Re: [PATCH 04/10] perf, tools, report: Add processing for cycle histograms

[Andi Kleen]

On Mon, Jun 01, 2015 at 03:02:23PM +0200, Jiri Olsa wrote:
> On Tue, May 26, 2015 at 07:37:30PM +0200, Andi Kleen wrote:
> > > hum, so this is assuming that having cycles fort 1st entry
> > > means there'll be for the rest?
> > > Also in that case why is there the '!= cycles' check within
> > > addr_map_symbol__account_cycles ?
> > >
> > It means there might be. It's just a short cut. But rarely
> > branches may still have 0 cycles, so it still needs to be
> > checked later.
> > 
> > In theory it could miss a valid one if the first happened
> > to be zero, but that seems very unlikely.
> 
> so having 'bs->entries[0].flags.cycles' is the only way
> of knowing that we have the feature enabled?

Yes.

In theory we could add caps in sysfs like the PT code,
but that's not implemented currently.

-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/