On Tue, 13 Jan 2004, Doug White wrote: > On Tue, 13 Jan 2004, Igor Sysoev wrote: > > > > > I've cvsuped the sources 5.2-BETA to 5.2-RELEASE. While rebuilding > > > > the system (make -j 3 buildworld) on 5.2-RELEASE kernel I've got the > > > > following panic: > > > > > > > > panic: bad pte <<< this is the problem > > > > cpuid = 0; > > > > boot() called on cpu#0 > > > > > > Hardware description? > > In a nutshell... 2x2.8GHz xeons in a serverworks board of some > construction. I have an IBM behind me here that while I haven't tested a > full buildworld on, I did install a few times and didn't get any panics. > Got a broken keyboard though :) > > I'm not aware of any outstanding pte-based issues. Perhaps the system has > bad memory or processor? I do not think that it's hardware problem. And I think that "panic: sleeping thread owns a non-sleepable lock" is the problem and "panic: bad pte" is a side effect because there is a single panic in back trace and this panic() has the same message: (kgdb) bt #0 doadump () at /usr/src/sys/kern/kern_shutdown.c:240 #1 0xc051067d in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:372 #2 0xc0510a4e in panic () at /usr/src/sys/kern/kern_shutdown.c:550 #3 0xc053595b in propagate_priority (td=0xc8252780) at /usr/src/sys/kern/subr_turnstile.c:176 [ ... ] (kgdb) up 3 0xc053595b in propagate_priority (td=0xc8252780) at /usr/src/sys/kern/subr_turnstile.c:176 176 KASSERT(!TD_IS_SLEEPING(td), (kgdb) l 171 * first thread to grab a slock of a sx lock. In that case 172 * it is possible for us to be at SSLEEP or some other 173 * weird state. We should probably just return if the state 174 * isn't SRUN or SLOCK. 175 */ 176 KASSERT(!TD_IS_SLEEPING(td), 177 ("sleeping thread (pid %d) owns a non-sleepable lock", 178 td->td_proc->p_pid)); 179 180 /* (kgdb) Igor Sysoev http://sysoev.ru/en/Received on Tue Jan 13 2004 - 12:32:52 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:37 UTC