]> git.openfabrics.org - ~shefty/rdma-dev.git/commitdiff
[PATCH] fix vsyscall settimeofday
authorDaniel Walker <dwalker@mvista.com>
Tue, 6 Mar 2007 09:42:11 +0000 (01:42 -0800)
committerLinus Torvalds <torvalds@woody.linux-foundation.org>
Tue, 6 Mar 2007 17:30:25 +0000 (09:30 -0800)
I've only seen this on x86_64.

The vsyscall state only gets updated when a timer interrupts comes in.  So
if the time is set long before the next timer, there will be a period when
a gettimeofday() won't reflect the correct time.

I added an explicit update_vsyscall() during the settimeofday(), that way
the vsyscall state doesn't get stale.

Signed-off-by: Daniel Walker <dwalker@mvista.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: John Stultz <johnstul@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kernel/timer.c

index ee0a2da4aab33e86df8d625e0cb9dcdd82a009e7..797cccb86431c1a423647141b62270f124c341dc 100644 (file)
@@ -862,6 +862,8 @@ int do_settimeofday(struct timespec *tv)
        clock->error = 0;
        ntp_clear();
 
+       update_vsyscall(&xtime, clock);
+
        write_sequnlock_irqrestore(&xtime_lock, flags);
 
        /* signal hrtimers about time change */