3 Broadcom BCM5700 Linux Driver
11 Copyright (c) 2000-2006 Broadcom Corporation
21 Installing Source RPM Package
22 Building Driver From TAR File
23 Patching Driver Into Kernel
24 Unloading and Removing Driver
33 This file describes the Linux driver for the Broadcom NetXtreme BCM5700
34 series 10/100/1000 Mbps PCI/PCI-X/PCI Express Ethernet Network Controllers.
40 The current version of the driver has been tested on the latest Red Hat, SuSE,
41 and other Linux distributions for i386, ia64, and x86_64 CPU architectures
42 using 2.4.x and 2.6.x kernels. The driver has been tested up to kernel
43 version 2.4.31 and 2.6.13.
45 The driver should work on other little endian or big endian CPU architectures,
46 but only very limited testing has been done on some of these machines. The
47 Makefile may have to be modified to include architecture specific compile
48 switches, and some minor changes in the source files may also be required.
49 On these machines, patching the driver into the kernel is recommended (see
50 below for instructions).
56 The driver is released in two packaging formats: source RPM and compressed tar
57 formats. The file names for the two packages are bcm5700-<version>.src.rpm and
58 bcm5700-<version>.tar.gz respectively. Identical source files to build the
59 driver are included in both packages. bcm_sup-<version> is a seperate tar file
60 that contains additional utilities such as patches and driver diskette images
61 for network installation.
64 Installing Source RPM Package
65 =============================
67 The following are general guidelines for installing the driver. Refer to
68 DISTRIB.TXT for additional installation notes for various Linux distributions.
70 1. Install the source RPM package:
72 rpm -ivh bcm5700-<version>.src.rpm
74 2. CD to the RPM path and build the binary driver for your kernel:
76 cd /usr/src/{redhat,OpenLinux,turbo,packages,rpm ..}
78 rpm -bb SPECS/bcm5700.spec
82 rpmbuild -bb SPECS/bcm5700.spec (for RPM version 4.x.x)
84 Note that the RPM path is different for different Linux distributions.
86 3. Install the newly built package (driver and man page):
88 rpm -ivh RPMS/i386/bcm5700-<version>.i386.rpm
90 Note that the --force option is needed if installing on Red Hat 7.2, 7.3,
91 2.1AS and others that already contain an older version of the driver.
93 The driver will be installed in the following path:
97 /lib/modules/<kernel_version>/net/bcm5700.o
101 /lib/modules/<kernel_version>/kernel/drivers/net/bcm5700.o
103 2.4.x kernels with bcm5700 as an addon driver (e.g. Red Hat 7.3, 2.1AS):
105 /lib/modules/<kernel_version>/kernel/drivers/net/bcm/bcm5700.o
109 /lib/modules/<kernel_version>/kernel/drivers/addon/bcm5700/bcm5700.o
113 /lib/modules/<kernel_version>/kernel/drivers/net/bcm5700.ko
115 2.6.x kernels with bcm5700 as an addon driver (e.g. SuSE 9 SLES):
117 /lib/modules/<kernel_version>/kernel/drivers/net/bcm/bcm5700.ko
123 If loading the driver on Red Hat 3 update 4, 2.1 AS or other newer kernels with the
124 tg3 driver, refer to the "Removing tg3 Driver" in DISTRIB.TXT before loading
127 5. To configure network protocol and address, refer to various Linux
131 Building Driver From TAR File
132 =============================
134 The following are general guidelines for installing the driver. Refer to
135 DISTRIB.TXT for additional installation notes for various Linux distributions.
137 1. Create a directory and extract the files:
139 tar xvzf bcm5700-<version>.tar.gz
141 2. Build the driver bcm5700.o as a loadable module for the running kernel:
146 3. Test the driver by loading it:
150 insmod bcm5700.ko (on 2.6.x kernels)
152 If loading the driver on Red Hat 3 update 4, 2.1 AS or other newer kernels with the
153 tg3 driver, refer to the "Removing tg3 Driver" in DISTRIB.TXT before loading
156 4. Install the driver and man page:
160 See RPM instructions above for the location of the installed driver.
162 5. To configure network protocol and address, refer to various Linux
166 Patching Driver Into Kernel (Optional)
167 ======================================
169 Patch files are included for patching the driver into some of the latest
170 2.4.x and 2.6.x kernel source trees. This step is optional and can be done
171 by users familiar with configuring and building the kernel. The patch
172 will modify the orginal kernel's source code.
174 The following steps may allow you to patch the driver into a kernel:
176 1. Select the patch file that matches your kernel and apply the patch:
178 patch -p1 -d <kernel_src_root> < bcm5700-<version>-2.4.<x>.patch
179 patch -p1 -d <kernel_src_root> < bcm5700-<version>-2.6.<x>.patch
181 where <version> is the version of the bcm5700 driver and 2.4.<x> is the
182 version of the kernel to patch (e.g. 2.4.31).
184 Note: <kernel_src_root> is usually /usr/src/linux or /usr/src/linux-2.4.<x>
185 Note: <kernel_src_root> is usually /usr/src/linux or /usr/src/linux-2.6.<x>
187 2. Configure the kernel to include the bcm5700 driver. It can be found
188 under Network Device Support ---> Ethernet (1000 Mbit) ---> Broadcom BCM5700
189 support when make menuconfig is run. Select built-in or module for the driver:
194 3. Compile the kernel:
202 Unloading and Removing Driver
203 =============================
205 To unload the driver, use ifconfig to bring down all eth# interfaces opened
206 by the driver, then do the following:
211 If the driver was installed using rpm, do the following to remove it:
216 If the driver was installed using make install from the tar file, the driver
217 bcm5700.o has to be manually deleted from the system. Refer to the section
218 "Installing Source RPM Package" for the location of the installed driver.
224 Optional parameters for the driver can be supplied as command line arguments
225 to the insmod command. Typically, these parameters are set in the file
226 /etc/modules.conf (see the man page for modules.conf). These parameters take
229 <parameter>=value[,value,...]
231 where the multiple values for the same parameter are for multiple NICs
232 installed in the system.
234 Note that default or other meaningful values will be used when invalid values
235 are selected. Some combinations of parameter values may conflict and lead to
236 failures. The driver cannot detect all such conflicting combinations.
238 All the parameters are listed below.
242 Selects the line speed of the link. This parameter is used together with
243 full_duplex and auto_speed to select the speed and duplexity of the link
244 and the setting of autonegotiation.
246 The valid values are:
248 0 Autonegotiate for highest speed supported by link partner (default)
253 If line_speed is set to 10, 100, or 1000, the NIC will autonegotiate for
254 the selected speed (and selected duplexity) if auto_speed is set to 1.
255 If auto_speed is set to 0, the selected speed and duplexity will be
256 set without autonegotiation. Note that 1000 Mbps must be negotiated for
257 copper twisted pair links.
261 Enables or disables autonegotiation. The valid values are:
263 0 Autonegotiation disabled
264 1 Autonegotiation enabled (default)
266 Note that this parameter is ignored and assumed 1 if line_speed is set
271 Selects the duplexity of the link. This paramter is used together with
272 line_speed to select the speed and duplexity of the link. Note that this
273 parameter is ignored if line_speed is 0.
275 The valid values are:
278 1 full duplex (default)
283 Enables or disables receiving flow control (pause) frames. This parameter
284 is used together with auto_flow_control. The valid values are:
286 0 pause receive disabled
287 1 pause receive enabled if auto_flow_control is set to 0, or
288 pause receive advertised if auto_flow_control is set to 1 (default)
292 Enables or disables transmitting flow control (pause) frames. This parameter
293 is used together with auto_flow_control. The valid values are:
295 0 pause transmit disabled
296 1 pause transmit enabled if auto_flow_control is set to 0, or
297 pause transmit advertised if auto_flow_control is set to 1 (default)
301 Enables or disables autonegotiation of flow control. This parameter is used
302 together with rx_flow_control and tx_flow_control to determine the
303 advertised flow control capability. The valid values are:
305 0 flow control autonegotiation disabled
306 1 flow control autonegotiation enabled with capability specified in
307 rx_flow_control and tx_flow_control (only valid if line_speed is
308 set to 0 or auto_speed is set to 1) (default)
312 Enables jumbo frames up to the specified MTU size. The valid range for
313 this parameter is 1500 to 9000. Default is 1500 which is standard
314 ethernet (non-jumbo) MTU size. Note that the MTU size excludes the
315 ethernet header size of 14 bytes. Actual frame size is MTU size + 14 bytes.
316 Jumbo MTU sizes are not supported on BCM5705 chips.
318 The MTU size can also be changed using ifconfig after the driver is loaded.
319 See the ifconfig man page for details.
323 Enables or disables hardware transmit TCP/UDP checksum. The valid values
327 1 checksum enabled (default)
331 Enables or disables hardware receive TCP/UDP checksum validation. The
335 1 checksum enabled (default)
339 Enables or disables scatter-gather and 64-bit DMA on x86. This option is
340 only useful when running on TUX-enabled kernels or newer kernels with
341 zero-copy TCP. The valid values are:
343 0 scatter-gather and 64-bit DMA on x86 disabled
344 1 scatter-gather and 64-bit DMA on x86 enabled (default)
348 Enables either NIC based or host based transmit buffer descriptors (Tx BDs).
349 NIC based Tx BDs may be slightly faster on certain machines on earlier
350 2.4 kernels where each transmit packet is usually entirely contiguous. On
351 later kernels with scatter-gather and TCP segmentation option, host based
352 Tx BDs using DMA transfer are usually faster. NIC based Tx BDs are not
353 supported on 5705 and later family controllers. The valid values are:
355 0 NIC based transmit buffer descriptors disabled (using host based
356 transmit buffer descriptors) (default)
357 1 NIC based transmit buffer descriptors enabled (not supported
358 on 5705 family controllers)
362 Configures the number of transmit descriptors. Default is 120. The
363 valid range is from 1 to 511. Note that the driver may not be able to
364 allocate the required amount of memory if this parameter is set too high.
365 Depending on kernel and CPU architecture, each descriptor may require up
366 to about 268 bytes. This parameter should not be set less than 80 if
367 adaptive_coalesce (see below) is enabled.
371 Configures the number of receive descriptors for frames up to 1528 bytes.
372 Default is 200. The valid range is from 1 to 511. This parameter should
373 not be set less than 80 on systems with high network traffic. Setting this
374 parameter higher allows the NIC to buffer larger bursts of network
375 traffic without dropping frames, especially on slower systems. Note that
376 the driver may not be able to allocate the required amount of memory if
377 this parameter is set too high. Depending on kernel and CPU architecture,
378 each descriptor may require up to about 268 bytes. Each descriptor also
379 requires a socket buffer of at least 1536 bytes. This parameter should not
380 be set less than 50 if adaptive_coalesce (see below) is enabled.
384 Configures the number of receive descriptors for jumbo frames larger
385 than 1528 bytes. Default is 128 and valid range is from 1 to 255.
386 When jumbo frames larger than 1528 bytes are used, this parameter should
387 not be set lower than 60 on systems with high network traffic. Setting
388 this parameter higher allows the NIC to buffer larger bursts of jumbo
389 traffic without dropping frames, especially on slower systems. Depending
390 on kernel and CPU architecture, each descriptor may require up to about
391 268 bytes. Each descriptor also requires a socket buffer the size of a
392 maximum jumbo frame. On systems with insufficient memory, it may be
393 necessary to reduce this parameter. This parameter should not be set less
394 than 50 if adaptive_coalesce (see below) is enabled. When the maximum
395 frame size is 1528 or smaller (MTU size 1514 or smaller), this parameter
396 is not used and is always 0.
400 Enables or disables adaptive adjustments to the various interrupt
401 coalescing parameters. Enabling it allows the driver to dynamically
402 adjust the interrupt coalescing parameters to achieve high throughput
403 during heavy traffic and low latency during light traffic.
404 rx_std_desc_cnt, (and rx_jumbo_desc_cnt if using jumbo frames) should not
405 be set less than 50, and tx_pkt_desc_cnt should not be set less than 80
406 when this parameter is enabled. Note that if the kernel supports the NAPI
407 receive polling mode, interrupt coalescing will be handled in a different
408 way and this parameter will not be used. The valid values are:
410 0 disabled (always disabled in NAPI mode)
415 Configures the number of 1 usec ticks before the NIC generates receive
416 interrupt after receiving a frame. This parameter works in conjunction
417 with the rx_max_coalesce_frames parameter. Interrupt will be generated
418 when either of these thresholds is exceeded. 0 means this parameter is
419 ignored and interrupt will be generated when the rx_max_coalesce_frames
420 threshold is reached. The valid range is from 0 to 500, and default is
421 60 (18 if using NAPI mode). This parameter is not used and will be
422 adjusted automatically if adaptive_coalesce is set to 1.
424 rx_max_coalesce_frames
426 Configures the number of received frames before the NIC generates receive
427 interrupt. The valid range is from 0 to 100, and default is 15 (6 if using
428 NAPI mode). This parameter and rx_coalesce_ticks cannot be both 0,
429 otherwise no receive interrupts will be generated. It should also be set
430 lower than rx_std_desc_cnt (and rx_jumbo_desc_cnt if using jumbo frames).
431 This parameter is not used and will be adjusted automatically if
432 adaptive_coalesce is set to 1.
436 Configures the number of 1 usec ticks before the NIC generates transmit
437 interrupt after transmitting a frame. This parameter works in conjunction
438 with the tx_max_coalesce_frames parameter. Interrupt will be generated
439 when either of these thresholds is exceeded. 0 means this
440 parameter is ignored and interrupt will be generated when the
441 tx_max_coalesce_frames threshold is reached. The valid range is from 0 to
442 500, and default is 200. This parameter is not used and will be adjusted
443 automatically if adaptive_coalesce is set to 1.
445 tx_max_coalesce_frames
447 Configures the number of transmitted frames before the NIC generates
448 transmit interrupt. The valid range is from 0 to 100, and default is 35.
449 This parameter and tx_coalesce_ticks cannot be both 0, otherwise no
450 transmit completion interrupt will be generated. This parameter should
451 always be set lower than tx_pkt_desc_cnt. This parameter is not used and
452 will be adjusted automatically if adaptive_coalesce is set to 1.
456 Configures the number of 1 usec ticks between periodic statistics updates
457 (DMAs). The valid range is from 100 to 3600000000, and default is 1000000
458 (1 sec.). 0 is also valid and is used to disable statistics updates.
459 This parameter is not used and will be set to default if adaptive_coalesce
460 is set to 1. Please note that the valid range may vary by kernel version.
464 Enables or disables magic packet Wake-On-LAN when the system is shutdown.
465 Note that not all systems support Wake-On-LAN. The valid values are:
467 0 magic packet Wake-On-LAN disabled (default)
468 1 magic packet Wake-On-LAN enabled
472 Enables or disables TCP Segmentation Option (TSO) when using kernels that
473 support it. This parameter is only defined on newer kernels that support
474 TSO. The valid values are:
477 1 TSO enabled (default)
481 Enables or disables message signal interrupts (MSI) when using kernels that
482 support it. This parameter is only defined on newer kernels that support
483 MSI. The valid values are:
486 0 MSI enabled (default)
487 Note: Only applies to the latest devices.
491 This parameter controls the stripping of VLAN tags on incoming packets,
492 and is used to allow VLAN tagged ASF or IPMI packets to be received
493 properly. The valid values are:
495 0 Auto mode (default)
499 In normal mode, VLAN tags are only stripped if VLANs are registered
500 by the 802.1q VLAN module or BASP. In forced strip mode, VLAN tags
501 are always stripped. Auto mode will select normal strip mode if ASF/IPMI
502 is disabled, or forced strip mode if ASF/IPMI is enabled.
506 If set to 1, this parameter will cause the driver to return
507 -EOPNOTSUPP when the SIOCGMIIREG or ETHTOOL_GLINK ioctls are called
508 during the first 6 seconds after driver reset. When the driver resets
509 the NIC during ifconfig, the link will drop and it may take several
510 seconds for the link to come up after autonegotiation completes. Some
511 applications, such as ifup, may not wait long enough for the link
512 before giving up. Setting this parameter to 1 may get around such
513 problems. The default value is 0, which means that the driver will
514 always return true link states to all ioctl calls, when applicable.
518 If set to 1, this parameter will cause the driver to never
519 put the device in D3Hot power state when the NIC is shutdown or
520 suspended. If set, this parameter will also disable the Wake-On-Lan
521 setting. A rare D3Hot related problem was seen during repeated
522 shutdown of PCI Express devices on systems running 2.6 kernels.
527 The following are the most common sample messages that may be logged in the file
528 /var/log/messages. Use dmesg -n <level> to control the level at which messages
529 will appear on the console. Most systems are set to level 6 by default.
531 Broadcom Gigabit Ethernet Driver bcm5700 with Broadcom NIC Extension (NICE)
532 ver. 8.3.17b (02/21/06)
537 eth#: Broadcom BCM5704 1000Base-T found at mem faff0000, IRQ 16, node addr 0010180402d8
538 eth#: Broadcom BCM5704 Integrated Copper transceiver found
539 eth#: Scatter-gather ON, 64-bit DMA ON, Tx Checksum ON, Rx Checksum ON, 802.1Q VLAN ON, TSO ON, NAPI ON
544 bcm5700: eth# NIC Link is Up, 1000 Mbps full duplex, receive and transmit flow control ON
546 Link up and speed indication
549 bcm5700: eth# NIC Link is Down
553 bcm5700: eth# Using Message Signaled Interrupt (MSI)
555 The driver is in MSI mode
557 bcm5700: eth# Using PCI INTX interrupt
559 The driver is in IRQ mode
564 Detailed statistics and configuration information can be viewed in the file
565 /proc/net/nicinfo/eth#.info.