Re: [PATCHv2] perf tools: Fix pipe check regression in attr event callback

From: Stephane Eranian
Date: Thu Jun 05 2014 - 17:35:03 EST


On Thu, Jun 5, 2014 at 10:41 PM, Jiri Olsa <jolsa@xxxxxxxxxx> wrote:
> On Thu, Jun 05, 2014 at 05:21:02PM -0300, Arnaldo Carvalho de Melo wrote:
>> Em Thu, Jun 05, 2014 at 07:00:11PM +0200, Jiri Olsa escreveu:
>> > On Thu, Jun 05, 2014 at 06:06:41PM +0200, Stephane Eranian wrote:
>> > > Hi Jiri,
>> > >
>> > > Somehow, I thought you had written a fix to avoid the problem below.
>> > > But when I try with tip.git, the problem is still there.
>> > > Could you push your fix ASAP.
>> > >
>> > > Thanks.
>> > >
>> > > $ perf record -o - noploop 2 | perf inject -b | perf report -i -
>> > > # To display the perf.data header info, please use
>> > > --header/--header-only options.
>> > > #
>> > > noploop for 2 seconds
>> > > 0x1bd0 [0x28]: failed to process type: 9
>> >
>> > hum, I remember fixing another issue.. this one is
>> > separated one, please try attached patch.
>> >
>> > thanks,
>> > jirka
>> >
>> >
>> > ---
>> > The file factoring in builtin-inject.c object introduced regression
>> > in attr event callback. The commit is:
>> > 3406912 perf inject: Handle output file via perf_data_file object
>> >
>> > Following hunk reversed the logic:
>> > - if (!inject->pipe_output)
>> > + if (&inject->output.is_pipe)
>>
>> Why do we need this '&'? this will always evaluate to true, as it will
>> get the address of a variable, also adding a ! before it will just it
>> eval to false always, or am I missing something? :-)
>>
>> I think this should really be:
>>
>> if (!inject->output.is_pipe)
>>
>> No?
>
> aaarhg.. yes ;-) v2 attached
>
Had the same reaction as Arnaldo. Now, it makes more sense...

> thanks,
> jirka
>
> ---
> The file factoring in builtin-inject.c object introduced regression
> in attr event callback. The commit is:
> 3406912 perf inject: Handle output file via perf_data_file object
>
> Following hunk reversed the logic:
> - if (!inject->pipe_output)
> + if (&inject->output.is_pipe)
>
> putting it back, following example now works:
> $ perf record -o - kill | perf inject -b | perf report -i -
>
> Plus removing extra '&' (kudos to Arnaldo)
>
> Reported-by: Stephane Eranian <eranian@xxxxxxxxxx>
> Cc: Arnaldo Carvalho de Melo <acme@xxxxxxxxxx>
> Cc: Corey Ashford <cjashfor@xxxxxxxxxxxxxxxxxx>
> Cc: David Ahern <dsahern@xxxxxxxxx>
> Cc: Frederic Weisbecker <fweisbec@xxxxxxxxx>
> Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> Cc: Namhyung Kim <namhyung@xxxxxxxxxx>
> Cc: Paul Mackerras <paulus@xxxxxxxxx>
> Cc: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
> Signed-off-by: Jiri Olsa <jolsa@xxxxxxxxxx>
> ---
> tools/perf/builtin-inject.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/tools/perf/builtin-inject.c b/tools/perf/builtin-inject.c
> index 6a3af00..16c7c11 100644
> --- a/tools/perf/builtin-inject.c
> +++ b/tools/perf/builtin-inject.c
> @@ -72,7 +72,7 @@ static int perf_event__repipe_attr(struct perf_tool *tool,
> if (ret)
> return ret;
>
> - if (&inject->output.is_pipe)
> + if (!inject->output.is_pipe)
> return 0;
>
> return perf_event__repipe_synth(tool, event);
> --
> 1.8.3.1
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/