[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <56A72D78.1040507@huawei.com>
Date: Tue, 26 Jan 2016 16:25:28 +0800
From: "Wangnan (F)" <wangnan0@...wei.com>
To: Alexei Starovoitov <ast@...nel.org>,
Arnaldo Carvalho de Melo <acme@...hat.com>
CC: Brendan Gregg <brendan.d.gregg@...il.com>,
Daniel Borkmann <daniel@...earbox.net>,
"David S. Miller" <davem@...emloft.net>,
He Kuang <hekuang@...wei.com>, Jiri Olsa <jolsa@...nel.org>,
Li Zefan <lizefan@...wei.com>,
Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
Namhyung Kim <namhyung@...nel.org>,
Peter Zijlstra <peterz@...radead.org>, <pi3orama@....com>,
Will Deacon <will.deacon@....com>,
<linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 50/54] perf record: Toggle overwrite ring buffer for reading
On 2016/1/25 17:56, Wang Nan wrote:
> Reading from a overwrite ring buffer is unrelible. perf_evsel__pause()
> should be called before reading from them.
>
> Toggel overwrite_evt_paused director after receiving done or switch
> output.
>
> Signed-off-by: Wang Nan <wangnan0@...wei.com>
> Signed-off-by: He Kuang <hekuang@...wei.com>
> Cc: Arnaldo Carvalho de Melo <acme@...hat.com>
> Cc: Jiri Olsa <jolsa@...nel.org>
> Cc: Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>
> Cc: Namhyung Kim <namhyung@...nel.org>
> Cc: Zefan Li <lizefan@...wei.com>
> Cc: pi3orama@....com
> ---
> tools/perf/builtin-record.c | 77 +++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 77 insertions(+)
[SNIP]
> +static void
> +record__toggle_overwrite_evsels(struct record *rec,
> + enum overwrite_evt_state state)
> +{
[SNIP]
> + rec->overwrite_evt_state = state;
> +
> + if (action == NONE)
> + return;
> +
> + evlist__for_each(evlist, pos) {
> + if (!pos->overwrite)
> + continue;
> + perf_evsel__pause(pos, action == PAUSE);
> + }
> +}
> +
This part is incorrect. We should pause ring buffers for each CPU
in a channel, not each evsel.
Already fixed at:
https://git.kernel.org/cgit/linux/kernel/git/pi3orama/linux.git/commit/?h=perf/overwrite&id=fe59d9c6621c60087ce7e6e269f2f15f152d6d71
Thank you.
Powered by blists - more mailing lists