lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [day] [month] [year] [list]
Date:	Mon, 2 Jul 2007 11:17:37 +0100
From:	"Daniel J Blueman" <daniel.blueman@...il.com>
To:	"Stephen Hemminger" <shemminger@...ux-foundation.org>
Cc:	netdev@...r.kernel.org
Subject: Re: sky2 hangs without any messages

Hi Stephen,

Tino Keitel <tino.keitel@...ei.de> wrote:
>> in the last 2 days, I had 2 outages of the NIC of my Mac mini Core Duo.
>> I checked the kernel log but I found nothing sky2 related. After
>> reloading the module, the interface worked again.

> There is a reproducible hang (due to missed IRQ) in sky2.  It is more reproducible
> with slower CPU's like Mac Mini.

This feels like the same problem I hit once every ~2 days on my server
platform or laptop [1]: both Yukon-ECs, rev 1 and 2, running NFSv4
over TCP over a gig-e switch, ia32, PCI-E.

Observations:

1. problem happens with both rev-1 and 2 Yukon-ECs, so less likely a silicon bug
2. reproducible with 1500 MTU on both ends
3. reproducible with and without receive offloading disabled (ie
'ethtool -K gig0 rx off') [check with tx offload too?]
4. reproducible with two different gig-e switches and different cables
5. reproducible with MSI interrupts [check with legacy interrupts?]

Is there anything that I can do to help debug this missing IRQ hang?
I'm thinking of moving to the vendor driver [2] to see if the issue is
entirely avoidable in software.

Thanks,
  Daniel

--- [1]

sky2 0000:02:00.0: v1.14 addr 0xd6000000 irq 16 Yukon-EC (0xb6) rev 2
sky2 gig0: addr 00:16:b6:70:d8:3d
sky2 gig0: enabling interface
sky2 gig0: ram buffer 48K
sky2 gig0: Link is up at 1000 Mbps, full duplex, flow control both
<NFS hang observed>
NETDEV WATCHDOG: gig0: transmit timed out
sky2 gig0: tx timeout
sky2 gig0: transmit ring 424 .. 401 report=427 done=427
sky2 gig0: disabling interface
sky2 gig0: enabling interface
sky2 gig0: ram buffer 48K
sky2 gig0: Link is up at 1000 Mbps, full duplex, flow control both
<NFS resumes after a few minutes>

$ grep gig0 /proc/interrupts
218:      27297          0   PCI-MSI-edge      gig0

--- [2]

http://www.syskonnect.de/e_en/products/adapters/pcie_server/sk-9exx/software/linux/driver/install-8_31.tar.bz2
-- 
Daniel J Blueman
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists