diff --git a/Documentation/perf_counter/Makefile b/Documentation/perf_counter/Makefile new file mode 100644 index 000000000000..b45749753fcb --- /dev/null +++ b/Documentation/perf_counter/Makefile @@ -0,0 +1,12 @@ +BINS = kerneltop perfstat + +all: $(BINS) + +kerneltop: kerneltop.c perfcounters.h + cc -O6 -Wall -lrt `pkg-config --cflags --libs glib-2.0` -o $@ $< + +perfstat: kerneltop + ln -sf kerneltop perfstat + +clean: + rm $(BINS) diff --git a/Documentation/perf_counter/kerneltop.c b/Documentation/perf_counter/kerneltop.c new file mode 100644 index 000000000000..cf0e30bab5d5 --- /dev/null +++ b/Documentation/perf_counter/kerneltop.c @@ -0,0 +1,956 @@ +/* + * kerneltop.c: show top kernel functions - performance counters showcase + + Build with: + + cc -O6 -Wall `pkg-config --cflags --libs glib-2.0` -o kerneltop kerneltop.c + + Sample output: + +------------------------------------------------------------------------------ + KernelTop: 2669 irqs/sec [NMI, cache-misses/cache-refs], (all, cpu: 2) +------------------------------------------------------------------------------ + + weight RIP kernel function + ______ ________________ _______________ + + 35.20 - ffffffff804ce74b : skb_copy_and_csum_dev + 33.00 - ffffffff804cb740 : sock_alloc_send_skb + 31.26 - ffffffff804ce808 : skb_push + 22.43 - ffffffff80510004 : tcp_established_options + 19.00 - ffffffff8027d250 : find_get_page + 15.76 - ffffffff804e4fc9 : eth_type_trans + 15.20 - ffffffff804d8baa : dst_release + 14.86 - ffffffff804cf5d8 : skb_release_head_state + 14.00 - ffffffff802217d5 : read_hpet + 12.00 - ffffffff804ffb7f : __ip_local_out + 11.97 - ffffffff804fc0c8 : ip_local_deliver_finish + 8.54 - ffffffff805001a3 : ip_queue_xmit + + Started by Ingo Molnar + + Improvements and fixes by: + + Arjan van de Ven + Yanmin Zhang + Mike Galbraith + + Released under the GPL v2. (and only v2, not any later version) + + */ +#define _GNU_SOURCE +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#include +#include +#include +#include +#include +#include + +#include + +#ifdef __x86_64__ +# define __NR_perf_counter_open 295 +#endif + +#ifdef __i386__ +# define __NR_perf_counter_open 333 +#endif + +/* + * Pick up some kernel type conventions: + */ +#define __user +#define asmlinkage + +typedef unsigned int __u32; +typedef unsigned long long __u64; +typedef long long __s64; + +/* + * User-space ABI bits: + */ + +/* + * Generalized performance counter event types, used by the hw_event.type + * parameter of the sys_perf_counter_open() syscall: + */ +enum hw_event_types { + /* + * Common hardware events, generalized by the kernel: + */ + PERF_COUNT_CPU_CYCLES = 0, + PERF_COUNT_INSTRUCTIONS = 1, + PERF_COUNT_CACHE_REFERENCES = 2, + PERF_COUNT_CACHE_MISSES = 3, + PERF_COUNT_BRANCH_INSTRUCTIONS = 4, + PERF_COUNT_BRANCH_MISSES = 5, + PERF_COUNT_BUS_CYCLES = 6, + + PERF_HW_EVENTS_MAX = 7, + + /* + * 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): + */ + PERF_COUNT_CPU_CLOCK = -1, + PERF_COUNT_TASK_CLOCK = -2, + PERF_COUNT_PAGE_FAULTS = -3, + PERF_COUNT_CONTEXT_SWITCHES = -4, + PERF_COUNT_CPU_MIGRATIONS = -5, + + PERF_SW_EVENTS_MIN = -6, +}; + +/* + * IRQ-notification data record type: + */ +enum perf_counter_record_type { + PERF_RECORD_SIMPLE = 0, + PERF_RECORD_IRQ = 1, + PERF_RECORD_GROUP = 2, +}; + +/* + * Hardware event to monitor via a performance monitoring counter: + */ +struct perf_counter_hw_event { + __s64 type; + + __u64 irq_period; + __u64 record_type; + __u64 read_format; + + __u64 disabled : 1, /* off by default */ + nmi : 1, /* NMI sampling */ + raw : 1, /* raw event type */ + 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 */ + + __reserved_1 : 54; + + __u32 extra_config_len; + __u32 __reserved_4; + + __u64 __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) + +asmlinkage int sys_perf_counter_open( + + struct perf_counter_hw_event *hw_event_uptr __user, + pid_t pid, + int cpu, + int group_fd, + unsigned long flags) +{ + int ret; + + ret = syscall( + __NR_perf_counter_open, hw_event_uptr, pid, cpu, group_fd, flags); +#if defined(__x86_64__) || defined(__i386__) + if (ret < 0 && ret > -4096) { + errno = -ret; + ret = -1; + } +#endif + return ret; +} + +const char *event_types [] = { + "CPU cycles", + "instructions", + "cache-refs", + "cache-misses", + "branches", + "branch-misses", + "bus cycles" +}; + +const unsigned int default_count[] = { + 1000000, + 1000000, + 10000, + 10000, + 1000000, + 10000, +}; + +/* + * prctl(PR_TASK_PERF_COUNTERS_DISABLE) will (cheaply) disable all + * counters in the current task. + */ +#define PR_TASK_PERF_COUNTERS_DISABLE 31 +#define PR_TASK_PERF_COUNTERS_ENABLE 32 + +#define MAX_COUNTERS 8 + +static int nr_counters = -1; + +static __u64 count_filter = 100; + +#define MAX_NR_CPUS 256 + +static int event_count[MAX_COUNTERS]; +static unsigned long event_id[MAX_COUNTERS]; +static int event_raw[MAX_COUNTERS]; + +static int tid = -1; +static int profile_cpu = -1; +static int nr_cpus = 0; +static int nmi = 1; +static int group = 0; + +static char *vmlinux; + +static char *sym_filter; +static unsigned long filter_start; +static unsigned long filter_end; + +static int delay_secs = 2; +static int zero; +static int dump_symtab; + +struct source_line { + uint64_t EIP; + unsigned long count; + char *line; +}; + +static GList *lines; + +static void display_help(void) +{ + printf( + "Usage: kerneltop []\n\n" + "KernelTop Options (up to %d event types can be specified at once):\n\n", + MAX_COUNTERS); + printf( + " -e EID --event_id=EID # event type ID [default: 0]\n" + " 0: CPU cycles\n" + " 1: instructions\n" + " 2: cache accesses\n" + " 3: cache misses\n" + " 4: branch instructions\n" + " 5: branch prediction misses\n" + " 6: bus cycles\n\n" + " rNNN: raw PMU events (eventsel+umask)\n\n" + " -c CNT --count=CNT # event period to sample\n\n" + " -C CPU --cpu=CPU # CPU (-1 for all) [default: -1]\n" + " -p PID --pid=PID # PID of sampled task (-1 for all) [default: -1]\n\n" + " -d delay --delay= # sampling/display delay [default: 2]\n" + " -f CNT --filter=CNT # min-event-count filter [default: 100]\n\n" + " -s symbol --symbol= # function to be showed annotated one-shot\n" + " -x path --vmlinux= # the vmlinux binary, required for -s use:\n" + " -z --zero # zero counts after display\n" + " -D --dump_symtab # dump symbol table to stderr on startup\n" + "\n"); + + exit(0); +} + +static void process_options(int argc, char *argv[]) +{ + int error = 0, counter; + + for (;;) { + int option_index = 0; + /** Options for getopt */ + static struct option long_options[] = { + {"count", required_argument, NULL, 'c'}, + {"cpu", required_argument, NULL, 'C'}, + {"delay", required_argument, NULL, 'd'}, + {"dump_symtab", no_argument, NULL, 'D'}, + {"event_id", required_argument, NULL, 'e'}, + {"filter", required_argument, NULL, 'f'}, + {"group", required_argument, NULL, 'g'}, + {"help", no_argument, NULL, 'h'}, + {"nmi", required_argument, NULL, 'n'}, + {"pid", required_argument, NULL, 'p'}, + {"vmlinux", required_argument, NULL, 'x'}, + {"symbol", required_argument, NULL, 's'}, + {"zero", no_argument, NULL, 'z'}, + {NULL, 0, NULL, 0 } + }; + int c = getopt_long(argc, argv, "c:C:d:De:f:g:hn:p:s:x:z", + long_options, &option_index); + if (c == -1) + break; + + switch (c) { + case 'c': + if (nr_counters == -1) + nr_counters = 0; + event_count[nr_counters] = atoi(optarg); break; + case 'C': + /* CPU and PID are mutually exclusive */ + if (tid != -1) { + printf("WARNING: CPU switch overriding PID\n"); + sleep(1); + tid = -1; + } + profile_cpu = atoi(optarg); break; + case 'd': delay_secs = atoi(optarg); break; + case 'D': dump_symtab = 1; break; + + case 'e': + nr_counters++; + if (nr_counters == MAX_COUNTERS) { + error = 1; + break; + } + if (*optarg == 'r') { + event_raw[nr_counters] = 1; + ++optarg; + } + event_id[nr_counters] = strtol(optarg, NULL, 16); + break; + + case 'f': count_filter = atoi(optarg); break; + case 'g': group = atoi(optarg); break; + case 'h': display_help(); break; + case 'n': nmi = atoi(optarg); break; + case 'p': + /* CPU and PID are mutually exclusive */ + if (profile_cpu != -1) { + printf("WARNING: PID switch overriding CPU\n"); + sleep(1); + profile_cpu = -1; + } + tid = atoi(optarg); break; + case 's': sym_filter = strdup(optarg); break; + case 'x': vmlinux = strdup(optarg); break; + case 'z': zero = 1; break; + default: error = 1; break; + } + } + if (error) + display_help(); + + nr_counters++; + if (nr_counters < 1) + nr_counters = 1; + + for (counter = 0; counter < nr_counters; counter++) { + if (event_count[counter]) + continue; + + if (event_id[counter] < PERF_HW_EVENTS_MAX) + event_count[counter] = default_count[event_id[counter]]; + else + event_count[counter] = 100000; + } +} + +static uint64_t min_ip; +static uint64_t max_ip = -1ll; + +struct sym_entry { + unsigned long long addr; + char *sym; + unsigned long count[MAX_COUNTERS]; + int skip; + GList *source; +}; + +#define MAX_SYMS 100000 + +static int sym_table_count; + +struct sym_entry *sym_filter_entry; + +static struct sym_entry sym_table[MAX_SYMS]; + +static void show_details(struct sym_entry *sym); + +/* + * Ordering weight: count-1 * count-1 * ... / count-n + */ +static double sym_weight(const struct sym_entry *sym) +{ + double weight; + int counter; + + weight = sym->count[0]; + + for (counter = 1; counter < nr_counters-1; counter++) + weight *= sym->count[counter]; + + weight /= (sym->count[counter] + 1); + + return weight; +} + +static int compare(const void *__sym1, const void *__sym2) +{ + const struct sym_entry *sym1 = __sym1, *sym2 = __sym2; + + return sym_weight(sym1) < sym_weight(sym2); +} + +static time_t last_refresh; +static long events; +static long userspace_events; +static const char CONSOLE_CLEAR[] = ""; + +static struct sym_entry tmp[MAX_SYMS]; + +static void print_sym_table(void) +{ + int i, printed; + int counter; + float events_per_sec = events/delay_secs; + float kevents_per_sec = (events-userspace_events)/delay_secs; + + memcpy(tmp, sym_table, sizeof(sym_table[0])*sym_table_count); + qsort(tmp, sym_table_count, sizeof(tmp[0]), compare); + + write(1, CONSOLE_CLEAR, strlen(CONSOLE_CLEAR)); + + printf( +"------------------------------------------------------------------------------\n"); + printf( " KernelTop:%8.0f irqs/sec kernel:%3.1f%% [%s, ", + events_per_sec, + 100.0 - (100.0*((events_per_sec-kevents_per_sec)/events_per_sec)), + nmi ? "NMI" : "IRQ"); + + if (nr_counters == 1) + printf("%d ", event_count[0]); + + for (counter = 0; counter < nr_counters; counter++) { + if (counter) + printf("/"); + + if (event_id[counter] < PERF_HW_EVENTS_MAX) + printf( "%s", event_types[event_id[counter]]); + else + printf( "raw:%04lx", event_id[counter]); + } + + printf( "], "); + + if (tid != -1) + printf(" (tid: %d", tid); + else + printf(" (all"); + + if (profile_cpu != -1) + printf(", cpu: %d)\n", profile_cpu); + else { + if (tid != -1) + printf(")\n"); + else + printf(", %d CPUs)\n", nr_cpus); + } + + printf("------------------------------------------------------------------------------\n\n"); + + if (nr_counters == 1) + printf(" events"); + else + printf(" weight events"); + + printf(" RIP kernel function\n" + " ______ ______ ________________ _______________\n\n" + ); + + printed = 0; + for (i = 0; i < sym_table_count; i++) { + int count; + + if (nr_counters == 1) { + if (printed <= 18 && + tmp[i].count[0] >= count_filter) { + printf("%19.2f - %016llx : %s\n", + sym_weight(tmp + i), tmp[i].addr, tmp[i].sym); + printed++; + } + } else { + if (printed <= 18 && + tmp[i].count[0] >= count_filter) { + printf("%8.1f %10ld - %016llx : %s\n", + sym_weight(tmp + i), + tmp[i].count[0], + tmp[i].addr, tmp[i].sym); + printed++; + } + } + /* + * Add decay to the counts: + */ + for (count = 0; count < nr_counters; count++) + sym_table[i].count[count] = zero ? 0 : sym_table[i].count[count] * 7 / 8; + } + + if (sym_filter_entry) + show_details(sym_filter_entry); + + last_refresh = time(NULL); + + { + struct pollfd stdin_poll = { .fd = 0, .events = POLLIN }; + + if (poll(&stdin_poll, 1, 0) == 1) { + printf("key pressed - exiting.\n"); + exit(0); + } + } +} + +static int read_symbol(FILE *in, struct sym_entry *s) +{ + static int filter_match = 0; + char *sym, stype; + char str[500]; + int rc, pos; + + rc = fscanf(in, "%llx %c %499s", &s->addr, &stype, str); + if (rc == EOF) + return -1; + + assert(rc == 3); + + /* skip until end of line: */ + pos = strlen(str); + do { + rc = fgetc(in); + if (rc == '\n' || rc == EOF || pos >= 499) + break; + str[pos] = rc; + pos++; + } while (1); + str[pos] = 0; + + sym = str; + + /* Filter out known duplicates and non-text symbols. */ + if (!strcmp(sym, "_text")) + return 1; + if (!min_ip && !strcmp(sym, "_stext")) + return 1; + if (!strcmp(sym, "_etext") || !strcmp(sym, "_sinittext")) + return 1; + if (stype != 'T' && stype != 't') + return 1; + if (!strncmp("init_module", sym, 11) || !strncmp("cleanup_module", sym, 14)) + return 1; + if (strstr(sym, "_text_start") || strstr(sym, "_text_end")) + return 1; + + s->sym = malloc(strlen(str)); + assert(s->sym); + + strcpy((char *)s->sym, str); + s->skip = 0; + + /* Tag events to be skipped. */ + if (!strcmp("default_idle", s->sym) || !strcmp("cpu_idle", s->sym)) + s->skip = 1; + if (!strcmp("enter_idle", s->sym) || !strcmp("exit_idle", s->sym)) + s->skip = 1; + + if (filter_match == 1) { + filter_end = s->addr; + filter_match = -1; + if (filter_end - filter_start > 10000) { + printf("hm, too large filter symbol <%s> - skipping.\n", + sym_filter); + printf("symbol filter start: %016lx\n", filter_start); + printf(" end: %016lx\n", filter_end); + filter_end = filter_start = 0; + sym_filter = NULL; + sleep(1); + } + } + if (filter_match == 0 && sym_filter && !strcmp(s->sym, sym_filter)) { + filter_match = 1; + filter_start = s->addr; + } + + return 0; +} + +int compare_addr(const void *__sym1, const void *__sym2) +{ + const struct sym_entry *sym1 = __sym1, *sym2 = __sym2; + + return sym1->addr > sym2->addr; +} + +static void sort_symbol_table(void) +{ + int i, dups; + + do { + qsort(sym_table, sym_table_count, sizeof(sym_table[0]), compare_addr); + for (i = 0, dups = 0; i < sym_table_count; i++) { + if (sym_table[i].addr == sym_table[i+1].addr) { + sym_table[i+1].addr = -1ll; + dups++; + } + } + sym_table_count -= dups; + } while(dups); +} + +static void parse_symbols(void) +{ + struct sym_entry *last; + + FILE *kallsyms = fopen("/proc/kallsyms", "r"); + + if (!kallsyms) { + printf("Could not open /proc/kallsyms - no CONFIG_KALLSYMS_ALL=y?\n"); + exit(-1); + } + + while (!feof(kallsyms)) { + if (read_symbol(kallsyms, &sym_table[sym_table_count]) == 0) { + sym_table_count++; + assert(sym_table_count <= MAX_SYMS); + } + } + + sort_symbol_table(); + min_ip = sym_table[0].addr; + max_ip = sym_table[sym_table_count-1].addr; + last = sym_table + sym_table_count++; + + last->addr = -1ll; + last->sym = ""; + + if (filter_end) { + int count; + for (count=0; count < sym_table_count; count ++) { + if (!strcmp(sym_table[count].sym, sym_filter)) { + sym_filter_entry = &sym_table[count]; + break; + } + } + } + if (dump_symtab) { + int i; + + for (i = 0; i < sym_table_count; i++) + fprintf(stderr, "%llx %s\n", + sym_table[i].addr, sym_table[i].sym); + } +} + + +static void parse_vmlinux(char *filename) +{ + FILE *file; + char command[PATH_MAX*2]; + if (!filename) + return; + + sprintf(command, "objdump --start-address=0x%016lx --stop-address=0x%016lx -dS %s", filter_start, filter_end, filename); + + file = popen(command, "r"); + if (!file) + return; + + while (!feof(file)) { + struct source_line *src; + size_t dummy = 0; + char *c; + + src = malloc(sizeof(struct source_line)); + assert(src != NULL); + memset(src, 0, sizeof(struct source_line)); + + if (getline(&src->line, &dummy, file) < 0) + break; + if (!src->line) + break; + + c = strchr(src->line, '\n'); + if (c) + *c = 0; + + lines = g_list_prepend(lines, src); + + if (strlen(src->line)>8 && src->line[8] == ':') + src->EIP = strtoull(src->line, NULL, 16); + if (strlen(src->line)>8 && src->line[16] == ':') + src->EIP = strtoull(src->line, NULL, 16); + } + pclose(file); + lines = g_list_reverse(lines); +} + +static void record_precise_ip(uint64_t ip) +{ + struct source_line *line; + GList *item; + + item = g_list_first(lines); + while (item) { + line = item->data; + if (line->EIP == ip) + line->count++; + if (line->EIP > ip) + break; + item = g_list_next(item); + } +} + +static void lookup_sym_in_vmlinux(struct sym_entry *sym) +{ + struct source_line *line; + GList *item; + char pattern[PATH_MAX]; + sprintf(pattern, "<%s>:", sym->sym); + + item = g_list_first(lines); + while (item) { + line = item->data; + if (strstr(line->line, pattern)) { + sym->source = item; + break; + } + item = g_list_next(item); + } +} + +void show_lines(GList *item_queue, int item_queue_count) +{ + int i; + struct source_line *line; + + for (i = 0; i < item_queue_count; i++) { + line = item_queue->data; + printf("%8li\t%s\n", line->count, line->line); + item_queue = g_list_next(item_queue); + } +} + +#define TRACE_COUNT 3 + +static void show_details(struct sym_entry *sym) +{ + struct source_line *line; + GList *item; + int displayed = 0; + GList *item_queue = NULL; + int item_queue_count = 0; + + if (!sym->source) + lookup_sym_in_vmlinux(sym); + if (!sym->source) + return; + + printf("Showing details for %s\n", sym->sym); + + item = sym->source; + while (item) { + line = item->data; + if (displayed && strstr(line->line, ">:")) + break; + + if (!item_queue_count) + item_queue = item; + item_queue_count ++; + + if (line->count >= count_filter) { + show_lines(item_queue, item_queue_count); + item_queue_count = 0; + item_queue = NULL; + } else if (item_queue_count > TRACE_COUNT) { + item_queue = g_list_next(item_queue); + item_queue_count --; + } + + line->count = 0; + displayed++; + if (displayed > 300) + break; + item = g_list_next(item); + } +} + +/* + * Binary search in the histogram table and record the hit: + */ +static void record_ip(uint64_t ip, int counter) +{ + int left_idx, middle_idx, right_idx, idx; + unsigned long left, middle, right; + + record_precise_ip(ip); + + left_idx = 0; + right_idx = sym_table_count-1; + assert(ip <= max_ip && ip >= min_ip); + + while (left_idx + 1 < right_idx) { + middle_idx = (left_idx + right_idx) / 2; + + left = sym_table[ left_idx].addr; + middle = sym_table[middle_idx].addr; + right = sym_table[ right_idx].addr; + + if (!(left <= middle && middle <= right)) { + printf("%016lx...\n%016lx...\n%016lx\n", left, middle, right); + printf("%d %d %d\n", left_idx, middle_idx, right_idx); + } + assert(left <= middle && middle <= right); + if (!(left <= ip && ip <= right)) { + printf(" left: %016lx\n", left); + printf(" ip: %016lx\n", ip); + printf("right: %016lx\n", right); + } + assert(left <= ip && ip <= right); + /* + * [ left .... target .... middle .... right ] + * => right := middle + */ + if (ip < middle) { + right_idx = middle_idx; + continue; + } + /* + * [ left .... middle ... target ... right ] + * => left := middle + */ + left_idx = middle_idx; + } + + idx = left_idx; + + if (!sym_table[idx].skip) + sym_table[idx].count[counter]++; + else events--; +} + +static void process_event(uint64_t ip, int counter) +{ + events++; + + if (ip < min_ip || ip > max_ip) { + userspace_events++; + return; + } + + record_ip(ip, counter); +} + +int main(int argc, char *argv[]) +{ + struct pollfd event_array[MAX_NR_CPUS][MAX_COUNTERS]; + struct perf_counter_hw_event hw_event; + int fd[MAX_NR_CPUS][MAX_COUNTERS]; + int i, counter, group_fd; + unsigned int cpu; + uint64_t ip; + ssize_t res; + int ret; + + process_options(argc, argv); + + nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); + if (tid != -1 || profile_cpu != -1) + nr_cpus = 1; + + assert(nr_cpus <= MAX_NR_CPUS); + + for (i = 0; i < nr_cpus; i++) { + group_fd = -1; + for (counter = 0; counter < nr_counters; counter++) { + + cpu = profile_cpu; + if (tid == -1 && profile_cpu == -1) + cpu = i; + + memset(&hw_event, 0, sizeof(hw_event)); + hw_event.type = event_id[counter]; + hw_event.raw = event_raw[counter]; + hw_event.irq_period = event_count[counter]; + hw_event.record_type = PERF_RECORD_IRQ; + hw_event.nmi = nmi; + + fd[i][counter] = sys_perf_counter_open(&hw_event, tid, cpu, group_fd, 0); + fcntl(fd[i][counter], F_SETFL, O_NONBLOCK); + if (fd[i][counter] < 0) { + printf("kerneltop error: syscall returned with %d (%s)\n", + fd[i][counter], strerror(-fd[i][counter])); + if (fd[i][counter] == -1) + printf("Are you root?\n"); + exit(-1); + } + assert(fd[i][counter] >= 0); + + /* + * First counter acts as the group leader: + */ + if (group && group_fd == -1) + group_fd = fd[i][counter]; + + event_array[i][counter].fd = fd[i][counter]; + event_array[i][counter].events = POLLIN; + } + } + + parse_symbols(); + if (vmlinux && sym_filter_entry) + parse_vmlinux(vmlinux); + + printf("KernelTop refresh period: %d seconds\n", delay_secs); + last_refresh = time(NULL); + + while (1) { + int hits = events; + + for (i = 0; i < nr_cpus; i++) { + for (counter = 0; counter < nr_counters; counter++) { + res = read(fd[i][counter], (char *) &ip, sizeof(ip)); + if (res > 0) { + assert(res == sizeof(ip)); + + process_event(ip, counter); + } + } + } + + if (time(NULL) >= last_refresh + delay_secs) { + print_sym_table(); + events = userspace_events = 0; + } + + if (hits == events) + ret = poll(event_array[0], nr_cpus, 1000); + hits = events; + } + + return 0; +} diff --git a/Documentation/perf_counter/perfstat.c b/Documentation/perf_counter/perfstat.c new file mode 100644 index 000000000000..9a5808fbcf90 --- /dev/null +++ b/Documentation/perf_counter/perfstat.c @@ -0,0 +1,521 @@ +/* + * perfstat: /usr/bin/time -alike performance counter statistics utility + * + * It summarizes the counter events of all tasks (and child tasks), + * covering all CPUs that the command (or workload) executes on. + * It only counts the per-task events of the workload started, + * independent of how many other tasks run on those CPUs. + * + * Build with: cc -O2 -g -lrt -Wall -W -o perfstat perfstat.c + * + * Sample output: + * + + $ ./perfstat -e 1 -e 3 -e 5 ls -lR /usr/include/ >/dev/null + + Performance counter stats for 'ls': + + 163516953 instructions + 2295 cache-misses + 2855182 branch-misses + + * + * Copyright (C) 2008, Red Hat Inc, Ingo Molnar + * + * Released under the GPLv2 (not later). + * + * Percpu counter support by: Yanmin Zhang + * Symbolic event options by: Wu Fengguang + */ +#define _GNU_SOURCE + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include + +#include + +#ifdef __x86_64__ +# define __NR_perf_counter_open 295 +#endif + +#ifdef __i386__ +# define __NR_perf_counter_open 333 +#endif + +#ifdef __powerpc__ +#define __NR_perf_counter_open 319 +#endif + +/* + * Pick up some kernel type conventions: + */ +#define __user +#define asmlinkage + +typedef unsigned int __u32; +typedef unsigned long long __u64; +typedef long long __s64; + +#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) + +/* + * User-space ABI bits: + */ + +/* + * Generalized performance counter event types, used by the hw_event.type + * parameter of the sys_perf_counter_open() syscall: + */ +enum hw_event_types { + /* + * Common hardware events, generalized by the kernel: + */ + PERF_COUNT_CPU_CYCLES = 0, + PERF_COUNT_INSTRUCTIONS = 1, + PERF_COUNT_CACHE_REFERENCES = 2, + PERF_COUNT_CACHE_MISSES = 3, + PERF_COUNT_BRANCH_INSTRUCTIONS = 4, + PERF_COUNT_BRANCH_MISSES = 5, + PERF_COUNT_BUS_CYCLES = 6, + + PERF_HW_EVENTS_MAX = 7, + + /* + * 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): + */ + PERF_COUNT_CPU_CLOCK = -1, + PERF_COUNT_TASK_CLOCK = -2, + PERF_COUNT_PAGE_FAULTS = -3, + PERF_COUNT_CONTEXT_SWITCHES = -4, + PERF_COUNT_CPU_MIGRATIONS = -5, + + PERF_SW_EVENTS_MIN = -6, +}; + +/* + * IRQ-notification data record type: + */ +enum perf_counter_record_type { + PERF_RECORD_SIMPLE = 0, + PERF_RECORD_IRQ = 1, + PERF_RECORD_GROUP = 2, +}; + +/* + * Hardware event to monitor via a performance monitoring counter: + */ +struct perf_counter_hw_event { + __s64 type; + + __u64 irq_period; + __u64 record_type; + __u64 read_format; + + __u64 disabled : 1, /* off by default */ + nmi : 1, /* NMI sampling */ + raw : 1, /* raw event type */ + 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 */ + + __reserved_1 : 54; + + __u32 extra_config_len; + __u32 __reserved_4; + + __u64 __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) + +asmlinkage int sys_perf_counter_open( + + struct perf_counter_hw_event *hw_event_uptr __user, + pid_t pid, + int cpu, + int group_fd, + unsigned long flags) +{ + int ret; + + ret = syscall( + __NR_perf_counter_open, hw_event_uptr, pid, cpu, group_fd, flags); +#if defined(__x86_64__) || defined(__i386__) + if (ret < 0 && ret > -4096) { + errno = -ret; + ret = -1; + } +#endif + return ret; +} + + +static char *hw_event_names [] = { + "CPU cycles", + "instructions", + "cache references", + "cache misses", + "branches", + "branch misses", + "bus cycles", +}; + +static char *sw_event_names [] = { + "cpu clock ticks", + "task clock ticks", + "pagefaults", + "context switches", + "CPU migrations", +}; + +struct event_symbol { + int event; + char *symbol; +}; + +static struct event_symbol event_symbols [] = { + {PERF_COUNT_CPU_CYCLES, "cpu-cycles", }, + {PERF_COUNT_CPU_CYCLES, "cycles", }, + {PERF_COUNT_INSTRUCTIONS, "instructions", }, + {PERF_COUNT_CACHE_REFERENCES, "cache-references", }, + {PERF_COUNT_CACHE_MISSES, "cache-misses", }, + {PERF_COUNT_BRANCH_INSTRUCTIONS, "branch-instructions", }, + {PERF_COUNT_BRANCH_INSTRUCTIONS, "branches", }, + {PERF_COUNT_BRANCH_MISSES, "branch-misses", }, + {PERF_COUNT_BUS_CYCLES, "bus-cycles", }, + {PERF_COUNT_CPU_CLOCK, "cpu-ticks", }, + {PERF_COUNT_CPU_CLOCK, "ticks", }, + {PERF_COUNT_TASK_CLOCK, "task-ticks", }, + {PERF_COUNT_PAGE_FAULTS, "page-faults", }, + {PERF_COUNT_PAGE_FAULTS, "faults", }, + {PERF_COUNT_CONTEXT_SWITCHES, "context-switches", }, + {PERF_COUNT_CONTEXT_SWITCHES, "cs", }, + {PERF_COUNT_CPU_MIGRATIONS, "cpu-migrations", }, + {PERF_COUNT_CPU_MIGRATIONS, "migrations", }, +}; + +#define MAX_COUNTERS 64 +#define MAX_NR_CPUS 256 + +static int nr_counters = 0; +static int nr_cpus = 0; + +static int event_id[MAX_COUNTERS] = + { -2, -5, -4, -3, 0, 1, 2, 3}; + +static int event_raw[MAX_COUNTERS]; + +static int system_wide = 0; + +static void display_help(void) +{ + unsigned int i; + int e; + + printf( + "Usage: perfstat [] \n\n" + "PerfStat Options (up to %d event types can be specified):\n\n", + MAX_COUNTERS); + printf( + " -e EVENT --event=EVENT # symbolic-name abbreviations"); + + for (i = 0, e = PERF_HW_EVENTS_MAX; i < ARRAY_SIZE(event_symbols); i++) { + if (e != event_symbols[i].event) { + e = event_symbols[i].event; + printf( + "\n %2d: %-20s", e, event_symbols[i].symbol); + } else + printf(" %s", event_symbols[i].symbol); + } + + printf("\n" + " rNNN: raw event type\n\n" + " -s # system-wide collection\n\n" + " -c --command= # command+arguments to be timed.\n" + "\n"); + exit(0); +} + +static int type_valid(int type) +{ + if (type >= PERF_HW_EVENTS_MAX) + return 0; + if (type <= PERF_SW_EVENTS_MIN) + return 0; + + return 1; +} + +static char *event_name(int ctr) +{ + int type = event_id[ctr]; + static char buf[32]; + + if (event_raw[ctr]) { + sprintf(buf, "raw 0x%x", type); + return buf; + } + if (!type_valid(type)) + return "unknown"; + + if (type >= 0) + return hw_event_names[type]; + + return sw_event_names[-type-1]; +} + +/* + * Each event can have multiple symbolic names. + * Symbolic names are (almost) exactly matched. + */ +static int match_event_symbols(char *str) +{ + unsigned int i; + + if (isdigit(str[0]) || str[0] == '-') + return atoi(str); + + for (i = 0; i < ARRAY_SIZE(event_symbols); i++) { + if (!strncmp(str, event_symbols[i].symbol, + strlen(event_symbols[i].symbol))) + return event_symbols[i].event; + } + + return PERF_HW_EVENTS_MAX; +} + +static void parse_events(char *str) +{ + int type, raw; + +again: + nr_counters++; + if (nr_counters == MAX_COUNTERS) + display_help(); + + raw = 0; + if (*str == 'r') { + raw = 1; + ++str; + type = strtol(str, NULL, 16); + } else { + type = match_event_symbols(str); + if (!type_valid(type)) + display_help(); + } + + event_id[nr_counters] = type; + event_raw[nr_counters] = raw; + + str = strstr(str, ","); + if (str) { + str++; + goto again; + } +} + +static void process_options(int argc, char *argv[]) +{ + for (;;) { + int option_index = 0; + /** Options for getopt */ + static struct option long_options[] = { + {"event", required_argument, NULL, 'e'}, + {"help", no_argument, NULL, 'h'}, + {"command", no_argument, NULL, 'c'}, + {NULL, 0, NULL, 0 } + }; + int c = getopt_long(argc, argv, "+:e:c:s", + long_options, &option_index); + if (c == -1) + break; + + switch (c) { + case 'c': + break; + case 's': + system_wide = 1; + break; + case 'e': + parse_events(optarg); + break; + default: + break; + } + } + if (optind == argc) + goto err; + + if (!nr_counters) + nr_counters = 8; + else + nr_counters++; + return; + +err: + display_help(); +} + +char fault_here[1000000]; + +#define PR_TASK_PERF_COUNTERS_DISABLE 31 +#define PR_TASK_PERF_COUNTERS_ENABLE 32 + +static int fd[MAX_NR_CPUS][MAX_COUNTERS]; + +static void create_counter(int counter) +{ + struct perf_counter_hw_event hw_event; + + memset(&hw_event, 0, sizeof(hw_event)); + hw_event.type = event_id[counter]; + hw_event.raw = event_raw[counter]; + hw_event.record_type = PERF_RECORD_SIMPLE; + hw_event.nmi = 0; + + if (system_wide) { + int cpu; + for (cpu = 0; cpu < nr_cpus; cpu ++) { + fd[cpu][counter] = sys_perf_counter_open(&hw_event, -1, cpu, -1, 0); + if (fd[cpu][counter] < 0) { + printf("perfstat error: syscall returned with %d (%s)\n", + fd[cpu][counter], strerror(errno)); + exit(-1); + } + + } + } else { + hw_event.inherit = 1; + hw_event.disabled = 1; + + fd[0][counter] = sys_perf_counter_open(&hw_event, 0, -1, -1, 0); + if (fd[0][counter] < 0) { + printf("perfstat error: syscall returned with %d (%s)\n", + fd[0][counter], strerror(errno)); + exit(-1); + } + } +} + + +#define rdclock() \ +({ \ + struct timespec ts; \ + \ + clock_gettime(CLOCK_MONOTONIC, &ts); \ + ts.tv_sec * 1000000000ULL + ts.tv_nsec; \ +}) + +int main(int argc, char *argv[]) +{ + unsigned long long t0, t1; + int counter; + ssize_t res; + int status; + int pid; + + process_options(argc, argv); + + if (system_wide) { + nr_cpus = sysconf(_SC_NPROCESSORS_ONLN); + assert(nr_cpus <= MAX_NR_CPUS); + assert(nr_cpus >= 0); + } else + nr_cpus = 1; + + for (counter = 0; counter < nr_counters; counter++) + create_counter(counter); + + argc -= optind; + argv += optind; + + /* + * Enable counters and exec the command: + */ + t0 = rdclock(); + prctl(PR_TASK_PERF_COUNTERS_ENABLE); + + if ((pid = fork()) < 0) + perror("failed to fork"); + if (!pid) { + if (execvp(argv[0], argv)) { + perror(argv[0]); + exit(-1); + } + } + while (wait(&status) >= 0) + ; + prctl(PR_TASK_PERF_COUNTERS_DISABLE); + t1 = rdclock(); + + fflush(stdout); + + fprintf(stderr, "\n"); + fprintf(stderr, " Performance counter stats for \'%s\':\n", + argv[0]); + fprintf(stderr, "\n"); + + for (counter = 0; counter < nr_counters; counter++) { + int cpu; + __u64 count, single_count; + + count = 0; + for (cpu = 0; cpu < nr_cpus; cpu ++) { + res = read(fd[cpu][counter], + (char *) &single_count, sizeof(single_count)); + assert(res == sizeof(single_count)); + count += single_count; + } + + if (!event_raw[counter] && + (event_id[counter] == PERF_COUNT_CPU_CLOCK || + event_id[counter] == PERF_COUNT_TASK_CLOCK)) { + + double msecs = (double)count / 1000000; + + fprintf(stderr, " %14.6f %-20s (msecs)\n", + msecs, event_name(counter)); + } else { + fprintf(stderr, " %14Ld %-20s (events)\n", + count, event_name(counter)); + } + if (!counter) + fprintf(stderr, "\n"); + } + fprintf(stderr, "\n"); + fprintf(stderr, " Wall-clock time elapsed: %12.6f msecs\n", + (double)(t1-t0)/1e6); + fprintf(stderr, "\n"); + + return 0; +}