freebsd stable - Apr 2015 - 10.1-STABLE bce: Watchdog timeout occurred

This may be caused by DMA alignment problems.
See 
https://docs.freebsd.org/cgi/getmsg.cgi?fetch=145859+0+archive/2015/freebsd-stable/20150419.freebsd-stable
for a recent thread about the msk driver.  The msk maintainer Yonghyeon 
Pyun has opted for super safe options of 32K alignment!

It's a long shot, but you could try increasing BCE_DMA_ALIGN and/or 
BCE_RX_BUF_ALIGN in the include file if_bcereg.h, say up to 4096, to see 
whether it makes any difference.

- Gareth.

On 21/04/2015 10:52, Alnis Morics wrote:
> On 04/21/2015 06:17 AM, Chris Ross wrote:
>>    I got a new [to me] system recently, a Dell PE 1950.  It has two 
>> bce parts on the motherboard that identify as:
>>
>> bce#: <QLogic NetXtreme II BCM5708 1000Base-T (B2)>
>>
>>    The OS I installed and kernel I'm running are from a download of
a
>> 10.1 STABLE ISO, r281235, April 7, 2015.
>>
>>    I had gone on to check out a newer stable from subversion, and 
>> build a custom kernel, but when I booted that one I got a bce0 that 
>> didn't seem to work, and kept emitting:
>>
>> bce0: /usr/src/sys/dev/bce/if_bce.c(7869): Watchdog timeout occurred, 
>> resetting!
>> bce0: link state changed to DOWN
>> bce0: link state changed to UP
>>
>>    So, I fell back.  But I've since noticed that even the original 
>> kernel seems to do this after booting.  I'm not yet running any 
>> notable amount of traffic through the system, but intend to make it 
>> an edge router, so certainly will be.
>>
>>    Is there any sort of issue noted in the bce driver in recent 
>> days/weeks/months?  Are other folks seeing this diagnostic/error?
>>
>>    I'll do a little more testing and see if I'm seeing it more
or
>> less often, but I know that in at least some cases the interface has 
>> flapped like this after boot for long enough that I was unable to get 
>> connected remotely, and resorted to a console login to reboot.
>>
>>                                                  - Chris
>>
> There are "Watchdog timeout" errors with some msk NICs. Both msk
and
> bce are dependent on MII bus code (see /usr/src/sys/amd64/conf/GENERIC)
>
> -Alnis
> _______________________________________________
> freebsd-stable at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-stable
> To unsubscribe, send any mail to "freebsd-stable-unsubscribe at
freebsd.org"

On Apr 21, 2015, at 10:10 , Gareth Wyn Roberts <g.w.roberts at
glyndwr.ac.uk> wrote:
> This may be caused by DMA alignment problems.
> See
https://docs.freebsd.org/cgi/getmsg.cgi?fetch=145859+0+archive/2015/freebsd-stable/20150419.freebsd-stable
for a recent thread about the msk driver.  The msk maintainer Yonghyeon Pyun has
opted for super safe options of 32K alignment!
> 
> It's a long shot, but you could try increasing BCE_DMA_ALIGN and/or
BCE_RX_BUF_ALIGN in the include file if_bcereg.h, say up to 4096, to see whether
it makes any difference.
  Well, after making that change, I was able to confirm that the problem
doesn't seem to occur.  However, in trying to verify the problem on an
unmodified kernel, I've rebooted a GENERIC from r281672 without that change,
and am also not seeing the problem.  :-/  I'm not sure whether the gremlins
have "fixed" something, or if I was just too critical in my initial
analysis.

  For now I'll take that change out of my tree and run without it.  If I see
the flapping again, I'll confirm that it's repeatable, then change the
alignments as suggested and see if I see a change.

  Thanks all...

                                         - Chris

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 842 bytes
Desc: Message signed with OpenPGP using GPGMail
URL:
<http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20150422/ec0739ac/attachment.sig>