[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d9389142-1065-0769-33b4-0f97e5d6eca4@linux.intel.com>
Date: Tue, 11 Aug 2020 11:12:39 +0300
From: Alexey Budankov <alexey.budankov@...ux.intel.com>
To: Peter Zijlstra <peterz@...radead.org>,
Andi Kleen <ak@...ux.intel.com>,
Alexander Shishkin <alexander.shishkin@...ux.intel.com>
Cc: Arnaldo Carvalho de Melo <acme@...hat.com>,
Ingo Molnar <mingo@...hat.com>, linux-kernel@...r.kernel.org,
Jiri Olsa <jolsa@...nel.org>, adrian.hunter@...el.com
Subject: Re: [PATCH 1/2] perf: Add closing sibling events' file descriptors
On 10.08.2020 23:36, Peter Zijlstra wrote:
> On Mon, Aug 10, 2020 at 07:45:18AM -0700, Andi Kleen wrote:
>
>> Unfortunately we're kind of stuck with the old NFILE=1024 default
>> even though it makes little sense on modern servers.
>
> Why can't that be changed? It seems to me all of userspace changes all
> the time; heck that system-doofus thing flushed 20+ years of sysadmin
> experience down the drain, just cause. Why can't we up a file limit?
>
If you asked me I would say that
Intel VTune HW micro architecture analysis would also be unblocked
by pushing RLIMIT_NOFILE limit beyond its current 1K default.
Currently rule of thumb for VTune users is to set the limit to
value == #cores X # events. The recommended value is 64K [1].
BR,
Alexei
[1] https://software.intel.com/content/www/us/en/develop/documentation/vtune-cookbook/top/configuration-recipes/profiling-hardware-without-sampling-drivers.html
Powered by blists - more mailing lists