From 889882bce2a5f69242c1f3acd840983f467499b9 Mon Sep 17 00:00:00 2001 From: Lukasz Odzioba Date: Tue, 4 Oct 2016 18:26:26 +0200 Subject: [PATCH 1/4] perf/x86/intel/cstate: Add C-state residency events for Knights Landing Although KNL does support C1,C6,PC2,PC3,PC6 states, the patch only supports C6,PC2,PC3,PC6, because there is no counter for C1. C6 residency counter MSR on KNL has a different address than other platforms which is handled as a new quirk flag. Signed-off-by: Lukasz Odzioba Acked-by: Peter Zijlstra Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Jiri Olsa Cc: Linus Torvalds Cc: Rafael J. Wysocki Cc: Stephane Eranian Cc: Thomas Gleixner Cc: Vince Weaver Cc: bp@suse.de Cc: dave.hansen@linux.intel.com Cc: kan.liang@intel.com Link: http://lkml.kernel.org/r/1475598386-19597-1-git-send-email-lukasz.odzioba@intel.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/cstate.c | 30 ++++++++++++++++++++++++++---- 1 file changed, 26 insertions(+), 4 deletions(-) diff --git a/arch/x86/events/intel/cstate.c b/arch/x86/events/intel/cstate.c index 3ca87b5a8677..4f5ac726335f 100644 --- a/arch/x86/events/intel/cstate.c +++ b/arch/x86/events/intel/cstate.c @@ -48,7 +48,8 @@ * Scope: Core * MSR_CORE_C6_RESIDENCY: CORE C6 Residency Counter * perf code: 0x02 - * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW + * SKL,KNL * Scope: Core * MSR_CORE_C7_RESIDENCY: CORE C7 Residency Counter * perf code: 0x03 @@ -56,15 +57,16 @@ * Scope: Core * MSR_PKG_C2_RESIDENCY: Package C2 Residency Counter. * perf code: 0x00 - * Available model: SNB,IVB,HSW,BDW,SKL + * Available model: SNB,IVB,HSW,BDW,SKL,KNL * Scope: Package (physical package) * MSR_PKG_C3_RESIDENCY: Package C3 Residency Counter. * perf code: 0x01 - * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Available model: NHM,WSM,SNB,IVB,HSW,BDW,SKL,KNL * Scope: Package (physical package) * MSR_PKG_C6_RESIDENCY: Package C6 Residency Counter. * perf code: 0x02 - * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW,SKL + * Available model: SLM,AMT,NHM,WSM,SNB,IVB,HSW,BDW + * SKL,KNL * Scope: Package (physical package) * MSR_PKG_C7_RESIDENCY: Package C7 Residency Counter. * perf code: 0x03 @@ -118,6 +120,7 @@ struct cstate_model { /* Quirk flags */ #define SLM_PKG_C6_USE_C7_MSR (1UL << 0) +#define KNL_CORE_C6_MSR (1UL << 1) struct perf_cstate_msr { u64 msr; @@ -488,6 +491,18 @@ static const struct cstate_model slm_cstates __initconst = { .quirks = SLM_PKG_C6_USE_C7_MSR, }; + +static const struct cstate_model knl_cstates __initconst = { + .core_events = BIT(PERF_CSTATE_CORE_C6_RES), + + .pkg_events = BIT(PERF_CSTATE_PKG_C2_RES) | + BIT(PERF_CSTATE_PKG_C3_RES) | + BIT(PERF_CSTATE_PKG_C6_RES), + .quirks = KNL_CORE_C6_MSR, +}; + + + #define X86_CSTATES_MODEL(model, states) \ { X86_VENDOR_INTEL, 6, model, X86_FEATURE_ANY, (unsigned long) &(states) } @@ -523,6 +538,8 @@ static const struct x86_cpu_id intel_cstates_match[] __initconst = { X86_CSTATES_MODEL(INTEL_FAM6_SKYLAKE_MOBILE, snb_cstates), X86_CSTATES_MODEL(INTEL_FAM6_SKYLAKE_DESKTOP, snb_cstates), + + X86_CSTATES_MODEL(INTEL_FAM6_XEON_PHI_KNL, knl_cstates), { }, }; MODULE_DEVICE_TABLE(x86cpu, intel_cstates_match); @@ -558,6 +575,11 @@ static int __init cstate_probe(const struct cstate_model *cm) if (cm->quirks & SLM_PKG_C6_USE_C7_MSR) pkg_msr[PERF_CSTATE_PKG_C6_RES].msr = MSR_PKG_C7_RESIDENCY; + /* KNL has different MSR for CORE C6 */ + if (cm->quirks & KNL_CORE_C6_MSR) + pkg_msr[PERF_CSTATE_CORE_C6_RES].msr = MSR_KNL_CORE_C6_RESIDENCY; + + has_cstate_core = cstate_probe_msr(cm->core_events, PERF_CSTATE_CORE_EVENT_MAX, core_msr, core_events_attrs); From 0933840acf7b65d6d30a5b6089d882afea57aca3 Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Thu, 20 Oct 2016 13:10:11 +0200 Subject: [PATCH 2/4] perf/core: Protect PMU device removal with a 'pmu_bus_running' check, to fix CONFIG_DEBUG_TEST_DRIVER_REMOVE=y kernel panic CAI Qian reported a crash in the PMU uncore device removal code, enabled by the CONFIG_DEBUG_TEST_DRIVER_REMOVE=y option: https://marc.info/?l=linux-kernel&m=147688837328451 The reason for the crash is that perf_pmu_unregister() tries to remove a PMU device which is not added at this point. We add PMU devices only after pmu_bus is registered, which happens in the perf_event_sysfs_init() call and sets the 'pmu_bus_running' flag. The fix is to get the 'pmu_bus_running' flag state at the point the PMU is taken out of the PMU list and remove the device later only if it's set. Reported-by: CAI Qian Tested-by: CAI Qian Signed-off-by: Jiri Olsa Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Greg Kroah-Hartman Cc: Jiri Olsa Cc: Kan Liang Cc: Linus Torvalds Cc: Peter Zijlstra Cc: Rob Herring Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20161020111011.GA13361@krava Signed-off-by: Ingo Molnar --- kernel/events/core.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/kernel/events/core.c b/kernel/events/core.c index c6e47e97b33f..a5d2e62faf7e 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -8855,7 +8855,10 @@ EXPORT_SYMBOL_GPL(perf_pmu_register); void perf_pmu_unregister(struct pmu *pmu) { + int remove_device; + mutex_lock(&pmus_lock); + remove_device = pmu_bus_running; list_del_rcu(&pmu->entry); mutex_unlock(&pmus_lock); @@ -8869,10 +8872,12 @@ void perf_pmu_unregister(struct pmu *pmu) free_percpu(pmu->pmu_disable_count); if (pmu->type >= PERF_TYPE_MAX) idr_remove(&pmu_idr, pmu->type); - if (pmu->nr_addr_filters) - device_remove_file(pmu->dev, &dev_attr_nr_addr_filters); - device_del(pmu->dev); - put_device(pmu->dev); + if (remove_device) { + if (pmu->nr_addr_filters) + device_remove_file(pmu->dev, &dev_attr_nr_addr_filters); + device_del(pmu->dev); + put_device(pmu->dev); + } free_pmu_context(pmu); } EXPORT_SYMBOL_GPL(perf_pmu_unregister); From 5aab90ce1ec449912a2ebc4d45e0c85dac29e9dd Mon Sep 17 00:00:00 2001 From: Jiri Olsa Date: Wed, 26 Oct 2016 11:48:24 +0200 Subject: [PATCH 3/4] perf/powerpc: Don't call perf_event_disable() from atomic context The trinity syscall fuzzer triggered following WARN() on powerpc: WARNING: CPU: 9 PID: 2998 at arch/powerpc/kernel/hw_breakpoint.c:278 ... NIP [c00000000093aedc] .hw_breakpoint_handler+0x28c/0x2b0 LR [c00000000093aed8] .hw_breakpoint_handler+0x288/0x2b0 Call Trace: [c0000002f7933580] [c00000000093aed8] .hw_breakpoint_handler+0x288/0x2b0 (unreliable) [c0000002f7933630] [c0000000000f671c] .notifier_call_chain+0x7c/0xf0 [c0000002f79336d0] [c0000000000f6abc] .__atomic_notifier_call_chain+0xbc/0x1c0 [c0000002f7933780] [c0000000000f6c40] .notify_die+0x70/0xd0 [c0000002f7933820] [c00000000001a74c] .do_break+0x4c/0x100 [c0000002f7933920] [c0000000000089fc] handle_dabr_fault+0x14/0x48 Followed by a lockdep warning: =============================== [ INFO: suspicious RCU usage. ] 4.8.0-rc5+ #7 Tainted: G W ------------------------------- ./include/linux/rcupdate.h:556 Illegal context switch in RCU read-side critical section! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 2 locks held by ls/2998: #0: (rcu_read_lock){......}, at: [] .__atomic_notifier_call_chain+0x0/0x1c0 #1: (rcu_read_lock){......}, at: [] .hw_breakpoint_handler+0x0/0x2b0 stack backtrace: CPU: 9 PID: 2998 Comm: ls Tainted: G W 4.8.0-rc5+ #7 Call Trace: [c0000002f7933150] [c00000000094b1f8] .dump_stack+0xe0/0x14c (unreliable) [c0000002f79331e0] [c00000000013c468] .lockdep_rcu_suspicious+0x138/0x180 [c0000002f7933270] [c0000000001005d8] .___might_sleep+0x278/0x2e0 [c0000002f7933300] [c000000000935584] .mutex_lock_nested+0x64/0x5a0 [c0000002f7933410] [c00000000023084c] .perf_event_ctx_lock_nested+0x16c/0x380 [c0000002f7933500] [c000000000230a80] .perf_event_disable+0x20/0x60 [c0000002f7933580] [c00000000093aeec] .hw_breakpoint_handler+0x29c/0x2b0 [c0000002f7933630] [c0000000000f671c] .notifier_call_chain+0x7c/0xf0 [c0000002f79336d0] [c0000000000f6abc] .__atomic_notifier_call_chain+0xbc/0x1c0 [c0000002f7933780] [c0000000000f6c40] .notify_die+0x70/0xd0 [c0000002f7933820] [c00000000001a74c] .do_break+0x4c/0x100 [c0000002f7933920] [c0000000000089fc] handle_dabr_fault+0x14/0x48 While it looks like the first WARN() is probably valid, the other one is triggered by disabling event via perf_event_disable() from atomic context. The event is disabled here in case we were not able to emulate the instruction that hit the breakpoint. By disabling the event we unschedule the event and make sure it's not scheduled back. But we can't call perf_event_disable() from atomic context, instead we need to use the event's pending_disable irq_work method to disable it. Reported-by: Jan Stancek Signed-off-by: Jiri Olsa Signed-off-by: Peter Zijlstra (Intel) Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Huang Ying Cc: Jiri Olsa Cc: Linus Torvalds Cc: Michael Neuling Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/20161026094824.GA21397@krava Signed-off-by: Ingo Molnar --- arch/powerpc/kernel/hw_breakpoint.c | 2 +- include/linux/perf_event.h | 1 + kernel/events/core.c | 10 ++++++++-- 3 files changed, 10 insertions(+), 3 deletions(-) diff --git a/arch/powerpc/kernel/hw_breakpoint.c b/arch/powerpc/kernel/hw_breakpoint.c index 9781c69eae57..03d089b3ed72 100644 --- a/arch/powerpc/kernel/hw_breakpoint.c +++ b/arch/powerpc/kernel/hw_breakpoint.c @@ -275,7 +275,7 @@ int hw_breakpoint_handler(struct die_args *args) if (!stepped) { WARN(1, "Unable to handle hardware breakpoint. Breakpoint at " "0x%lx will be disabled.", info->address); - perf_event_disable(bp); + perf_event_disable_inatomic(bp); goto out; } /* diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index 060d0ede88df..4741ecdb9817 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1257,6 +1257,7 @@ extern u64 perf_swevent_set_period(struct perf_event *event); extern void perf_event_enable(struct perf_event *event); extern void perf_event_disable(struct perf_event *event); extern void perf_event_disable_local(struct perf_event *event); +extern void perf_event_disable_inatomic(struct perf_event *event); extern void perf_event_task_tick(void); #else /* !CONFIG_PERF_EVENTS: */ static inline void * diff --git a/kernel/events/core.c b/kernel/events/core.c index a5d2e62faf7e..0e292132efac 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -1960,6 +1960,12 @@ void perf_event_disable(struct perf_event *event) } EXPORT_SYMBOL_GPL(perf_event_disable); +void perf_event_disable_inatomic(struct perf_event *event) +{ + event->pending_disable = 1; + irq_work_queue(&event->pending); +} + static void perf_set_shadow_time(struct perf_event *event, struct perf_event_context *ctx, u64 tstamp) @@ -7075,8 +7081,8 @@ static int __perf_event_overflow(struct perf_event *event, if (events && atomic_dec_and_test(&event->event_limit)) { ret = 1; event->pending_kill = POLL_HUP; - event->pending_disable = 1; - irq_work_queue(&event->pending); + + perf_event_disable_inatomic(event); } READ_ONCE(event->overflow_handler)(event, data, regs); From f92b7604149a55cb601fc0b52911b1e11f0f2514 Mon Sep 17 00:00:00 2001 From: Imre Palik Date: Fri, 21 Oct 2016 01:18:59 -0700 Subject: [PATCH 4/4] perf/x86/intel: Honour the CPUID for number of fixed counters in hypervisors perf doesn't seem to honour the number of fixed counters specified by CPUID leaf 0xa. It always assumes that Intel CPUs have at least 3 fixed counters. So if some of the fixed counters are masked out by the hypervisor, it still tries to check/set them. This patch makes perf behave nicer when the kernel is running under a hypervisor that doesn't expose all the counters. This patch contains some ideas from Matt Wilson. Signed-off-by: Imre Palik Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Andi Kleen Cc: Alexander Kozyrev Cc: Alexander Shishkin Cc: Arnaldo Carvalho de Melo Cc: Artyom Kuanbekov Cc: David Carrillo-Cisneros Cc: David Woodhouse Cc: H. Peter Anvin Cc: Jiri Olsa Cc: Kan Liang Cc: Linus Torvalds Cc: Matt Wilson Cc: Peter Zijlstra Cc: Stephane Eranian Cc: Thomas Gleixner Link: http://lkml.kernel.org/r/1477037939-15605-1-git-send-email-imrep.amz@gmail.com Signed-off-by: Ingo Molnar --- arch/x86/events/intel/core.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index eab0915f5995..a74a2dbc0180 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -3607,10 +3607,14 @@ __init int intel_pmu_init(void) /* * Quirk: v2 perfmon does not report fixed-purpose events, so - * assume at least 3 events: + * assume at least 3 events, when not running in a hypervisor: */ - if (version > 1) - x86_pmu.num_counters_fixed = max((int)edx.split.num_counters_fixed, 3); + if (version > 1) { + int assume = 3 * !boot_cpu_has(X86_FEATURE_HYPERVISOR); + + x86_pmu.num_counters_fixed = + max((int)edx.split.num_counters_fixed, assume); + } if (boot_cpu_has(X86_FEATURE_PDCM)) { u64 capabilities;