Le Friday 24 June 2005 17:03, Kris Kennaway a écrit : > On Fri, Jun 24, 2005 at 04:26:55PM +0200, Thierry Herbelot wrote: > > This is with an SMP machine (oldish BP6) > > Looks like you forgot to include the full panic...wasn't there a "last > used by FOO" also displayed? If so, you should configure > DEBUG_MEMGUARD to watch this malloc type (some assembly required, but > it's pretty straightforward -- see Bosko Milekic's post to this > mailing list when he committed support describing how to do it) and > try to replicate the panic. sorry if I forgot anything : this is the only message I had on the serial console : [traditional dmesg startup messages] Local package initialization:. Additional TCP options:. Starting background file system checks in 60 seconds. Fri Jun 24 09:10:10 CEST 2005 FreeBSD/i386 (multi-cur.herbelot.nom) (ttyd0) login: Jun 24 09:11:02 multi-cur su: thierry.herbelot to root on /dev/ttyp0 panic: Memory modified after free 0xc15c1400(256) val=0 _at_ 0xc15c1400 cpuid = 0 KDB: enter: panic [thread pid 520 tid 100070 ] Stopped at kdb_enter+0x2b: nop db> where Tracing pid 520 tid 100070 td 0xc1508780 kdb_enter(c084ecf5) at kdb_enter+0x2b panic(c08695fc,c15c1400,100,0,c15c1400) at panic+0x127 trash_ctor(c15c1400,100,cc89db40,1,c104a9d8) at trash_ctor+0x2f mb_ctor_mbuf(c15c1400,100,cc89db40,1) at mb_ctor_mbuf+0x18 uma_zalloc_arg(c104a9a0,cc89db40,1) at uma_zalloc_arg+0x10f [rest of the ddb session] I forgot to add : this is with a straight GENERIC kernel, built from sources cvsupped around yesterday Jun 23 20:00:00 CEST 2005 the full dmesg is accessible at : http://herbelot.tfh.free.fr/050624.trace Cheers TfH > > KrisReceived on Fri Jun 24 2005 - 13:38:05 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:37 UTC