Andy Bavier [Wed, 27 Oct 2004 21:09:03 +0000 (21:09 +0000)]
Commented out the 'interactive bonus' and 'cvt saving' when updating cvt
Andy Bavier [Wed, 27 Oct 2004 21:07:25 +0000 (21:07 +0000)]
Commented out the contribution of 'urgency' to a class's effective priority
Andy Bavier [Wed, 27 Oct 2004 21:06:11 +0000 (21:06 +0000)]
Increased size of CLASSQUEUE from 128 to 1024 to support CPU shares of 4 or greater
Steve Muir [Wed, 27 Oct 2004 20:12:10 +0000 (20:12 +0000)]
Permit hard-linking across bind-mounts as long as the source dir is not
read-only [PL #2624]
Mark Huang [Wed, 27 Oct 2004 15:09:37 +0000 (15:09 +0000)]
add miscellaneous tags
Marc Fiuczynski [Sat, 23 Oct 2004 20:13:55 +0000 (20:13 +0000)]
Disabled surplus allocation feature of the CKRM cpu scheduler.
See PL2642 for further details.
Marc Fiuczynski [Fri, 22 Oct 2004 06:26:45 +0000 (06:26 +0000)]
Increased CKRM_SHARE_ACCURACY from (1<<10) to (1<<13). This resolves PL2620.
Marc Fiuczynski [Thu, 21 Oct 2004 17:54:19 +0000 (17:54 +0000)]
Fix to prevent kernel from crashing in the memory controller on shutdown.
Mark Huang [Wed, 20 Oct 2004 02:33:41 +0000 (02:33 +0000)]
- PL2525 fix: implement SO_SETXID (reset the context ID of an existing
socket). Callable by root context only. To avoid having to register a
new sockopt number, #define SO_SETXID to be SO_PASSCRED (which is not
usually callable by setsockopt()).
Mark Huang [Wed, 20 Oct 2004 02:33:15 +0000 (02:33 +0000)]
- PL2525 fix: implement SO_SETXID (reset the context ID of an existing
socket). Callable by root context only. To avoid having to register a
new sockopt number, #define SO_SETXID to be SO_PASSCRED (which is not
usually callable by setsockopt()):
#if !defined(SO_SETXID)
#define SO_SETXID SO_PASSCRED
#endif
int xid;
setsockopt(fd, SOL_SOCKET, SO_SETXID, &xid, sizeof(xid));
Marc Fiuczynski [Tue, 19 Oct 2004 23:21:35 +0000 (23:21 +0000)]
have been renamed to ckrm_numtasks by the CKRM folks
Marc Fiuczynski [Tue, 19 Oct 2004 23:05:37 +0000 (23:05 +0000)]
Reenabled CKRM memory controller.
Marc Fiuczynski [Tue, 19 Oct 2004 22:55:10 +0000 (22:55 +0000)]
disabled code for CKRM memory class shrinking
Mark Huang [Mon, 18 Oct 2004 17:45:31 +0000 (17:45 +0000)]
- fix printing of conntrack owners
Marc Fiuczynski [Mon, 18 Oct 2004 13:24:29 +0000 (13:24 +0000)]
version 9.1 of the ckrm cpu scheduler.
Marc Fiuczynski [Fri, 15 Oct 2004 20:55:24 +0000 (20:55 +0000)]
commented out unnecessary BUG_ON statement
Marc Fiuczynski [Fri, 15 Oct 2004 20:53:04 +0000 (20:53 +0000)]
typo
Marc Fiuczynski [Fri, 15 Oct 2004 20:51:06 +0000 (20:51 +0000)]
enabled ckrm memory controller
Marc Fiuczynski [Mon, 11 Oct 2004 20:26:18 +0000 (20:26 +0000)]
Config file to be used for "kernel-desktop" rpm package. It includes a
bunch of stuff -- such as USB, cpufreq, and laptop support that we
currently do not wish to add to the standard
kernel-2.6.8-i686-planetlab.config file.
Mark Huang [Sun, 10 Oct 2004 22:24:53 +0000 (22:24 +0000)]
PL2340 workaround: disable CKRM memory controller
Mark Huang [Fri, 1 Oct 2004 20:10:17 +0000 (20:10 +0000)]
- initialize connection contexts to -1 (== unknown)
Mark Huang [Fri, 1 Oct 2004 20:09:49 +0000 (20:09 +0000)]
- Fix thinko. Bypass xid checks if socket was created in the root
context, like the freaking comment says.
Mark Huang [Fri, 1 Oct 2004 17:59:41 +0000 (17:59 +0000)]
- panic on OOM condition instead of killing processes
- compile hangcheck-timer into kernel
Mark Huang [Fri, 1 Oct 2004 17:55:04 +0000 (17:55 +0000)]
- reboot by default
Mark Huang [Fri, 1 Oct 2004 17:54:48 +0000 (17:54 +0000)]
- Just panic() instead of the default behavior of selecting processes
for death.
Mark Huang [Fri, 1 Oct 2004 17:53:49 +0000 (17:53 +0000)]
- add option to panic when OOM instead of killing processes
Mark Huang [Fri, 1 Oct 2004 17:52:44 +0000 (17:52 +0000)]
- delay 15 minutes before rebooting the machine after a panic
Mark Huang [Fri, 1 Oct 2004 17:47:57 +0000 (17:47 +0000)]
- export vx_debug_nid and vx_debug_xid which are necessary if, for
instance, af_packet.o is compiled as a module
Marc Fiuczynski [Thu, 30 Sep 2004 12:45:42 +0000 (12:45 +0000)]
CKRM cpu controller version 8.2.
Marc Fiuczynski [Tue, 28 Sep 2004 16:53:02 +0000 (16:53 +0000)]
Fixed compilation warnings in the CKRM cpu controller.
Marc Fiuczynski [Tue, 28 Sep 2004 16:52:36 +0000 (16:52 +0000)]
Merged to version 4 of the CKRM io controller.
Fixed some compilation warnings in the CKRM cpu controller.
Marc Fiuczynski [Tue, 28 Sep 2004 07:49:30 +0000 (07:49 +0000)]
version 1 of the CKRM memory controller.
Marc Fiuczynski [Tue, 28 Sep 2004 06:48:25 +0000 (06:48 +0000)]
Initial revision
Marc Fiuczynski [Tue, 28 Sep 2004 06:25:53 +0000 (06:25 +0000)]
CKRM CPU MONITOR now part of CKCRM CPU SCHEDULE
Marc Fiuczynski [Tue, 28 Sep 2004 06:14:31 +0000 (06:14 +0000)]
Merge in the version 7 cpu controller from CKRM.
- Had to shuffle the runqueue datastructure from sched.h to sched.c.
- Fixed arithmetic fault that would occur early in boot due to uninitialized
CKRM cpu controller data structure. Odd that the CKRM folks didn't have
this bug, too.
Marc Fiuczynski [Sat, 25 Sep 2004 10:49:11 +0000 (10:49 +0000)]
CKRM E16rc1 memory controller configuration
Marc Fiuczynski [Fri, 24 Sep 2004 20:37:00 +0000 (20:37 +0000)]
CKRM e16 memory controller.
This contains a fix I made to make it work with initrd. A variant of
this fix will be incorporated in the next release of the CKRM memory
controller.
Marc Fiuczynski [Thu, 23 Sep 2004 19:26:45 +0000 (19:26 +0000)]
CKRM E16rc1 io controller support.
This version only implements 20 classes of I/O priority. The next release
candidate will support limits, which can then be used as a means to express
proportional sharing between classes (vservers).
The cfq-iosched.c function was hacked by me to work around an issue with
Fedora Core 2 that does not appear in the vanilla 2.6.8.1 kernel. The
CKRM folks are aware of this and will fix this problem (hopefully in the
next release, too).
Andy Bavier [Thu, 23 Sep 2004 19:23:55 +0000 (19:23 +0000)]
Build cls_fw module for bandwidth scheduler
Marc Fiuczynski [Thu, 23 Sep 2004 13:16:46 +0000 (13:16 +0000)]
accidentally imported
Marc Fiuczynski [Thu, 23 Sep 2004 11:47:02 +0000 (11:47 +0000)]
Initial revision
Marc Fiuczynski [Thu, 23 Sep 2004 00:19:46 +0000 (00:19 +0000)]
ckrm_E16 release candidate 1 merge
Marc Fiuczynski [Wed, 22 Sep 2004 12:12:05 +0000 (12:12 +0000)]
This commit was generated by cvs2svn to compensate for changes in r180,
which included commits to RCS files with non-trunk default branches.
Marc Fiuczynski [Wed, 22 Sep 2004 12:12:03 +0000 (12:12 +0000)]
Initial revision
Mark Huang [Mon, 20 Sep 2004 14:46:49 +0000 (14:46 +0000)]
- include the Fedora release number so that Fedora kernels don't defeat
ours in a yum update cage match
Marc Fiuczynski [Sat, 18 Sep 2004 11:50:22 +0000 (11:50 +0000)]
enabled CKRM
Marc Fiuczynski [Sat, 18 Sep 2004 11:48:31 +0000 (11:48 +0000)]
CKRM e15 IO controller framework --- cfq-iosched.c remains unchanged for now, hence the controller does nothing until we get a new, ported version of this file
Marc Fiuczynski [Sat, 18 Sep 2004 10:53:07 +0000 (10:53 +0000)]
bind mount patch (patch-2.6.8.1-bme0.05.1.diff)
Mark Huang [Fri, 17 Sep 2004 19:17:39 +0000 (19:17 +0000)]
- don't require kernel-utils yet until we move to FC3
Mark Huang [Fri, 17 Sep 2004 18:08:38 +0000 (18:08 +0000)]
Dammit, UML patch keeps adding generated files
Mark Huang [Fri, 17 Sep 2004 18:06:14 +0000 (18:06 +0000)]
- fix printing of conntrack xid tags
Mark Huang [Fri, 17 Sep 2004 18:03:39 +0000 (18:03 +0000)]
- remove warning, VServer 1.9.2 patch appears to be aware of this change
now
Mark Huang [Fri, 17 Sep 2004 18:00:35 +0000 (18:00 +0000)]
Merge to User Mode Linux 2.6.8.1-1um (uml-patch-2.6.8.1-1)
Mark Huang [Fri, 17 Sep 2004 15:28:58 +0000 (15:28 +0000)]
Initial revision
Mark Huang [Fri, 17 Sep 2004 04:02:26 +0000 (04:02 +0000)]
Add default 2.6.8 configuration for PlanetLab Linux
Mark Huang [Fri, 17 Sep 2004 02:53:31 +0000 (02:53 +0000)]
Merge to VServer 1.9.2 (patch-2.6.8.1-vs1.9.2.diff)
Mark Huang [Thu, 16 Sep 2004 18:47:16 +0000 (18:47 +0000)]
Initial revision
Mark Huang [Thu, 16 Sep 2004 18:16:15 +0000 (18:16 +0000)]
Merge to Fedora Core 2 kernel-2.6.8-1.521
Marc Fiuczynski [Wed, 15 Sep 2004 20:47:57 +0000 (20:47 +0000)]
Initial revision
Marc Fiuczynski [Wed, 15 Sep 2004 19:47:30 +0000 (19:47 +0000)]
Initial revision
Marc Fiuczynski [Wed, 15 Sep 2004 17:11:56 +0000 (17:11 +0000)]
This commit was generated by cvs2svn to compensate for changes in r152,
which included commits to RCS files with non-trunk default branches.
Marc Fiuczynski [Wed, 15 Sep 2004 17:11:54 +0000 (17:11 +0000)]
This commit was generated by cvs2svn to compensate for changes in r149,
which included commits to RCS files with non-trunk default branches.
Marc Fiuczynski [Wed, 15 Sep 2004 17:11:52 +0000 (17:11 +0000)]
This commit was generated by cvs2svn to compensate for changes in r146,
which included commits to RCS files with non-trunk default branches.
Marc Fiuczynski [Wed, 15 Sep 2004 17:11:48 +0000 (17:11 +0000)]
This commit was generated by cvs2svn to compensate for changes in r141,
which included commits to RCS files with non-trunk default branches.
Marc Fiuczynski [Wed, 15 Sep 2004 17:00:00 +0000 (17:00 +0000)]
Initial revision
Mark Huang [Tue, 14 Sep 2004 15:02:02 +0000 (15:02 +0000)]
- add /boot/Kerntypes-x.x.x to the files list
Mark Huang [Mon, 13 Sep 2004 22:29:33 +0000 (22:29 +0000)]
- install /boot/Kerntypes for LKCD
Marc Fiuczynski [Mon, 13 Sep 2004 03:06:06 +0000 (03:06 +0000)]
Initial revision
Mark Huang [Sun, 12 Sep 2004 04:58:31 +0000 (04:58 +0000)]
- fix merge of LKCD patch
Mark Huang [Sun, 12 Sep 2004 04:37:35 +0000 (04:37 +0000)]
- Be more like FC2 w.r.t. BSD process accounting support, kallsyms
support, misc binary support, optimization, kernel debugging options
- enable hangcheck-timer instead of softdog; hangcheck-timer uses the
TSC intead of jiffies (heavy I/O can infinitely delay jiffy count),
and requires no user mode support (beyond loading the module)
- enable basic (437, ISO-8859-1, UTF8) NLS support to quell warnings
Mark Huang [Sun, 12 Sep 2004 03:12:54 +0000 (03:12 +0000)]
Compile LKCD support
Mark Huang [Sun, 12 Sep 2004 03:11:14 +0000 (03:11 +0000)]
Merge LKCD 2.6 tree at :pserver:anonymous@cvs.sourceforge.net:/cvsroot/lkcd/2.6 as of
20040910
Mark Huang [Fri, 10 Sep 2004 19:18:37 +0000 (19:18 +0000)]
Initial revision
Mark Huang [Fri, 10 Sep 2004 15:15:24 +0000 (15:15 +0000)]
Kernel crashes, back out ckrm-e15-bugfixes-merge for now
Marc Fiuczynski [Wed, 8 Sep 2004 19:40:51 +0000 (19:40 +0000)]
Initial revision
Mark Huang [Wed, 8 Sep 2004 18:48:46 +0000 (18:48 +0000)]
- disable loopback crypto block device so that we can disable in-kernel
crypto. Let's maintain some semblance of exportability here.
Mark Huang [Wed, 8 Sep 2004 18:33:34 +0000 (18:33 +0000)]
- enable support for tracing thread mode (not everyone has SKAS patched
into their host kernel like I do...)
- HumFS/HPPFS renamed to ExternFS
- disable Ethernet, NE2000, and RAID support (not sure why this was in
there in the first place)
- disable in-kernel crypto support
- enable IPOD by default
- enable /proc/config.gz
- disable NLS support
- remove more junk that we don't support or configure: IP
multicast/MBONE, TCP SYN cookies, all QoS/queueing modules except HTB
Mark Huang [Wed, 8 Sep 2004 18:20:38 +0000 (18:20 +0000)]
- remove more junk that we don't support or configure: IP
multicast/MBONE, TCP SYN cookies, all QoS/queueing modules except HTB
Mark Huang [Wed, 8 Sep 2004 17:57:51 +0000 (17:57 +0000)]
- enable IPOD by default
- enable /proc/config.gz
- disable NLS support
Mark Huang [Wed, 8 Sep 2004 17:46:51 +0000 (17:46 +0000)]
- port, cleanup Emulab ICMP Ping Of Death (IPOD) patch from 2.4.x
Mark Huang [Wed, 8 Sep 2004 17:45:22 +0000 (17:45 +0000)]
- fix a pretty fundamental bug in how context IDs are associated with
connections. In the loopback case, you need to associate each loopback
connection with two contexts, the original sender's and the original
receiver's (or, equivalently, the replying sender's, which is why you
see !dir in the NF_IP_LOCAL_IN case). In the common case (sending from
a local process to a remote machine), the code need only refer to the
original sender's context ID.
Marc Fiuczynski [Tue, 7 Sep 2004 19:54:07 +0000 (19:54 +0000)]
various fixes to ckrm core and the cpu controller
Mark Huang [Wed, 25 Aug 2004 18:40:36 +0000 (18:40 +0000)]
Silently drop if VNET is active (if INET bind() has been
overridden) and the context is not entitled to read the
packet.
If a socket has been closed, the xid is set to -1; in this case,
consider the socket to be owned by root and allow the packet to pass.
Mark Huang [Wed, 25 Aug 2004 18:21:49 +0000 (18:21 +0000)]
- hard-code builder as support@planet-lab.org
Mark Huang [Fri, 20 Aug 2004 19:33:57 +0000 (19:33 +0000)]
Merge to uml-2.6.7-2um
Mark Huang [Fri, 20 Aug 2004 18:50:31 +0000 (18:50 +0000)]
Initial revision
Mark Huang [Wed, 18 Aug 2004 19:44:51 +0000 (19:44 +0000)]
- The bootcd environment does not have device mapper enabled, and newer
mkinitrd scripts depend upon the existence of "/dev/mapper" somewhere in
the rootdev name to enable root LVM support. The initrd image generated
by the initial kernel RPM installation thus ends up lacking root LVM
support.
Mark Huang [Tue, 17 Aug 2004 20:26:46 +0000 (20:26 +0000)]
- for some reason, conf can't find libkconfig.so in a chroot jail. Help
it along.
Mark Huang [Tue, 17 Aug 2004 03:42:12 +0000 (03:42 +0000)]
- force BuildArchitecture to i686
Mark Huang [Wed, 11 Aug 2004 21:24:43 +0000 (21:24 +0000)]
- nuke all_x86
Mark Huang [Wed, 11 Aug 2004 21:21:20 +0000 (21:21 +0000)]
- restore accidentally removed "%define signmodules 0"
- building on Mandrake unsupported for now anyway...
- too much trouble to figure out how to specify the correct --target
argument to RPM on different kinds of build machines. PlanetLab is
i686 only anyway, who cares what arch RPM thinks the package should
be.
- make buildcheck works on Fedora Core 2. Go figure.
- %post: make PlanetLab links
Mark Huang [Tue, 10 Aug 2004 19:40:59 +0000 (19:40 +0000)]
- too many problems cross-building this on Mandrake and RH9. Restore all
the FC2 prereqs, and require that this RPM be built on FC2 (build on
jeeves in the /export0/fedora chroot jail)
- build in BUILD/linux-2.6.7 instead of BUILD/kernel-2.6.7/linux-2.6.7
Mark Huang [Tue, 10 Aug 2004 16:53:26 +0000 (16:53 +0000)]
- try removing module-init-tools prereq
Mark Huang [Mon, 9 Aug 2004 21:30:26 +0000 (21:30 +0000)]
- first cut at an adapted FC2 kernel-2.6 spec file
Mark Huang [Mon, 9 Aug 2004 21:22:34 +0000 (21:22 +0000)]
- just use a single PlanetLab release number
Mark Huang [Thu, 5 Aug 2004 19:43:47 +0000 (19:43 +0000)]
remove for now; configs in this directory must be in the form kernel-%{_target_cpu}-[other stuff].config
Mark Huang [Wed, 4 Aug 2004 20:25:34 +0000 (20:25 +0000)]
remove VServer ipv4root checks in favor of VNET limitations
Mark Huang [Wed, 4 Aug 2004 20:10:45 +0000 (20:10 +0000)]
- rename uml config to better fit spec file expectations
Mark Huang [Wed, 4 Aug 2004 16:44:56 +0000 (16:44 +0000)]
Merge to official uml-patch-2.6.7-1um
Mark Huang [Wed, 4 Aug 2004 15:32:38 +0000 (15:32 +0000)]
Initial revision