Re: [RFC 00/14] perf tools: Introduce new 'ftrace' command

[Namhyung Kim <namhyung@kernel.org>]

Hi Steve,

On Tue, 23 Apr 2013 11:58:01 -0400, Steven Rostedt wrote:
> On Tue, 2013-04-23 at 17:30 +0900, Namhyung Kim wrote:
>> Hello,
>> 
>> This patchset implements a front-end tool for kernel's ftrace.  It
>> uses function_graph tracer by default and normal function tracer is
>> also supported.  (Of course you need to enable those tracers in your
>> kernel first.)
>> 
>> NOTE: It's in very early stage, so may contain many rough edges.
>> 
>
> Very nice Namhyung, thanks for doing this. I did a quick run through of
> the patches and I have no complaints about them. I'm not sure how the
> others will feel about it.

Thanks!

>
> Also, have you given thought on how to execute both ftrace and the pmu
> counters? That is, to get a way to interleave the data?

I didn't think about it yet.  I just thought enabling event tracing in
ftrace and get the data along with the function tracing.  I'm not sure
how they interfere each other when enabled at the same time.

If that's not a valid concern, I think it's doable.


> I added a 'perf' clock to ftrace. I didn't see that being checked,
> although I didn't look too hard. If possible, you can use that, as well
> as the multi-buffers that are coming in 3.10.

Will look at it later.

>
> I would also like to add a system call to allow you to get to the ftrace
> data without the need for debugfs. I'm not sure if we should piggy back
> on the perf system call, or add a new one. I'm thinking we should add a
> new one so that it doesn't get too confusing.

Could you elaborate on it more?  I cannot see how it'll look like..

>
> Hope others have comments,

Thanks for your kind review.
Namhyung