ALi SATA controller

[Søren Schmidt]

Bruce M. Simpson wrote:
> Hi,
>
> I have experienced the same regression since 6.2 was branched on an 
> ASUS amd64 Vintage-PE1 box.
>
> Juraj Lutter wrote:
>>
>> Feb 26 21:45:25 river kernel: atapci0: <AcerLabs M5229 UDMA133
>> controller> port 0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xff00-0xff0f at
>> device 31.0 on pci0
>> Feb 26 21:45:25 river kernel: ata0: <ATA channel 0> on atapci0
>> Feb 26 21:45:25 river kernel: ata1: <ATA channel 1> on atapci0
>> Feb 26 21:45:25 river kernel: atapci1: <AcerLabs M5287 SATA150
>> controller> port
>> 0xc400-0xc40f,0xc000-0xc007,0xb800-0xb80f,0xb400-0xb407,0xb000-0xb01f
>> mem 0xfe8ff400-0xfe8ff7ff irq 21 at device 31.1 on pci0
>> Feb 26 21:45:25 river kernel: atapci1: AHCI controller reset failure
>> Feb 26 21:45:25 river kernel: device_attach: atapci1 attach returned 6
> I'm sure I reported this to the list around 4th December 2006.
>
> My workaround was to install another SATA controller in the PCI-e slot 
> and use that instead for the root disk. I think CURRENT also suffers 
> from this problem. I think the last release which worked OK was 6.1.
>
The problem is that your BIOS registers the resources used for AHCI 
operation but apparently hasn't really enabled them.

Look for an option in the BIOS to turn on/off AHCI mode.

-Søren