linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Ingo Molnar <mingo@kernel.org>, Thomas Gleixner <tglx@linutronix.de>
Cc: Jiri Olsa <jolsa@kernel.org>, Namhyung Kim <namhyung@kernel.org>,
	Clark Williams <williams@redhat.com>,
	linux-kernel@vger.kernel.org, linux-perf-users@vger.kernel.org,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: [PATCH 05/73] perf tools: Speed up report for perf compiled with linwunwind
Date: Fri, 17 May 2019 16:35:03 -0300	[thread overview]
Message-ID: <20190517193611.4974-6-acme@kernel.org> (raw)
In-Reply-To: <20190517193611.4974-1-acme@kernel.org>

From: Jiri Olsa <jolsa@kernel.org>

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@kernel.org>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20190426073804.17238-1-jolsa@kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.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 @@ int unwind__prepare_access(struct thread *thread, struct map *map,
 
 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);
 }
-- 
2.20.1


  parent reply	other threads:[~2019-05-17 19:36 UTC|newest]

Thread overview: 80+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-17 19:34 [GIT PULL] perf/core improvements and fixes Arnaldo Carvalho de Melo
2019-05-17 19:34 ` [PATCH 01/73] perf annotate: Remove hist__account_cycles() from callback Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 02/73] perf test: Fix spelling mistake "leadking" -> "leaking" Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 03/73] csky: Add support for libdw Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 04/73] tools lib traceevent: Remove hard coded install paths from pkg-config file Arnaldo Carvalho de Melo
2019-05-17 19:35 ` Arnaldo Carvalho de Melo [this message]
2019-05-17 19:35 ` [PATCH 06/73] tools arch: Update arch/x86/lib/memcpy_64.S copy used in 'perf bench mem memcpy' Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 07/73] tools arch uapi: Sync the x86 kvm.h copy Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 08/73] tools x86 uapi asm: Sync the pt_regs.h copy with the kernel sources Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 09/73] tools pci: Do not delete pcitest.sh in 'make clean' Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 10/73] perf record: Fix suggestion to get list of registers usable with --user-regs and --intr-regs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 11/73] perf parse-regs: Improve error output when faced with unknown register name Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 12/73] perf tools x86: Add support for recording and printing XMM registers Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 13/73] perf scripts python: exported-sql-viewer.py: Move view creation Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 14/73] perf scripts python: exported-sql-viewer.py: Fix error when shrinking / enlarging font Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 15/73] perf scripts python: exported-sql-viewer.py: Add tree level Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 16/73] perf scripts python: exported-sql-viewer.py: Add copy to clipboard Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 17/73] perf scripts python: exported-sql-viewer.py: Add context menu Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 18/73] perf scripts python: exported-sql-viewer.py: Add 'About' dialog box Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 19/73] perf vendor events intel: Add uncore_upi JSON support Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 20/73] perf machine: Null-terminate version char array upon fgets(/proc/version) error Arnaldo Carvalho de Melo
2019-05-18  0:05   ` Donald Yandt
2019-05-20 14:46     ` Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 21/73] tools lib traceevent: Introduce man pages Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 22/73] tools lib traceevent: Add support for man pages with multiple names Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 23/73] tools lib traceevent: Man pages for tep_handler related APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 24/73] tools lib traceevent: Man page for header_page APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 25/73] tools lib traceevent: Man page for get/set cpus APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 26/73] tools lib traceevent: Man page for file endian APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 27/73] tools lib traceevent: Man page for host " Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 28/73] tools lib traceevent: Man page for page size APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 29/73] tools lib traceevent: Man page for tep_strerror() Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 30/73] tools lib traceevent: Man pages for event handler APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 31/73] tools lib traceevent: Man pages for function related libtraceevent APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 32/73] tools lib traceevent: Man pages for registering print function Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 33/73] tools lib traceevent: Man page for tep_read_number() Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 34/73] tools lib traceevent: Man pages for event find APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 35/73] tools lib traceevent: Man page for list events APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 36/73] tools lib traceevent: Man pages for libtraceevent event get APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 37/73] tools lib traceevent: Man pages for find field APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 38/73] tools lib traceevent: Man pages for get field value APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 39/73] tools lib traceevent: Man pages for print field APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 40/73] tools lib traceevent: Man page for tep_read_number_field() Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 41/73] tools lib traceevent: Man pages for event fields APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 42/73] tools lib traceevent: Man pages for event filter APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 43/73] tools lib traceevent: Man pages for parse event APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 44/73] tools lib traceevent: Man page for tep_parse_header_page() Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 45/73] tools lib traceevent: Man pages for APIs used to extract common fields from a record Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 46/73] tools lib traceevent: Man pages for trace sequences APIs Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 47/73] perf build tests: Add NO_LIBZSTD=1 to make_minimal Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 48/73] perf session: Define 'bytes_transferred' and 'bytes_compressed' metrics Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 49/73] perf record: Implement COMPRESSED event record and its attributes Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 50/73] perf mmap: Implement dedicated memory buffer for data compression Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 51/73] perf tools: Introduce Zstd streaming based compression API Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 52/73] perf record: Implement compression for serial trace streaming Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 53/73] perf record: Implement compression for AIO " Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 54/73] perf report: Add stub processing of compressed events for -D Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 55/73] perf record: Implement -z,--compression_level[=<n>] option Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 56/73] perf report: Implement perf.data record decompression Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 57/73] perf inject: Enable COMPRESSED " Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 58/73] perf tests: Implement Zstd comp/decomp integration test Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 59/73] perf test zstd: Fixup verbose mode output Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 60/73] perf jevents: Remove unused variable Arnaldo Carvalho de Melo
2019-05-17 19:35 ` [PATCH 61/73] perf vendor events arm64: Remove [[:xdigit:]] wildcard Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 62/73] perf vendor events arm64: Map Brahma-B53 CPUID to cortex-a53 events Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 63/73] perf vendor events arm64: Add Cortex-A57 and Cortex-A72 events Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 64/73] perf parse-regs: Split parse_regs Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 65/73] perf parse-regs: Add generic support for arch__intr/user_reg_mask() Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 66/73] perf regs x86: Add X86 specific arch__intr_reg_mask() Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 67/73] perf intel-pt: Fix instructions sampling rate Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 68/73] perf intel-pt: Fix improved sample timestamp Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 69/73] perf intel-pt: Fix sample timestamp wrt non-taken branches Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 70/73] perf docs: Add description for stderr Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 71/73] perf tools: Add a 'percore' event qualifier Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 72/73] perf stat: Factor out aggregate counts printing Arnaldo Carvalho de Melo
2019-05-17 19:36 ` [PATCH 73/73] perf stat: Support 'percore' event qualifier Arnaldo Carvalho de Melo
2019-05-18  8:27 ` [GIT PULL] perf/core improvements and fixes Ingo Molnar
2019-05-18  8:42 ` [PATCH] tools/headers: Synchronize kernel ABI headers Ingo Molnar
2019-05-18 13:39   ` Arnaldo Carvalho de Melo
2019-05-18 17:12     ` 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=20190517193611.4974-6-acme@kernel.org \
    --to=acme@kernel.org \
    --cc=acme@redhat.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=williams@redhat.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;
as well as URLs for NNTP newsgroup(s).