Re: [PATCH] perf: make perf.data more self-descriptive (v5)

[Stephane Eranian]

On Fri, Sep 23, 2011 at 4:09 PM, David Ahern <dsahern@xxxxxxxxx> wrote:
>
>
> On 09/23/2011 07:40 AM, Stephane Eranian wrote:
> >
> >
> > On Fri, Sep 23, 2011 at 3:36 PM, David Ahern <dsahern@xxxxxxxxx
> > <mailto:dsahern@xxxxxxxxx>> wrote:
> >
> >
> >
> > Â Â On 09/23/2011 03:04 AM, Stephane Eranian wrote:
> > Â Â > On Fri, Sep 23, 2011 at 10:03 AM, Pekka Enberg
> > Â Â <penberg@xxxxxxxxxxxxxx <mailto:penberg@xxxxxxxxxxxxxx>> wrote:
> > Â Â >> Hi Stephane!
> > Â Â >>
> > Â Â >> On Fri, Sep 23, 2011 at 10:31 AM, Stephane Eranian
> > Â Â <eranian@xxxxxxxxxx <mailto:eranian@xxxxxxxxxx>> wrote:
> > Â Â >>>> So how important is this information? The output is going to be
> > Â Â somewhat
> > Â Â >>>> awkward for very large CPU counts... :-)
> > Â Â >>>>
> > Â Â >>> It is useful to determine how CPUs share caches for instance.
> > Â Â >>> It can get large but large, but the meta-data header is not
> > Â Â printed by
> > Â Â >>> default, you need to request it with the -I option.
> > Â Â >>
> > Â Â >> Well, sure but it blocks rest of the interesting information too.
> > Â Â It seems to me
> > Â Â >> that the CPU information could be truncated to some sane limit by
> > Â Â default and
> > Â Â >> introduce a command line option for users that really want to see
> > Â Â all of it.
> > Â Â >>
> > Â Â > Ok, so here is a proposal:
> > Â Â > - reorder the info so one liners appear first
> > Â Â > - display the "truncated" info by default (no option)
> > Â Â > - truncated: numa topo, cpu topo, stop after 4 cpus/nodes, print msg
> >
> > Â Â Earlier I gave an example for a 2 socket, quad-core with hyperthreading
> > Â Â (16 cpus total):
> > Â Â Â Âhttps://lkml.org/lkml/2011/9/6/355
> > Â Â The information is repetitive. It would be better to devise a way to
> > Â Â reduce the repetition versus truncate the information.
> >
> >
> > I have modified the patch to NOT print the CPU, NUMA topology by
> > default (but mentioned they are available with the -I option). The
> > other bits of information are displayed systematically (no truncation).
> >
> > What ways would you propose to still print the info is a less-verbose
> > fashion?
> >
>
> for example, sibling cores:
> # CPU0 sibling cores Â: 0,2,4,6,8,10,12,14
>
> you don't need to print that info for CPUs 0,2,4,6,8,10,12,14. Just
> once. Maybe just:
>
> # CPU sibling cores Â: 0,2,4,6,8,10,12,14
>
> Similarly for the threads:
> # CPU sibling threads: 0,8
>
>
> That drops the output from 32 lines to 10.
>
> # CPU sibling cores Â: 0,2,4,6,8,10,12,14
> # CPU sibling cores Â: 1,3,5,7,9,11,13,15
> # CPU sibling threads: 0,8
> # CPU sibling threads: 1,9
> # CPU sibling threads: 2,10
> # CPU sibling threads: 3,11
> # CPU sibling threads: 4,12
> # CPU sibling threads: 5,13
> # CPU sibling threads: 6,14
> # CPU sibling threads: 7,15
>
Ok, I will look into this.
--
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/