From: tip-bot for Kirill Smelkov <kirr@landau.phys.spbu.ru>
To: linux-tip-commits@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, acme@redhat.com, hpa@zytor.com,
mingo@redhat.com, kirr@landau.phys.spbu.ru, efault@gmx.de,
tglx@linutronix.de, mingo@elte.hu
Subject: [tip:perf/core] perf top: Teach it to autolocate vmlinux
Date: Thu, 4 Feb 2010 09:54:36 GMT [thread overview]
Message-ID: <tip-6cff0e8dbaa4d5d822a814e5028683d7e71c3291@git.kernel.org> (raw)
In-Reply-To: <1265223128-11786-9-git-send-email-acme@infradead.org>
Commit-ID: 6cff0e8dbaa4d5d822a814e5028683d7e71c3291
Gitweb: http://git.kernel.org/tip/6cff0e8dbaa4d5d822a814e5028683d7e71c3291
Author: Kirill Smelkov <kirr@landau.phys.spbu.ru>
AuthorDate: Wed, 3 Feb 2010 16:52:08 -0200
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Thu, 4 Feb 2010 09:33:28 +0100
perf top: Teach it to autolocate vmlinux
By relying on logic in dso__load_kernel_sym(), we can
automatically load vmlinux.
The only thing which needs to be adjusted, is how --sym-annotate
option is handled - now we can't rely on vmlinux been loaded
until full successful pass of dso__load_vmlinux(), but that's
not the case if we'll do sym_filter_entry setup in
symbol_filter().
So move this step right after event__process_sample() where we
know the whole dso__load_kernel_sym() pass is done.
By the way, though conceptually similar `perf top` still can't
annotate userspace - see next patches with fixes.
Signed-off-by: Kirill Smelkov <kirr@landau.phys.spbu.ru>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Cc: Mike Galbraith <efault@gmx.de>
LKML-Reference: <1265223128-11786-9-git-send-email-acme@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
tools/perf/Documentation/perf-top.txt | 2 +-
tools/perf/builtin-top.c | 39 +++++++++++++++++++-------------
2 files changed, 24 insertions(+), 17 deletions(-)
diff --git a/tools/perf/Documentation/perf-top.txt b/tools/perf/Documentation/perf-top.txt
index 4a7d558..785b9fc 100644
--- a/tools/perf/Documentation/perf-top.txt
+++ b/tools/perf/Documentation/perf-top.txt
@@ -74,7 +74,7 @@ OPTIONS
-s <symbol>::
--sym-annotate=<symbol>::
- Annotate this symbol. Requires -k option.
+ Annotate this symbol.
-v::
--verbose::
diff --git a/tools/perf/builtin-top.c b/tools/perf/builtin-top.c
index 1fc018e..83c09c8 100644
--- a/tools/perf/builtin-top.c
+++ b/tools/perf/builtin-top.c
@@ -94,6 +94,7 @@ struct source_line {
static char *sym_filter = NULL;
struct sym_entry *sym_filter_entry = NULL;
+struct sym_entry *sym_filter_entry_sched = NULL;
static int sym_pcnt_filter = 5;
static int sym_counter = 0;
static int display_weighted = -1;
@@ -695,11 +696,9 @@ static void print_mapped_keys(void)
fprintf(stdout, "\t[f] profile display filter (count). \t(%d)\n", count_filter);
- if (symbol_conf.vmlinux_name) {
- fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
- fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL");
- fprintf(stdout, "\t[S] stop annotation.\n");
- }
+ fprintf(stdout, "\t[F] annotate display filter (percent). \t(%d%%)\n", sym_pcnt_filter);
+ fprintf(stdout, "\t[s] annotate symbol. \t(%s)\n", name?: "NULL");
+ fprintf(stdout, "\t[S] stop annotation.\n");
if (nr_counters > 1)
fprintf(stdout, "\t[w] toggle display weighted/count[E]r. \t(%d)\n", display_weighted ? 1 : 0);
@@ -725,14 +724,13 @@ static int key_mapped(int c)
case 'Q':
case 'K':
case 'U':
+ case 'F':
+ case 's':
+ case 'S':
return 1;
case 'E':
case 'w':
return nr_counters > 1 ? 1 : 0;
- case 'F':
- case 's':
- case 'S':
- return symbol_conf.vmlinux_name ? 1 : 0;
default:
break;
}
@@ -910,8 +908,12 @@ static int symbol_filter(struct map *map, struct symbol *sym)
syme = symbol__priv(sym);
syme->map = map;
syme->src = NULL;
- if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter))
- sym_filter_entry = syme;
+
+ if (!sym_filter_entry && sym_filter && !strcmp(name, sym_filter)) {
+ /* schedule initial sym_filter_entry setup */
+ sym_filter_entry_sched = syme;
+ sym_filter = NULL;
+ }
for (i = 0; skip_symbols[i]; i++) {
if (!strcmp(skip_symbols[i], name)) {
@@ -976,6 +978,13 @@ static void event__process_sample(const event_t *self,
return;
}
+ /* let's see, whether we need to install initial sym_filter_entry */
+ if (sym_filter_entry_sched) {
+ sym_filter_entry = sym_filter_entry_sched;
+ sym_filter_entry_sched = NULL;
+ parse_source(sym_filter_entry);
+ }
+
syme = symbol__priv(al.sym);
if (!syme->skip) {
syme->count[counter]++;
@@ -1270,7 +1279,7 @@ static const struct option options[] = {
OPT_BOOLEAN('i', "inherit", &inherit,
"child tasks inherit counters"),
OPT_STRING('s', "sym-annotate", &sym_filter, "symbol name",
- "symbol to annotate - requires -k option"),
+ "symbol to annotate"),
OPT_BOOLEAN('z', "zero", &zero,
"zero history across updates"),
OPT_INTEGER('F', "freq", &freq,
@@ -1306,16 +1315,14 @@ int cmd_top(int argc, const char **argv, const char *prefix __used)
symbol_conf.priv_size = (sizeof(struct sym_entry) +
(nr_counters + 1) * sizeof(unsigned long));
- if (symbol_conf.vmlinux_name == NULL)
- symbol_conf.try_vmlinux_path = true;
+
+ symbol_conf.try_vmlinux_path = (symbol_conf.vmlinux_name == NULL);
if (symbol__init() < 0)
return -1;
if (delay_secs < 1)
delay_secs = 1;
- parse_source(sym_filter_entry);
-
/*
* User specified count overrides default frequency.
*/
next prev parent reply other threads:[~2010-02-04 9:56 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-02-03 18:52 [PATCH 1/9] perf symbols: Remove perf_session usage in symbols layer Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 2/9] perf symbols: Fixup vsyscall maps Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 3/9] perf symbols: Ditch vdso global variable Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 4/9] perf probe: Don't use a perf_session instance just to resolve symbols Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 5/9] perf build-id: Move the routine to find DSOs with hits to the lib Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 6/9] perf record: Stop intercepting events, use postprocessing to get build-ids Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 7/9] perf tools: Adjust some verbosity levels Arnaldo Carvalho de Melo
2010-02-03 18:52 ` [PATCH 8/9] perf annotate: fix it for non-prelinked *.so Arnaldo Carvalho de Melo
2010-02-04 6:31 ` Mike Galbraith
2010-02-04 9:54 ` [tip:perf/core] perf annotate: Fix perf top module symbol annotation tip-bot for Mike Galbraith
2010-02-04 19:34 ` [PATCH 8/9] perf annotate: fix it for non-prelinked *.so Kirill Smelkov
2010-02-04 19:48 ` Arnaldo Carvalho de Melo
2010-02-05 6:54 ` Mike Galbraith
2010-02-04 9:54 ` [tip:perf/core] perf annotate: Fix " tip-bot for Kirill Smelkov
2010-02-03 18:52 ` [PATCH 9/9] perf top: teach it to autolocate vmlinux Arnaldo Carvalho de Melo
2010-02-04 9:54 ` tip-bot for Kirill Smelkov [this message]
2010-02-04 9:26 ` [PATCH 1/9] perf symbols: Remove perf_session usage in symbols layer Ingo Molnar
2010-02-04 13:04 ` 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=tip-6cff0e8dbaa4d5d822a814e5028683d7e71c3291@git.kernel.org \
--to=kirr@landau.phys.spbu.ru \
--cc=acme@redhat.com \
--cc=efault@gmx.de \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
/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;
as well as URLs for NNTP newsgroup(s).