Re: CF flash PATA on libata failure to attach

[Mark Lord <liml@rtr.ca>]

Andrew Hall wrote:
>..
>> Signed-off-by:  Mark Lord <mlord@pobox.com>
>> ---
>> --- linux/drivers/ata/ata_piix.c.orig	2007-06-10 18:58:27.000000000
>> -0400
>> +++ linux/drivers/ata/ata_piix.c	2007-06-28 21:09:04.000000000 -0400
>> @@ -537,7 +537,7 @@
>>  		.flags		= PIIX_SATA_FLAGS | PIIX_FLAG_SCR |
>>  				  PIIX_FLAG_AHCI,
>>  		.pio_mask	= 0x1f,	/* pio0-4 */
>> -		.mwdma_mask	= 0x07, /* mwdma0-2 */
>> +		.mwdma_mask	= 0x00, /* mwdma0-2 */
>>  		.udma_mask	= 0x7f,	/* udma0-6 */
>>  		.port_ops	= &piix_sata_ops,
>>  	},
> 
> That worked a treat! CF is flagged as PIO4 and HDD is now UDMA. I can't tell
> you how grateful I am that you were able to point out the fix / modification
> but I can certainly now sleep a bit easier.
> 
> Furthermore, if there is anything else I can do or would like me to test
> while we have this type of hardware, please let me know.
> 
> Thanks again, Mark..

You can certainly also thank Tejun and Jeff,
for making libata so easy to tune with a one-liner liner like this!

Per my other email -- did you try the legacy IDE driver
instead of libata?  Can you provide a boot log from that for Tejun?

Cheers!