From: Namhyung Kim <namhyung@gmail.com>
To: Don Zickus <dzickus@redhat.com>
Cc: acme@ghostprotocols.net, LKML <linux-kernel@vger.kernel.org>,
jolsa@redhat.com, jmario@redhat.com, fowles@inreach.com,
peterz@infradead.org, eranian@google.com, andi.kleen@intel.com
Subject: Re: [PATCH 10/15 V3] perf, c2c: Display cacheline HITM analysis to stdout
Date: Wed, 09 Apr 2014 08:46:29 +0900 [thread overview]
Message-ID: <87vbujv0yy.fsf@sejong.aot.lge.com> (raw)
In-Reply-To: <1395689826-215033-11-git-send-email-dzickus@redhat.com> (Don Zickus's message of "Mon, 24 Mar 2014 15:37:01 -0400")
Hi Don,
On Mon, 24 Mar 2014 15:37:01 -0400, Don Zickus wrote:
> + sprintf(summary, "%4d %5.1f%% %5.1f%% %5.1f%% %5.1f%% %8d %8d %8d %8d %18s %6s\n",
> + record,
> + tot_dist * 100.,
> + tot_cumm * 100.,
> + ld_dist * 100.,
> + ld_cumm * 100.,
I think adding a 0 after decimal point helps readability little bit.
> + h->stats.t.rmt_hitm,
> + h->stats.t.lcl_hitm,
> + h->stats.t.st_l1hit,
> + h->stats.t.st_l1miss,
> + addrstr,
> + pidstr);
> +
> + for (j = 0; j < (int)strlen(summary); j++) printf("-");
> + printf("\n");
> + printf("%s", summary);
> + for (j = 0; j < (int)strlen(summary); j++) printf("-");
> + printf("\n");
> +}
> +
> +static void print_socket_stats_str(struct c2c_hit *clo,
> + struct c2c_stats *node_stats)
> +{
> + int i, j;
> +
> + if (!node_stats)
> + return;
> +
> + for (i = 0; i < max_node_num; i++) {
> + struct c2c_stats *stats = &node_stats[i];
> + int num = CPU_COUNT(&stats->cpuset);
> +
> + if (!num) {
> + /* pad align socket info */
> + for (j = 0; j < 21; j++)
> + printf(" ");
> + continue;
> + }
> +
> + printf("%2d{%2d ", i, num);
> +
> + if (clo->stats.t.rmt_hitm > 0)
> + printf("%5.1f%% ", 100. * ((double)stats->t.rmt_hitm / (double) clo->stats.t.rmt_hitm));
Just for your information, most perf tools show 2 digit after decimal
point so it'd be "%5.2f%%" but it's not a strict rule.
Anyway I think multiplying 100.0 first can remove the casts. So code
below will do the same thing..
100.0 * stats->t.rmt_hitm / clo->stats.t.rmt_hitm
The same goes to other percentage calculations too.
Thanks,
Namhyung
> + else
> + printf("%6s ", "n/a");
> +
> + if (clo->stats.t.store > 0)
> + printf("%5.1f%%} ", 100. * ((double)stats->t.store / (double)clo->stats.t.store));
> + else
> + printf("%6s} ", "n/a");
> + }
> +}
next prev parent reply other threads:[~2014-04-08 23:46 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-24 19:36 [PATCH 00/15 V3] perf, c2c: Add new tool to analyze cacheline contention on NUMA systems Don Zickus
2014-03-24 19:36 ` [PATCH 01/15 V3] perf: Fix stddev calculation Don Zickus
2014-03-24 19:36 ` [PATCH 02/15 V3] perf, callchain: Add generic callchain print handler for stdio Don Zickus
2014-03-24 19:36 ` [PATCH 03/15 V3] perf c2c: Shared data analyser Don Zickus
2014-04-08 6:59 ` Namhyung Kim
2014-04-08 14:22 ` Don Zickus
2014-04-09 0:58 ` Namhyung Kim
2014-04-09 1:29 ` Andi Kleen
2014-04-08 14:23 ` Don Zickus
2014-03-24 19:36 ` [PATCH 04/15 V3] perf c2c: Dump raw records, decode data_src bits Don Zickus
2014-04-08 7:09 ` Namhyung Kim
2014-03-24 19:36 ` [PATCH 05/15 V3] perf, c2c: Rework setup code to prepare for features Don Zickus
2014-03-29 17:10 ` Jiri Olsa
2014-04-01 2:52 ` Don Zickus
2014-04-08 7:41 ` Namhyung Kim
2014-04-08 14:11 ` Don Zickus
2014-04-09 1:12 ` Namhyung Kim
2014-04-09 1:36 ` Don Zickus
2014-04-11 14:57 ` Jiri Olsa
2014-04-08 7:18 ` Namhyung Kim
2014-03-24 19:36 ` [PATCH 06/15 V3] perf, c2c: Add in new options to configure latency and stores Don Zickus
2014-03-29 17:11 ` Jiri Olsa
2014-04-01 2:55 ` Don Zickus
2014-04-06 13:14 ` Jiri Olsa
2014-04-07 18:16 ` Don Zickus
2014-04-09 0:17 ` Namhyung Kim
2014-04-08 7:37 ` Namhyung Kim
2014-04-08 7:31 ` Namhyung Kim
2014-03-24 19:36 ` [PATCH 07/15 V3] perf, c2c: Add in sort on physid Don Zickus
2014-04-08 7:56 ` Namhyung Kim
2014-04-08 14:17 ` Don Zickus
2014-04-09 1:30 ` Namhyung Kim
2014-04-09 1:56 ` Don Zickus
2014-03-24 19:36 ` [PATCH 08/15 V3] perf, c2c: Add stats to track data source bits and cpu to node maps Don Zickus
2014-04-08 8:05 ` Namhyung Kim
2014-03-24 19:37 ` [PATCH 09/15 V3] perf, c2c: Sort based on hottest cache line Don Zickus
2014-04-08 8:23 ` Namhyung Kim
2014-03-24 19:37 ` [PATCH 10/15 V3] perf, c2c: Display cacheline HITM analysis to stdout Don Zickus
2014-04-08 8:26 ` Namhyung Kim
2014-04-08 23:46 ` Namhyung Kim [this message]
2014-03-24 19:37 ` [PATCH 11/15 V3] perf, c2c: Add callchain support Don Zickus
2014-03-24 19:37 ` [PATCH 12/15 V3] perf, c2c: Output summary stats Don Zickus
2014-03-24 19:37 ` [PATCH 13/15 V3] perf, c2c: Dump rbtree for debugging Don Zickus
2014-03-24 19:37 ` [PATCH 14/15 V3] perf, c2c: Add symbol count table Don Zickus
2014-03-24 19:37 ` [PATCH 15/15 V3] perf, c2c: Add shared cachline summary table Don Zickus
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=87vbujv0yy.fsf@sejong.aot.lge.com \
--to=namhyung@gmail.com \
--cc=acme@ghostprotocols.net \
--cc=andi.kleen@intel.com \
--cc=dzickus@redhat.com \
--cc=eranian@google.com \
--cc=fowles@inreach.com \
--cc=jmario@redhat.com \
--cc=jolsa@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=peterz@infradead.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
Powered by JetHome