From: Arnaldo Carvalho de Melo <acme@infradead.org>
To: Ingo Molnar <mingo@elte.hu>
Cc: linux-kernel@vger.kernel.org,
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>,
2nddept-manager@sdl.hitachi.co.jp,
Franck Bui-Huu <fbuihuu@gmail.com>, Ingo Molnar <mingo@elte.hu>,
Paul Mackerras <paulus@samba.org>,
Peter Zijlstra <a.p.zijlstra@chello.nl>,
Srikar Dronamraju <srikar@linux.vnet.ibm.com>,
Steven Rostedt <rostedt@goodmis.org>,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 23/29] perf probe: Enable to put probe inline function call site
Date: Tue, 25 Jan 2011 15:21:46 -0200 [thread overview]
Message-ID: <1295976112-2796-24-git-send-email-acme@infradead.org> (raw)
In-Reply-To: <1295976112-2796-1-git-send-email-acme@infradead.org>
From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Enable to put probe inline function call site. This will increase line-based
probe-ability.
<Without this patch>
$ ./perf probe -L schedule:48
<schedule:48>
pre_schedule(rq, prev);
50 if (unlikely(!rq->nr_running))
idle_balance(cpu, rq);
put_prev_task(rq, prev);
next = pick_next_task(rq);
56 if (likely(prev != next)) {
sched_info_switch(prev, next);
trace_sched_switch_out(prev, next);
perf_event_task_sched_out(prev, next);
<With this patch>
$ ./perf probe -L schedule:48
<schedule:48>
48 pre_schedule(rq, prev);
50 if (unlikely(!rq->nr_running))
51 idle_balance(cpu, rq);
53 put_prev_task(rq, prev);
54 next = pick_next_task(rq);
56 if (likely(prev != next)) {
57 sched_info_switch(prev, next);
58 trace_sched_switch_out(prev, next);
59 perf_event_task_sched_out(prev, next);
Cc: 2nddept-manager@sdl.hitachi.co.jp
Cc: Franck Bui-Huu <fbuihuu@gmail.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
LKML-Reference: <20110113124604.22426.48873.stgit@ltc236.sdl.hitachi.co.jp>
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
---
tools/perf/util/probe-finder.c | 56 ++++++++++++++++++++++++++++++++++-----
1 files changed, 48 insertions(+), 8 deletions(-)
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 508c017..69215bf 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -280,6 +280,19 @@ static bool die_compare_name(Dwarf_Die *dw_die, const char *tname)
return name ? (strcmp(tname, name) == 0) : false;
}
+/* Get callsite line number of inline-function instance */
+static int die_get_call_lineno(Dwarf_Die *in_die)
+{
+ Dwarf_Attribute attr;
+ Dwarf_Word ret;
+
+ if (!dwarf_attr(in_die, DW_AT_call_line, &attr))
+ return -ENOENT;
+
+ dwarf_formudata(&attr, &ret);
+ return (int)ret;
+}
+
/* Get type die */
static Dwarf_Die *die_get_type(Dwarf_Die *vr_die, Dwarf_Die *die_mem)
{
@@ -463,27 +476,54 @@ typedef int (* line_walk_handler_t) (const char *fname, int lineno,
Dwarf_Addr addr, void *data);
struct __line_walk_param {
+ const char *fname;
line_walk_handler_t handler;
void *data;
int retval;
};
-/* Walk on decl lines in given DIE */
+static int __die_walk_funclines_cb(Dwarf_Die *in_die, void *data)
+{
+ struct __line_walk_param *lw = data;
+ Dwarf_Addr addr;
+ int lineno;
+
+ if (dwarf_tag(in_die) == DW_TAG_inlined_subroutine) {
+ lineno = die_get_call_lineno(in_die);
+ if (lineno > 0 && dwarf_entrypc(in_die, &addr) == 0) {
+ lw->retval = lw->handler(lw->fname, lineno, addr,
+ lw->data);
+ if (lw->retval != 0)
+ return DIE_FIND_CB_FOUND;
+ }
+ }
+ return DIE_FIND_CB_SIBLING;
+}
+
+/* Walk on lines of blocks included in given DIE */
static int __die_walk_funclines(Dwarf_Die *sp_die,
line_walk_handler_t handler, void *data)
{
- const char *fname;
+ struct __line_walk_param lw = {
+ .handler = handler,
+ .data = data,
+ .retval = 0,
+ };
+ Dwarf_Die die_mem;
Dwarf_Addr addr;
- int lineno, ret = 0;
+ int lineno;
/* Handle function declaration line */
- fname = dwarf_decl_file(sp_die);
- if (fname && dwarf_decl_line(sp_die, &lineno) == 0 &&
+ lw.fname = dwarf_decl_file(sp_die);
+ if (lw.fname && dwarf_decl_line(sp_die, &lineno) == 0 &&
dwarf_entrypc(sp_die, &addr) == 0) {
- ret = handler(fname, lineno, addr, data);
+ lw.retval = handler(lw.fname, lineno, addr, data);
+ if (lw.retval != 0)
+ goto done;
}
-
- return ret;
+ die_find_child(sp_die, __die_walk_funclines_cb, &lw, &die_mem);
+done:
+ return lw.retval;
}
static int __die_walk_culines_cb(Dwarf_Die *sp_die, void *data)
--
1.6.2.5
next prev parent reply other threads:[~2011-01-25 17:22 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-25 17:21 [GIT PULL 00/29] perf/core improvements for 2.6.39 Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 01/29] perf evsel: Introduce perf_evlist Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 02/29] perf evlist: Adopt the pollfd array Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 03/29] perf evsel: Support event groups Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 04/29] perf evsel: Allow specifying if the inherit bit should be set Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 05/29] perf top: Use perf_evsel__open Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 06/29] perf record: " Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 07/29] perf evsel: Introduce mmap support Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 08/29] perf record: Use struct perf_mmap and helpers Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 09/29] perf record: Move perf_mmap__write_tail to perf.h Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 10/29] perf evlist: Move the mmap array from perf_evsel Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 11/29] perf record: Use perf_evlist__mmap Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 12/29] perf tools: Add missing cpu_map__delete() Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 13/29] perf test: Check counts on all cpus in test__open_syscall_event_on_all_cpus Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 14/29] perf test: check if cpu_map__new() return NULL Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 15/29] perf evlist: Steal mmap reading routine from 'perf top' Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 16/29] perf test: Add test for the evlist mmap routines Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 17/29] perf callchain: Feed callchains into a cursor Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 18/29] perf callchain: Rename cumul_hits into callchain_cumul_hits Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 19/29] perf callchain: Rename register_callchain_param into callchain_register_param Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 20/29] perf callchain: Don't give arbitrary gender to callchain tree nodes Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 21/29] perf top: Add native_safe_halt to skip symbols Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 22/29] perf probe: Introduce lines walker interface Arnaldo Carvalho de Melo
2011-01-25 17:21 ` Arnaldo Carvalho de Melo [this message]
2011-01-25 17:21 ` [PATCH 24/29] perf probe: Add --funcs to show available functions in symtab Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 25/29] perf record: auto detect when stdout is a pipe Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 26/29] perf tools: Pass the struct opt to the wildcard parsing routine Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 27/29] perf threads: Move thread_map to separate file Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 28/29] perf tools: Move event__parse_sample to evsel.c Arnaldo Carvalho de Melo
2011-01-25 17:21 ` [PATCH 29/29] perf evsel: Introduce perf_evsel__{in,ex}it Arnaldo Carvalho de Melo
2011-01-26 6:52 ` [GIT PULL 00/29] perf/core improvements for 2.6.39 Ingo Molnar
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=1295976112-2796-24-git-send-email-acme@infradead.org \
--to=acme@infradead.org \
--cc=2nddept-manager@sdl.hitachi.co.jp \
--cc=a.p.zijlstra@chello.nl \
--cc=acme@redhat.com \
--cc=fbuihuu@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=masami.hiramatsu.pt@hitachi.com \
--cc=mingo@elte.hu \
--cc=paulus@samba.org \
--cc=rostedt@goodmis.org \
--cc=srikar@linux.vnet.ibm.com \
/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
all inboxes | Powered by JetHome®