All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexey Budankov <alexey.budankov@linux.intel.com>
To: Arnaldo Carvalho de Melo <acme@kernel.org>
Cc: Jiri Olsa <jolsa@redhat.com>, Namhyung Kim <namhyung@kernel.org>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andi Kleen <ak@linux.intel.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v1 00/15] Introduce threaded trace streaming for basic perf record operation
Date: Wed, 14 Oct 2020 15:15:53 +0300	[thread overview]
Message-ID: <3ea3ee09-9b59-f172-22a5-d5196498caf5@linux.intel.com> (raw)
In-Reply-To: <20201013162006.GJ1063281@kernel.org>


On 13.10.2020 19:20, Arnaldo Carvalho de Melo wrote:
> Em Mon, Oct 12, 2020 at 11:50:29AM +0300, Alexey Budankov escreveu:
>>
>> Patch set provides threaded trace streaming for base perf record
>> operation. Provided streaming mode (--threads) mitigates profiling
>> data losses and resolves scalability issues of serial and asynchronous
>> (--aio) trace streaming modes on multicore server systems. The patch
>> set is based on the prototype [1], [2] and the most closely relates
>> to mode 3) "mode that creates thread for every monitored memory map".
>>
>> The threaded mode executes one-to-one mapping of trace streaming threads
>> to mapped data buffers and streaming into per-CPU trace files located
>> at data directory. The data buffers and threads are affined to NUMA
>> nodes and monitored CPUs according to system topology. --cpu option
>> can be used to specify exact CPUs to be monitored.
>>
>> Basic analysis of data directories is provided for perf report mode.
>> Raw dump (-D) and aggregated reports are available for data directories,
>> still with no memory consumption optimizations. However data directories
>> collected with --compression-level option enabled can be analyzed with
>> little less memory because trace files are unmaped from tool process
>> memory after loading collected data.
>>
>> Provided streaming mode is available with Zstd compression/decompression
>> (--compression-level) and handling of external commands (--control).
>> AUX area tracing, related and derived modes like --snapshot or
>> --aux-sample are not enabled. --switch-output, --switch-output-event, 
>> --switch-max-files and --timestamp-filename options are not enabled.
> 
> Would be interesting to spell out what are the difficulties to have
> those options working with this threaded mode, as I expect that once
> this is all reviewed and tested we should switch to it by default, no?

At the moment I am not sure about this as the default mode. It all depends
on specifics of HW configuration and workload to be monitored and analyzed.
On middle and small sized systems --aio could still fit better from HW/OS
resource consumption perspective.

Initial intent to enable AUX area tracing faced the need to define some
(optimal?) way to store index data at data directory, thus left aside of
this first step to bring threaded trace streaming into Perf tool.
--switch-output-* and --timestamp-filename use cases are not yet clear
for data directories and thus look like features of the second order.

Addressing all that issues in a single patch set looks too much. Proper
way to have it all in is step-by-step. Also I should say that it is aside
of the scope of current Intel VTune specific needs.

Alexei

  reply	other threads:[~2020-10-14 12:16 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-12  8:50 [PATCH v1 00/15] Introduce threaded trace streaming for basic perf record operation Alexey Budankov
2020-10-12  8:53 ` [PATCH v1 01/15] perf session: introduce trace file path to be shown in raw trace dump Alexey Budankov
2020-10-12  8:54 ` [PATCH v1 02/15] perf report: output trace file name " Alexey Budankov
2020-10-12 16:01   ` Andi Kleen
2020-10-12 17:06     ` Alexey Budankov
2020-10-20 16:31       ` Alexey Budankov
2020-10-20 17:10         ` Alexey Budankov
2020-10-20 20:29           ` Andi Kleen
2020-10-13 19:54   ` Jiri Olsa
2020-10-13 21:23     ` Alexey Budankov
2020-10-12  8:55 ` [PATCH v1 03/15] perf data: open data directory in read access mode Alexey Budankov
2020-10-12 16:03   ` Andi Kleen
2020-10-12 16:52     ` Alexey Budankov
2020-10-13 16:22       ` Arnaldo Carvalho de Melo
2020-10-14 10:39         ` Alexey Budankov
2020-10-12  8:56 ` [PATCH v1 04/15] perf session: move reader object definition to header file Alexey Budankov
2020-10-12  8:57 ` [PATCH v1 05/15] perf session: introduce decompressor into trace reader object Alexey Budankov
2020-10-12 16:05   ` Andi Kleen
2020-10-12 16:50     ` Alexey Budankov
2020-10-12  8:58 ` [PATCH v1 06/15] perf session: load data directory into tool process memory Alexey Budankov
2020-10-12 16:09   ` Andi Kleen
2020-10-12 16:49     ` Alexey Budankov
2020-10-13 11:30       ` Alexey Budankov
2020-10-21  6:54         ` Namhyung Kim
2020-10-21 10:25           ` Alexey Budankov
2020-10-21 10:57             ` Namhyung Kim
2020-10-21 13:01               ` Alexey Budankov
2020-10-12  8:59 ` [PATCH v1 07/15] perf record: introduce trace file, compressor and stats in mmap object Alexey Budankov
2020-10-12  9:01 ` [PATCH v1 08/15] perf record: write trace data into mmap trace files Alexey Budankov
2020-10-14 10:52   ` Namhyung Kim
2020-10-14 12:09     ` Alexey Budankov
2020-10-21  7:34       ` Namhyung Kim
2020-10-21 10:24         ` Alexey Budankov
2020-10-21 10:51           ` Namhyung Kim
2020-10-12  9:02 ` [PATCH v1 09/15] perf record: introduce thread specific objects for trace streaming Alexey Budankov
2020-10-12  9:03 ` [PATCH v1 10/15] perf record: manage thread specific data array Alexey Budankov
2020-10-12  9:05 ` [PATCH v1 11/15] perf evlist: introduce evlist__ctlfd_update() to update ctl fd status Alexey Budankov
2020-10-12  9:05 ` [PATCH v1 12/15] perf record: introduce thread local variable for trace streaming Alexey Budankov
2020-10-12  9:10 ` [PATCH v1 13/15] perf record: stop threads in the end of " Alexey Budankov
2020-10-12  9:11 ` [PATCH v1 14/15] perf record: start threads in the beginning " Alexey Budankov
2020-10-12  9:13 ` [PATCH v1 15/15] perf record: introduce --threads command line option Alexey Budankov
2020-10-13 16:20 ` [PATCH v1 00/15] Introduce threaded trace streaming for basic perf record operation Arnaldo Carvalho de Melo
2020-10-14 12:15   ` Alexey Budankov [this message]
2020-10-14 17:27 ` Ingo Molnar
2020-10-15 10:35   ` Alexey Budankov

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=3ea3ee09-9b59-f172-22a5-d5196498caf5@linux.intel.com \
    --to=alexey.budankov@linux.intel.com \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=ak@linux.intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=jolsa@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=namhyung@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 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.