Bug#618341: bnx2: MSI-X hang on 5709 NICs - nic dead
Package: linux-2.6
Version: 2.6.32-bpo.5
Severity: critical
Tags: squeeze patch lenny
Justification: breaks the whole system
http://marc.info/?l=linux-netdevm=127240304211909w=2
here is the description from the mail , repeated:
It has been reported that under certain heavy traffic conditions
in MSI-X
mode, the driver can lose an MSI-X vector causing all packets in
the
associated rx/tx ring pair to be dropped. The problem is caused
by
the chip dropping the write to unmask the MSI-X vector by the
kernel
(when migrating the IRQ for example).
This can be prevented by increasing the GRC timeout value for
these
register read and write operations.
Thanks to Dell for helping us debug this problem.
Signed-off-by: Michael Chan mc...@broadcom.com
we are using the 2.6.32 kernel and are bitten by this frequently.
vers/net/bnx2.c |6 +-
1 files changed, 5 insertions(+), 1 deletions(-)
diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c
index a257bab..4c1e51e 100644
--- a/drivers/net/bnx2.c
+++ b/drivers/net/bnx2.c
@@ -4759,8 +4759,12 @@ bnx2_reset_chip(struct bnx2 *bp, u32 reset_code)
rc = bnx2_alloc_bad_rbuf(bp);
}
- if (bp-flags BNX2_FLAG_USING_MSIX)
+ if (bp-flags BNX2_FLAG_USING_MSIX) {
bnx2_setup_msix_tbl(bp);
+ /* Prevent MSIX table reads and write from timing out */
+ REG_WR(bp, BNX2_MISC_ECO_HW_CTL,
+ BNX2_MISC_ECO_HW_CTL_LARGE_GRC_TMOUT_EN);
+ }
return rc;
}
--
1.6.4.GIT
-- System Information:
Debian Release: 6.0
APT prefers proposed-updates
APT policy: (500, 'proposed-updates'), (500, 'stable'), (1, 'experimental')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.37-1-amd64 (SMP w/2 CPU cores)
Locale: LANG=de_DE.UTF-8, LC_CTYPE=de_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/bash
--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive:
http://lists.debian.org/20110314135717.11072.32078.report...@jesaja.schuldei.org
Bug#573406: src:linux-2.6: please use CONFIG_IP_FIB_TRIE
Package: src:linux-2.6 Severity: wishlist please use CONFIG_IP_FIB_TRIE in the future. This improves lookup performance if you have a large number of routes. LC-trie is a longest matching prefix lookup algorithm which performs better than FIB_HASH for large routing tables. But, it consumes more memory and is more complex. LC-trie is described in: IP-address lookup using LC-tries. Stefan Nilsson and Gunnar Karlsson IEEE Journal on Selected Areas in Communications, 17(6):1083-1092, June 1999 An experimental study of compression methods for dynamic tries Stefan Nilsson and Matti Tikkanen. Algorithmica, 33(1):19-33, 2002. http://www.nada.kth.se/~snilsson/public/papers/dyntrie2/ it has been in the kernel since 2.6.13, is actively maintained and mature. while it does not hurt the run-of-the-mill desktop or server setup in any way it will drasticly improve performance for routers. -- System Information: Debian Release: 5.0.4 APT prefers proposed-updates APT policy: (500, 'proposed-updates'), (500, 'stable') Architecture: amd64 (x86_64) Kernel: Linux 2.6.30-1-amd64 (SMP w/2 CPU cores) Locale: LANG=en_US.UTF-8, LC_CTYPE=en_US.UTF-8 (charmap=UTF-8) Shell: /bin/sh linked to /bin/bash -- To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/20100311080727.7136.58188.report...@jesaja.schuldei.org
# CONFIG_SOFTWARE_SUSPEND is not set
hi! mkinitramfs tried to echo the major and minor number of the resume partition into the /sys/power/resume file. however # CONFIG_SOFTWARE_SUSPEND is not set and the file does not exist. that is why i can not suspend to disk with my debian kernel. can this be activated again? signature.asc Description: Digital signature
Re: # CONFIG_SOFTWARE_SUSPEND is not set
Sorry i forgot to reply to the list in my earlier mail. * Maximilian Attems [EMAIL PROTECTED] [2005-11-03 10:54:27]: On Thu, Nov 03, 2005 at 10:37:05AM +0100, Andreas Schuldei wrote: * Maximilian Attems [EMAIL PROTECTED] [2005-11-03 10:11:06]: On Thu, Nov 03, 2005 at 09:31:08AM +0100, Andreas Schuldei wrote: mkinitramfs tried to echo the major and minor number of the resume partition into the /sys/power/resume file. however # CONFIG_SOFTWARE_SUSPEND is not set and the file does not exist. that is why i can not suspend to disk with my debian kernel. can this be activated again? indeed it was meant to be activated 2.6.14, where there problems with 13 and 14? meant = 2.6.14, yes saw lots of fixes between those releases. didn't knew it was enabled in official debian builds for 2.6.13. in 13 it is disabled. and still if your box is sata (scsi subsys) you won't be able to suspend nor if you are using firewire with latest git. not speaking of strange drivers.. those you can unload before suspend. i could suspend my notebook just fine with my own 2.6.8 kernel signature.asc Description: Digital signature
