linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Arnaldo Carvalho de Melo <acme@infradead.org>
To: Adrian Hunter <adrian.hunter@intel.com>
Cc: Ingo Molnar <mingo@kernel.org>,
	linux-kernel@vger.kernel.org, David Ahern <dsahern@gmail.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Jiri Olsa <jolsa@redhat.com>, Mike Galbraith <efault@gmx.de>,
	Namhyung Kim <namhyung@kernel.org>,
	Paul Mackerras <paulus@samba.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Stephane Eranian <eranian@google.com>
Subject: Re: [PATCH 10/10] perf tests: Use lower sample_freq in sw clock event period test
Date: Tue, 12 Nov 2013 11:41:01 -0300	[thread overview]
Message-ID: <20131112144101.GF11239@ghostprotocols.net> (raw)
In-Reply-To: <5281D3B8.2030104@intel.com>

Em Tue, Nov 12, 2013 at 09:07:36AM +0200, Adrian Hunter escreveu:
> On 11/11/13 22:22, Arnaldo Carvalho de Melo wrote:
> > We were using it at 10 kHz, which doesn't work in machines where somehow
> > the max freq was auto reduced by the kernel:
> >
> > [root@ssdandy ~]# perf test 19
> > 19: Test software clock events have valid period values    : FAILED!
> > [root@ssdandy ~]# perf test -v 19
> > 19: Test software clock events have valid period values    :
> > --- start ---
> > Couldn't open evlist: Invalid argument
> > ---- end ----
> > Test software clock events have valid period values: FAILED!
> > [root@ssdandy ~]#
> >
> > [root@ssdandy ~]# cat /proc/sys/kernel/perf_event_max_sample_rate
> > 7000
> >
> > Reducing it to 500 Hz should be good enough for this test and also
> > shouldn't affect what it is testing.
> >
> > But warn the user if it fails, informing the knob and the freq tried.
> 
> Doesn't work for me:
> 
> ./perf test -v 19
> 19: Test software clock events have valid period values    :
> --- start ---
> mmap size 528384B
> mmap size 528384B
> All (0) samples have period value of 1!
> ---- end ----
> Test software clock events have valid period values: FAILED!
> 
> But this fixes it:
> 
> diff --git a/tools/perf/tests/sw-clock.c b/tools/perf/tests/sw-clock.c
> index 93a7139..6664a7c 100644
> --- a/tools/perf/tests/sw-clock.c
> +++ b/tools/perf/tests/sw-clock.c
> @@ -9,7 +9,7 @@
>  #include "util/cpumap.h"
>  #include "util/thread_map.h"
>  
> -#define NR_LOOPS  1000000
> +#define NR_LOOPS  10000000

Lower frequency, need to generate more noise, ugh. Adding that, but I
think this test needs to be reworked, Namhyung?

- Arnaldo

  parent reply	other threads:[~2013-11-12 14:41 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-11 20:22 [GIT PULL 00/10] perf/core improvements and fixes Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 01/10] perf ui tui progress: Don't force a refresh during progress update Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 02/10] perf evsel: Remove idx parm from constructor Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 03/10] perf record: Synthesize non-exec MMAP records when --data used Arnaldo Carvalho de Melo
2013-11-11 20:38   ` David Ahern
2013-11-11 21:02     ` Vince Weaver
2013-11-11 20:22 ` [PATCH 04/10] perf machine: Introduce synthesize_threads method out of open coded equivalent Arnaldo Carvalho de Melo
2013-11-11 20:40   ` David Ahern
2013-11-11 20:50     ` Ingo Molnar
2013-11-12 11:34       ` Arnaldo Carvalho de Melo
2013-11-12 12:44         ` Ingo Molnar
2013-11-12 21:57         ` [tip:perf/urgent] perf target: Shorten perf_target__ to target__ tip-bot for Arnaldo Carvalho de Melo
2014-01-12 18:33         ` [tip:perf/core] perf tools: Rename 'perf_record_opts' to ' record_opts tip-bot for Arnaldo Carvalho de Melo
2014-01-12 18:35         ` [tip:perf/core] perf report: Rename 'perf_report' to 'report' tip-bot for Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 05/10] perf machine: Simplify synthesize_threads method Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 06/10] perf tools: Prevent condition that all sort keys are elided Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 07/10] perf record: Use correct return type for write() Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 08/10] perf record: Move existing write_output into helper function Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 09/10] perf tests: Check return of perf_evlist__open sw clock event period test Arnaldo Carvalho de Melo
2013-11-11 20:22 ` [PATCH 10/10] perf tests: Use lower sample_freq in " Arnaldo Carvalho de Melo
2013-11-11 20:45   ` David Ahern
2013-11-12  7:07   ` Adrian Hunter
2013-11-12  8:40     ` Namhyung Kim
2013-11-12 11:59       ` Arnaldo Carvalho de Melo
2013-11-12 14:41     ` Arnaldo Carvalho de Melo [this message]
2013-11-15  6:03       ` Namhyung Kim
2013-11-12 21:56     ` [tip:perf/urgent] perf tests: Compensate lower sample freq with longer test loop tip-bot for Adrian Hunter

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=20131112144101.GF11239@ghostprotocols.net \
    --to=acme@infradead.org \
    --cc=adrian.hunter@intel.com \
    --cc=dsahern@gmail.com \
    --cc=efault@gmx.de \
    --cc=eranian@google.com \
    --cc=fweisbec@gmail.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@kernel.org \
    --cc=namhyung@kernel.org \
    --cc=paulus@samba.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;
as well as URLs for NNTP newsgroup(s).