Attachment 603552 Details for Bug 836803 – RHEL PATCH 4/7

[patch] RHEL PATCH 4/7

0004-hrtimers-Move-lock-held-region-in-hrtimer_interrupt.patch (text/plain), 2.99 KB, created by Prarit Bhargava on 2012-08-10 13:57:01 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Prarit Bhargava

Created: 2012-08-10 13:57:01 UTC

Size: 2.99 KB

patch

obsolete

>From 71f391ac3917ccb858d009900125be9fabb4ee08 Mon Sep 17 00:00:00 2001
>From: Prarit Bhargava <prarit@redhat.com>
>Date: Tue, 24 Jul 2012 10:25:01 -0400
>Subject: hrtimers: Move lock held region in hrtimer_interrupt()
>
>commit 196951e91262fccda81147d2bcf7fdab08668b40
>Author: Thomas Gleixner <tglx@linutronix.de>
>Date:   Tue Jul 10 18:43:23 2012 -0400
>
>    hrtimers: Move lock held region in hrtimer_interrupt()
>
>    We need to update the base offsets from this code and we need to do
>    that under base->lock. Move the lock held region around the
>    ktime_get() calls. The ktime_get() calls are going to be replaced with
>    a function which gets the time and the offsets atomically.
>
>    Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
>    Reviewed-by: Ingo Molnar <mingo@kernel.org>
>    Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
>    Acked-by: Prarit Bhargava <prarit@redhat.com>
>    Cc: stable@vger.kernel.org
>    Signed-off-by: John Stultz <johnstul@us.ibm.com>
>    Link: http://lkml.kernel.org/r/1341960205-56738-6-git-send-email-johnstul@us.ibm.com
>    Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
>
>Note, because retrigger_next_event() is RHEL6 specific at this point, I
>did not modify the declaration
>
>	struct hrtimer_cpu_base *base;
>
>in the RHEL6 backport of linux-2.6 commit
>b12a03ce4880bd13786a98db6de494a3e0123129.  It has now been modified to
>
>	struct hrtimer_cpu_base *base = &__get_cpu_var(hrtimer_bases);
>
>so this code matches upstream and compiles.
>
>Additionally, note again, that RHEL6 still uses CLOCK_REALTIME, where the
>upstream kernel uses HRTIMER_BASE_REALTIME and HRTIMER_BASE_BOOTTIME.
>---
> kernel/hrtimer.c |    7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
>diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
>index 6703d49..87bb8a5 100644
>--- a/kernel/hrtimer.c
>+++ b/kernel/hrtimer.c
>@@ -707,7 +707,7 @@ static inline void hrtimer_init_timer_hres(struct hrtimer *timer) { }
>  */
> static void retrigger_next_event(void *arg)
> {
>-	struct hrtimer_cpu_base *base;
>+	struct hrtimer_cpu_base *base = &__get_cpu_var(hrtimer_bases);
> 	struct timespec realtime_offset, wtm, sleep;
> 
> 	if (!hrtimer_hres_active())
>@@ -1244,11 +1244,10 @@ void hrtimer_interrupt(struct clock_event_device *dev)
> 	cpu_base->nr_events++;
> 	dev->next_event.tv64 = KTIME_MAX;
> 
>+	spin_lock(&cpu_base->lock);
> 	entry_time = now = ktime_get();
> retry:
> 	expires_next.tv64 = KTIME_MAX;
>-
>-	spin_lock(&cpu_base->lock);
> 	/*
> 	 * We set expires_next to KTIME_MAX here with cpu_base->lock
> 	 * held to prevent that a timer is enqueued in our queue via
>@@ -1323,6 +1322,7 @@ retry:
> 	 * interrupt routine. We give it 3 attempts to avoid
> 	 * overreacting on some spurious event.
> 	 */
>+	spin_lock(&cpu_base->lock);
> 	now = ktime_get();
> 	cpu_base->nr_retries++;
> 	if (++retries < 3)
>@@ -1335,6 +1335,7 @@ retry:
> 	 */
> 	cpu_base->nr_hangs++;
> 	cpu_base->hang_detected = 1;
>+	spin_unlock(&cpu_base->lock);
> 	delta = ktime_sub(now, entry_time);
> 	if (delta.tv64 > cpu_base->max_hang_time.tv64)
> 		cpu_base->max_hang_time = delta;
>-- 
>1.7.9.3
>

Actions: View | Diff

Attachments on bug 836803: 597663 | 603549 | 603550 | 603551 | 603552 | 603553 | 603554 | 603555