Re: mmotm 2008-09-08-18-32 uploaded

[Andrew Morton <akpm@linux-foundation.org>]

On Wed, 10 Sep 2008 10:01:42 +0900 KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:

> On Tue, 9 Sep 2008 10:37:41 -0700
> Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> > > Following is comparison with a mmtom based on rc4.
> > > ===
> > > [rc4mm1]
> > > Execl Throughput                           3004.4 lps   (29.6 secs, 3 samples)
> > > C Compiler Throughput                      1017.9 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (1 concurrent)               5726.3 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (8 concurrent)               1124.3 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (16 concurrent)               576.0 lpm   (60.0 secs, 3 samples)
> > > Dc: sqrt(2) to 99 decimal places         125446.5 lpm   (30.0 secs, 3 samples)
> > > 
> > > [rc5mm1]
> > > Execl Throughput                           3006.5 lps   (29.8 secs, 3 samples)
> > > C Compiler Throughput                      1006.7 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (1 concurrent)               4863.7 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (8 concurrent)                943.7 lpm   (60.0 secs, 3 samples)
> > > Shell Scripts (16 concurrent)               482.7 lpm   (60.0 secs, 3 samples)
> > > Dc: sqrt(2) to 99 decimal places         124804.9 lpm   (30.0 secs, 3 samples)
> > > ==
> > > 15% down in shell script test. Any idea ? (scheduler ?)
> > 
> > Dunno.  There is a largel number of debugging patches at the tail of
> > the series so they should be the first thing to eliminate.
> > 
> But I compare mmtom and mmtom, both includes debug series at the tail of series.

oh, sorry, I didn't read closely enough.

Yes, someone broke something.

> I'll revisit this later.

I'd suggest testing linux-next as a first step.