From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Naohiro Aota <naota@elisp.net>,
Peter Zijlstra <peterz@infradead.org>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
David Ahern <dsahern@gmail.com>,
namhyung@kernel.org, Jiri Olsa <jolsa@redhat.com>,
Ingo Molnar <mingo@kernel.org>
Subject: [PATCH perf/core ] perf probe: Speed up perf probe --list by caching debuginfo
Date: Wed, 17 Jun 2015 23:58:54 +0900 [thread overview]
Message-ID: <20150617145854.19715.15314.stgit@localhost.localdomain> (raw)
In-Reply-To: <55818B0A.4030506@hitachi.com>
Speed up the "perf probe --list" by caching the last used
debuginfo. perf probe --list always open and load debuginfo
for each entry of probe list. This takes very a long time.
E.g. with vfs_* events (total 96 probes)
----
[root@localhost perf]# time ./perf probe -l &> /dev/null
real 0m25.376s
user 0m24.381s
sys 0m1.012s
----
To solve this issue, this adds debuginfo_cache to cache the
last used debuginfo on memory.
With this fix, the perf-probe --list significantly improves
its speed.
----
[root@localhost perf]# time ./perf probe -l &> /dev/null
real 0m0.161s
user 0m0.136s
sys 0m0.025s
----
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
---
tools/perf/util/probe-event.c | 48 ++++++++++++++++++++++++++++++++++++++---
1 file changed, 44 insertions(+), 4 deletions(-)
diff --git a/tools/perf/util/probe-event.c b/tools/perf/util/probe-event.c
index 65a1c82..076527b 100644
--- a/tools/perf/util/probe-event.c
+++ b/tools/perf/util/probe-event.c
@@ -429,6 +429,41 @@ static struct debuginfo *open_debuginfo(const char *module, bool silent)
return ret;
}
+/* For caching the last debuginfo */
+static struct debuginfo *debuginfo_cache;
+static char *debuginfo_cache_path;
+
+static struct debuginfo *debuginfo_cache__open(const char *module, bool silent)
+{
+ if ((debuginfo_cache_path && !strcmp(debuginfo_cache_path, module)) ||
+ (!debuginfo_cache_path && !module && debuginfo_cache))
+ goto out;
+
+ /* Copy module path */
+ free(debuginfo_cache_path);
+ if (module) {
+ debuginfo_cache_path = strdup(module);
+ if (!debuginfo_cache_path) {
+ debuginfo__delete(debuginfo_cache);
+ debuginfo_cache = NULL;
+ goto out;
+ }
+ }
+
+ debuginfo_cache = open_debuginfo(module, silent);
+ if (!debuginfo_cache)
+ zfree(&debuginfo_cache_path);
+out:
+ return debuginfo_cache;
+}
+
+static void debuginfo_cache__exit(void)
+{
+ debuginfo__delete(debuginfo_cache);
+ debuginfo_cache = NULL;
+ zfree(&debuginfo_cache_path);
+}
+
static int get_text_start_address(const char *exec, unsigned long *address)
{
@@ -490,12 +525,11 @@ static int find_perf_probe_point_from_dwarf(struct probe_trace_point *tp,
pr_debug("try to find information at %" PRIx64 " in %s\n", addr,
tp->module ? : "kernel");
- dinfo = open_debuginfo(tp->module, verbose == 0);
- if (dinfo) {
+ dinfo = debuginfo_cache__open(tp->module, verbose == 0);
+ if (dinfo)
ret = debuginfo__find_probe_point(dinfo,
(unsigned long)addr, pp);
- debuginfo__delete(dinfo);
- } else
+ else
ret = -ENOENT;
if (ret > 0) {
@@ -930,6 +964,10 @@ out:
#else /* !HAVE_DWARF_SUPPORT */
+static void debuginfo_cache__exit(void)
+{
+}
+
static int
find_perf_probe_point_from_dwarf(struct probe_trace_point *tp __maybe_unused,
struct perf_probe_point *pp __maybe_unused,
@@ -2266,6 +2304,8 @@ next:
break;
}
strlist__delete(rawlist);
+ /* Cleanup cached debuginfo if needed */
+ debuginfo_cache__exit();
return ret;
}
next prev parent reply other threads:[~2015-06-17 15:02 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-16 11:50 [PATCH perf/core v3 0/3] perf-probe bugfixes Masami Hiramatsu
2015-06-16 11:50 ` [PATCH perf/core v3 1/3] [BUGFIX] perf probe: List probes in stdout Masami Hiramatsu
2015-06-16 11:50 ` [PATCH perf/core v3 2/3] [BUGFIX] perf probe: Fix to return error if no-probe is added Masami Hiramatsu
2015-06-18 8:15 ` [tip:perf/core] perf probe: Fix to return error if no probe " tip-bot for Masami Hiramatsu
2015-06-16 11:50 ` [PATCH perf/core v3 3/3] [BUGFIX] perf probe: Show usage even if the last event is skipped Masami Hiramatsu
2015-06-16 14:46 ` Arnaldo Carvalho de Melo
2015-06-17 5:31 ` Masami Hiramatsu
2015-06-17 14:58 ` Masami Hiramatsu
2015-06-17 14:58 ` Masami Hiramatsu [this message]
2015-06-18 8:17 ` [tip:perf/core] perf probe: Speed up perf probe --list by caching debuginfo tip-bot for Masami Hiramatsu
2015-06-17 15:00 ` Re: [PATCH perf/core v3 3/3] [BUGFIX] perf probe: Show usage even if the last event is skipped Arnaldo Carvalho de Melo
2015-06-18 8:16 ` [tip:perf/core] " tip-bot for Masami Hiramatsu
2015-06-16 13:48 ` [PATCH perf/core v3 0/3] perf-probe bugfixes Arnaldo Carvalho de Melo
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=20150617145854.19715.15314.stgit@localhost.localdomain \
--to=masami.hiramatsu.pt@hitachi.com \
--cc=acme@kernel.org \
--cc=dsahern@gmail.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=namhyung@kernel.org \
--cc=naota@elisp.net \
--cc=peterz@infradead.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.