diff options
author | Oleg Nesterov <oleg@tv-sign.ru> | 2005-10-24 10:29:58 -0400 |
---|---|---|
committer | Linus Torvalds <torvalds@g5.osdl.org> | 2005-10-24 11:13:14 -0400 |
commit | a69ac4a78d8bd9e1ec478bd7297d4f047fcd44a8 (patch) | |
tree | a56edf4f69cc01c6d4b7b3eeee3646c7ea9810df /kernel/posix-cpu-timers.c | |
parent | ca531a0a5e01e5122f67cb6aca8fcbfc70e18e0b (diff) |
[PATCH] posix-timers: fix posix_cpu_timer_set() vs run_posix_cpu_timers() race
This might be harmless, but looks like a race from code inspection (I
was unable to trigger it). I must admit, I don't understand why we
can't return TIMER_RETRY after 'spin_unlock(&p->sighand->siglock)'
without doing bump_cpu_timer(), but this is what original code does.
posix_cpu_timer_set:
read_lock(&tasklist_lock);
spin_lock(&p->sighand->siglock);
list_del_init(&timer->it.cpu.entry);
spin_unlock(&p->sighand->siglock);
We are probaly deleting the timer from run_posix_cpu_timers's 'firing'
local list_head while run_posix_cpu_timers() does list_for_each_safe.
Various bad things can happen, for example we can just delete this timer
so that list_for_each() will not notice it and run_posix_cpu_timers()
will not reset '->firing' flag. In that case,
....
if (timer->it.cpu.firing) {
read_unlock(&tasklist_lock);
timer->it.cpu.firing = -1;
return TIMER_RETRY;
}
sys_timer_settime() goes to 'retry:', calls posix_cpu_timer_set() again,
it returns TIMER_RETRY ...
Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'kernel/posix-cpu-timers.c')
-rw-r--r-- | kernel/posix-cpu-timers.c | 12 |
1 files changed, 8 insertions, 4 deletions
diff --git a/kernel/posix-cpu-timers.c b/kernel/posix-cpu-timers.c index 92a038064628..b15462b17a58 100644 --- a/kernel/posix-cpu-timers.c +++ b/kernel/posix-cpu-timers.c | |||
@@ -730,9 +730,15 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, | |||
730 | * Disarm any old timer after extracting its expiry time. | 730 | * Disarm any old timer after extracting its expiry time. |
731 | */ | 731 | */ |
732 | BUG_ON(!irqs_disabled()); | 732 | BUG_ON(!irqs_disabled()); |
733 | |||
734 | ret = 0; | ||
733 | spin_lock(&p->sighand->siglock); | 735 | spin_lock(&p->sighand->siglock); |
734 | old_expires = timer->it.cpu.expires; | 736 | old_expires = timer->it.cpu.expires; |
735 | list_del_init(&timer->it.cpu.entry); | 737 | if (unlikely(timer->it.cpu.firing)) { |
738 | timer->it.cpu.firing = -1; | ||
739 | ret = TIMER_RETRY; | ||
740 | } else | ||
741 | list_del_init(&timer->it.cpu.entry); | ||
736 | spin_unlock(&p->sighand->siglock); | 742 | spin_unlock(&p->sighand->siglock); |
737 | 743 | ||
738 | /* | 744 | /* |
@@ -780,7 +786,7 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, | |||
780 | } | 786 | } |
781 | } | 787 | } |
782 | 788 | ||
783 | if (unlikely(timer->it.cpu.firing)) { | 789 | if (unlikely(ret)) { |
784 | /* | 790 | /* |
785 | * We are colliding with the timer actually firing. | 791 | * We are colliding with the timer actually firing. |
786 | * Punt after filling in the timer's old value, and | 792 | * Punt after filling in the timer's old value, and |
@@ -788,8 +794,6 @@ int posix_cpu_timer_set(struct k_itimer *timer, int flags, | |||
788 | * it as an overrun (thanks to bump_cpu_timer above). | 794 | * it as an overrun (thanks to bump_cpu_timer above). |
789 | */ | 795 | */ |
790 | read_unlock(&tasklist_lock); | 796 | read_unlock(&tasklist_lock); |
791 | timer->it.cpu.firing = -1; | ||
792 | ret = TIMER_RETRY; | ||
793 | goto out; | 797 | goto out; |
794 | } | 798 | } |
795 | 799 | ||