Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Sender: Ingo Molnar <mingo.kernel.org@gmail.com>
Date:   Wed, 14 Oct 2020 19:27:00 +0200
From:   Ingo Molnar <mingo@kernel.org>
To:     Alexey Budankov <alexey.budankov@linux.intel.com>
Cc:     Arnaldo Carvalho de Melo <acme@kernel.org>,
        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
Message-ID: <20201014172700.GA3595702@gmail.com>
References: <810f3a69-0004-9dff-a911-b7ff97220ae0@linux.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <810f3a69-0004-9dff-a911-b7ff97220ae0@linux.intel.com>
Precedence: bulk


* Alexey Budankov <alexey.budankov@linux.intel.com> wrote:

> 
> 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.

Yay! This should really be the default trace capture model everywhere 
possible.

Can we do this for perf top too? It's really struggling with lots of cores.

If on a 64-core system I run just a moderately higher frequency 'perf top' 
of 1 kHz:

  perf top -e cycles -F 1000

perf stays stuck forever in 'Collecting samples...', and I also get a lot 
of:

  [548112.871089] Uhhuh. NMI received for unknown reason 31 on CPU 25.
  [548112.871089] Do you have a strange power saving mode enabled?

Thanks,

	Ingo