| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Wed, 27 Apr 2011 16:02:14 GMT
From: tip-bot for Ingo Molnar <mingo@...e.hu>
To: linux-tip-commits@...r.kernel.org
Cc: linux-kernel@...r.kernel.org, acme@...hat.com, hpa@...or.com,
mingo@...hat.com, a.p.zijlstra@...llo.nl, fweisbec@...il.com,
tglx@...utronix.de, mingo@...e.hu
Subject: [tip:perf/stat] perf stat: Add stalled cycles to the default output
Commit-ID: 1fc570ad89e55dc32dfa4dda1311948b38f26524
Gitweb: http://git.kernel.org/tip/1fc570ad89e55dc32dfa4dda1311948b38f26524
Author: Ingo Molnar <mingo@...e.hu>
AuthorDate: Wed, 27 Apr 2011 05:20:22 +0200
Committer: Ingo Molnar <mingo@...e.hu>
CommitDate: Tue, 26 Apr 2011 20:04:57 +0200
perf stat: Add stalled cycles to the default output
The new default output looks like this:
Performance counter stats for './loop_1b_instructions':
236.010686 task-clock # 0.996 CPUs utilized
0 context-switches # 0.000 M/sec
0 CPU-migrations # 0.000 M/sec
99 page-faults # 0.000 M/sec
756,487,646 cycles # 3.205 GHz
354,938,996 stalled-cycles # 46.92% of all cycles are idle
1,001,403,797 instructions # 1.32 insns per cycle
# 0.35 stalled cycles per insn
100,279,773 branches # 424.895 M/sec
12,646 branch-misses # 0.013 % of all branches
0.236902540 seconds time elapsed
We dropped cache-refs and cache-misses and added stalled-cycles - this is a
more generic "how well utilized is the CPU" metric.
If the stalled-cycles ratio is too high then more specific measurements can be
taken to figure out the source of the inefficiency.
Acked-by: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Acked-by: Arnaldo Carvalho de Melo <acme@...hat.com>
Cc: Frederic Weisbecker <fweisbec@...il.com>
Link: http://lkml.kernel.org/n/tip-pbpl2l4mn797s69bclfpwkwn@git.kernel.org
Signed-off-by: Ingo Molnar <mingo@...e.hu>
---
tools/perf/builtin-stat.c | 5 ++---
tools/perf/util/parse-events.c | 11 ++++++-----
2 files changed, 8 insertions(+), 8 deletions(-)
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index e881c20..924d18c 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -65,11 +65,10 @@ static struct perf_event_attr default_attrs[] = {
{ .type = PERF_TYPE_SOFTWARE, .config = PERF_COUNT_SW_PAGE_FAULTS },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CPU_CYCLES },
+ { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_STALLED_CYCLES },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_INSTRUCTIONS },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_INSTRUCTIONS },
{ .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_BRANCH_MISSES },
- { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_REFERENCES },
- { .type = PERF_TYPE_HARDWARE, .config = PERF_COUNT_HW_CACHE_MISSES },
};
@@ -468,7 +467,7 @@ static void abs_printout(int cpu, struct perf_evsel *evsel, double avg)
if (total)
ratio = avg * 100 / total;
- fprintf(stderr, " # %8.3f %% of all branches", ratio);
+ fprintf(stderr, " # %5.2f %% of all branches ", ratio);
} else if (perf_evsel__match(evsel, HARDWARE, HW_CACHE_MISSES) &&
runtime_cacherefs_stats[cpu].n != 0) {
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index b5bfef1..bbbb735 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -32,13 +32,13 @@ char debugfs_path[MAXPATHLEN];
static struct event_symbol event_symbols[] = {
{ CHW(CPU_CYCLES), "cpu-cycles", "cycles" },
+ { CHW(STALLED_CYCLES), "stalled-cycles", "idle-cycles" },
{ CHW(INSTRUCTIONS), "instructions", "" },
{ CHW(CACHE_REFERENCES), "cache-references", "" },
{ CHW(CACHE_MISSES), "cache-misses", "" },
{ CHW(BRANCH_INSTRUCTIONS), "branch-instructions", "branches" },
{ CHW(BRANCH_MISSES), "branch-misses", "" },
{ CHW(BUS_CYCLES), "bus-cycles", "" },
- { CHW(STALLED_CYCLES), "stalled-cycles", "" },
{ CSW(CPU_CLOCK), "cpu-clock", "" },
{ CSW(TASK_CLOCK), "task-clock", "" },
@@ -54,9 +54,9 @@ static struct event_symbol event_symbols[] = {
#define __PERF_EVENT_FIELD(config, name) \
((config & PERF_EVENT_##name##_MASK) >> PERF_EVENT_##name##_SHIFT)
-#define PERF_EVENT_RAW(config) __PERF_EVENT_FIELD(config, RAW)
+#define PERF_EVENT_RAW(config) __PERF_EVENT_FIELD(config, RAW)
#define PERF_EVENT_CONFIG(config) __PERF_EVENT_FIELD(config, CONFIG)
-#define PERF_EVENT_TYPE(config) __PERF_EVENT_FIELD(config, TYPE)
+#define PERF_EVENT_TYPE(config) __PERF_EVENT_FIELD(config, TYPE)
#define PERF_EVENT_ID(config) __PERF_EVENT_FIELD(config, EVENT)
static const char *hw_event_names[] = {
@@ -67,6 +67,7 @@ static const char *hw_event_names[] = {
"branches",
"branch-misses",
"bus-cycles",
+ "stalled-cycles",
};
static const char *sw_event_names[] = {
@@ -308,7 +309,7 @@ const char *__event_name(int type, u64 config)
switch (type) {
case PERF_TYPE_HARDWARE:
- if (config < PERF_COUNT_HW_MAX)
+ if (config < PERF_COUNT_HW_MAX && hw_event_names[config])
return hw_event_names[config];
return "unknown-hardware";
@@ -334,7 +335,7 @@ const char *__event_name(int type, u64 config)
}
case PERF_TYPE_SOFTWARE:
- if (config < PERF_COUNT_SW_MAX)
+ if (config < PERF_COUNT_SW_MAX && sw_event_names[config])
return sw_event_names[config];
return "unknown-software";
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists