[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <169783679858.2217970.17247677172655395443.b4-ty@kernel.org>
Date: Fri, 20 Oct 2023 14:21:16 -0700
From: Namhyung Kim <namhyung@...nel.org>
To: adrian.hunter@...el.com, alexander.shishkin@...ux.intel.com,
irogers@...gle.com, peterz@...radead.org,
linux-kernel@...r.kernel.org, mark.rutland@....com,
Yang Jihong <yangjihong1@...wei.com>, mingo@...hat.com,
jolsa@...nel.org, linux-perf-users@...r.kernel.org, acme@...nel.org
Subject: Re: [PATCH] perf data: Increase RLIMIT_NOFILE limit when open too many files in perf_data__create_dir()
On Fri, 13 Oct 2023 07:59:45 +0000, Yang Jihong wrote:
> If using parallel threads to collect data, perf record needs at least 6 fds
> per CPU. (one for sys_perf_event_open, four for pipe msg and ack of the
> pipe, see record__thread_data_open_pipes(), and one for open perf.data.XXX)
> For an environment with more than 100 cores, if perf record uses both
> `-a` and `--threads` options, it is easy to exceed the upper limit of the
> file descriptor number, when we run out of them try to increase the limits.
>
> [...]
Applied to perf-tools-next, thanks!
Powered by blists - more mailing lists