public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Andi Kleen <andi@firstfloor.org>
To: acme@kernel.org
Cc: jolsa@kernel.org, linux-perf-users@vger.kernel.org,
	linux-kernel@vger.kernel.org, Andi Kleen <ak@linux.intel.com>
Subject: [PATCH v1 10/10] perf, tools, stat: Improve scaling
Date: Mon, 11 Mar 2019 13:24:46 -0700	[thread overview]
Message-ID: <20190311202446.10210-11-andi@firstfloor.org> (raw)
In-Reply-To: <20190311202446.10210-1-andi@firstfloor.org>

From: Andi Kleen <ak@linux.intel.com>

The multiplexing scaling in perf stat mysteriously adds 0.5 to the
value. This dates back to the original perf tool. Other scaling
code doesn't use that strange convention. Remove the extra 0.5.

Before:

$ perf stat -e 'cycles,cycles,cycles,cycles,cycles,cycles' grep -rq foo

 Performance counter stats for 'grep -rq foo':

         6,403,580      cycles                                                        (81.62%)
         6,404,341      cycles                                                        (81.64%)
         6,402,983      cycles                                                        (81.62%)
         6,399,941      cycles                                                        (81.63%)
         6,399,451      cycles                                                        (81.62%)
         6,436,105      cycles                                                        (91.87%)

       0.005843799 seconds time elapsed

       0.002905000 seconds user
       0.002902000 seconds sys

After:

$ ./perf stat -e 'cycles,cycles,cycles,cycles,cycles,cycles' grep -rq foo

 Performance counter stats for 'grep -rq foo':

         6,422,704      cycles                                                        (81.68%)
         6,401,842      cycles                                                        (81.68%)
         6,398,432      cycles                                                        (81.68%)
         6,397,098      cycles                                                        (81.68%)
         6,396,074      cycles                                                        (81.67%)
         6,434,980      cycles                                                        (91.62%)

       0.005884437 seconds time elapsed

       0.003580000 seconds user
       0.002356000 seconds sys

Signed-off-by: Andi Kleen <ak@linux.intel.com>
---
 tools/perf/util/evsel.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 60450087f198..e425f342fe81 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -1342,7 +1342,7 @@ void perf_counts_values__scale(struct perf_counts_values *count,
 			count->val = 0;
 		} else if (count->run < count->ena) {
 			scaled = 1;
-			count->val = (u64)((double) count->val * count->ena / count->run + 0.5);
+			count->val = (u64)((double) count->val * count->ena / count->run);
 		}
 	}
 
-- 
2.20.1


  parent reply	other threads:[~2019-03-11 20:24 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-11 20:24 Misc improvements and bug fixes for perf Andi Kleen
2019-03-11 20:24 ` [PATCH v1 01/10] perf, tools, list: Filter metrics too Andi Kleen
2019-03-11 20:24 ` [PATCH v1 02/10] perf, tools, stat: Avoid memory overrun with -r Andi Kleen
2019-03-11 20:28   ` Andi Kleen
2019-03-11 20:24 ` [PATCH v1 03/10] perf, tools, record: Allow to limit number of reported perf.data files Andi Kleen
2019-03-11 20:24 ` [PATCH v1 04/10] perf, tools, record: Clarify help for --switch-output Andi Kleen
2019-03-12 10:30   ` Jiri Olsa
2019-03-11 20:24 ` [PATCH v1 05/10] perf, report: Show all sort keys in help output Andi Kleen
2019-03-12 10:30   ` Jiri Olsa
2019-03-12 16:43     ` Andi Kleen
2019-03-11 20:24 ` [PATCH v1 06/10] perf, tools, report: Print better message for JITed code Andi Kleen
2019-03-11 20:33   ` Arnaldo Carvalho de Melo
2019-03-11 20:48     ` Andi Kleen
2019-03-11 20:58       ` Arnaldo Carvalho de Melo
2019-03-11 20:24 ` [PATCH v1 07/10] perf, tools, report: Indicate JITed code better in report Andi Kleen
2019-03-11 20:24 ` [PATCH v1 08/10] perf, tools, script: Support relative time Andi Kleen
2019-03-11 20:24 ` [PATCH v1 09/10] perf, tools, stat: Fix --no-scale Andi Kleen
2019-03-11 20:24 ` Andi Kleen [this message]
2019-03-12 10:31 ` Misc improvements and bug fixes for perf Jiri Olsa

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=20190311202446.10210-11-andi@firstfloor.org \
    --to=andi@firstfloor.org \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=jolsa@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox