Re: IDE-DMA strangeness (another one)

Andre M. Hedrick (hedrick@astro.dyer.vanderbilt.edu)
Tue, 1 Dec 1998 22:10:22 -0600 (CST)


On Tue, 1 Dec 1998, Alex Buell wrote:

> On Tue, 1 Dec 1998, Rik van Riel wrote:
>
> > With this move I saw in improvement (hdparm -t) from 3
> > to 8.4 (!) MB/second... Now I wonder what kind of bug
> > in the Linux I/O subsystem could hold performance back
> > _that_ badly.
>
> Makes a lot of sense if your old IDE disk was PIO only and the newer IDE
> uses DMA.
>
> My machine exhibited the same speedup when I swapped around my old PIO IDE
> and my spankin' new UMDA IDE hard disks. (It's stuck in DMA though as
> PIIX3 chipset doesn't support UDMA - guess I'll have to buy a Promise
> UDMA33 controller someday!)

Look at ACARD's AEC6210UF aka SIIG CN2449 with BIOS it is a modified SCSI
chipset for all UDMA/DMA/PIO mixing. It has a FLASH 1/2 Meg BIOS for
upgrades.

Either way...........boot of the UDMA card by calling "pci=reverse".

Promise is a lot more code demanding..............

There is a new card IDE-EXPRESS by HighPoint Techologies. It is DVD/ATA3
and other junk..............

> hdparm -t reports 9.6MB/s throughput for /dev/hda although hdparm -T
> reports 33MB/s throughput, I don't quite believe that one though. :o)
>
> Cheers,
> Alex.
>
> ---
> /\_/\ Legalise cannabis now!
> ( o.o ) Grow some cannabis today!
> > ^ < Peace, Love, Unity and Respect to all.
>
> Check out http://www.tahallah.demon.co.uk
> Linux lo-pc3035a 2.1.130 #10 Mon Nov 30 10:49:50 EST 1998
> One Intel Pentium 166MHz processor, 66.36 total bogomips, 16M RAM
> System library 5.4.44
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/
>

Cheers,
Andre Hedrick
The IDE-FNG for Linux
The APC UPS Specialist for Linux

http://www.dyer.vanderbilt.edu/server/udma/
http://www.dyer.vanderbilt.edu/server/apcupsd/apcupsd-3.5.0.bin.tar.gz

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/