| 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: Sat, 18 May 2019 01:49:06 -0700
From: tip-bot for Jiri Olsa <tipbot@...or.com>
To: linux-tip-commits@...r.kernel.org
Cc: tglx@...utronix.de, linux-kernel@...r.kernel.org,
namhyung@...nel.org, alexander.shishkin@...ux.intel.com,
acme@...hat.com, jolsa@...nel.org, mingo@...nel.org, hpa@...or.com,
peterz@...radead.org
Subject: [tip:perf/core] perf tools: Speed up report for perf compiled with
linwunwind
Commit-ID: 382619c07ff6491b33d54fccff7407336ddcb6d4
Gitweb: https://git.kernel.org/tip/382619c07ff6491b33d54fccff7407336ddcb6d4
Author: Jiri Olsa <jolsa@...nel.org>
AuthorDate: Fri, 26 Apr 2019 09:38:04 +0200
Committer: Arnaldo Carvalho de Melo <acme@...hat.com>
CommitDate: Wed, 15 May 2019 16:36:46 -0300
perf tools: Speed up report for perf compiled with linwunwind
When compiled with libunwind, perf does some preparatory work when
processing side-band events. This is not needed when report actually
don't unwind dwarf callchains, so it's disabled with
dwarf_callchain_users bool.
However we could move that check to higher level and shield more
unwanted code for normal report processing, giving us following speed up
on kernel build profile:
Before:
$ perf record make -j40
...
$ ll ../../perf.data
-rw-------. 1 jolsa jolsa 461783932 Apr 26 09:11 perf.data
$ perf stat -e cycles:u,instructions:u perf report -i perf.data > out
Performance counter stats for 'perf report -i perf.data':
78,669,920,155 cycles:u
99,076,431,951 instructions:u # 1.26 insn per cycle
55.382823668 seconds time elapsed
27.512341000 seconds user
27.712871000 seconds sys
After:
$ perf stat -e cycles:u,instructions:u perf report -i perf.data > out
Performance counter stats for 'perf report -i perf.data':
59,626,798,904 cycles:u
88,583,575,849 instructions:u # 1.49 insn per cycle
21.296935559 seconds time elapsed
20.010191000 seconds user
1.202935000 seconds sys
The speed is higher with profile having many side-band events,
because these trigger libunwind preparatory code.
This does not apply for perf compiled with libdw for dwarf unwind,
only for build with libunwind.
Signed-off-by: Jiri Olsa <jolsa@...nel.org>
Cc: Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Cc: Namhyung Kim <namhyung@...nel.org>
Cc: Peter Zijlstra <peterz@...radead.org>
Link: http://lkml.kernel.org/r/20190426073804.17238-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@...hat.com>
---
tools/perf/util/thread.c | 3 ++-
tools/perf/util/unwind-libunwind-local.c | 6 ------
tools/perf/util/unwind-libunwind.c | 10 ++++++++++
3 files changed, 12 insertions(+), 7 deletions(-)
diff --git a/tools/perf/util/thread.c b/tools/perf/util/thread.c
index 50678d318185..403045a2bbea 100644
--- a/tools/perf/util/thread.c
+++ b/tools/perf/util/thread.c
@@ -15,6 +15,7 @@
#include "map.h"
#include "symbol.h"
#include "unwind.h"
+#include "callchain.h"
#include <api/fs/fs.h>
@@ -327,7 +328,7 @@ static int thread__prepare_access(struct thread *thread)
{
int err = 0;
- if (symbol_conf.use_callchain)
+ if (dwarf_callchain_users)
err = __thread__prepare_access(thread);
return err;
diff --git a/tools/perf/util/unwind-libunwind-local.c b/tools/perf/util/unwind-libunwind-local.c
index f3c666a84e4d..25e1406b1f8b 100644
--- a/tools/perf/util/unwind-libunwind-local.c
+++ b/tools/perf/util/unwind-libunwind-local.c
@@ -617,8 +617,6 @@ static unw_accessors_t accessors = {
static int _unwind__prepare_access(struct thread *thread)
{
- if (!dwarf_callchain_users)
- return 0;
thread->addr_space = unw_create_addr_space(&accessors, 0);
if (!thread->addr_space) {
pr_err("unwind: Can't create unwind address space.\n");
@@ -631,15 +629,11 @@ static int _unwind__prepare_access(struct thread *thread)
static void _unwind__flush_access(struct thread *thread)
{
- if (!dwarf_callchain_users)
- return;
unw_flush_cache(thread->addr_space, 0, 0);
}
static void _unwind__finish_access(struct thread *thread)
{
- if (!dwarf_callchain_users)
- return;
unw_destroy_addr_space(thread->addr_space);
}
diff --git a/tools/perf/util/unwind-libunwind.c b/tools/perf/util/unwind-libunwind.c
index 9778b3133b77..c0811977d7d5 100644
--- a/tools/perf/util/unwind-libunwind.c
+++ b/tools/perf/util/unwind-libunwind.c
@@ -5,6 +5,7 @@
#include "session.h"
#include "debug.h"
#include "env.h"
+#include "callchain.h"
struct unwind_libunwind_ops __weak *local_unwind_libunwind_ops;
struct unwind_libunwind_ops __weak *x86_32_unwind_libunwind_ops;
@@ -24,6 +25,9 @@ int unwind__prepare_access(struct thread *thread, struct map *map,
struct unwind_libunwind_ops *ops = local_unwind_libunwind_ops;
int err;
+ if (!dwarf_callchain_users)
+ return 0;
+
if (thread->addr_space) {
pr_debug("unwind: thread map already set, dso=%s\n",
map->dso->name);
@@ -65,12 +69,18 @@ out_register:
void unwind__flush_access(struct thread *thread)
{
+ if (!dwarf_callchain_users)
+ return;
+
if (thread->unwind_libunwind_ops)
thread->unwind_libunwind_ops->flush_access(thread);
}
void unwind__finish_access(struct thread *thread)
{
+ if (!dwarf_callchain_users)
+ return;
+
if (thread->unwind_libunwind_ops)
thread->unwind_libunwind_ops->finish_access(thread);
}
Powered by blists - more mailing lists