On Sun, 18 Jul 2004, Willem Jan Withagen wrote: > > > Even more reason to do some (ab)using test with the em0 card??? > > > > Yeah, that would be very helpful if you could. > > Well uptill now it was less than 30 minutes to get a crash.... No > counting the ones where it directly crashed whilest booting. > > So we could get to > 20 crashes/sunday :) Now there's a productivity boost :-). > I now got a kernel with no IPv6. First abuse that with bge0. But if it > is what you think it is, it'll make no difference. Yeah, we'll see -- do you have much IPv6 activity on your network? > Note that I can get annottated traces with param-values and line numbers > if I coredump the panics. But dumping core and saving it, takes almost > just as long as causing one. I should probably go read the if_bge code in some more detail, and the TCP code. With IPv6 disabled, we should be able to use stronger inpcb assertions in TCP, which may help diagnose locking problems. Robert N M Watson FreeBSD Core Team, TrustedBSD Projects robert_at_fledge.watson.org Principal Research Scientist, McAfee ResearchReceived on Sun Jul 18 2004 - 17:52:56 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:02 UTC