From: kan.liang@intel.com
To: a.p.zijlstra@chello.nl, acme@kernel.org, linux-kernel@vger.kernel.org
Cc: ak@linux.intel.com, Kan Liang <kan.liang@intel.com>
Subject: [PATCH 3/5] perf,tools: change perf stat to use event's cpu map
Date: Tue, 3 Mar 2015 03:54:44 -0500 [thread overview]
Message-ID: <1425372886-33397-3-git-send-email-kan.liang@intel.com> (raw)
In-Reply-To: <1425372886-33397-1-git-send-email-kan.liang@intel.com>
From: Kan Liang <kan.liang@intel.com>
With cpu_list, perf stat should use event's cpu map.
Different events may have different cpu map. So the same index in cpu
map does not point to the same CPU id. perf_evsel__get_cpumap_index is
introduced to find out the CPU id's index for a given cpu map.
Signed-off-by: Kan Liang <kan.liang@intel.com>
---
tools/perf/builtin-stat.c | 5 +++--
tools/perf/util/evsel.h | 24 ++++++++++++++++++++++++
2 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/tools/perf/builtin-stat.c b/tools/perf/builtin-stat.c
index e598e4e..5ae8686 100644
--- a/tools/perf/builtin-stat.c
+++ b/tools/perf/builtin-stat.c
@@ -165,7 +165,7 @@ static inline void diff_timespec(struct timespec *r, struct timespec *a,
static inline struct cpu_map *perf_evsel__cpus(struct perf_evsel *evsel)
{
- return (evsel->cpus && !target.cpu_list) ? evsel->cpus : evsel_list->cpus;
+ return evsel->cpus ? evsel->cpus : evsel_list->cpus;
}
static inline int perf_evsel__nr_cpus(struct perf_evsel *evsel)
@@ -302,7 +302,7 @@ static int create_perf_stat_counter(struct perf_evsel *evsel)
attr->inherit = !no_inherit;
if (target__has_cpu(&target))
- return perf_evsel__open_per_cpu(evsel, perf_evsel__cpus(evsel));
+ return perf_evsel__open_per_cpu(evsel, evsel_list->cpus);
if (!target__has_task(&target) && perf_evsel__is_group_leader(evsel)) {
attr->disabled = 1;
@@ -1218,6 +1218,7 @@ static void print_aggr(char *prefix)
nr = 0;
for (cpu = 0; cpu < perf_evsel__nr_cpus(counter); cpu++) {
cpu2 = perf_evsel__cpus(counter)->map[cpu];
+ cpu2 = perf_evsel__get_cpumap_index(NULL, cpu2, evsel_list->cpus);
s2 = aggr_get_id(evsel_list->cpus, cpu2);
if (s2 != id)
continue;
diff --git a/tools/perf/util/evsel.h b/tools/perf/util/evsel.h
index dcf202a..a633320 100644
--- a/tools/perf/util/evsel.h
+++ b/tools/perf/util/evsel.h
@@ -8,6 +8,7 @@
#include <linux/types.h>
#include "xyarray.h"
#include "symbol.h"
+#include "cpumap.h"
struct perf_counts_values {
union {
@@ -359,4 +360,27 @@ static inline bool has_branch_callstack(struct perf_evsel *evsel)
{
return evsel->attr.branch_sample_type & PERF_SAMPLE_BRANCH_CALL_STACK;
}
+
+static inline int perf_evsel__get_cpumap_index(struct perf_evsel *evsel,
+ int cpu,
+ struct cpu_map *evlist_cpus)
+{
+ struct cpu_map *cpus;
+ int i;
+
+ if (evlist_cpus == NULL)
+ return -1;
+
+ if (evsel && evsel->cpus)
+ cpus = evsel->cpus;
+ else
+ cpus = evlist_cpus;
+
+ for (i = 0; i < cpus->nr; i++) {
+ if (cpu == cpus->map[i])
+ return i;
+ }
+ return -1;
+}
+
#endif /* __PERF_EVSEL_H */
--
1.8.3.1
next prev parent reply other threads:[~2015-03-03 16:04 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-03 8:54 [PATCH 1/5] perf,core: allow invalid context events to be part of sw/hw groups kan.liang
2015-03-03 8:54 ` [PATCH 2/5] perf,tools: check and re-organize evsel cpu maps kan.liang
2015-03-03 16:09 ` Arnaldo Carvalho de Melo
2015-03-03 16:11 ` Arnaldo Carvalho de Melo
2015-03-03 17:09 ` Liang, Kan
2015-03-04 0:15 ` Arnaldo Carvalho de Melo
2015-03-18 12:31 ` Liang, Kan
2015-03-03 8:54 ` kan.liang [this message]
2015-03-03 8:54 ` [PATCH 4/5] perf,tools: open/mmap event according to event's cpu map not evlist's kan.liang
2015-03-03 8:54 ` [PATCH 5/5] perf/x86/intel/uncore: do not implicitly set uncore event cpu kan.liang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1425372886-33397-3-git-send-email-kan.liang@intel.com \
--to=kan.liang@intel.com \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@kernel.org \
--cc=ak@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
Powered by JetHome