perf/x86: Warn to early_printk() in case irq_work is too slow

On Mon, Feb 10, 2014 at 08:45:16AM -0800, Dave Hansen wrote:
> The reason I coded this up was that NMIs were firing off so fast that
> nothing else was getting a chance to run.  With this patch, at least the
> printk() would come out and I'd have some idea what was going on.

It will start spewing to early_printk() (which is a lot nicer to use
from NMI context too) when it fails to queue the IRQ-work because its
already enqueued.

It does have the false-positive for when two CPUs trigger the warn
concurrently, but that should be rare and some extra clutter on the
early printk shouldn't be a problem.

Cc: hpa@zytor.com
Cc: tglx@linutronix.de
Cc: dzickus@redhat.com
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: mingo@kernel.org
Fixes: 6a02ad66b2 ("perf/x86: Push the duration-logging printk() to IRQ context")
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20140211150116.GO27965@twins.programming.kicks-ass.net
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
This commit is contained in:
Peter Zijlstra 2014-02-11 16:01:16 +01:00 committed by Thomas Gleixner
parent 90ed5b0fa5
commit cd578abb24
3 changed files with 12 additions and 5 deletions

View file

@ -32,7 +32,7 @@ void init_irq_work(struct irq_work *work, void (*func)(struct irq_work *))
#define DEFINE_IRQ_WORK(name, _f) struct irq_work name = { .func = (_f), } #define DEFINE_IRQ_WORK(name, _f) struct irq_work name = { .func = (_f), }
void irq_work_queue(struct irq_work *work); bool irq_work_queue(struct irq_work *work);
void irq_work_run(void); void irq_work_run(void);
void irq_work_sync(struct irq_work *work); void irq_work_sync(struct irq_work *work);

View file

@ -243,7 +243,7 @@ static void perf_duration_warn(struct irq_work *w)
printk_ratelimited(KERN_WARNING printk_ratelimited(KERN_WARNING
"perf interrupt took too long (%lld > %lld), lowering " "perf interrupt took too long (%lld > %lld), lowering "
"kernel.perf_event_max_sample_rate to %d\n", "kernel.perf_event_max_sample_rate to %d\n",
avg_local_sample_len, allowed_ns, avg_local_sample_len, allowed_ns >> 1,
sysctl_perf_event_sample_rate); sysctl_perf_event_sample_rate);
} }
@ -283,7 +283,12 @@ void perf_sample_event_took(u64 sample_len_ns)
update_perf_cpu_limits(); update_perf_cpu_limits();
irq_work_queue(&perf_duration_work); if (!irq_work_queue(&perf_duration_work)) {
early_printk("perf interrupt took too long (%lld > %lld), lowering "
"kernel.perf_event_max_sample_rate to %d\n",
avg_local_sample_len, allowed_ns >> 1,
sysctl_perf_event_sample_rate);
}
} }
static atomic64_t perf_event_id; static atomic64_t perf_event_id;

View file

@ -61,11 +61,11 @@ void __weak arch_irq_work_raise(void)
* *
* Can be re-enqueued while the callback is still in progress. * Can be re-enqueued while the callback is still in progress.
*/ */
void irq_work_queue(struct irq_work *work) bool irq_work_queue(struct irq_work *work)
{ {
/* Only queue if not already pending */ /* Only queue if not already pending */
if (!irq_work_claim(work)) if (!irq_work_claim(work))
return; return false;
/* Queue the entry and raise the IPI if needed. */ /* Queue the entry and raise the IPI if needed. */
preempt_disable(); preempt_disable();
@ -83,6 +83,8 @@ void irq_work_queue(struct irq_work *work)
} }
preempt_enable(); preempt_enable();
return true;
} }
EXPORT_SYMBOL_GPL(irq_work_queue); EXPORT_SYMBOL_GPL(irq_work_queue);