Date: Thu, 1 Oct 2015 21:54:45 +0000 (UTC) From: John Baldwin <jhb@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-9@freebsd.org Subject: svn commit: r288463 - in stable: 10/sys/kern 9/sys/kern Message-ID: <201510012154.t91LsjM3018844@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: jhb Date: Thu Oct 1 21:54:43 2015 New Revision: 288463 URL: https://svnweb.freebsd.org/changeset/base/288463 Log: MFC 286256: kgdb uses td_oncpu to determine if a thread is running and should use a pcb from stoppcbs[] rather than the thread's PCB. However, exited threads retained td_oncpu from the last time they ran, and newborn threads had their CPU fields cleared to zero during fork and thread creation since they are in the set of fields zeroed when threads are setup. To fix, explicitly update the CPU fields for exiting threads in sched_throw() to reflect the switch out and reset the CPU fields for new threads in sched_fork_thread() to NOCPU. Modified: stable/9/sys/kern/sched_4bsd.c stable/9/sys/kern/sched_ule.c Directory Properties: stable/9/sys/ (props changed) Changes in other areas also in this revision: Modified: stable/10/sys/kern/sched_4bsd.c stable/10/sys/kern/sched_ule.c Directory Properties: stable/10/ (props changed) Modified: stable/9/sys/kern/sched_4bsd.c ============================================================================== --- stable/9/sys/kern/sched_4bsd.c Thu Oct 1 21:52:25 2015 (r288462) +++ stable/9/sys/kern/sched_4bsd.c Thu Oct 1 21:54:43 2015 (r288463) @@ -793,6 +793,8 @@ sched_fork_thread(struct thread *td, str { struct td_sched *ts; + childtd->td_oncpu = NOCPU; + childtd->td_lastcpu = NOCPU; childtd->td_estcpu = td->td_estcpu; childtd->td_lock = &sched_lock; childtd->td_cpuset = cpuset_ref(td->td_cpuset); @@ -1671,6 +1673,8 @@ sched_throw(struct thread *td) } else { lock_profile_release_lock(&sched_lock.lock_object); MPASS(td->td_lock == &sched_lock); + td->td_lastcpu = td->td_oncpu; + td->td_oncpu = NOCPU; } mtx_assert(&sched_lock, MA_OWNED); KASSERT(curthread->td_md.md_spinlock_count == 1, ("invalid count")); Modified: stable/9/sys/kern/sched_ule.c ============================================================================== --- stable/9/sys/kern/sched_ule.c Thu Oct 1 21:52:25 2015 (r288462) +++ stable/9/sys/kern/sched_ule.c Thu Oct 1 21:54:43 2015 (r288463) @@ -2035,6 +2035,8 @@ sched_fork_thread(struct thread *td, str */ ts = td->td_sched; ts2 = child->td_sched; + child->td_oncpu = NOCPU; + child->td_lastcpu = NOCPU; child->td_lock = TDQ_LOCKPTR(TDQ_SELF()); child->td_cpuset = cpuset_ref(td->td_cpuset); ts2->ts_cpu = ts->ts_cpu; @@ -2656,6 +2658,8 @@ sched_throw(struct thread *td) MPASS(td->td_lock == TDQ_LOCKPTR(tdq)); tdq_load_rem(tdq, td); lock_profile_release_lock(&TDQ_LOCKPTR(tdq)->lock_object); + td->td_lastcpu = td->td_oncpu; + td->td_oncpu = NOCPU; } KASSERT(curthread->td_md.md_spinlock_count == 1, ("invalid count")); newtd = choosethread();
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201510012154.t91LsjM3018844>