Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1462698 > unrolled thread

Re: [PATCH 2/2] perf script: Don't disable use_callchain if input is pipe

Started byHekuang <hekuang@huawei.com>
First post2016-08-15 12:30 +0200
Last post2016-08-15 18:30 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH 2/2] perf script: Don't disable use_callchain if input is  pipe Hekuang <hekuang@huawei.com> - 2016-08-15 12:30 +0200
    Re: [PATCH 2/2] perf script: Don't disable use_callchain if input is  pipe Arnaldo Carvalho de Melo <acme@kernel.org> - 2016-08-15 18:30 +0200

#1462698 — Re: [PATCH 2/2] perf script: Don't disable use_callchain if input is pipe

FromHekuang <hekuang@huawei.com>
Date2016-08-15 12:30 +0200
SubjectRe: [PATCH 2/2] perf script: Don't disable use_callchain if input is pipe
Message-ID<s6njA-4ly-9@gated-at.bofh.it>
ping.

在 2016/8/4 19:25, He Kuang 写道:
> Because perf data from pipe do not have a header with evsel attr, we
> should not check that and disable symbol_conf.use_callchain. Otherwise,
> perf script won't show callchains even if the data stream contains
> callchain.
>
> Before:
>    $ perf record -g -o - uname |perf script
>    Linux
>    [ perf record: Woken up 1 times to write data ]
>    [ perf record: Captured and wrote 0.000 MB - ]
>    uname  1828 182630.186578:  250000 cpu-clock:  ..b9499 setup_arg_pages
>    uname  1828 182630.186850:  250000 cpu-clock:  ..83b20 ___might_sleep
>    uname  1828 182630.187153:  250000 cpu-clock:  ..4b6be file_map_prot_ch
>    ...
>
> After:
>    $ perf record -g -o - uname |perf script
>    Linux
>    [ perf record: Woken up 1 times to write data ]
>    [ perf record: Captured and wrote 0.000 MB - ]
>    uname  1833 182675.927099:     250000 cpu-clock:
>                    ba5520 _raw_spin_lock+0xfe200040 ([kernel.kallsyms])
>                    389dd4 expand_downwards+0xfe200154 ([kernel.kallsyms])
>                    389f34 expand_stack+0xfe200024 ([kernel.kallsyms])
>                    3b957e setup_arg_pages+0xfe20019e ([kernel.kallsyms])
>                    40c80f load_elf_binary+0xfe20042f ([kernel.kallsyms])
>                    ...
>
> Signed-off-by: He Kuang <hekuang@huawei.com>
> ---
>   tools/perf/builtin-script.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
> index d47aef9..ec8df8f 100644
> --- a/tools/perf/builtin-script.c
> +++ b/tools/perf/builtin-script.c
> @@ -371,14 +371,16 @@ static int perf_session__check_output_opt(struct perf_session *session)
>   
>   	if (!no_callchain) {
>   		bool use_callchain = false;
> +		bool not_pipe = false;
>   
>   		evlist__for_each_entry(session->evlist, evsel) {
> +			not_pipe = true;
>   			if (evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
>   				use_callchain = true;
>   				break;
>   			}
>   		}
> -		if (!use_callchain)
> +		if (not_pipe && !use_callchain)
>   			symbol_conf.use_callchain = false;
>   	}
>   

[toc] | [next] | [standalone]


#1462975

FromArnaldo Carvalho de Melo <acme@kernel.org>
Date2016-08-15 18:30 +0200
Message-ID<s6sVY-7S0-13@gated-at.bofh.it>
In reply to#1462698
Em Mon, Aug 15, 2016 at 06:25:15PM +0800, Hekuang escreveu:
> ping.

Thanks, tested and applied both patches.

- Arnaldo
 
> 在 2016/8/4 19:25, He Kuang 写道:
> > Because perf data from pipe do not have a header with evsel attr, we
> > should not check that and disable symbol_conf.use_callchain. Otherwise,
> > perf script won't show callchains even if the data stream contains
> > callchain.
> > 
> > Before:
> >    $ perf record -g -o - uname |perf script
> >    Linux
> >    [ perf record: Woken up 1 times to write data ]
> >    [ perf record: Captured and wrote 0.000 MB - ]
> >    uname  1828 182630.186578:  250000 cpu-clock:  ..b9499 setup_arg_pages
> >    uname  1828 182630.186850:  250000 cpu-clock:  ..83b20 ___might_sleep
> >    uname  1828 182630.187153:  250000 cpu-clock:  ..4b6be file_map_prot_ch
> >    ...
> > 
> > After:
> >    $ perf record -g -o - uname |perf script
> >    Linux
> >    [ perf record: Woken up 1 times to write data ]
> >    [ perf record: Captured and wrote 0.000 MB - ]
> >    uname  1833 182675.927099:     250000 cpu-clock:
> >                    ba5520 _raw_spin_lock+0xfe200040 ([kernel.kallsyms])
> >                    389dd4 expand_downwards+0xfe200154 ([kernel.kallsyms])
> >                    389f34 expand_stack+0xfe200024 ([kernel.kallsyms])
> >                    3b957e setup_arg_pages+0xfe20019e ([kernel.kallsyms])
> >                    40c80f load_elf_binary+0xfe20042f ([kernel.kallsyms])
> >                    ...
> > 
> > Signed-off-by: He Kuang <hekuang@huawei.com>
> > ---
> >   tools/perf/builtin-script.c | 4 +++-
> >   1 file changed, 3 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tools/perf/builtin-script.c b/tools/perf/builtin-script.c
> > index d47aef9..ec8df8f 100644
> > --- a/tools/perf/builtin-script.c
> > +++ b/tools/perf/builtin-script.c
> > @@ -371,14 +371,16 @@ static int perf_session__check_output_opt(struct perf_session *session)
> >   	if (!no_callchain) {
> >   		bool use_callchain = false;
> > +		bool not_pipe = false;
> >   		evlist__for_each_entry(session->evlist, evsel) {
> > +			not_pipe = true;
> >   			if (evsel->attr.sample_type & PERF_SAMPLE_CALLCHAIN) {
> >   				use_callchain = true;
> >   				break;
> >   			}
> >   		}
> > -		if (!use_callchain)
> > +		if (not_pipe && !use_callchain)
> >   			symbol_conf.use_callchain = false;
> >   	}
> 

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web