Re: libata machine check on Alpha

[Albert Lee <albertcc@tw.ibm.com>]

Carlos Pardo wrote:
> The 3611 bridge does not support the ATAPI DMA direction bit in hardware.(software support only)  
> 
> The 3811 bridge supports ATAPI DMA direction bit in both software and hardware.
> 
> The 646 chip is ancient and he should upgrade.  The behavior of this device is unknown since this part was EOL'd years ago.
> 
> 

Hi Carlos,

Does this mean that only the 3611 bridge needs the DMA_DIR bit set by host software, while 3811
can detect the ATAPI DMA direction by hardware and doesn't need the DMA_DIR bit?
(I guess that 3811 detects the direction by looking into the SCSI CDB. If the SCSI command is
unknown to it, 3811 still needs the DMA_DIR bit set by software.)
Is there anyway that we could detect whether the DMA_DIR needs to be set by host software?
Is there anyway that we could detect whether the bridge is a 3611 or a 3811 chip automatically?

Per Jeff's comment, there is a document on t13 (http://www.t13.org/docs2003/e03131r0.pdf).
The documents reads:
"Eg. – Word 50, bit 13: Set to 1 for devices that support DMAIN bit in Features Register for Packet Command".
However, this looks more for the ATAPI device than for the bridge?
Also it looks not a t13 standard yet?
Can we use this word 50 to identify 3611/3811?

Thanks,

Albert