Re: data corruption with current (maybe sis chipset related?)

From: Heiko Schaefer <hschaefer_at_fto.de>
Date: Thu, 8 May 2003 17:12:26 +0200 (CEST)
Hello Vallo,

> > > > options		DISABLE_PSE
> > > > options		DISABLE_PG_G
> >
> > ok. i have done my copying and checksumming orgy with these options in the
> > kernel - and it seems that i do not get any corruption anymore.
> > i will re-run my test once again now, just to be safe.
> >
> > after that: does it make sense to single out which of the two options is
> > the relevant one ?
> >
> > i'm also clueless what exactly these options do, and what exactly we've
> > ruled out now. could the lack of data-corruption just be some side-effect?
>
> I'm also forced to use these options on my P4 workstation, otherwise
> every parallel buildworld over -j4 gave me gcc internal errors. Can
> you please give your testing procedure, as I have two 120GB ATA
> disks and plenty of free space. The system I'll be testing on is
> 2x2400+ MP system with 1G of ECC memory.

here's what i do:

i have one 30gb disk and one 60gb disk (both ide, one is udma66, the other
udma100), i'll call them A and B.

A is filled up with lots of files of approx 5mb size. i've had them
checksummed in the beginning (i use 'cfv', which can be found in ports -
but i'm sure any checksumming program will do the trick)

i then copy all these files from A onto B ("cp -Rp"), twofold (into two
subdirs), just to move a lot of data. sometimes i did that in parallel,
sometimes serially - it doesn't seem to make any difference.

then i let cfv test the checksums of the two copied directories on B.
usually i got one or two checksum mismatches. occasionally in addition,
the filesystem on B was damaged (i noticed this when 'wc' told me
different size than ls -l for one of the mismatching files. fsck found
something to repair in that case).

regards,

Heiko

-- 
Free Software. Why put up with inferior code and antisocial corporations?
http://www.gnu.org/philosophy/why-free.html
Received on Thu May 08 2003 - 06:12:32 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:07 UTC