[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <ec3bb670-9af9-e974-e260-8e22c7f7d116@linux.intel.com>
Date: Thu, 28 Feb 2019 23:19:44 +0300
From: Alexey Budankov <alexey.budankov@...ux.intel.com>
To: Arnaldo Carvalho de Melo <arnaldo.melo@...il.com>
Cc: Jiri Olsa <jolsa@...hat.com>, Namhyung Kim <namhyung@...nel.org>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
Peter Zijlstra <peterz@...radead.org>,
Ingo Molnar <mingo@...hat.com>,
Andi Kleen <ak@...ux.intel.com>,
linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v4 02/10] perf record: implement
-f,--mmap-flush=<threshold> option
On 28.02.2019 21:44, Arnaldo Carvalho de Melo wrote:
> Em Thu, Feb 28, 2019 at 12:00:41PM +0300, Alexey Budankov escreveu:
>>
>> Implemented -f,--mmap-flush option that specifies threshold to postpone
>> and/or trigger the move of data from mmaped kernel buffers to a storage.
>> The option can be used to avoid capturing every single byte of data into
>> the stored trace. The default option value is 1.
>
> Can you add something here explain more clearly what this means, and
> also on the tools/perf/Documentation/perf-record.txt file? Something
> like a paragraph explaining when is that there is a mmap flush normally,
> or rephrase what you wrote if you think you said that.
Will try to rephrase and may be provide some description of
cases that this option can specifically address.
Thanks,
Alexey
>
> - Arnaldo
>
>
<SNIP>
Powered by blists - more mailing lists