From 2b7fcbc5a9c719a306af1c4986a9f5c2cbfcec65 Mon Sep 17 00:00:00 2001 From: Arnaldo Carvalho de Melo Date: Tue, 11 Sep 2012 19:29:17 -0300 Subject: [PATCH] perf sched: Use perf_evsel__{int,str}val This patch also stops reading the common fields, as they were not being used except for one ->common_pid case that was replaced by sample->tid, i.e. the info is already in the perf_sample struct. Also it only fills the _event structures when there is a handler. [root@sandy ~]# perf sched record sleep 30s [ perf record: Woken up 1 times to write data ] [ perf record: Captured and wrote 8.585 MB perf.data (~375063 samples) ] Before: [root@sandy ~]# perf stat -r 10 perf sched lat > /dev/null Performance counter stats for 'perf sched lat' (10 runs): 129.117838 task-clock # 0.994 CPUs utilized ( +- 0.28% ) 14 context-switches # 0.111 K/sec ( +- 2.10% ) 0 cpu-migrations # 0.002 K/sec ( +- 66.67% ) 7,654 page-faults # 0.059 M/sec ( +- 0.67% ) 438,121,661 cycles # 3.393 GHz ( +- 0.06% ) [83.06%] 150,808,605 stalled-cycles-frontend # 34.42% frontend cycles idle ( +- 0.14% ) [83.10%] 80,748,941 stalled-cycles-backend # 18.43% backend cycles idle ( +- 0.64% ) [66.73%] 758,605,879 instructions # 1.73 insns per cycle # 0.20 stalled cycles per insn ( +- 0.08% ) [83.54%] 162,164,321 branches # 1255.940 M/sec ( +- 0.10% ) [83.70%] 1,609,903 branch-misses # 0.99% of all branches ( +- 0.08% ) [83.62%] 0.129949153 seconds time elapsed ( +- 0.28% ) After: [root@sandy ~]# perf stat -r 10 perf sched lat > /dev/null Performance counter stats for 'perf sched lat' (10 runs): 103.592215 task-clock # 0.993 CPUs utilized ( +- 0.33% ) 12 context-switches # 0.114 K/sec ( +- 3.29% ) 0 cpu-migrations # 0.000 K/sec 7,605 page-faults # 0.073 M/sec ( +- 0.00% ) 345,796,112 cycles # 3.338 GHz ( +- 0.07% ) [82.90%] 106,876,796 stalled-cycles-frontend # 30.91% frontend cycles idle ( +- 0.38% ) [83.23%] 62,060,877 stalled-cycles-backend # 17.95% backend cycles idle ( +- 0.80% ) [67.14%] 628,246,586 instructions # 1.82 insns per cycle # 0.17 stalled cycles per insn ( +- 0.04% ) [83.64%] 134,962,057 branches # 1302.820 M/sec ( +- 0.10% ) [83.64%] 1,233,037 branch-misses # 0.91% of all branches ( +- 0.29% ) [83.41%] 0.104333272 seconds time elapsed ( +- 0.33% ) [root@sandy ~]# Cc: David Ahern Cc: Frederic Weisbecker Cc: Jiri Olsa Cc: Mike Galbraith Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/n/tip-weu9t63zkrfrazkn0gxj48xy@git.kernel.org Signed-off-by: Arnaldo Carvalho de Melo --- tools/perf/builtin-sched.c | 249 ++++++++++++++----------------------- 1 file changed, 90 insertions(+), 159 deletions(-) diff --git a/tools/perf/builtin-sched.c b/tools/perf/builtin-sched.c index 79f88fa3f7a..0df5e7a08c6 100644 --- a/tools/perf/builtin-sched.c +++ b/tools/perf/builtin-sched.c @@ -98,82 +98,40 @@ struct work_atoms { typedef int (*sort_fn_t)(struct work_atoms *, struct work_atoms *); struct trace_switch_event { - u32 size; - - u16 common_type; - u8 common_flags; - u8 common_preempt_count; - u32 common_pid; - u32 common_tgid; - - char prev_comm[16]; + char *prev_comm; u32 prev_pid; u32 prev_prio; u64 prev_state; - char next_comm[16]; + char *next_comm; u32 next_pid; u32 next_prio; }; struct trace_runtime_event { - u32 size; - - u16 common_type; - u8 common_flags; - u8 common_preempt_count; - u32 common_pid; - u32 common_tgid; - - char comm[16]; + char *comm; u32 pid; u64 runtime; u64 vruntime; }; struct trace_wakeup_event { - u32 size; - - u16 common_type; - u8 common_flags; - u8 common_preempt_count; - u32 common_pid; - u32 common_tgid; - - char comm[16]; + char *comm; u32 pid; - u32 prio; u32 success; u32 cpu; }; struct trace_fork_event { - u32 size; - - u16 common_type; - u8 common_flags; - u8 common_preempt_count; - u32 common_pid; - u32 common_tgid; - - char parent_comm[16]; + char *parent_comm; u32 parent_pid; - char child_comm[16]; + char *child_comm; u32 child_pid; }; struct trace_migrate_task_event { - u32 size; - - u16 common_type; - u8 common_flags; - u8 common_preempt_count; - u32 common_pid; - u32 common_tgid; - - char comm[16]; + char *comm; u32 pid; - u32 prio; u32 cpu; }; @@ -184,7 +142,7 @@ struct trace_sched_handler { int (*switch_event)(struct perf_sched *sched, struct trace_switch_event *event, struct machine *machine, - struct event_format *tp_format, + struct perf_evsel *evsel, struct perf_sample *sample); int (*runtime_event)(struct perf_sched *sched, @@ -195,12 +153,12 @@ struct trace_sched_handler { int (*wakeup_event)(struct perf_sched *sched, struct trace_wakeup_event *event, struct machine *machine, - struct event_format *tp_format, + struct perf_evsel *evsel, struct perf_sample *sample); int (*fork_event)(struct perf_sched *sched, struct trace_fork_event *event, - struct event_format *tp_format); + struct perf_evsel *evsel); int (*migrate_task_event)(struct perf_sched *sched, struct trace_migrate_task_event *event, @@ -740,40 +698,22 @@ static void test_calibrations(struct perf_sched *sched) printf("the sleep test took %" PRIu64 " nsecs\n", T1 - T0); } -#define FILL_FIELD(ptr, field, event, data) \ - ptr.field = (typeof(ptr.field)) raw_field_value(event, #field, data) - -#define FILL_ARRAY(ptr, array, event, data) \ -do { \ - void *__array = raw_field_ptr(event, #array, data); \ - memcpy(ptr.array, __array, sizeof(ptr.array)); \ -} while(0) - -#define FILL_COMMON_FIELDS(ptr, event, data) \ -do { \ - FILL_FIELD(ptr, common_type, event, data); \ - FILL_FIELD(ptr, common_flags, event, data); \ - FILL_FIELD(ptr, common_preempt_count, event, data); \ - FILL_FIELD(ptr, common_pid, event, data); \ - FILL_FIELD(ptr, common_tgid, event, data); \ -} while (0) - static int replay_wakeup_event(struct perf_sched *sched, struct trace_wakeup_event *wakeup_event, struct machine *machine __maybe_unused, - struct event_format *event, struct perf_sample *sample) + struct perf_evsel *evsel, struct perf_sample *sample) { struct task_desc *waker, *wakee; if (verbose) { - printf("sched_wakeup event %p\n", event); + printf("sched_wakeup event %p\n", evsel); printf(" ... pid %d woke up %s/%d\n", - wakeup_event->common_pid, wakeup_event->comm, wakeup_event->pid); + sample->tid, wakeup_event->comm, wakeup_event->pid); } - waker = register_pid(sched, wakeup_event->common_pid, ""); + waker = register_pid(sched, sample->tid, ""); wakee = register_pid(sched, wakeup_event->pid, wakeup_event->comm); add_sched_event_wakeup(sched, waker, sample->time, wakee); @@ -784,7 +724,7 @@ static int replay_switch_event(struct perf_sched *sched, struct trace_switch_event *switch_event, struct machine *machine __maybe_unused, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample) { struct task_desc *prev, __maybe_unused *next; @@ -793,7 +733,7 @@ replay_switch_event(struct perf_sched *sched, s64 delta; if (verbose) - printf("sched_switch event %p\n", event); + printf("sched_switch event %p\n", evsel); if (cpu >= MAX_CPUS || cpu < 0) return 0; @@ -829,10 +769,10 @@ replay_switch_event(struct perf_sched *sched, static int replay_fork_event(struct perf_sched *sched, struct trace_fork_event *fork_event, - struct event_format *event) + struct perf_evsel *evsel) { if (verbose) { - printf("sched_fork event %p\n", event); + printf("sched_fork event %p\n", evsel); printf("... parent: %s/%d\n", fork_event->parent_comm, fork_event->parent_pid); printf("... child: %s/%d\n", fork_event->child_comm, fork_event->child_pid); } @@ -931,7 +871,7 @@ static int thread_atoms_insert(struct perf_sched *sched, struct thread *thread) static int latency_fork_event(struct perf_sched *sched __maybe_unused, struct trace_fork_event *fork_event __maybe_unused, - struct event_format *event __maybe_unused) + struct perf_evsel *evsel __maybe_unused) { /* should insert the newcomer */ return 0; @@ -1015,7 +955,7 @@ static int latency_switch_event(struct perf_sched *sched, struct trace_switch_event *switch_event, struct machine *machine, - struct event_format *event __maybe_unused, + struct perf_evsel *evsel __maybe_unused, struct perf_sample *sample) { struct work_atoms *out_events, *in_events; @@ -1106,7 +1046,7 @@ static int latency_wakeup_event(struct perf_sched *sched, struct trace_wakeup_event *wakeup_event, struct machine *machine, - struct event_format *event __maybe_unused, + struct perf_evsel *evsel __maybe_unused, struct perf_sample *sample) { struct work_atoms *atoms; @@ -1350,34 +1290,32 @@ static void perf_sched__sort_lat(struct perf_sched *sched) } static int process_sched_wakeup_event(struct perf_tool *tool, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine) { struct perf_sched *sched = container_of(tool, struct perf_sched, tool); - void *data = sample->raw_data; - struct trace_wakeup_event wakeup_event; - int err = 0; - - FILL_COMMON_FIELDS(wakeup_event, event, data); - FILL_ARRAY(wakeup_event, comm, event, data); - FILL_FIELD(wakeup_event, pid, event, data); - FILL_FIELD(wakeup_event, prio, event, data); - FILL_FIELD(wakeup_event, success, event, data); - FILL_FIELD(wakeup_event, cpu, event, data); + if (sched->tp_handler->wakeup_event) { + struct trace_wakeup_event event = { + .comm = perf_evsel__strval(evsel, sample, "comm"), + .pid = perf_evsel__intval(evsel, sample, "pid"), + .prio = perf_evsel__intval(evsel, sample, "prio"), + .success = perf_evsel__intval(evsel, sample, "success"), + .cpu = perf_evsel__intval(evsel, sample, "cpu"), + }; - if (sched->tp_handler->wakeup_event) - err = sched->tp_handler->wakeup_event(sched, &wakeup_event, machine, event, sample); + return sched->tp_handler->wakeup_event(sched, &event, machine, evsel, sample); + } - return err; + return 0; } static int map_switch_event(struct perf_sched *sched, struct trace_switch_event *switch_event, struct machine *machine, - struct event_format *event __maybe_unused, + struct perf_evsel *evsel __maybe_unused, struct perf_sample *sample) { struct thread *sched_out __maybe_unused, *sched_in; @@ -1455,120 +1393,113 @@ map_switch_event(struct perf_sched *sched, } static int process_sched_switch_event(struct perf_tool *tool, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine) { struct perf_sched *sched = container_of(tool, struct perf_sched, tool); int this_cpu = sample->cpu, err = 0; - void *data = sample->raw_data; - struct trace_switch_event switch_event; - - FILL_COMMON_FIELDS(switch_event, event, data); - - FILL_ARRAY(switch_event, prev_comm, event, data); - FILL_FIELD(switch_event, prev_pid, event, data); - FILL_FIELD(switch_event, prev_prio, event, data); - FILL_FIELD(switch_event, prev_state, event, data); - FILL_ARRAY(switch_event, next_comm, event, data); - FILL_FIELD(switch_event, next_pid, event, data); - FILL_FIELD(switch_event, next_prio, event, data); + u32 prev_pid = perf_evsel__intval(evsel, sample, "prev_pid"), + next_pid = perf_evsel__intval(evsel, sample, "next_pid"); if (sched->curr_pid[this_cpu] != (u32)-1) { /* * Are we trying to switch away a PID that is * not current? */ - if (sched->curr_pid[this_cpu] != switch_event.prev_pid) + if (sched->curr_pid[this_cpu] != prev_pid) sched->nr_context_switch_bugs++; } - if (sched->tp_handler->switch_event) - err = sched->tp_handler->switch_event(sched, &switch_event, machine, event, sample); - sched->curr_pid[this_cpu] = switch_event.next_pid; + if (sched->tp_handler->switch_event) { + struct trace_switch_event event = { + .prev_comm = perf_evsel__strval(evsel, sample, "prev_comm"), + .prev_pid = prev_pid, + .prev_prio = perf_evsel__intval(evsel, sample, "prev_prio"), + .prev_state = perf_evsel__intval(evsel, sample, "prev_state"), + .next_comm = perf_evsel__strval(evsel, sample, "next_comm"), + .next_pid = next_pid, + .next_prio = perf_evsel__intval(evsel, sample, "next_prio"), + }; + + err = sched->tp_handler->switch_event(sched, &event, machine, evsel, sample); + } + + sched->curr_pid[this_cpu] = next_pid; return err; } static int process_sched_runtime_event(struct perf_tool *tool, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine) { struct perf_sched *sched = container_of(tool, struct perf_sched, tool); - void *data = sample->raw_data; - struct trace_runtime_event runtime_event; - int err = 0; - FILL_ARRAY(runtime_event, comm, event, data); - FILL_FIELD(runtime_event, pid, event, data); - FILL_FIELD(runtime_event, runtime, event, data); - FILL_FIELD(runtime_event, vruntime, event, data); - - if (sched->tp_handler->runtime_event) - err = sched->tp_handler->runtime_event(sched, &runtime_event, machine, sample); + if (sched->tp_handler->runtime_event) { + struct trace_runtime_event event = { + .comm = perf_evsel__strval(evsel, sample, "comm"), + .pid = perf_evsel__intval(evsel, sample, "pid"), + .runtime = perf_evsel__intval(evsel, sample, "runtime"), + .vruntime = perf_evsel__intval(evsel, sample, "vruntime"), + }; + return sched->tp_handler->runtime_event(sched, &event, machine, sample); + } - return err; + return 0; } static int process_sched_fork_event(struct perf_tool *tool, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine __maybe_unused) { struct perf_sched *sched = container_of(tool, struct perf_sched, tool); - void *data = sample->raw_data; - struct trace_fork_event fork_event; - int err = 0; - - FILL_COMMON_FIELDS(fork_event, event, data); - FILL_ARRAY(fork_event, parent_comm, event, data); - FILL_FIELD(fork_event, parent_pid, event, data); - FILL_ARRAY(fork_event, child_comm, event, data); - FILL_FIELD(fork_event, child_pid, event, data); - - if (sched->tp_handler->fork_event) - err = sched->tp_handler->fork_event(sched, &fork_event, event); + if (sched->tp_handler->fork_event) { + struct trace_fork_event event = { + .parent_comm = perf_evsel__strval(evsel, sample, "parent_comm"), + .child_comm = perf_evsel__strval(evsel, sample, "child_comm"), + .parent_pid = perf_evsel__intval(evsel, sample, "parent_pid"), + .child_pid = perf_evsel__intval(evsel, sample, "child_pid"), + }; + return sched->tp_handler->fork_event(sched, &event, evsel); + } - return err; + return 0; } static int process_sched_exit_event(struct perf_tool *tool __maybe_unused, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample __maybe_unused, struct machine *machine __maybe_unused) { - if (verbose) - printf("sched_exit event %p\n", event); - + pr_debug("sched_exit event %p\n", evsel); return 0; } static int process_sched_migrate_task_event(struct perf_tool *tool, - struct event_format *event, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine) { struct perf_sched *sched = container_of(tool, struct perf_sched, tool); - void *data = sample->raw_data; - struct trace_migrate_task_event migrate_task_event; - int err = 0; - FILL_COMMON_FIELDS(migrate_task_event, event, data); - - FILL_ARRAY(migrate_task_event, comm, event, data); - FILL_FIELD(migrate_task_event, pid, event, data); - FILL_FIELD(migrate_task_event, prio, event, data); - FILL_FIELD(migrate_task_event, cpu, event, data); - - if (sched->tp_handler->migrate_task_event) - err = sched->tp_handler->migrate_task_event(sched, &migrate_task_event, machine, sample); + if (sched->tp_handler->migrate_task_event) { + struct trace_migrate_task_event event = { + .comm = perf_evsel__strval(evsel, sample, "comm"), + .pid = perf_evsel__intval(evsel, sample, "pid"), + .prio = perf_evsel__intval(evsel, sample, "prio"), + .cpu = perf_evsel__intval(evsel, sample, "cpu"), + }; + return sched->tp_handler->migrate_task_event(sched, &event, machine, sample); + } - return err; + return 0; } typedef int (*tracepoint_handler)(struct perf_tool *tool, - struct event_format *tp_format, + struct perf_evsel *evsel, struct perf_sample *sample, struct machine *machine); @@ -1592,7 +1523,7 @@ static int perf_sched__process_tracepoint_sample(struct perf_tool *tool __maybe_ if (evsel->handler.func != NULL) { tracepoint_handler f = evsel->handler.func; - err = f(tool, evsel->tp_format, sample, machine); + err = f(tool, evsel, sample, machine); } return err; -- 2.46.0