lkml.org 
[lkml]   [1997]   [Nov]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Triton DMA
Alan Cox wrote:
>
> > >someone who gets filesystem corruption when copying large amounts of data
> > >from an ext2fs to an vfat filesystem. The problem was perfectly
> > >reproducible. After disabling the Triton DMA support the problem was gone.

Mmm... first I've heard of any such problem.

Note that Triton chipsets are the most common chipsets for Pentium
machines today, by far. As such, odds are good that anyone having
*any* problem with a system is likely to notice it on a triton board.

All problems I've heard of thus far have been tracked to loose cables,
cables too long (IDE spec says 18" max, many boxes use 24"..),
or PIO/DMA settings too fast for the drive.

> a) Some BIOSes turn DMA on when they shouldnt

Rare, and Linux makes its own decision anyway.

> b) DMA mode tests your RAM and especially cache harder so if you've got
> a problem you'll find it

Quite possible.

> c) Most such transactions are faster and IDE cables are low quality and
> without parity so a slightly iffy cable is death

DMA transactions are *exactly* the same speed/timing as PIO transactions
(thanks to the same cycle time of 120ns, plus pipelining/prefetch in the
chipsets for the PIO transfers). Except for Ultra33 DMA, which *does*
include 32-bit CRC checking (far superior to parity).
--
mlord@pobox.com
The Linux IDE guy

\
 
 \ /
  Last update: 2005-03-22 13:40    [W:0.053 / U:0.828 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site