Re: [RFC PATCH 1/1] proc: introduce /proc/<pid>/lbr_stack

From: Andi Kleen
Date: Fri Feb 27 2015 - 12:54:39 EST


On Fri, Feb 27, 2015 at 08:58:29AM +0100, Peter Zijlstra wrote:
> On Mon, Feb 23, 2015 at 09:44:48AM -0800, Andi Kleen wrote:
> > On Mon, Feb 23, 2015 at 05:49:57PM +0100, Peter Zijlstra wrote:
> > > On Mon, Feb 23, 2015 at 03:43:41AM +0000, kan.liang@xxxxxxxxx wrote:
> > > > From: Kan Liang <kan.liang@xxxxxxxxx>
> > > >
> > > > Haswell has a new feature that utilizes the existing Last Branch Record
> > > > facility to record call chains. It has been implemented in perf. The
> > > > call chains information is saved during perf event context.
> > > >
> > > > This patch exposes a /proc/<pid>/lbr_stack file that shows the saved LBR
> > > > call chain information.
> > >
> > > But why? I mean, this thing is only useful if you have a concurrently
> > > running perf record that selects the LBR-stack stuff.
> > >
> > > And if you have that, you might as well look at its output instead. Why
> > > add this unconditional proc file that doesn't function on its own?
> >
> > perf record doesn't show where you're currently blocked.
>
> Of course it does; look at perf inject -s.

Trace points don't support the LBR stack.

-Andi
--
ak@xxxxxxxxxxxxxxx -- Speaking for myself only
--
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/