diff options
author | Ingo Molnar <mingo@elte.hu> | 2009-09-21 10:20:38 (GMT) |
---|---|---|
committer | Ingo Molnar <mingo@elte.hu> | 2009-09-21 12:34:11 (GMT) |
commit | 57c0c15b5244320065374ad2c54f4fbec77a6428 (patch) | |
tree | 35369d817f5925aca09b083bba47c437b91386d9 | |
parent | cdd6c482c9ff9c55475ee7392ec8f672eddb7be6 (diff) | |
download | linux-57c0c15b5244320065374ad2c54f4fbec77a6428.tar.xz |
perf: Tidy up after the big rename
- provide compatibility Kconfig entry for existing PERF_COUNTERS .config's
- provide courtesy copy of old perf_counter.h, for user-space projects
- small indentation fixups
- fix up MAINTAINERS
- fix small x86 printout fallout
- fix up small PowerPC comment fallout (use 'counter' as in register)
Reviewed-by: Arjan van de Ven <arjan@linux.intel.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
LKML-Reference: <new-submission>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
-rw-r--r-- | MAINTAINERS | 2 | ||||
-rw-r--r-- | arch/powerpc/include/asm/paca.h | 2 | ||||
-rw-r--r-- | arch/powerpc/kernel/perf_event.c | 12 | ||||
-rw-r--r-- | arch/x86/kernel/cpu/perf_event.c | 14 | ||||
-rw-r--r-- | include/linux/perf_counter.h | 441 | ||||
-rw-r--r-- | include/linux/perf_event.h | 98 | ||||
-rw-r--r-- | init/Kconfig | 37 | ||||
-rw-r--r-- | kernel/perf_event.c | 4 |
8 files changed, 534 insertions, 76 deletions
diff --git a/MAINTAINERS b/MAINTAINERS index 43761a0..751a307 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4000,7 +4000,7 @@ S: Maintained F: include/linux/delayacct.h F: kernel/delayacct.c -PERFORMANCE COUNTER SUBSYSTEM +PERFORMANCE EVENTS SUBSYSTEM M: Peter Zijlstra <a.p.zijlstra@chello.nl> M: Paul Mackerras <paulus@samba.org> M: Ingo Molnar <mingo@elte.hu> diff --git a/arch/powerpc/include/asm/paca.h b/arch/powerpc/include/asm/paca.h index 154f405..7d8514c 100644 --- a/arch/powerpc/include/asm/paca.h +++ b/arch/powerpc/include/asm/paca.h @@ -122,7 +122,7 @@ struct paca_struct { u8 soft_enabled; /* irq soft-enable flag */ u8 hard_enabled; /* set if irqs are enabled in MSR */ u8 io_sync; /* writel() needs spin_unlock sync */ - u8 perf_event_pending; /* PM interrupt while soft-disabled */ + u8 perf_event_pending; /* PM interrupt while soft-disabled */ /* Stuff for accurate time accounting */ u64 user_time; /* accumulated usermode TB ticks */ diff --git a/arch/powerpc/kernel/perf_event.c b/arch/powerpc/kernel/perf_event.c index c98321f..197b7d9 100644 --- a/arch/powerpc/kernel/perf_event.c +++ b/arch/powerpc/kernel/perf_event.c @@ -41,7 +41,7 @@ DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events); struct power_pmu *ppmu; /* - * Normally, to ignore kernel events we set the FCS (freeze events + * Normally, to ignore kernel events we set the FCS (freeze counters * in supervisor mode) bit in MMCR0, but if the kernel runs with the * hypervisor bit set in the MSR, or if we are running on a processor * where the hypervisor bit is forced to 1 (as on Apple G5 processors), @@ -159,7 +159,7 @@ void perf_event_print_debug(void) } /* - * Read one performance monitor event (PMC). + * Read one performance monitor counter (PMC). */ static unsigned long read_pmc(int idx) { @@ -409,7 +409,7 @@ static void power_pmu_read(struct perf_event *event) val = read_pmc(event->hw.idx); } while (atomic64_cmpxchg(&event->hw.prev_count, prev, val) != prev); - /* The events are only 32 bits wide */ + /* The counters are only 32 bits wide */ delta = (val - prev) & 0xfffffffful; atomic64_add(delta, &event->count); atomic64_sub(delta, &event->hw.period_left); @@ -543,7 +543,7 @@ void hw_perf_disable(void) } /* - * Set the 'freeze events' bit. + * Set the 'freeze counters' bit. * The barrier is to make sure the mtspr has been * executed and the PMU has frozen the events * before we return. @@ -1124,7 +1124,7 @@ const struct pmu *hw_perf_event_init(struct perf_event *event) } /* - * A event has overflowed; update its count and record + * A counter has overflowed; update its count and record * things if requested. Note that interrupts are hard-disabled * here so there is no possibility of being interrupted. */ @@ -1271,7 +1271,7 @@ static void perf_event_interrupt(struct pt_regs *regs) /* * Reset MMCR0 to its normal value. This will set PMXE and - * clear FC (freeze events) and PMAO (perf mon alert occurred) + * clear FC (freeze counters) and PMAO (perf mon alert occurred) * and thus allow interrupts to occur again. * XXX might want to use MSR.PM to keep the events frozen until * we get back out of this interrupt. diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 0d03629..a3c7adb 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -2081,13 +2081,13 @@ void __init init_hw_perf_events(void) perf_events_lapic_init(); register_die_notifier(&perf_event_nmi_notifier); - pr_info("... version: %d\n", x86_pmu.version); - pr_info("... bit width: %d\n", x86_pmu.event_bits); - pr_info("... generic events: %d\n", x86_pmu.num_events); - pr_info("... value mask: %016Lx\n", x86_pmu.event_mask); - pr_info("... max period: %016Lx\n", x86_pmu.max_period); - pr_info("... fixed-purpose events: %d\n", x86_pmu.num_events_fixed); - pr_info("... event mask: %016Lx\n", perf_event_mask); + 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); + pr_info("... value mask: %016Lx\n", x86_pmu.event_mask); + pr_info("... max period: %016Lx\n", x86_pmu.max_period); + pr_info("... fixed-purpose events: %d\n", x86_pmu.num_events_fixed); + pr_info("... event mask: %016Lx\n", perf_event_mask); } static inline void x86_pmu_read(struct perf_event *event) diff --git a/include/linux/perf_counter.h b/include/linux/perf_counter.h new file mode 100644 index 0000000..368bd70 --- /dev/null +++ b/include/linux/perf_counter.h @@ -0,0 +1,441 @@ +/* + * NOTE: this file will be removed in a future kernel release, it is + * provided as a courtesy copy of user-space code that relies on the + * old (pre-rename) symbols and constants. + * + * Performance events: + * + * Copyright (C) 2008-2009, Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2008-2009, Red Hat, Inc., Ingo Molnar + * Copyright (C) 2008-2009, Red Hat, Inc., Peter Zijlstra + * + * Data type definitions, declarations, prototypes. + * + * Started by: Thomas Gleixner and Ingo Molnar + * + * For licencing details see kernel-base/COPYING + */ +#ifndef _LINUX_PERF_COUNTER_H +#define _LINUX_PERF_COUNTER_H + +#include <linux/types.h> +#include <linux/ioctl.h> +#include <asm/byteorder.h> + +/* + * User-space ABI bits: + */ + +/* + * attr.type + */ +enum perf_type_id { + PERF_TYPE_HARDWARE = 0, + PERF_TYPE_SOFTWARE = 1, + PERF_TYPE_TRACEPOINT = 2, + PERF_TYPE_HW_CACHE = 3, + PERF_TYPE_RAW = 4, + + PERF_TYPE_MAX, /* non-ABI */ +}; + +/* + * Generalized performance counter event types, used by the + * attr.event_id parameter of the sys_perf_counter_open() + * syscall: + */ +enum perf_hw_id { + /* + * Common hardware events, generalized by the kernel: + */ + PERF_COUNT_HW_CPU_CYCLES = 0, + PERF_COUNT_HW_INSTRUCTIONS = 1, + PERF_COUNT_HW_CACHE_REFERENCES = 2, + PERF_COUNT_HW_CACHE_MISSES = 3, + PERF_COUNT_HW_BRANCH_INSTRUCTIONS = 4, + PERF_COUNT_HW_BRANCH_MISSES = 5, + PERF_COUNT_HW_BUS_CYCLES = 6, + + PERF_COUNT_HW_MAX, /* non-ABI */ +}; + +/* + * Generalized hardware cache counters: + * + * { L1-D, L1-I, LLC, ITLB, DTLB, BPU } x + * { read, write, prefetch } x + * { accesses, misses } + */ +enum perf_hw_cache_id { + PERF_COUNT_HW_CACHE_L1D = 0, + PERF_COUNT_HW_CACHE_L1I = 1, + PERF_COUNT_HW_CACHE_LL = 2, + PERF_COUNT_HW_CACHE_DTLB = 3, + PERF_COUNT_HW_CACHE_ITLB = 4, + PERF_COUNT_HW_CACHE_BPU = 5, + + PERF_COUNT_HW_CACHE_MAX, /* non-ABI */ +}; + +enum perf_hw_cache_op_id { + PERF_COUNT_HW_CACHE_OP_READ = 0, + PERF_COUNT_HW_CACHE_OP_WRITE = 1, + PERF_COUNT_HW_CACHE_OP_PREFETCH = 2, + + PERF_COUNT_HW_CACHE_OP_MAX, /* non-ABI */ +}; + +enum perf_hw_cache_op_result_id { + PERF_COUNT_HW_CACHE_RESULT_ACCESS = 0, + PERF_COUNT_HW_CACHE_RESULT_MISS = 1, + + PERF_COUNT_HW_CACHE_RESULT_MAX, /* non-ABI */ +}; + +/* + * Special "software" counters provided by the kernel, even if the hardware + * does not support performance counters. These counters measure various + * physical and sw events of the kernel (and allow the profiling of them as + * well): + */ +enum perf_sw_ids { + PERF_COUNT_SW_CPU_CLOCK = 0, + PERF_COUNT_SW_TASK_CLOCK = 1, + PERF_COUNT_SW_PAGE_FAULTS = 2, + PERF_COUNT_SW_CONTEXT_SWITCHES = 3, + PERF_COUNT_SW_CPU_MIGRATIONS = 4, + PERF_COUNT_SW_PAGE_FAULTS_MIN = 5, + PERF_COUNT_SW_PAGE_FAULTS_MAJ = 6, + + PERF_COUNT_SW_MAX, /* non-ABI */ +}; + +/* + * Bits that can be set in attr.sample_type to request information + * in the overflow packets. + */ +enum perf_counter_sample_format { + PERF_SAMPLE_IP = 1U << 0, + PERF_SAMPLE_TID = 1U << 1, + PERF_SAMPLE_TIME = 1U << 2, + PERF_SAMPLE_ADDR = 1U << 3, + PERF_SAMPLE_READ = 1U << 4, + PERF_SAMPLE_CALLCHAIN = 1U << 5, + PERF_SAMPLE_ID = 1U << 6, + PERF_SAMPLE_CPU = 1U << 7, + PERF_SAMPLE_PERIOD = 1U << 8, + PERF_SAMPLE_STREAM_ID = 1U << 9, + PERF_SAMPLE_RAW = 1U << 10, + + PERF_SAMPLE_MAX = 1U << 11, /* non-ABI */ +}; + +/* + * The format of the data returned by read() on a perf counter fd, + * as specified by attr.read_format: + * + * struct read_format { + * { u64 value; + * { u64 time_enabled; } && PERF_FORMAT_ENABLED + * { u64 time_running; } && PERF_FORMAT_RUNNING + * { u64 id; } && PERF_FORMAT_ID + * } && !PERF_FORMAT_GROUP + * + * { u64 nr; + * { u64 time_enabled; } && PERF_FORMAT_ENABLED + * { u64 time_running; } && PERF_FORMAT_RUNNING + * { u64 value; + * { u64 id; } && PERF_FORMAT_ID + * } cntr[nr]; + * } && PERF_FORMAT_GROUP + * }; + */ +enum perf_counter_read_format { + PERF_FORMAT_TOTAL_TIME_ENABLED = 1U << 0, + PERF_FORMAT_TOTAL_TIME_RUNNING = 1U << 1, + PERF_FORMAT_ID = 1U << 2, + PERF_FORMAT_GROUP = 1U << 3, + + PERF_FORMAT_MAX = 1U << 4, /* non-ABI */ +}; + +#define PERF_ATTR_SIZE_VER0 64 /* sizeof first published struct */ + +/* + * Hardware event to monitor via a performance monitoring counter: + */ +struct perf_counter_attr { + + /* + * Major type: hardware/software/tracepoint/etc. + */ + __u32 type; + + /* + * Size of the attr structure, for fwd/bwd compat. + */ + __u32 size; + + /* + * Type specific configuration information. + */ + __u64 config; + + union { + __u64 sample_period; + __u64 sample_freq; + }; + + __u64 sample_type; + __u64 read_format; + + __u64 disabled : 1, /* off by default */ + inherit : 1, /* children inherit it */ + pinned : 1, /* must always be on PMU */ + exclusive : 1, /* only group on PMU */ + exclude_user : 1, /* don't count user */ + exclude_kernel : 1, /* ditto kernel */ + exclude_hv : 1, /* ditto hypervisor */ + exclude_idle : 1, /* don't count when idle */ + mmap : 1, /* include mmap data */ + comm : 1, /* include comm data */ + freq : 1, /* use freq, not period */ + inherit_stat : 1, /* per task counts */ + enable_on_exec : 1, /* next exec enables */ + task : 1, /* trace fork/exit */ + watermark : 1, /* wakeup_watermark */ + + __reserved_1 : 49; + + union { + __u32 wakeup_events; /* wakeup every n events */ + __u32 wakeup_watermark; /* bytes before wakeup */ + }; + __u32 __reserved_2; + + __u64 __reserved_3; +}; + +/* + * Ioctls that can be done on a perf counter fd: + */ +#define PERF_COUNTER_IOC_ENABLE _IO ('$', 0) +#define PERF_COUNTER_IOC_DISABLE _IO ('$', 1) +#define PERF_COUNTER_IOC_REFRESH _IO ('$', 2) +#define PERF_COUNTER_IOC_RESET _IO ('$', 3) +#define PERF_COUNTER_IOC_PERIOD _IOW('$', 4, u64) +#define PERF_COUNTER_IOC_SET_OUTPUT _IO ('$', 5) + +enum perf_counter_ioc_flags { + PERF_IOC_FLAG_GROUP = 1U << 0, +}; + +/* + * Structure of the page that can be mapped via mmap + */ +struct perf_counter_mmap_page { + __u32 version; /* version number of this structure */ + __u32 compat_version; /* lowest version this is compat with */ + + /* + * Bits needed to read the hw counters in user-space. + * + * u32 seq; + * s64 count; + * + * do { + * seq = pc->lock; + * + * barrier() + * if (pc->index) { + * count = pmc_read(pc->index - 1); + * count += pc->offset; + * } else + * goto regular_read; + * + * barrier(); + * } while (pc->lock != seq); + * + * NOTE: for obvious reason this only works on self-monitoring + * processes. + */ + __u32 lock; /* seqlock for synchronization */ + __u32 index; /* hardware counter identifier */ + __s64 offset; /* add to hardware counter value */ + __u64 time_enabled; /* time counter active */ + __u64 time_running; /* time counter on cpu */ + + /* + * Hole for extension of the self monitor capabilities + */ + + __u64 __reserved[123]; /* align to 1k */ + + /* + * Control data for the mmap() data buffer. + * + * User-space reading the @data_head value should issue an rmb(), on + * SMP capable platforms, after reading this value -- see + * perf_counter_wakeup(). + * + * When the mapping is PROT_WRITE the @data_tail value should be + * written by userspace to reflect the last read data. In this case + * the kernel will not over-write unread data. + */ + __u64 data_head; /* head in the data section */ + __u64 data_tail; /* user-space written tail */ +}; + +#define PERF_EVENT_MISC_CPUMODE_MASK (3 << 0) +#define PERF_EVENT_MISC_CPUMODE_UNKNOWN (0 << 0) +#define PERF_EVENT_MISC_KERNEL (1 << 0) +#define PERF_EVENT_MISC_USER (2 << 0) +#define PERF_EVENT_MISC_HYPERVISOR (3 << 0) + +struct perf_event_header { + __u32 type; + __u16 misc; + __u16 size; +}; + +enum perf_event_type { + + /* + * The MMAP events record the PROT_EXEC mappings so that we can + * correlate userspace IPs to code. They have the following structure: + * + * struct { + * struct perf_event_header header; + * + * u32 pid, tid; + * u64 addr; + * u64 len; + * u64 pgoff; + * char filename[]; + * }; + */ + PERF_EVENT_MMAP = 1, + + /* + * struct { + * struct perf_event_header header; + * u64 id; + * u64 lost; + * }; + */ + PERF_EVENT_LOST = 2, + + /* + * struct { + * struct perf_event_header header; + * + * u32 pid, tid; + * char comm[]; + * }; + */ + PERF_EVENT_COMM = 3, + + /* + * struct { + * struct perf_event_header header; + * u32 pid, ppid; + * u32 tid, ptid; + * u64 time; + * }; + */ + PERF_EVENT_EXIT = 4, + + /* + * struct { + * struct perf_event_header header; + * u64 time; + * u64 id; + * u64 stream_id; + * }; + */ + PERF_EVENT_THROTTLE = 5, + PERF_EVENT_UNTHROTTLE = 6, + + /* + * struct { + * struct perf_event_header header; + * u32 pid, ppid; + * u32 tid, ptid; + * { u64 time; } && PERF_SAMPLE_TIME + * }; + */ + PERF_EVENT_FORK = 7, + + /* + * struct { + * struct perf_event_header header; + * u32 pid, tid; + * + * struct read_format values; + * }; + */ + PERF_EVENT_READ = 8, + + /* + * struct { + * struct perf_event_header header; + * + * { u64 ip; } && PERF_SAMPLE_IP + * { u32 pid, tid; } && PERF_SAMPLE_TID + * { u64 time; } && PERF_SAMPLE_TIME + * { u64 addr; } && PERF_SAMPLE_ADDR + * { u64 id; } && PERF_SAMPLE_ID + * { u64 stream_id;} && PERF_SAMPLE_STREAM_ID + * { u32 cpu, res; } && PERF_SAMPLE_CPU + * { u64 period; } && PERF_SAMPLE_PERIOD + * + * { struct read_format values; } && PERF_SAMPLE_READ + * + * { u64 nr, + * u64 ips[nr]; } && PERF_SAMPLE_CALLCHAIN + * + * # + * # The RAW record below is opaque data wrt the ABI + * # + * # That is, the ABI doesn't make any promises wrt to + * # the stability of its content, it may vary depending + * # on event, hardware, kernel version and phase of + * # the moon. + * # + * # In other words, PERF_SAMPLE_RAW contents are not an ABI. + * # + * + * { u32 size; + * char data[size];}&& PERF_SAMPLE_RAW + * }; + */ + PERF_EVENT_SAMPLE = 9, + + PERF_EVENT_MAX, /* non-ABI */ +}; + +enum perf_callchain_context { + PERF_CONTEXT_HV = (__u64)-32, + PERF_CONTEXT_KERNEL = (__u64)-128, + PERF_CONTEXT_USER = (__u64)-512, + + PERF_CONTEXT_GUEST = (__u64)-2048, + PERF_CONTEXT_GUEST_KERNEL = (__u64)-2176, + PERF_CONTEXT_GUEST_USER = (__u64)-2560, + + PERF_CONTEXT_MAX = (__u64)-4095, +}; + +#define PERF_FLAG_FD_NO_GROUP (1U << 0) +#define PERF_FLAG_FD_OUTPUT (1U << 1) + +/* + * In case some app still references the old symbols: + */ + +#define __NR_perf_counter_open __NR_perf_event_open + +#define PR_TASK_PERF_COUNTERS_DISABLE PR_TASK_PERF_EVENTS_DISABLE +#define PR_TASK_PERF_COUNTERS_ENABLE PR_TASK_PERF_EVENTS_ENABLE + +#endif /* _LINUX_PERF_COUNTER_H */ diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h index ae9d9ed..acefaf7 100644 --- a/include/linux/perf_event.h +++ b/include/linux/perf_event.h @@ -1,15 +1,15 @@ /* - * Performance events: + * Performance events: * * Copyright (C) 2008-2009, Thomas Gleixner <tglx@linutronix.de> * Copyright (C) 2008-2009, Red Hat, Inc., Ingo Molnar * Copyright (C) 2008-2009, Red Hat, Inc., Peter Zijlstra * - * Data type definitions, declarations, prototypes. + * Data type definitions, declarations, prototypes. * * Started by: Thomas Gleixner and Ingo Molnar * - * For licencing details see kernel-base/COPYING + * For licencing details see kernel-base/COPYING */ #ifndef _LINUX_PERF_EVENT_H #define _LINUX_PERF_EVENT_H @@ -131,19 +131,19 @@ enum perf_event_sample_format { * as specified by attr.read_format: * * struct read_format { - * { u64 value; - * { u64 time_enabled; } && PERF_FORMAT_ENABLED - * { u64 time_running; } && PERF_FORMAT_RUNNING - * { u64 id; } && PERF_FORMAT_ID - * } && !PERF_FORMAT_GROUP + * { u64 value; + * { u64 time_enabled; } && PERF_FORMAT_ENABLED + * { u64 time_running; } && PERF_FORMAT_RUNNING + * { u64 id; } && PERF_FORMAT_ID + * } && !PERF_FORMAT_GROUP * - * { u64 nr; - * { u64 time_enabled; } && PERF_FORMAT_ENABLED - * { u64 time_running; } && PERF_FORMAT_RUNNING - * { u64 value; - * { u64 id; } && PERF_FORMAT_ID - * } cntr[nr]; - * } && PERF_FORMAT_GROUP + * { u64 nr; + * { u64 time_enabled; } && PERF_FORMAT_ENABLED + * { u64 time_running; } && PERF_FORMAT_RUNNING + * { u64 value; + * { u64 id; } && PERF_FORMAT_ID + * } cntr[nr]; + * } && PERF_FORMAT_GROUP * }; */ enum perf_event_read_format { @@ -152,7 +152,7 @@ enum perf_event_read_format { PERF_FORMAT_ID = 1U << 2, PERF_FORMAT_GROUP = 1U << 3, - PERF_FORMAT_MAX = 1U << 4, /* non-ABI */ + PERF_FORMAT_MAX = 1U << 4, /* non-ABI */ }; #define PERF_ATTR_SIZE_VER0 64 /* sizeof first published struct */ @@ -216,8 +216,8 @@ struct perf_event_attr { * Ioctls that can be done on a perf event fd: */ #define PERF_EVENT_IOC_ENABLE _IO ('$', 0) -#define PERF_EVENT_IOC_DISABLE _IO ('$', 1) -#define PERF_EVENT_IOC_REFRESH _IO ('$', 2) +#define PERF_EVENT_IOC_DISABLE _IO ('$', 1) +#define PERF_EVENT_IOC_REFRESH _IO ('$', 2) #define PERF_EVENT_IOC_RESET _IO ('$', 3) #define PERF_EVENT_IOC_PERIOD _IOW('$', 4, u64) #define PERF_EVENT_IOC_SET_OUTPUT _IO ('$', 5) @@ -314,9 +314,9 @@ enum perf_event_type { /* * struct { - * struct perf_event_header header; - * u64 id; - * u64 lost; + * struct perf_event_header header; + * u64 id; + * u64 lost; * }; */ PERF_RECORD_LOST = 2, @@ -383,23 +383,23 @@ enum perf_event_type { * { u64 id; } && PERF_SAMPLE_ID * { u64 stream_id;} && PERF_SAMPLE_STREAM_ID * { u32 cpu, res; } && PERF_SAMPLE_CPU - * { u64 period; } && PERF_SAMPLE_PERIOD + * { u64 period; } && PERF_SAMPLE_PERIOD * * { struct read_format values; } && PERF_SAMPLE_READ * * { u64 nr, * u64 ips[nr]; } && PERF_SAMPLE_CALLCHAIN * - * # - * # The RAW record below is opaque data wrt the ABI - * # - * # That is, the ABI doesn't make any promises wrt to - * # the stability of its content, it may vary depending - * # on event_id, hardware, kernel version and phase of - * # the moon. - * # - * # In other words, PERF_SAMPLE_RAW contents are not an ABI. - * # + * # + * # The RAW record below is opaque data wrt the ABI + * # + * # That is, the ABI doesn't make any promises wrt to + * # the stability of its content, it may vary depending + * # on event, hardware, kernel version and phase of + * # the moon. + * # + * # In other words, PERF_SAMPLE_RAW contents are not an ABI. + * # * * { u32 size; * char data[size];}&& PERF_SAMPLE_RAW @@ -503,10 +503,10 @@ struct pmu { * enum perf_event_active_state - the states of a event */ enum perf_event_active_state { - PERF_EVENT_STATE_ERROR = -2, + PERF_EVENT_STATE_ERROR = -2, PERF_EVENT_STATE_OFF = -1, PERF_EVENT_STATE_INACTIVE = 0, - PERF_EVENT_STATE_ACTIVE = 1, + PERF_EVENT_STATE_ACTIVE = 1, }; struct file; @@ -529,7 +529,7 @@ struct perf_mmap_data { long watermark; /* wakeup watermark */ - struct perf_event_mmap_page *user_page; + struct perf_event_mmap_page *user_page; void *data_pages[0]; }; @@ -694,14 +694,14 @@ struct perf_cpu_context { }; struct perf_output_handle { - struct perf_event *event; - struct perf_mmap_data *data; - unsigned long head; - unsigned long offset; - int nmi; - int sample; - int locked; - unsigned long flags; + struct perf_event *event; + struct perf_mmap_data *data; + unsigned long head; + unsigned long offset; + int nmi; + int sample; + int locked; + unsigned long flags; }; #ifdef CONFIG_PERF_EVENTS @@ -829,22 +829,22 @@ static inline void perf_event_task_sched_out(struct task_struct *task, struct task_struct *next, int cpu) { } static inline void -perf_event_task_tick(struct task_struct *task, int cpu) { } +perf_event_task_tick(struct task_struct *task, int cpu) { } static inline int perf_event_init_task(struct task_struct *child) { return 0; } static inline void perf_event_exit_task(struct task_struct *child) { } static inline void perf_event_free_task(struct task_struct *task) { } -static inline void perf_event_do_pending(void) { } -static inline void perf_event_print_debug(void) { } +static inline void perf_event_do_pending(void) { } +static inline void perf_event_print_debug(void) { } static inline void perf_disable(void) { } static inline void perf_enable(void) { } -static inline int perf_event_task_disable(void) { return -EINVAL; } -static inline int perf_event_task_enable(void) { return -EINVAL; } +static inline int perf_event_task_disable(void) { return -EINVAL; } +static inline int perf_event_task_enable(void) { return -EINVAL; } static inline void perf_sw_event(u32 event_id, u64 nr, int nmi, struct pt_regs *regs, u64 addr) { } -static inline void perf_event_mmap(struct vm_area_struct *vma) { } +static inline void perf_event_mmap(struct vm_area_struct *vma) { } static inline void perf_event_comm(struct task_struct *tsk) { } static inline void perf_event_fork(struct task_struct *tsk) { } static inline void perf_event_init(void) { } diff --git a/init/Kconfig b/init/Kconfig index cfdf5c3..706728b 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -920,26 +920,31 @@ config HAVE_PERF_EVENTS help See tools/perf/design.txt for details. -menu "Performance Counters" +menu "Kernel Performance Events And Counters" config PERF_EVENTS - bool "Kernel Performance Counters" - default y if PROFILING + bool "Kernel performance events and counters" + default y if (PROFILING || PERF_COUNTERS) depends on HAVE_PERF_EVENTS select ANON_INODES help - Enable kernel support for performance counter hardware. + Enable kernel support for various performance events provided + by software and hardware. - Performance counters are special hardware registers available - on most modern CPUs. These registers count the number of certain + Software events are supported either build-in or via the + use of generic tracepoints. + + Most modern CPUs support performance events via performance + counter registers. These registers count the number of certain types of hw events: such as instructions executed, cachemisses suffered, or branches mis-predicted - without slowing down the kernel or applications. These registers can also trigger interrupts when a threshold number of events have passed - and can thus be used to profile the code that runs on that CPU. - The Linux Performance Counter subsystem provides an abstraction of - these hardware capabilities, available via a system call. It + The Linux Performance Event subsystem provides an abstraction of + these software and hardware cevent apabilities, available via a + system call and used by the "perf" utility in tools/perf/. It provides per task and per CPU counters, and it provides event capabilities on top of those. @@ -950,14 +955,26 @@ config EVENT_PROFILE depends on PERF_EVENTS && EVENT_TRACING default y help - Allow the use of tracepoints as software performance counters. + Allow the use of tracepoints as software performance events. - When this is enabled, you can create perf counters based on + When this is enabled, you can create perf events based on tracepoints using PERF_TYPE_TRACEPOINT and the tracepoint ID found in debugfs://tracing/events/*/*/id. (The -e/--events option to the perf tool can parse and interpret symbolic tracepoints, in the subsystem:tracepoint_name format.) +config PERF_COUNTERS + bool "Kernel performance counters (old config option)" + depends on HAVE_PERF_EVENTS + help + This config has been obsoleted by the PERF_EVENTS + config option - please see that one for details. + + It has no effect on the kernel whether you enable + it or not, it is a compatibility placeholder. + + Say N if unsure. + endmenu config VM_EVENT_COUNTERS diff --git a/kernel/perf_event.c b/kernel/perf_event.c index 6e8b99a..76ac4db 100644 --- a/kernel/perf_event.c +++ b/kernel/perf_event.c @@ -1,12 +1,12 @@ /* - * Performance event core code + * Performance events core code: * * Copyright (C) 2008 Thomas Gleixner <tglx@linutronix.de> * Copyright (C) 2008-2009 Red Hat, Inc., Ingo Molnar * Copyright (C) 2008-2009 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com> * Copyright © 2009 Paul Mackerras, IBM Corp. <paulus@au1.ibm.com> * - * For licensing details see kernel-base/COPYING + * For licensing details see kernel-base/COPYING */ #include <linux/fs.h> |