Joe Feise wrote (ao):
> Sander wrote on 01/29/06 23:25:
> > Andrew Morton wrote (ao):
> >> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.16-rc1/2.6.16-rc1-mm4/
> >
> >> +reiser4-big-update-bug-fix-for-readpage-fix.patch
> >> +reiser4-warnings-cleanup.patch
> >> +reiser4-do-not-use-get_user_pages-and-do-not-check.patch
> >>
> >> reiser4 fixes and cleanups
> >
> > Jumping from 2.6.15-rc1-mm1 to 2.6.16-rc1-mm4.
> >
> > For me the Reiser4 slowdown[1] reported on the Reiserfs mailinglist are
> > gone with this kernel.
> >
> > [1] http://humilis.net/reiser4slowdown.html
>
>
> FWIW, I just ran these tests on 2.6.16-rc1-mm3:
> kernel: 2.6.16-rc1-mm3
> OS: Slackware 10.2 x86
> Compiler: gcc (GCC) 3.3.6
> Disk: SCSI on Adaptec 29160
> single disk, no raid/lvm/etc.
> # for i in `seq 4`; do echo "foo" > test && time vim +"s/foo/bar/" +"wq" test
> &>
> /dev/null ; done
>
> real 0m2.637s
> user 0m0.032s
> sys 0m0.040s
>
> real 0m2.139s
> user 0m0.016s
> sys 0m0.008s
>
> real 0m2.137s
> user 0m0.020s
> sys 0m0.016s
>
> real 0m2.137s
> user 0m0.020s
> sys 0m0.008s
>
> # strace -T -e sync,fsync ./reiser4-fsync
> sync() = 0 <2.308290>
> fsync(3) = 0 <0.071363>
So maybe it was gone already :-)
For some reason, it didn't hit everybody though, but it can very well be
that it was fixed already.
Btw, in my previous mail I forgot a big THANKS to all involved.
Kind regards, Sander
--
Humilis IT Services and Solutions
http://www.humilis.net
