On Thu, 30 Nov 2000, Zdenek Kabelac wrote:
> > On Thu, Nov 30, 2000 at 10:00:34AM +0100, thus said Zdenek Kabelac:
> >
> > > test11 is hardly broken - DO NOT USE THIS KERNEL
> > > try test12pre3 which fixies some problem - though not all.
> >
> > Umm, what is broken in it? I have been running it 2 days without a
> > problem. Getting much less APIC errors than before, for the record. Should I
> > be scared of something evil like (eek) file corruption? What has changed in
> > test12?
>
> Please read linux kernel mailing list - several main kernel developers are
> reporting file corruption with test11 - with test12pre3 there
> are several fixies for this problem.
>
> This is what happend to me:
>
> $ tar zxf linux-2.4.0-test11.tar.gz
>
> gzip: stdin: invalid compressed data--format violated
> tar: Unexpected EOF in archive
> tar: Child returned status 1
> tar: Error exit delayed from previous errors
>
> After reboot this file was correct again...
I can testify to this. In the test I'm doing (dump/restore to copy a
filesystem), I run a diff after the copy and find many files that are different.
I narrowed down some stuff last night (I spent hours on it.. I'm beat). Here's
my setup:
hda: CDROM drive
hdc: Quantum 12GB (ATA/33 mode)
hde: IBM 20GB (ATA/33 mode)
hdg: IBM 46GB (ATA/100 in ATA/66 mode)
DMA mode is enabled on all.
All I have to do to hang the system (on 2.4.0-test11 and 2.2.17) is: (by the
way, I also made a uniprocessor kernel and this still would hang)
dd if=/dev/hdg of=/dev/null
Just before it does, the kernel prints either:
hdg: timeout waiting for DMA
or
ida_dmaproc: chipset_supported ide_dma_timeout func only: 14
I'm using the QQ BIOS. I have no card in slot 3 (I had one in during earlier
tests and once I moved it, the frequency of hangs dropped).
I checked my mobo and I don't have the Unisen voltage regulator.
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
