From 3d083407a16698de86b42aee0da2ffb280b5cb7e Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sat, 27 Feb 2010 17:24:15 +0100 Subject: x86/hw-breakpoints: Remove the name field Remove the name field from the arch_hw_breakpoint. We never deal with target symbols in the arch level, neither do we need to ever store it. It's a legacy for the previous version of the x86 breakpoint backend. Let's remove it. Signed-off-by: Frederic Weisbecker Cc: K.Prasad Cc: Linus Torvalds --- arch/x86/include/asm/hw_breakpoint.h | 1 - arch/x86/kernel/hw_breakpoint.c | 7 ------- 2 files changed, 8 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/hw_breakpoint.h b/arch/x86/include/asm/hw_breakpoint.h index 0675a7c4c20e..2a1bd8f4f23a 100644 --- a/arch/x86/include/asm/hw_breakpoint.h +++ b/arch/x86/include/asm/hw_breakpoint.h @@ -10,7 +10,6 @@ * (display/resolving) */ struct arch_hw_breakpoint { - char *name; /* Contains name of the symbol to set bkpt */ unsigned long address; u8 len; u8 type; diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index dca2802c666f..41e08dff0161 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -343,13 +343,6 @@ int arch_validate_hwbkpt_settings(struct perf_event *bp, return ret; } - /* - * For kernel-addresses, either the address or symbol name can be - * specified. - */ - if (info->name) - info->address = (unsigned long) - kallsyms_lookup_name(info->name); /* * Check that the low-order bits of the address are appropriate * for the alignment implied by len. -- cgit v1.2.3 From 1e259e0a9982078896f3404240096cbea01daca4 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Sun, 28 Feb 2010 20:51:15 +0100 Subject: hw-breakpoints: Remove stub unthrottle callback We support event unthrottling in breakpoint events. It means that if we have more than sysctl_perf_event_sample_rate/HZ, perf will throttle, ignoring subsequent events until the next tick. So if ptrace exceeds this max rate, it will omit events, which breaks the ptrace determinism that is supposed to report every triggered breakpoints. This is likely to happen if we set sysctl_perf_event_sample_rate to 1. This patch removes support for unthrottling in breakpoint events to break throttling and restore ptrace determinism. Signed-off-by: Frederic Weisbecker Cc: 2.6.33.x Cc: Peter Zijlstra Cc: K.Prasad Cc: Paul Mackerras --- arch/x86/kernel/hw_breakpoint.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/hw_breakpoint.c b/arch/x86/kernel/hw_breakpoint.c index bb6006e3e295..1e8ceadc0d6a 100644 --- a/arch/x86/kernel/hw_breakpoint.c +++ b/arch/x86/kernel/hw_breakpoint.c @@ -531,8 +531,3 @@ void hw_breakpoint_pmu_read(struct perf_event *bp) { /* TODO */ } - -void hw_breakpoint_pmu_unthrottle(struct perf_event *bp) -{ - /* TODO */ -} -- cgit v1.2.3 From 1d6040f17d12a65b9f7ab4cb9fd6d721206b79ec Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Thu, 25 Feb 2010 19:40:46 +0100 Subject: perf, x86: make IBS macros available in perf_event.h This patch moves code from oprofile to perf_event.h to make it also available for usage by perf. Signed-off-by: Robert Richter --- arch/x86/include/asm/perf_event.h | 10 ++++++++++ arch/x86/oprofile/op_model_amd.c | 11 ----------- 2 files changed, 10 insertions(+), 11 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index befd172c82ad..4933ccde96c4 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -117,6 +117,16 @@ union cpuid10_edx { */ #define X86_PMC_IDX_FIXED_BTS (X86_PMC_IDX_FIXED + 16) +/* IbsFetchCtl bits/masks */ +#define IBS_FETCH_RAND_EN (1ULL<<57) +#define IBS_FETCH_VAL (1ULL<<49) +#define IBS_FETCH_ENABLE (1ULL<<48) +#define IBS_FETCH_CNT_MASK 0xFFFF0000ULL + +/* IbsOpCtl bits */ +#define IBS_OP_CNT_CTL (1ULL<<19) +#define IBS_OP_VAL (1ULL<<18) +#define IBS_OP_ENABLE (1ULL<<17) #ifdef CONFIG_PERF_EVENTS extern void init_hw_perf_events(void); diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index 6a58256dce9f..c67174917305 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -46,17 +46,6 @@ static unsigned long reset_value[NUM_VIRT_COUNTERS]; -/* IbsFetchCtl bits/masks */ -#define IBS_FETCH_RAND_EN (1ULL<<57) -#define IBS_FETCH_VAL (1ULL<<49) -#define IBS_FETCH_ENABLE (1ULL<<48) -#define IBS_FETCH_CNT_MASK 0xFFFF0000ULL - -/* IbsOpCtl bits */ -#define IBS_OP_CNT_CTL (1ULL<<19) -#define IBS_OP_VAL (1ULL<<18) -#define IBS_OP_ENABLE (1ULL<<17) - #define IBS_FETCH_SIZE 6 #define IBS_OP_SIZE 12 -- cgit v1.2.3 From a163b1099dc7016704043c7fc572ae42519f08f7 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Thu, 25 Feb 2010 19:43:07 +0100 Subject: perf, x86: add some IBS macros to perf_event.h Signed-off-by: Robert Richter --- arch/x86/include/asm/perf_event.h | 4 +++- arch/x86/oprofile/op_model_amd.c | 6 +++--- 2 files changed, 6 insertions(+), 4 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index 4933ccde96c4..c7f60e1297ab 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -121,12 +121,14 @@ union cpuid10_edx { #define IBS_FETCH_RAND_EN (1ULL<<57) #define IBS_FETCH_VAL (1ULL<<49) #define IBS_FETCH_ENABLE (1ULL<<48) -#define IBS_FETCH_CNT_MASK 0xFFFF0000ULL +#define IBS_FETCH_CNT 0xFFFF0000ULL +#define IBS_FETCH_MAX_CNT 0x0000FFFFULL /* IbsOpCtl bits */ #define IBS_OP_CNT_CTL (1ULL<<19) #define IBS_OP_VAL (1ULL<<18) #define IBS_OP_ENABLE (1ULL<<17) +#define IBS_OP_MAX_CNT 0x0000FFFFULL #ifdef CONFIG_PERF_EVENTS extern void init_hw_perf_events(void); diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index c67174917305..8ddb9fa9c1b2 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -279,7 +279,7 @@ op_amd_handle_ibs(struct pt_regs * const regs, oprofile_write_commit(&entry); /* reenable the IRQ */ - ctl &= ~(IBS_FETCH_VAL | IBS_FETCH_CNT_MASK); + ctl &= ~(IBS_FETCH_VAL | IBS_FETCH_CNT); ctl |= IBS_FETCH_ENABLE; wrmsrl(MSR_AMD64_IBSFETCHCTL, ctl); } @@ -319,7 +319,7 @@ static inline void op_amd_start_ibs(void) return; if (ibs_config.fetch_enabled) { - val = (ibs_config.max_cnt_fetch >> 4) & 0xFFFF; + val = (ibs_config.max_cnt_fetch >> 4) & IBS_FETCH_MAX_CNT; val |= ibs_config.rand_en ? IBS_FETCH_RAND_EN : 0; val |= IBS_FETCH_ENABLE; wrmsrl(MSR_AMD64_IBSFETCHCTL, val); @@ -341,7 +341,7 @@ static inline void op_amd_start_ibs(void) * avoid underflows. */ ibs_op_ctl = min(ibs_op_ctl + IBS_RANDOM_MAXCNT_OFFSET, - 0xFFFFULL); + IBS_OP_MAX_CNT); } if (ibs_caps & IBS_CAPS_OPCNT && ibs_config.dispatched_ops) ibs_op_ctl |= IBS_OP_CNT_CTL; -- cgit v1.2.3 From bb1165d6882f423f90fc7007a88c6c993b7c2ac4 Mon Sep 17 00:00:00 2001 From: Robert Richter Date: Mon, 1 Mar 2010 14:21:23 +0100 Subject: perf, x86: rename macro in ARCH_PERFMON_EVENTSEL_ENABLE For consistency reasons this patch renames ARCH_PERFMON_EVENTSEL0_ENABLE to ARCH_PERFMON_EVENTSEL_ENABLE. The following is performed: $ sed -i -e s/ARCH_PERFMON_EVENTSEL0_ENABLE/ARCH_PERFMON_EVENTSEL_ENABLE/g \ arch/x86/include/asm/perf_event.h arch/x86/kernel/cpu/perf_event.c \ arch/x86/kernel/cpu/perf_event_p6.c \ arch/x86/kernel/cpu/perfctr-watchdog.c \ arch/x86/oprofile/op_model_amd.c arch/x86/oprofile/op_model_ppro.c Signed-off-by: Robert Richter --- arch/x86/include/asm/perf_event.h | 2 +- arch/x86/kernel/cpu/perf_event.c | 8 ++++---- arch/x86/kernel/cpu/perf_event_p6.c | 8 ++++---- arch/x86/kernel/cpu/perfctr-watchdog.c | 2 +- arch/x86/oprofile/op_model_amd.c | 6 +++--- arch/x86/oprofile/op_model_ppro.c | 6 +++--- 6 files changed, 16 insertions(+), 16 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index c7f60e1297ab..80e693684f18 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -18,7 +18,7 @@ #define MSR_ARCH_PERFMON_EVENTSEL0 0x186 #define MSR_ARCH_PERFMON_EVENTSEL1 0x187 -#define ARCH_PERFMON_EVENTSEL0_ENABLE (1 << 22) +#define ARCH_PERFMON_EVENTSEL_ENABLE (1 << 22) #define ARCH_PERFMON_EVENTSEL_ANY (1 << 21) #define ARCH_PERFMON_EVENTSEL_INT (1 << 20) #define ARCH_PERFMON_EVENTSEL_OS (1 << 17) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 641ccb9dddbc..6531b4bdb22d 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -553,9 +553,9 @@ static void x86_pmu_disable_all(void) if (!test_bit(idx, cpuc->active_mask)) continue; rdmsrl(x86_pmu.eventsel + idx, val); - if (!(val & ARCH_PERFMON_EVENTSEL0_ENABLE)) + if (!(val & ARCH_PERFMON_EVENTSEL_ENABLE)) continue; - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(x86_pmu.eventsel + idx, val); } } @@ -590,7 +590,7 @@ static void x86_pmu_enable_all(void) continue; val = event->hw.config; - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(x86_pmu.eventsel + idx, val); } } @@ -853,7 +853,7 @@ void hw_perf_enable(void) static inline void __x86_pmu_enable_event(struct hw_perf_event *hwc, int idx) { (void)checking_wrmsrl(hwc->config_base + idx, - hwc->config | ARCH_PERFMON_EVENTSEL0_ENABLE); + hwc->config | ARCH_PERFMON_EVENTSEL_ENABLE); } static inline void x86_pmu_disable_event(struct hw_perf_event *hwc, int idx) diff --git a/arch/x86/kernel/cpu/perf_event_p6.c b/arch/x86/kernel/cpu/perf_event_p6.c index 1ca5ba078afd..a4e67b99d91c 100644 --- a/arch/x86/kernel/cpu/perf_event_p6.c +++ b/arch/x86/kernel/cpu/perf_event_p6.c @@ -62,7 +62,7 @@ static void p6_pmu_disable_all(void) /* p6 only has one enable register */ rdmsrl(MSR_P6_EVNTSEL0, val); - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(MSR_P6_EVNTSEL0, val); } @@ -72,7 +72,7 @@ static void p6_pmu_enable_all(void) /* p6 only has one enable register */ rdmsrl(MSR_P6_EVNTSEL0, val); - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(MSR_P6_EVNTSEL0, val); } @@ -83,7 +83,7 @@ p6_pmu_disable_event(struct hw_perf_event *hwc, int idx) u64 val = P6_NOP_EVENT; if (cpuc->enabled) - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; (void)checking_wrmsrl(hwc->config_base + idx, val); } @@ -95,7 +95,7 @@ static void p6_pmu_enable_event(struct hw_perf_event *hwc, int idx) val = hwc->config; if (cpuc->enabled) - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; (void)checking_wrmsrl(hwc->config_base + idx, val); } diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index 74f4e85a5727..fb329e9f8494 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -680,7 +680,7 @@ static int setup_intel_arch_watchdog(unsigned nmi_hz) cpu_nmi_set_wd_enabled(); apic_write(APIC_LVTPC, APIC_DM_NMI); - evntsel |= ARCH_PERFMON_EVENTSEL0_ENABLE; + evntsel |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsr(evntsel_msr, evntsel, 0); intel_arch_wd_ops.checkbit = 1ULL << (eax.split.bit_width - 1); return 1; diff --git a/arch/x86/oprofile/op_model_amd.c b/arch/x86/oprofile/op_model_amd.c index 8ddb9fa9c1b2..090cbbec7dbd 100644 --- a/arch/x86/oprofile/op_model_amd.c +++ b/arch/x86/oprofile/op_model_amd.c @@ -171,7 +171,7 @@ static void op_amd_setup_ctrs(struct op_x86_model_spec const *model, continue; } rdmsrl(msrs->controls[i].addr, val); - if (val & ARCH_PERFMON_EVENTSEL0_ENABLE) + if (val & ARCH_PERFMON_EVENTSEL_ENABLE) op_x86_warn_in_use(i); val &= model->reserved; wrmsrl(msrs->controls[i].addr, val); @@ -398,7 +398,7 @@ static void op_amd_start(struct op_msrs const * const msrs) if (!reset_value[op_x86_phys_to_virt(i)]) continue; rdmsrl(msrs->controls[i].addr, val); - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(msrs->controls[i].addr, val); } @@ -418,7 +418,7 @@ static void op_amd_stop(struct op_msrs const * const msrs) if (!reset_value[op_x86_phys_to_virt(i)]) continue; rdmsrl(msrs->controls[i].addr, val); - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(msrs->controls[i].addr, val); } diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index 5d1727ba409e..2bf90fafa7b5 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -88,7 +88,7 @@ static void ppro_setup_ctrs(struct op_x86_model_spec const *model, continue; } rdmsrl(msrs->controls[i].addr, val); - if (val & ARCH_PERFMON_EVENTSEL0_ENABLE) + if (val & ARCH_PERFMON_EVENTSEL_ENABLE) op_x86_warn_in_use(i); val &= model->reserved; wrmsrl(msrs->controls[i].addr, val); @@ -166,7 +166,7 @@ static void ppro_start(struct op_msrs const * const msrs) for (i = 0; i < num_counters; ++i) { if (reset_value[i]) { rdmsrl(msrs->controls[i].addr, val); - val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + val |= ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(msrs->controls[i].addr, val); } } @@ -184,7 +184,7 @@ static void ppro_stop(struct op_msrs const * const msrs) if (!reset_value[i]) continue; rdmsrl(msrs->controls[i].addr, val); - val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + val &= ~ARCH_PERFMON_EVENTSEL_ENABLE; wrmsrl(msrs->controls[i].addr, val); } } -- cgit v1.2.3 From 320ebf09cbb6d01954c9a060266aa8e0d27f4638 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Tue, 2 Mar 2010 12:35:37 +0100 Subject: perf, x86: Restrict the ANY flag The ANY flag can show SMT data of another task (like 'top'), so we want to disable it when system-wide profiling is disabled. Signed-off-by: Peter Zijlstra LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/perf_event.c | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch') diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 6531b4bdb22d..aab2e1ce9dee 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -503,6 +503,9 @@ static int __hw_perf_event_init(struct perf_event *event) */ if (attr->type == PERF_TYPE_RAW) { hwc->config |= x86_pmu.raw_event(attr->config); + if ((hwc->config & ARCH_PERFMON_EVENTSEL_ANY) && + perf_paranoid_cpu() && !capable(CAP_SYS_ADMIN)) + return -EACCES; return 0; } -- cgit v1.2.3 From b622d644c7d61a5cb95b74e7b143c263bed21f0a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 1 Feb 2010 15:36:30 +0100 Subject: perf_events, x86: Fixup fixed counter constraints Patch 1da53e0230 ("perf_events, x86: Improve x86 event scheduling") lost us one of the fixed purpose counters and then ed8777fc13 ("perf_events, x86: Fix event constraint masks") broke it even further. Widen the fixed event mask to event+umask and specify the full config for each of the 3 fixed purpose counters. Then let the init code fill out the placement for the GP regs based on the cpuid info. Signed-off-by: Peter Zijlstra Cc: Stephane Eranian LKML-Reference: Signed-off-by: Ingo Molnar --- arch/x86/include/asm/perf_event.h | 2 +- arch/x86/kernel/cpu/perf_event.c | 25 ++++++++++++++++++------- arch/x86/kernel/cpu/perf_event_intel.c | 31 +++++++++++++++++++++---------- 3 files changed, 40 insertions(+), 18 deletions(-) (limited to 'arch') diff --git a/arch/x86/include/asm/perf_event.h b/arch/x86/include/asm/perf_event.h index 80e693684f18..db6109a885a7 100644 --- a/arch/x86/include/asm/perf_event.h +++ b/arch/x86/include/asm/perf_event.h @@ -50,7 +50,7 @@ INTEL_ARCH_INV_MASK| \ INTEL_ARCH_EDGE_MASK|\ INTEL_ARCH_UNIT_MASK|\ - INTEL_ARCH_EVTSEL_MASK) + INTEL_ARCH_EVENT_MASK) #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL 0x3c #define ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK (0x00 << 8) diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index aab2e1ce9dee..bfc43fa208bc 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -73,10 +73,10 @@ struct debug_store { struct event_constraint { union { unsigned long idxmsk[BITS_TO_LONGS(X86_PMC_IDX_MAX)]; - u64 idxmsk64[1]; + u64 idxmsk64; }; - int code; - int cmask; + u64 code; + u64 cmask; int weight; }; @@ -103,7 +103,7 @@ struct cpu_hw_events { }; #define __EVENT_CONSTRAINT(c, n, m, w) {\ - { .idxmsk64[0] = (n) }, \ + { .idxmsk64 = (n) }, \ .code = (c), \ .cmask = (m), \ .weight = (w), \ @@ -116,7 +116,7 @@ struct cpu_hw_events { EVENT_CONSTRAINT(c, n, INTEL_ARCH_EVTSEL_MASK) #define FIXED_EVENT_CONSTRAINT(c, n) \ - EVENT_CONSTRAINT(c, n, INTEL_ARCH_FIXED_MASK) + EVENT_CONSTRAINT(c, (1ULL << (32+n)), INTEL_ARCH_FIXED_MASK) #define EVENT_CONSTRAINT_END \ EVENT_CONSTRAINT(0, 0, 0) @@ -615,8 +615,8 @@ static int x86_schedule_events(struct cpu_hw_events *cpuc, int n, int *assign) bitmap_zero(used_mask, X86_PMC_IDX_MAX); for (i = 0; i < n; i++) { - constraints[i] = - x86_pmu.get_event_constraints(cpuc, cpuc->event_list[i]); + c = x86_pmu.get_event_constraints(cpuc, cpuc->event_list[i]); + constraints[i] = c; } /* @@ -1350,6 +1350,7 @@ static void __init pmu_check_apic(void) void __init init_hw_perf_events(void) { + struct event_constraint *c; int err; pr_info("Performance Events: "); @@ -1398,6 +1399,16 @@ void __init init_hw_perf_events(void) __EVENT_CONSTRAINT(0, (1ULL << x86_pmu.num_events) - 1, 0, x86_pmu.num_events); + if (x86_pmu.event_constraints) { + for_each_event_constraint(c, x86_pmu.event_constraints) { + if (c->cmask != INTEL_ARCH_FIXED_MASK) + continue; + + c->idxmsk64 |= (1ULL << x86_pmu.num_events) - 1; + c->weight += x86_pmu.num_events; + } + } + pr_info("... version: %d\n", x86_pmu.version); pr_info("... bit width: %d\n", x86_pmu.event_bits); pr_info("... generic registers: %d\n", x86_pmu.num_events); diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index cf6590cf4a5f..4fbdfe5708d9 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -1,7 +1,7 @@ #ifdef CONFIG_CPU_SUP_INTEL /* - * Intel PerfMon v3. Used on Core2 and later. + * Intel PerfMon, used on Core and later. */ static const u64 intel_perfmon_event_map[] = { @@ -27,8 +27,14 @@ static struct event_constraint intel_core_event_constraints[] = static struct event_constraint intel_core2_event_constraints[] = { - FIXED_EVENT_CONSTRAINT(0xc0, (0x3|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */ - FIXED_EVENT_CONSTRAINT(0x3c, (0x3|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* + * Core2 has Fixed Counter 2 listed as CPU_CLK_UNHALTED.REF and event + * 0x013c as CPU_CLK_UNHALTED.BUS and specifies there is a fixed + * ratio between these counters. + */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ INTEL_EVENT_CONSTRAINT(0x10, 0x1), /* FP_COMP_OPS_EXE */ INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */ INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */ @@ -37,14 +43,16 @@ static struct event_constraint intel_core2_event_constraints[] = INTEL_EVENT_CONSTRAINT(0x18, 0x1), /* IDLE_DURING_DIV */ INTEL_EVENT_CONSTRAINT(0x19, 0x2), /* DELAYED_BYPASS */ INTEL_EVENT_CONSTRAINT(0xa1, 0x1), /* RS_UOPS_DISPATCH_CYCLES */ + INTEL_EVENT_CONSTRAINT(0xc9, 0x1), /* ITLB_MISS_RETIRED (T30-9) */ INTEL_EVENT_CONSTRAINT(0xcb, 0x1), /* MEM_LOAD_RETIRED */ EVENT_CONSTRAINT_END }; static struct event_constraint intel_nehalem_event_constraints[] = { - FIXED_EVENT_CONSTRAINT(0xc0, (0xf|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */ - FIXED_EVENT_CONSTRAINT(0x3c, (0xf|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ INTEL_EVENT_CONSTRAINT(0x40, 0x3), /* L1D_CACHE_LD */ INTEL_EVENT_CONSTRAINT(0x41, 0x3), /* L1D_CACHE_ST */ INTEL_EVENT_CONSTRAINT(0x42, 0x3), /* L1D_CACHE_LOCK */ @@ -58,8 +66,9 @@ static struct event_constraint intel_nehalem_event_constraints[] = static struct event_constraint intel_westmere_event_constraints[] = { - FIXED_EVENT_CONSTRAINT(0xc0, (0xf|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */ - FIXED_EVENT_CONSTRAINT(0x3c, (0xf|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ INTEL_EVENT_CONSTRAINT(0x51, 0x3), /* L1D */ INTEL_EVENT_CONSTRAINT(0x60, 0x1), /* OFFCORE_REQUESTS_OUTSTANDING */ INTEL_EVENT_CONSTRAINT(0x63, 0x3), /* CACHE_LOCK_CYCLES */ @@ -68,8 +77,9 @@ static struct event_constraint intel_westmere_event_constraints[] = static struct event_constraint intel_gen_event_constraints[] = { - FIXED_EVENT_CONSTRAINT(0xc0, (0x3|(1ULL<<32))), /* INSTRUCTIONS_RETIRED */ - FIXED_EVENT_CONSTRAINT(0x3c, (0x3|(1ULL<<33))), /* UNHALTED_CORE_CYCLES */ + FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ + FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ + /* FIXED_EVENT_CONSTRAINT(0x013c, 2), CPU_CLK_UNHALTED.REF */ EVENT_CONSTRAINT_END }; @@ -935,7 +945,7 @@ static __init int intel_pmu_init(void) x86_pmu.event_constraints = intel_nehalem_event_constraints; pr_cont("Nehalem/Corei7 events, "); break; - case 28: + case 28: /* Atom */ memcpy(hw_cache_event_ids, atom_hw_cache_event_ids, sizeof(hw_cache_event_ids)); @@ -951,6 +961,7 @@ static __init int intel_pmu_init(void) x86_pmu.event_constraints = intel_westmere_event_constraints; pr_cont("Westmere events, "); break; + default: /* * default constraints for v2 and up -- cgit v1.2.3 From 29044ad1509ecc229f1d5a31aeed7a8dc61a71c4 Mon Sep 17 00:00:00 2001 From: Frederic Weisbecker Date: Wed, 3 Mar 2010 02:25:22 +0100 Subject: x86/stacktrace: Don't dereference bad frame pointers Callers of a stacktrace might pass bad frame pointers. Those are usually checked for safety in stack walking helpers before any dereferencing, but this is not the case when we need to go through one more frame pointer that backlinks the irq stack to the previous one, as we don't have any reliable address boudaries to compare this frame pointer against. This raises crashes when we record callchains for ftrace events with perf because we don't use the right helpers to capture registers there. We get wrong frame pointers as we call task_pt_regs() even on kernel threads, which is a wrong thing as it gives us the initial state of any kernel threads freshly created. This is even not what we want for user tasks. What we want is a hot snapshot of registers when the ftrace event triggers, not the state before a task entered the kernel. This requires more thoughts to do it correctly though. So first put a guardian to ensure the given frame pointer can be dereferenced to avoid crashes. We'll think about how to fix the callers in a subsequent patch. Signed-off-by: Frederic Weisbecker Cc: Ingo Molnar Cc: Thomas Gleixner Cc: H. Peter Anvin Cc: Peter Zijlstra Cc: Paul Mackerras Cc: Steven Rostedt Cc: 2.6.33.x Cc: Arnaldo Carvalho de Melo --- arch/x86/kernel/dumpstack_64.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'arch') diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index 0ad9597073f5..a6c906c9b193 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -125,9 +125,15 @@ fixup_bp_irq_link(unsigned long bp, unsigned long *stack, { #ifdef CONFIG_FRAME_POINTER struct stack_frame *frame = (struct stack_frame *)bp; + unsigned long next; - if (!in_irq_stack(stack, irq_stack, irq_stack_end)) - return (unsigned long)frame->next_frame; + if (!in_irq_stack(stack, irq_stack, irq_stack_end)) { + if (!probe_kernel_address(&frame->next_frame, next)) + return next; + else + WARN_ONCE(1, "Perf: bad frame pointer = %p in " + "callchain\n", &frame->next_frame); + } #endif return bp; } -- cgit v1.2.3 From dc1d628a67a8f042e711ea5accc0beedc3ef0092 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 3 Mar 2010 15:55:04 +0100 Subject: perf: Provide generic perf_sample_data initialization This makes it easier to extend perf_sample_data and fixes a bug on arm and sparc, which failed to set ->raw to NULL, which can cause crashes when combined with PERF_SAMPLE_RAW. It also optimizes PowerPC and tracepoint, because the struct initialization is forced to zero out the whole structure. Signed-off-by: Peter Zijlstra Acked-by: Jean Pihet Reviewed-by: Frederic Weisbecker Acked-by: David S. Miller Cc: Jamie Iles Cc: Paul Mackerras Cc: Stephane Eranian Cc: stable@kernel.org LKML-Reference: <20100304140100.315416040@chello.nl> Signed-off-by: Ingo Molnar --- arch/arm/kernel/perf_event.c | 4 ++-- arch/powerpc/kernel/perf_event.c | 8 ++++---- arch/sparc/kernel/perf_event.c | 2 +- arch/x86/kernel/cpu/perf_event.c | 3 +-- arch/x86/kernel/cpu/perf_event_intel.c | 6 ++---- 5 files changed, 10 insertions(+), 13 deletions(-) (limited to 'arch') diff --git a/arch/arm/kernel/perf_event.c b/arch/arm/kernel/perf_event.c index c54ceb3d1f97..3875d99cc40f 100644 --- a/arch/arm/kernel/perf_event.c +++ b/arch/arm/kernel/perf_event.c @@ -965,7 +965,7 @@ armv6pmu_handle_irq(int irq_num, */ armv6_pmcr_write(pmcr); - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx <= armpmu->num_events; ++idx) { @@ -1945,7 +1945,7 @@ static irqreturn_t armv7pmu_handle_irq(int irq_num, void *dev) */ regs = get_irq_regs(); - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); for (idx = 0; idx <= armpmu->num_events; ++idx) { diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index b6cf8f1f4d35..5120bd44f69a 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -1164,10 +1164,10 @@ static void record_and_restart(struct perf_event *event, unsigned long val, * Finally record data if requested. */ if (record) { - struct perf_sample_data data = { - .addr = ~0ULL, - .period = event->hw.last_period, - }; + struct perf_sample_data data; + + perf_sample_data_init(&data, ~0ULL); + data.period = event->hw.last_period; if (event->attr.sample_type & PERF_SAMPLE_ADDR) perf_get_data_addr(regs, &data.addr); diff --git a/arch/sparc/kernel/perf_event.c b/arch/sparc/kernel/perf_event.c index 9f2b2bac8b2b..6504208f375f 100644 --- a/arch/sparc/kernel/perf_event.c +++ b/arch/sparc/kernel/perf_event.c @@ -1189,7 +1189,7 @@ static int __kprobes perf_event_nmi_handler(struct notifier_block *self, regs = args->regs; - data.addr = 0; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 97cddbf32936..42aafd11e170 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -1097,8 +1097,7 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) int idx, handled = 0; u64 val; - data.addr = 0; - data.raw = NULL; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 73102df8bfc1..44b60c852107 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -590,10 +590,9 @@ static void intel_pmu_drain_bts_buffer(void) ds->bts_index = ds->bts_buffer_base; + perf_sample_data_init(&data, 0); data.period = event->hw.last_period; - data.addr = 0; - data.raw = NULL; regs.ip = 0; /* @@ -742,8 +741,7 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) int bit, loops; u64 ack, status; - data.addr = 0; - data.raw = NULL; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); -- cgit v1.2.3