Re: 2.1.111: IDE DMA disabled...BLAH...BLAH...

Ragnar Hojland Espinosa (
Thu, 30 Jul 1998 00:35:25 +0200 (MET DST)

On Wed, 29 Jul 1998, Mark Lord wrote:

> The very few cases I have ever seen reported of "corruption that goes
> away" when DMA is turned off, have thus far been totally bogus.

Bogus me? :) This deserves some little comment.

> I get a heck of a lot more email on this stuff than anyone else,
> your godness included. The pattern is always thus:

On DMA related corruption, I did report corruption. I did check the cables
for each HD drive to see if they were following the specs. My (UP) CPU isnt
overclocked. Fans are fine and clean. Nothing else is broken. hdparm
didnt solve anything. I did change the memory, and when I thought the
problem was gone.. wrong. Tried setting conservative values on the BIOS.
Even changed the box to make sure there wasnt some sort of weird power
supply problem.

> 6. Since the final exchanges were in private email (reducing noise
> linux-kernel), Linus never sees the resolution, and thus assumes
> that Gadi and I are shitty coders.

Well, thats a communication problem between you and Linus which shouldnt
be happening IMO.

> I have yet to see a case where turning off IDE DMA
> actually cures a disk corruption problem.

Here. I bet my head I can nuke my fs by running a kernel with DMA support,
and in fact I know when fs corruption is DMA-like here. If something is
wrong with this system, I'll be more than happy to know what. But right
now I still think it's the SiS chipset (a working GA586S2, anyone?) but
wont be able to know for sure until I change it.

Again, if anyone has any constructive suggestion, please email. I'm not
afraid of admitting I overlooked something (and actually don't know why
one would be)

____/|  Ragnar Hojland  (      Fingerprint  94C4B
\ o.O|                                                   2F0D27DE025BE2302C
 =(_)=  "Thou shalt not follow the NULL pointer for      104B78C56 B72F0822
   U     chaos and madness await thee at its end."       hkp://

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to Please read the FAQ at