aboutsummaryrefslogtreecommitdiffstats
path: root/kernel/signal.c
diff options
context:
space:
mode:
authorRoland McGrath <roland@redhat.com>2005-10-21 18:03:29 -0400
committerLinus Torvalds <torvalds@g5.osdl.org>2005-10-21 18:38:08 -0400
commit25f407f0b668f5e4ebd5d13e1fb4306ba6427ead (patch)
tree10d8661419da9e0d6be6d22ef319582d052c4a26 /kernel/signal.c
parent9465bee863bc4c6cf1566c12d6f92a8133e3da5c (diff)
[PATCH] Call exit_itimers from do_exit, not __exit_signal
When I originally moved exit_itimers into __exit_signal, that was the only place where we could reliably know it was the last thread in the group dying, without races. Since then we've gotten the signal_struct.live counter, and do_exit can reliably do group-wide cleanup work. This patch moves the call to do_exit, where it's made without locks. This avoids the deadlock issues that the old __exit_signal code's comment talks about, and the one that Oleg found recently with process CPU timers. [ This replaces e03d13e985d48ac4885382c9e3b1510c78bd047f, which is why it was just reverted. ] Signed-off-by: Roland McGrath <roland@redhat.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'kernel/signal.c')
-rw-r--r--kernel/signal.c14
1 files changed, 1 insertions, 13 deletions
diff --git a/kernel/signal.c b/kernel/signal.c
index 50c992643771..f2b96b08fb44 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -397,20 +397,8 @@ void __exit_signal(struct task_struct *tsk)
397 flush_sigqueue(&tsk->pending); 397 flush_sigqueue(&tsk->pending);
398 if (sig) { 398 if (sig) {
399 /* 399 /*
400 * We are cleaning up the signal_struct here. We delayed 400 * We are cleaning up the signal_struct here.
401 * calling exit_itimers until after flush_sigqueue, just in
402 * case our thread-local pending queue contained a queued
403 * timer signal that would have been cleared in
404 * exit_itimers. When that called sigqueue_free, it would
405 * attempt to re-take the tasklist_lock and deadlock. This
406 * can never happen if we ensure that all queues the
407 * timer's signal might be queued on have been flushed
408 * first. The shared_pending queue, and our own pending
409 * queue are the only queues the timer could be on, since
410 * there are no other threads left in the group and timer
411 * signals are constrained to threads inside the group.
412 */ 401 */
413 exit_itimers(sig);
414 exit_thread_group_keys(sig); 402 exit_thread_group_keys(sig);
415 kmem_cache_free(signal_cachep, sig); 403 kmem_cache_free(signal_cachep, sig);
416 } 404 }