Skip to contentSkip to conversation

sched_clock.c change to -stable?

DISCUSSION David Miller 2 participants 3 messages Jan 16, 2010 – Jan 18, 2010

lore
Share & tools
lore
#0 lore raw

Ingo, I need commit:

commit b9f8fcd55bbdb037e5332dbdb7b494f0b70861ac

Author: David Miller <[email protected]>
Date:   Sun Dec 13 18:25:02 2009 -0800

sched: Fix cpu_clock() in NMIs, on !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK

as a dependency on some sparc perf bug fixes I'd like to submit to -stable.

Therefore, do you mind if I send this commit to -stable?

Thanks.

Acked-by Peter Zijlstra
#1 lore raw
Quoted 7 lines · David Miller

Ingo, I need commit:

commit b9f8fcd55bbdb037e5332dbdb7b494f0b70861ac

Author: David Miller <[email protected]>
Date:   Sun Dec 13 18:25:02 2009 -0800
Quoted 8 lines

sched: Fix cpu_clock() in NMIs, on !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK

as a dependency on some sparc perf bug fixes I'd like to submit to -stable.

Therefore, do you mind if I send this commit to -stable?

Thanks.

Sure!

I've attached the commit for the -stable team's convenience.

Ingo

------------------>

Quoted 1 line

From b9f8fcd55bbdb037e5332dbdb7b494f0b70861ac Mon Sep 17 00:00:00 2001

From: David Miller <[email protected]>
Date: Sun, 13 Dec 2009 18:25:02 -0800
Subject: [PATCH] sched: Fix cpu_clock() in NMIs, on !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK

Relax stable-sched-clock architectures to not save/disable/restore hardirqs in cpu_clock().

The background is that I was trying to resolve a sparc64 perf issue when I discovered this problem.

On sparc64 I implement pseudo NMIs by simply running the kernel at IRQ level 14 when local_irq_disable() is called, this allows performance counter events to still come in at IRQ level 15.

This doesn't work if any code in an NMI handler does local_irq_save() or local_irq_disable() since the "disable" will kick us back to cpu IRQ level 14 thus letting NMIs back in and we recurse.

The only path which that does that in the perf event IRQ handling path is the code supporting frequency based events. It uses cpu_clock().

cpu_clock() simply invokes sched_clock() with IRQs disabled.

And that's a fundamental bug all on it's own, particularly for the HAVE_UNSTABLE_SCHED_CLOCK case. NMIs can thus get into the sched_clock() code interrupting the local IRQ disable code sections of it.

Furthermore, for the not-HAVE_UNSTABLE_SCHED_CLOCK case, the IRQ disabling done by cpu_clock() is just pure overhead and completely unnecessary.

So the core problem is that sched_clock() is not NMI safe, but we are invoking it from NMI contexts in the perf events code (via cpu_clock()).

A less important issue is the overhead of IRQ disabling when it isn't necessary in cpu_clock().

CONFIG_HAVE_UNSTABLE_SCHED_CLOCK architectures are not affected by this patch.

LKML-Reference: <[email protected]>

---
 kernel/sched_clock.c |   23 +++++++++++++++--------
 1 files changed, 15 insertions(+), 8 deletions(-)
kernel/sched_clock.c +15 -8
index 479ce56..5b49613 100644
@@ -236,6 +236,18 @@ void sched_clock_idle_wakeup_event(u64 delta_ns)
 }
 EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event);
 
+unsigned long long cpu_clock(int cpu)
+{
+	unsigned long long clock;
+	unsigned long flags;
+
+	local_irq_save(flags);
+	clock = sched_clock_cpu(cpu);
+	local_irq_restore(flags);
+
+	return clock;
+}
+
 #else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
 
 void sched_clock_init(void)
@@ -251,17 +263,12 @@ u64 sched_clock_cpu(int cpu)
 	return sched_clock();
 }
 
-#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
 
 unsigned long long cpu_clock(int cpu)
 {
-	unsigned long long clock;
-	unsigned long flags;
+	return sched_clock_cpu(cpu);
+}
 
-	local_irq_save(flags);
-	clock = sched_clock_cpu(cpu);
-	local_irq_restore(flags);
+#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */
 
-	return clock;
-}
 EXPORT_SYMBOL_GPL(cpu_clock);
#2 lore raw

From: Ingo Molnar <[email protected]>
Date: Sat, 16 Jan 2010 12:36:47 +0100

Quoted 7 lines
Quoted 3 lines

Therefore, do you mind if I send this commit to -stable?

Thanks.

Sure!

I've attached the commit for the -stable team's convenience.

Great, I'll submit it also as part of my sparc -stable submission to make sure they really get it :-)

Thanks!