Re: PATCH 2.3.26: kmalloc GFP_ZERO

Jes Sorensen (Jes.Sorensen@cern.ch)
09 Nov 1999 18:00:49 +0100


>>>>> "yodaiken" == yodaiken <yodaiken@chelm.cs.nmt.edu> writes:

yodaiken> On Tue, Nov 09, 1999 at 05:17:30PM +0100, Jes Sorensen
yodaiken> wrote:
yodaiken> PPC turns cache off and zeros during idle task. This is a
yodaiken> major performance win. We measured.
>> And Jakub just told us it wasn't a win under the same conditions
>> on the Sparc64.

yodaiken> It depends on how you do it, on the architecture, and on
yodaiken> what you measure. Lmbench is an invaluable hint, but it is
yodaiken> not definitive. If kernel compiles speed up by 20%, you
yodaiken> knpw something.

Now DaveM already posted good reason why lmbench was a bad bench for
this ... now did you in fact see a 20% speedup on kernel compiles? I
for some reason strongly doubt that.

Jes

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/