Boot problem: "ata0: resetting devices"

[Florent DANIEL]

Le Jeudi 29 Mai 2003 07:05, Rich Morin a écrit :
> I recently purchased a new motherboard (ECS EliteGroup Mainboard P4S5A/DX+;
> AMI BIOS) and a matching CPU (Celeron, 478 pin, 2.1 GHz) from outpost.com.
> I stuffed these into an old chassis, adding a 500 W power supply and a pair
> of 512 MB DDR DIMMs.
>
> After walking through the configuration (accepting "Optimal Settings" for
> everything), I tried booting up the machine on an IDE drive that was loaded
> with FreeBSD and had been used most recently with a 200 MHz Pentium II.
>
> The boot sequence proceeded without problems for quite a while, checking
> devices, etc.  Then, however, it halted with a pair of nastygrams:
>
>    ad0: READ command timeout tag=0 serv=0 resetting
>    ata0: resetting devices
>
> I tried putting the ATA drive and the CDROM on separate cables, but that
> didn't seem to help.  My suspicion is that the drive is too slow for the
> ATA controller, but I don't see any way to compensate for this.  Help?

Hello,

Deactivate the DMA mode in the BIOS, and it it should boot. Which version of 
FreeBSD do you use ? What is the chipset used by the motherboard ? If you're 
on 4.8, with a SiS 651, here is a possible fix to use the DMA modes :

http://minilien.com/?BNI4EdKfXa

" >Fix:
Here is the patch to /sys/dev/ata/ata-dma.c

*** ata-dma.c.orig      Wed Oct  2 23:13:38 2002
--- ata-dma.c   Tue Feb 25 10:20:08 2003
***************
*** 504,509 ****
--- 504,510 ----
            ata_find_dev(parent, 0x06401039, 0) ||      /* SiS 640 */
            ata_find_dev(parent, 0x06451039, 0) ||      /* SiS 645 */
            ata_find_dev(parent, 0x06501039, 0) ||      /* SiS 650 */
+           ata_find_dev(parent, 0x06511039, 0) ||      /* SiS 651 */
            ata_find_dev(parent, 0x07301039, 0) ||      /* SiS 730 */
            ata_find_dev(parent, 0x07331039, 0) ||      /* SiS 733 */
            ata_find_dev(parent, 0x07351039, 0) ||      /* SiS 735 */

It worked fine for this 3 monthes on my machine.
I'm afraid the same problem may occur on SiS 648, 655, 746 and 755-based 
machines... "

Florent