_statefuns.c index 9e58144f4..66e9c5b27 100644 --- a/net/sctp/sm_statefuns.c +++ b/net/sctp/sm_statefuns.c @@ -4026,18 +4026,12 @@ sctp_disposition_t sctp_sf_do_9_1_prm_abort( * from its upper layer, but retransmits data to the far end * if necessary to fill gaps. */ - struct msghdr *msg = arg; - struct sctp_chunk *abort; + struct sctp_chunk *abort = arg; sctp_disposition_t retval; retval = SCTP_DISPOSITION_CONSUME; - /* Generate ABORT chunk to send the peer. */ - abort = sctp_make_abort_user(asoc, NULL, msg); - if (!abort) - retval = SCTP_DISPOSITION_NOMEM; - else - sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); /* Even if we can't send the ABORT due to low memory delete the * TCB. This is a departure from our typical NOMEM handling. @@ -4161,8 +4155,7 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_abort( void *arg, sctp_cmd_seq_t *commands) { - struct msghdr *msg = arg; - struct sctp_chunk *abort; + struct sctp_chunk *abort = arg; sctp_disposition_t retval; /* Stop T1-init timer */ @@ -4170,12 +4163,7 @@ sctp_disposition_t sctp_sf_cookie_wait_prm_abort( SCTP_TO(SCTP_EVENT_TIMEOUT_T1_INIT)); retval = SCTP_DISPOSITION_CONSUME; - /* Generate ABORT chunk to send the peer */ - abort = sctp_make_abort_user(asoc, NULL, msg); - if (!abort) - retval = SCTP_DISPOSITION_NOMEM; - else - sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); + sctp_add_cmd_sf(commands, SCTP_CMD_REPLY, SCTP_CHUNK(abort)); sctp_add_cmd_sf(commands, SCTP_CMD_NEW_STATE, SCTP_STATE(SCTP_STATE_CLOSED)); diff --git a/net/sctp/socket.c b/net/sctp/socket.c index b811691c3..5b1c8375c 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -1246,9 +1246,13 @@ SCTP_STATIC void sctp_close(struct sock *sk, long timeout) } } - if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) - sctp_primitive_ABORT(asoc, NULL); - else + if (sock_flag(sk, SOCK_LINGER) && !sk->sk_lingertime) { + struct sctp_chunk *chunk; + + chunk = sctp_make_abort_user(asoc, NULL, 0); + if (chunk) + sctp_primitive_ABORT(asoc, chunk); + } else sctp_primitive_SHUTDOWN(asoc, NULL); } @@ -1477,8 +1481,16 @@ SCTP_STATIC int sctp_sendmsg(struct kiocb *iocb, struct sock *sk, goto out_unlock; } if (sinfo_flags & SCTP_ABORT) { + struct sctp_chunk *chunk; + + chunk = sctp_make_abort_user(asoc, msg, msg_len); + if (!chunk) { + err = -ENOMEM; + goto out_unlock; + } + SCTP_DEBUG_PRINTK("Aborting association: %p\n", asoc); - sctp_primitive_ABORT(asoc, msg); + sctp_primitive_ABORT(asoc, chunk); err = 0; goto out_unlock; } diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c index 7026b0866..00cb388ec 100644 --- a/net/sunrpc/cache.c +++ b/net/sunrpc/cache.c @@ -71,7 +71,12 @@ struct cache_head *sunrpc_cache_lookup(struct cache_detail *detail, new = detail->alloc(); if (!new) return NULL; + /* must fully initialise 'new', else + * we might get lose if we need to + * cache_put it soon. + */ cache_init(new); + detail->init(new, key); write_lock(&detail->hash_lock); @@ -85,7 +90,6 @@ struct cache_head *sunrpc_cache_lookup(struct cache_detail *detail, return tmp; } } - detail->init(new, key); new->next = *head; *head = new; detail->entries++; diff --git a/sound/core/oss/pcm_oss.c b/sound/core/oss/pcm_oss.c index 7f393a024..0691aca51 100644 --- a/sound/core/oss/pcm_oss.c +++ b/sound/core/oss/pcm_oss.c @@ -1745,6 +1745,8 @@ static int snd_pcm_oss_open_file(struct file *file, for (idx = 0; idx < 2; idx++) { if (setup[idx].disable) continue; + if (! pcm->streams[idx].substream_count) + continue; /* no matching substream */ if (idx == SNDRV_PCM_STREAM_PLAYBACK) { if (! (f_mode & FMODE_WRITE)) continue; diff --git a/sound/core/timer.c b/sound/core/timer.c index cdeeb639b..4585600ba 100644 --- a/sound/core/timer.c +++ b/sound/core/timer.c @@ -628,8 +628,9 @@ static void snd_timer_tasklet(unsigned long arg) struct snd_timer_instance *ti; struct list_head *p; unsigned long resolution, ticks; + unsigned long flags; - spin_lock(&timer->lock); + spin_lock_irqsave(&timer->lock, flags); /* now process all callbacks */ while (!list_empty(&timer->sack_list_head)) { p = timer->sack_list_head.next; /* get first item */ @@ -649,7 +650,7 @@ static void snd_timer_tasklet(unsigned long arg) spin_lock(&timer->lock); ti->flags &= ~SNDRV_TIMER_IFLG_CALLBACK; } - spin_unlock(&timer->lock); + spin_unlock_irqrestore(&timer->lock, flags); } /* diff --git a/sound/isa/cs423x/Makefile b/sound/isa/cs423x/Makefile index d2afaea30..2fb4f7409 100644 --- a/sound/isa/cs423x/Makefile +++ b/sound/isa/cs423x/Makefile @@ -11,6 +11,7 @@ snd-cs4236-objs := cs4236.o # Toplevel Module Dependency obj-$(CONFIG_SND_AZT2320) += snd-cs4231-lib.o +obj-$(CONFIG_SND_MIRO) += snd-cs4231-lib.o obj-$(CONFIG_SND_OPL3SA2) += snd-cs4231-lib.o obj-$(CONFIG_SND_CS4231) += snd-cs4231.o snd-cs4231-lib.o obj-$(CONFIG_SND_CS4232) += snd-cs4232.o snd-cs4231-lib.o diff --git a/sound/pci/Kconfig b/sound/pci/Kconfig index a2081803a..8f34986dd 100644 --- a/sound/pci/Kconfig +++ b/sound/pci/Kconfig @@ -318,17 +318,19 @@ config SND_FM801 To compile this driver as a module, choose M here: the module will be called snd-fm801. -config SND_FM801_TEA575X - tristate "ForteMedia FM801 + TEA5757 tuner" +config SND_FM801_TEA575X_BOOL + bool "ForteMedia FM801 + TEA5757 tuner" depends on SND_FM801 - select VIDEO_DEV help Say Y here to include support for soundcards based on the ForteMedia FM801 chip with a TEA5757 tuner connected to GPIO1-3 pins (Media - Forte SF256-PCS-02). + Forte SF256-PCS-02) into the snd-fm801 driver. - To compile this driver as a module, choose M here: the module - will be called snd-fm801-tea575x. +config SND_FM801_TEA575X + tristate + depends on SND_FM801_TEA575X_BOOL + default SND_FM801 + select VIDEO_DEV config SND_HDA_INTEL tristate "Intel HD Audio" diff --git a/sound/pci/au88x0/au88x0_mpu401.c b/sound/pci/au88x0/au88x0_mpu401.c index 873f486b0..118dcc71e 100644 --- a/sound/pci/au88x0/au88x0_mpu401.c +++ b/sound/pci/au88x0/au88x0_mpu401.c @@ -47,7 +47,7 @@ static int __devinit snd_vortex_midi(vortex_t * vortex) struct snd_rawmidi *rmidi; int temp, mode; struct snd_mpu401 *mpu; - int port; + unsigned long port; #ifdef VORTEX_MPU401_LEGACY /* EnableHardCodedMPU401Port() */ diff --git a/sound/pci/fm801.c b/sound/pci/fm801.c index d72fc28c5..09a2885ca 100644 --- a/sound/pci/fm801.c +++ b/sound/pci/fm801.c @@ -35,7 +35,7 @@ #include -#if (defined(CONFIG_SND_FM801_TEA575X) || defined(CONFIG_SND_FM801_TEA575X_MODULE)) && (defined(CONFIG_VIDEO_DEV) || defined(CONFIG_VIDEO_DEV_MODULE)) +#ifdef CONFIG_SND_FM801_TEA575X_BOOL #include #define TEA575X_RADIO 1 #endif diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 959953ca3..9dd541df6 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -37,35 +37,55 @@ #include #include #include +#include #include #include #include #include #include +#include #include #include #include "hda_codec.h" -static int index[SNDRV_CARDS] = SNDRV_DEFAULT_IDX; -static int enable[SNDRV_CARDS] = SNDRV_DEFAULT_ENABLE_PNP; -static char *id[SNDRV_CARDS] = SNDRV_DEFAULT_STR; -static char *model[SNDRV_CARDS]; +static int index = SNDRV_DEFAULT_IDX1; +static char *id = SNDRV_DEFAULT_STR1; +static char *model; +static int position_fix; +static int probe_mask = -1; +static int single_cmd; -module_param_array(index, int, NULL, 0444); +module_param(index, int, 0444); MODULE_PARM_DESC(index, "Index value for Intel HD audio interface."); -module_param_array(id, charp, NULL, 0444); +module_param(id, charp, 0444); MODULE_PARM_DESC(id, "ID string for Intel HD audio interface."); -module_param_array(enable, bool, NULL, 0444); -MODULE_PARM_DESC(enable, "Enable Intel HD audio interface."); -module_param_array(model, charp, NULL, 0444); +module_param(model, charp, 0444); MODULE_PARM_DESC(model, "Use the given board model."); +module_param(position_fix, int, 0444); +MODULE_PARM_DESC(position_fix, "Fix DMA pointer (0 = auto, 1 = none, 2 = POSBUF, 3 = FIFO size)."); +module_param(probe_mask, int, 0444); +MODULE_PARM_DESC(probe_mask, "Bitmask to probe codecs (default = -1)."); +module_param(single_cmd, bool, 0444); +MODULE_PARM_DESC(single_cmd, "Use single command to communicate with codecs (for debugging only)."); + + +/* just for backward compatibility */ +static int enable; +module_param(enable, bool, 0444); MODULE_LICENSE("GPL"); MODULE_SUPPORTED_DEVICE("{{Intel, ICH6}," "{Intel, ICH6M}," "{Intel, ICH7}," - "{Intel, ESB2}}"); + "{Intel, ESB2}," + "{Intel, ICH8}," + "{ATI, SB450}," + "{ATI, SB600}," + "{VIA, VT8251}," + "{VIA, VT8237A}," + "{SiS, SIS966}," + "{ULI, M5461}}"); MODULE_DESCRIPTION("Intel HDA driver"); #define SFX "hda-intel: " @@ -135,13 +155,30 @@ enum { SDI0, SDI1, SDI2, SDI3, SDO0, SDO1, SDO2, SDO3 }; */ /* max number of SDs */ -#define MAX_ICH6_DEV 8 +/* ICH, ATI and VIA have 4 playback and 4 capture */ +#define ICH6_CAPTURE_INDEX 0 +#define ICH6_NUM_CAPTURE 4 +#define ICH6_PLAYBACK_INDEX 4 +#define ICH6_NUM_PLAYBACK 4 + +/* ULI has 6 playback and 5 capture */ +#define ULI_CAPTURE_INDEX 0 +#define ULI_NUM_CAPTURE 5 +#define ULI_PLAYBACK_INDEX 5 +#define ULI_NUM_PLAYBACK 6 + +/* this number is statically defined for simplicity */ +#define MAX_AZX_DEV 16 + /* max number of fragments - we may use more if allocating more pages for BDL */ -#define AZX_MAX_FRAG (PAGE_SIZE / (MAX_ICH6_DEV * 16)) +#define BDL_SIZE PAGE_ALIGN(8192) +#define AZX_MAX_FRAG (BDL_SIZE / (MAX_AZX_DEV * 16)) /* max buffer size - no h/w limit, you can increase as you like */ #define AZX_MAX_BUF_SIZE (1024*1024*1024) /* max number of PCM devics per card */ -#define AZX_MAX_PCMS 8 +#define AZX_MAX_AUDIO_PCMS 6 +#define AZX_MAX_MODEM_PCMS 2 +#define AZX_MAX_PCMS (AZX_MAX_AUDIO_PCMS + AZX_MAX_MODEM_PCMS) /* RIRB int mask: overrun[2], response[0] */ #define RIRB_INT_RESPONSE 0x01 @@ -150,7 +187,7 @@ enum { SDI0, SDI1, SDI2, SDI3, SDO0, SDO1, SDO2, SDO3 }; /* STATESTS int mask: SD2,SD1,SD0 */ #define STATESTS_INT_MASK 0x07 -#define AZX_MAX_CODECS 3 +#define AZX_MAX_CODECS 4 /* SD_CTL bits */ #define SD_CTL_STREAM_RESET 0x01 /* stream reset bit */ @@ -172,6 +209,9 @@ enum { SDI0, SDI1, SDI2, SDI3, SDO0, SDO1, SDO2, SDO3 }; #define ICH6_INT_CTRL_EN 0x40000000 /* controller interrupt enable bit */ #define ICH6_INT_GLOBAL_EN 0x80000000 /* global interrupt enable bit */ +/* GCTL unsolicited response enable bit */ +#define ICH6_GCTL_UREN (1<<8) + /* GCTL reset bit */ #define ICH6_GCTL_RESET (1<<0) @@ -183,27 +223,26 @@ enum { SDI0, SDI1, SDI2, SDI3, SDO0, SDO1, SDO2, SDO3 }; #define ICH6_MAX_CORB_ENTRIES 256 #define ICH6_MAX_RIRB_ENTRIES 256 +/* position fix mode */ +enum { + POS_FIX_AUTO, + POS_FIX_NONE, + POS_FIX_POSBUF, + POS_FIX_FIFO, +}; -/* - * Use CORB/RIRB for communication from/to codecs. - * This is the way recommended by Intel (see below). - */ -#define USE_CORB_RIRB +/* Defines for ATI HD Audio support in SB450 south bridge */ +#define ATI_SB450_HDAUDIO_MISC_CNTR2_ADDR 0x42 +#define ATI_SB450_HDAUDIO_ENABLE_SNOOP 0x02 -/* - * Define this if use the position buffer instead of reading SD_LPIB - * It's not used as default since SD_LPIB seems to give more accurate position - */ -/* #define USE_POSBUF */ +/* Defines for Nvidia HDA support */ +#define NVIDIA_HDA_TRANSREG_ADDR 0x4e +#define NVIDIA_HDA_ENABLE_COHBITS 0x0f /* */ -typedef struct snd_azx azx_t; -typedef struct snd_azx_rb azx_rb_t; -typedef struct snd_azx_dev azx_dev_t; - -struct snd_azx_dev { +struct azx_dev { u32 *bdl; /* virtual address of the BDL */ dma_addr_t bdl_addr; /* physical address of the BDL */ volatile u32 *posbuf; /* position buffer pointer */ @@ -218,17 +257,19 @@ struct snd_azx_dev { u32 sd_int_sta_mask; /* stream int status mask */ /* pcm support */ - snd_pcm_substream_t *substream; /* assigned substream, set in PCM open */ + struct snd_pcm_substream *substream; /* assigned substream, set in PCM open */ unsigned int format_val; /* format value to be set in the controller and the codec */ unsigned char stream_tag; /* assigned stream */ unsigned char index; /* stream index */ + /* for sanity check of position buffer */ + unsigned int period_intr; unsigned int opened: 1; unsigned int running: 1; }; /* CORB/RIRB */ -struct snd_azx_rb { +struct azx_rb { u32 *buf; /* CORB/RIRB buffer * Each CORB entry is 4byte, RIRB is 8byte */ @@ -239,10 +280,18 @@ struct snd_azx_rb { u32 res; /* last read value */ }; -struct snd_azx { - snd_card_t *card; +struct azx { + struct snd_card *card; struct pci_dev *pci; + /* chip type specific */ + int driver_type; + int playback_streams; + int playback_index_offset; + int capture_streams; + int capture_index_offset; + int num_streams; + /* pci resources */ unsigned long addr; void __iomem *remap_addr; @@ -250,27 +299,51 @@ struct snd_azx { /* locks */ spinlock_t reg_lock; - struct semaphore open_mutex; + struct mutex open_mutex; - /* streams */ - azx_dev_t azx_dev[MAX_ICH6_DEV]; + /* streams (x num_streams) */ + struct azx_dev *azx_dev; /* PCM */ unsigned int pcm_devs; - snd_pcm_t *pcm[AZX_MAX_PCMS]; + struct snd_pcm *pcm[AZX_MAX_PCMS]; /* HD codec */ unsigned short codec_mask; struct hda_bus *bus; /* CORB/RIRB */ - azx_rb_t corb; - azx_rb_t rirb; + struct azx_rb corb; + struct azx_rb rirb; /* BDL, CORB/RIRB and position buffers */ struct snd_dma_buffer bdl; struct snd_dma_buffer rb; struct snd_dma_buffer posbuf; + + /* flags */ + int position_fix; + unsigned int initialized: 1; + unsigned int single_cmd: 1; +}; + +/* driver types */ +enum { + AZX_DRIVER_ICH, + AZX_DRIVER_ATI, + AZX_DRIVER_VIA, + AZX_DRIVER_SIS, + AZX_DRIVER_ULI, + AZX_DRIVER_NVIDIA, +}; + +static char *driver_short_names[] __devinitdata = { + [AZX_DRIVER_ICH] = "HDA Intel", + [AZX_DRIVER_ATI] = "HDA ATI SB", + [AZX_DRIVER_VIA] = "HDA VIA VT82xx", + [AZX_DRIVER_SIS] = "HDA SIS966", + [AZX_DRIVER_ULI] = "HDA ULI M5461", + [AZX_DRIVER_NVIDIA] = "HDA NVidia", }; /* @@ -303,7 +376,7 @@ struct snd_azx { readb((dev)->sd_addr + ICH6_REG_##reg) /* for pcm support */ -#define get_azx_dev(substream) (azx_dev_t*)(substream->runtime->private_data) +#define get_azx_dev(substream) (substream->runtime->private_data) /* Get the upper 32bit of the given dma_addr_t * Compiler should optimize and eliminate the code if dma_addr_t is 32bit @@ -315,11 +388,10 @@ struct snd_azx { * Interface for HD codec */ -#ifdef USE_CORB_RIRB /* * CORB / RIRB interface */ -static int azx_alloc_cmd_io(azx_t *chip) +static int azx_alloc_cmd_io(struct azx *chip) { int err; @@ -333,7 +405,7 @@ static int azx_alloc_cmd_io(azx_t *chip) return 0; } -static void azx_init_cmd_io(azx_t *chip) +static void azx_init_cmd_io(struct azx *chip) { /* CORB set up */ chip->corb.addr = chip->rb.addr; @@ -341,6 +413,8 @@ static void azx_init_cmd_io(azx_t *chip) azx_writel(chip, CORBLBASE, (u32)chip->corb.addr); azx_writel(chip, CORBUBASE, upper_32bit(chip->corb.addr)); + /* set the corb size to 256 entries (ULI requires explicitly) */ + azx_writeb(chip, CORBSIZE, 0x02); /* set the corb write pointer to 0 */ azx_writew(chip, CORBWP, 0); /* reset the corb hw read pointer */ @@ -354,20 +428,18 @@ static void azx_init_cmd_io(azx_t *chip) azx_writel(chip, RIRBLBASE, (u32)chip->rirb.addr); azx_writel(chip, RIRBUBASE, upper_32bit(chip->rirb.addr)); + /* set the rirb size to 256 entries (ULI requires explicitly) */ + azx_writeb(chip, RIRBSIZE, 0x02); /* reset the rirb hw write pointer */ azx_writew(chip, RIRBWP, ICH6_RBRWP_CLR); /* set N=1, get RIRB response interrupt for new entry */ azx_writew(chip, RINTCNT, 1); /* enable rirb dma and response irq */ -#ifdef USE_CORB_RIRB azx_writeb(chip, RIRBCTL, ICH6_RBCTL_DMA_EN | ICH6_RBCTL_IRQ_EN); -#else - azx_writeb(chip, RIRBCTL, ICH6_RBCTL_DMA_EN); -#endif chip->rirb.rp = chip->rirb.cmds = 0; } -static void azx_free_cmd_io(azx_t *chip) +static void azx_free_cmd_io(struct azx *chip) { /* disable ringbuffer DMAs */ azx_writeb(chip, RIRBCTL, 0); @@ -375,10 +447,10 @@ static void azx_free_cmd_io(azx_t *chip) } /* send a command */ -static int azx_send_cmd(struct hda_codec *codec, hda_nid_t nid, int direct, - unsigned int verb, unsigned int para) +static int azx_corb_send_cmd(struct hda_codec *codec, hda_nid_t nid, int direct, + unsigned int verb, unsigned int para) { - azx_t *chip = codec->bus->private_data; + struct azx *chip = codec->bus->private_data; unsigned int wp; u32 val; @@ -405,7 +477,7 @@ static int azx_send_cmd(struct hda_codec *codec, hda_nid_t nid, int direct, #define ICH6_RIRB_EX_UNSOL_EV (1<<4) /* retrieve RIRB entry - called from interrupt handler */ -static void azx_update_rirb(azx_t *chip) +static void azx_update_rirb(struct azx *chip) { unsigned int rp, wp; u32 res, res_ex; @@ -432,16 +504,21 @@ static void azx_update_rirb(azx_t *chip) } /* receive a response */ -static unsigned int azx_get_response(struct hda_codec *codec) +static unsigned int azx_rirb_get_response(struct hda_codec *codec) { - azx_t *chip = codec->bus->private_data; + struct azx *chip = codec->bus->private_data; int timeout = 50; while (chip->rirb.cmds) { if (! --timeout) { - snd_printk(KERN_ERR "azx_get_response timeout\n"); + snd_printk(KERN_ERR + "hda_intel: azx_get_response timeout, " + "switching to single_cmd mode...\n"); chip->rirb.rp = azx_readb(chip, RIRBWP); chip->rirb.cmds = 0; + /* switch to single_cmd mode */ + chip->single_cmd = 1; + azx_free_cmd_io(chip); return -1; } msleep(1); @@ -449,7 +526,6 @@ static unsigned int azx_get_response(struct hda_codec *codec) return chip->rirb.res; /* the last value */ } -#else /* * Use the single immediate command instead of CORB/RIRB for simplicity * @@ -460,15 +536,12 @@ static unsigned int azx_get_response(struct hda_codec *codec) * I left the codes, however, for debugging/testing purposes. */ -#define azx_alloc_cmd_io(chip) 0 -#define azx_init_cmd_io(chip) -#define azx_free_cmd_io(chip) - /* send a command */ -static int azx_send_cmd(struct hda_codec *codec, hda_nid_t nid, int direct, - unsigned int verb, unsigned int para) +static int azx_single_send_cmd(struct hda_codec *codec, hda_nid_t nid, + int direct, unsigned int verb, + unsigned int para) { - azx_t *chip = codec->bus->private_data; + struct azx *chip = codec->bus->private_data; u32 val; int timeout = 50; @@ -494,9 +567,9 @@ static int azx_send_cmd(struct hda_codec *codec, hda_nid_t nid, int direct, } /* receive a response */ -static unsigned int azx_get_response(struct hda_codec *codec) +static unsigned int azx_single_get_response(struct hda_codec *codec) { - azx_t *chip = codec->bus->private_data; + struct azx *chip = codec->bus->private_data; int timeout = 50; while (timeout--) { @@ -509,12 +582,38 @@ static unsigned int azx_get_response(struct hda_codec *codec) return (unsigned int)-1; } -#define azx_update_rirb(chip) +/* + * The below are the main callbacks from hda_codec. + * + * They are just the skeleton to call sub-callbacks according to the + * current setting of chip->single_cmd. + */ + +/* send a command */ +static int azx_send_cmd(struct hda_codec *codec, hda_nid_t nid, + int direct, unsigned int verb, + unsigned int para) +{ + struct azx *chip = codec->bus->private_data; + if (chip->single_cmd) + return azx_single_send_cmd(codec, nid, direct, verb, para); + else + return azx_corb_send_cmd(codec, nid, direct, verb, para); +} + +/* get a response */ +static unsigned int azx_get_response(struct hda_codec *codec) +{ + struct azx *chip = codec->bus->private_data; + if (chip->single_cmd) + return azx_single_get_response(codec); + else + return azx_rirb_get_response(codec); +} -#endif /* USE_CORB_RIRB */ /* reset codec link */ -static int azx_reset(azx_t *chip) +static int azx_reset(struct azx *chip) { int count; @@ -546,6 +645,9 @@ static int azx_reset(azx_t *chip) return -EBUSY; } + /* Accept unsolicited responses */ + azx_writel(chip, GCTL, azx_readl(chip, GCTL) | ICH6_GCTL_UREN); + /* detect codecs */ if (! chip->codec_mask) { chip->codec_mask = azx_readw(chip, STATESTS); @@ -561,7 +663,7 @@ static int azx_reset(azx_t *chip) */ /* enable interrupts */ -static void azx_int_enable(azx_t *chip) +static void azx_int_enable(struct azx *chip) { /* enable controller CIE and GIE */ azx_writel(chip, INTCTL, azx_readl(chip, INTCTL) | @@ -569,13 +671,13 @@ static void azx_int_enable(azx_t *chip) } /* disable interrupts */ -static void azx_int_disable(azx_t *chip) +static void azx_int_disable(struct azx *chip) { int i; /* disable interrupts in stream descriptor */ - for (i = 0; i < MAX_ICH6_DEV; i++) { - azx_dev_t *azx_dev = &chip->azx_dev[i]; + for (i = 0; i < chip->num_streams; i++) { + struct azx_dev *azx_dev = &chip->azx_dev[i]; azx_sd_writeb(azx_dev, SD_CTL, azx_sd_readb(azx_dev, SD_CTL) & ~SD_INT_MASK); } @@ -589,13 +691,13 @@ static void azx_int_disable(azx_t *chip) } /* clear interrupts */ -static void azx_int_clear(azx_t *chip) +static void azx_int_clear(struct azx *chip) { int i; /* clear stream status */ - for (i = 0; i < MAX_ICH6_DEV; i++) { - azx_dev_t *azx_dev = &chip->azx_dev[i]; + for (i = 0; i < chip->num_streams; i++) { + struct azx_dev *azx_dev = &chip->azx_dev[i]; azx_sd_writeb(azx_dev, SD_STS, SD_INT_MASK); } @@ -610,7 +712,7 @@ static void azx_int_clear(azx_t *chip) } /* start a stream */ -static void azx_stream_start(azx_t *chip, azx_dev_t *azx_dev) +static void azx_stream_start(struct azx *chip, struct azx_dev *azx_dev) { /* enable SIE */ azx_writeb(chip, INTCTL, @@ -621,7 +723,7 @@ static void azx_stream_start(azx_t *chip, azx_dev_t *azx_dev) } /* stop a stream */ -static void azx_stream_stop(azx_t *chip, azx_dev_t *azx_dev) +static void azx_stream_stop(struct azx *chip, struct azx_dev *azx_dev) { /* stop DMA */ azx_sd_writeb(azx_dev, SD_CTL, azx_sd_readb(azx_dev, SD_CTL) & @@ -636,16 +738,16 @@ static void azx_stream_stop(azx_t *chip, azx_dev_t *azx_dev) /* * initialize the chip */ -static void azx_init_chip(azx_t *chip) +static void azx_init_chip(struct azx *chip) { - unsigned char tcsel_reg; + unsigned char reg; /* Clear bits 0-2 of PCI register TCSEL (at offset 0x44) * TCSEL == Traffic Class Select Register, which sets PCI express QOS * Ensuring these bits are 0 clears playback static on some HD Audio codecs */ - pci_read_config_byte (chip->pci, ICH6_PCIREG_TCSEL, &tcsel_reg); - pci_write_config_byte(chip->pci, ICH6_PCIREG_TCSEL, tcsel_reg & 0xf8); + pci_read_config_byte (chip->pci, ICH6_PCIREG_TCSEL, ®); + pci_write_config_byte(chip->pci, ICH6_PCIREG_TCSEL, reg & 0xf8); /* reset controller */ azx_reset(chip); @@ -655,13 +757,28 @@ static void azx_init_chip(azx_t *chip) azx_int_enable(chip); /* initialize the codec command I/O */ - azx_init_cmd_io(chip); + if (! chip->single_cmd) + azx_init_cmd_io(chip); -#ifdef USE_POSBUF /* program the position buffer */ azx_writel(chip, DPLBASE, (u32)chip->posbuf.addr); azx_writel(chip, DPUBASE, upper_32bit(chip->posbuf.addr)); -#endif + + switch (chip->driver_type) { + case AZX_DRIVER_ATI: + /* For ATI SB450 azalia HD audio, we need to enable snoop */ + pci_read_config_byte(chip->pci, ATI_SB450_HDAUDIO_MISC_CNTR2_ADDR, + ®); + pci_write_config_byte(chip->pci, ATI_SB450_HDAUDIO_MISC_CNTR2_ADDR, + (reg & 0xf8) | ATI_SB450_HDAUDIO_ENABLE_SNOOP); + break; + case AZX_DRIVER_NVIDIA: + /* For NVIDIA HDA, enable snoop */ + pci_read_config_byte(chip->pci,NVIDIA_HDA_TRANSREG_ADDR, ®); + pci_write_config_byte(chip->pci,NVIDIA_HDA_TRANSREG_ADDR, + (reg & 0xf0) | NVIDIA_HDA_ENABLE_COHBITS); + break; + } } @@ -670,8 +787,8 @@ static void azx_init_chip(azx_t *chip) */ static irqreturn_t azx_interrupt(int irq, void* dev_id, struct pt_regs *regs) { - azx_t *chip = dev_id; - azx_dev_t *azx_dev; + struct azx *chip = dev_id; + struct azx_dev *azx_dev; u32 status; int i; @@ -683,11 +800,12 @@ static irqreturn_t azx_interrupt(int irq, void* dev_id, struct pt_regs *regs) return IRQ_NONE; } - for (i = 0; i < MAX_ICH6_DEV; i++) { + for (i = 0; i < chip->num_streams; i++) { azx_dev = &chip->azx_dev[i]; if (status & azx_dev->sd_int_sta_mask) { azx_sd_writeb(azx_dev, SD_STS, SD_INT_MASK); if (azx_dev->substream && azx_dev->running) { + azx_dev->period_intr++; spin_unlock(&chip->reg_lock); snd_pcm_period_elapsed(azx_dev->substream); spin_lock(&chip->reg_lock); @@ -698,7 +816,7 @@ static irqreturn_t azx_interrupt(int irq, void* dev_id, struct pt_regs *regs) /* clear rirb int */ status = azx_readb(chip, RIRBSTS); if (status & RIRB_INT_MASK) { - if (status & RIRB_INT_RESPONSE) + if (! chip->single_cmd && (status & RIRB_INT_RESPONSE)) azx_update_rirb(chip); azx_writeb(chip, RIRBSTS, RIRB_INT_MASK); } @@ -717,7 +835,7 @@ static irqreturn_t azx_interrupt(int irq, void* dev_id, struct pt_regs *regs) /* * set up BDL entries */ -static void azx_setup_periods(azx_dev_t *azx_dev) +static void azx_setup_periods(struct azx_dev *azx_dev) { u32 *bdl = azx_dev->bdl; dma_addr_t dma_addr = azx_dev->substream->runtime->dma_addr; @@ -746,7 +864,7 @@ static void azx_setup_periods(azx_dev_t *azx_dev) /* * set up the SD for streaming */ -static int azx_setup_controller(azx_t *chip, azx_dev_t *azx_dev) +static int azx_setup_controller(struct azx *chip, struct azx_dev *azx_dev) { unsigned char val; int timeout; @@ -791,11 +909,10 @@ static int azx_setup_controller(azx_t *chip, azx_dev_t *azx_dev) /* upper BDL address */ azx_sd_writel(azx_dev, SD_BDLPU, upper_32bit(azx_dev->bdl_addr)); -#ifdef USE_POSBUF /* enable the position buffer */ if (! (azx_readl(chip, DPLBASE) & ICH6_DPLBASE_ENABLE)) azx_writel(chip, DPLBASE, (u32)chip->posbuf.addr | ICH6_DPLBASE_ENABLE); -#endif + /* set the interrupt enable bits in the descriptor control register */ azx_sd_writel(azx_dev, SD_CTL, azx_sd_readl(azx_dev, SD_CTL) | SD_INT_MASK); @@ -807,7 +924,7 @@ static int azx_setup_controller(azx_t *chip, azx_dev_t *azx_dev) * Codec initialization */ -static int __devinit azx_codec_create(azx_t *chip, const char *model) +static int __devinit azx_codec_create(struct azx *chip, const char *model) { struct hda_bus_template bus_temp; int c, codecs, err; @@ -824,7 +941,7 @@ static int __devinit azx_codec_create(azx_t *chip, const char *model) codecs = 0; for (c = 0; c < AZX_MAX_CODECS; c++) { - if (chip->codec_mask & (1 << c)) { + if ((chip->codec_mask & (1 << c)) & probe_mask) { err = snd_hda_codec_new(chip->bus, c, NULL); if (err < 0) continue; @@ -845,11 +962,17 @@ static int __devinit azx_codec_create(azx_t *chip, const char *model) */ /* assign a stream for the PCM */ -static inline azx_dev_t *azx_assign_device(azx_t *chip, int stream) +static inline struct azx_dev *azx_assign_device(struct azx *chip, int stream) { - int dev, i; - dev = stream == SNDRV_PCM_STREAM_PLAYBACK ? 4 : 0; - for (i = 0; i < 4; i++, dev++) + int dev, i, nums; + if (stream == SNDRV_PCM_STREAM_PLAYBACK) { + dev = chip->playback_index_offset; + nums = chip->playback_streams; + } else { + dev = chip->capture_index_offset; + nums = chip->capture_streams; + } + for (i = 0; i < nums; i++, dev++) if (! chip->azx_dev[dev].opened) { chip->azx_dev[dev].opened = 1; return &chip->azx_dev[dev]; @@ -858,17 +981,17 @@ static inline azx_dev_t *azx_assign_device(azx_t *chip, int stream) } /* release the assigned stream */ -static inline void azx_release_device(azx_dev_t *azx_dev) +static inline void azx_release_device(struct azx_dev *azx_dev) { azx_dev->opened = 0; } -static snd_pcm_hardware_t azx_pcm_hw = { +static struct snd_pcm_hardware azx_pcm_hw = { .info = (SNDRV_PCM_INFO_MMAP | SNDRV_PCM_INFO_INTERLEAVED | SNDRV_PCM_INFO_BLOCK_TRANSFER | SNDRV_PCM_INFO_MMAP_VALID | - SNDRV_PCM_INFO_PAUSE | - SNDRV_PCM_INFO_RESUME), + SNDRV_PCM_INFO_PAUSE /*|*/ + /*SNDRV_PCM_INFO_RESUME*/), .formats = SNDRV_PCM_FMTBIT_S16_LE, .rates = SNDRV_PCM_RATE_48000, .rate_min = 48000, @@ -884,25 +1007,25 @@ static snd_pcm_hardware_t azx_pcm_hw = { }; struct azx_pcm { - azx_t *chip; + struct azx *chip; struct hda_codec *codec; struct hda_pcm_stream *hinfo[2]; }; -static int azx_pcm_open(snd_pcm_substream_t *substream) +static int azx_pcm_open(struct snd_pcm_substream *substream) { struct azx_pcm *apcm = snd_pcm_substream_chip(substream); struct hda_pcm_stream *hinfo = apcm->hinfo[substream->stream]; - azx_t *chip = apcm->chip; - azx_dev_t *azx_dev; - snd_pcm_runtime_t *runtime = substream->runtime; + struct azx *chip = apcm->chip; + struct azx_dev *azx_dev; + struct snd_pcm_runtime *runtime = substream->runtime; unsigned long flags; int err; - down(&chip->open_mutex); + mutex_lock(&chip->open_mutex); azx_dev = azx_assign_device(chip, substream->stream); if (azx_dev == NULL) { - up(&chip->open_mutex); + mutex_unlock(&chip->open_mutex); return -EBUSY; } runtime->hw = azx_pcm_hw; @@ -914,7 +1037,7 @@ static int azx_pcm_open(snd_pcm_substream_t *substream) snd_pcm_hw_constraint_integer(runtime, SNDRV_PCM_HW_PARAM_PERIODS); if ((err = hinfo->ops.open(hinfo, apcm->codec, substream)) < 0) { azx_release_device(azx_dev); - up(&chip->open_mutex); + mutex_unlock(&chip->open_mutex); return err; } spin_lock_irqsave(&chip->reg_lock, flags); @@ -923,38 +1046,38 @@ static int azx_pcm_open(snd_pcm_substream_t *substream) spin_unlock_irqrestore(&chip->reg_lock, flags); runtime->private_data = azx_dev; - up(&chip->open_mutex); + mutex_unlock(&chip->open_mutex); return 0; } -static int azx_pcm_close(snd_pcm_substream_t *substream) +static int azx_pcm_close(struct snd_pcm_substream *substream) { struct azx_pcm *apcm = snd_pcm_substream_chip(substream); struct hda_pcm_stream *hinfo = apcm->hinfo[substream->stream]; - azx_t *chip = apcm->chip; - azx_dev_t *azx_dev = get_azx_dev(substream); + struct azx *chip = apcm->chip; + struct azx_dev *azx_dev = get_azx_dev(substream); unsigned long flags; - down(&chip->open_mutex); + mutex_lock(&chip->open_mutex); spin_lock_irqsave(&chip->reg_lock, flags); azx_dev->substream = NULL; azx_dev->running = 0; spin_unlock_irqrestore(&chip->reg_lock, flags); azx_release_device(azx_dev); hinfo->ops.close(hinfo, apcm->codec, substream); - up(&chip->open_mutex); + mutex_unlock(&chip->open_mutex); return 0; } -static int azx_pcm_hw_params(snd_pcm_substream_t *substream, snd_pcm_hw_params_t *hw_params) +static int azx_pcm_hw_params(struct snd_pcm_substream *substream, struct snd_pcm_hw_params *hw_params) { return snd_pcm_lib_malloc_pages(substream, params_buffer_bytes(hw_params)); } -static int azx_pcm_hw_free(snd_pcm_substream_t *substream) +static int azx_pcm_hw_free(struct snd_pcm_substream *substream) { struct azx_pcm *apcm = snd_pcm_substream_chip(substream); - azx_dev_t *azx_dev = get_azx_dev(substream); + struct azx_dev *azx_dev = get_azx_dev(substream); struct hda_pcm_stream *hinfo = apcm->hinfo[substream->stream]; /* reset BDL address */ @@ -967,13 +1090,13 @@ static int azx_pcm_hw_free(snd_pcm_substream_t *substream) return snd_pcm_lib_free_pages(substream); } -static int azx_pcm_prepare(snd_pcm_substream_t *substream) +static int azx_pcm_prepare(struct snd_pcm_substream *substream) { struct azx_pcm *apcm = snd_pcm_substream_chip(substream); - azx_t *chip = apcm->chip; - azx_dev_t *azx_dev = get_azx_dev(substream); + struct azx *chip = apcm->chip; + struct azx_dev *azx_dev = get_azx_dev(substream); struct hda_pcm_stream *hinfo = apcm->hinfo[substream->stream]; - snd_pcm_runtime_t *runtime = substream->runtime; + struct snd_pcm_runtime *runtime = substream->runtime; azx_dev->bufsize = snd_pcm_lib_buffer_bytes(substream); azx_dev->fragsize = snd_pcm_lib_period_bytes(substream); @@ -1001,11 +1124,11 @@ static int azx_pcm_prepare(snd_pcm_substream_t *substream) azx_dev->format_val, substream); } -static int azx_pcm_trigger(snd_pcm_substream_t *substream, int cmd) +static int azx_pcm_trigger(struct snd_pcm_substream *substream, int cmd) { struct azx_pcm *apcm = snd_pcm_substream_chip(substream); - azx_dev_t *azx_dev = get_azx_dev(substream); - azx_t *chip = apcm->chip; + struct azx_dev *azx_dev = get_azx_dev(substream); + struct azx *chip = apcm->chip; int err = 0; spin_lock(&chip->reg_lock); @@ -1017,6 +1140,7 @@ static int azx_pcm_trigger(snd_pcm_substream_t *substream, int cmd) azx_dev->running = 1; break; case SNDRV_PCM_TRIGGER_PAUSE_PUSH: + case SNDRV_PCM_TRIGGER_SUSPEND: case SNDRV_PCM_TRIGGER_STOP: azx_stream_stop(chip, azx_dev); azx_dev->running = 0; @@ -1026,6 +1150,7 @@ static int azx_pcm_trigger(snd_pcm_substream_t *substream, int cmd) } spin_unlock(&chip->reg_lock); if (cmd == SNDRV_PCM_TRIGGER_PAUSE_PUSH || + cmd == SNDRV_PCM_TRIGGER_SUSPEND || cmd == SNDRV_PCM_TRIGGER_STOP) { int timeout = 5000; while (azx_sd_readb(azx_dev, SD_CTL) & SD_CTL_DMA_START && --timeout) @@ -1034,24 +1159,38 @@ static int azx_pcm_trigger(snd_pcm_substream_t *substream, int cmd) return err; } -static snd_pcm_uframes_t azx_pcm_pointer(snd_pcm_substream_t *substream) +static snd_pcm_uframes_t azx_pcm_pointer(struct snd_pcm_substream *substream) { - azx_dev_t *azx_dev = get_azx_dev(substream); + struct azx_pcm *apcm = snd_pcm_substream_chip(substream); + struct azx *chip = apcm->chip; + struct azx_dev *azx_dev = get_azx_dev(substream); unsigned int pos; -#ifdef USE_POSBUF - /* use the position buffer */ - pos = *azx_dev->posbuf; -#else - /* read LPIB */ - pos = azx_sd_readl(azx_dev, SD_LPIB) + azx_dev->fifo_size; -#endif + if (chip->position_fix == POS_FIX_POSBUF || + chip->position_fix == POS_FIX_AUTO) { + /* use the position buffer */ + pos = *azx_dev->posbuf; + if (chip->position_fix == POS_FIX_AUTO && + azx_dev->period_intr == 1 && ! pos) { + printk(KERN_WARNING + "hda-intel: Invalid position buffer, " + "using LPIB read method instead.\n"); + chip->position_fix = POS_FIX_NONE; + goto read_lpib; + } + } else { + read_lpib: + /* read LPIB */ + pos = azx_sd_readl(azx_dev, SD_LPIB); + if (chip->position_fix == POS_FIX_FIFO) + pos += azx_dev->fifo_size; + } if (pos >= azx_dev->bufsize) pos = 0; return bytes_to_frames(substream->runtime, pos); } -static snd_pcm_ops_t azx_pcm_ops = { +static struct snd_pcm_ops azx_pcm_ops = { .open = azx_pcm_open, .close = azx_pcm_close, .ioctl = snd_pcm_lib_ioctl, @@ -1062,16 +1201,16 @@ static snd_pcm_ops_t azx_pcm_ops = { .pointer = azx_pcm_pointer, }; -static void azx_pcm_free(snd_pcm_t *pcm) +static void azx_pcm_free(struct snd_pcm *pcm) { kfree(pcm->private_data); } -static int __devinit create_codec_pcm(azx_t *chip, struct hda_codec *codec, +static int __devinit create_codec_pcm(struct azx *chip, struct hda_codec *codec, struct hda_pcm *cpcm, int pcm_dev) { int err; - snd_pcm_t *pcm; + struct snd_pcm *pcm; struct azx_pcm *apcm; snd_assert(cpcm->stream[0].substreams || cpcm->stream[1].substreams, return -EINVAL); @@ -1100,11 +1239,12 @@ static int __devinit create_codec_pcm(azx_t *chip, struct hda_codec *codec, snd_dma_pci_data(chip->pci), 1024 * 64, 1024 * 128); chip->pcm[pcm_dev] = pcm; + chip->pcm_devs = pcm_dev + 1; return 0; } -static int __devinit azx_pcm_create(azx_t *chip) +static int __devinit azx_pcm_create(struct azx *chip) { struct list_head *p; struct hda_codec *codec; @@ -1114,17 +1254,39 @@ static int __devinit azx_pcm_create(azx_t *chip) if ((err = snd_hda_build_pcms(chip->bus)) < 0) return err; + /* create audio PCMs */ pcm_dev = 0; list_for_each(p, &chip->bus->codec_list) { codec = list_entry(p, struct hda_codec, list); for (c = 0; c < codec->num_pcms; c++) { + if (codec->pcm_info[c].is_modem) + continue; /* create later */ + if (pcm_dev >= AZX_MAX_AUDIO_PCMS) { + snd_printk(KERN_ERR SFX "Too many audio PCMs\n"); + return -EINVAL; + } + err = create_codec_pcm(chip, codec, &codec->pcm_info[c], pcm_dev); + if (err < 0) + return err; + pcm_dev++; + } + } + + /* create modem PCMs */ + pcm_dev = AZX_MAX_AUDIO_PCMS; + list_for_each(p, &chip->bus->codec_list) { + codec = list_entry(p, struct hda_codec, list); + for (c = 0; c < codec->num_pcms; c++) { + if (! codec->pcm_info[c].is_modem) + continue; /* already created */ if (pcm_dev >= AZX_MAX_PCMS) { - snd_printk(KERN_ERR SFX "Too many PCMs\n"); + snd_printk(KERN_ERR SFX "Too many modem PCMs\n"); return -EINVAL; } err = create_codec_pcm(chip, codec, &codec->pcm_info[c], pcm_dev); if (err < 0) return err; + chip->pcm[pcm_dev]->dev_class = SNDRV_PCM_CLASS_MODEM; pcm_dev++; } } @@ -1134,7 +1296,7 @@ static int __devinit azx_pcm_create(azx_t *chip) /* * mixer creation - all stuff is implemented in hda module */ -static int __devinit azx_mixer_create(azx_t *chip) +static int __devinit azx_mixer_create(struct azx *chip) { return snd_hda_build_controls(chip->bus); } @@ -1143,21 +1305,19 @@ static int __devinit azx_mixer_create(azx_t *chip) /* * initialize SD streams */ -static int __devinit azx_init_stream(azx_t *chip) +static int __devinit azx_init_stream(struct azx *chip) { int i; /* initialize each stream (aka device) * assign the starting bdl address to each stream (device) and initialize */ - for (i = 0; i < MAX_ICH6_DEV; i++) { + for (i = 0; i < chip->num_streams; i++) { unsigned int off = sizeof(u32) * (i * AZX_MAX_FRAG * 4); - azx_dev_t *azx_dev = &chip->azx_dev[i]; + struct azx_dev *azx_dev = &chip->azx_dev[i]; azx_dev->bdl = (u32 *)(chip->bdl.area + off); azx_dev->bdl_addr = chip->bdl.addr + off; -#ifdef USE_POSBUF azx_dev->posbuf = (volatile u32 *)(chip->posbuf.area + i * 8); -#endif /* offset: SDI0=0x80, SDI1=0xa0, ... SDO3=0x160 */ azx_dev->sd_addr = chip->remap_addr + (0x20 * i + 0x80); /* int mask: SDI0=0x01, SDI1=0x02, ... SDO3=0x80 */ @@ -1175,28 +1335,33 @@ static int __devinit azx_init_stream(azx_t *chip) /* * power management */ -static int azx_suspend(snd_card_t *card, pm_message_t state) +static int azx_suspend(struct pci_dev *pci, pm_message_t state) { - azx_t *chip = card->pm_private_data; + struct snd_card *card = pci_get_drvdata(pci); + struct azx *chip = card->private_data; int i; + snd_power_change_state(card, SNDRV_CTL_POWER_D3hot); for (i = 0; i < chip->pcm_devs; i++) - if (chip->pcm[i]) - snd_pcm_suspend_all(chip->pcm[i]); + snd_pcm_suspend_all(chip->pcm[i]); snd_hda_suspend(chip->bus, state); azx_free_cmd_io(chip); - pci_disable_device(chip->pci); + pci_disable_device(pci); + pci_save_state(pci); return 0; } -static int azx_resume(snd_card_t *card) +static int azx_resume(struct pci_dev *pci) { - azx_t *chip = card->pm_private_data; + struct snd_card *card = pci_get_drvdata(pci); + struct azx *chip = card->private_data; - pci_enable_device(chip->pci); - pci_set_master(chip->pci); + pci_restore_state(pci); + pci_enable_device(pci); + pci_set_master(pci); azx_init_chip(chip); snd_hda_resume(chip->bus); + snd_power_change_state(card, SNDRV_CTL_POWER_D0); return 0; } #endif /* CONFIG_PM */ @@ -1205,12 +1370,12 @@ static int azx_resume(snd_card_t *card) /* * destructor */ -static int azx_free(azx_t *chip) +static int azx_free(struct azx *chip) { - if (chip->remap_addr) { + if (chip->initialized) { int i; - for (i = 0; i < MAX_ICH6_DEV; i++) + for (i = 0; i < chip->num_streams; i++) azx_stream_stop(chip, &chip->azx_dev[i]); /* disable interrupts */ @@ -1226,29 +1391,28 @@ static int azx_free(azx_t *chip) /* wait a little for interrupts to finish */ msleep(1); - - iounmap(chip->remap_addr); } if (chip->irq >= 0) free_irq(chip->irq, (void*)chip); + if (chip->remap_addr) + iounmap(chip->remap_addr); if (chip->bdl.area) snd_dma_free_pages(&chip->bdl); if (chip->rb.area) snd_dma_free_pages(&chip->rb); -#ifdef USE_POSBUF if (chip->posbuf.area) snd_dma_free_pages(&chip->posbuf); -#endif pci_release_regions(chip->pci); pci_disable_device(chip->pci); + kfree(chip->azx_dev); kfree(chip); return 0; } -static int azx_dev_free(snd_device_t *device) +static int azx_dev_free(struct snd_device *device) { return azx_free(device->device_data); } @@ -1256,11 +1420,13 @@ static int azx_dev_free(snd_device_t *device) /* * constructor */ -static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **rchip) +static int __devinit azx_create(struct snd_card *card, struct pci_dev *pci, + int driver_type, + struct azx **rchip) { - azx_t *chip; + struct azx *chip; int err = 0; - static snd_device_ops_t ops = { + static struct snd_device_ops ops = { .dev_free = azx_dev_free, }; @@ -1269,7 +1435,7 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r if ((err = pci_enable_device(pci)) < 0) return err; - chip = kcalloc(1, sizeof(*chip), GFP_KERNEL); + chip = kzalloc(sizeof(*chip), GFP_KERNEL); if (NULL == chip) { snd_printk(KERN_ERR SFX "cannot allocate chip\n"); @@ -1278,10 +1444,24 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r } spin_lock_init(&chip->reg_lock); - init_MUTEX(&chip->open_mutex); + mutex_init(&chip->open_mutex); chip->card = card; chip->pci = pci; chip->irq = -1; + chip->driver_type = driver_type; + + chip->position_fix = position_fix; + chip->single_cmd = single_cmd; + +#if BITS_PER_LONG != 64 + /* Fix up base address on ULI M5461 */ + if (chip->driver_type == AZX_DRIVER_ULI) { + u16 tmp3; + pci_read_config_word(pci, 0x40, &tmp3); + pci_write_config_word(pci, 0x40, tmp3 | 0x10); + pci_write_config_dword(pci, PCI_BASE_ADDRESS_1, 0); + } +#endif if ((err = pci_request_regions(pci, "ICH HD audio")) < 0) { kfree(chip); @@ -1308,23 +1488,43 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r pci_set_master(pci); synchronize_irq(chip->irq); + switch (chip->driver_type) { + case AZX_DRIVER_ULI: + chip->playback_streams = ULI_NUM_PLAYBACK; + chip->capture_streams = ULI_NUM_CAPTURE; + chip->playback_index_offset = ULI_PLAYBACK_INDEX; + chip->capture_index_offset = ULI_CAPTURE_INDEX; + break; + default: + chip->playback_streams = ICH6_NUM_PLAYBACK; + chip->capture_streams = ICH6_NUM_CAPTURE; + chip->playback_index_offset = ICH6_PLAYBACK_INDEX; + chip->capture_index_offset = ICH6_CAPTURE_INDEX; + break; + } + chip->num_streams = chip->playback_streams + chip->capture_streams; + chip->azx_dev = kcalloc(chip->num_streams, sizeof(*chip->azx_dev), GFP_KERNEL); + if (! chip->azx_dev) { + snd_printk(KERN_ERR "cannot malloc azx_dev\n"); + goto errout; + } + /* allocate memory for the BDL for each stream */ if ((err = snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(chip->pci), - PAGE_SIZE, &chip->bdl)) < 0) { + BDL_SIZE, &chip->bdl)) < 0) { snd_printk(KERN_ERR SFX "cannot allocate BDL\n"); goto errout; } -#ifdef USE_POSBUF /* allocate memory for the position buffer */ if ((err = snd_dma_alloc_pages(SNDRV_DMA_TYPE_DEV, snd_dma_pci_data(chip->pci), - MAX_ICH6_DEV * 8, &chip->posbuf)) < 0) { + chip->num_streams * 8, &chip->posbuf)) < 0) { snd_printk(KERN_ERR SFX "cannot allocate posbuf\n"); goto errout; } -#endif /* allocate CORB/RIRB */ - if ((err = azx_alloc_cmd_io(chip)) < 0) - goto errout; + if (! chip->single_cmd) + if ((err = azx_alloc_cmd_io(chip)) < 0) + goto errout; /* initialize streams */ azx_init_stream(chip); @@ -1332,6 +1532,8 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r /* initialize chip */ azx_init_chip(chip); + chip->initialized = 1; + /* codec detection */ if (! chip->codec_mask) { snd_printk(KERN_ERR SFX "no codecs found!\n"); @@ -1344,6 +1546,10 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r goto errout; } + strcpy(card->driver, "HDA-Intel"); + strcpy(card->shortname, driver_short_names[chip->driver_type]); + sprintf(card->longname, "%s at 0x%lx irq %i", card->shortname, chip->addr, chip->irq); + *rchip = chip; return 0; @@ -1354,35 +1560,25 @@ static int __devinit azx_create(snd_card_t *card, struct pci_dev *pci, azx_t **r static int __devinit azx_probe(struct pci_dev *pci, const struct pci_device_id *pci_id) { - static int dev; - snd_card_t *card; - azx_t *chip; + struct snd_card *card; + struct azx *chip; int err = 0; - if (dev >= SNDRV_CARDS) - return -ENODEV; - if (! enable[dev]) { - dev++; - return -ENOENT; - } - - card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0); + card = snd_card_new(index, id, THIS_MODULE, 0); if (NULL == card) { snd_printk(KERN_ERR SFX "Error creating card!\n"); return -ENOMEM; } - if ((err = azx_create(card, pci, &chip)) < 0) { + if ((err = azx_create(card, pci, pci_id->driver_data, + &chip)) < 0) { snd_card_free(card); return err; } - - strcpy(card->driver, "HDA-Intel"); - strcpy(card->shortname, "HDA Intel"); - sprintf(card->longname, "%s at 0x%lx irq %i", card->shortname, chip->addr, chip->irq); + card->private_data = chip; /* create codec instances */ - if ((err = azx_codec_create(chip, model[dev])) < 0) { + if ((err = azx_codec_create(chip, model)) < 0) { snd_card_free(card); return err; } @@ -1399,7 +1595,6 @@ static int __devinit azx_probe(struct pci_dev *pci, const struct pci_device_id * return err; } - snd_card_set_pm_callback(card, azx_suspend, azx_resume, chip); snd_card_set_dev(card, &pci->dev); if ((err = snd_card_register(card)) < 0) { @@ -1408,7 +1603,6 @@ static int __devinit azx_probe(struct pci_dev *pci, const struct pci_device_id * } pci_set_drvdata(pci, card); - dev++; return err; } @@ -1420,10 +1614,18 @@ static void __devexit azx_remove(struct pci_dev *pci) } /* PCI IDs */ -static struct pci_device_id azx_ids[] = { - { 0x8086, 0x2668, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH6 */ - { 0x8086, 0x27d8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ICH7 */ - { 0x8086, 0x269a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0 }, /* ESB2 */ +static struct pci_device_id azx_ids[] __devinitdata = { + { 0x8086, 0x2668, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ICH }, /* ICH6 */ + { 0x8086, 0x27d8, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ICH }, /* ICH7 */ + { 0x8086, 0x269a, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ICH }, /* ESB2 */ + { 0x8086, 0x284b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ICH }, /* ICH8 */ + { 0x1002, 0x437b, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATI }, /* ATI SB450 */ + { 0x1002, 0x4383, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ATI }, /* ATI SB600 */ + { 0x1106, 0x3288, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_VIA }, /* VIA VT8251/VT8237A */ + { 0x1039, 0x7502, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_SIS }, /* SIS966 */ + { 0x10b9, 0x5461, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_ULI }, /* ULI M5461 */ + { 0x10de, 0x026c, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_NVIDIA }, /* NVIDIA 026c */ + { 0x10de, 0x0371, PCI_ANY_ID, PCI_ANY_ID, 0, 0, AZX_DRIVER_NVIDIA }, /* NVIDIA 0371 */ { 0, } }; MODULE_DEVICE_TABLE(pci, azx_ids); @@ -1434,12 +1636,15 @@ static struct pci_driver driver = { .id_table = azx_ids, .probe = azx_probe, .remove = __devexit_p(azx_remove), - SND_PCI_PM_CALLBACKS +#ifdef CONFIG_PM + .suspend = azx_suspend, + .resume = azx_resume, +#endif }; static int __init alsa_card_azx_init(void) { - return pci_module_init(&driver); + return pci_register_driver(&driver); } static void __exit alsa_card_azx_exit(void) diff --git a/sound/pci/hda/patch_analog.c b/sound/pci/hda/patch_analog.c index 75d23849f..d7343dc82 100644 --- a/sound/pci/hda/patch_analog.c +++ b/sound/pci/hda/patch_analog.c @@ -1,5 +1,5 @@ /* - * HD audio interface patch for AD1986A + * HD audio interface patch for AD1981HD, AD1983, AD1986A, AD1988 * * Copyright (c) 2005 Takashi Iwai * @@ -23,17 +23,375 @@ #include #include #include +#include + #include #include "hda_codec.h" #include "hda_local.h" -struct ad1986a_spec { - struct semaphore amp_mutex; /* PCM volume/mute control mutex */ - struct hda_multi_out multiout; /* playback */ - unsigned int cur_mux; /* capture source */ - struct hda_pcm pcm_rec[2]; /* PCM information */ +struct ad198x_spec { + struct snd_kcontrol_new *mixers[5]; + int num_mixers; + + const struct hda_verb *init_verbs[5]; /* initialization verbs + * don't forget NULL termination! + */ + unsigned int num_init_verbs; + + /* playback */ + struct hda_multi_out multiout; /* playback set-up + * max_channels, dacs must be set + * dig_out_nid and hp_nid are optional + */ + unsigned int cur_eapd; + unsigned int need_dac_fix; + + /* capture */ + unsigned int num_adc_nids; + hda_nid_t *adc_nids; + hda_nid_t dig_in_nid; /* digital-in NID; optional */ + + /* capture source */ + const struct hda_input_mux *input_mux; + hda_nid_t *capsrc_nids; + unsigned int cur_mux[3]; + + /* channel model */ + const struct hda_channel_mode *channel_mode; + int num_channel_mode; + + /* PCM information */ + struct hda_pcm pcm_rec[2]; /* used in alc_build_pcms() */ + + struct mutex amp_mutex; /* PCM volume/mute control mutex */ + unsigned int spdif_route; + + /* dynamic controls, init_verbs and input_mux */ + struct auto_pin_cfg autocfg; + unsigned int num_kctl_alloc, num_kctl_used; + struct snd_kcontrol_new *kctl_alloc; + struct hda_input_mux private_imux; + hda_nid_t private_dac_nids[4]; +}; + +/* + * input MUX handling (common part) + */ +static int ad198x_mux_enum_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + + return snd_hda_input_mux_info(spec->input_mux, uinfo); +} + +static int ad198x_mux_enum_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + unsigned int adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + + ucontrol->value.enumerated.item[0] = spec->cur_mux[adc_idx]; + return 0; +} + +static int ad198x_mux_enum_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + unsigned int adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); + + return snd_hda_input_mux_put(codec, spec->input_mux, ucontrol, + spec->capsrc_nids[adc_idx], + &spec->cur_mux[adc_idx]); +} + +/* + * initialization (common callbacks) + */ +static int ad198x_init(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + int i; + + for (i = 0; i < spec->num_init_verbs; i++) + snd_hda_sequence_write(codec, spec->init_verbs[i]); + return 0; +} + +static int ad198x_build_controls(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + unsigned int i; + int err; + + for (i = 0; i < spec->num_mixers; i++) { + err = snd_hda_add_new_ctls(codec, spec->mixers[i]); + if (err < 0) + return err; + } + if (spec->multiout.dig_out_nid) { + err = snd_hda_create_spdif_out_ctls(codec, spec->multiout.dig_out_nid); + if (err < 0) + return err; + } + if (spec->dig_in_nid) { + err = snd_hda_create_spdif_in_ctls(codec, spec->dig_in_nid); + if (err < 0) + return err; + } + return 0; +} + +/* + * Analog playback callbacks + */ +static int ad198x_playback_pcm_open(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + return snd_hda_multi_out_analog_open(codec, &spec->multiout, substream); +} + +static int ad198x_playback_pcm_prepare(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + unsigned int stream_tag, + unsigned int format, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + return snd_hda_multi_out_analog_prepare(codec, &spec->multiout, stream_tag, + format, substream); +} + +static int ad198x_playback_pcm_cleanup(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + return snd_hda_multi_out_analog_cleanup(codec, &spec->multiout); +} + +/* + * Digital out + */ +static int ad198x_dig_playback_pcm_open(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + return snd_hda_multi_out_dig_open(codec, &spec->multiout); +} + +static int ad198x_dig_playback_pcm_close(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + return snd_hda_multi_out_dig_close(codec, &spec->multiout); +} + +/* + * Analog capture + */ +static int ad198x_capture_pcm_prepare(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + unsigned int stream_tag, + unsigned int format, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + snd_hda_codec_setup_stream(codec, spec->adc_nids[substream->number], + stream_tag, 0, format); + return 0; +} + +static int ad198x_capture_pcm_cleanup(struct hda_pcm_stream *hinfo, + struct hda_codec *codec, + struct snd_pcm_substream *substream) +{ + struct ad198x_spec *spec = codec->spec; + snd_hda_codec_setup_stream(codec, spec->adc_nids[substream->number], + 0, 0, 0); + return 0; +} + + +/* + */ +static struct hda_pcm_stream ad198x_pcm_analog_playback = { + .substreams = 1, + .channels_min = 2, + .channels_max = 6, /* changed later */ + .nid = 0, /* fill later */ + .ops = { + .open = ad198x_playback_pcm_open, + .prepare = ad198x_playback_pcm_prepare, + .cleanup = ad198x_playback_pcm_cleanup + }, +}; + +static struct hda_pcm_stream ad198x_pcm_analog_capture = { + .substreams = 1, + .channels_min = 2, + .channels_max = 2, + .nid = 0, /* fill later */ + .ops = { + .prepare = ad198x_capture_pcm_prepare, + .cleanup = ad198x_capture_pcm_cleanup + }, +}; + +static struct hda_pcm_stream ad198x_pcm_digital_playback = { + .substreams = 1, + .channels_min = 2, + .channels_max = 2, + .nid = 0, /* fill later */ + .ops = { + .open = ad198x_dig_playback_pcm_open, + .close = ad198x_dig_playback_pcm_close + }, +}; + +static struct hda_pcm_stream ad198x_pcm_digital_capture = { + .substreams = 1, + .channels_min = 2, + .channels_max = 2, + /* NID is set in alc_build_pcms */ +}; + +static int ad198x_build_pcms(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + struct hda_pcm *info = spec->pcm_rec; + + codec->num_pcms = 1; + codec->pcm_info = info; + + info->name = "AD198x Analog"; + info->stream[SNDRV_PCM_STREAM_PLAYBACK] = ad198x_pcm_analog_playback; + info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max = spec->multiout.max_channels; + info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = spec->multiout.dac_nids[0]; + info->stream[SNDRV_PCM_STREAM_CAPTURE] = ad198x_pcm_analog_capture; + info->stream[SNDRV_PCM_STREAM_CAPTURE].substreams = spec->num_adc_nids; + info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->adc_nids[0]; + + if (spec->multiout.dig_out_nid) { + info++; + codec->num_pcms++; + info->name = "AD198x Digital"; + info->stream[SNDRV_PCM_STREAM_PLAYBACK] = ad198x_pcm_digital_playback; + info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = spec->multiout.dig_out_nid; + if (spec->dig_in_nid) { + info->stream[SNDRV_PCM_STREAM_CAPTURE] = ad198x_pcm_digital_capture; + info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->dig_in_nid; + } + } + + return 0; +} + +static void ad198x_free(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + unsigned int i; + + if (spec->kctl_alloc) { + for (i = 0; i < spec->num_kctl_used; i++) + kfree(spec->kctl_alloc[i].name); + kfree(spec->kctl_alloc); + } + kfree(codec->spec); +} + +#ifdef CONFIG_PM +static int ad198x_resume(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + int i; + + codec->patch_ops.init(codec); + for (i = 0; i < spec->num_mixers; i++) + snd_hda_resume_ctls(codec, spec->mixers[i]); + if (spec->multiout.dig_out_nid) + snd_hda_resume_spdif_out(codec); + if (spec->dig_in_nid) + snd_hda_resume_spdif_in(codec); + return 0; +} +#endif + +static struct hda_codec_ops ad198x_patch_ops = { + .build_controls = ad198x_build_controls, + .build_pcms = ad198x_build_pcms, + .init = ad198x_init, + .free = ad198x_free, +#ifdef CONFIG_PM + .resume = ad198x_resume, +#endif }; + +/* + * EAPD control + * the private value = nid | (invert << 8) + */ +static int ad198x_eapd_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN; + uinfo->count = 1; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = 1; + return 0; +} + +static int ad198x_eapd_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + int invert = (kcontrol->private_value >> 8) & 1; + if (invert) + ucontrol->value.integer.value[0] = ! spec->cur_eapd; + else + ucontrol->value.integer.value[0] = spec->cur_eapd; + return 0; +} + +static int ad198x_eapd_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + int invert = (kcontrol->private_value >> 8) & 1; + hda_nid_t nid = kcontrol->private_value & 0xff; + unsigned int eapd; + eapd = ucontrol->value.integer.value[0]; + if (invert) + eapd = !eapd; + if (eapd == spec->cur_eapd && ! codec->in_resume) + return 0; + spec->cur_eapd = eapd; + snd_hda_codec_write(codec, nid, + 0, AC_VERB_SET_EAPD_BTLENABLE, + eapd ? 0x02 : 0x00); + return 1; +} + +static int ad198x_ch_mode_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo); +static int ad198x_ch_mode_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol); +static int ad198x_ch_mode_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol); + + +/* + * AD1986A specific + */ + #define AD1986A_SPDIF_OUT 0x02 #define AD1986A_FRONT_DAC 0x03 #define AD1986A_SURR_DAC 0x04 @@ -43,6 +401,8 @@ struct ad1986a_spec { static hda_nid_t ad1986a_dac_nids[3] = { AD1986A_FRONT_DAC, AD1986A_SURR_DAC, AD1986A_CLFE_DAC }; +static hda_nid_t ad1986a_adc_nids[1] = { AD1986A_ADC }; +static hda_nid_t ad1986a_capsrc_nids[1] = { 0x12 }; static struct hda_input_mux ad1986a_capture_source = { .num_items = 7, @@ -65,92 +425,66 @@ static struct hda_input_mux ad1986a_capture_source = { #define ad1986a_pcm_amp_vol_info snd_hda_mixer_amp_volume_info -static int ad1986a_pcm_amp_vol_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int ad1986a_pcm_amp_vol_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *ad = codec->spec; + struct ad198x_spec *ad = codec->spec; - down(&ad->amp_mutex); + mutex_lock(&ad->amp_mutex); snd_hda_mixer_amp_volume_get(kcontrol, ucontrol); - up(&ad->amp_mutex); + mutex_unlock(&ad->amp_mutex); return 0; } -static int ad1986a_pcm_amp_vol_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int ad1986a_pcm_amp_vol_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *ad = codec->spec; + struct ad198x_spec *ad = codec->spec; int i, change = 0; - down(&ad->amp_mutex); + mutex_lock(&ad->amp_mutex); for (i = 0; i < ARRAY_SIZE(ad1986a_dac_nids); i++) { kcontrol->private_value = HDA_COMPOSE_AMP_VAL(ad1986a_dac_nids[i], 3, 0, HDA_OUTPUT); change |= snd_hda_mixer_amp_volume_put(kcontrol, ucontrol); } kcontrol->private_value = HDA_COMPOSE_AMP_VAL(AD1986A_FRONT_DAC, 3, 0, HDA_OUTPUT); - up(&ad->amp_mutex); + mutex_unlock(&ad->amp_mutex); return change; } -#define ad1986a_pcm_amp_sw_info snd_hda_mixer_amp_volume_info +#define ad1986a_pcm_amp_sw_info snd_hda_mixer_amp_switch_info -static int ad1986a_pcm_amp_sw_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int ad1986a_pcm_amp_sw_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *ad = codec->spec; + struct ad198x_spec *ad = codec->spec; - down(&ad->amp_mutex); + mutex_lock(&ad->amp_mutex); snd_hda_mixer_amp_switch_get(kcontrol, ucontrol); - up(&ad->amp_mutex); + mutex_unlock(&ad->amp_mutex); return 0; } -static int ad1986a_pcm_amp_sw_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int ad1986a_pcm_amp_sw_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *ad = codec->spec; + struct ad198x_spec *ad = codec->spec; int i, change = 0; - down(&ad->amp_mutex); + mutex_lock(&ad->amp_mutex); for (i = 0; i < ARRAY_SIZE(ad1986a_dac_nids); i++) { kcontrol->private_value = HDA_COMPOSE_AMP_VAL(ad1986a_dac_nids[i], 3, 0, HDA_OUTPUT); change |= snd_hda_mixer_amp_switch_put(kcontrol, ucontrol); } kcontrol->private_value = HDA_COMPOSE_AMP_VAL(AD1986A_FRONT_DAC, 3, 0, HDA_OUTPUT); - up(&ad->amp_mutex); + mutex_unlock(&ad->amp_mutex); return change; } -/* - * input MUX handling - */ -static int ad1986a_mux_enum_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo) -{ - return snd_hda_input_mux_info(&ad1986a_capture_source, uinfo); -} - -static int ad1986a_mux_enum_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) -{ - struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *spec = codec->spec; - - ucontrol->value.enumerated.item[0] = spec->cur_mux; - return 0; -} - -static int ad1986a_mux_enum_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) -{ - struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct ad1986a_spec *spec = codec->spec; - - return snd_hda_input_mux_put(codec, &ad1986a_capture_source, ucontrol, - AD1986A_ADC, &spec->cur_mux); -} - /* * mixers */ -static snd_kcontrol_new_t ad1986a_mixers[] = { +static struct snd_kcontrol_new ad1986a_mixers[] = { { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = "PCM Playback Volume", @@ -194,14 +528,151 @@ static snd_kcontrol_new_t ad1986a_mixers[] = { { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = "Capture Source", - .info = ad1986a_mux_enum_info, - .get = ad1986a_mux_enum_get, - .put = ad1986a_mux_enum_put, + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, }, HDA_CODEC_MUTE("Stereo Downmix Switch", 0x09, 0x0, HDA_OUTPUT), { } /* end */ }; +/* additional mixers for 3stack mode */ +static struct snd_kcontrol_new ad1986a_3st_mixers[] = { + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = ad198x_ch_mode_info, + .get = ad198x_ch_mode_get, + .put = ad198x_ch_mode_put, + }, + { } /* end */ +}; + +/* laptop model - 2ch only */ +static hda_nid_t ad1986a_laptop_dac_nids[1] = { AD1986A_FRONT_DAC }; + +static struct snd_kcontrol_new ad1986a_laptop_mixers[] = { + HDA_CODEC_VOLUME("PCM Playback Volume", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Master Playback Volume", 0x1b, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Master Playback Switch", 0x1b, 0x0, HDA_OUTPUT), + /* HDA_CODEC_VOLUME("Headphone Playback Volume", 0x1a, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x1a, 0x0, HDA_OUTPUT), */ + HDA_CODEC_VOLUME("CD Playback Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x17, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x17, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Aux Playback Volume", 0x16, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Aux Playback Switch", 0x16, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x13, 0x0, HDA_OUTPUT), + /* HDA_CODEC_VOLUME("PC Speaker Playback Volume", 0x18, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PC Speaker Playback Switch", 0x18, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mono Playback Volume", 0x1e, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mono Playback Switch", 0x1e, 0x0, HDA_OUTPUT), */ + HDA_CODEC_VOLUME("Capture Volume", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x12, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + { } /* end */ +}; + +/* laptop-eapd model - 2ch only */ + +/* master controls both pins 0x1a and 0x1b */ +static int ad1986a_laptop_master_vol_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + long *valp = ucontrol->value.integer.value; + int change; + + change = snd_hda_codec_amp_update(codec, 0x1a, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + change |= snd_hda_codec_amp_update(codec, 0x1a, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + snd_hda_codec_amp_update(codec, 0x1b, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + snd_hda_codec_amp_update(codec, 0x1b, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + return change; +} + +static int ad1986a_laptop_master_sw_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + long *valp = ucontrol->value.integer.value; + int change; + + change = snd_hda_codec_amp_update(codec, 0x1a, 0, HDA_OUTPUT, 0, + 0x80, valp[0] ? 0 : 0x80); + change |= snd_hda_codec_amp_update(codec, 0x1a, 1, HDA_OUTPUT, 0, + 0x80, valp[1] ? 0 : 0x80); + snd_hda_codec_amp_update(codec, 0x1b, 0, HDA_OUTPUT, 0, + 0x80, valp[0] ? 0 : 0x80); + snd_hda_codec_amp_update(codec, 0x1b, 1, HDA_OUTPUT, 0, + 0x80, valp[1] ? 0 : 0x80); + return change; +} + +static struct hda_input_mux ad1986a_laptop_eapd_capture_source = { + .num_items = 3, + .items = { + { "Mic", 0x0 }, + { "Internal Mic", 0x4 }, + { "Mix", 0x5 }, + }, +}; + +static struct snd_kcontrol_new ad1986a_laptop_eapd_mixers[] = { + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Volume", + .info = snd_hda_mixer_amp_volume_info, + .get = snd_hda_mixer_amp_volume_get, + .put = ad1986a_laptop_master_vol_put, + .private_value = HDA_COMPOSE_AMP_VAL(0x1a, 3, 0, HDA_OUTPUT), + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Switch", + .info = snd_hda_mixer_amp_switch_info, + .get = snd_hda_mixer_amp_switch_get, + .put = ad1986a_laptop_master_sw_put, + .private_value = HDA_COMPOSE_AMP_VAL(0x1a, 3, 0, HDA_OUTPUT), + }, + HDA_CODEC_VOLUME("PCM Playback Volume", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x17, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x17, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Capture Volume", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x12, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "External Amplifier", + .info = ad198x_eapd_info, + .get = ad198x_eapd_get, + .put = ad198x_eapd_put, + .private_value = 0x1b | (1 << 8), /* port-D, inversed */ + }, + { } /* end */ +}; + /* * initialization verbs */ @@ -241,205 +712,1924 @@ static struct hda_verb ad1986a_init_verbs[] = { {0x1c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, {0x1d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* HP Pin */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0 }, + /* Front, Surround, CLFE Pins */ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + {0x1d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* Mono Pin */ + {0x1e, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* Mic Pin */ + {0x1f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* Line, Aux, CD, Beep-In Pin */ + {0x20, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + {0x21, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + {0x22, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + {0x23, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + {0x24, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, { } /* end */ }; +/* additional verbs for 3-stack model */ +static struct hda_verb ad1986a_3st_init_verbs[] = { + /* Mic and line-in selectors */ + {0x0f, AC_VERB_SET_CONNECT_SEL, 0x2}, + {0x10, AC_VERB_SET_CONNECT_SEL, 0x1}, + { } /* end */ +}; -static int ad1986a_init(struct hda_codec *codec) -{ - snd_hda_sequence_write(codec, ad1986a_init_verbs); - return 0; -} +static struct hda_verb ad1986a_ch2_init[] = { + /* Surround out -> Line In */ + { 0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + { 0x1c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* CLFE -> Mic in */ + { 0x1d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + { 0x1d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + { } /* end */ +}; -static int ad1986a_build_controls(struct hda_codec *codec) -{ - int err; +static struct hda_verb ad1986a_ch4_init[] = { + /* Surround out -> Surround */ + { 0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x1c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* CLFE -> Mic in */ + { 0x1d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + { 0x1d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + { } /* end */ +}; - err = snd_hda_add_new_ctls(codec, ad1986a_mixers); - if (err < 0) - return err; - err = snd_hda_create_spdif_out_ctls(codec, AD1986A_SPDIF_OUT); - if (err < 0) - return err; - return 0; -} +static struct hda_verb ad1986a_ch6_init[] = { + /* Surround out -> Surround out */ + { 0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x1c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* CLFE -> CLFE */ + { 0x1d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x1d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + { } /* end */ +}; + +static struct hda_channel_mode ad1986a_modes[3] = { + { 2, ad1986a_ch2_init }, + { 4, ad1986a_ch4_init }, + { 6, ad1986a_ch6_init }, +}; + +/* eapd initialization */ +static struct hda_verb ad1986a_eapd_init_verbs[] = { + {0x1b, AC_VERB_SET_EAPD_BTLENABLE, 0x00}, + {} +}; + +/* models */ +enum { AD1986A_6STACK, AD1986A_3STACK, AD1986A_LAPTOP, AD1986A_LAPTOP_EAPD }; + +static struct hda_board_config ad1986a_cfg_tbl[] = { + { .modelname = "6stack", .config = AD1986A_6STACK }, + { .modelname = "3stack", .config = AD1986A_3STACK }, + { .pci_subvendor = 0x10de, .pci_subdevice = 0xcb84, + .config = AD1986A_3STACK }, /* ASUS A8N-VM CSM */ + { .modelname = "laptop", .config = AD1986A_LAPTOP }, + { .pci_subvendor = 0x144d, .pci_subdevice = 0xc01e, + .config = AD1986A_LAPTOP }, /* FSC V2060 */ + { .pci_subvendor = 0x17c0, .pci_subdevice = 0x2017, + .config = AD1986A_LAPTOP }, /* Samsung M50 */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x818f, + .config = AD1986A_LAPTOP }, /* ASUS P5GV-MX */ + { .modelname = "laptop-eapd", .config = AD1986A_LAPTOP_EAPD }, + { .pci_subvendor = 0x144d, .pci_subdevice = 0xc024, + .config = AD1986A_LAPTOP_EAPD }, /* Samsung R65-T2300 Charis */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1153, + .config = AD1986A_LAPTOP_EAPD }, /* ASUS M9 */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1213, + .config = AD1986A_LAPTOP_EAPD }, /* ASUS A6J */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x11f7, + .config = AD1986A_LAPTOP_EAPD }, /* ASUS U5A */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1297, + .config = AD1986A_LAPTOP_EAPD }, /* ASUS Z62F */ + { .pci_subvendor = 0x103c, .pci_subdevice = 0x30af, + .config = AD1986A_LAPTOP_EAPD }, /* HP Compaq Presario B2800 */ + {} +}; + +static int patch_ad1986a(struct hda_codec *codec) +{ + struct ad198x_spec *spec; + int board_config; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + mutex_init(&spec->amp_mutex); + codec->spec = spec; + + spec->multiout.max_channels = 6; + spec->multiout.num_dacs = ARRAY_SIZE(ad1986a_dac_nids); + spec->multiout.dac_nids = ad1986a_dac_nids; + spec->multiout.dig_out_nid = AD1986A_SPDIF_OUT; + spec->num_adc_nids = 1; + spec->adc_nids = ad1986a_adc_nids; + spec->capsrc_nids = ad1986a_capsrc_nids; + spec->input_mux = &ad1986a_capture_source; + spec->num_mixers = 1; + spec->mixers[0] = ad1986a_mixers; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1986a_init_verbs; + + codec->patch_ops = ad198x_patch_ops; + + /* override some parameters */ + board_config = snd_hda_check_board_config(codec, ad1986a_cfg_tbl); + switch (board_config) { + case AD1986A_3STACK: + spec->num_mixers = 2; + spec->mixers[1] = ad1986a_3st_mixers; + spec->num_init_verbs = 3; + spec->init_verbs[1] = ad1986a_3st_init_verbs; + spec->init_verbs[2] = ad1986a_ch2_init; + spec->channel_mode = ad1986a_modes; + spec->num_channel_mode = ARRAY_SIZE(ad1986a_modes); + spec->need_dac_fix = 1; + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = 1; + break; + case AD1986A_LAPTOP: + spec->mixers[0] = ad1986a_laptop_mixers; + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = 1; + spec->multiout.dac_nids = ad1986a_laptop_dac_nids; + break; + case AD1986A_LAPTOP_EAPD: + spec->mixers[0] = ad1986a_laptop_eapd_mixers; + spec->num_init_verbs = 2; + spec->init_verbs[1] = ad1986a_eapd_init_verbs; + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = 1; + spec->multiout.dac_nids = ad1986a_laptop_dac_nids; + spec->multiout.dig_out_nid = 0; + spec->input_mux = &ad1986a_laptop_eapd_capture_source; + break; + } + + return 0; +} /* - * Analog playback callbacks + * AD1983 specific */ -static int ad1986a_playback_pcm_open(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - snd_pcm_substream_t *substream) + +#define AD1983_SPDIF_OUT 0x02 +#define AD1983_DAC 0x03 +#define AD1983_ADC 0x04 + +static hda_nid_t ad1983_dac_nids[1] = { AD1983_DAC }; +static hda_nid_t ad1983_adc_nids[1] = { AD1983_ADC }; +static hda_nid_t ad1983_capsrc_nids[1] = { 0x15 }; + +static struct hda_input_mux ad1983_capture_source = { + .num_items = 4, + .items = { + { "Mic", 0x0 }, + { "Line", 0x1 }, + { "Mix", 0x2 }, + { "Mix Mono", 0x3 }, + }, +}; + +/* + * SPDIF playback route + */ +static int ad1983_spdif_route_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) { - struct ad1986a_spec *spec = codec->spec; - return snd_hda_multi_out_analog_open(codec, &spec->multiout, substream); + static char *texts[] = { "PCM", "ADC" }; + + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 2; + if (uinfo->value.enumerated.item > 1) + uinfo->value.enumerated.item = 1; + strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); + return 0; } -static int ad1986a_playback_pcm_prepare(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - unsigned int stream_tag, - unsigned int format, - snd_pcm_substream_t *substream) +static int ad1983_spdif_route_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { - struct ad1986a_spec *spec = codec->spec; - return snd_hda_multi_out_analog_prepare(codec, &spec->multiout, stream_tag, - format, substream); + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + + ucontrol->value.enumerated.item[0] = spec->spdif_route; + return 0; } -static int ad1986a_playback_pcm_cleanup(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - snd_pcm_substream_t *substream) +static int ad1983_spdif_route_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { - struct ad1986a_spec *spec = codec->spec; - return snd_hda_multi_out_analog_cleanup(codec, &spec->multiout); + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + + if (spec->spdif_route != ucontrol->value.enumerated.item[0]) { + spec->spdif_route = ucontrol->value.enumerated.item[0]; + snd_hda_codec_write(codec, spec->multiout.dig_out_nid, 0, + AC_VERB_SET_CONNECT_SEL, spec->spdif_route); + return 1; + } + return 0; } +static struct snd_kcontrol_new ad1983_mixers[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x05, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Playback Switch", 0x05, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x07, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("Mono Playback Switch", 0x07, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("PCM Playback Volume", 0x11, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x11, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("PC Speaker Playback Volume", 0x10, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("PC Speaker Playback Switch", 0x10, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Boost", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Capture Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x15, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = SNDRV_CTL_NAME_IEC958("",PLAYBACK,NONE) "Route", + .info = ad1983_spdif_route_info, + .get = ad1983_spdif_route_get, + .put = ad1983_spdif_route_put, + }, + { } /* end */ +}; + +static struct hda_verb ad1983_init_verbs[] = { + /* Front, HP, Mono; mute as default */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* Beep, PCM, Mic, Line-In: mute */ + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* Front, HP selectors; from Mix */ + {0x05, AC_VERB_SET_CONNECT_SEL, 0x01}, + {0x06, AC_VERB_SET_CONNECT_SEL, 0x01}, + /* Mono selector; from Mix */ + {0x0b, AC_VERB_SET_CONNECT_SEL, 0x03}, + /* Mic selector; Mic */ + {0x0c, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Line-in selector: Line-in */ + {0x0d, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Mic boost: 0dB */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* Record selector: mic */ + {0x15, AC_VERB_SET_CONNECT_SEL, 0x0}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* SPDIF route: PCM */ + {0x02, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Front Pin */ + {0x05, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* HP Pin */ + {0x06, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0 }, + /* Mono Pin */ + {0x07, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* Mic Pin */ + {0x08, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* Line Pin */ + {0x09, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + { } /* end */ +}; + + +static int patch_ad1983(struct hda_codec *codec) +{ + struct ad198x_spec *spec; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + mutex_init(&spec->amp_mutex); + codec->spec = spec; + + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = ARRAY_SIZE(ad1983_dac_nids); + spec->multiout.dac_nids = ad1983_dac_nids; + spec->multiout.dig_out_nid = AD1983_SPDIF_OUT; + spec->num_adc_nids = 1; + spec->adc_nids = ad1983_adc_nids; + spec->capsrc_nids = ad1983_capsrc_nids; + spec->input_mux = &ad1983_capture_source; + spec->num_mixers = 1; + spec->mixers[0] = ad1983_mixers; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1983_init_verbs; + spec->spdif_route = 0; + + codec->patch_ops = ad198x_patch_ops; + + return 0; +} + + /* - * Digital out + * AD1981 HD specific */ -static int ad1986a_dig_playback_pcm_open(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - snd_pcm_substream_t *substream) + +#define AD1981_SPDIF_OUT 0x02 +#define AD1981_DAC 0x03 +#define AD1981_ADC 0x04 + +static hda_nid_t ad1981_dac_nids[1] = { AD1981_DAC }; +static hda_nid_t ad1981_adc_nids[1] = { AD1981_ADC }; +static hda_nid_t ad1981_capsrc_nids[1] = { 0x15 }; + +/* 0x0c, 0x09, 0x0e, 0x0f, 0x19, 0x05, 0x18, 0x17 */ +static struct hda_input_mux ad1981_capture_source = { + .num_items = 7, + .items = { + { "Front Mic", 0x0 }, + { "Line", 0x1 }, + { "Mix", 0x2 }, + { "Mix Mono", 0x3 }, + { "CD", 0x4 }, + { "Mic", 0x6 }, + { "Aux", 0x7 }, + }, +}; + +static struct snd_kcontrol_new ad1981_mixers[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x05, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Playback Switch", 0x05, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x07, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("Mono Playback Switch", 0x07, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("PCM Playback Volume", 0x11, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x11, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Aux Playback Volume", 0x1b, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Aux Playback Switch", 0x1b, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x1c, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x1c, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x1d, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x1d, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("PC Speaker Playback Volume", 0x0d, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("PC Speaker Playback Switch", 0x0d, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Front Mic Boost", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Boost", 0x18, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Capture Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x15, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + /* identical with AD1983 */ + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = SNDRV_CTL_NAME_IEC958("",PLAYBACK,NONE) "Route", + .info = ad1983_spdif_route_info, + .get = ad1983_spdif_route_get, + .put = ad1983_spdif_route_put, + }, + { } /* end */ +}; + +static struct hda_verb ad1981_init_verbs[] = { + /* Front, HP, Mono; mute as default */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* Beep, PCM, Front Mic, Line, Rear Mic, Aux, CD-In: mute */ + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x1c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x1d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* Front, HP selectors; from Mix */ + {0x05, AC_VERB_SET_CONNECT_SEL, 0x01}, + {0x06, AC_VERB_SET_CONNECT_SEL, 0x01}, + /* Mono selector; from Mix */ + {0x0b, AC_VERB_SET_CONNECT_SEL, 0x03}, + /* Mic Mixer; select Front Mic */ + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + {0x1f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* Mic boost: 0dB */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* Record selector: Front mic */ + {0x15, AC_VERB_SET_CONNECT_SEL, 0x0}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + /* SPDIF route: PCM */ + {0x02, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Front Pin */ + {0x05, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* HP Pin */ + {0x06, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0 }, + /* Mono Pin */ + {0x07, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* Front & Rear Mic Pins */ + {0x08, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* Line Pin */ + {0x09, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* Digital Beep */ + {0x0d, AC_VERB_SET_CONNECT_SEL, 0x00}, + /* Line-Out as Input: disabled */ + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + { } /* end */ +}; + +/* + * Patch for HP nx6320 + * + * nx6320 uses EAPD in the reserve way - EAPD-on means the internal + * speaker output enabled _and_ mute-LED off. + */ + +#define AD1981_HP_EVENT 0x37 +#define AD1981_MIC_EVENT 0x38 + +static struct hda_verb ad1981_hp_init_verbs[] = { + {0x05, AC_VERB_SET_EAPD_BTLENABLE, 0x00 }, /* default off */ + /* pin sensing on HP and Mic jacks */ + {0x06, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | AD1981_HP_EVENT}, + {0x08, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | AD1981_MIC_EVENT}, + {} +}; + +/* turn on/off EAPD (+ mute HP) as a master switch */ +static int ad1981_hp_master_sw_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { - struct ad1986a_spec *spec = codec->spec; - return snd_hda_multi_out_dig_open(codec, &spec->multiout); + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + + if (! ad198x_eapd_put(kcontrol, ucontrol)) + return 0; + + /* toggle HP mute appropriately */ + snd_hda_codec_amp_update(codec, 0x06, 0, HDA_OUTPUT, 0, + 0x80, spec->cur_eapd ? 0 : 0x80); + snd_hda_codec_amp_update(codec, 0x06, 1, HDA_OUTPUT, 0, + 0x80, spec->cur_eapd ? 0 : 0x80); + return 1; } -static int ad1986a_dig_playback_pcm_close(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - snd_pcm_substream_t *substream) +/* bind volumes of both NID 0x05 and 0x06 */ +static int ad1981_hp_master_vol_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) { - struct ad1986a_spec *spec = codec->spec; - return snd_hda_multi_out_dig_close(codec, &spec->multiout); + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + long *valp = ucontrol->value.integer.value; + int change; + + change = snd_hda_codec_amp_update(codec, 0x05, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + change |= snd_hda_codec_amp_update(codec, 0x05, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + snd_hda_codec_amp_update(codec, 0x06, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + snd_hda_codec_amp_update(codec, 0x06, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + return change; } -/* - * Analog capture - */ -static int ad1986a_capture_pcm_prepare(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - unsigned int stream_tag, - unsigned int format, - snd_pcm_substream_t *substream) +/* mute internal speaker if HP is plugged */ +static void ad1981_hp_automute(struct hda_codec *codec) +{ + unsigned int present; + + present = snd_hda_codec_read(codec, 0x06, 0, + AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + snd_hda_codec_amp_update(codec, 0x05, 0, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); + snd_hda_codec_amp_update(codec, 0x05, 1, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); +} + +/* toggle input of built-in and mic jack appropriately */ +static void ad1981_hp_automic(struct hda_codec *codec) { - snd_hda_codec_setup_stream(codec, AD1986A_ADC, stream_tag, 0, format); + static struct hda_verb mic_jack_on[] = { + {0x1f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + {} + }; + static struct hda_verb mic_jack_off[] = { + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x1f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + {} + }; + unsigned int present; + + present = snd_hda_codec_read(codec, 0x08, 0, + AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + if (present) + snd_hda_sequence_write(codec, mic_jack_on); + else + snd_hda_sequence_write(codec, mic_jack_off); +} + +/* unsolicited event for HP jack sensing */ +static void ad1981_hp_unsol_event(struct hda_codec *codec, + unsigned int res) +{ + res >>= 26; + switch (res) { + case AD1981_HP_EVENT: + ad1981_hp_automute(codec); + break; + case AD1981_MIC_EVENT: + ad1981_hp_automic(codec); + break; + } +} + +static struct hda_input_mux ad1981_hp_capture_source = { + .num_items = 3, + .items = { + { "Mic", 0x0 }, + { "Docking-Station", 0x1 }, + { "Mix", 0x2 }, + }, +}; + +static struct snd_kcontrol_new ad1981_hp_mixers[] = { + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Volume", + .info = snd_hda_mixer_amp_volume_info, + .get = snd_hda_mixer_amp_volume_get, + .put = ad1981_hp_master_vol_put, + .private_value = HDA_COMPOSE_AMP_VAL(0x05, 3, 0, HDA_OUTPUT), + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Switch", + .info = ad198x_eapd_info, + .get = ad198x_eapd_get, + .put = ad1981_hp_master_sw_put, + .private_value = 0x05, + }, + HDA_CODEC_VOLUME("PCM Playback Volume", 0x11, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x11, 0x0, HDA_OUTPUT), +#if 0 + /* FIXME: analog mic/line loopback doesn't work with my tests... + * (although recording is OK) + */ + HDA_CODEC_VOLUME("Mic Playback Volume", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x12, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Docking-Station Playback Volume", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Docking-Station Playback Switch", 0x13, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x1c, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x1c, 0x0, HDA_OUTPUT), + /* FIXME: does this laptop have analog CD connection? */ + HDA_CODEC_VOLUME("CD Playback Volume", 0x1d, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x1d, 0x0, HDA_OUTPUT), +#endif + HDA_CODEC_VOLUME("Mic Boost", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Internal Mic Boost", 0x18, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Capture Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x15, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + { } /* end */ +}; + +/* initialize jack-sensing, too */ +static int ad1981_hp_init(struct hda_codec *codec) +{ + ad198x_init(codec); + ad1981_hp_automute(codec); + ad1981_hp_automic(codec); return 0; } -static int ad1986a_capture_pcm_cleanup(struct hda_pcm_stream *hinfo, - struct hda_codec *codec, - snd_pcm_substream_t *substream) +/* models */ +enum { AD1981_BASIC, AD1981_HP }; + +static struct hda_board_config ad1981_cfg_tbl[] = { + { .modelname = "hp", .config = AD1981_HP }, + /* All HP models */ + { .pci_subvendor = 0x103c, .config = AD1981_HP }, + { .modelname = "basic", .config = AD1981_BASIC }, + {} +}; + +static int patch_ad1981(struct hda_codec *codec) { - snd_hda_codec_setup_stream(codec, AD1986A_ADC, 0, 0, 0); + struct ad198x_spec *spec; + int board_config; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + mutex_init(&spec->amp_mutex); + codec->spec = spec; + + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = ARRAY_SIZE(ad1981_dac_nids); + spec->multiout.dac_nids = ad1981_dac_nids; + spec->multiout.dig_out_nid = AD1981_SPDIF_OUT; + spec->num_adc_nids = 1; + spec->adc_nids = ad1981_adc_nids; + spec->capsrc_nids = ad1981_capsrc_nids; + spec->input_mux = &ad1981_capture_source; + spec->num_mixers = 1; + spec->mixers[0] = ad1981_mixers; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1981_init_verbs; + spec->spdif_route = 0; + + codec->patch_ops = ad198x_patch_ops; + + /* override some parameters */ + board_config = snd_hda_check_board_config(codec, ad1981_cfg_tbl); + switch (board_config) { + case AD1981_HP: + spec->mixers[0] = ad1981_hp_mixers; + spec->num_init_verbs = 2; + spec->init_verbs[1] = ad1981_hp_init_verbs; + spec->multiout.dig_out_nid = 0; + spec->input_mux = &ad1981_hp_capture_source; + + codec->patch_ops.init = ad1981_hp_init; + codec->patch_ops.unsol_event = ad1981_hp_unsol_event; + break; + } + return 0; } /* + * AD1988 + * + * Output pins and routes + * + * Pin Mix Sel DAC (*) + * port-A 0x11 (mute/hp) <- 0x22 <- 0x37 <- 03/04/06 + * port-B 0x14 (mute/hp) <- 0x2b <- 0x30 <- 03/04/06 + * port-C 0x15 (mute) <- 0x2c <- 0x31 <- 05/0a + * port-D 0x12 (mute/hp) <- 0x29 <- 04 + * port-E 0x17 (mute/hp) <- 0x26 <- 0x32 <- 05/0a + * port-F 0x16 (mute) <- 0x2a <- 06 + * port-G 0x24 (mute) <- 0x27 <- 05 + * port-H 0x25 (mute) <- 0x28 <- 0a + * mono 0x13 (mute/amp)<- 0x1e <- 0x36 <- 03/04/06 + * + * DAC0 = 03h, DAC1 = 04h, DAC2 = 05h, DAC3 = 06h, DAC4 = 0ah + * (*) DAC2/3/4 are swapped to DAC3/4/2 on AD198A rev.2 due to a h/w bug. + * + * Input pins and routes + * + * pin boost mix input # / adc input # + * port-A 0x11 -> 0x38 -> mix 2, ADC 0 + * port-B 0x14 -> 0x39 -> mix 0, ADC 1 + * port-C 0x15 -> 0x3a -> 33:0 - mix 1, ADC 2 + * port-D 0x12 -> 0x3d -> mix 3, ADC 8 + * port-E 0x17 -> 0x3c -> 34:0 - mix 4, ADC 4 + * port-F 0x16 -> 0x3b -> mix 5, ADC 3 + * port-G 0x24 -> N/A -> 33:1 - mix 1, 34:1 - mix 4, ADC 6 + * port-H 0x25 -> N/A -> 33:2 - mix 1, 34:2 - mix 4, ADC 7 + * + * + * DAC assignment + * 6stack - front/surr/CLFE/side/opt DACs - 04/06/05/0a/03 + * 3stack - front/surr/CLFE/opt DACs - 04/05/0a/03 + * + * Inputs of Analog Mix (0x20) + * 0:Port-B (front mic) + * 1:Port-C/G/H (line-in) + * 2:Port-A + * 3:Port-D (line-in/2) + * 4:Port-E/G/H (mic-in) + * 5:Port-F (mic2-in) + * 6:CD + * 7:Beep + * + * ADC selection + * 0:Port-A + * 1:Port-B (front mic-in) + * 2:Port-C (line-in) + * 3:Port-F (mic2-in) + * 4:Port-E (mic-in) + * 5:CD + * 6:Port-G + * 7:Port-H + * 8:Port-D (line-in/2) + * 9:Mix + * + * Proposed pin assignments by the datasheet + * + * 6-stack + * Port-A front headphone + * B front mic-in + * C rear line-in + * D rear front-out + * E rear mic-in + * F rear surround + * G rear CLFE + * H rear side + * + * 3-stack + * Port-A front headphone + * B front mic + * C rear line-in/surround + * D rear front-out + * E rear mic-in/CLFE + * + * laptop + * Port-A headphone + * B mic-in + * C docking station + * D internal speaker (with EAPD) + * E/F quad mic array + */ + + +/* models */ +enum { + AD1988_6STACK, + AD1988_6STACK_DIG, + AD1988_3STACK, + AD1988_3STACK_DIG, + AD1988_LAPTOP, + AD1988_LAPTOP_DIG, + AD1988_AUTO, + AD1988_MODEL_LAST, +}; + +/* reivision id to check workarounds */ +#define AD1988A_REV2 0x100200 + +#define is_rev2(codec) \ + ((codec)->vendor_id == 0x11d41988 && \ + (codec)->revision_id == AD1988A_REV2) + +/* + * mixers */ -static struct hda_pcm_stream ad1986a_pcm_analog_playback = { - .substreams = 1, - .channels_min = 2, - .channels_max = 6, - .nid = AD1986A_FRONT_DAC, /* NID to query formats and rates */ - .ops = { - .open = ad1986a_playback_pcm_open, - .prepare = ad1986a_playback_pcm_prepare, - .cleanup = ad1986a_playback_pcm_cleanup + +static hda_nid_t ad1988_6stack_dac_nids[4] = { + 0x04, 0x06, 0x05, 0x0a +}; + +static hda_nid_t ad1988_3stack_dac_nids[3] = { + 0x04, 0x05, 0x0a +}; + +/* for AD1988A revision-2, DAC2-4 are swapped */ +static hda_nid_t ad1988_6stack_dac_nids_rev2[4] = { + 0x04, 0x05, 0x0a, 0x06 +}; + +static hda_nid_t ad1988_3stack_dac_nids_rev2[3] = { + 0x04, 0x0a, 0x06 +}; + +static hda_nid_t ad1988_adc_nids[3] = { + 0x08, 0x09, 0x0f +}; + +static hda_nid_t ad1988_capsrc_nids[3] = { + 0x0c, 0x0d, 0x0e +}; + +#define AD1988_SPDIF_OUT 0x02 +#define AD1988_SPDIF_IN 0x07 + +static struct hda_input_mux ad1988_6stack_capture_source = { + .num_items = 5, + .items = { + { "Front Mic", 0x0 }, + { "Line", 0x1 }, + { "Mic", 0x4 }, + { "CD", 0x5 }, + { "Mix", 0x9 }, }, }; -static struct hda_pcm_stream ad1986a_pcm_analog_capture = { - .substreams = 2, - .channels_min = 2, - .channels_max = 2, - .nid = AD1986A_ADC, /* NID to query formats and rates */ - .ops = { - .prepare = ad1986a_capture_pcm_prepare, - .cleanup = ad1986a_capture_pcm_cleanup +static struct hda_input_mux ad1988_laptop_capture_source = { + .num_items = 3, + .items = { + { "Mic/Line", 0x0 }, + { "CD", 0x5 }, + { "Mix", 0x9 }, }, }; -static struct hda_pcm_stream ad1986a_pcm_digital_playback = { - .substreams = 1, - .channels_min = 2, - .channels_max = 2, - .nid = AD1986A_SPDIF_OUT, - .ops = { - .open = ad1986a_dig_playback_pcm_open, - .close = ad1986a_dig_playback_pcm_close +/* + */ +static int ad198x_ch_mode_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + return snd_hda_ch_mode_info(codec, uinfo, spec->channel_mode, + spec->num_channel_mode); +} + +static int ad198x_ch_mode_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + return snd_hda_ch_mode_get(codec, ucontrol, spec->channel_mode, + spec->num_channel_mode, spec->multiout.max_channels); +} + +static int ad198x_ch_mode_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct ad198x_spec *spec = codec->spec; + if (spec->need_dac_fix) + spec->multiout.num_dacs = spec->multiout.max_channels / 2; + return snd_hda_ch_mode_put(codec, ucontrol, spec->channel_mode, + spec->num_channel_mode, &spec->multiout.max_channels); +} + +/* 6-stack mode */ +static struct snd_kcontrol_new ad1988_6stack_mixers1[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x04, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x05, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x05, 2, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Side Playback Volume", 0x0a, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new ad1988_6stack_mixers1_rev2[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x04, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x05, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0a, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0a, 2, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Side Playback Volume", 0x06, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new ad1988_6stack_mixers2[] = { + HDA_BIND_MUTE("Front Playback Switch", 0x29, 2, HDA_INPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x2a, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x27, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x27, 2, 2, HDA_INPUT), + HDA_BIND_MUTE("Side Playback Switch", 0x28, 2, HDA_INPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x22, 2, HDA_INPUT), + HDA_BIND_MUTE("Mono Playback Switch", 0x1e, 2, HDA_INPUT), + + HDA_CODEC_VOLUME("CD Playback Volume", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x20, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x20, 0x1, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x20, 0x4, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x20, 0x4, HDA_INPUT), + + HDA_CODEC_VOLUME("Beep Playback Volume", 0x10, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x10, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Analog Mix Playback Volume", 0x21, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Analog Mix Playback Switch", 0x21, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Front Mic Boost", 0x39, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Boost", 0x3c, 0x0, HDA_OUTPUT), + + { } /* end */ +}; + +/* 3-stack mode */ +static struct snd_kcontrol_new ad1988_3stack_mixers1[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x04, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0a, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x05, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x05, 2, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new ad1988_3stack_mixers1_rev2[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x04, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0a, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x06, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x06, 2, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new ad1988_3stack_mixers2[] = { + HDA_BIND_MUTE("Front Playback Switch", 0x29, 2, HDA_INPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x2c, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x26, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x26, 2, 2, HDA_INPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x22, 2, HDA_INPUT), + HDA_BIND_MUTE("Mono Playback Switch", 0x1e, 2, HDA_INPUT), + + HDA_CODEC_VOLUME("CD Playback Volume", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x20, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x20, 0x1, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x20, 0x4, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x20, 0x4, HDA_INPUT), + + HDA_CODEC_VOLUME("Beep Playback Volume", 0x10, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x10, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Analog Mix Playback Volume", 0x21, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Analog Mix Playback Switch", 0x21, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Front Mic Boost", 0x39, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Mic Boost", 0x3c, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = ad198x_ch_mode_info, + .get = ad198x_ch_mode_get, + .put = ad198x_ch_mode_put, }, + + { } /* end */ +}; + +/* laptop mode */ +static struct snd_kcontrol_new ad1988_laptop_mixers[] = { + HDA_CODEC_VOLUME("PCM Playback Volume", 0x04, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("PCM Playback Switch", 0x29, 0x0, HDA_INPUT), + HDA_BIND_MUTE("Mono Playback Switch", 0x1e, 2, HDA_INPUT), + + HDA_CODEC_VOLUME("CD Playback Volume", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x20, 0x6, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x20, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x20, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x20, 0x1, HDA_INPUT), + + HDA_CODEC_VOLUME("Beep Playback Volume", 0x10, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x10, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Analog Mix Playback Volume", 0x21, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Analog Mix Playback Switch", 0x21, 0x0, HDA_OUTPUT), + + HDA_CODEC_VOLUME("Mic Boost", 0x39, 0x0, HDA_OUTPUT), + + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "External Amplifier", + .info = ad198x_eapd_info, + .get = ad198x_eapd_get, + .put = ad198x_eapd_put, + .private_value = 0x12 | (1 << 8), /* port-D, inversed */ + }, + + { } /* end */ +}; + +/* capture */ +static struct snd_kcontrol_new ad1988_capture_mixers[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Capture Switch", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x0d, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x0d, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 2, 0x0e, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 2, 0x0e, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 3, + .info = ad198x_mux_enum_info, + .get = ad198x_mux_enum_get, + .put = ad198x_mux_enum_put, + }, + { } /* end */ }; -static int ad1986a_build_pcms(struct hda_codec *codec) +static int ad1988_spdif_playback_source_info(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_info *uinfo) { - struct ad1986a_spec *spec = codec->spec; - struct hda_pcm *info = spec->pcm_rec; + static char *texts[] = { + "PCM", "ADC1", "ADC2", "ADC3" + }; + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 4; + if (uinfo->value.enumerated.item >= 4) + uinfo->value.enumerated.item = 3; + strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); + return 0; +} - codec->num_pcms = 2; - codec->pcm_info = info; +static int ad1988_spdif_playback_source_get(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + unsigned int sel; - info->name = "AD1986A Analog"; - info->stream[SNDRV_PCM_STREAM_PLAYBACK] = ad1986a_pcm_analog_playback; - info->stream[SNDRV_PCM_STREAM_CAPTURE] = ad1986a_pcm_analog_capture; - info++; + sel = snd_hda_codec_read(codec, 0x02, 0, AC_VERB_GET_CONNECT_SEL, 0); + if (sel > 0) { + sel = snd_hda_codec_read(codec, 0x0b, 0, AC_VERB_GET_CONNECT_SEL, 0); + if (sel <= 3) + sel++; + else + sel = 0; + } + ucontrol->value.enumerated.item[0] = sel; + return 0; +} - info->name = "AD1986A Digital"; - info->stream[SNDRV_PCM_STREAM_PLAYBACK] = ad1986a_pcm_digital_playback; +static int ad1988_spdif_playback_source_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + unsigned int sel; + int change; + + sel = snd_hda_codec_read(codec, 0x02, 0, AC_VERB_GET_CONNECT_SEL, 0); + if (! ucontrol->value.enumerated.item[0]) { + change = sel != 0; + if (change) + snd_hda_codec_write(codec, 0x02, 0, AC_VERB_SET_CONNECT_SEL, 0); + } else { + change = sel == 0; + if (change) + snd_hda_codec_write(codec, 0x02, 0, AC_VERB_SET_CONNECT_SEL, 1); + sel = snd_hda_codec_read(codec, 0x0b, 0, AC_VERB_GET_CONNECT_SEL, 0) + 1; + change |= sel == ucontrol->value.enumerated.item[0]; + if (change) + snd_hda_codec_write(codec, 0x02, 0, AC_VERB_SET_CONNECT_SEL, + ucontrol->value.enumerated.item[0] - 1); + } + return change; +} + +static struct snd_kcontrol_new ad1988_spdif_out_mixers[] = { + HDA_CODEC_VOLUME("IEC958 Playback Volume", 0x1b, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "IEC958 Playback Source", + .info = ad1988_spdif_playback_source_info, + .get = ad1988_spdif_playback_source_get, + .put = ad1988_spdif_playback_source_put, + }, + { } /* end */ +}; + +static struct snd_kcontrol_new ad1988_spdif_in_mixers[] = { + HDA_CODEC_VOLUME("IEC958 Capture Volume", 0x1c, 0x0, HDA_INPUT), + { } /* end */ +}; + + +/* + * initialization verbs + */ + +/* + * for 6-stack (+dig) + */ +static struct hda_verb ad1988_6stack_init_verbs[] = { + /* Front, Surround, CLFE, side DAC; unmute as default */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Port-A front headphon path */ + {0x37, AC_VERB_SET_CONNECT_SEL, 0x01}, /* DAC1:04h */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* Port-D line-out path */ + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Port-F surround path */ + {0x2a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x2a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Port-G CLFE path */ + {0x27, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x27, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x24, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Port-H side path */ + {0x28, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x28, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x25, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x25, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Mono out path */ + {0x36, AC_VERB_SET_CONNECT_SEL, 0x1}, /* DAC1:04h */ + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, 0xb01f}, /* unmute, 0dB */ + /* Port-B front mic-in path */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x39, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* Port-C line-in path */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x3a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x33, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Port-E mic-in path */ + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x3c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x34, AC_VERB_SET_CONNECT_SEL, 0x0}, + + { } +}; + +static struct hda_verb ad1988_capture_init_verbs[] = { + /* mute analog mix */ + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, + /* select ADCs - front-mic */ + {0x0c, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0d, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0e, AC_VERB_SET_CONNECT_SEL, 0x1}, + /* ADCs; muted */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + + { } +}; + +static struct hda_verb ad1988_spdif_init_verbs[] = { + /* SPDIF out sel */ + {0x02, AC_VERB_SET_CONNECT_SEL, 0x0}, /* PCM */ + {0x0b, AC_VERB_SET_CONNECT_SEL, 0x0}, /* ADC1 */ + {0x1d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x1d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + /* SPDIF out pin */ + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE | 0x27}, /* 0dB */ + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0) | 0x17}, /* 0dB */ + + { } +}; +/* + * verbs for 3stack (+dig) + */ +static struct hda_verb ad1988_3stack_ch2_init[] = { + /* set port-C to line-in */ + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { 0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN }, + /* set port-E to mic-in */ + { 0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { 0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80 }, + { } /* end */ +}; + +static struct hda_verb ad1988_3stack_ch6_init[] = { + /* set port-C to surround out */ + { 0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + /* set port-E to CLFE out */ + { 0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + { } /* end */ +}; + +static struct hda_channel_mode ad1988_3stack_modes[2] = { + { 2, ad1988_3stack_ch2_init }, + { 6, ad1988_3stack_ch6_init }, +}; + +static struct hda_verb ad1988_3stack_init_verbs[] = { + /* Front, Surround, CLFE, side DAC; unmute as default */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Port-A front headphon path */ + {0x37, AC_VERB_SET_CONNECT_SEL, 0x01}, /* DAC1:04h */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* Port-D line-out path */ + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Mono out path */ + {0x36, AC_VERB_SET_CONNECT_SEL, 0x1}, /* DAC1:04h */ + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, 0xb01f}, /* unmute, 0dB */ + /* Port-B front mic-in path */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x39, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* Port-C line-in/surround path - 6ch mode as default */ + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x3a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x31, AC_VERB_SET_CONNECT_SEL, 0x0}, /* output sel: DAC 0x05 */ + {0x33, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* Port-E mic-in/CLFE path - 6ch mode as default */ + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x3c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x32, AC_VERB_SET_CONNECT_SEL, 0x1}, /* output sel: DAC 0x0a */ + {0x34, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* mute analog mix */ + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, + /* select ADCs - front-mic */ + {0x0c, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0d, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0e, AC_VERB_SET_CONNECT_SEL, 0x1}, + /* ADCs; muted */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + { } +}; + +/* + * verbs for laptop mode (+dig) + */ +static struct hda_verb ad1988_laptop_hp_on[] = { + /* unmute port-A and mute port-D */ + { 0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + { 0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { } /* end */ +}; +static struct hda_verb ad1988_laptop_hp_off[] = { + /* mute port-A and unmute port-D */ + { 0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { 0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + { } /* end */ +}; + +#define AD1988_HP_EVENT 0x01 + +static struct hda_verb ad1988_laptop_init_verbs[] = { + /* Front, Surround, CLFE, side DAC; unmute as default */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Port-A front headphon path */ + {0x37, AC_VERB_SET_CONNECT_SEL, 0x01}, /* DAC1:04h */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* unsolicited event for pin-sense */ + {0x11, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | AD1988_HP_EVENT }, + /* Port-D line-out path + EAPD */ + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x29, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x12, AC_VERB_SET_EAPD_BTLENABLE, 0x00}, /* EAPD-off */ + /* Mono out path */ + {0x36, AC_VERB_SET_CONNECT_SEL, 0x1}, /* DAC1:04h */ + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x1e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, 0xb01f}, /* unmute, 0dB */ + /* Port-B mic-in path */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x39, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* Port-C docking station - try to output */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x3a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x33, AC_VERB_SET_CONNECT_SEL, 0x0}, + /* mute analog mix */ + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, + {0x20, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, + /* select ADCs - mic */ + {0x0c, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0d, AC_VERB_SET_CONNECT_SEL, 0x1}, + {0x0e, AC_VERB_SET_CONNECT_SEL, 0x1}, + /* ADCs; muted */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + { } +}; + +static void ad1988_laptop_unsol_event(struct hda_codec *codec, unsigned int res) +{ + if ((res >> 26) != AD1988_HP_EVENT) + return; + if (snd_hda_codec_read(codec, 0x11, 0, AC_VERB_GET_PIN_SENSE, 0) & (1 << 31)) + snd_hda_sequence_write(codec, ad1988_laptop_hp_on); + else + snd_hda_sequence_write(codec, ad1988_laptop_hp_off); +} + + +/* + * Automatic parse of I/O pins from the BIOS configuration + */ + +#define NUM_CONTROL_ALLOC 32 +#define NUM_VERB_ALLOC 32 + +enum { + AD_CTL_WIDGET_VOL, + AD_CTL_WIDGET_MUTE, + AD_CTL_BIND_MUTE, +}; +static struct snd_kcontrol_new ad1988_control_templates[] = { + HDA_CODEC_VOLUME(NULL, 0, 0, 0), + HDA_CODEC_MUTE(NULL, 0, 0, 0), + HDA_BIND_MUTE(NULL, 0, 0, 0), +}; + +/* add dynamic controls */ +static int add_control(struct ad198x_spec *spec, int type, const char *name, + unsigned long val) +{ + struct snd_kcontrol_new *knew; + + if (spec->num_kctl_used >= spec->num_kctl_alloc) { + int num = spec->num_kctl_alloc + NUM_CONTROL_ALLOC; + + knew = kcalloc(num + 1, sizeof(*knew), GFP_KERNEL); /* array + terminator */ + if (! knew) + return -ENOMEM; + if (spec->kctl_alloc) { + memcpy(knew, spec->kctl_alloc, sizeof(*knew) * spec->num_kctl_alloc); + kfree(spec->kctl_alloc); + } + spec->kctl_alloc = knew; + spec->num_kctl_alloc = num; + } + + knew = &spec->kctl_alloc[spec->num_kctl_used]; + *knew = ad1988_control_templates[type]; + knew->name = kstrdup(name, GFP_KERNEL); + if (! knew->name) + return -ENOMEM; + knew->private_value = val; + spec->num_kctl_used++; return 0; } -static void ad1986a_free(struct hda_codec *codec) +#define AD1988_PIN_CD_NID 0x18 +#define AD1988_PIN_BEEP_NID 0x10 + +static hda_nid_t ad1988_mixer_nids[8] = { + /* A B C D E F G H */ + 0x22, 0x2b, 0x2c, 0x29, 0x26, 0x2a, 0x27, 0x28 +}; + +static inline hda_nid_t ad1988_idx_to_dac(struct hda_codec *codec, int idx) { - kfree(codec->spec); + static hda_nid_t idx_to_dac[8] = { + /* A B C D E F G H */ + 0x04, 0x06, 0x05, 0x04, 0x0a, 0x06, 0x05, 0x0a + }; + static hda_nid_t idx_to_dac_rev2[8] = { + /* A B C D E F G H */ + 0x04, 0x05, 0x0a, 0x04, 0x06, 0x05, 0x0a, 0x06 + }; + if (is_rev2(codec)) + return idx_to_dac_rev2[idx]; + else + return idx_to_dac[idx]; } -#ifdef CONFIG_PM -static int ad1986a_resume(struct hda_codec *codec) +static hda_nid_t ad1988_boost_nids[8] = { + 0x38, 0x39, 0x3a, 0x3d, 0x3c, 0x3b, 0, 0 +}; + +static int ad1988_pin_idx(hda_nid_t nid) { - ad1986a_init(codec); - snd_hda_resume_ctls(codec, ad1986a_mixers); - snd_hda_resume_spdif_out(codec); + static hda_nid_t ad1988_io_pins[8] = { + 0x11, 0x14, 0x15, 0x12, 0x17, 0x16, 0x24, 0x25 + }; + int i; + for (i = 0; i < ARRAY_SIZE(ad1988_io_pins); i++) + if (ad1988_io_pins[i] == nid) + return i; + return 0; /* should be -1 */ +} + +static int ad1988_pin_to_loopback_idx(hda_nid_t nid) +{ + static int loopback_idx[8] = { + 2, 0, 1, 3, 4, 5, 1, 4 + }; + switch (nid) { + case AD1988_PIN_CD_NID: + return 6; + default: + return loopback_idx[ad1988_pin_idx(nid)]; + } +} + +static int ad1988_pin_to_adc_idx(hda_nid_t nid) +{ + static int adc_idx[8] = { + 0, 1, 2, 8, 4, 3, 6, 7 + }; + switch (nid) { + case AD1988_PIN_CD_NID: + return 5; + default: + return adc_idx[ad1988_pin_idx(nid)]; + } +} + +/* fill in the dac_nids table from the parsed pin configuration */ +static int ad1988_auto_fill_dac_nids(struct hda_codec *codec, + const struct auto_pin_cfg *cfg) +{ + struct ad198x_spec *spec = codec->spec; + int i, idx; + + spec->multiout.dac_nids = spec->private_dac_nids; + + /* check the pins hardwired to audio widget */ + for (i = 0; i < cfg->line_outs; i++) { + idx = ad1988_pin_idx(cfg->line_out_pins[i]); + spec->multiout.dac_nids[i] = ad1988_idx_to_dac(codec, idx); + } + spec->multiout.num_dacs = cfg->line_outs; return 0; } -#endif -static struct hda_codec_ops ad1986a_patch_ops = { - .build_controls = ad1986a_build_controls, - .build_pcms = ad1986a_build_pcms, - .init = ad1986a_init, - .free = ad1986a_free, -#ifdef CONFIG_PM - .resume = ad1986a_resume, -#endif +/* add playback controls from the parsed DAC table */ +static int ad1988_auto_create_multi_out_ctls(struct ad198x_spec *spec, + const struct auto_pin_cfg *cfg) +{ + char name[32]; + static const char *chname[4] = { "Front", "Surround", NULL /*CLFE*/, "Side" }; + hda_nid_t nid; + int i, err; + + for (i = 0; i < cfg->line_outs; i++) { + hda_nid_t dac = spec->multiout.dac_nids[i]; + if (! dac) + continue; + nid = ad1988_mixer_nids[ad1988_pin_idx(cfg->line_out_pins[i])]; + if (i == 2) { + /* Center/LFE */ + err = add_control(spec, AD_CTL_WIDGET_VOL, + "Center Playback Volume", + HDA_COMPOSE_AMP_VAL(dac, 1, 0, HDA_OUTPUT)); + if (err < 0) + return err; + err = add_control(spec, AD_CTL_WIDGET_VOL, + "LFE Playback Volume", + HDA_COMPOSE_AMP_VAL(dac, 2, 0, HDA_OUTPUT)); + if (err < 0) + return err; + err = add_control(spec, AD_CTL_BIND_MUTE, + "Center Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 1, 2, HDA_INPUT)); + if (err < 0) + return err; + err = add_control(spec, AD_CTL_BIND_MUTE, + "LFE Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 2, 2, HDA_INPUT)); + if (err < 0) + return err; + } else { + sprintf(name, "%s Playback Volume", chname[i]); + err = add_control(spec, AD_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(dac, 3, 0, HDA_OUTPUT)); + if (err < 0) + return err; + sprintf(name, "%s Playback Switch", chname[i]); + err = add_control(spec, AD_CTL_BIND_MUTE, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 2, HDA_INPUT)); + if (err < 0) + return err; + } + } + return 0; +} + +/* add playback controls for speaker and HP outputs */ +static int ad1988_auto_create_extra_out(struct hda_codec *codec, hda_nid_t pin, + const char *pfx) +{ + struct ad198x_spec *spec = codec->spec; + hda_nid_t nid; + int idx, err; + char name[32]; + + if (! pin) + return 0; + + idx = ad1988_pin_idx(pin); + nid = ad1988_idx_to_dac(codec, idx); + /* specify the DAC as the extra output */ + if (! spec->multiout.hp_nid) + spec->multiout.hp_nid = nid; + else + spec->multiout.extra_out_nid[0] = nid; + /* control HP volume/switch on the output mixer amp */ + sprintf(name, "%s Playback Volume", pfx); + if ((err = add_control(spec, AD_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + nid = ad1988_mixer_nids[idx]; + sprintf(name, "%s Playback Switch", pfx); + if ((err = add_control(spec, AD_CTL_BIND_MUTE, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 2, HDA_INPUT))) < 0) + return err; + return 0; +} + +/* create input playback/capture controls for the given pin */ +static int new_analog_input(struct ad198x_spec *spec, hda_nid_t pin, + const char *ctlname, int boost) +{ + char name[32]; + int err, idx; + + sprintf(name, "%s Playback Volume", ctlname); + idx = ad1988_pin_to_loopback_idx(pin); + if ((err = add_control(spec, AD_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(0x20, 3, idx, HDA_INPUT))) < 0) + return err; + sprintf(name, "%s Playback Switch", ctlname); + if ((err = add_control(spec, AD_CTL_WIDGET_MUTE, name, + HDA_COMPOSE_AMP_VAL(0x20, 3, idx, HDA_INPUT))) < 0) + return err; + if (boost) { + hda_nid_t bnid; + idx = ad1988_pin_idx(pin); + bnid = ad1988_boost_nids[idx]; + if (bnid) { + sprintf(name, "%s Boost", ctlname); + return add_control(spec, AD_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(bnid, 3, idx, HDA_OUTPUT)); + + } + } + return 0; +} + +/* create playback/capture controls for input pins */ +static int ad1988_auto_create_analog_input_ctls(struct ad198x_spec *spec, + const struct auto_pin_cfg *cfg) +{ + struct hda_input_mux *imux = &spec->private_imux; + int i, err; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + err = new_analog_input(spec, cfg->input_pins[i], + auto_pin_cfg_labels[i], + i <= AUTO_PIN_FRONT_MIC); + if (err < 0) + return err; + imux->items[imux->num_items].label = auto_pin_cfg_labels[i]; + imux->items[imux->num_items].index = ad1988_pin_to_adc_idx(cfg->input_pins[i]); + imux->num_items++; + } + imux->items[imux->num_items].label = "Mix"; + imux->items[imux->num_items].index = 9; + imux->num_items++; + + if ((err = add_control(spec, AD_CTL_WIDGET_VOL, + "Analog Mix Playback Volume", + HDA_COMPOSE_AMP_VAL(0x21, 3, 0x0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, AD_CTL_WIDGET_MUTE, + "Analog Mix Playback Switch", + HDA_COMPOSE_AMP_VAL(0x21, 3, 0x0, HDA_OUTPUT))) < 0) + return err; + + return 0; +} + +static void ad1988_auto_set_output_and_unmute(struct hda_codec *codec, + hda_nid_t nid, int pin_type, + int dac_idx) +{ + /* set as output */ + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, pin_type); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE); + switch (nid) { + case 0x11: /* port-A - DAC 04 */ + snd_hda_codec_write(codec, 0x37, 0, AC_VERB_SET_CONNECT_SEL, 0x01); + break; + case 0x14: /* port-B - DAC 06 */ + snd_hda_codec_write(codec, 0x30, 0, AC_VERB_SET_CONNECT_SEL, 0x02); + break; + case 0x15: /* port-C - DAC 05 */ + snd_hda_codec_write(codec, 0x31, 0, AC_VERB_SET_CONNECT_SEL, 0x00); + break; + case 0x17: /* port-E - DAC 0a */ + snd_hda_codec_write(codec, 0x32, 0, AC_VERB_SET_CONNECT_SEL, 0x01); + break; + case 0x13: /* mono - DAC 04 */ + snd_hda_codec_write(codec, 0x36, 0, AC_VERB_SET_CONNECT_SEL, 0x01); + break; + } +} + +static void ad1988_auto_init_multi_out(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + int i; + + for (i = 0; i < spec->autocfg.line_outs; i++) { + hda_nid_t nid = spec->autocfg.line_out_pins[i]; + ad1988_auto_set_output_and_unmute(codec, nid, PIN_OUT, i); + } +} + +static void ad1988_auto_init_extra_out(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + hda_nid_t pin; + + pin = spec->autocfg.speaker_pins[0]; + if (pin) /* connect to front */ + ad1988_auto_set_output_and_unmute(codec, pin, PIN_OUT, 0); + pin = spec->autocfg.hp_pin; + if (pin) /* connect to front */ + ad1988_auto_set_output_and_unmute(codec, pin, PIN_HP, 0); +} + +static void ad1988_auto_init_analog_input(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + int i, idx; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + hda_nid_t nid = spec->autocfg.input_pins[i]; + if (! nid) + continue; + switch (nid) { + case 0x15: /* port-C */ + snd_hda_codec_write(codec, 0x33, 0, AC_VERB_SET_CONNECT_SEL, 0x0); + break; + case 0x17: /* port-E */ + snd_hda_codec_write(codec, 0x34, 0, AC_VERB_SET_CONNECT_SEL, 0x0); + break; + } + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + i <= AUTO_PIN_FRONT_MIC ? PIN_VREF80 : PIN_IN); + if (nid != AD1988_PIN_CD_NID) + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_MUTE); + idx = ad1988_pin_idx(nid); + if (ad1988_boost_nids[idx]) + snd_hda_codec_write(codec, ad1988_boost_nids[idx], 0, + AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_ZERO); + } +} + +/* parse the BIOS configuration and set up the alc_spec */ +/* return 1 if successful, 0 if the proper config is not found, or a negative error code */ +static int ad1988_parse_auto_config(struct hda_codec *codec) +{ + struct ad198x_spec *spec = codec->spec; + int err; + + if ((err = snd_hda_parse_pin_def_config(codec, &spec->autocfg, NULL)) < 0) + return err; + if ((err = ad1988_auto_fill_dac_nids(codec, &spec->autocfg)) < 0) + return err; + if (! spec->autocfg.line_outs) + return 0; /* can't find valid BIOS pin config */ + if ((err = ad1988_auto_create_multi_out_ctls(spec, &spec->autocfg)) < 0 || + (err = ad1988_auto_create_extra_out(codec, + spec->autocfg.speaker_pins[0], + "Speaker")) < 0 || + (err = ad1988_auto_create_extra_out(codec, spec->autocfg.hp_pin, + "Headphone")) < 0 || + (err = ad1988_auto_create_analog_input_ctls(spec, &spec->autocfg)) < 0) + return err; + + spec->multiout.max_channels = spec->multiout.num_dacs * 2; + + if (spec->autocfg.dig_out_pin) + spec->multiout.dig_out_nid = AD1988_SPDIF_OUT; + if (spec->autocfg.dig_in_pin) + spec->dig_in_nid = AD1988_SPDIF_IN; + + if (spec->kctl_alloc) + spec->mixers[spec->num_mixers++] = spec->kctl_alloc; + + spec->init_verbs[spec->num_init_verbs++] = ad1988_6stack_init_verbs; + + spec->input_mux = &spec->private_imux; + + return 1; +} + +/* init callback for auto-configuration model -- overriding the default init */ +static int ad1988_auto_init(struct hda_codec *codec) +{ + ad198x_init(codec); + ad1988_auto_init_multi_out(codec); + ad1988_auto_init_extra_out(codec); + ad1988_auto_init_analog_input(codec); + return 0; +} + + +/* + */ + +static struct hda_board_config ad1988_cfg_tbl[] = { + { .modelname = "6stack", .config = AD1988_6STACK }, + { .modelname = "6stack-dig", .config = AD1988_6STACK_DIG }, + { .modelname = "3stack", .config = AD1988_3STACK }, + { .modelname = "3stack-dig", .config = AD1988_3STACK_DIG }, + { .modelname = "laptop", .config = AD1988_LAPTOP }, + { .modelname = "laptop-dig", .config = AD1988_LAPTOP_DIG }, + { .modelname = "auto", .config = AD1988_AUTO }, + {} }; -static int patch_ad1986a(struct hda_codec *codec) +static int patch_ad1988(struct hda_codec *codec) { - struct ad1986a_spec *spec; + struct ad198x_spec *spec; + int board_config; - spec = kcalloc(1, sizeof(*spec), GFP_KERNEL); + spec = kzalloc(sizeof(*spec), GFP_KERNEL); if (spec == NULL) return -ENOMEM; - init_MUTEX(&spec->amp_mutex); + mutex_init(&spec->amp_mutex); codec->spec = spec; - spec->multiout.max_channels = 6; - spec->multiout.num_dacs = ARRAY_SIZE(ad1986a_dac_nids); - spec->multiout.dac_nids = ad1986a_dac_nids; - spec->multiout.dig_out_nid = AD1986A_SPDIF_OUT; + if (is_rev2(codec)) + snd_printk(KERN_INFO "patch_analog: AD1988A rev.2 is detected, enable workarounds\n"); + + board_config = snd_hda_check_board_config(codec, ad1988_cfg_tbl); + if (board_config < 0 || board_config >= AD1988_MODEL_LAST) { + printk(KERN_INFO "hda_codec: Unknown model for AD1988, trying auto-probe from BIOS...\n"); + board_config = AD1988_AUTO; + } - codec->patch_ops = ad1986a_patch_ops; + if (board_config == AD1988_AUTO) { + /* automatic parse from the BIOS config */ + int err = ad1988_parse_auto_config(codec); + if (err < 0) { + ad198x_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using 6-stack mode...\n"); + board_config = AD1988_6STACK; + } + } + + switch (board_config) { + case AD1988_6STACK: + case AD1988_6STACK_DIG: + spec->multiout.max_channels = 8; + spec->multiout.num_dacs = 4; + if (is_rev2(codec)) + spec->multiout.dac_nids = ad1988_6stack_dac_nids_rev2; + else + spec->multiout.dac_nids = ad1988_6stack_dac_nids; + spec->input_mux = &ad1988_6stack_capture_source; + spec->num_mixers = 2; + if (is_rev2(codec)) + spec->mixers[0] = ad1988_6stack_mixers1_rev2; + else + spec->mixers[0] = ad1988_6stack_mixers1; + spec->mixers[1] = ad1988_6stack_mixers2; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1988_6stack_init_verbs; + if (board_config == AD1988_6STACK_DIG) { + spec->multiout.dig_out_nid = AD1988_SPDIF_OUT; + spec->dig_in_nid = AD1988_SPDIF_IN; + } + break; + case AD1988_3STACK: + case AD1988_3STACK_DIG: + spec->multiout.max_channels = 6; + spec->multiout.num_dacs = 3; + if (is_rev2(codec)) + spec->multiout.dac_nids = ad1988_3stack_dac_nids_rev2; + else + spec->multiout.dac_nids = ad1988_3stack_dac_nids; + spec->input_mux = &ad1988_6stack_capture_source; + spec->channel_mode = ad1988_3stack_modes; + spec->num_channel_mode = ARRAY_SIZE(ad1988_3stack_modes); + spec->num_mixers = 2; + if (is_rev2(codec)) + spec->mixers[0] = ad1988_3stack_mixers1_rev2; + else + spec->mixers[0] = ad1988_3stack_mixers1; + spec->mixers[1] = ad1988_3stack_mixers2; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1988_3stack_init_verbs; + if (board_config == AD1988_3STACK_DIG) + spec->multiout.dig_out_nid = AD1988_SPDIF_OUT; + break; + case AD1988_LAPTOP: + case AD1988_LAPTOP_DIG: + spec->multiout.max_channels = 2; + spec->multiout.num_dacs = 1; + spec->multiout.dac_nids = ad1988_3stack_dac_nids; + spec->input_mux = &ad1988_laptop_capture_source; + spec->num_mixers = 1; + spec->mixers[0] = ad1988_laptop_mixers; + spec->num_init_verbs = 1; + spec->init_verbs[0] = ad1988_laptop_init_verbs; + if (board_config == AD1988_LAPTOP_DIG) + spec->multiout.dig_out_nid = AD1988_SPDIF_OUT; + break; + } + + spec->num_adc_nids = ARRAY_SIZE(ad1988_adc_nids); + spec->adc_nids = ad1988_adc_nids; + spec->capsrc_nids = ad1988_capsrc_nids; + spec->mixers[spec->num_mixers++] = ad1988_capture_mixers; + spec->init_verbs[spec->num_init_verbs++] = ad1988_capture_init_verbs; + if (spec->multiout.dig_out_nid) { + spec->mixers[spec->num_mixers++] = ad1988_spdif_out_mixers; + spec->init_verbs[spec->num_init_verbs++] = ad1988_spdif_init_verbs; + } + if (spec->dig_in_nid) + spec->mixers[spec->num_mixers++] = ad1988_spdif_in_mixers; + + codec->patch_ops = ad198x_patch_ops; + switch (board_config) { + case AD1988_AUTO: + codec->patch_ops.init = ad1988_auto_init; + break; + case AD1988_LAPTOP: + case AD1988_LAPTOP_DIG: + codec->patch_ops.unsol_event = ad1988_laptop_unsol_event; + break; + } return 0; } + /* * patch entries */ struct hda_codec_preset snd_hda_preset_analog[] = { + { .id = 0x11d41981, .name = "AD1981", .patch = patch_ad1981 }, + { .id = 0x11d41983, .name = "AD1983", .patch = patch_ad1983 }, { .id = 0x11d41986, .name = "AD1986A", .patch = patch_ad1986a }, + { .id = 0x11d41988, .name = "AD1988", .patch = patch_ad1988 }, + { .id = 0x11d4198b, .name = "AD1988B", .patch = patch_ad1988 }, {} /* terminator */ }; diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c index 17c506242..94cf292d9 100644 --- a/sound/pci/hda/patch_realtek.c +++ b/sound/pci/hda/patch_realtek.c @@ -3,8 +3,10 @@ * * HD audio interface patch for ALC 260/880/882 codecs * - * Copyright (c) 2004 PeiSen Hou + * Copyright (c) 2004 Kailang Yang + * PeiSen Hou * Takashi Iwai + * Jonathan Woithe * * This driver is free software; you can redistribute it and/or modify * it under the terms of the GNU General Public License as published by @@ -33,115 +35,166 @@ /* ALC880 board config type */ enum { - ALC880_MINIMAL, ALC880_3ST, ALC880_3ST_DIG, ALC880_5ST, ALC880_5ST_DIG, ALC880_W810, + ALC880_Z71V, + ALC880_6ST, + ALC880_6ST_DIG, + ALC880_F1734, + ALC880_ASUS, + ALC880_ASUS_DIG, + ALC880_ASUS_W1V, + ALC880_ASUS_DIG2, + ALC880_UNIWILL_DIG, + ALC880_CLEVO, + ALC880_TCL_S700, + ALC880_LG, + ALC880_LG_LW, +#ifdef CONFIG_SND_DEBUG + ALC880_TEST, +#endif + ALC880_AUTO, + ALC880_MODEL_LAST /* last tag */ +}; + +/* ALC260 models */ +enum { + ALC260_BASIC, + ALC260_HP, + ALC260_HP_3013, + ALC260_FUJITSU_S702X, + ALC260_ACER, +#ifdef CONFIG_SND_DEBUG + ALC260_TEST, +#endif + ALC260_AUTO, + ALC260_MODEL_LAST /* last tag */ +}; + +/* ALC262 models */ +enum { + ALC262_BASIC, + ALC262_FUJITSU, + ALC262_AUTO, + ALC262_MODEL_LAST /* last tag */ }; +/* ALC861 models */ +enum { + ALC861_3ST, + ALC861_3ST_DIG, + ALC861_6ST_DIG, + ALC861_AUTO, + ALC861_MODEL_LAST, +}; + +/* ALC882 models */ +enum { + ALC882_3ST_DIG, + ALC882_6ST_DIG, + ALC882_AUTO, + ALC882_MODEL_LAST, +}; + +/* for GPIO Poll */ +#define GPIO_MASK 0x03 + struct alc_spec { /* codec parameterization */ - unsigned int front_panel: 1; - - snd_kcontrol_new_t* mixers[2]; + struct snd_kcontrol_new *mixers[5]; /* mixer arrays */ unsigned int num_mixers; - struct hda_verb *init_verbs; + const struct hda_verb *init_verbs[5]; /* initialization verbs + * don't forget NULL termination! + */ + unsigned int num_init_verbs; - char* stream_name_analog; + char *stream_name_analog; /* analog PCM stream */ struct hda_pcm_stream *stream_analog_playback; struct hda_pcm_stream *stream_analog_capture; - char* stream_name_digital; + char *stream_name_digital; /* digital PCM stream */ struct hda_pcm_stream *stream_digital_playback; struct hda_pcm_stream *stream_digital_capture; /* playback */ - struct hda_multi_out multiout; + struct hda_multi_out multiout; /* playback set-up + * max_channels, dacs must be set + * dig_out_nid and hp_nid are optional + */ /* capture */ unsigned int num_adc_nids; hda_nid_t *adc_nids; - hda_nid_t dig_in_nid; + hda_nid_t dig_in_nid; /* digital-in NID; optional */ /* capture source */ + unsigned int num_mux_defs; const struct hda_input_mux *input_mux; unsigned int cur_mux[3]; /* channel model */ - const struct alc_channel_mode *channel_mode; + const struct hda_channel_mode *channel_mode; int num_channel_mode; /* PCM information */ - struct hda_pcm pcm_rec[2]; -}; - -/* DAC/ADC assignment */ - -static hda_nid_t alc880_dac_nids[4] = { - /* front, rear, clfe, rear_surr */ - 0x02, 0x05, 0x04, 0x03 -}; - -static hda_nid_t alc880_w810_dac_nids[3] = { - /* front, rear/surround, clfe */ - 0x02, 0x03, 0x04 -}; - -static hda_nid_t alc880_adc_nids[3] = { - /* ADC0-2 */ - 0x07, 0x08, 0x09, -}; - -#define ALC880_DIGOUT_NID 0x06 -#define ALC880_DIGIN_NID 0x0a - -static hda_nid_t alc260_dac_nids[1] = { - /* front */ - 0x02, -}; - -static hda_nid_t alc260_adc_nids[2] = { - /* ADC0-1 */ - 0x04, 0x05, + struct hda_pcm pcm_rec[3]; /* used in alc_build_pcms() */ + + /* dynamic controls, init_verbs and input_mux */ + struct auto_pin_cfg autocfg; + unsigned int num_kctl_alloc, num_kctl_used; + struct snd_kcontrol_new *kctl_alloc; + struct hda_input_mux private_imux; + hda_nid_t private_dac_nids[5]; + + /* hooks */ + void (*init_hook)(struct hda_codec *codec); + void (*unsol_event)(struct hda_codec *codec, unsigned int res); + + /* for pin sensing */ + unsigned int sense_updated: 1; + unsigned int jack_present: 1; }; -#define ALC260_DIGOUT_NID 0x03 -#define ALC260_DIGIN_NID 0x06 - -static struct hda_input_mux alc880_capture_source = { - .num_items = 4, - .items = { - { "Mic", 0x0 }, - { "Front Mic", 0x3 }, - { "Line", 0x2 }, - { "CD", 0x4 }, - }, +/* + * configuration template - to be copied to the spec instance + */ +struct alc_config_preset { + struct snd_kcontrol_new *mixers[5]; /* should be identical size with spec */ + const struct hda_verb *init_verbs[5]; + unsigned int num_dacs; + hda_nid_t *dac_nids; + hda_nid_t dig_out_nid; /* optional */ + hda_nid_t hp_nid; /* optional */ + unsigned int num_adc_nids; + hda_nid_t *adc_nids; + hda_nid_t dig_in_nid; + unsigned int num_channel_mode; + const struct hda_channel_mode *channel_mode; + unsigned int num_mux_defs; + const struct hda_input_mux *input_mux; + void (*unsol_event)(struct hda_codec *, unsigned int); + void (*init_hook)(struct hda_codec *); }; -static struct hda_input_mux alc260_capture_source = { - .num_items = 4, - .items = { - { "Mic", 0x0 }, - { "Front Mic", 0x1 }, - { "Line", 0x2 }, - { "CD", 0x4 }, - }, -}; /* * input MUX handling */ -static int alc_mux_enum_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo) +static int alc_mux_enum_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); struct alc_spec *spec = codec->spec; - return snd_hda_input_mux_info(spec->input_mux, uinfo); + unsigned int mux_idx = snd_ctl_get_ioffidx(kcontrol, &uinfo->id); + if (mux_idx >= spec->num_mux_defs) + mux_idx = 0; + return snd_hda_input_mux_info(&spec->input_mux[mux_idx], uinfo); } -static int alc_mux_enum_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int alc_mux_enum_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); struct alc_spec *spec = codec->spec; @@ -151,205 +204,392 @@ static int alc_mux_enum_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucon return 0; } -static int alc_mux_enum_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int alc_mux_enum_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); struct alc_spec *spec = codec->spec; unsigned int adc_idx = snd_ctl_get_ioffidx(kcontrol, &ucontrol->id); - return snd_hda_input_mux_put(codec, spec->input_mux, ucontrol, + unsigned int mux_idx = adc_idx >= spec->num_mux_defs ? 0 : adc_idx; + return snd_hda_input_mux_put(codec, &spec->input_mux[mux_idx], ucontrol, spec->adc_nids[adc_idx], &spec->cur_mux[adc_idx]); } + /* * channel mode setting */ -struct alc_channel_mode { - int channels; - const struct hda_verb *sequence; -}; - +static int alc_ch_mode_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct alc_spec *spec = codec->spec; + return snd_hda_ch_mode_info(codec, uinfo, spec->channel_mode, + spec->num_channel_mode); +} -/* - * channel source setting (2/6 channel selection for 3-stack) - */ +static int alc_ch_mode_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct alc_spec *spec = codec->spec; + return snd_hda_ch_mode_get(codec, ucontrol, spec->channel_mode, + spec->num_channel_mode, spec->multiout.max_channels); +} -/* - * set the path ways for 2 channel output - * need to set the codec line out and mic 1 pin widgets to inputs - */ -static struct hda_verb alc880_threestack_ch2_init[] = { - /* set pin widget 1Ah (line in) for input */ - { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, - /* set pin widget 18h (mic1) for input, for mic also enable the vref */ - { 0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, - /* mute the output for Line In PW */ - { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080 }, - /* mute for Mic1 PW */ - { 0x18, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080 }, - { } /* end */ -}; +static int alc_ch_mode_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + struct alc_spec *spec = codec->spec; + return snd_hda_ch_mode_put(codec, ucontrol, spec->channel_mode, + spec->num_channel_mode, &spec->multiout.max_channels); +} /* - * 6ch mode - * need to set the codec line out and mic 1 pin widgets to outputs + * Control the mode of pin widget settings via the mixer. "pc" is used + * instead of "%" to avoid consequences of accidently treating the % as + * being part of a format specifier. Maximum allowed length of a value is + * 63 characters plus NULL terminator. + * + * Note: some retasking pin complexes seem to ignore requests for input + * states other than HiZ (eg: PIN_VREFxx) and revert to HiZ if any of these + * are requested. Therefore order this list so that this behaviour will not + * cause problems when mixer clients move through the enum sequentially. + * NIDs 0x0f and 0x10 have been observed to have this behaviour as of + * March 2006. */ -static struct hda_verb alc880_threestack_ch6_init[] = { - /* set pin widget 1Ah (line in) for output */ - { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, - /* set pin widget 18h (mic1) for output */ - { 0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, - /* unmute the output for Line In PW */ - { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000 }, - /* unmute for Mic1 PW */ - { 0x18, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000 }, - /* for rear channel output using Line In 1 - * set select widget connection (nid = 0x12) - to summer node - * for rear NID = 0x0f...offset 3 in connection list - */ - { 0x12, AC_VERB_SET_CONNECT_SEL, 0x3 }, - /* for Mic1 - retask for center/lfe */ - /* set select widget connection (nid = 0x10) - to summer node for - * front CLFE NID = 0x0e...offset 2 in connection list - */ - { 0x10, AC_VERB_SET_CONNECT_SEL, 0x2 }, - { } /* end */ +static char *alc_pin_mode_names[] = { + "Mic 50pc bias", "Mic 80pc bias", + "Line in", "Line out", "Headphone out", }; - -static struct alc_channel_mode alc880_threestack_modes[2] = { - { 2, alc880_threestack_ch2_init }, - { 6, alc880_threestack_ch6_init }, +static unsigned char alc_pin_mode_values[] = { + PIN_VREF50, PIN_VREF80, PIN_IN, PIN_OUT, PIN_HP, }; - - -/* - * channel source setting (6/8 channel selection for 5-stack) +/* The control can present all 5 options, or it can limit the options based + * in the pin being assumed to be exclusively an input or an output pin. In + * addition, "input" pins may or may not process the mic bias option + * depending on actual widget capability (NIDs 0x0f and 0x10 don't seem to + * accept requests for bias as of chip versions up to March 2006) and/or + * wiring in the computer. */ - -/* set the path ways for 6 channel output - * need to set the codec line out and mic 1 pin widgets to inputs +#define ALC_PIN_DIR_IN 0x00 +#define ALC_PIN_DIR_OUT 0x01 +#define ALC_PIN_DIR_INOUT 0x02 +#define ALC_PIN_DIR_IN_NOMICBIAS 0x03 +#define ALC_PIN_DIR_INOUT_NOMICBIAS 0x04 + +/* Info about the pin modes supported by the different pin direction modes. + * For each direction the minimum and maximum values are given. */ -static struct hda_verb alc880_fivestack_ch6_init[] = { - /* set pin widget 1Ah (line in) for input */ - { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, - /* mute the output for Line In PW */ - { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080 }, - { } /* end */ +static signed char alc_pin_mode_dir_info[5][2] = { + { 0, 2 }, /* ALC_PIN_DIR_IN */ + { 3, 4 }, /* ALC_PIN_DIR_OUT */ + { 0, 4 }, /* ALC_PIN_DIR_INOUT */ + { 2, 2 }, /* ALC_PIN_DIR_IN_NOMICBIAS */ + { 2, 4 }, /* ALC_PIN_DIR_INOUT_NOMICBIAS */ }; +#define alc_pin_mode_min(_dir) (alc_pin_mode_dir_info[_dir][0]) +#define alc_pin_mode_max(_dir) (alc_pin_mode_dir_info[_dir][1]) +#define alc_pin_mode_n_items(_dir) \ + (alc_pin_mode_max(_dir)-alc_pin_mode_min(_dir)+1) -/* need to set the codec line out and mic 1 pin widgets to outputs */ -static struct hda_verb alc880_fivestack_ch8_init[] = { - /* set pin widget 1Ah (line in) for output */ - { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, - /* unmute the output for Line In PW */ - { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000 }, - /* output for surround channel output using Line In 1 */ - /* set select widget connection (nid = 0x12) - to summer node - * for surr_rear NID = 0x0d...offset 1 in connection list - */ - { 0x12, AC_VERB_SET_CONNECT_SEL, 0x1 }, - { } /* end */ -}; +static int alc_pin_mode_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + unsigned int item_num = uinfo->value.enumerated.item; + unsigned char dir = (kcontrol->private_value >> 16) & 0xff; -static struct alc_channel_mode alc880_fivestack_modes[2] = { - { 6, alc880_fivestack_ch6_init }, - { 8, alc880_fivestack_ch8_init }, -}; + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = alc_pin_mode_n_items(dir); -/* - * channel source setting for W810 system - * - * W810 has rear IO for: - * Front (DAC 02) - * Surround (DAC 03) - * Center/LFE (DAC 04) - * Digital out (06) - * - * The system also has a pair of internal speakers, and a headphone jack. - * These are both connected to Line2 on the codec, hence to DAC 02. - * - * There is a variable resistor to control the speaker or headphone - * volume. This is a hardware-only device without a software API. - * - * Plugging headphones in will disable the internal speakers. This is - * implemented in hardware, not via the driver using jack sense. In - * a similar fashion, plugging into the rear socket marked "front" will - * disable both the speakers and headphones. - * - * For input, there's a microphone jack, and an "audio in" jack. - * These may not do anything useful with this driver yet, because I - * haven't setup any initialization verbs for these yet... - */ + if (item_numalc_pin_mode_max(dir)) + item_num = alc_pin_mode_min(dir); + strcpy(uinfo->value.enumerated.name, alc_pin_mode_names[item_num]); + return 0; +} -static struct alc_channel_mode alc880_w810_modes[1] = { - { 6, NULL } -}; +static int alc_pin_mode_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + unsigned int i; + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char dir = (kcontrol->private_value >> 16) & 0xff; + long *valp = ucontrol->value.integer.value; + unsigned int pinctl = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_PIN_WIDGET_CONTROL,0x00); + + /* Find enumerated value for current pinctl setting */ + i = alc_pin_mode_min(dir); + while (alc_pin_mode_values[i]!=pinctl && i<=alc_pin_mode_max(dir)) + i++; + *valp = i<=alc_pin_mode_max(dir)?i:alc_pin_mode_min(dir); + return 0; +} -/* - */ -static int alc880_ch_mode_info(snd_kcontrol_t *kcontrol, snd_ctl_elem_info_t *uinfo) +static int alc_pin_mode_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { + signed int change; struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct alc_spec *spec = codec->spec; + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char dir = (kcontrol->private_value >> 16) & 0xff; + long val = *ucontrol->value.integer.value; + unsigned int pinctl = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_PIN_WIDGET_CONTROL,0x00); + + if (valalc_pin_mode_max(dir)) + val = alc_pin_mode_min(dir); + + change = pinctl != alc_pin_mode_values[val]; + if (change) { + /* Set pin mode to that requested */ + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_PIN_WIDGET_CONTROL, + alc_pin_mode_values[val]); + + /* Also enable the retasking pin's input/output as required + * for the requested pin mode. Enum values of 2 or less are + * input modes. + * + * Dynamically switching the input/output buffers probably + * reduces noise slightly (particularly on input) so we'll + * do it. However, having both input and output buffers + * enabled simultaneously doesn't seem to be problematic if + * this turns out to be necessary in the future. + */ + if (val <= 2) { + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_MUTE); + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_AMP_GAIN_MUTE, + AMP_IN_UNMUTE(0)); + } else { + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_AMP_GAIN_MUTE, + AMP_IN_MUTE(0)); + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_UNMUTE); + } + } + return change; +} - snd_assert(spec->channel_mode, return -ENXIO); - uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; +#define ALC_PIN_MODE(xname, nid, dir) \ + { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, .index = 0, \ + .info = alc_pin_mode_info, \ + .get = alc_pin_mode_get, \ + .put = alc_pin_mode_put, \ + .private_value = nid | (dir<<16) } + +/* A switch control for ALC260 GPIO pins. Multiple GPIOs can be ganged + * together using a mask with more than one bit set. This control is + * currently used only by the ALC260 test model. At this stage they are not + * needed for any "production" models. + */ +#ifdef CONFIG_SND_DEBUG +static int alc_gpio_data_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN; uinfo->count = 1; - uinfo->value.enumerated.items = 2; - if (uinfo->value.enumerated.item >= 2) - uinfo->value.enumerated.item = 1; - sprintf(uinfo->value.enumerated.name, "%dch", - spec->channel_mode[uinfo->value.enumerated.item].channels); + uinfo->value.integer.min = 0; + uinfo->value.integer.max = 1; return 0; -} - -static int alc880_ch_mode_get(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +} +static int alc_gpio_data_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct alc_spec *spec = codec->spec; + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char mask = (kcontrol->private_value >> 16) & 0xff; + long *valp = ucontrol->value.integer.value; + unsigned int val = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_GPIO_DATA,0x00); - snd_assert(spec->channel_mode, return -ENXIO); - ucontrol->value.enumerated.item[0] = - (spec->multiout.max_channels == spec->channel_mode[0].channels) ? 0 : 1; + *valp = (val & mask) != 0; return 0; } +static int alc_gpio_data_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + signed int change; + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char mask = (kcontrol->private_value >> 16) & 0xff; + long val = *ucontrol->value.integer.value; + unsigned int gpio_data = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_GPIO_DATA,0x00); + + /* Set/unset the masked GPIO bit(s) as needed */ + change = (val==0?0:mask) != (gpio_data & mask); + if (val==0) + gpio_data &= ~mask; + else + gpio_data |= mask; + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_GPIO_DATA,gpio_data); + + return change; +} +#define ALC_GPIO_DATA_SWITCH(xname, nid, mask) \ + { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, .index = 0, \ + .info = alc_gpio_data_info, \ + .get = alc_gpio_data_get, \ + .put = alc_gpio_data_put, \ + .private_value = nid | (mask<<16) } +#endif /* CONFIG_SND_DEBUG */ + +/* A switch control to allow the enabling of the digital IO pins on the + * ALC260. This is incredibly simplistic; the intention of this control is + * to provide something in the test model allowing digital outputs to be + * identified if present. If models are found which can utilise these + * outputs a more complete mixer control can be devised for those models if + * necessary. + */ +#ifdef CONFIG_SND_DEBUG +static int alc_spdif_ctrl_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + uinfo->type = SNDRV_CTL_ELEM_TYPE_BOOLEAN; + uinfo->count = 1; + uinfo->value.integer.min = 0; + uinfo->value.integer.max = 1; + return 0; +} +static int alc_spdif_ctrl_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char mask = (kcontrol->private_value >> 16) & 0xff; + long *valp = ucontrol->value.integer.value; + unsigned int val = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_DIGI_CONVERT,0x00); -static int alc880_ch_mode_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) + *valp = (val & mask) != 0; + return 0; +} +static int alc_spdif_ctrl_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { + signed int change; struct hda_codec *codec = snd_kcontrol_chip(kcontrol); - struct alc_spec *spec = codec->spec; - int mode; + hda_nid_t nid = kcontrol->private_value & 0xffff; + unsigned char mask = (kcontrol->private_value >> 16) & 0xff; + long val = *ucontrol->value.integer.value; + unsigned int ctrl_data = snd_hda_codec_read(codec,nid,0,AC_VERB_GET_DIGI_CONVERT,0x00); + + /* Set/unset the masked control bit(s) as needed */ + change = (val==0?0:mask) != (ctrl_data & mask); + if (val==0) + ctrl_data &= ~mask; + else + ctrl_data |= mask; + snd_hda_codec_write(codec,nid,0,AC_VERB_SET_DIGI_CONVERT_1,ctrl_data); + + return change; +} +#define ALC_SPDIF_CTRL_SWITCH(xname, nid, mask) \ + { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = xname, .index = 0, \ + .info = alc_spdif_ctrl_info, \ + .get = alc_spdif_ctrl_get, \ + .put = alc_spdif_ctrl_put, \ + .private_value = nid | (mask<<16) } +#endif /* CONFIG_SND_DEBUG */ - snd_assert(spec->channel_mode, return -ENXIO); - mode = ucontrol->value.enumerated.item[0] ? 1 : 0; - if (spec->multiout.max_channels == spec->channel_mode[mode].channels && - ! codec->in_resume) - return 0; +/* + * set up from the preset table + */ +static void setup_preset(struct alc_spec *spec, const struct alc_config_preset *preset) +{ + int i; - /* change the current channel setting */ - spec->multiout.max_channels = spec->channel_mode[mode].channels; - if (spec->channel_mode[mode].sequence) - snd_hda_sequence_write(codec, spec->channel_mode[mode].sequence); + for (i = 0; i < ARRAY_SIZE(preset->mixers) && preset->mixers[i]; i++) + spec->mixers[spec->num_mixers++] = preset->mixers[i]; + for (i = 0; i < ARRAY_SIZE(preset->init_verbs) && preset->init_verbs[i]; i++) + spec->init_verbs[spec->num_init_verbs++] = preset->init_verbs[i]; + + spec->channel_mode = preset->channel_mode; + spec->num_channel_mode = preset->num_channel_mode; - return 1; -} + spec->multiout.max_channels = spec->channel_mode[0].channels; + spec->multiout.num_dacs = preset->num_dacs; + spec->multiout.dac_nids = preset->dac_nids; + spec->multiout.dig_out_nid = preset->dig_out_nid; + spec->multiout.hp_nid = preset->hp_nid; + + spec->num_mux_defs = preset->num_mux_defs; + if (! spec->num_mux_defs) + spec->num_mux_defs = 1; + spec->input_mux = preset->input_mux; + + spec->num_adc_nids = preset->num_adc_nids; + spec->adc_nids = preset->adc_nids; + spec->dig_in_nid = preset->dig_in_nid; + + spec->unsol_event = preset->unsol_event; + spec->init_hook = preset->init_hook; +} /* + * ALC880 3-stack model + * + * DAC: Front = 0x02 (0x0c), Surr = 0x05 (0x0f), CLFE = 0x04 (0x0e) + * Pin assignment: Front = 0x14, Line-In/Surr = 0x1a, Mic/CLFE = 0x18, F-Mic = 0x1b + * HP = 0x19 */ -/* 3-stack mode - * Pin assignment: Front=0x14, Line-In/Rear=0x1a, Mic/CLFE=0x18, F-Mic=0x1b - * HP=0x19 +static hda_nid_t alc880_dac_nids[4] = { + /* front, rear, clfe, rear_surr */ + 0x02, 0x05, 0x04, 0x03 +}; + +static hda_nid_t alc880_adc_nids[3] = { + /* ADC0-2 */ + 0x07, 0x08, 0x09, +}; + +/* The datasheet says the node 0x07 is connected from inputs, + * but it shows zero connection in the real implementation on some devices. + * Note: this is a 915GAV bug, fixed on 915GLV */ -static snd_kcontrol_new_t alc880_base_mixer[] = { +static hda_nid_t alc880_adc_nids_alt[2] = { + /* ADC1-2 */ + 0x08, 0x09, +}; + +#define ALC880_DIGOUT_NID 0x06 +#define ALC880_DIGIN_NID 0x0a + +static struct hda_input_mux alc880_capture_source = { + .num_items = 4, + .items = { + { "Mic", 0x0 }, + { "Front Mic", 0x3 }, + { "Line", 0x2 }, + { "CD", 0x4 }, + }, +}; + +/* channel source setting (2/6 channel selection for 3-stack) */ +/* 2ch mode */ +static struct hda_verb alc880_threestack_ch2_init[] = { + /* set line-in to input, mute it */ + { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN }, + { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + /* set mic-in to input vref 80%, mute it */ + { 0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80 }, + { 0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { } /* end */ +}; + +/* 6ch mode */ +static struct hda_verb alc880_threestack_ch6_init[] = { + /* set line-in to output, unmute it */ + { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + /* set mic-in to output, unmute it */ + { 0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + { } /* end */ +}; + +static struct hda_channel_mode alc880_threestack_modes[2] = { + { 2, alc880_threestack_ch2_init }, + { 6, alc880_threestack_ch6_init }, +}; + +static struct snd_kcontrol_new alc880_three_stack_mixer[] = { HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), HDA_CODEC_VOLUME("Surround Playback Volume", 0x0f, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Surround Playback Switch", 0x1a, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0f, 2, HDA_INPUT), HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0e, 1, 0x0, HDA_OUTPUT), HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x18, 1, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x18, 2, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0e, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0e, 2, 2, HDA_INPUT), HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x02, HDA_INPUT), @@ -360,12 +600,25 @@ static snd_kcontrol_new_t alc880_base_mixer[] = { HDA_CODEC_MUTE("Front Mic Playback Switch", 0x0b, 0x3, HDA_INPUT), HDA_CODEC_VOLUME("PC Speaker Playback Volume", 0x0b, 0x05, HDA_INPUT), HDA_CODEC_MUTE("PC Speaker Playback Switch", 0x0b, 0x05, HDA_INPUT), - HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0d, 0x0, HDA_OUTPUT), HDA_CODEC_MUTE("Headphone Playback Switch", 0x19, 0x0, HDA_OUTPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + }, + { } /* end */ +}; + +/* capture mixer elements */ +static struct snd_kcontrol_new alc880_capture_mixer[] = { HDA_CODEC_VOLUME("Capture Volume", 0x07, 0x0, HDA_INPUT), HDA_CODEC_MUTE("Capture Switch", 0x07, 0x0, HDA_INPUT), HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x08, 0x0, HDA_INPUT), HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 2, 0x09, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 2, 0x09, 0x0, HDA_INPUT), { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, /* The multiple "Capture Source" controls confuse alsamixer @@ -374,91 +627,313 @@ static snd_kcontrol_new_t alc880_base_mixer[] = { */ /* .name = "Capture Source", */ .name = "Input Source", - .count = 2, + .count = 3, .info = alc_mux_enum_info, .get = alc_mux_enum_get, .put = alc_mux_enum_put, }, + { } /* end */ +}; + +/* capture mixer elements (in case NID 0x07 not available) */ +static struct snd_kcontrol_new alc880_capture_alt_mixer[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x09, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x09, 0x0, HDA_INPUT), { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, - .name = "Channel Mode", - .info = alc880_ch_mode_info, - .get = alc880_ch_mode_get, - .put = alc880_ch_mode_put, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 2, + .info = alc_mux_enum_info, + .get = alc_mux_enum_get, + .put = alc_mux_enum_put, }, { } /* end */ }; -/* 5-stack mode - * Pin assignment: Front=0x14, Rear=0x17, CLFE=0x16 - * Line-In/Side=0x1a, Mic=0x18, F-Mic=0x1b, HP=0x19 - */ -static snd_kcontrol_new_t alc880_five_stack_mixer[] = { - HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT), - HDA_CODEC_VOLUME("Surround Playback Volume", 0x0f, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Surround Playback Switch", 0x17, 0x0, HDA_OUTPUT), + + +/* + * ALC880 5-stack model + * + * DAC: Front = 0x02 (0x0c), Surr = 0x05 (0x0f), CLFE = 0x04 (0x0d), Side = 0x02 (0xd) + * Pin assignment: Front = 0x14, Surr = 0x17, CLFE = 0x16 + * Line-In/Side = 0x1a, Mic = 0x18, F-Mic = 0x1b, HP = 0x19 + */ + +/* additional mixers to alc880_three_stack_mixer */ +static struct snd_kcontrol_new alc880_five_stack_mixer[] = { + HDA_CODEC_VOLUME("Side Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Side Playback Switch", 0x0d, 2, HDA_INPUT), + { } /* end */ +}; + +/* channel source setting (6/8 channel selection for 5-stack) */ +/* 6ch mode */ +static struct hda_verb alc880_fivestack_ch6_init[] = { + /* set line-in to input, mute it */ + { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN }, + { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE }, + { } /* end */ +}; + +/* 8ch mode */ +static struct hda_verb alc880_fivestack_ch8_init[] = { + /* set line-in to output, unmute it */ + { 0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE }, + { } /* end */ +}; + +static struct hda_channel_mode alc880_fivestack_modes[2] = { + { 6, alc880_fivestack_ch6_init }, + { 8, alc880_fivestack_ch8_init }, +}; + + +/* + * ALC880 6-stack model + * + * DAC: Front = 0x02 (0x0c), Surr = 0x03 (0x0d), CLFE = 0x04 (0x0e), Side = 0x05 (0x0f) + * Pin assignment: Front = 0x14, Surr = 0x15, CLFE = 0x16, Side = 0x17, + * Mic = 0x18, F-Mic = 0x19, Line = 0x1a, HP = 0x1b + */ + +static hda_nid_t alc880_6st_dac_nids[4] = { + /* front, rear, clfe, rear_surr */ + 0x02, 0x03, 0x04, 0x05 +}; + +static struct hda_input_mux alc880_6stack_capture_source = { + .num_items = 4, + .items = { + { "Mic", 0x0 }, + { "Front Mic", 0x1 }, + { "Line", 0x2 }, + { "CD", 0x4 }, + }, +}; + +/* fixed 8-channels */ +static struct hda_channel_mode alc880_sixstack_modes[1] = { + { 8, NULL }, +}; + +static struct snd_kcontrol_new alc880_six_stack_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0d, 2, HDA_INPUT), HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0e, 1, 0x0, HDA_OUTPUT), HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x16, 1, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x16, 2, 0x0, HDA_OUTPUT), - HDA_CODEC_VOLUME("Side Playback Volume", 0x0d, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Side Playback Switch", 0x1a, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0e, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0e, 2, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Side Playback Volume", 0x0f, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Side Playback Switch", 0x0f, 2, HDA_INPUT), HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x02, HDA_INPUT), HDA_CODEC_MUTE("Line Playback Switch", 0x0b, 0x02, HDA_INPUT), HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), - HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x0b, 0x3, HDA_INPUT), - HDA_CODEC_MUTE("Front Mic Playback Switch", 0x0b, 0x3, HDA_INPUT), + HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x0b, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x0b, 0x1, HDA_INPUT), HDA_CODEC_VOLUME("PC Speaker Playback Volume", 0x0b, 0x05, HDA_INPUT), HDA_CODEC_MUTE("PC Speaker Playback Switch", 0x0b, 0x05, HDA_INPUT), - HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0d, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Headphone Playback Switch", 0x19, 0x0, HDA_OUTPUT), - HDA_CODEC_VOLUME("Capture Volume", 0x07, 0x0, HDA_INPUT), - HDA_CODEC_MUTE("Capture Switch", 0x07, 0x0, HDA_INPUT), - HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x08, 0x0, HDA_INPUT), - HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x08, 0x0, HDA_INPUT), - { - .iface = SNDRV_CTL_ELEM_IFACE_MIXER, - /* The multiple "Capture Source" controls confuse alsamixer - * So call somewhat different.. - * FIXME: the controls appear in the "playback" view! - */ - /* .name = "Capture Source", */ - .name = "Input Source", - .count = 2, - .info = alc_mux_enum_info, - .get = alc_mux_enum_get, - .put = alc_mux_enum_put, - }, { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, .name = "Channel Mode", - .info = alc880_ch_mode_info, - .get = alc880_ch_mode_get, - .put = alc880_ch_mode_put, + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, }, { } /* end */ }; -static snd_kcontrol_new_t alc880_w810_base_mixer[] = { + +/* + * ALC880 W810 model + * + * W810 has rear IO for: + * Front (DAC 02) + * Surround (DAC 03) + * Center/LFE (DAC 04) + * Digital out (06) + * + * The system also has a pair of internal speakers, and a headphone jack. + * These are both connected to Line2 on the codec, hence to DAC 02. + * + * There is a variable resistor to control the speaker or headphone + * volume. This is a hardware-only device without a software API. + * + * Plugging headphones in will disable the internal speakers. This is + * implemented in hardware, not via the driver using jack sense. In + * a similar fashion, plugging into the rear socket marked "front" will + * disable both the speakers and headphones. + * + * For input, there's a microphone jack, and an "audio in" jack. + * These may not do anything useful with this driver yet, because I + * haven't setup any initialization verbs for these yet... + */ + +static hda_nid_t alc880_w810_dac_nids[3] = { + /* front, rear/surround, clfe */ + 0x02, 0x03, 0x04 +}; + +/* fixed 6 channels */ +static struct hda_channel_mode alc880_w810_modes[1] = { + { 6, NULL } +}; + +/* Pin assignment: Front = 0x14, Surr = 0x15, CLFE = 0x16, HP = 0x1b */ +static struct snd_kcontrol_new alc880_w810_base_mixer[] = { HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), HDA_CODEC_VOLUME("Surround Playback Volume", 0x0d, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Surround Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0d, 2, HDA_INPUT), HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0e, 1, 0x0, HDA_OUTPUT), HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x16, 1, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x16, 2, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0e, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0e, 2, 2, HDA_INPUT), HDA_CODEC_MUTE("Headphone Playback Switch", 0x1b, 0x0, HDA_OUTPUT), - HDA_CODEC_VOLUME("Capture Volume", 0x07, 0x0, HDA_INPUT), - HDA_CODEC_MUTE("Capture Switch", 0x07, 0x0, HDA_INPUT), - HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x08, 0x0, HDA_INPUT), - HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x08, 0x0, HDA_INPUT), - HDA_CODEC_VOLUME_IDX("Capture Volume", 2, 0x09, 0x0, HDA_INPUT), - HDA_CODEC_MUTE_IDX("Capture Switch", 2, 0x09, 0x0, HDA_INPUT), + { } /* end */ +}; + + +/* + * Z710V model + * + * DAC: Front = 0x02 (0x0c), HP = 0x03 (0x0d) + * Pin assignment: Front = 0x14, HP = 0x15, Mic = 0x18, Mic2 = 0x19(?), Line = 0x1a + */ + +static hda_nid_t alc880_z71v_dac_nids[1] = { + 0x02 +}; +#define ALC880_Z71V_HP_DAC 0x03 + +/* fixed 2 channels */ +static struct hda_channel_mode alc880_2_jack_modes[1] = { + { 2, NULL } +}; + +static struct snd_kcontrol_new alc880_z71v_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x0d, 2, HDA_INPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + { } /* end */ +}; + + +/* FIXME! */ +/* + * ALC880 F1734 model + * + * DAC: HP = 0x02 (0x0c), Front = 0x03 (0x0d) + * Pin assignment: HP = 0x14, Front = 0x15, Mic = 0x18 + */ + +static hda_nid_t alc880_f1734_dac_nids[1] = { + 0x03 +}; +#define ALC880_F1734_HP_DAC 0x02 + +static struct snd_kcontrol_new alc880_f1734_mixer[] = { + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Internal Speaker Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Internal Speaker Playback Switch", 0x0d, 2, HDA_INPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + { } /* end */ +}; + + +/* FIXME! */ +/* + * ALC880 ASUS model + * + * DAC: HP/Front = 0x02 (0x0c), Surr = 0x03 (0x0d), CLFE = 0x04 (0x0e) + * Pin assignment: HP/Front = 0x14, Surr = 0x15, CLFE = 0x16, + * Mic = 0x18, Line = 0x1a + */ + +#define alc880_asus_dac_nids alc880_w810_dac_nids /* identical with w810 */ +#define alc880_asus_modes alc880_threestack_modes /* 2/6 channel mode */ + +static struct snd_kcontrol_new alc880_asus_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0d, 2, HDA_INPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0e, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0e, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0e, 2, 2, HDA_INPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x0b, 0x02, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + }, + { } /* end */ +}; + +/* FIXME! */ +/* + * ALC880 ASUS W1V model + * + * DAC: HP/Front = 0x02 (0x0c), Surr = 0x03 (0x0d), CLFE = 0x04 (0x0e) + * Pin assignment: HP/Front = 0x14, Surr = 0x15, CLFE = 0x16, + * Mic = 0x18, Line = 0x1a, Line2 = 0x1b + */ + +/* additional mixers to alc880_asus_mixer */ +static struct snd_kcontrol_new alc880_asus_w1v_mixer[] = { + HDA_CODEC_VOLUME("Line2 Playback Volume", 0x0b, 0x03, HDA_INPUT), + HDA_CODEC_MUTE("Line2 Playback Switch", 0x0b, 0x03, HDA_INPUT), + { } /* end */ +}; + +/* additional mixers to alc880_asus_mixer */ +static struct snd_kcontrol_new alc880_pcbeep_mixer[] = { + HDA_CODEC_VOLUME("PC Speaker Playback Volume", 0x0b, 0x05, HDA_INPUT), + HDA_CODEC_MUTE("PC Speaker Playback Switch", 0x0b, 0x05, HDA_INPUT), + { } /* end */ +}; + +/* TCL S700 */ +static struct snd_kcontrol_new alc880_tcl_s700_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Playback Switch", 0x1b, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x14, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0B, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0B, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0B, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0B, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, /* The multiple "Capture Source" controls confuse alsamixer @@ -467,7 +942,7 @@ static snd_kcontrol_new_t alc880_w810_base_mixer[] = { */ /* .name = "Capture Source", */ .name = "Input Source", - .count = 3, + .count = 1, .info = alc_mux_enum_info, .get = alc_mux_enum_get, .put = alc_mux_enum_put, @@ -476,6 +951,7 @@ static snd_kcontrol_new_t alc880_w810_base_mixer[] = { }; /* + * build control elements */ static int alc_build_controls(struct hda_codec *codec) { @@ -502,230 +978,576 @@ static int alc_build_controls(struct hda_codec *codec) return 0; } + /* * initialize the codec volumes, etc */ -static struct hda_verb alc880_init_verbs_three_stack[] = { - /* Line In pin widget for input */ - {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, - /* CD pin widget for input */ - {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, - /* Mic1 (rear panel) pin widget for input and vref at 80% */ - {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* Mic2 (front panel) pin widget for input and vref at 80% */ - {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* unmute amp left and right */ - {0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, - /* set connection select to line in (default select for this ADC) */ - {0x07, AC_VERB_SET_CONNECT_SEL, 0x02}, - /* unmute front mixer amp left (volume = 0) */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* unmute rear mixer amp left and right (volume = 0) */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* unmute rear mixer amp left and right (volume = 0) */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x18, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - - /* using rear surround as the path for headphone output */ - /* unmute rear surround mixer amp left and right (volume = 0) */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* PASD 3 stack boards use the Mic 2 as the headphone output */ - /* need to program the selector associated with the Mic 2 pin widget to - * surround path (index 0x01) for headphone output */ - {0x11, AC_VERB_SET_CONNECT_SEL, 0x01}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x19, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* need to retask the Mic 2 pin widget to output */ - {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - - /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) for mixer widget(nid=0x0B) - * to support the input path of analog loopback +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc880_volume_init_verbs[] = { + /* + * Unmute ADC0-2 and set the default input to mic-in + */ + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) of the analog-loopback + * mixer widget * Note: PASD motherboards uses the Line In 2 as the input for front panel * mic (mic 2) */ - /* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03 */ - /* unmute CD */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x04 << 8))}, - /* unmute Line In */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x02 << 8))}, - /* unmute Mic 1 */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - /* unmute Line In 2 (for PASD boards Mic 2) */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x03 << 8))}, - - /* Unmute input amps for the line out paths to support the output path of - * analog loopback - * the mixers on the output path has 2 inputs, one from the DAC and one - * from the mixer + /* Amp Indices: Mic1 = 0, Mic2 = 1, Line1 = 2, Line2 = 3, CD = 4 */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + + /* + * Set up output mixers (0x0c - 0x0f) */ - /* Amp Indexes: DAC = 0x01 & mixer = 0x00 */ - /* Unmute Front out path */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Unmute Surround (used as HP) out path */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Unmute C/LFE out path */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x01 << 8))}, /* mute */ - /* Unmute rear Surround out path */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + /* set vol=0 to output mixers */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* set up input amps for analog loopback */ + /* Amp Indices: DAC = 0, mixer = 1 */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, { } }; -static struct hda_verb alc880_init_verbs_five_stack[] = { - /* Line In pin widget for input */ - {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, - /* CD pin widget for input */ - {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, - /* Mic1 (rear panel) pin widget for input and vref at 80% */ - {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* Mic2 (front panel) pin widget for input and vref at 80% */ - {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* unmute amp left and right */ - {0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, - /* set connection select to line in (default select for this ADC) */ - {0x07, AC_VERB_SET_CONNECT_SEL, 0x02}, - /* unmute front mixer amp left and right (volume = 0) */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* five rear and clfe */ - /* unmute rear mixer amp left and right (volume = 0) */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x17, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* unmute clfe mixer amp left and right (volume = 0) */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x16, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - - /* using rear surround as the path for headphone output */ - /* unmute rear surround mixer amp left and right (volume = 0) */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* PASD 3 stack boards use the Mic 2 as the headphone output */ - /* need to program the selector associated with the Mic 2 pin widget to - * surround path (index 0x01) for headphone output +/* + * 3-stack pin configuration: + * front = 0x14, mic/clfe = 0x18, HP = 0x19, line/surr = 0x1a, f-mic = 0x1b + */ +static struct hda_verb alc880_pin_3stack_init_verbs[] = { + /* + * preset connection lists of input pins + * 0 = front, 1 = rear_surr, 2 = CLFE, 3 = surround */ - {0x11, AC_VERB_SET_CONNECT_SEL, 0x01}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x19, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* need to retask the Mic 2 pin widget to output */ - {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x10, AC_VERB_SET_CONNECT_SEL, 0x02}, /* mic/clfe */ + {0x11, AC_VERB_SET_CONNECT_SEL, 0x00}, /* HP */ + {0x12, AC_VERB_SET_CONNECT_SEL, 0x03}, /* line/surround */ - /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) for mixer - * widget(nid=0x0B) to support the input path of analog loopback - */ - /* Note: PASD motherboards uses the Line In 2 as the input for front panel mic (mic 2) */ - /* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03*/ - /* unmute CD */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x04 << 8))}, - /* unmute Line In */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x02 << 8))}, - /* unmute Mic 1 */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - /* unmute Line In 2 (for PASD boards Mic 2) */ - {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x03 << 8))}, - - /* Unmute input amps for the line out paths to support the output path of - * analog loopback - * the mixers on the output path has 2 inputs, one from the DAC and - * one from the mixer + /* + * Set pin mode and muting */ - /* Amp Indexes: DAC = 0x01 & mixer = 0x00 */ - /* Unmute Front out path */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Unmute Surround (used as HP) out path */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Unmute C/LFE out path */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x01 << 8))}, /* mute */ - /* Unmute rear Surround out path */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + /* set front pin widgets 0x14 for output */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Mic1 (rear panel) pin widget for input and vref at 80% */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Mic2 (as headphone out) for HP output */ + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Line In pin widget for input */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Line2 (as front mic) pin widget for input and vref at 80% */ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* CD pin widget for input */ + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, { } }; -static struct hda_verb alc880_w810_init_verbs[] = { - /* front channel selector/amp: input 0: DAC: unmuted, (no volume selection) */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, +/* + * 5-stack pin configuration: + * front = 0x14, surround = 0x17, clfe = 0x16, mic = 0x18, HP = 0x19, + * line-in/side = 0x1a, f-mic = 0x1b + */ +static struct hda_verb alc880_pin_5stack_init_verbs[] = { + /* + * preset connection lists of input pins + * 0 = front, 1 = rear_surr, 2 = CLFE, 3 = surround + */ + {0x11, AC_VERB_SET_CONNECT_SEL, 0x00}, /* HP */ + {0x12, AC_VERB_SET_CONNECT_SEL, 0x01}, /* line/side */ - /* front channel selector/amp: input 1: capture mix: muted, (no volume selection) */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0x7180}, + /* + * Set pin mode and muting + */ + /* set pin widgets 0x14-0x17 for output */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* unmute pins for output (no gain on this amp) */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, - /* front channel selector/amp: output 0: unmuted, max volume */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* Mic1 (rear panel) pin widget for input and vref at 80% */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Mic2 (as headphone out) for HP output */ + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Line In pin widget for input */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Line2 (as front mic) pin widget for input and vref at 80% */ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* CD pin widget for input */ + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, - /* front out pin: muted, (no volume selection) */ - {0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + { } +}; - /* front out pin: NOT headphone enable, out enable, vref disabled */ - {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, +/* + * W810 pin configuration: + * front = 0x14, surround = 0x15, clfe = 0x16, HP = 0x1b + */ +static struct hda_verb alc880_pin_w810_init_verbs[] = { + /* hphone/speaker input selector: front DAC */ + {0x13, AC_VERB_SET_CONNECT_SEL, 0x0}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, - /* surround channel selector/amp: input 0: DAC: unmuted, (no volume selection) */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, - /* surround channel selector/amp: input 1: capture mix: muted, (no volume selection) */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0x7180}, + { } +}; - /* surround channel selector/amp: output 0: unmuted, max volume */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, +/* + * Z71V pin configuration: + * Speaker-out = 0x14, HP = 0x15, Mic = 0x18, Line-in = 0x1a, Mic2 = 0x1b (?) + */ +static struct hda_verb alc880_pin_z71v_init_verbs[] = { + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, - /* surround out pin: muted, (no volume selection) */ - {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, - /* surround out pin: NOT headphone enable, out enable, vref disabled */ - {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + { } +}; +/* + * 6-stack pin configuration: + * front = 0x14, surr = 0x15, clfe = 0x16, side = 0x17, mic = 0x18, f-mic = 0x19, + * line = 0x1a, HP = 0x1b + */ +static struct hda_verb alc880_pin_6stack_init_verbs[] = { + {0x13, AC_VERB_SET_CONNECT_SEL, 0x00}, /* HP */ + + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + + { } +}; - /* c/lfe channel selector/amp: input 0: DAC: unmuted, (no volume selection) */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, +/* FIXME! */ +/* + * F1734 pin configuration: + * HP = 0x14, speaker-out = 0x15, mic = 0x18 + */ +static struct hda_verb alc880_pin_f1734_init_verbs[] = { + {0x10, AC_VERB_SET_CONNECT_SEL, 0x02}, + {0x11, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x12, AC_VERB_SET_CONNECT_SEL, 0x01}, + {0x13, AC_VERB_SET_CONNECT_SEL, 0x00}, + + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, - /* c/lfe channel selector/amp: input 1: capture mix: muted, (no volume selection) */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0x7180}, + { } +}; - /* c/lfe channel selector/amp: output 0: unmuted, max volume */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, +/* FIXME! */ +/* + * ASUS pin configuration: + * HP/front = 0x14, surr = 0x15, clfe = 0x16, mic = 0x18, line = 0x1a + */ +static struct hda_verb alc880_pin_asus_init_verbs[] = { + {0x10, AC_VERB_SET_CONNECT_SEL, 0x02}, + {0x11, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x12, AC_VERB_SET_CONNECT_SEL, 0x01}, + {0x13, AC_VERB_SET_CONNECT_SEL, 0x00}, + + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + + { } +}; - /* c/lfe out pin: muted, (no volume selection) */ - {0x16, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, +/* Enable GPIO mask and set output */ +static struct hda_verb alc880_gpio1_init_verbs[] = { + {0x01, AC_VERB_SET_GPIO_MASK, 0x01}, + {0x01, AC_VERB_SET_GPIO_DIRECTION, 0x01}, + {0x01, AC_VERB_SET_GPIO_DATA, 0x01}, - /* c/lfe out pin: NOT headphone enable, out enable, vref disabled */ - {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + { } +}; +/* Enable GPIO mask and set output */ +static struct hda_verb alc880_gpio2_init_verbs[] = { + {0x01, AC_VERB_SET_GPIO_MASK, 0x02}, + {0x01, AC_VERB_SET_GPIO_DIRECTION, 0x02}, + {0x01, AC_VERB_SET_GPIO_DATA, 0x02}, - /* hphone/speaker input selector: front DAC */ - {0x13, AC_VERB_SET_CONNECT_SEL, 0x0}, + { } +}; + +/* Clevo m520g init */ +static struct hda_verb alc880_pin_clevo_init_verbs[] = { + /* headphone output */ + {0x11, AC_VERB_SET_CONNECT_SEL, 0x01}, + /* line-out */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Line-in */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* CD */ + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Mic1 (rear panel) */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Mic2 (front panel) */ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* headphone */ + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* change to EAPD mode */ + {0x20, AC_VERB_SET_COEF_INDEX, 0x07}, + {0x20, AC_VERB_SET_PROC_COEF, 0x3060}, + + { } +}; + +static struct hda_verb alc880_pin_tcl_S700_init_verbs[] = { + /* Headphone output */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* Front output*/ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x1b, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Line In pin widget for input */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + /* CD pin widget for input */ + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + /* Mic1 (rear panel) pin widget for input and vref at 80% */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + + /* change to EAPD mode */ + {0x20, AC_VERB_SET_COEF_INDEX, 0x07}, + {0x20, AC_VERB_SET_PROC_COEF, 0x3070}, + + { } +}; + +/* + * LG m1 express dual + * + * Pin assignment: + * Rear Line-In/Out (blue): 0x14 + * Build-in Mic-In: 0x15 + * Speaker-out: 0x17 + * HP-Out (green): 0x1b + * Mic-In/Out (red): 0x19 + * SPDIF-Out: 0x1e + */ + +/* To make 5.1 output working (green=Front, blue=Surr, red=CLFE) */ +static hda_nid_t alc880_lg_dac_nids[3] = { + 0x05, 0x02, 0x03 +}; + +/* seems analog CD is not working */ +static struct hda_input_mux alc880_lg_capture_source = { + .num_items = 3, + .items = { + { "Mic", 0x1 }, + { "Line", 0x5 }, + { "Internal Mic", 0x6 }, + }, +}; + +/* 2,4,6 channel modes */ +static struct hda_verb alc880_lg_ch2_init[] = { + /* set line-in and mic-in to input */ + { 0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN }, + { 0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80 }, + { } +}; + +static struct hda_verb alc880_lg_ch4_init[] = { + /* set line-in to out and mic-in to input */ + { 0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP }, + { 0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80 }, + { } +}; + +static struct hda_verb alc880_lg_ch6_init[] = { + /* set line-in and mic-in to output */ + { 0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP }, + { 0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP }, + { } +}; + +static struct hda_channel_mode alc880_lg_ch_modes[3] = { + { 2, alc880_lg_ch2_init }, + { 4, alc880_lg_ch4_init }, + { 6, alc880_lg_ch6_init }, +}; + +static struct snd_kcontrol_new alc880_lg_mixer[] = { + /* FIXME: it's not really "master" but front channels */ + HDA_CODEC_VOLUME("Master Playback Volume", 0x0f, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Master Playback Switch", 0x0f, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0d, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0d, 2, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0d, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0d, 2, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x1, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x06, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x0b, 0x06, HDA_INPUT), + HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x0b, 0x07, HDA_INPUT), + HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x0b, 0x07, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + }, + { } /* end */ +}; - /* hphone/speaker out pin: muted, (no volume selection) */ - {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, +static struct hda_verb alc880_lg_init_verbs[] = { + /* set capture source to mic-in */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + /* mute all amp mixer inputs */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(5)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(6)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(7)}, + /* line-in to input */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* built-in mic */ + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* speaker-out */ + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* mic-in to input */ + {0x11, AC_VERB_SET_CONNECT_SEL, 0x01}, + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* HP-out */ + {0x13, AC_VERB_SET_CONNECT_SEL, 0x03}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* jack sense */ + {0x1b, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | 0x1}, + { } +}; + +/* toggle speaker-output according to the hp-jack state */ +static void alc880_lg_automute(struct hda_codec *codec) +{ + unsigned int present; + + present = snd_hda_codec_read(codec, 0x1b, 0, + AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + snd_hda_codec_amp_update(codec, 0x17, 0, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); + snd_hda_codec_amp_update(codec, 0x17, 1, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); +} + +static void alc880_lg_unsol_event(struct hda_codec *codec, unsigned int res) +{ + /* Looks like the unsol event is incompatible with the standard + * definition. 4bit tag is placed at 28 bit! + */ + if ((res >> 28) == 0x01) + alc880_lg_automute(codec); +} + +/* + * LG LW20 + * + * Pin assignment: + * Speaker-out: 0x14 + * Mic-In: 0x18 + * Built-in Mic-In: 0x19 (?) + * HP-Out: 0x1b + * SPDIF-Out: 0x1e + */ - /* hphone/speaker out pin: NOT headphone enable, out enable, vref disabled */ - {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, +/* seems analog CD is not working */ +static struct hda_input_mux alc880_lg_lw_capture_source = { + .num_items = 2, + .items = { + { "Mic", 0x0 }, + { "Internal Mic", 0x1 }, + }, +}; +static struct snd_kcontrol_new alc880_lg_lw_mixer[] = { + HDA_CODEC_VOLUME("Master Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Master Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Internal Mic Playback Volume", 0x0b, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Internal Mic Playback Switch", 0x0b, 0x01, HDA_INPUT), + { } /* end */ +}; +static struct hda_verb alc880_lg_lw_init_verbs[] = { + /* set capture source to mic-in */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(7)}, + /* speaker-out */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* HP-out */ + {0x13, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* mic-in to input */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* built-in mic */ + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* jack sense */ + {0x1b, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | 0x1}, { } }; +/* toggle speaker-output according to the hp-jack state */ +static void alc880_lg_lw_automute(struct hda_codec *codec) +{ + unsigned int present; + + present = snd_hda_codec_read(codec, 0x1b, 0, + AC_VERB_GET_PIN_SENSE, 0) & 0x80000000; + snd_hda_codec_amp_update(codec, 0x14, 0, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); + snd_hda_codec_amp_update(codec, 0x14, 1, HDA_OUTPUT, 0, + 0x80, present ? 0x80 : 0); +} + +static void alc880_lg_lw_unsol_event(struct hda_codec *codec, unsigned int res) +{ + /* Looks like the unsol event is incompatible with the standard + * definition. 4bit tag is placed at 28 bit! + */ + if ((res >> 28) == 0x01) + alc880_lg_lw_automute(codec); +} + +/* + * Common callbacks + */ + static int alc_init(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; - snd_hda_sequence_write(codec, spec->init_verbs); + unsigned int i; + + for (i = 0; i < spec->num_init_verbs; i++) + snd_hda_sequence_write(codec, spec->init_verbs[i]); + + if (spec->init_hook) + spec->init_hook(codec); + return 0; } +static void alc_unsol_event(struct hda_codec *codec, unsigned int res) +{ + struct alc_spec *spec = codec->spec; + + if (spec->unsol_event) + spec->unsol_event(codec, res); +} + #ifdef CONFIG_PM /* * resume @@ -736,9 +1558,8 @@ static int alc_resume(struct hda_codec *codec) int i; alc_init(codec); - for (i = 0; i < spec->num_mixers; i++) { + for (i = 0; i < spec->num_mixers; i++) snd_hda_resume_ctls(codec, spec->mixers[i]); - } if (spec->multiout.dig_out_nid) snd_hda_resume_spdif_out(codec); if (spec->dig_in_nid) @@ -753,7 +1574,7 @@ static int alc_resume(struct hda_codec *codec) */ static int alc880_playback_pcm_open(struct hda_pcm_stream *hinfo, struct hda_codec *codec, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; return snd_hda_multi_out_analog_open(codec, &spec->multiout, substream); @@ -763,7 +1584,7 @@ static int alc880_playback_pcm_prepare(struct hda_pcm_stream *hinfo, struct hda_codec *codec, unsigned int stream_tag, unsigned int format, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; return snd_hda_multi_out_analog_prepare(codec, &spec->multiout, stream_tag, @@ -772,7 +1593,7 @@ static int alc880_playback_pcm_prepare(struct hda_pcm_stream *hinfo, static int alc880_playback_pcm_cleanup(struct hda_pcm_stream *hinfo, struct hda_codec *codec, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; return snd_hda_multi_out_analog_cleanup(codec, &spec->multiout); @@ -783,7 +1604,7 @@ static int alc880_playback_pcm_cleanup(struct hda_pcm_stream *hinfo, */ static int alc880_dig_playback_pcm_open(struct hda_pcm_stream *hinfo, struct hda_codec *codec, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; return snd_hda_multi_out_dig_open(codec, &spec->multiout); @@ -791,7 +1612,7 @@ static int alc880_dig_playback_pcm_open(struct hda_pcm_stream *hinfo, static int alc880_dig_playback_pcm_close(struct hda_pcm_stream *hinfo, struct hda_codec *codec, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; return snd_hda_multi_out_dig_close(codec, &spec->multiout); @@ -804,7 +1625,7 @@ static int alc880_capture_pcm_prepare(struct hda_pcm_stream *hinfo, struct hda_codec *codec, unsigned int stream_tag, unsigned int format, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; @@ -815,7 +1636,7 @@ static int alc880_capture_pcm_prepare(struct hda_pcm_stream *hinfo, static int alc880_capture_pcm_cleanup(struct hda_pcm_stream *hinfo, struct hda_codec *codec, - snd_pcm_substream_t *substream) + struct snd_pcm_substream *substream) { struct alc_spec *spec = codec->spec; @@ -830,7 +1651,7 @@ static struct hda_pcm_stream alc880_pcm_analog_playback = { .substreams = 1, .channels_min = 2, .channels_max = 8, - .nid = 0x02, /* NID to query formats and rates */ + /* NID is set in alc_build_pcms */ .ops = { .open = alc880_playback_pcm_open, .prepare = alc880_playback_pcm_prepare, @@ -842,7 +1663,7 @@ static struct hda_pcm_stream alc880_pcm_analog_capture = { .substreams = 2, .channels_min = 2, .channels_max = 2, - .nid = 0x07, /* NID to query formats and rates */ + /* NID is set in alc_build_pcms */ .ops = { .prepare = alc880_capture_pcm_prepare, .cleanup = alc880_capture_pcm_cleanup @@ -867,6 +1688,13 @@ static struct hda_pcm_stream alc880_pcm_digital_capture = { /* NID is set in alc_build_pcms */ }; +/* Used by alc_build_pcms to flag that a PCM has no playback stream */ +static struct hda_pcm_stream alc_pcm_null_playback = { + .substreams = 0, + .channels_min = 0, + .channels_max = 0, +}; + static int alc_build_pcms(struct hda_codec *codec) { struct alc_spec *spec = codec->spec; @@ -877,13 +1705,40 @@ static int alc_build_pcms(struct hda_codec *codec) codec->pcm_info = info; info->name = spec->stream_name_analog; - info->stream[SNDRV_PCM_STREAM_PLAYBACK] = *(spec->stream_analog_playback); - info->stream[SNDRV_PCM_STREAM_CAPTURE] = *(spec->stream_analog_capture); + if (spec->stream_analog_playback) { + snd_assert(spec->multiout.dac_nids, return -EINVAL); + info->stream[SNDRV_PCM_STREAM_PLAYBACK] = *(spec->stream_analog_playback); + info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = spec->multiout.dac_nids[0]; + } + if (spec->stream_analog_capture) { + snd_assert(spec->adc_nids, return -EINVAL); + info->stream[SNDRV_PCM_STREAM_CAPTURE] = *(spec->stream_analog_capture); + info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->adc_nids[0]; + } - info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max = 0; - for (i = 0; i < spec->num_channel_mode; i++) { - if (spec->channel_mode[i].channels > info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max) { - info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max = spec->channel_mode[i].channels; + if (spec->channel_mode) { + info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max = 0; + for (i = 0; i < spec->num_channel_mode; i++) { + if (spec->channel_mode[i].channels > info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max) { + info->stream[SNDRV_PCM_STREAM_PLAYBACK].channels_max = spec->channel_mode[i].channels; + } + } + } + + /* If the use of more than one ADC is requested for the current + * model, configure a second analog capture-only PCM. + */ + if (spec->num_adc_nids > 1) { + codec->num_pcms++; + info++; + info->name = spec->stream_name_analog; + /* No playback stream for second PCM */ + info->stream[SNDRV_PCM_STREAM_PLAYBACK] = alc_pcm_null_playback; + info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = 0; + if (spec->stream_analog_capture) { + snd_assert(spec->adc_nids, return -EINVAL); + info->stream[SNDRV_PCM_STREAM_CAPTURE] = *(spec->stream_analog_capture); + info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->adc_nids[1]; } } @@ -891,11 +1746,13 @@ static int alc_build_pcms(struct hda_codec *codec) codec->num_pcms++; info++; info->name = spec->stream_name_digital; - if (spec->multiout.dig_out_nid) { + if (spec->multiout.dig_out_nid && + spec->stream_digital_playback) { info->stream[SNDRV_PCM_STREAM_PLAYBACK] = *(spec->stream_digital_playback); info->stream[SNDRV_PCM_STREAM_PLAYBACK].nid = spec->multiout.dig_out_nid; } - if (spec->dig_in_nid) { + if (spec->dig_in_nid && + spec->stream_digital_capture) { info->stream[SNDRV_PCM_STREAM_CAPTURE] = *(spec->stream_digital_capture); info->stream[SNDRV_PCM_STREAM_CAPTURE].nid = spec->dig_in_nid; } @@ -906,7 +1763,18 @@ static int alc_build_pcms(struct hda_codec *codec) static void alc_free(struct hda_codec *codec) { - kfree(codec->spec); + struct alc_spec *spec = codec->spec; + unsigned int i; + + if (! spec) + return; + + if (spec->kctl_alloc) { + for (i = 0; i < spec->num_kctl_used; i++) + kfree(spec->kctl_alloc[i].name); + kfree(spec->kctl_alloc); + } + kfree(spec); } /* @@ -916,211 +1784,1139 @@ static struct hda_codec_ops alc_patch_ops = { .build_pcms = alc_build_pcms, .init = alc_init, .free = alc_free, + .unsol_event = alc_unsol_event, #ifdef CONFIG_PM .resume = alc_resume, #endif }; + +/* + * Test configuration for debugging + * + * Almost all inputs/outputs are enabled. I/O pins can be configured via + * enum controls. + */ +#ifdef CONFIG_SND_DEBUG +static hda_nid_t alc880_test_dac_nids[4] = { + 0x02, 0x03, 0x04, 0x05 +}; + +static struct hda_input_mux alc880_test_capture_source = { + .num_items = 7, + .items = { + { "In-1", 0x0 }, + { "In-2", 0x1 }, + { "In-3", 0x2 }, + { "In-4", 0x3 }, + { "CD", 0x4 }, + { "Front", 0x5 }, + { "Surround", 0x6 }, + }, +}; + +static struct hda_channel_mode alc880_test_modes[4] = { + { 2, NULL }, + { 4, NULL }, + { 6, NULL }, + { 8, NULL }, +}; + +static int alc_test_pin_ctl_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + static char *texts[] = { + "N/A", "Line Out", "HP Out", + "In Hi-Z", "In 50%", "In Grd", "In 80%", "In 100%" + }; + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 8; + if (uinfo->value.enumerated.item >= 8) + uinfo->value.enumerated.item = 7; + strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); + return 0; +} + +static int alc_test_pin_ctl_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = (hda_nid_t)kcontrol->private_value; + unsigned int pin_ctl, item = 0; + + pin_ctl = snd_hda_codec_read(codec, nid, 0, + AC_VERB_GET_PIN_WIDGET_CONTROL, 0); + if (pin_ctl & AC_PINCTL_OUT_EN) { + if (pin_ctl & AC_PINCTL_HP_EN) + item = 2; + else + item = 1; + } else if (pin_ctl & AC_PINCTL_IN_EN) { + switch (pin_ctl & AC_PINCTL_VREFEN) { + case AC_PINCTL_VREF_HIZ: item = 3; break; + case AC_PINCTL_VREF_50: item = 4; break; + case AC_PINCTL_VREF_GRD: item = 5; break; + case AC_PINCTL_VREF_80: item = 6; break; + case AC_PINCTL_VREF_100: item = 7; break; + } + } + ucontrol->value.enumerated.item[0] = item; + return 0; +} + +static int alc_test_pin_ctl_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = (hda_nid_t)kcontrol->private_value; + static unsigned int ctls[] = { + 0, AC_PINCTL_OUT_EN, AC_PINCTL_OUT_EN | AC_PINCTL_HP_EN, + AC_PINCTL_IN_EN | AC_PINCTL_VREF_HIZ, + AC_PINCTL_IN_EN | AC_PINCTL_VREF_50, + AC_PINCTL_IN_EN | AC_PINCTL_VREF_GRD, + AC_PINCTL_IN_EN | AC_PINCTL_VREF_80, + AC_PINCTL_IN_EN | AC_PINCTL_VREF_100, + }; + unsigned int old_ctl, new_ctl; + + old_ctl = snd_hda_codec_read(codec, nid, 0, + AC_VERB_GET_PIN_WIDGET_CONTROL, 0); + new_ctl = ctls[ucontrol->value.enumerated.item[0]]; + if (old_ctl != new_ctl) { + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, new_ctl); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, + ucontrol->value.enumerated.item[0] >= 3 ? 0xb080 : 0xb000); + return 1; + } + return 0; +} + +static int alc_test_pin_src_info(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_info *uinfo) +{ + static char *texts[] = { + "Front", "Surround", "CLFE", "Side" + }; + uinfo->type = SNDRV_CTL_ELEM_TYPE_ENUMERATED; + uinfo->count = 1; + uinfo->value.enumerated.items = 4; + if (uinfo->value.enumerated.item >= 4) + uinfo->value.enumerated.item = 3; + strcpy(uinfo->value.enumerated.name, texts[uinfo->value.enumerated.item]); + return 0; +} + +static int alc_test_pin_src_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = (hda_nid_t)kcontrol->private_value; + unsigned int sel; + + sel = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_CONNECT_SEL, 0); + ucontrol->value.enumerated.item[0] = sel & 3; + return 0; +} + +static int alc_test_pin_src_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + hda_nid_t nid = (hda_nid_t)kcontrol->private_value; + unsigned int sel; + + sel = snd_hda_codec_read(codec, nid, 0, AC_VERB_GET_CONNECT_SEL, 0) & 3; + if (ucontrol->value.enumerated.item[0] != sel) { + sel = ucontrol->value.enumerated.item[0] & 3; + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_CONNECT_SEL, sel); + return 1; + } + return 0; +} + +#define PIN_CTL_TEST(xname,nid) { \ + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, \ + .name = xname, \ + .info = alc_test_pin_ctl_info, \ + .get = alc_test_pin_ctl_get, \ + .put = alc_test_pin_ctl_put, \ + .private_value = nid \ + } + +#define PIN_SRC_TEST(xname,nid) { \ + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, \ + .name = xname, \ + .info = alc_test_pin_src_info, \ + .get = alc_test_pin_src_get, \ + .put = alc_test_pin_src_put, \ + .private_value = nid \ + } + +static struct snd_kcontrol_new alc880_test_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Surround Playback Volume", 0x0d, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("CLFE Playback Volume", 0x0e, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Side Playback Volume", 0x0f, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0d, 2, HDA_INPUT), + HDA_BIND_MUTE("CLFE Playback Switch", 0x0e, 2, HDA_INPUT), + HDA_BIND_MUTE("Side Playback Switch", 0x0f, 2, HDA_INPUT), + PIN_CTL_TEST("Front Pin Mode", 0x14), + PIN_CTL_TEST("Surround Pin Mode", 0x15), + PIN_CTL_TEST("CLFE Pin Mode", 0x16), + PIN_CTL_TEST("Side Pin Mode", 0x17), + PIN_CTL_TEST("In-1 Pin Mode", 0x18), + PIN_CTL_TEST("In-2 Pin Mode", 0x19), + PIN_CTL_TEST("In-3 Pin Mode", 0x1a), + PIN_CTL_TEST("In-4 Pin Mode", 0x1b), + PIN_SRC_TEST("In-1 Pin Source", 0x18), + PIN_SRC_TEST("In-2 Pin Source", 0x19), + PIN_SRC_TEST("In-3 Pin Source", 0x1a), + PIN_SRC_TEST("In-4 Pin Source", 0x1b), + HDA_CODEC_VOLUME("In-1 Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("In-1 Playback Switch", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("In-2 Playback Volume", 0x0b, 0x1, HDA_INPUT), + HDA_CODEC_MUTE("In-2 Playback Switch", 0x0b, 0x1, HDA_INPUT), + HDA_CODEC_VOLUME("In-3 Playback Volume", 0x0b, 0x2, HDA_INPUT), + HDA_CODEC_MUTE("In-3 Playback Switch", 0x0b, 0x2, HDA_INPUT), + HDA_CODEC_VOLUME("In-4 Playback Volume", 0x0b, 0x3, HDA_INPUT), + HDA_CODEC_MUTE("In-4 Playback Switch", 0x0b, 0x3, HDA_INPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x4, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x4, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + }, + { } /* end */ +}; + +static struct hda_verb alc880_test_init_verbs[] = { + /* Unmute inputs of 0x0c - 0x0f */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + /* Vol output for 0x0c-0x0f */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* Set output pins 0x14-0x17 */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Unmute output pins 0x14-0x17 */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Set input pins 0x18-0x1c */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + /* Mute input pins 0x18-0x1b */ + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* ADC set up */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, + /* Analog input/passthru */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, + { } +}; +#endif + /* */ static struct hda_board_config alc880_cfg_tbl[] = { /* Back 3 jack, front 2 jack */ { .modelname = "3stack", .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe200, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe201, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe202, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe203, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe204, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe205, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe206, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe207, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe208, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe209, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20a, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20b, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20c, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20d, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20e, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe20f, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe210, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe211, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe214, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe302, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe303, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe304, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe306, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe307, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xe404, .config = ALC880_3ST }, - { .pci_vendor = 0x8086, .pci_device = 0xa101, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x3031, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4036, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4037, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4038, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4040, .config = ALC880_3ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4041, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe200, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe201, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe202, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe203, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe204, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe205, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe206, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe207, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe208, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe209, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20a, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20b, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20c, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20d, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20e, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe20f, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe210, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe211, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe214, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe302, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe303, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe304, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe306, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe307, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe404, .config = ALC880_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xa101, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x3031, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4036, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4037, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4038, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4040, .config = ALC880_3ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4041, .config = ALC880_3ST }, + /* TCL S700 */ + { .pci_subvendor = 0x19db, .pci_subdevice = 0x4188, .config = ALC880_TCL_S700 }, /* Back 3 jack, front 2 jack (Internal add Aux-In) */ - { .pci_vendor = 0x1025, .pci_device = 0xe310, .config = ALC880_3ST }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0xe310, .config = ALC880_3ST }, + { .pci_subvendor = 0x104d, .pci_subdevice = 0x81d6, .config = ALC880_3ST }, + { .pci_subvendor = 0x104d, .pci_subdevice = 0x81a0, .config = ALC880_3ST }, /* Back 3 jack plus 1 SPDIF out jack, front 2 jack */ { .modelname = "3stack-digout", .config = ALC880_3ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xe308, .config = ALC880_3ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe308, .config = ALC880_3ST_DIG }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0x0070, .config = ALC880_3ST_DIG }, + /* Clevo m520G NB */ + { .pci_subvendor = 0x1558, .pci_subdevice = 0x0520, .config = ALC880_CLEVO }, /* Back 3 jack plus 1 SPDIF out jack, front 2 jack (Internal add Aux-In)*/ - { .pci_vendor = 0x8086, .pci_device = 0xe305, .config = ALC880_3ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xd402, .config = ALC880_3ST_DIG }, - { .pci_vendor = 0x1025, .pci_device = 0xe309, .config = ALC880_3ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe305, .config = ALC880_3ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xd402, .config = ALC880_3ST_DIG }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0xe309, .config = ALC880_3ST_DIG }, /* Back 5 jack, front 2 jack */ { .modelname = "5stack", .config = ALC880_5ST }, - { .pci_vendor = 0x107b, .pci_device = 0x3033, .config = ALC880_5ST }, - { .pci_vendor = 0x107b, .pci_device = 0x4039, .config = ALC880_5ST }, - { .pci_vendor = 0x107b, .pci_device = 0x3032, .config = ALC880_5ST }, - { .pci_vendor = 0x103c, .pci_device = 0x2a09, .config = ALC880_5ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x3033, .config = ALC880_5ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x4039, .config = ALC880_5ST }, + { .pci_subvendor = 0x107b, .pci_subdevice = 0x3032, .config = ALC880_5ST }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x2a09, .config = ALC880_5ST }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x814e, .config = ALC880_5ST }, /* Back 5 jack plus 1 SPDIF out jack, front 2 jack */ { .modelname = "5stack-digout", .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xe224, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xe400, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xe401, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xe402, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xd400, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xd401, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x8086, .pci_device = 0xa100, .config = ALC880_5ST_DIG }, - { .pci_vendor = 0x1565, .pci_device = 0x8202, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe224, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe400, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe401, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xe402, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xd400, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xd401, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xa100, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x1565, .pci_subdevice = 0x8202, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0x1019, .pci_subdevice = 0xa880, .config = ALC880_5ST_DIG }, + { .pci_subvendor = 0xa0a0, .pci_subdevice = 0x0560, + .config = ALC880_5ST_DIG }, /* Aopen i915GMm-HFS */ + /* { .pci_subvendor = 0x1019, .pci_subdevice = 0xa884, .config = ALC880_5ST_DIG }, */ /* conflict with 6stack */ + { .pci_subvendor = 0x1695, .pci_subdevice = 0x400d, .config = ALC880_5ST_DIG }, + /* note subvendor = 0 below */ + /* { .pci_subvendor = 0x0000, .pci_subdevice = 0x8086, .config = ALC880_5ST_DIG }, */ { .modelname = "w810", .config = ALC880_W810 }, - { .pci_vendor = 0x161f, .pci_device = 0x203d, .config = ALC880_W810 }, + { .pci_subvendor = 0x161f, .pci_subdevice = 0x203d, .config = ALC880_W810 }, + + { .modelname = "z71v", .config = ALC880_Z71V }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1964, .config = ALC880_Z71V }, + + { .modelname = "6stack", .config = ALC880_6ST }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x8196, .config = ALC880_6ST }, /* ASUS P5GD1-HVM */ + { .pci_subvendor = 0x1043, .pci_subdevice = 0x81b4, .config = ALC880_6ST }, + { .pci_subvendor = 0x1019, .pci_subdevice = 0xa884, .config = ALC880_6ST }, /* Acer APFV */ + { .pci_subvendor = 0x1458, .pci_subdevice = 0xa102, .config = ALC880_6ST }, /* Gigabyte K8N51 */ + + { .modelname = "6stack-digout", .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x2668, .pci_subdevice = 0x8086, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0x2668, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1462, .pci_subdevice = 0x1150, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0xe803, .pci_subdevice = 0x1019, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1039, .pci_subdevice = 0x1234, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0x0077, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0x0078, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0x0087, .config = ALC880_6ST_DIG }, + { .pci_subvendor = 0x1297, .pci_subdevice = 0xc790, .config = ALC880_6ST_DIG }, /* Shuttle ST20G5 */ + { .pci_subvendor = 0x1509, .pci_subdevice = 0x925d, .config = ALC880_6ST_DIG }, /* FIC P4M-915GD1 */ + { .pci_subvendor = 0x1695, .pci_subdevice = 0x4012, .config = ALC880_5ST_DIG }, /* Epox EP-5LDA+ GLi */ + + { .modelname = "asus", .config = ALC880_ASUS }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1964, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1973, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x19b3, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1113, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1173, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1993, .config = ALC880_ASUS }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x10c3, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1133, .config = ALC880_ASUS }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1123, .config = ALC880_ASUS_DIG }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x1143, .config = ALC880_ASUS }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x10b3, .config = ALC880_ASUS_W1V }, + { .pci_subvendor = 0x1043, .pci_subdevice = 0x8181, .config = ALC880_ASUS_DIG }, /* ASUS P4GPL-X */ + { .pci_subvendor = 0x1558, .pci_subdevice = 0x5401, .config = ALC880_ASUS_DIG2 }, + + { .modelname = "uniwill", .config = ALC880_UNIWILL_DIG }, + { .pci_subvendor = 0x1584, .pci_subdevice = 0x9050, .config = ALC880_UNIWILL_DIG }, + + { .modelname = "F1734", .config = ALC880_F1734 }, + { .pci_subvendor = 0x1734, .pci_subdevice = 0x107c, .config = ALC880_F1734 }, + { .pci_subvendor = 0x1584, .pci_subdevice = 0x9054, .config = ALC880_F1734 }, + + { .modelname = "lg", .config = ALC880_LG }, + { .pci_subvendor = 0x1854, .pci_subdevice = 0x003b, .config = ALC880_LG }, + + { .modelname = "lg-lw", .config = ALC880_LG_LW }, + { .pci_subvendor = 0x1854, .pci_subdevice = 0x0018, .config = ALC880_LG_LW }, + +#ifdef CONFIG_SND_DEBUG + { .modelname = "test", .config = ALC880_TEST }, +#endif + { .modelname = "auto", .config = ALC880_AUTO }, {} }; -static int patch_alc880(struct hda_codec *codec) -{ - struct alc_spec *spec; - int board_config; - - spec = kcalloc(1, sizeof(*spec), GFP_KERNEL); - if (spec == NULL) - return -ENOMEM; +/* + * ALC880 codec presets + */ +static struct alc_config_preset alc880_presets[] = { + [ALC880_3ST] = { + .mixers = { alc880_three_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_3stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .num_channel_mode = ARRAY_SIZE(alc880_threestack_modes), + .channel_mode = alc880_threestack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_3ST_DIG] = { + .mixers = { alc880_three_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_3stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_threestack_modes), + .channel_mode = alc880_threestack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_TCL_S700] = { + .mixers = { alc880_tcl_s700_mixer }, + .init_verbs = { alc880_volume_init_verbs, + alc880_pin_tcl_S700_init_verbs, + alc880_gpio2_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .hp_nid = 0x03, + .num_channel_mode = ARRAY_SIZE(alc880_2_jack_modes), + .channel_mode = alc880_2_jack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_5ST] = { + .mixers = { alc880_three_stack_mixer, alc880_five_stack_mixer}, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_5stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .num_channel_mode = ARRAY_SIZE(alc880_fivestack_modes), + .channel_mode = alc880_fivestack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_5ST_DIG] = { + .mixers = { alc880_three_stack_mixer, alc880_five_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_5stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_fivestack_modes), + .channel_mode = alc880_fivestack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_6ST] = { + .mixers = { alc880_six_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_6stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_6st_dac_nids), + .dac_nids = alc880_6st_dac_nids, + .num_channel_mode = ARRAY_SIZE(alc880_sixstack_modes), + .channel_mode = alc880_sixstack_modes, + .input_mux = &alc880_6stack_capture_source, + }, + [ALC880_6ST_DIG] = { + .mixers = { alc880_six_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_6stack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_6st_dac_nids), + .dac_nids = alc880_6st_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_sixstack_modes), + .channel_mode = alc880_sixstack_modes, + .input_mux = &alc880_6stack_capture_source, + }, + [ALC880_W810] = { + .mixers = { alc880_w810_base_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_w810_init_verbs, + alc880_gpio2_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_w810_dac_nids), + .dac_nids = alc880_w810_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_w810_modes), + .channel_mode = alc880_w810_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_Z71V] = { + .mixers = { alc880_z71v_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_z71v_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_z71v_dac_nids), + .dac_nids = alc880_z71v_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .hp_nid = 0x03, + .num_channel_mode = ARRAY_SIZE(alc880_2_jack_modes), + .channel_mode = alc880_2_jack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_F1734] = { + .mixers = { alc880_f1734_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_f1734_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_f1734_dac_nids), + .dac_nids = alc880_f1734_dac_nids, + .hp_nid = 0x02, + .num_channel_mode = ARRAY_SIZE(alc880_2_jack_modes), + .channel_mode = alc880_2_jack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_ASUS] = { + .mixers = { alc880_asus_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_asus_init_verbs, + alc880_gpio1_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_asus_dac_nids), + .dac_nids = alc880_asus_dac_nids, + .num_channel_mode = ARRAY_SIZE(alc880_asus_modes), + .channel_mode = alc880_asus_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_ASUS_DIG] = { + .mixers = { alc880_asus_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_asus_init_verbs, + alc880_gpio1_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_asus_dac_nids), + .dac_nids = alc880_asus_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_asus_modes), + .channel_mode = alc880_asus_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_ASUS_DIG2] = { + .mixers = { alc880_asus_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_asus_init_verbs, + alc880_gpio2_init_verbs }, /* use GPIO2 */ + .num_dacs = ARRAY_SIZE(alc880_asus_dac_nids), + .dac_nids = alc880_asus_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_asus_modes), + .channel_mode = alc880_asus_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_ASUS_W1V] = { + .mixers = { alc880_asus_mixer, alc880_asus_w1v_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_asus_init_verbs, + alc880_gpio1_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_asus_dac_nids), + .dac_nids = alc880_asus_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_asus_modes), + .channel_mode = alc880_asus_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_UNIWILL_DIG] = { + .mixers = { alc880_asus_mixer, alc880_pcbeep_mixer }, + .init_verbs = { alc880_volume_init_verbs, alc880_pin_asus_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_asus_dac_nids), + .dac_nids = alc880_asus_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_asus_modes), + .channel_mode = alc880_asus_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_CLEVO] = { + .mixers = { alc880_three_stack_mixer }, + .init_verbs = { alc880_volume_init_verbs, + alc880_pin_clevo_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_dac_nids), + .dac_nids = alc880_dac_nids, + .hp_nid = 0x03, + .num_channel_mode = ARRAY_SIZE(alc880_threestack_modes), + .channel_mode = alc880_threestack_modes, + .input_mux = &alc880_capture_source, + }, + [ALC880_LG] = { + .mixers = { alc880_lg_mixer }, + .init_verbs = { alc880_volume_init_verbs, + alc880_lg_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_lg_dac_nids), + .dac_nids = alc880_lg_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_lg_ch_modes), + .channel_mode = alc880_lg_ch_modes, + .input_mux = &alc880_lg_capture_source, + .unsol_event = alc880_lg_unsol_event, + .init_hook = alc880_lg_automute, + }, + [ALC880_LG_LW] = { + .mixers = { alc880_lg_lw_mixer }, + .init_verbs = { alc880_volume_init_verbs, + alc880_lg_lw_init_verbs }, + .num_dacs = 1, + .dac_nids = alc880_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_2_jack_modes), + .channel_mode = alc880_2_jack_modes, + .input_mux = &alc880_lg_lw_capture_source, + .unsol_event = alc880_lg_lw_unsol_event, + .init_hook = alc880_lg_lw_automute, + }, +#ifdef CONFIG_SND_DEBUG + [ALC880_TEST] = { + .mixers = { alc880_test_mixer }, + .init_verbs = { alc880_test_init_verbs }, + .num_dacs = ARRAY_SIZE(alc880_test_dac_nids), + .dac_nids = alc880_test_dac_nids, + .dig_out_nid = ALC880_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc880_test_modes), + .channel_mode = alc880_test_modes, + .input_mux = &alc880_test_capture_source, + }, +#endif +}; - codec->spec = spec; +/* + * Automatic parse of I/O pins from the BIOS configuration + */ - board_config = snd_hda_check_board_config(codec, alc880_cfg_tbl); - if (board_config < 0) { - snd_printd(KERN_INFO "hda_codec: Unknown model for ALC880\n"); - board_config = ALC880_MINIMAL; - } +#define NUM_CONTROL_ALLOC 32 +#define NUM_VERB_ALLOC 32 - switch (board_config) { - case ALC880_W810: - spec->mixers[spec->num_mixers] = alc880_w810_base_mixer; - spec->num_mixers++; - break; - case ALC880_5ST: - case ALC880_5ST_DIG: - spec->mixers[spec->num_mixers] = alc880_five_stack_mixer; - spec->num_mixers++; - break; - default: - spec->mixers[spec->num_mixers] = alc880_base_mixer; - spec->num_mixers++; - break; - } +enum { + ALC_CTL_WIDGET_VOL, + ALC_CTL_WIDGET_MUTE, + ALC_CTL_BIND_MUTE, +}; +static struct snd_kcontrol_new alc880_control_templates[] = { + HDA_CODEC_VOLUME(NULL, 0, 0, 0), + HDA_CODEC_MUTE(NULL, 0, 0, 0), + HDA_BIND_MUTE(NULL, 0, 0, 0), +}; - switch (board_config) { - case ALC880_3ST_DIG: - case ALC880_5ST_DIG: - case ALC880_W810: - spec->multiout.dig_out_nid = ALC880_DIGOUT_NID; - break; - default: - break; - } +/* add dynamic controls */ +static int add_control(struct alc_spec *spec, int type, const char *name, unsigned long val) +{ + struct snd_kcontrol_new *knew; - switch (board_config) { - case ALC880_3ST: - case ALC880_3ST_DIG: - case ALC880_5ST: - case ALC880_5ST_DIG: - case ALC880_W810: - spec->front_panel = 1; - break; - default: - break; - } + if (spec->num_kctl_used >= spec->num_kctl_alloc) { + int num = spec->num_kctl_alloc + NUM_CONTROL_ALLOC; - switch (board_config) { - case ALC880_5ST: - case ALC880_5ST_DIG: - spec->init_verbs = alc880_init_verbs_five_stack; - spec->channel_mode = alc880_fivestack_modes; - spec->num_channel_mode = ARRAY_SIZE(alc880_fivestack_modes); - break; - case ALC880_W810: - spec->init_verbs = alc880_w810_init_verbs; - spec->channel_mode = alc880_w810_modes; - spec->num_channel_mode = ARRAY_SIZE(alc880_w810_modes); - break; - default: - spec->init_verbs = alc880_init_verbs_three_stack; - spec->channel_mode = alc880_threestack_modes; - spec->num_channel_mode = ARRAY_SIZE(alc880_threestack_modes); - break; + knew = kcalloc(num + 1, sizeof(*knew), GFP_KERNEL); /* array + terminator */ + if (! knew) + return -ENOMEM; + if (spec->kctl_alloc) { + memcpy(knew, spec->kctl_alloc, sizeof(*knew) * spec->num_kctl_alloc); + kfree(spec->kctl_alloc); + } + spec->kctl_alloc = knew; + spec->num_kctl_alloc = num; } - spec->stream_name_analog = "ALC880 Analog"; - spec->stream_analog_playback = &alc880_pcm_analog_playback; - spec->stream_analog_capture = &alc880_pcm_analog_capture; - - spec->stream_name_digital = "ALC880 Digital"; - spec->stream_digital_playback = &alc880_pcm_digital_playback; - spec->stream_digital_capture = &alc880_pcm_digital_capture; - - spec->multiout.max_channels = spec->channel_mode[0].channels; + knew = &spec->kctl_alloc[spec->num_kctl_used]; + *knew = alc880_control_templates[type]; + knew->name = kstrdup(name, GFP_KERNEL); + if (! knew->name) + return -ENOMEM; + knew->private_value = val; + spec->num_kctl_used++; + return 0; +} - switch (board_config) { - case ALC880_W810: - spec->multiout.num_dacs = ARRAY_SIZE(alc880_w810_dac_nids); - spec->multiout.dac_nids = alc880_w810_dac_nids; - // No dedicated headphone socket - it's shared with built-in speakers. - break; - default: - spec->multiout.num_dacs = ARRAY_SIZE(alc880_dac_nids); - spec->multiout.dac_nids = alc880_dac_nids; - spec->multiout.hp_nid = 0x03; /* rear-surround NID */ - break; +#define alc880_is_fixed_pin(nid) ((nid) >= 0x14 && (nid) <= 0x17) +#define alc880_fixed_pin_idx(nid) ((nid) - 0x14) +#define alc880_is_multi_pin(nid) ((nid) >= 0x18) +#define alc880_multi_pin_idx(nid) ((nid) - 0x18) +#define alc880_is_input_pin(nid) ((nid) >= 0x18) +#define alc880_input_pin_idx(nid) ((nid) - 0x18) +#define alc880_idx_to_dac(nid) ((nid) + 0x02) +#define alc880_dac_to_idx(nid) ((nid) - 0x02) +#define alc880_idx_to_mixer(nid) ((nid) + 0x0c) +#define alc880_idx_to_selector(nid) ((nid) + 0x10) +#define ALC880_PIN_CD_NID 0x1c + +/* fill in the dac_nids table from the parsed pin configuration */ +static int alc880_auto_fill_dac_nids(struct alc_spec *spec, const struct auto_pin_cfg *cfg) +{ + hda_nid_t nid; + int assigned[4]; + int i, j; + + memset(assigned, 0, sizeof(assigned)); + spec->multiout.dac_nids = spec->private_dac_nids; + + /* check the pins hardwired to audio widget */ + for (i = 0; i < cfg->line_outs; i++) { + nid = cfg->line_out_pins[i]; + if (alc880_is_fixed_pin(nid)) { + int idx = alc880_fixed_pin_idx(nid); + spec->multiout.dac_nids[i] = alc880_idx_to_dac(idx); + assigned[idx] = 1; + } } - - spec->input_mux = &alc880_capture_source; - spec->num_adc_nids = ARRAY_SIZE(alc880_adc_nids); - spec->adc_nids = alc880_adc_nids; - - codec->patch_ops = alc_patch_ops; - + /* left pins can be connect to any audio widget */ + for (i = 0; i < cfg->line_outs; i++) { + nid = cfg->line_out_pins[i]; + if (alc880_is_fixed_pin(nid)) + continue; + /* search for an empty channel */ + for (j = 0; j < cfg->line_outs; j++) { + if (! assigned[j]) { + spec->multiout.dac_nids[i] = alc880_idx_to_dac(j); + assigned[j] = 1; + break; + } + } + } + spec->multiout.num_dacs = cfg->line_outs; return 0; } -/* +/* add playback controls from the parsed DAC table */ +static int alc880_auto_create_multi_out_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + char name[32]; + static const char *chname[4] = { "Front", "Surround", NULL /*CLFE*/, "Side" }; + hda_nid_t nid; + int i, err; + + for (i = 0; i < cfg->line_outs; i++) { + if (! spec->multiout.dac_nids[i]) + continue; + nid = alc880_idx_to_mixer(alc880_dac_to_idx(spec->multiout.dac_nids[i])); + if (i == 2) { + /* Center/LFE */ + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, "Center Playback Volume", + HDA_COMPOSE_AMP_VAL(nid, 1, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, "LFE Playback Volume", + HDA_COMPOSE_AMP_VAL(nid, 2, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, "Center Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 1, 2, HDA_INPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, "LFE Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 2, 2, HDA_INPUT))) < 0) + return err; + } else { + sprintf(name, "%s Playback Volume", chname[i]); + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + sprintf(name, "%s Playback Switch", chname[i]); + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 2, HDA_INPUT))) < 0) + return err; + } + } + return 0; +} + +/* add playback controls for speaker and HP outputs */ +static int alc880_auto_create_extra_out(struct alc_spec *spec, hda_nid_t pin, + const char *pfx) +{ + hda_nid_t nid; + int err; + char name[32]; + + if (! pin) + return 0; + + if (alc880_is_fixed_pin(pin)) { + nid = alc880_idx_to_dac(alc880_fixed_pin_idx(pin)); + /* specify the DAC as the extra output */ + if (! spec->multiout.hp_nid) + spec->multiout.hp_nid = nid; + else + spec->multiout.extra_out_nid[0] = nid; + /* control HP volume/switch on the output mixer amp */ + nid = alc880_idx_to_mixer(alc880_fixed_pin_idx(pin)); + sprintf(name, "%s Playback Volume", pfx); + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + sprintf(name, "%s Playback Switch", pfx); + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 2, HDA_INPUT))) < 0) + return err; + } else if (alc880_is_multi_pin(pin)) { + /* set manual connection */ + /* we have only a switch on HP-out PIN */ + sprintf(name, "%s Playback Switch", pfx); + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, name, + HDA_COMPOSE_AMP_VAL(pin, 3, 0, HDA_OUTPUT))) < 0) + return err; + } + return 0; +} + +/* create input playback/capture controls for the given pin */ +static int new_analog_input(struct alc_spec *spec, hda_nid_t pin, const char *ctlname, + int idx, hda_nid_t mix_nid) +{ + char name[32]; + int err; + + sprintf(name, "%s Playback Volume", ctlname); + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, name, + HDA_COMPOSE_AMP_VAL(mix_nid, 3, idx, HDA_INPUT))) < 0) + return err; + sprintf(name, "%s Playback Switch", ctlname); + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, name, + HDA_COMPOSE_AMP_VAL(mix_nid, 3, idx, HDA_INPUT))) < 0) + return err; + return 0; +} + +/* create playback/capture controls for input pins */ +static int alc880_auto_create_analog_input_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + struct hda_input_mux *imux = &spec->private_imux; + int i, err, idx; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + if (alc880_is_input_pin(cfg->input_pins[i])) { + idx = alc880_input_pin_idx(cfg->input_pins[i]); + err = new_analog_input(spec, cfg->input_pins[i], + auto_pin_cfg_labels[i], + idx, 0x0b); + if (err < 0) + return err; + imux->items[imux->num_items].label = auto_pin_cfg_labels[i]; + imux->items[imux->num_items].index = alc880_input_pin_idx(cfg->input_pins[i]); + imux->num_items++; + } + } + return 0; +} + +static void alc880_auto_set_output_and_unmute(struct hda_codec *codec, + hda_nid_t nid, int pin_type, + int dac_idx) +{ + /* set as output */ + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, pin_type); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE); + /* need the manual connection? */ + if (alc880_is_multi_pin(nid)) { + struct alc_spec *spec = codec->spec; + int idx = alc880_multi_pin_idx(nid); + snd_hda_codec_write(codec, alc880_idx_to_selector(idx), 0, + AC_VERB_SET_CONNECT_SEL, + alc880_dac_to_idx(spec->multiout.dac_nids[dac_idx])); + } +} + +static void alc880_auto_init_multi_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < spec->autocfg.line_outs; i++) { + hda_nid_t nid = spec->autocfg.line_out_pins[i]; + alc880_auto_set_output_and_unmute(codec, nid, PIN_OUT, i); + } +} + +static void alc880_auto_init_extra_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + hda_nid_t pin; + + pin = spec->autocfg.speaker_pins[0]; + if (pin) /* connect to front */ + alc880_auto_set_output_and_unmute(codec, pin, PIN_OUT, 0); + pin = spec->autocfg.hp_pin; + if (pin) /* connect to front */ + alc880_auto_set_output_and_unmute(codec, pin, PIN_HP, 0); +} + +static void alc880_auto_init_analog_input(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + hda_nid_t nid = spec->autocfg.input_pins[i]; + if (alc880_is_input_pin(nid)) { + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + i <= AUTO_PIN_FRONT_MIC ? PIN_VREF80 : PIN_IN); + if (nid != ALC880_PIN_CD_NID) + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_MUTE); + } + } +} + +/* parse the BIOS configuration and set up the alc_spec */ +/* return 1 if successful, 0 if the proper config is not found, or a negative error code */ +static int alc880_parse_auto_config(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int err; + static hda_nid_t alc880_ignore[] = { 0x1d, 0 }; + + if ((err = snd_hda_parse_pin_def_config(codec, &spec->autocfg, + alc880_ignore)) < 0) + return err; + if (! spec->autocfg.line_outs) + return 0; /* can't find valid BIOS pin config */ + + if ((err = alc880_auto_fill_dac_nids(spec, &spec->autocfg)) < 0 || + (err = alc880_auto_create_multi_out_ctls(spec, &spec->autocfg)) < 0 || + (err = alc880_auto_create_extra_out(spec, + spec->autocfg.speaker_pins[0], + "Speaker")) < 0 || + (err = alc880_auto_create_extra_out(spec, spec->autocfg.hp_pin, + "Headphone")) < 0 || + (err = alc880_auto_create_analog_input_ctls(spec, &spec->autocfg)) < 0) + return err; + + spec->multiout.max_channels = spec->multiout.num_dacs * 2; + + if (spec->autocfg.dig_out_pin) + spec->multiout.dig_out_nid = ALC880_DIGOUT_NID; + if (spec->autocfg.dig_in_pin) + spec->dig_in_nid = ALC880_DIGIN_NID; + + if (spec->kctl_alloc) + spec->mixers[spec->num_mixers++] = spec->kctl_alloc; + + spec->init_verbs[spec->num_init_verbs++] = alc880_volume_init_verbs; + + spec->num_mux_defs = 1; + spec->input_mux = &spec->private_imux; + + return 1; +} + +/* additional initialization for auto-configuration model */ +static void alc880_auto_init(struct hda_codec *codec) +{ + alc880_auto_init_multi_out(codec); + alc880_auto_init_extra_out(codec); + alc880_auto_init_analog_input(codec); +} + +/* + * OK, here we have finally the patch for ALC880 + */ + +static int patch_alc880(struct hda_codec *codec) +{ + struct alc_spec *spec; + int board_config; + int err; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + codec->spec = spec; + + board_config = snd_hda_check_board_config(codec, alc880_cfg_tbl); + if (board_config < 0 || board_config >= ALC880_MODEL_LAST) { + printk(KERN_INFO "hda_codec: Unknown model for ALC880, trying auto-probe from BIOS...\n"); + board_config = ALC880_AUTO; + } + + if (board_config == ALC880_AUTO) { + /* automatic parse from the BIOS config */ + err = alc880_parse_auto_config(codec); + if (err < 0) { + alc_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using 3-stack mode...\n"); + board_config = ALC880_3ST; + } + } + + if (board_config != ALC880_AUTO) + setup_preset(spec, &alc880_presets[board_config]); + + spec->stream_name_analog = "ALC880 Analog"; + spec->stream_analog_playback = &alc880_pcm_analog_playback; + spec->stream_analog_capture = &alc880_pcm_analog_capture; + + spec->stream_name_digital = "ALC880 Digital"; + spec->stream_digital_playback = &alc880_pcm_digital_playback; + spec->stream_digital_capture = &alc880_pcm_digital_capture; + + if (! spec->adc_nids && spec->input_mux) { + /* check whether NID 0x07 is valid */ + unsigned int wcap = get_wcaps(codec, alc880_adc_nids[0]); + wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT; /* get type */ + if (wcap != AC_WID_AUD_IN) { + spec->adc_nids = alc880_adc_nids_alt; + spec->num_adc_nids = ARRAY_SIZE(alc880_adc_nids_alt); + spec->mixers[spec->num_mixers] = alc880_capture_alt_mixer; + spec->num_mixers++; + } else { + spec->adc_nids = alc880_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc880_adc_nids); + spec->mixers[spec->num_mixers] = alc880_capture_mixer; + spec->num_mixers++; + } + } + + codec->patch_ops = alc_patch_ops; + if (board_config == ALC880_AUTO) + spec->init_hook = alc880_auto_init; + + return 0; +} + + +/* * ALC260 support */ +static hda_nid_t alc260_dac_nids[1] = { + /* front */ + 0x02, +}; + +static hda_nid_t alc260_adc_nids[1] = { + /* ADC0 */ + 0x04, +}; + +static hda_nid_t alc260_adc_nids_alt[1] = { + /* ADC1 */ + 0x05, +}; + +static hda_nid_t alc260_hp_adc_nids[2] = { + /* ADC1, 0 */ + 0x05, 0x04 +}; + +/* NIDs used when simultaneous access to both ADCs makes sense. Note that + * alc260_capture_mixer assumes ADC0 (nid 0x04) is the first ADC. + */ +static hda_nid_t alc260_dual_adc_nids[2] = { + /* ADC0, ADC1 */ + 0x04, 0x05 +}; + +#define ALC260_DIGOUT_NID 0x03 +#define ALC260_DIGIN_NID 0x06 + +static struct hda_input_mux alc260_capture_source = { + .num_items = 4, + .items = { + { "Mic", 0x0 }, + { "Front Mic", 0x1 }, + { "Line", 0x2 }, + { "CD", 0x4 }, + }, +}; + +/* On Fujitsu S702x laptops capture only makes sense from Mic/LineIn jack, + * headphone jack and the internal CD lines since these are the only pins at + * which audio can appear. For flexibility, also allow the option of + * recording the mixer output on the second ADC (ADC0 doesn't have a + * connection to the mixer output). + */ +static struct hda_input_mux alc260_fujitsu_capture_sources[2] = { + { + .num_items = 3, + .items = { + { "Mic/Line", 0x0 }, + { "CD", 0x4 }, + { "Headphone", 0x2 }, + }, + }, + { + .num_items = 4, + .items = { + { "Mic/Line", 0x0 }, + { "CD", 0x4 }, + { "Headphone", 0x2 }, + { "Mixer", 0x5 }, + }, + }, + +}; + +/* Acer TravelMate(/Extensa/Aspire) notebooks have similar configuration to + * the Fujitsu S702x, but jacks are marked differently. + */ +static struct hda_input_mux alc260_acer_capture_sources[2] = { + { + .num_items = 4, + .items = { + { "Mic", 0x0 }, + { "Line", 0x2 }, + { "CD", 0x4 }, + { "Headphone", 0x5 }, + }, + }, + { + .num_items = 5, + .items = { + { "Mic", 0x0 }, + { "Line", 0x2 }, + { "CD", 0x4 }, + { "Headphone", 0x6 }, + { "Mixer", 0x5 }, + }, + }, +}; /* * This is just place-holder, so there's something for alc_build_pcms to look * at when it calculates the maximum number of channels. ALC260 has no mixer * element which allows changing the channel mode, so the verb list is * never used. */ -static struct alc_channel_mode alc260_modes[1] = { +static struct hda_channel_mode alc260_modes[1] = { { 2, NULL }, }; -snd_kcontrol_new_t alc260_base_mixer[] = { + +/* Mixer combinations + * + * basic: base_output + input + pc_beep + capture + * HP: base_output + input + capture_alt + * HP_3013: hp_3013 + input + capture + * fujitsu: fujitsu + capture + * acer: acer + capture + */ + +static struct snd_kcontrol_new alc260_base_output_mixer[] = { HDA_CODEC_VOLUME("Front Playback Volume", 0x08, 0x0, HDA_OUTPUT), - /* use LINE2 for the output */ - /* HDA_CODEC_MUTE("Front Playback Switch", 0x0f, 0x0, HDA_OUTPUT), */ - HDA_CODEC_MUTE("Front Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x08, 2, HDA_INPUT), + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x09, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x09, 2, HDA_INPUT), + HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x0a, 1, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Mono Playback Switch", 0x0a, 1, 2, HDA_INPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new alc260_input_mixer[] = { HDA_CODEC_VOLUME("CD Playback Volume", 0x07, 0x04, HDA_INPUT), HDA_CODEC_MUTE("CD Playback Switch", 0x07, 0x04, HDA_INPUT), HDA_CODEC_VOLUME("Line Playback Volume", 0x07, 0x02, HDA_INPUT), @@ -1129,17 +2925,96 @@ snd_kcontrol_new_t alc260_base_mixer[] = { HDA_CODEC_MUTE("Mic Playback Switch", 0x07, 0x0, HDA_INPUT), HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x07, 0x01, HDA_INPUT), HDA_CODEC_MUTE("Front Mic Playback Switch", 0x07, 0x01, HDA_INPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new alc260_pc_beep_mixer[] = { HDA_CODEC_VOLUME("PC Speaker Playback Volume", 0x07, 0x05, HDA_INPUT), HDA_CODEC_MUTE("PC Speaker Playback Switch", 0x07, 0x05, HDA_INPUT), - HDA_CODEC_VOLUME("Headphone Playback Volume", 0x09, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Headphone Playback Switch", 0x10, 0x0, HDA_OUTPUT), - HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x0a, 1, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("Mono Playback Switch", 0x11, 1, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +static struct snd_kcontrol_new alc260_hp_3013_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x09, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Playback Switch", 0x10, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("Aux-In Playback Volume", 0x07, 0x06, HDA_INPUT), + HDA_CODEC_MUTE("Aux-In Playback Switch", 0x07, 0x06, HDA_INPUT), + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x08, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("iSpeaker Playback Volume", 0x0a, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("iSpeaker Playback Switch", 0x11, 1, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +/* Fujitsu S702x series laptops. ALC260 pin usage: Mic/Line jack = 0x12, + * HP jack = 0x14, CD audio = 0x16, internal speaker = 0x10. + */ +static struct snd_kcontrol_new alc260_fujitsu_mixer[] = { + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x08, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Headphone Playback Switch", 0x08, 2, HDA_INPUT), + ALC_PIN_MODE("Headphone Jack Mode", 0x14, ALC_PIN_DIR_INOUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic/Line Playback Volume", 0x07, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic/Line Playback Switch", 0x07, 0x0, HDA_INPUT), + ALC_PIN_MODE("Mic/Line Jack Mode", 0x12, ALC_PIN_DIR_IN), + HDA_CODEC_VOLUME("Beep Playback Volume", 0x07, 0x05, HDA_INPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x07, 0x05, HDA_INPUT), + HDA_CODEC_VOLUME("Internal Speaker Playback Volume", 0x09, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Internal Speaker Playback Switch", 0x09, 2, HDA_INPUT), + { } /* end */ +}; + +/* Mixer for Acer TravelMate(/Extensa/Aspire) notebooks. Note that current + * versions of the ALC260 don't act on requests to enable mic bias from NID + * 0x0f (used to drive the headphone jack in these laptops). The ALC260 + * datasheet doesn't mention this restriction. At this stage it's not clear + * whether this behaviour is intentional or is a hardware bug in chip + * revisions available in early 2006. Therefore for now allow the + * "Headphone Jack Mode" control to span all choices, but if it turns out + * that the lack of mic bias for this NID is intentional we could change the + * mode from ALC_PIN_DIR_INOUT to ALC_PIN_DIR_INOUT_NOMICBIAS. + * + * In addition, Acer TravelMate(/Extensa/Aspire) notebooks in early 2006 + * don't appear to make the mic bias available from the "line" jack, even + * though the NID used for this jack (0x14) can supply it. The theory is + * that perhaps Acer have included blocking capacitors between the ALC260 + * and the output jack. If this turns out to be the case for all such + * models the "Line Jack Mode" mode could be changed from ALC_PIN_DIR_INOUT + * to ALC_PIN_DIR_INOUT_NOMICBIAS. + */ +static struct snd_kcontrol_new alc260_acer_mixer[] = { + HDA_CODEC_VOLUME("Master Playback Volume", 0x08, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Master Playback Switch", 0x08, 2, HDA_INPUT), + ALC_PIN_MODE("Headphone Jack Mode", 0x0f, ALC_PIN_DIR_INOUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x07, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x07, 0x0, HDA_INPUT), + ALC_PIN_MODE("Mic Jack Mode", 0x12, ALC_PIN_DIR_IN), + HDA_CODEC_VOLUME("Line Playback Volume", 0x07, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x07, 0x02, HDA_INPUT), + ALC_PIN_MODE("Line Jack Mode", 0x14, ALC_PIN_DIR_INOUT), + HDA_CODEC_VOLUME("Beep Playback Volume", 0x07, 0x05, HDA_INPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x07, 0x05, HDA_INPUT), + { } /* end */ +}; + +/* capture mixer elements */ +static struct snd_kcontrol_new alc260_capture_mixer[] = { HDA_CODEC_VOLUME("Capture Volume", 0x04, 0x0, HDA_INPUT), HDA_CODEC_MUTE("Capture Switch", 0x04, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x05, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x05, 0x0, HDA_INPUT), { .iface = SNDRV_CTL_ELEM_IFACE_MIXER, - .name = "Capture Source", + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 2, .info = alc_mux_enum_info, .get = alc_mux_enum_get, .put = alc_mux_enum_put, @@ -1147,25 +3022,104 @@ snd_kcontrol_new_t alc260_base_mixer[] = { { } /* end */ }; +static struct snd_kcontrol_new alc260_capture_alt_mixer[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x05, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x05, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 1, + .info = alc_mux_enum_info, + .get = alc_mux_enum_get, + .put = alc_mux_enum_put, + }, + { } /* end */ +}; + +/* + * initialization verbs + */ static struct hda_verb alc260_init_verbs[] = { /* Line In pin widget for input */ - {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, /* CD pin widget for input */ - {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, /* Mic1 (rear panel) pin widget for input and vref at 80% */ - {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, /* Mic2 (front panel) pin widget for input and vref at 80% */ - {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, /* LINE-2 is used for line-out in rear */ - {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, /* select line-out */ - {0x15, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x0e, AC_VERB_SET_CONNECT_SEL, 0x00}, /* LINE-OUT pin */ - {0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, /* enable HP */ - {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, /* enable Mono */ + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* mute capture amp left and right */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + /* set connection select to line in (default select for this ADC) */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x02}, + /* mute capture amp left and right */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + /* set connection select to line in (default select for this ADC) */ + {0x05, AC_VERB_SET_CONNECT_SEL, 0x02}, + /* set vol=0 Line-Out mixer amp left and right */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* unmute pin widget amp left and right (no gain on this amp) */ + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* set vol=0 HP mixer amp left and right */ + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* unmute pin widget amp left and right (no gain on this amp) */ + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* set vol=0 Mono mixer amp left and right */ + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* unmute pin widget amp left and right (no gain on this amp) */ + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* unmute LINE-2 out pin */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03 */ + /* mute CD */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + /* mute Line In */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + /* mute Mic */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + /* Amp Indexes: DAC = 0x01 & mixer = 0x00 */ + /* mute Front out path */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + /* mute Headphone out path */ + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + /* mute Mono out path */ + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + { } +}; + +static struct hda_verb alc260_hp_init_verbs[] = { + /* Headphone and output */ + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0}, + /* mono output */ {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + /* Mic1 (rear panel) pin widget for input and vref at 80% */ + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + /* Mic2 (front panel) pin widget for input and vref at 80% */ + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + /* Line In pin widget for input */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + /* Line-2 pin widget for output */ + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + /* CD pin widget for input */ + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, /* unmute amp left and right */ {0x04, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, /* set connection select to line in (default select for this ADC) */ @@ -1173,17 +3127,11 @@ static struct hda_verb alc260_init_verbs[] = { /* unmute Line-Out mixer amp left and right (volume = 0) */ {0x08, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, /* mute pin widget amp left and right (no gain on this amp) */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, /* unmute HP mixer amp left and right (volume = 0) */ {0x09, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, /* mute pin widget amp left and right (no gain on this amp) */ - {0x10, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, - /* unmute Mono mixer amp left and right (volume = 0) */ - {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - /* mute pin widget amp left and right (no gain on this amp) */ - {0x11, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, - /* mute LINE-2 out */ - {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080}, + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, /* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03 */ /* unmute CD */ {0x07, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x04 << 8))}, @@ -1204,54 +3152,829 @@ static struct hda_verb alc260_init_verbs[] = { { } }; -static struct hda_pcm_stream alc260_pcm_analog_playback = { - .substreams = 1, - .channels_min = 2, - .channels_max = 2, - .nid = 0x2, -}; - -static struct hda_pcm_stream alc260_pcm_analog_capture = { - .substreams = 1, - .channels_min = 2, - .channels_max = 2, - .nid = 0x4, -}; - -static int patch_alc260(struct hda_codec *codec) -{ - struct alc_spec *spec; - - spec = kcalloc(1, sizeof(*spec), GFP_KERNEL); - if (spec == NULL) - return -ENOMEM; - - codec->spec = spec; - - spec->mixers[spec->num_mixers] = alc260_base_mixer; +static struct hda_verb alc260_hp_3013_init_verbs[] = { + /* Line out and output */ + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + /* mono output */ + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + /* Mic1 (rear panel) pin widget for input and vref at 80% */ + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + /* Mic2 (front panel) pin widget for input and vref at 80% */ + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + /* Line In pin widget for input */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + /* Headphone pin widget for output */ + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0}, + /* CD pin widget for input */ + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + /* unmute amp left and right */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, + /* set connection select to line in (default select for this ADC) */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x02}, + /* unmute Line-Out mixer amp left and right (volume = 0) */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* mute pin widget amp left and right (no gain on this amp) */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + /* unmute HP mixer amp left and right (volume = 0) */ + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, + /* mute pin widget amp left and right (no gain on this amp) */ + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + /* Amp Indexes: CD = 0x04, Line In 1 = 0x02, Mic 1 = 0x00 & Line In 2 = 0x03 */ + /* unmute CD */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x04 << 8))}, + /* unmute Line In */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x02 << 8))}, + /* unmute Mic */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + /* Amp Indexes: DAC = 0x01 & mixer = 0x00 */ + /* Unmute Front out path */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + /* Unmute Headphone out path */ + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + /* Unmute Mono out path */ + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + { } +}; + +/* Initialisation sequence for ALC260 as configured in Fujitsu S702x + * laptops. ALC260 pin usage: Mic/Line jack = 0x12, HP jack = 0x14, CD + * audio = 0x16, internal speaker = 0x10. + */ +static struct hda_verb alc260_fujitsu_init_verbs[] = { + /* Disable all GPIOs */ + {0x01, AC_VERB_SET_GPIO_MASK, 0}, + /* Internal speaker is connected to headphone pin */ + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* Headphone/Line-out jack connects to Line1 pin; make it an output */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + /* Mic/Line-in jack is connected to mic1 pin, so make it an input */ + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + /* Ensure all other unused pins are disabled and muted. */ + {0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + + /* Disable digital (SPDIF) pins */ + {0x03, AC_VERB_SET_DIGI_CONVERT_1, 0}, + {0x06, AC_VERB_SET_DIGI_CONVERT_1, 0}, + + /* Ensure Line1 pin widget takes its input from the OUT1 sum bus + * when acting as an output. + */ + {0x0d, AC_VERB_SET_CONNECT_SEL, 0}, + + /* Start with output sum widgets muted and their output gains at min */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + + /* Unmute HP pin widget amp left and right (no equiv mixer ctrl) */ + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Unmute Line1 pin widget output buffer since it starts as an output. + * If the pin mode is changed by the user the pin mode control will + * take care of enabling the pin's input/output buffers as needed. + * Therefore there's no need to enable the input buffer at this + * stage. + */ + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Unmute input buffer of pin widget used for Line-in (no equiv + * mixer ctrl) + */ + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Mute capture amp left and right */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + /* Set ADC connection select to match default mixer setting - line + * in (on mic1 pin) + */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Do the same for the second ADC: mute capture input amp and + * set ADC connection to line in (on mic1 pin) + */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x05, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Mute all inputs to mixer widget (even unconnected ones) */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, /* mic1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* mic2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, /* line1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, /* line2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, /* CD pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, /* Beep-gen pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, /* Line-out pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, /* HP-pin pin */ + + { } +}; + +/* Initialisation sequence for ALC260 as configured in Acer TravelMate and + * similar laptops (adapted from Fujitsu init verbs). + */ +static struct hda_verb alc260_acer_init_verbs[] = { + /* On TravelMate laptops, GPIO 0 enables the internal speaker and + * the headphone jack. Turn this on and rely on the standard mute + * methods whenever the user wants to turn these outputs off. + */ + {0x01, AC_VERB_SET_GPIO_MASK, 0x01}, + {0x01, AC_VERB_SET_GPIO_DIRECTION, 0x01}, + {0x01, AC_VERB_SET_GPIO_DATA, 0x01}, + /* Internal speaker/Headphone jack is connected to Line-out pin */ + {0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + /* Internal microphone/Mic jack is connected to Mic1 pin */ + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF50}, + /* Line In jack is connected to Line1 pin */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + /* Ensure all other unused pins are disabled and muted. */ + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + /* Disable digital (SPDIF) pins */ + {0x03, AC_VERB_SET_DIGI_CONVERT_1, 0}, + {0x06, AC_VERB_SET_DIGI_CONVERT_1, 0}, + + /* Ensure Mic1 and Line1 pin widgets take input from the OUT1 sum + * bus when acting as outputs. + */ + {0x0b, AC_VERB_SET_CONNECT_SEL, 0}, + {0x0d, AC_VERB_SET_CONNECT_SEL, 0}, + + /* Start with output sum widgets muted and their output gains at min */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + + /* Unmute Line-out pin widget amp left and right (no equiv mixer ctrl) */ + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Unmute Mic1 and Line1 pin widget input buffers since they start as + * inputs. If the pin mode is changed by the user the pin mode control + * will take care of enabling the pin's input/output buffers as needed. + * Therefore there's no need to enable the input buffer at this + * stage. + */ + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Mute capture amp left and right */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + /* Set ADC connection select to match default mixer setting - mic + * (on mic1 pin) + */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Do similar with the second ADC: mute capture input amp and + * set ADC connection to mic to match ALSA's default state. + */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x05, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Mute all inputs to mixer widget (even unconnected ones) */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, /* mic1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* mic2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, /* line1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, /* line2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, /* CD pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, /* Beep-gen pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, /* Line-out pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, /* HP-pin pin */ + + { } +}; + +/* Test configuration for debugging, modelled after the ALC880 test + * configuration. + */ +#ifdef CONFIG_SND_DEBUG +static hda_nid_t alc260_test_dac_nids[1] = { + 0x02, +}; +static hda_nid_t alc260_test_adc_nids[2] = { + 0x04, 0x05, +}; +/* For testing the ALC260, each input MUX needs its own definition since + * the signal assignments are different. This assumes that the first ADC + * is NID 0x04. + */ +static struct hda_input_mux alc260_test_capture_sources[2] = { + { + .num_items = 7, + .items = { + { "MIC1 pin", 0x0 }, + { "MIC2 pin", 0x1 }, + { "LINE1 pin", 0x2 }, + { "LINE2 pin", 0x3 }, + { "CD pin", 0x4 }, + { "LINE-OUT pin", 0x5 }, + { "HP-OUT pin", 0x6 }, + }, + }, + { + .num_items = 8, + .items = { + { "MIC1 pin", 0x0 }, + { "MIC2 pin", 0x1 }, + { "LINE1 pin", 0x2 }, + { "LINE2 pin", 0x3 }, + { "CD pin", 0x4 }, + { "Mixer", 0x5 }, + { "LINE-OUT pin", 0x6 }, + { "HP-OUT pin", 0x7 }, + }, + }, +}; +static struct snd_kcontrol_new alc260_test_mixer[] = { + /* Output driver widgets */ + HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x0a, 1, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Mono Playback Switch", 0x0a, 1, 2, HDA_INPUT), + HDA_CODEC_VOLUME("LOUT2 Playback Volume", 0x09, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("LOUT2 Playback Switch", 0x09, 2, HDA_INPUT), + HDA_CODEC_VOLUME("LOUT1 Playback Volume", 0x08, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("LOUT1 Playback Switch", 0x08, 2, HDA_INPUT), + + /* Modes for retasking pin widgets + * Note: the ALC260 doesn't seem to act on requests to enable mic + * bias from NIDs 0x0f and 0x10. The ALC260 datasheet doesn't + * mention this restriction. At this stage it's not clear whether + * this behaviour is intentional or is a hardware bug in chip + * revisions available at least up until early 2006. Therefore for + * now allow the "HP-OUT" and "LINE-OUT" Mode controls to span all + * choices, but if it turns out that the lack of mic bias for these + * NIDs is intentional we could change their modes from + * ALC_PIN_DIR_INOUT to ALC_PIN_DIR_INOUT_NOMICBIAS. + */ + ALC_PIN_MODE("HP-OUT pin mode", 0x10, ALC_PIN_DIR_INOUT), + ALC_PIN_MODE("LINE-OUT pin mode", 0x0f, ALC_PIN_DIR_INOUT), + ALC_PIN_MODE("LINE2 pin mode", 0x15, ALC_PIN_DIR_INOUT), + ALC_PIN_MODE("LINE1 pin mode", 0x14, ALC_PIN_DIR_INOUT), + ALC_PIN_MODE("MIC2 pin mode", 0x13, ALC_PIN_DIR_INOUT), + ALC_PIN_MODE("MIC1 pin mode", 0x12, ALC_PIN_DIR_INOUT), + + /* Loopback mixer controls */ + HDA_CODEC_VOLUME("MIC1 Playback Volume", 0x07, 0x00, HDA_INPUT), + HDA_CODEC_MUTE("MIC1 Playback Switch", 0x07, 0x00, HDA_INPUT), + HDA_CODEC_VOLUME("MIC2 Playback Volume", 0x07, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("MIC2 Playback Switch", 0x07, 0x01, HDA_INPUT), + HDA_CODEC_VOLUME("LINE1 Playback Volume", 0x07, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("LINE1 Playback Switch", 0x07, 0x02, HDA_INPUT), + HDA_CODEC_VOLUME("LINE2 Playback Volume", 0x07, 0x03, HDA_INPUT), + HDA_CODEC_MUTE("LINE2 Playback Switch", 0x07, 0x03, HDA_INPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x07, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Beep Playback Volume", 0x07, 0x05, HDA_INPUT), + HDA_CODEC_MUTE("Beep Playback Switch", 0x07, 0x05, HDA_INPUT), + HDA_CODEC_VOLUME("LINE-OUT loopback Playback Volume", 0x07, 0x06, HDA_INPUT), + HDA_CODEC_MUTE("LINE-OUT loopback Playback Switch", 0x07, 0x06, HDA_INPUT), + HDA_CODEC_VOLUME("HP-OUT loopback Playback Volume", 0x07, 0x7, HDA_INPUT), + HDA_CODEC_MUTE("HP-OUT loopback Playback Switch", 0x07, 0x7, HDA_INPUT), + + /* Controls for GPIO pins, assuming they are configured as outputs */ + ALC_GPIO_DATA_SWITCH("GPIO pin 0", 0x01, 0x01), + ALC_GPIO_DATA_SWITCH("GPIO pin 1", 0x01, 0x02), + ALC_GPIO_DATA_SWITCH("GPIO pin 2", 0x01, 0x04), + ALC_GPIO_DATA_SWITCH("GPIO pin 3", 0x01, 0x08), + + /* Switches to allow the digital IO pins to be enabled. The datasheet + * is ambigious as to which NID is which; testing on laptops which + * make this output available should provide clarification. + */ + ALC_SPDIF_CTRL_SWITCH("SPDIF Playback Switch", 0x03, 0x01), + ALC_SPDIF_CTRL_SWITCH("SPDIF Capture Switch", 0x06, 0x01), + + { } /* end */ +}; +static struct hda_verb alc260_test_init_verbs[] = { + /* Enable all GPIOs as outputs with an initial value of 0 */ + {0x01, AC_VERB_SET_GPIO_DIRECTION, 0x0f}, + {0x01, AC_VERB_SET_GPIO_DATA, 0x00}, + {0x01, AC_VERB_SET_GPIO_MASK, 0x0f}, + + /* Enable retasking pins as output, initially without power amp */ + {0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x13, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x12, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + + /* Disable digital (SPDIF) pins initially, but users can enable + * them via a mixer switch. In the case of SPDIF-out, this initverb + * payload also sets the generation to 0, output to be in "consumer" + * PCM format, copyright asserted, no pre-emphasis and no validity + * control. + */ + {0x03, AC_VERB_SET_DIGI_CONVERT_1, 0}, + {0x06, AC_VERB_SET_DIGI_CONVERT_1, 0}, + + /* Ensure mic1, mic2, line1 and line2 pin widgets take input from the + * OUT1 sum bus when acting as an output. + */ + {0x0b, AC_VERB_SET_CONNECT_SEL, 0}, + {0x0c, AC_VERB_SET_CONNECT_SEL, 0}, + {0x0d, AC_VERB_SET_CONNECT_SEL, 0}, + {0x0e, AC_VERB_SET_CONNECT_SEL, 0}, + + /* Start with output sum widgets muted and their output gains at min */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + + /* Unmute retasking pin widget output buffers since the default + * state appears to be output. As the pin mode is changed by the + * user the pin mode control will take care of enabling the pin's + * input/output buffers as needed. + */ + {0x10, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x13, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x12, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + /* Also unmute the mono-out pin widget */ + {0x11, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + /* Mute capture amp left and right */ + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + /* Set ADC connection select to match default mixer setting (mic1 + * pin) + */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Do the same for the second ADC: mute capture input amp and + * set ADC connection to mic1 pin + */ + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x05, AC_VERB_SET_CONNECT_SEL, 0x00}, + + /* Mute all inputs to mixer widget (even unconnected ones) */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, /* mic1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* mic2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, /* line1 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, /* line2 pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, /* CD pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(5)}, /* Beep-gen pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(6)}, /* Line-out pin */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(7)}, /* HP-pin pin */ + + { } +}; +#endif + +static struct hda_pcm_stream alc260_pcm_analog_playback = { + .substreams = 1, + .channels_min = 2, + .channels_max = 2, +}; + +static struct hda_pcm_stream alc260_pcm_analog_capture = { + .substreams = 1, + .channels_min = 2, + .channels_max = 2, +}; + +#define alc260_pcm_digital_playback alc880_pcm_digital_playback +#define alc260_pcm_digital_capture alc880_pcm_digital_capture + +/* + * for BIOS auto-configuration + */ + +static int alc260_add_playback_controls(struct alc_spec *spec, hda_nid_t nid, + const char *pfx) +{ + hda_nid_t nid_vol; + unsigned long vol_val, sw_val; + char name[32]; + int err; + + if (nid >= 0x0f && nid < 0x11) { + nid_vol = nid - 0x7; + vol_val = HDA_COMPOSE_AMP_VAL(nid_vol, 3, 0, HDA_OUTPUT); + sw_val = HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT); + } else if (nid == 0x11) { + nid_vol = nid - 0x7; + vol_val = HDA_COMPOSE_AMP_VAL(nid_vol, 2, 0, HDA_OUTPUT); + sw_val = HDA_COMPOSE_AMP_VAL(nid, 2, 0, HDA_OUTPUT); + } else if (nid >= 0x12 && nid <= 0x15) { + nid_vol = 0x08; + vol_val = HDA_COMPOSE_AMP_VAL(nid_vol, 3, 0, HDA_OUTPUT); + sw_val = HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT); + } else + return 0; /* N/A */ + + snprintf(name, sizeof(name), "%s Playback Volume", pfx); + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, name, vol_val)) < 0) + return err; + snprintf(name, sizeof(name), "%s Playback Switch", pfx); + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, name, sw_val)) < 0) + return err; + return 1; +} + +/* add playback controls from the parsed DAC table */ +static int alc260_auto_create_multi_out_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + hda_nid_t nid; + int err; + + spec->multiout.num_dacs = 1; + spec->multiout.dac_nids = spec->private_dac_nids; + spec->multiout.dac_nids[0] = 0x02; + + nid = cfg->line_out_pins[0]; + if (nid) { + err = alc260_add_playback_controls(spec, nid, "Front"); + if (err < 0) + return err; + } + + nid = cfg->speaker_pins[0]; + if (nid) { + err = alc260_add_playback_controls(spec, nid, "Speaker"); + if (err < 0) + return err; + } + + nid = cfg->hp_pin; + if (nid) { + err = alc260_add_playback_controls(spec, nid, "Headphone"); + if (err < 0) + return err; + } + return 0; +} + +/* create playback/capture controls for input pins */ +static int alc260_auto_create_analog_input_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + struct hda_input_mux *imux = &spec->private_imux; + int i, err, idx; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + if (cfg->input_pins[i] >= 0x12) { + idx = cfg->input_pins[i] - 0x12; + err = new_analog_input(spec, cfg->input_pins[i], + auto_pin_cfg_labels[i], idx, 0x07); + if (err < 0) + return err; + imux->items[imux->num_items].label = auto_pin_cfg_labels[i]; + imux->items[imux->num_items].index = idx; + imux->num_items++; + } + if ((cfg->input_pins[i] >= 0x0f) && (cfg->input_pins[i] <= 0x10)){ + idx = cfg->input_pins[i] - 0x09; + err = new_analog_input(spec, cfg->input_pins[i], + auto_pin_cfg_labels[i], idx, 0x07); + if (err < 0) + return err; + imux->items[imux->num_items].label = auto_pin_cfg_labels[i]; + imux->items[imux->num_items].index = idx; + imux->num_items++; + } + } + return 0; +} + +static void alc260_auto_set_output_and_unmute(struct hda_codec *codec, + hda_nid_t nid, int pin_type, + int sel_idx) +{ + /* set as output */ + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, pin_type); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE); + /* need the manual connection? */ + if (nid >= 0x12) { + int idx = nid - 0x12; + snd_hda_codec_write(codec, idx + 0x0b, 0, + AC_VERB_SET_CONNECT_SEL, sel_idx); + + } +} + +static void alc260_auto_init_multi_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + hda_nid_t nid; + + nid = spec->autocfg.line_out_pins[0]; + if (nid) + alc260_auto_set_output_and_unmute(codec, nid, PIN_OUT, 0); + + nid = spec->autocfg.speaker_pins[0]; + if (nid) + alc260_auto_set_output_and_unmute(codec, nid, PIN_OUT, 0); + + nid = spec->autocfg.hp_pin; + if (nid) + alc260_auto_set_output_and_unmute(codec, nid, PIN_OUT, 0); +} + +#define ALC260_PIN_CD_NID 0x16 +static void alc260_auto_init_analog_input(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + hda_nid_t nid = spec->autocfg.input_pins[i]; + if (nid >= 0x12) { + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + i <= AUTO_PIN_FRONT_MIC ? PIN_VREF80 : PIN_IN); + if (nid != ALC260_PIN_CD_NID) + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_MUTE); + } + } +} + +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc260_volume_init_verbs[] = { + /* + * Unmute ADC0-1 and set the default input to mic-in + */ + {0x04, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x05, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) of the analog-loopback + * mixer widget + * Note: PASD motherboards uses the Line In 2 as the input for front panel + * mic (mic 2) + */ + /* Amp Indices: Mic1 = 0, Mic2 = 1, Line1 = 2, Line2 = 3, CD = 4 */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + + /* + * Set up output mixers (0x08 - 0x0a) + */ + /* set vol=0 to output mixers */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* set up input amps for analog loopback */ + /* Amp Indices: DAC = 0, mixer = 1 */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + { } +}; + +static int alc260_parse_auto_config(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + unsigned int wcap; + int err; + static hda_nid_t alc260_ignore[] = { 0x17, 0 }; + + if ((err = snd_hda_parse_pin_def_config(codec, &spec->autocfg, + alc260_ignore)) < 0) + return err; + if ((err = alc260_auto_create_multi_out_ctls(spec, &spec->autocfg)) < 0) + return err; + if (! spec->kctl_alloc) + return 0; /* can't find valid BIOS pin config */ + if ((err = alc260_auto_create_analog_input_ctls(spec, &spec->autocfg)) < 0) + return err; + + spec->multiout.max_channels = 2; + + if (spec->autocfg.dig_out_pin) + spec->multiout.dig_out_nid = ALC260_DIGOUT_NID; + if (spec->kctl_alloc) + spec->mixers[spec->num_mixers++] = spec->kctl_alloc; + + spec->init_verbs[spec->num_init_verbs++] = alc260_volume_init_verbs; + + spec->num_mux_defs = 1; + spec->input_mux = &spec->private_imux; + + /* check whether NID 0x04 is valid */ + wcap = get_wcaps(codec, 0x04); + wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT; /* get type */ + if (wcap != AC_WID_AUD_IN) { + spec->adc_nids = alc260_adc_nids_alt; + spec->num_adc_nids = ARRAY_SIZE(alc260_adc_nids_alt); + spec->mixers[spec->num_mixers] = alc260_capture_alt_mixer; + } else { + spec->adc_nids = alc260_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc260_adc_nids); + spec->mixers[spec->num_mixers] = alc260_capture_mixer; + } spec->num_mixers++; - spec->init_verbs = alc260_init_verbs; - spec->channel_mode = alc260_modes; - spec->num_channel_mode = ARRAY_SIZE(alc260_modes); + return 1; +} + +/* additional initialization for auto-configuration model */ +static void alc260_auto_init(struct hda_codec *codec) +{ + alc260_auto_init_multi_out(codec); + alc260_auto_init_analog_input(codec); +} + +/* + * ALC260 configurations + */ +static struct hda_board_config alc260_cfg_tbl[] = { + { .modelname = "basic", .config = ALC260_BASIC }, + { .pci_subvendor = 0x104d, .pci_subdevice = 0x81bb, + .config = ALC260_BASIC }, /* Sony VAIO */ + { .pci_subvendor = 0x152d, .pci_subdevice = 0x0729, + .config = ALC260_BASIC }, /* CTL Travel Master U553W */ + { .modelname = "hp", .config = ALC260_HP }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3010, .config = ALC260_HP }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3011, .config = ALC260_HP }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3012, .config = ALC260_HP_3013 }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3013, .config = ALC260_HP_3013 }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3014, .config = ALC260_HP }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3015, .config = ALC260_HP }, + { .pci_subvendor = 0x103c, .pci_subdevice = 0x3016, .config = ALC260_HP }, + { .modelname = "fujitsu", .config = ALC260_FUJITSU_S702X }, + { .pci_subvendor = 0x10cf, .pci_subdevice = 0x1326, .config = ALC260_FUJITSU_S702X }, + { .modelname = "acer", .config = ALC260_ACER }, + { .pci_subvendor = 0x1025, .pci_subdevice = 0x008f, .config = ALC260_ACER }, +#ifdef CONFIG_SND_DEBUG + { .modelname = "test", .config = ALC260_TEST }, +#endif + { .modelname = "auto", .config = ALC260_AUTO }, + {} +}; + +static struct alc_config_preset alc260_presets[] = { + [ALC260_BASIC] = { + .mixers = { alc260_base_output_mixer, + alc260_input_mixer, + alc260_pc_beep_mixer, + alc260_capture_mixer }, + .init_verbs = { alc260_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_dac_nids), + .dac_nids = alc260_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_adc_nids), + .adc_nids = alc260_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .input_mux = &alc260_capture_source, + }, + [ALC260_HP] = { + .mixers = { alc260_base_output_mixer, + alc260_input_mixer, + alc260_capture_alt_mixer }, + .init_verbs = { alc260_hp_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_dac_nids), + .dac_nids = alc260_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_hp_adc_nids), + .adc_nids = alc260_hp_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .input_mux = &alc260_capture_source, + }, + [ALC260_HP_3013] = { + .mixers = { alc260_hp_3013_mixer, + alc260_input_mixer, + alc260_capture_alt_mixer }, + .init_verbs = { alc260_hp_3013_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_dac_nids), + .dac_nids = alc260_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_hp_adc_nids), + .adc_nids = alc260_hp_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .input_mux = &alc260_capture_source, + }, + [ALC260_FUJITSU_S702X] = { + .mixers = { alc260_fujitsu_mixer, + alc260_capture_mixer }, + .init_verbs = { alc260_fujitsu_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_dac_nids), + .dac_nids = alc260_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_dual_adc_nids), + .adc_nids = alc260_dual_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .num_mux_defs = ARRAY_SIZE(alc260_fujitsu_capture_sources), + .input_mux = alc260_fujitsu_capture_sources, + }, + [ALC260_ACER] = { + .mixers = { alc260_acer_mixer, + alc260_capture_mixer }, + .init_verbs = { alc260_acer_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_dac_nids), + .dac_nids = alc260_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_dual_adc_nids), + .adc_nids = alc260_dual_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .num_mux_defs = ARRAY_SIZE(alc260_acer_capture_sources), + .input_mux = alc260_acer_capture_sources, + }, +#ifdef CONFIG_SND_DEBUG + [ALC260_TEST] = { + .mixers = { alc260_test_mixer, + alc260_capture_mixer }, + .init_verbs = { alc260_test_init_verbs }, + .num_dacs = ARRAY_SIZE(alc260_test_dac_nids), + .dac_nids = alc260_test_dac_nids, + .num_adc_nids = ARRAY_SIZE(alc260_test_adc_nids), + .adc_nids = alc260_test_adc_nids, + .num_channel_mode = ARRAY_SIZE(alc260_modes), + .channel_mode = alc260_modes, + .num_mux_defs = ARRAY_SIZE(alc260_test_capture_sources), + .input_mux = alc260_test_capture_sources, + }, +#endif +}; + +static int patch_alc260(struct hda_codec *codec) +{ + struct alc_spec *spec; + int err, board_config; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + codec->spec = spec; + + board_config = snd_hda_check_board_config(codec, alc260_cfg_tbl); + if (board_config < 0 || board_config >= ALC260_MODEL_LAST) { + snd_printd(KERN_INFO "hda_codec: Unknown model for ALC260\n"); + board_config = ALC260_AUTO; + } + + if (board_config == ALC260_AUTO) { + /* automatic parse from the BIOS config */ + err = alc260_parse_auto_config(codec); + if (err < 0) { + alc_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using base mode...\n"); + board_config = ALC260_BASIC; + } + } + + if (board_config != ALC260_AUTO) + setup_preset(spec, &alc260_presets[board_config]); spec->stream_name_analog = "ALC260 Analog"; spec->stream_analog_playback = &alc260_pcm_analog_playback; spec->stream_analog_capture = &alc260_pcm_analog_capture; - spec->multiout.max_channels = spec->channel_mode[0].channels; - spec->multiout.num_dacs = ARRAY_SIZE(alc260_dac_nids); - spec->multiout.dac_nids = alc260_dac_nids; - - spec->input_mux = &alc260_capture_source; - spec->num_adc_nids = ARRAY_SIZE(alc260_adc_nids); - spec->adc_nids = alc260_adc_nids; + spec->stream_name_digital = "ALC260 Digital"; + spec->stream_digital_playback = &alc260_pcm_digital_playback; + spec->stream_digital_capture = &alc260_pcm_digital_capture; codec->patch_ops = alc_patch_ops; + if (board_config == ALC260_AUTO) + spec->init_hook = alc260_auto_init; return 0; } + /* * ALC882 support * @@ -1263,8 +3986,10 @@ static int patch_alc260(struct hda_codec *codec) * In addition, an independent DAC for the multi-playback (not used in this * driver yet). */ +#define ALC882_DIGOUT_NID 0x06 +#define ALC882_DIGIN_NID 0x0a -static struct alc_channel_mode alc882_ch_modes[1] = { +static struct hda_channel_mode alc882_ch_modes[1] = { { 8, NULL } }; @@ -1273,10 +3998,9 @@ static hda_nid_t alc882_dac_nids[4] = { 0x02, 0x03, 0x04, 0x05 }; -static hda_nid_t alc882_adc_nids[3] = { - /* ADC0-2 */ - 0x07, 0x08, 0x09, -}; +/* identical with ALC880 */ +#define alc882_adc_nids alc880_adc_nids +#define alc882_adc_nids_alt alc880_adc_nids_alt /* input MUX */ /* FIXME: should be a matrix-type input source selection */ @@ -1290,11 +4014,10 @@ static struct hda_input_mux alc882_capture_source = { { "CD", 0x4 }, }, }; - #define alc882_mux_enum_info alc_mux_enum_info #define alc882_mux_enum_get alc_mux_enum_get -static int alc882_mux_enum_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *ucontrol) +static int alc882_mux_enum_put(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { struct hda_codec *codec = snd_kcontrol_chip(kcontrol); struct alc_spec *spec = codec->spec; @@ -1319,20 +4042,47 @@ static int alc882_mux_enum_put(snd_kcontrol_t *kcontrol, snd_ctl_elem_value_t *u return 1; } +/* + * 6ch mode + */ +static struct hda_verb alc882_sixstack_ch6_init[] = { + { 0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x00 }, + { 0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { } /* end */ +}; + +/* + * 8ch mode + */ +static struct hda_verb alc882_sixstack_ch8_init[] = { + { 0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { 0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT }, + { } /* end */ +}; + +static struct hda_channel_mode alc882_sixstack_modes[2] = { + { 6, alc882_sixstack_ch6_init }, + { 8, alc882_sixstack_ch8_init }, +}; + /* Pin assignment: Front=0x14, Rear=0x15, CLFE=0x16, Side=0x17 * Mic=0x18, Front Mic=0x19, Line-In=0x1a, HP=0x1b */ -static snd_kcontrol_new_t alc882_base_mixer[] = { +static struct snd_kcontrol_new alc882_base_mixer[] = { HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Front Playback Switch", 0x0c, 2, HDA_INPUT), HDA_CODEC_VOLUME("Surround Playback Volume", 0x0d, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Surround Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Surround Playback Switch", 0x0d, 2, HDA_INPUT), HDA_CODEC_VOLUME_MONO("Center Playback Volume", 0x0e, 1, 0x0, HDA_OUTPUT), HDA_CODEC_VOLUME_MONO("LFE Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x16, 1, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x16, 2, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE_MONO("Center Playback Switch", 0x0e, 1, 2, HDA_INPUT), + HDA_BIND_MUTE_MONO("LFE Playback Switch", 0x0e, 2, 2, HDA_INPUT), HDA_CODEC_VOLUME("Side Playback Volume", 0x0f, 0x0, HDA_OUTPUT), - HDA_CODEC_MUTE("Side Playback Switch", 0x17, 0x0, HDA_OUTPUT), + HDA_BIND_MUTE("Side Playback Switch", 0x0f, 2, HDA_INPUT), HDA_CODEC_MUTE("Headphone Playback Switch", 0x1b, 0x0, HDA_OUTPUT), HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), @@ -1362,133 +4112,1602 @@ static snd_kcontrol_new_t alc882_base_mixer[] = { { } /* end */ }; +static struct snd_kcontrol_new alc882_chmode_mixer[] = { + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + }, + { } /* end */ +}; + static struct hda_verb alc882_init_verbs[] = { /* Front mixer: unmute input/output amp left and right (volume = 0) */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* Rear mixer */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* CLFE mixer */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, /* Side mixer */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, 0xb000}, - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, - /* Front Pin: to output mode */ - {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - /* Front Pin: mute amp left and right (no volume) */ - {0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* select Front mixer (0x0c, index 0) */ + /* Front Pin: output 0 (0x0c) */ + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, {0x14, AC_VERB_SET_CONNECT_SEL, 0x00}, - /* Rear Pin */ - {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - /* Rear Pin: mute amp left and right (no volume) */ - {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* select Rear mixer (0x0d, index 1) */ + /* Rear Pin: output 1 (0x0d) */ + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, {0x15, AC_VERB_SET_CONNECT_SEL, 0x01}, - /* CLFE Pin */ - {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - /* CLFE Pin: mute amp left and right (no volume) */ - {0x16, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* select CLFE mixer (0x0e, index 2) */ + /* CLFE Pin: output 2 (0x0e) */ + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, {0x16, AC_VERB_SET_CONNECT_SEL, 0x02}, - /* Side Pin */ - {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - /* Side Pin: mute amp left and right (no volume) */ - {0x17, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* select Side mixer (0x0f, index 3) */ + /* Side Pin: output 3 (0x0f) */ + {0x17, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_OUT}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, {0x17, AC_VERB_SET_CONNECT_SEL, 0x03}, - /* Headphone Pin */ - {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, - /* Headphone Pin: mute amp left and right (no volume) */ - {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, - /* select Front mixer (0x0c, index 0) */ + /* Mic (rear) pin: input vref at 80% */ + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Front Mic pin: input vref at 80% */ + {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_VREF80}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Line In pin: input */ + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + /* Line-2 In: Headphone output (output 0 - 0x0c) */ + {0x1b, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, {0x1b, AC_VERB_SET_CONNECT_SEL, 0x00}, - /* Mic (rear) pin widget for input and vref at 80% */ - {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* Front Mic pin widget for input and vref at 80% */ - {0x19, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, - /* Line In pin widget for input */ - {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, /* CD pin widget for input */ - {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_IN}, /* FIXME: use matrix-type input source selection */ /* Mixer elements: 0x18, 19, 1a, 1b, 1c, 1d, 14, 15, 16, 17, 0b */ /* Input mixer1: unmute Mic, F-Mic, Line, CD inputs */ - {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, - {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, - {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, /* Input mixer2 */ - {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, - {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, - {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, /* Input mixer3 */ - {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, - {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, - {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, - {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, - /* ADC1: unmute amp left and right */ - {0x07, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, - /* ADC2: unmute amp left and right */ - {0x08, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, - /* ADC3: unmute amp left and right */ - {0x08, AC_VERB_SET_AMP_GAIN_MUTE, 0x7000}, - - /* Unmute front loopback */ - {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Unmute rear loopback */ - {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, - /* Mute CLFE loopback */ - {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x01 << 8))}, - /* Unmute side loopback */ - {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(3)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(2)}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(4)}, + /* ADC1: mute amp left and right */ + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + /* ADC2: mute amp left and right */ + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + /* ADC3: mute amp left and right */ + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, { } }; -static int patch_alc882(struct hda_codec *codec) -{ +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc882_auto_init_verbs[] = { + /* + * Unmute ADC0-2 and set the default input to mic-in + */ + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) of the analog-loopback + * mixer widget + * Note: PASD motherboards uses the Line In 2 as the input for front panel + * mic (mic 2) + */ + /* Amp Indices: Mic1 = 0, Mic2 = 1, Line1 = 2, Line2 = 3, CD = 4 */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + + /* + * Set up output mixers (0x0c - 0x0f) + */ + /* set vol=0 to output mixers */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* set up input amps for analog loopback */ + /* Amp Indices: DAC = 0, mixer = 1 */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0f, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x26, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x26, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* FIXME: use matrix-type input source selection */ + /* Mixer elements: 0x18, 19, 1a, 1b, 1c, 1d, 14, 15, 16, 17, 0b */ + /* Input mixer1: unmute Mic, F-Mic, Line, CD inputs */ + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer2 */ + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer3 */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + + { } +}; + +/* capture mixer elements */ +static struct snd_kcontrol_new alc882_capture_alt_mixer[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x09, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x09, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 2, + .info = alc882_mux_enum_info, + .get = alc882_mux_enum_get, + .put = alc882_mux_enum_put, + }, + { } /* end */ +}; + +static struct snd_kcontrol_new alc882_capture_mixer[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x07, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x07, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 1, 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 1, 0x08, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME_IDX("Capture Volume", 2, 0x09, 0x0, HDA_INPUT), + HDA_CODEC_MUTE_IDX("Capture Switch", 2, 0x09, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + * FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 3, + .info = alc882_mux_enum_info, + .get = alc882_mux_enum_get, + .put = alc882_mux_enum_put, + }, + { } /* end */ +}; + +/* pcm configuration: identiacal with ALC880 */ +#define alc882_pcm_analog_playback alc880_pcm_analog_playback +#define alc882_pcm_analog_capture alc880_pcm_analog_capture +#define alc882_pcm_digital_playback alc880_pcm_digital_playback +#define alc882_pcm_digital_capture alc880_pcm_digital_capture + +/* + * configuration and preset + */ +static struct hda_board_config alc882_cfg_tbl[] = { + { .modelname = "3stack-dig", .config = ALC882_3ST_DIG }, + { .modelname = "6stack-dig", .config = ALC882_6ST_DIG }, + { .pci_subvendor = 0x1462, .pci_subdevice = 0x6668, .config = ALC882_6ST_DIG }, /* MSI */ + { .pci_subvendor = 0x105b, .pci_subdevice = 0x6668, .config = ALC882_6ST_DIG }, /* Foxconn */ + { .pci_subvendor = 0x1019, .pci_subdevice = 0x6668, .config = ALC882_6ST_DIG }, /* ECS */ + { .modelname = "auto", .config = ALC882_AUTO }, + {} +}; + +static struct alc_config_preset alc882_presets[] = { + [ALC882_3ST_DIG] = { + .mixers = { alc882_base_mixer }, + .init_verbs = { alc882_init_verbs }, + .num_dacs = ARRAY_SIZE(alc882_dac_nids), + .dac_nids = alc882_dac_nids, + .dig_out_nid = ALC882_DIGOUT_NID, + .num_adc_nids = ARRAY_SIZE(alc882_adc_nids), + .adc_nids = alc882_adc_nids, + .dig_in_nid = ALC882_DIGIN_NID, + .num_channel_mode = ARRAY_SIZE(alc882_ch_modes), + .channel_mode = alc882_ch_modes, + .input_mux = &alc882_capture_source, + }, + [ALC882_6ST_DIG] = { + .mixers = { alc882_base_mixer, alc882_chmode_mixer }, + .init_verbs = { alc882_init_verbs }, + .num_dacs = ARRAY_SIZE(alc882_dac_nids), + .dac_nids = alc882_dac_nids, + .dig_out_nid = ALC882_DIGOUT_NID, + .num_adc_nids = ARRAY_SIZE(alc882_adc_nids), + .adc_nids = alc882_adc_nids, + .dig_in_nid = ALC882_DIGIN_NID, + .num_channel_mode = ARRAY_SIZE(alc882_sixstack_modes), + .channel_mode = alc882_sixstack_modes, + .input_mux = &alc882_capture_source, + }, +}; + + +/* + * BIOS auto configuration + */ +static void alc882_auto_set_output_and_unmute(struct hda_codec *codec, + hda_nid_t nid, int pin_type, + int dac_idx) +{ + /* set as output */ + struct alc_spec *spec = codec->spec; + int idx; + + if (spec->multiout.dac_nids[dac_idx] == 0x25) + idx = 4; + else + idx = spec->multiout.dac_nids[dac_idx] - 2; + + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, pin_type); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE); + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_CONNECT_SEL, idx); + +} + +static void alc882_auto_init_multi_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i <= HDA_SIDE; i++) { + hda_nid_t nid = spec->autocfg.line_out_pins[i]; + if (nid) + alc882_auto_set_output_and_unmute(codec, nid, PIN_OUT, i); + } +} + +static void alc882_auto_init_hp_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + hda_nid_t pin; + + pin = spec->autocfg.hp_pin; + if (pin) /* connect to front */ + alc882_auto_set_output_and_unmute(codec, pin, PIN_HP, 0); /* use dac 0 */ +} + +#define alc882_is_input_pin(nid) alc880_is_input_pin(nid) +#define ALC882_PIN_CD_NID ALC880_PIN_CD_NID + +static void alc882_auto_init_analog_input(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + hda_nid_t nid = spec->autocfg.input_pins[i]; + if (alc882_is_input_pin(nid)) { + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + i <= AUTO_PIN_FRONT_MIC ? PIN_VREF80 : PIN_IN); + if (nid != ALC882_PIN_CD_NID) + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_AMP_GAIN_MUTE, + AMP_OUT_MUTE); + } + } +} + +/* almost identical with ALC880 parser... */ +static int alc882_parse_auto_config(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int err = alc880_parse_auto_config(codec); + + if (err < 0) + return err; + else if (err > 0) + /* hack - override the init verbs */ + spec->init_verbs[0] = alc882_auto_init_verbs; + return err; +} + +/* additional initialization for auto-configuration model */ +static void alc882_auto_init(struct hda_codec *codec) +{ + alc882_auto_init_multi_out(codec); + alc882_auto_init_hp_out(codec); + alc882_auto_init_analog_input(codec); +} + +/* + * ALC882 Headphone poll in 3.5.1a or 3.5.2 + */ + +static int patch_alc882(struct hda_codec *codec) +{ + struct alc_spec *spec; + int err, board_config; + + spec = kzalloc(sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + codec->spec = spec; + + board_config = snd_hda_check_board_config(codec, alc882_cfg_tbl); + + if (board_config < 0 || board_config >= ALC882_MODEL_LAST) { + printk(KERN_INFO "hda_codec: Unknown model for ALC882, trying auto-probe from BIOS...\n"); + board_config = ALC882_AUTO; + } + + if (board_config == ALC882_AUTO) { + /* automatic parse from the BIOS config */ + err = alc882_parse_auto_config(codec); + if (err < 0) { + alc_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using base mode...\n"); + board_config = ALC882_3ST_DIG; + } + } + + if (board_config != ALC882_AUTO) + setup_preset(spec, &alc882_presets[board_config]); + + spec->stream_name_analog = "ALC882 Analog"; + spec->stream_analog_playback = &alc882_pcm_analog_playback; + spec->stream_analog_capture = &alc882_pcm_analog_capture; + + spec->stream_name_digital = "ALC882 Digital"; + spec->stream_digital_playback = &alc882_pcm_digital_playback; + spec->stream_digital_capture = &alc882_pcm_digital_capture; + + if (! spec->adc_nids && spec->input_mux) { + /* check whether NID 0x07 is valid */ + unsigned int wcap = get_wcaps(codec, 0x07); + wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT; /* get type */ + if (wcap != AC_WID_AUD_IN) { + spec->adc_nids = alc882_adc_nids_alt; + spec->num_adc_nids = ARRAY_SIZE(alc882_adc_nids_alt); + spec->mixers[spec->num_mixers] = alc882_capture_alt_mixer; + spec->num_mixers++; + } else { + spec->adc_nids = alc882_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc882_adc_nids); + spec->mixers[spec->num_mixers] = alc882_capture_mixer; + spec->num_mixers++; + } + } + + codec->patch_ops = alc_patch_ops; + if (board_config == ALC882_AUTO) + spec->init_hook = alc882_auto_init; + + return 0; +} + +/* + * ALC262 support + */ + +#define ALC262_DIGOUT_NID ALC880_DIGOUT_NID +#define ALC262_DIGIN_NID ALC880_DIGIN_NID + +#define alc262_dac_nids alc260_dac_nids +#define alc262_adc_nids alc882_adc_nids +#define alc262_adc_nids_alt alc882_adc_nids_alt + +#define alc262_modes alc260_modes +#define alc262_capture_source alc882_capture_source + +static struct snd_kcontrol_new alc262_base_mixer[] = { + HDA_CODEC_VOLUME("Front Playback Volume", 0x0c, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Front Playback Switch", 0x14, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x0b, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x0b, 0x02, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Front Mic Playback Volume", 0x0b, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x0b, 0x01, HDA_INPUT), + /* HDA_CODEC_VOLUME("PC Beep Playback Volume", 0x0b, 0x05, HDA_INPUT), + HDA_CODEC_MUTE("PC Beelp Playback Switch", 0x0b, 0x05, HDA_INPUT), */ + HDA_CODEC_VOLUME("Headphone Playback Volume", 0x0D, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_VOLUME_MONO("Mono Playback Volume", 0x0e, 2, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("Mono Playback Switch", 0x16, 2, 0x0, HDA_OUTPUT), + { } /* end */ +}; + +#define alc262_capture_mixer alc882_capture_mixer +#define alc262_capture_alt_mixer alc882_capture_alt_mixer + +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc262_init_verbs[] = { + /* + * Unmute ADC0-2 and set the default input to mic-in + */ + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) of the analog-loopback + * mixer widget + * Note: PASD motherboards uses the Line In 2 as the input for front panel + * mic (mic 2) + */ + /* Amp Indices: Mic1 = 0, Mic2 = 1, Line1 = 2, Line2 = 3, CD = 4 */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + + /* + * Set up output mixers (0x0c - 0x0e) + */ + /* set vol=0 to output mixers */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + /* set up input amps for analog loopback */ + /* Amp Indices: DAC = 0, mixer = 1 */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x15, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0}, + {0x16, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40}, + {0x18, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24}, + {0x1a, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + {0x1c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20}, + + {0x14, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, 0x0000}, + + {0x14, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x15, AC_VERB_SET_CONNECT_SEL, 0x01}, + + /* FIXME: use matrix-type input source selection */ + /* Mixer elements: 0x18, 19, 1a, 1b, 1c, 1d, 14, 15, 16, 17, 0b */ + /* Input mixer1: unmute Mic, F-Mic, Line, CD inputs */ + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer2 */ + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer3 */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + + { } +}; + +/* + * fujitsu model + * 0x14 = headphone/spdif-out, 0x15 = internal speaker + */ + +#define ALC_HP_EVENT 0x37 + +static struct hda_verb alc262_fujitsu_unsol_verbs[] = { + {0x14, AC_VERB_SET_UNSOLICITED_ENABLE, AC_USRSP_EN | ALC_HP_EVENT}, + {0x14, AC_VERB_SET_PIN_WIDGET_CONTROL, PIN_HP}, + {} +}; + +static struct hda_input_mux alc262_fujitsu_capture_source = { + .num_items = 2, + .items = { + { "Mic", 0x0 }, + { "CD", 0x4 }, + }, +}; + +/* mute/unmute internal speaker according to the hp jack and mute state */ +static void alc262_fujitsu_automute(struct hda_codec *codec, int force) +{ + struct alc_spec *spec = codec->spec; + unsigned int mute; + + if (force || ! spec->sense_updated) { + unsigned int present; + /* need to execute and sync at first */ + snd_hda_codec_read(codec, 0x14, 0, AC_VERB_SET_PIN_SENSE, 0); + present = snd_hda_codec_read(codec, 0x14, 0, + AC_VERB_GET_PIN_SENSE, 0); + spec->jack_present = (present & 0x80000000) != 0; + spec->sense_updated = 1; + } + if (spec->jack_present) { + /* mute internal speaker */ + snd_hda_codec_amp_update(codec, 0x15, 0, HDA_OUTPUT, 0, + 0x80, 0x80); + snd_hda_codec_amp_update(codec, 0x15, 1, HDA_OUTPUT, 0, + 0x80, 0x80); + } else { + /* unmute internal speaker if necessary */ + mute = snd_hda_codec_amp_read(codec, 0x14, 0, HDA_OUTPUT, 0); + snd_hda_codec_amp_update(codec, 0x15, 0, HDA_OUTPUT, 0, + 0x80, mute & 0x80); + mute = snd_hda_codec_amp_read(codec, 0x14, 1, HDA_OUTPUT, 0); + snd_hda_codec_amp_update(codec, 0x15, 1, HDA_OUTPUT, 0, + 0x80, mute & 0x80); + } +} + +/* unsolicited event for HP jack sensing */ +static void alc262_fujitsu_unsol_event(struct hda_codec *codec, + unsigned int res) +{ + if ((res >> 26) != ALC_HP_EVENT) + return; + alc262_fujitsu_automute(codec, 1); +} + +/* bind volumes of both NID 0x0c and 0x0d */ +static int alc262_fujitsu_master_vol_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + long *valp = ucontrol->value.integer.value; + int change; + + change = snd_hda_codec_amp_update(codec, 0x0c, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + change |= snd_hda_codec_amp_update(codec, 0x0c, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + snd_hda_codec_amp_update(codec, 0x0d, 0, HDA_OUTPUT, 0, + 0x7f, valp[0] & 0x7f); + snd_hda_codec_amp_update(codec, 0x0d, 1, HDA_OUTPUT, 0, + 0x7f, valp[1] & 0x7f); + return change; +} + +/* bind hp and internal speaker mute (with plug check) */ +static int alc262_fujitsu_master_sw_put(struct snd_kcontrol *kcontrol, + struct snd_ctl_elem_value *ucontrol) +{ + struct hda_codec *codec = snd_kcontrol_chip(kcontrol); + long *valp = ucontrol->value.integer.value; + int change; + + change = snd_hda_codec_amp_update(codec, 0x14, 0, HDA_OUTPUT, 0, + 0x80, valp[0] ? 0 : 0x80); + change |= snd_hda_codec_amp_update(codec, 0x14, 1, HDA_OUTPUT, 0, + 0x80, valp[1] ? 0 : 0x80); + if (change || codec->in_resume) + alc262_fujitsu_automute(codec, codec->in_resume); + return change; +} + +static struct snd_kcontrol_new alc262_fujitsu_mixer[] = { + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Volume", + .info = snd_hda_mixer_amp_volume_info, + .get = snd_hda_mixer_amp_volume_get, + .put = alc262_fujitsu_master_vol_put, + .private_value = HDA_COMPOSE_AMP_VAL(0x0c, 3, 0, HDA_OUTPUT), + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Master Playback Switch", + .info = snd_hda_mixer_amp_switch_info, + .get = snd_hda_mixer_amp_switch_get, + .put = alc262_fujitsu_master_sw_put, + .private_value = HDA_COMPOSE_AMP_VAL(0x14, 3, 0, HDA_OUTPUT), + }, + HDA_CODEC_VOLUME("CD Playback Volume", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x0b, 0x04, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Boost", 0x18, 0, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x0b, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x0b, 0x0, HDA_INPUT), + { } /* end */ +}; + +/* add playback controls from the parsed DAC table */ +static int alc262_auto_create_multi_out_ctls(struct alc_spec *spec, const struct auto_pin_cfg *cfg) +{ + hda_nid_t nid; + int err; + + spec->multiout.num_dacs = 1; /* only use one dac */ + spec->multiout.dac_nids = spec->private_dac_nids; + spec->multiout.dac_nids[0] = 2; + + nid = cfg->line_out_pins[0]; + if (nid) { + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, "Front Playback Volume", + HDA_COMPOSE_AMP_VAL(0x0c, 3, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Front Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + } + + nid = cfg->speaker_pins[0]; + if (nid) { + if (nid == 0x16) { + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, "Speaker Playback Volume", + HDA_COMPOSE_AMP_VAL(0x0e, 2, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Speaker Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 2, 0, HDA_OUTPUT))) < 0) + return err; + } else { + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Speaker Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + } + } + nid = cfg->hp_pin; + if (nid) { + /* spec->multiout.hp_nid = 2; */ + if (nid == 0x16) { + if ((err = add_control(spec, ALC_CTL_WIDGET_VOL, "Headphone Playback Volume", + HDA_COMPOSE_AMP_VAL(0x0e, 2, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Headphone Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 2, 0, HDA_OUTPUT))) < 0) + return err; + } else { + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Headphone Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + } + } + return 0; +} + +/* identical with ALC880 */ +#define alc262_auto_create_analog_input_ctls alc880_auto_create_analog_input_ctls + +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc262_volume_init_verbs[] = { + /* + * Unmute ADC0-2 and set the default input to mic-in + */ + {0x07, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x09, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x09, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute input amps (CD, Line In, Mic 1 & Mic 2) of the analog-loopback + * mixer widget + * Note: PASD motherboards uses the Line In 2 as the input for front panel + * mic (mic 2) + */ + /* Amp Indices: Mic1 = 0, Mic2 = 1, Line1 = 2, Line2 = 3, CD = 4 */ + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x0b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(4)}, + + /* + * Set up output mixers (0x0c - 0x0f) + */ + /* set vol=0 to output mixers */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_ZERO}, + + /* set up input amps for analog loopback */ + /* Amp Indices: DAC = 0, mixer = 1 */ + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0d, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x0e, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* FIXME: use matrix-type input source selection */ + /* Mixer elements: 0x18, 19, 1a, 1b, 1c, 1d, 14, 15, 16, 17, 0b */ + /* Input mixer1: unmute Mic, F-Mic, Line, CD inputs */ + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x24, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer2 */ + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x23, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + /* Input mixer3 */ + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x00 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x03 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8))}, + {0x22, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x04 << 8))}, + + { } +}; + +/* pcm configuration: identiacal with ALC880 */ +#define alc262_pcm_analog_playback alc880_pcm_analog_playback +#define alc262_pcm_analog_capture alc880_pcm_analog_capture +#define alc262_pcm_digital_playback alc880_pcm_digital_playback +#define alc262_pcm_digital_capture alc880_pcm_digital_capture + +/* + * BIOS auto configuration + */ +static int alc262_parse_auto_config(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int err; + static hda_nid_t alc262_ignore[] = { 0x1d, 0 }; + + if ((err = snd_hda_parse_pin_def_config(codec, &spec->autocfg, + alc262_ignore)) < 0) + return err; + if (! spec->autocfg.line_outs) + return 0; /* can't find valid BIOS pin config */ + if ((err = alc262_auto_create_multi_out_ctls(spec, &spec->autocfg)) < 0 || + (err = alc262_auto_create_analog_input_ctls(spec, &spec->autocfg)) < 0) + return err; + + spec->multiout.max_channels = spec->multiout.num_dacs * 2; + + if (spec->autocfg.dig_out_pin) + spec->multiout.dig_out_nid = ALC262_DIGOUT_NID; + if (spec->autocfg.dig_in_pin) + spec->dig_in_nid = ALC262_DIGIN_NID; + + if (spec->kctl_alloc) + spec->mixers[spec->num_mixers++] = spec->kctl_alloc; + + spec->init_verbs[spec->num_init_verbs++] = alc262_volume_init_verbs; + spec->num_mux_defs = 1; + spec->input_mux = &spec->private_imux; + + return 1; +} + +#define alc262_auto_init_multi_out alc882_auto_init_multi_out +#define alc262_auto_init_hp_out alc882_auto_init_hp_out +#define alc262_auto_init_analog_input alc882_auto_init_analog_input + + +/* init callback for auto-configuration model -- overriding the default init */ +static void alc262_auto_init(struct hda_codec *codec) +{ + alc262_auto_init_multi_out(codec); + alc262_auto_init_hp_out(codec); + alc262_auto_init_analog_input(codec); +} + +/* + * configuration and preset + */ +static struct hda_board_config alc262_cfg_tbl[] = { + { .modelname = "basic", .config = ALC262_BASIC }, + { .modelname = "fujitsu", .config = ALC262_FUJITSU }, + { .pci_subvendor = 0x10cf, .pci_subdevice = 0x1397, .config = ALC262_FUJITSU }, + { .modelname = "auto", .config = ALC262_AUTO }, + {} +}; + +static struct alc_config_preset alc262_presets[] = { + [ALC262_BASIC] = { + .mixers = { alc262_base_mixer }, + .init_verbs = { alc262_init_verbs }, + .num_dacs = ARRAY_SIZE(alc262_dac_nids), + .dac_nids = alc262_dac_nids, + .hp_nid = 0x03, + .num_channel_mode = ARRAY_SIZE(alc262_modes), + .channel_mode = alc262_modes, + .input_mux = &alc262_capture_source, + }, + [ALC262_FUJITSU] = { + .mixers = { alc262_fujitsu_mixer }, + .init_verbs = { alc262_init_verbs, alc262_fujitsu_unsol_verbs }, + .num_dacs = ARRAY_SIZE(alc262_dac_nids), + .dac_nids = alc262_dac_nids, + .hp_nid = 0x03, + .dig_out_nid = ALC262_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc262_modes), + .channel_mode = alc262_modes, + .input_mux = &alc262_fujitsu_capture_source, + .unsol_event = alc262_fujitsu_unsol_event, + }, +}; + +static int patch_alc262(struct hda_codec *codec) +{ struct alc_spec *spec; + int board_config; + int err; spec = kcalloc(1, sizeof(*spec), GFP_KERNEL); if (spec == NULL) return -ENOMEM; codec->spec = spec; +#if 0 + /* pshou 07/11/05 set a zero PCM sample to DAC when FIFO is under-run */ + { + int tmp; + snd_hda_codec_write(codec, 0x1a, 0, AC_VERB_SET_COEF_INDEX, 7); + tmp = snd_hda_codec_read(codec, 0x20, 0, AC_VERB_GET_PROC_COEF, 0); + snd_hda_codec_write(codec, 0x1a, 0, AC_VERB_SET_COEF_INDEX, 7); + snd_hda_codec_write(codec, 0x1a, 0, AC_VERB_SET_PROC_COEF, tmp | 0x80); + } +#endif + + board_config = snd_hda_check_board_config(codec, alc262_cfg_tbl); + if (board_config < 0 || board_config >= ALC262_MODEL_LAST) { + printk(KERN_INFO "hda_codec: Unknown model for ALC262, trying auto-probe from BIOS...\n"); + board_config = ALC262_AUTO; + } + + if (board_config == ALC262_AUTO) { + /* automatic parse from the BIOS config */ + err = alc262_parse_auto_config(codec); + if (err < 0) { + alc_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using base mode...\n"); + board_config = ALC262_BASIC; + } + } + + if (board_config != ALC262_AUTO) + setup_preset(spec, &alc262_presets[board_config]); + + spec->stream_name_analog = "ALC262 Analog"; + spec->stream_analog_playback = &alc262_pcm_analog_playback; + spec->stream_analog_capture = &alc262_pcm_analog_capture; + + spec->stream_name_digital = "ALC262 Digital"; + spec->stream_digital_playback = &alc262_pcm_digital_playback; + spec->stream_digital_capture = &alc262_pcm_digital_capture; + + if (! spec->adc_nids && spec->input_mux) { + /* check whether NID 0x07 is valid */ + unsigned int wcap = get_wcaps(codec, 0x07); + + wcap = (wcap & AC_WCAP_TYPE) >> AC_WCAP_TYPE_SHIFT; /* get type */ + if (wcap != AC_WID_AUD_IN) { + spec->adc_nids = alc262_adc_nids_alt; + spec->num_adc_nids = ARRAY_SIZE(alc262_adc_nids_alt); + spec->mixers[spec->num_mixers] = alc262_capture_alt_mixer; + spec->num_mixers++; + } else { + spec->adc_nids = alc262_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc262_adc_nids); + spec->mixers[spec->num_mixers] = alc262_capture_mixer; + spec->num_mixers++; + } + } + + codec->patch_ops = alc_patch_ops; + if (board_config == ALC262_AUTO) + spec->init_hook = alc262_auto_init; + + return 0; +} + + +/* + * ALC861 channel source setting (2/6 channel selection for 3-stack) + */ + +/* + * set the path ways for 2 channel output + * need to set the codec line out and mic 1 pin widgets to inputs + */ +static struct hda_verb alc861_threestack_ch2_init[] = { + /* set pin widget 1Ah (line in) for input */ + { 0x0c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* set pin widget 18h (mic1/2) for input, for mic also enable the vref */ + { 0x0d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb00c }, + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x01 << 8)) }, //mic + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, (0x7000 | (0x02 << 8)) }, //line in + { } /* end */ +}; +/* + * 6ch mode + * need to set the codec line out and mic 1 pin widgets to outputs + */ +static struct hda_verb alc861_threestack_ch6_init[] = { + /* set pin widget 1Ah (line in) for output (Back Surround)*/ + { 0x0c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + /* set pin widget 18h (mic1) for output (CLFE)*/ + { 0x0d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + + { 0x0c, AC_VERB_SET_CONNECT_SEL, 0x00 }, + { 0x0d, AC_VERB_SET_CONNECT_SEL, 0x00 }, + + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb080 }, + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x01 << 8)) }, //mic + { 0x15, AC_VERB_SET_AMP_GAIN_MUTE, (0x7080 | (0x02 << 8)) }, //line in + { } /* end */ +}; + +static struct hda_channel_mode alc861_threestack_modes[2] = { + { 2, alc861_threestack_ch2_init }, + { 6, alc861_threestack_ch6_init }, +}; + +/* patch-ALC861 */ + +static struct snd_kcontrol_new alc861_base_mixer[] = { + /* output mixer control */ + HDA_CODEC_MUTE("Front Playback Switch", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Surround Playback Switch", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x05, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x05, 2, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Side Playback Switch", 0x04, 0x0, HDA_OUTPUT), + + /*Input mixer control */ + /* HDA_CODEC_VOLUME("Input Playback Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Input Playback Switch", 0x15, 0x0, HDA_OUTPUT), */ + HDA_CODEC_VOLUME("CD Playback Volume", 0x15, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x15, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x15, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x15, 0x02, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x15, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x15, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x10, 0x01, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x1a, 0x03, HDA_INPUT), + + /* Capture mixer control */ + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .count = 1, + .info = alc_mux_enum_info, + .get = alc_mux_enum_get, + .put = alc_mux_enum_put, + }, + { } /* end */ +}; + +static struct snd_kcontrol_new alc861_3ST_mixer[] = { + /* output mixer control */ + HDA_CODEC_MUTE("Front Playback Switch", 0x03, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Surround Playback Switch", 0x06, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("Center Playback Switch", 0x05, 1, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE_MONO("LFE Playback Switch", 0x05, 2, 0x0, HDA_OUTPUT), + /*HDA_CODEC_MUTE("Side Playback Switch", 0x04, 0x0, HDA_OUTPUT), */ + + /* Input mixer control */ + /* HDA_CODEC_VOLUME("Input Playback Volume", 0x15, 0x0, HDA_OUTPUT), + HDA_CODEC_MUTE("Input Playback Switch", 0x15, 0x0, HDA_OUTPUT), */ + HDA_CODEC_VOLUME("CD Playback Volume", 0x15, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("CD Playback Switch", 0x15, 0x0, HDA_INPUT), + HDA_CODEC_VOLUME("Line Playback Volume", 0x15, 0x02, HDA_INPUT), + HDA_CODEC_MUTE("Line Playback Switch", 0x15, 0x02, HDA_INPUT), + HDA_CODEC_VOLUME("Mic Playback Volume", 0x15, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Mic Playback Switch", 0x15, 0x01, HDA_INPUT), + HDA_CODEC_MUTE("Front Mic Playback Switch", 0x10, 0x01, HDA_OUTPUT), + HDA_CODEC_MUTE("Headphone Playback Switch", 0x1a, 0x03, HDA_INPUT), + + /* Capture mixer control */ + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Capture Source", + .count = 1, + .info = alc_mux_enum_info, + .get = alc_mux_enum_get, + .put = alc_mux_enum_put, + }, + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + .name = "Channel Mode", + .info = alc_ch_mode_info, + .get = alc_ch_mode_get, + .put = alc_ch_mode_put, + .private_value = ARRAY_SIZE(alc861_threestack_modes), + }, + { } /* end */ +}; + +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc861_base_init_verbs[] = { + /* + * Unmute ADC0 and set the default input to mic-in + */ + /* port-A for surround (rear panel) */ + { 0x0e, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x0e, AC_VERB_SET_CONNECT_SEL, 0x00 }, + /* port-B for mic-in (rear panel) with vref */ + { 0x0d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* port-C for line-in (rear panel) */ + { 0x0c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* port-D for Front */ + { 0x0b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x0b, AC_VERB_SET_CONNECT_SEL, 0x00 }, + /* port-E for HP out (front panel) */ + { 0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0 }, + /* route front PCM to HP */ + { 0x0f, AC_VERB_SET_CONNECT_SEL, 0x01 }, + /* port-F for mic-in (front panel) with vref */ + { 0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* port-G for CLFE (rear panel) */ + { 0x1f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x1f, AC_VERB_SET_CONNECT_SEL, 0x00 }, + /* port-H for side (rear panel) */ + { 0x20, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x20, AC_VERB_SET_CONNECT_SEL, 0x00 }, + /* CD-in */ + { 0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* route front mic to ADC1*/ + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute DAC0~3 & spdif out*/ + {0x03, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + /* Unmute Mixer 14 (mic) 1c (Line in)*/ + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* Unmute Stereo Mixer 15 */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb00c }, //Output 0~12 step + + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, // hp used DAC 3 (Front) + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + + { } +}; + +static struct hda_verb alc861_threestack_init_verbs[] = { + /* + * Unmute ADC0 and set the default input to mic-in + */ + /* port-A for surround (rear panel) */ + { 0x0e, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x00 }, + /* port-B for mic-in (rear panel) with vref */ + { 0x0d, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* port-C for line-in (rear panel) */ + { 0x0c, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* port-D for Front */ + { 0x0b, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x40 }, + { 0x0b, AC_VERB_SET_CONNECT_SEL, 0x00 }, + /* port-E for HP out (front panel) */ + { 0x0f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0xc0 }, + /* route front PCM to HP */ + { 0x0f, AC_VERB_SET_CONNECT_SEL, 0x01 }, + /* port-F for mic-in (front panel) with vref */ + { 0x10, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x24 }, + /* port-G for CLFE (rear panel) */ + { 0x1f, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x00 }, + /* port-H for side (rear panel) */ + { 0x20, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x00 }, + /* CD-in */ + { 0x11, AC_VERB_SET_PIN_WIDGET_CONTROL, 0x20 }, + /* route front mic to ADC1*/ + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + /* Unmute DAC0~3 & spdif out*/ + {0x03, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + /* Unmute Mixer 14 (mic) 1c (Line in)*/ + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* Unmute Stereo Mixer 15 */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb00c }, //Output 0~12 step + + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, // hp used DAC 3 (Front) + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + { } +}; +/* + * generic initialization of ADC, input mixers and output mixers + */ +static struct hda_verb alc861_auto_init_verbs[] = { + /* + * Unmute ADC0 and set the default input to mic-in + */ +// {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, + {0x08, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + + /* Unmute DAC0~3 & spdif out*/ + {0x03, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x04, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x05, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x06, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_MUTE}, + {0x07, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE}, + + /* Unmute Mixer 14 (mic) 1c (Line in)*/ + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x014, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x01c, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + /* Unmute Stereo Mixer 15 */ + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x15, AC_VERB_SET_AMP_GAIN_MUTE, 0xb00c}, + + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x16, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x17, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x18, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(0)}, + {0x19, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(1)}, + + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x1a, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(0)}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_MUTE(1)}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(2)}, + {0x1b, AC_VERB_SET_AMP_GAIN_MUTE, AMP_IN_UNMUTE(3)}, + + {0x08, AC_VERB_SET_CONNECT_SEL, 0x00}, // set Mic 1 + + { } +}; + +/* pcm configuration: identiacal with ALC880 */ +#define alc861_pcm_analog_playback alc880_pcm_analog_playback +#define alc861_pcm_analog_capture alc880_pcm_analog_capture +#define alc861_pcm_digital_playback alc880_pcm_digital_playback +#define alc861_pcm_digital_capture alc880_pcm_digital_capture + + +#define ALC861_DIGOUT_NID 0x07 + +static struct hda_channel_mode alc861_8ch_modes[1] = { + { 8, NULL } +}; + +static hda_nid_t alc861_dac_nids[4] = { + /* front, surround, clfe, side */ + 0x03, 0x06, 0x05, 0x04 +}; + +static hda_nid_t alc861_adc_nids[1] = { + /* ADC0-2 */ + 0x08, +}; + +static struct hda_input_mux alc861_capture_source = { + .num_items = 5, + .items = { + { "Mic", 0x0 }, + { "Front Mic", 0x3 }, + { "Line", 0x1 }, + { "CD", 0x4 }, + { "Mixer", 0x5 }, + }, +}; + +/* fill in the dac_nids table from the parsed pin configuration */ +static int alc861_auto_fill_dac_nids(struct alc_spec *spec, const struct auto_pin_cfg *cfg) +{ + int i; + hda_nid_t nid; + + spec->multiout.dac_nids = spec->private_dac_nids; + for (i = 0; i < cfg->line_outs; i++) { + nid = cfg->line_out_pins[i]; + if (nid) { + if (i >= ARRAY_SIZE(alc861_dac_nids)) + continue; + spec->multiout.dac_nids[i] = alc861_dac_nids[i]; + } + } + spec->multiout.num_dacs = cfg->line_outs; + return 0; +} + +/* add playback controls from the parsed DAC table */ +static int alc861_auto_create_multi_out_ctls(struct alc_spec *spec, + const struct auto_pin_cfg *cfg) +{ + char name[32]; + static const char *chname[4] = { "Front", "Surround", NULL /*CLFE*/, "Side" }; + hda_nid_t nid; + int i, idx, err; + + for (i = 0; i < cfg->line_outs; i++) { + nid = spec->multiout.dac_nids[i]; + if (! nid) + continue; + if (nid == 0x05) { + /* Center/LFE */ + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, "Center Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 1, 0, HDA_OUTPUT))) < 0) + return err; + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, "LFE Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 2, 0, HDA_OUTPUT))) < 0) + return err; + } else { + for (idx = 0; idx < ARRAY_SIZE(alc861_dac_nids) - 1; idx++) + if (nid == alc861_dac_nids[idx]) + break; + sprintf(name, "%s Playback Switch", chname[idx]); + if ((err = add_control(spec, ALC_CTL_BIND_MUTE, name, + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + } + } + return 0; +} + +static int alc861_auto_create_hp_ctls(struct alc_spec *spec, hda_nid_t pin) +{ + int err; + hda_nid_t nid; + + if (! pin) + return 0; + + if ((pin >= 0x0b && pin <= 0x10) || pin == 0x1f || pin == 0x20) { + nid = 0x03; + if ((err = add_control(spec, ALC_CTL_WIDGET_MUTE, "Headphone Playback Switch", + HDA_COMPOSE_AMP_VAL(nid, 3, 0, HDA_OUTPUT))) < 0) + return err; + spec->multiout.hp_nid = nid; + } + return 0; +} + +/* create playback/capture controls for input pins */ +static int alc861_auto_create_analog_input_ctls(struct alc_spec *spec, const struct auto_pin_cfg *cfg) +{ + struct hda_input_mux *imux = &spec->private_imux; + int i, err, idx, idx1; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + switch(cfg->input_pins[i]) { + case 0x0c: + idx1 = 1; + idx = 2; // Line In + break; + case 0x0f: + idx1 = 2; + idx = 2; // Line In + break; + case 0x0d: + idx1 = 0; + idx = 1; // Mic In + break; + case 0x10: + idx1 = 3; + idx = 1; // Mic In + break; + case 0x11: + idx1 = 4; + idx = 0; // CD + break; + default: + continue; + } + + err = new_analog_input(spec, cfg->input_pins[i], + auto_pin_cfg_labels[i], idx, 0x15); + if (err < 0) + return err; + + imux->items[imux->num_items].label = auto_pin_cfg_labels[i]; + imux->items[imux->num_items].index = idx1; + imux->num_items++; + } + return 0; +} + +static struct snd_kcontrol_new alc861_capture_mixer[] = { + HDA_CODEC_VOLUME("Capture Volume", 0x08, 0x0, HDA_INPUT), + HDA_CODEC_MUTE("Capture Switch", 0x08, 0x0, HDA_INPUT), + + { + .iface = SNDRV_CTL_ELEM_IFACE_MIXER, + /* The multiple "Capture Source" controls confuse alsamixer + * So call somewhat different.. + *FIXME: the controls appear in the "playback" view! + */ + /* .name = "Capture Source", */ + .name = "Input Source", + .count = 1, + .info = alc_mux_enum_info, + .get = alc_mux_enum_get, + .put = alc_mux_enum_put, + }, + { } /* end */ +}; + +static void alc861_auto_set_output_and_unmute(struct hda_codec *codec, hda_nid_t nid, + int pin_type, int dac_idx) +{ + /* set as output */ + + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, pin_type); + snd_hda_codec_write(codec, dac_idx, 0, AC_VERB_SET_AMP_GAIN_MUTE, AMP_OUT_UNMUTE); + +} + +static void alc861_auto_init_multi_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < spec->autocfg.line_outs; i++) { + hda_nid_t nid = spec->autocfg.line_out_pins[i]; + if (nid) + alc861_auto_set_output_and_unmute(codec, nid, PIN_OUT, spec->multiout.dac_nids[i]); + } +} + +static void alc861_auto_init_hp_out(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + hda_nid_t pin; + + pin = spec->autocfg.hp_pin; + if (pin) /* connect to front */ + alc861_auto_set_output_and_unmute(codec, pin, PIN_HP, spec->multiout.dac_nids[0]); +} + +static void alc861_auto_init_analog_input(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int i; + + for (i = 0; i < AUTO_PIN_LAST; i++) { + hda_nid_t nid = spec->autocfg.input_pins[i]; + if ((nid>=0x0c) && (nid <=0x11)) { + snd_hda_codec_write(codec, nid, 0, AC_VERB_SET_PIN_WIDGET_CONTROL, + i <= AUTO_PIN_FRONT_MIC ? PIN_VREF80 : PIN_IN); + } + } +} + +/* parse the BIOS configuration and set up the alc_spec */ +/* return 1 if successful, 0 if the proper config is not found, or a negative error code */ +static int alc861_parse_auto_config(struct hda_codec *codec) +{ + struct alc_spec *spec = codec->spec; + int err; + static hda_nid_t alc861_ignore[] = { 0x1d, 0 }; + + if ((err = snd_hda_parse_pin_def_config(codec, &spec->autocfg, + alc861_ignore)) < 0) + return err; + if (! spec->autocfg.line_outs) + return 0; /* can't find valid BIOS pin config */ + + if ((err = alc861_auto_fill_dac_nids(spec, &spec->autocfg)) < 0 || + (err = alc861_auto_create_multi_out_ctls(spec, &spec->autocfg)) < 0 || + (err = alc861_auto_create_hp_ctls(spec, spec->autocfg.hp_pin)) < 0 || + (err = alc861_auto_create_analog_input_ctls(spec, &spec->autocfg)) < 0) + return err; + + spec->multiout.max_channels = spec->multiout.num_dacs * 2; + + if (spec->autocfg.dig_out_pin) + spec->multiout.dig_out_nid = ALC861_DIGOUT_NID; + + if (spec->kctl_alloc) + spec->mixers[spec->num_mixers++] = spec->kctl_alloc; + + spec->init_verbs[spec->num_init_verbs++] = alc861_auto_init_verbs; - spec->mixers[spec->num_mixers] = alc882_base_mixer; + spec->num_mux_defs = 1; + spec->input_mux = &spec->private_imux; + + spec->adc_nids = alc861_adc_nids; + spec->num_adc_nids = ARRAY_SIZE(alc861_adc_nids); + spec->mixers[spec->num_mixers] = alc861_capture_mixer; spec->num_mixers++; - spec->multiout.dig_out_nid = ALC880_DIGOUT_NID; - spec->dig_in_nid = ALC880_DIGIN_NID; - spec->front_panel = 1; - spec->init_verbs = alc882_init_verbs; - spec->channel_mode = alc882_ch_modes; - spec->num_channel_mode = ARRAY_SIZE(alc882_ch_modes); + return 1; +} - spec->stream_name_analog = "ALC882 Analog"; - spec->stream_analog_playback = &alc880_pcm_analog_playback; - spec->stream_analog_capture = &alc880_pcm_analog_capture; +/* additional initialization for auto-configuration model */ +static void alc861_auto_init(struct hda_codec *codec) +{ + alc861_auto_init_multi_out(codec); + alc861_auto_init_hp_out(codec); + alc861_auto_init_analog_input(codec); +} - spec->stream_name_digital = "ALC882 Digital"; - spec->stream_digital_playback = &alc880_pcm_digital_playback; - spec->stream_digital_capture = &alc880_pcm_digital_capture; - spec->multiout.max_channels = spec->channel_mode[0].channels; - spec->multiout.num_dacs = ARRAY_SIZE(alc882_dac_nids); - spec->multiout.dac_nids = alc882_dac_nids; +/* + * configuration and preset + */ +static struct hda_board_config alc861_cfg_tbl[] = { + { .modelname = "3stack", .config = ALC861_3ST }, + { .pci_subvendor = 0x8086, .pci_subdevice = 0xd600, .config = ALC861_3ST }, + { .modelname = "3stack-dig", .config = ALC861_3ST_DIG }, + { .modelname = "6stack-dig", .config = ALC861_6ST_DIG }, + { .modelname = "auto", .config = ALC861_AUTO }, + {} +}; - spec->input_mux = &alc882_capture_source; - spec->num_adc_nids = ARRAY_SIZE(alc882_adc_nids); - spec->adc_nids = alc882_adc_nids; +static struct alc_config_preset alc861_presets[] = { + [ALC861_3ST] = { + .mixers = { alc861_3ST_mixer }, + .init_verbs = { alc861_threestack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc861_dac_nids), + .dac_nids = alc861_dac_nids, + .num_channel_mode = ARRAY_SIZE(alc861_threestack_modes), + .channel_mode = alc861_threestack_modes, + .num_adc_nids = ARRAY_SIZE(alc861_adc_nids), + .adc_nids = alc861_adc_nids, + .input_mux = &alc861_capture_source, + }, + [ALC861_3ST_DIG] = { + .mixers = { alc861_base_mixer }, + .init_verbs = { alc861_threestack_init_verbs }, + .num_dacs = ARRAY_SIZE(alc861_dac_nids), + .dac_nids = alc861_dac_nids, + .dig_out_nid = ALC861_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc861_threestack_modes), + .channel_mode = alc861_threestack_modes, + .num_adc_nids = ARRAY_SIZE(alc861_adc_nids), + .adc_nids = alc861_adc_nids, + .input_mux = &alc861_capture_source, + }, + [ALC861_6ST_DIG] = { + .mixers = { alc861_base_mixer }, + .init_verbs = { alc861_base_init_verbs }, + .num_dacs = ARRAY_SIZE(alc861_dac_nids), + .dac_nids = alc861_dac_nids, + .dig_out_nid = ALC861_DIGOUT_NID, + .num_channel_mode = ARRAY_SIZE(alc861_8ch_modes), + .channel_mode = alc861_8ch_modes, + .num_adc_nids = ARRAY_SIZE(alc861_adc_nids), + .adc_nids = alc861_adc_nids, + .input_mux = &alc861_capture_source, + }, +}; - codec->patch_ops = alc_patch_ops; +static int patch_alc861(struct hda_codec *codec) +{ + struct alc_spec *spec; + int board_config; + int err; + + spec = kcalloc(1, sizeof(*spec), GFP_KERNEL); + if (spec == NULL) + return -ENOMEM; + + codec->spec = spec; + + board_config = snd_hda_check_board_config(codec, alc861_cfg_tbl); + if (board_config < 0 || board_config >= ALC861_MODEL_LAST) { + printk(KERN_INFO "hda_codec: Unknown model for ALC861, trying auto-probe from BIOS...\n"); + board_config = ALC861_AUTO; + } + + if (board_config == ALC861_AUTO) { + /* automatic parse from the BIOS config */ + err = alc861_parse_auto_config(codec); + if (err < 0) { + alc_free(codec); + return err; + } else if (! err) { + printk(KERN_INFO "hda_codec: Cannot set up configuration from BIOS. Using base mode...\n"); + board_config = ALC861_3ST_DIG; + } + } + + if (board_config != ALC861_AUTO) + setup_preset(spec, &alc861_presets[board_config]); + + spec->stream_name_analog = "ALC861 Analog"; + spec->stream_analog_playback = &alc861_pcm_analog_playback; + spec->stream_analog_capture = &alc861_pcm_analog_capture; + + spec->stream_name_digital = "ALC861 Digital"; + spec->stream_digital_playback = &alc861_pcm_digital_playback; + spec->stream_digital_capture = &alc861_pcm_digital_capture; + + codec->patch_ops = alc_patch_ops; + if (board_config == ALC861_AUTO) + spec->init_hook = alc861_auto_init; + return 0; } @@ -1497,7 +5716,11 @@ static int patch_alc882(struct hda_codec *codec) */ struct hda_codec_preset snd_hda_preset_realtek[] = { { .id = 0x10ec0260, .name = "ALC260", .patch = patch_alc260 }, + { .id = 0x10ec0262, .name = "ALC262", .patch = patch_alc262 }, { .id = 0x10ec0880, .name = "ALC880", .patch = patch_alc880 }, { .id = 0x10ec0882, .name = "ALC882", .patch = patch_alc882 }, + { .id = 0x10ec0883, .name = "ALC883", .patch = patch_alc882 }, + { .id = 0x10ec0885, .name = "ALC885", .patch = patch_alc882 }, + { .id = 0x10ec0861, .name = "ALC861", .patch = patch_alc861 }, {} /* terminator */ }; diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c index 8c440fb98..d8622951c 100644 --- a/sound/pci/hda/patch_sigmatel.c +++ b/sound/pci/hda/patch_sigmatel.c @@ -1262,13 +1262,13 @@ static int vaio_master_sw_put(struct snd_kcontrol *kcontrol, int change; change = snd_hda_codec_amp_update(codec, 0x02, 0, HDA_OUTPUT, 0, - 0x80, valp[0] & 0x80); + 0x80, (valp[0] ? 0 : 0x80)); change |= snd_hda_codec_amp_update(codec, 0x02, 1, HDA_OUTPUT, 0, - 0x80, valp[1] & 0x80); + 0x80, (valp[1] ? 0 : 0x80)); snd_hda_codec_amp_update(codec, 0x05, 0, HDA_OUTPUT, 0, - 0x80, valp[0] & 0x80); + 0x80, (valp[0] ? 0 : 0x80)); snd_hda_codec_amp_update(codec, 0x05, 1, HDA_OUTPUT, 0, - 0x80, valp[1] & 0x80); + 0x80, (valp[1] ? 0 : 0x80)); return change; } diff --git a/sound/pci/rme9652/hdsp.c b/sound/pci/rme9652/hdsp.c index 61f82f0d5..10586e48a 100644 --- a/sound/pci/rme9652/hdsp.c +++ b/sound/pci/rme9652/hdsp.c @@ -389,7 +389,7 @@ MODULE_SUPPORTED_DEVICE("{{RME Hammerfall-DSP}," /* use hotplug firmeare loader? */ #if defined(CONFIG_FW_LOADER) || defined(CONFIG_FW_LOADER_MODULE) -#ifndef HDSP_USE_HWDEP_LOADER +#if !defined(HDSP_USE_HWDEP_LOADER) && !defined(CONFIG_SND_HDSP) #define HDSP_FW_LOADER #endif #endif @@ -3169,9 +3169,10 @@ snd_hdsp_proc_read(struct snd_info_entry *entry, struct snd_info_buffer *buffer) char *clock_source; int x; - if (hdsp_check_for_iobox (hdsp)) + if (hdsp_check_for_iobox (hdsp)) { snd_iprintf(buffer, "No I/O box connected.\nPlease connect one and upload firmware.\n"); return; + } if (hdsp_check_for_firmware(hdsp, 0)) { if (hdsp->state & HDSP_FirmwareCached) { -- 2.47.0