Hi Dave, Stephane, Knut, Peter or Will, 14c63f17b1fde5a575a28e96547a22b451c71fb5 perf: Drop sample rate when sampling is too slow e5302920da9ef23f9d19d4e9ac85704cc25bee7a perf: Fix interrupt handler timing harness 723478c8a471403c53cf144999701f6e0c4bbd11 perf: Enforce 1 as lower limit for perf_event_max_sample_rate 5f5092e72cc25a6a5785308270e0085b2b2772cc ARM: perf: hook up perf_sample_event_took around pmu irq These patches look applicable to stable-3.10, these patches prevent an interrupt storm when perf sampling is too slow on x86 and arm, an interrupt storm may trigger softlockup. Patch e5302920da9e and 723478c8a471 are bug fix for 14c63f17b1fd. The last one patch can't be applied directly, I've adjusted the context manually. Now they were built successful for me. What do you think? the statistics of these patches： Documentation/sysctl/kernel.txt | 26 +++++++++++ arch/arm/kernel/perf_event.c | 11 ++++- arch/x86/kernel/cpu/perf_event.c | 12 ++++- include/linux/perf_event.h | 7 +++ kernel/events/core.c | 94 +++++++++++++++++++++++++++++++++++++--- kernel/sysctl.c | 10 +++++ 6 files changed, 152 insertions(+), 8 deletions(-) A detailed discussion can see the following link: http://www.gossamer-threads.com/lists/linux/kernel/1839307?do=post_view_threaded#1839307 -------------------------------------------------------- From: Will Deacon <will.deacon@xxxxxxx> Subject: [PATCH] ARM: perf: hook up perf_sample_event_took around pmu irq handling commit 5f5092e72cc25a6a5785308270e0085b2b2772cc upstream. Since we indirect all of our PMU IRQ handling through a dispatcher, it's trivial to hook up perf_sample_event_took to prevent applications such as oprofile from generating interrupt storms due to an unrealisticly low sample period. Reported-by: Robert Richter <rric@xxxxxxxxxx> Signed-off-by: Will Deacon <will.deacon@xxxxxxx> [wml: Backported to 3.10: context adjust ] Signed-off-by: Weng Meiling <wengmeiling.weng@xxxxxxxxxx> --- arch/arm/kernel/perf_event.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index e19edc6..ace0ce8 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -303,11 +303,18 @@ static irqreturn_t armpmu_dispatch_irq(int irq, void *dev) struct arm_pmu *armpmu = (struct arm_pmu *) dev; struct platform_device *plat_device = armpmu->plat_device; struct arm_pmu_platdata *plat = dev_get_platdata(&plat_device->dev); + int ret; + u64 start_clock, finish_clock; + start_clock = sched_clock(); if (plat && plat->handle_irq) - return plat->handle_irq(irq, dev, armpmu->handle_irq); + ret = plat->handle_irq(irq, dev, armpmu->handle_irq); else - return armpmu->handle_irq(irq, dev); + ret = armpmu->handle_irq(irq, dev); + finish_clock = sched_clock(); + + perf_sample_event_took(finish_clock - start_clock); + return ret; } static void -- 1.8.2.2 -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html