Re: Time for a pre-patch-2.0.31-3?
Dr. Werner Fink (werner@suse.de)
Thu, 3 Jul 1997 13:25:10 +0200
> > And with your latest patch with pre-patch 2.0.31-2
> >
> > -------Sequential Output-------- ---Sequential Input-- --Random--
> > -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
> > Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
> > 100 2262 76.8 4841 30.6 1362 15.5 1445 52.7 1568 7.6 120.1 5.3
> >
> > Note the drastic fall in read performance. :( These results are consistent.
>
> I rememember the original port of the 2.1 buffer.c into 2.0.30 was inspired
> by this read performance problem. Leonard Zubkoff presented a patch for it
> at this time. Maybe he knows how to fix it. It *should* be fixed for 2.0.31.
>
> -Andi
I found only a answer on mail of Leonard Zubkoff in
http://www.linuxhq.com/patch/20-p0408.html ... smething similar is included.
Leonards fix on the old buffer code of 2.0.29 is useless in my changes
because there is no such code anymore :-)
The Linux 2.0 SMP Lockup (deadlock) patch of Leonard is a really interesting
one for smp users and it is NOT included in my changes.
Werner