On Wednesday 12 April 2006 23:03, John Baldwin wrote: > On Wednesday 12 April 2006 09:52, Csaba Henk wrote: > > Hi! > > > > With my CURRENT installation from Apr 1, I get: > > > > panic: Assertion !(curthread->td_flags & TDF_SINTR) failed at > > /usr/src/sys/kern/subr_sleepqueue.c:529 cpuid = 0 > > KDB: enter: panic > > The sleepq_remove() in msleep() to handle this nested sleep case isn't > clearing the TDF_SINTR from the previous tsleep. I suspect this changed > in the recent fixes as I think they changed the sleepqueue code to clear > TDF_SINTR when the thread resumed, rather than clearing it when the thread > was removed from the queue IIRC. > Can you test following patch ? I think the TDF_SINTR was never cleared in sleepq_remove or sleepq_resume_thread, the functions were not touched in my previous fixes.:-) David Xu Index: subr_sleepqueue.c =================================================================== RCS file: /home/ncvs/src/sys/kern/subr_sleepqueue.c,v retrieving revision 1.26 diff -u -r1.26 subr_sleepqueue.c --- subr_sleepqueue.c 23 Feb 2006 03:42:17 -0000 1.26 +++ subr_sleepqueue.c 13 Apr 2006 00:59:01 -0000 _at__at_ -633,6 +633,7 _at__at_ td->td_wmesg = NULL; td->td_wchan = NULL; + td->td_flags &= ~TDF_SINTR; /* * Note that thread td might not be sleeping if it is runningReceived on Wed Apr 12 2006 - 23:07:54 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:54 UTC