Re: VT82C686A corruption with 2.4.x

David Raufeisen (david@fortyoz.org)
Tue, 30 Jan 2001 12:53:18 -0800


bash-2.04# dmesg | grep -i udma
VP_IDE: VIA vt82c686a IDE UDMA66 controller on pci0:7.1
hda: 30015216 sectors (15368 MB) w/2048KiB Cache, CHS=1868/255/63, UDMA(66)
hdb: 80041248 sectors (40981 MB) w/2048KiB Cache, CHS=4982/255/63, UDMA(66)
hdc: ATAPI 52X CD-ROM drive, 192kB Cache, UDMA(33)

bash-2.04# uname -a
Linux prototype 2.4.1 #1 Tue Jan 30 01:45:38 PST 2001 i686 unknown

bash-2.04# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/hda2 14G 3.7G 10G 26% /
/dev/hdb1 38G 20G 19G 51% /storage

I dunno what you wanted to do by:

> dd if=/dev/hda4 of=/tmp/testing.img bs=1024M count=2k

Becuase it won't read 1024M into memory at once, says memory exhausted..

So I did..

bash-2.04# cd /storage
bash-2.04# dd if=/dev/hda2 of=testing.img bs=1024k count=2000
2000+0 records in
2000+0 records out
bash-2.04# ls -sh testing.img
2.0G testing.img
bash-2.04#

No problems here (nothing in dmesg, no crashing..) Course interactivity in X was total shit while dd was
running =\.

On Tuesday, 30 January 2001, at 11:51:58 (-0800),
David D.W. Downey wrote:

>
> Actually what rumors are you hearing?
>
> Right now I can tell you from personal experience that the VIA VT82C686A
> chipset is causing kernel deaths, corrupted data on my drives, and UDMA
> issues (meaning that when I enable the UDMA support the kernel
> CONSISTENTLY crashes.)
>
> This is all pre patch-2.4.1-pre11. I've not tested the new patch as of yet
> as I was in a car accident and have not felt well enough to mess with
> things. I will however be testing the new patch in about a half hour. The
> test bed will be the SMP box I've been talking about on the list, Red Hat
> 7.0 (only one that will install on the machine without dying instantly
> during installation) and using the KGCC rather than the GCC that comes
> with RH7.
>
> Supposedly there are a couple of other patches available to add as well,
> but I'm not sure where they are exactly. I just downloaded the patch that
> Voj gave (v3.20) for the VIA chipsets.
>
>
> Nicholas, give me a bit and I'll let you know what's going on with my
> tests here. I can consistently get the current kernel to die simply by
> running
>
> dd if=/dev/hda4 of=/tmp/testing.img bs=1024M count=2k
>
>
> TRy this on your box with the patch-2.4.1-pre11 added to the kernel source
> and let me know what you get. Maybe we can work on this together.
>
>
> David D.W. Downey
>

-- 
David Raufeisen <david@fortyoz.org>
Cell: (604) 818-3596
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/