cvs commit: src/sys/dev/bge if_bge.c

[Jung-uk Kim]

On Wednesday 13 December 2006 07:23 pm, Scott Long wrote:
> Jung-uk Kim wrote:
> > On Wednesday 13 December 2006 03:51 pm, Scott Long wrote:
> >> scottl      2006-12-13 20:51:51 UTC
> >>
> >>   FreeBSD src repository
> >>
> >>   Modified files:
> >>     sys/dev/bge          if_bge.c
> >>   Log:
> >>   Remove a redundant write of the firmware reset magic number. 
> >> It looks to have been added erroneously, and it causes problems
> >> on some chips.  A larger change is needed to do this write at a
> >> more appropriate place, but that change requires reworking the
> >> ASF logic.  That will be worked on in the future.
> >>
> >>   Submitted by: Bruce Evans
> >
> > I am still getting firmware handshake timeouts and/or watchdog
> > timeouts.  Most importantly it panics or get witness warnings
> > (lots of 'memory modified after free').  Panic goes like this
> > (while kldunload if_bge with dhclient enabled):
> >
> > brgphy0: detached
> > miibus0: detached
> > bge0: firmware handshake timed out, found 0x4b657654
> > bge0: firmware handshake timed out, found 0x4b657654
> > bge0: detached
> > bge0: <Broadcom BCM5705 A3, ASIC rev. 0x3003> mem ...
> > bge0: firmware handshake timed out, found 0x4b657654
> > bge0: firmware handshake timed out, found 0x4b657654
> > miibus0: <MII bus> on bge0
> > brgphy0: <BCM5705 1/100/1000baseTX PHY> on miibus0
> > brgphy0:  10baseT, ...
> > bge0: Ethernet address: ...
> > bge0: firmware handshake timed out, found 0x4b657654
> > bge0: firmware handshake timed out, found 0x4b657654
> > bge0: discard frame w/o leading ethernet header (len 4294967192
> > pkt len 42949672)
> > Kernel page fault with the following non-sleepable locks held:
> > exclusive sleep mutex bge0 (network driver) r = 0
> > (0xffffffff80e81010) locked @
> > /usr/src/sys/modules/bge/../../dev/bge/if_bge.c:2828 ...
> >
> > As you can see the packet length is really bogus.  I am wondering
> > if the chip is actually initialized correctly to begin with.
> >
> > Jung-uk Kim
>
> Does it work correctly before you do the unload?

I believe it does.  However, some times I get watchdog timeout by 
doing ifconfig down/up.  It does not recover from it and it spits out 
lots of 'memory modified after free' warnings.

Jung-uk Kim