Re: [PATCHSET 00/21] perf tools: Add support to accumulate hist periods (v8)

From: Namhyung Kim
Date: Sun Feb 09 2014 - 21:43:47 EST


Hi Rodrigo,

On Fri, 7 Feb 2014 03:31:10 +0000, Rodrigo Campos wrote:
> On Fri, Feb 07, 2014 at 10:35:02AM +0900, Namhyung Kim wrote:
>>
>> Currently the perf enables both of --call-graph and --children when it
>> finds callchains in the samples. While this is useful for TUI or GTK,
>> I'm not sure for stdio as it'd consume so much lines.
>>
>> It does not handle all kind of cases like event groups and annotations
>> yet, but I really want to release it and get reviews.
>>
>> You can also get this series on 'perf/cumulate-v8' branch in my tree at:
>>
>> git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git
>
>
> I've modified a little your example to link against two libraries dynamically
> and see different cases of how is self vs children attributed when running with
> "--sort dso" and it seems great.

Thanks for testing!

>
> I've tried simple examples just as: calling a lib1 function that calls lib2
> function which uses lot of CPU, calling lib1 function that uses lot of CPU and
> also calls lib2 func that uses lot of CPU. And things like that. Probably using
> one lib was enough (I first played with only one) as it's a different symbol
> from the main, but just in case some weird bug (was really easy) I tested with
> two :-)

Nice.

>
> I might be able to test with more real world scenarios on Saturday, although I'm
> sick now and I'm not sure how I will feel :S

Oh, sorry to hear that. Hope you get very well soon!

Thanks,
Namhyung
--
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/