* Borislav Petkov <b...@alien8.de> wrote:
> On Mon, Jul 01, 2013 at 03:35:47PM -0700, Wedson Almeida Filho wrote:
> > On Mon, Jul 1, 2013 at 3:28 PM, Borislav Petkov <b...@alien8.de> wrote:
> > >
> > > perf stat --repeat 10 -a --sync --pre 'make -s clean; echo 1 >
> > > /proc/sys/vm/drop_caches' make -s -j64 bzImage
> >
> > How many CPUs do you have in your system? Maybe -j64 vs -jNUM_CPUs
> > affects your measurements as well.
>
> 8. But that shouldn't matter since I made the non-differing measurements
> two mails back with -j64.
>
> Also -j9, i.e. -j$(($NUM_CPUS+1)) gives "121.613217871 seconds time
> elapsed" because with -j9 the probability of some core not executing a
> make thread for whatever reason is higher than with -j64. But it is only
> as high as an additional 1s with this workload.
>
> I think with -j64 Ingo meant to saturate the scheduler to make sure
> there always are runnable threads more than cores available so that we
> can maximize the core utilization with threads running our workload.
Yeah - I didn't know your CPU count, -j64 is what I use.
Also, just in case it wasn't clear: thanks for the measurements - and I'd
be in favor of merging this patch if it shows any improvement or if
measurements lie within noise, because per asm review the change should be
a win.
Thanks,
Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
