ATA regression with 5.4-PRERELEASE

[Mars Trading]

I guess this is related to the ATA - HDD matching problem some users 
have been experiencing since 5.3-RELEASE - Errors look similar.

Briefly:
System unstable when ATA controller - HDD UDMA mismatched.
e.g.: UDMA 66 controller & UDMA 100 HDDrive.

Late model Seagate drives can be configured to use lower UDMA access 
using their uata100d.exe utility (used it, myself).  Don't know if 
other brands have a similar utility though.

Hope this gets fixed before 5.4-RELEASE.

Dennis

On Monday 14 March 2005 04:07, Oleg V. Nauman wrote:
>    I have some regression (or anti-regression?) with ATA on my
> home system (RELENG_5 from last Friday)
> I have had bad IDE 80-conductor cable, so I temporary switched
> two my ATA devices with old 40-conductor cable on the secondary
> IDE channel.
> Now I have something strange: my SAMSUNG SP0411N HDD on the
> secondary IDE channel detected again as UDMA100 device, but works
> with errors (as expected anyway in this situation).
> ATA controller on my system (from dmesg output):
> atapci0: <VIA 82C686B UDMA100 controller> port
> 0xd000-0xd00f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 7.1
> on pci0 and my devices:
> ad0: 19130MB <SAMSUNG SV2001H/QN200-05> [38869/16/63] at
> ata0-master UDMA100 ad1: 38166MB <ST340016A/7.18> [77545/16/63]
> at ata0-slave UDMA100 acd0: CDROM <SAMSUNG CD-ROM SC-152L/C100>
> at ata1-master PIO4 ad3: 38204MB <SAMSUNG SP0411N/TW100-11>
> [77622/16/63] at ata1-slave UDMA100 ad3: WARNING - READ_DMA UDMA
> ICRC error (retrying request) LBA=63 ad3: WARNING - READ_DMA UDMA
> ICRC error (retrying request) LBA=63 ad3: FAILURE - READ_DMA
> status=51<READY,DSC,ERROR> error=84<ICRC,ABORTED> LBA=63
>
> 	So: where is my UDMA33 or PIO4? :)

-- 
Mars Trading
9620 Kamagong St., Makati,
Manila, Philippines
Phone: (632) 8952241
FAX: (632) 8996124
email: marstrade at gmail.com