All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kalesh Singh <kaleshsingh@google.com>
To: unlisted-recipients:; (no To-header on input)
Cc: surenb@google.com, hridya@google.com, namhyung@kernel.org,
	kernel-team@android.com, Kalesh Singh <kaleshsingh@google.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Jonathan Corbet <corbet@lwn.net>, Ingo Molnar <mingo@redhat.com>,
	Shuah Khan <shuah@kernel.org>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	Tom Zanussi <zanussi@kernel.org>,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-kselftest@vger.kernel.org
Subject: [PATCH v4 6/8] tracing/histogram: Optimize division by a power of 2
Date: Mon, 25 Oct 2021 13:08:38 -0700	[thread overview]
Message-ID: <20211025200852.3002369-7-kaleshsingh@google.com> (raw)
In-Reply-To: <20211025200852.3002369-1-kaleshsingh@google.com>

The division is a slow operation. If the divisor is a power of 2, use a
shift instead.

Results were obtained using Android's version of perf (simpleperf[1]) as
described below:

1. hist_field_div() is modified to call 2 test functions:
   test_hist_field_div_[not]_optimized(); passing them the
   same args. Use noinline and volatile to ensure these are
   not optimized out by the compiler.
2. Create a hist event trigger that uses division:
      events/kmem/rss_stat$ echo 'hist:keys=common_pid:x=size/<divisor>'
         >> trigger
      events/kmem/rss_stat$ echo 'hist:keys=common_pid:vals=$x'
         >> trigger
3. Run Android's lmkd_test[2] to generate rss_stat events, and
   record CPU samples with Android's simpleperf:
      simpleperf record -a --exclude-perf --post-unwind=yes -m 16384 -g
         -f 2000 -o perf.data

== Results ==

Divisor is a power of 2 (divisor == 32):

   test_hist_field_div_not_optimized  | 8,717,091 cpu-cycles
   test_hist_field_div_optimized      | 1,643,137 cpu-cycles

If the divisor is a power of 2, the optimized version is ~5.3x faster.

Divisor is not a power of 2 (divisor == 33):

   test_hist_field_div_not_optimized  | 4,444,324 cpu-cycles
   test_hist_field_div_optimized      | 5,497,958 cpu-cycles

If the divisor is not a power of 2, as expected, the optimized version is
slightly slower (~24% slower).

[1] https://android.googlesource.com/platform/system/extras/+/master/simpleperf/doc/README.md
[2] https://cs.android.com/android/platform/superproject/+/master:system/memory/lmkd/tests/lmkd_test.cpp

Signed-off-by: Kalesh Singh <kaleshsingh@google.com>
Suggested-by: Steven Rostedt <rostedt@goodmis.org>

---
 kernel/trace/trace_events_hist.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/kernel/trace/trace_events_hist.c b/kernel/trace/trace_events_hist.c
index db28bcf976f4..364cb3091789 100644
--- a/kernel/trace/trace_events_hist.c
+++ b/kernel/trace/trace_events_hist.c
@@ -304,6 +304,10 @@ static u64 hist_field_div(struct hist_field *hist_field,
 	if (!val2)
 		return -1;
 
+	/* Use shift if the divisor is a power of 2 */
+	if (!(val2 & (val2 - 1)))
+		return val1 >> __ffs64(val2);
+
 	return div64_u64(val1, val2);
 }
 
-- 
2.33.0.1079.g6e70778dc9-goog


  parent reply	other threads:[~2021-10-25 20:24 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-25 20:08 [PATCH v4 0/8] tracing: Extend histogram triggers expression parsing Kalesh Singh
2021-10-25 20:08 ` [PATCH v4 1/8] tracing: Add support for creating hist trigger variables from literal Kalesh Singh
2021-10-25 20:08 ` [PATCH v4 2/8] tracing: Add division and multiplication support for hist triggers Kalesh Singh
2021-10-25 20:08 ` [PATCH v4 3/8] tracing: Fix operator precedence for hist triggers expression Kalesh Singh
2021-10-25 20:08 ` [PATCH v4 4/8] tracing/histogram: Simplify handling of .sym-offset in expressions Kalesh Singh
2021-10-25 20:08 ` [PATCH v4 5/8] tracing/histogram: Covert expr to const if both operands are constants Kalesh Singh
2021-10-25 20:08 ` Kalesh Singh [this message]
2021-10-26 19:14   ` [PATCH v4 6/8] tracing/histogram: Optimize division by a power of 2 Steven Rostedt
2021-10-26 23:39     ` Kalesh Singh
2021-10-27  0:18       ` Steven Rostedt
2021-10-27  1:09         ` Kalesh Singh
2021-10-27  1:15           ` Steven Rostedt
2021-10-27  1:31             ` Kalesh Singh
2021-10-27  2:21               ` Steven Rostedt
2021-10-27  3:15                 ` Steven Rostedt
2021-10-27  4:04                   ` Kalesh Singh
2021-10-27 14:06                     ` Steven Rostedt
2021-10-25 20:08 ` [PATCH v4 7/8] tracing/selftests: Add tests for hist trigger expression parsing Kalesh Singh
2021-10-26 12:43   ` Masami Hiramatsu
2021-10-26 14:28     ` Kalesh Singh
2021-10-26 21:44       ` Steven Rostedt
2021-10-26 23:36         ` Kalesh Singh
2021-10-27  0:20           ` Steven Rostedt
2021-10-27  1:15             ` Kalesh Singh
2021-10-27  3:14               ` Masami Hiramatsu
2021-10-27  4:27                 ` Kalesh Singh
2021-10-27 14:31                   ` Steven Rostedt
2021-10-27 14:52                     ` Masami Hiramatsu
2021-10-27 15:01                       ` Steven Rostedt
2021-10-27 15:50                         ` Steven Rostedt
2021-10-27 15:55                           ` Kalesh Singh
2021-10-27 17:17                             ` Steven Rostedt
2021-10-27  2:34       ` Masami Hiramatsu
2021-10-27 17:36         ` Steven Rostedt
2021-10-26 15:07     ` Steven Rostedt
2021-10-29  6:48   ` [tracing/selftests] cfece71411: kernel-selftests.ftrace.event_trigger_-_test_inter-event_histogram_trigger_onchange_action.fail kernel test robot
2021-10-29  6:48     ` kernel test robot
2021-10-29 12:00     ` Masami Hiramatsu
2021-10-29 12:00       ` Masami Hiramatsu
2021-10-29 13:10       ` Steven Rostedt
2021-10-29 13:10         ` Steven Rostedt
2021-11-01  3:43       ` [LKP] " Li Zhijian
2021-11-01  3:43         ` Li Zhijian
2021-10-25 20:08 ` [PATCH v4 8/8] tracing/histogram: Document expression arithmetic and constants Kalesh Singh

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=20211025200852.3002369-7-kaleshsingh@google.com \
    --to=kaleshsingh@google.com \
    --cc=corbet@lwn.net \
    --cc=hridya@google.com \
    --cc=kernel-team@android.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=mhiramat@kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@kernel.org \
    --cc=rostedt@goodmis.org \
    --cc=shuah@kernel.org \
    --cc=surenb@google.com \
    --cc=zanussi@kernel.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.