Re: [PATCH 1/4] libata: consolidate ata_dev_classify()

[Tejun Heo <tj@kernel.org>]

Hello, Hannes.

Sorry about the delay.

On Wed, Jul 30, 2014 at 09:55:08AM +0200, Hannes Reinecke wrote:
> ata_dev_classify() just uses the 'lbah' and 'lbam'
> fields from the taskfile, so we can as well use those
> as arguments and rip out the custom code from sas_ata.

I wonder whether it'd easier to just make sas code pass in
ata_taskfile instead?  The interface which takes three consecutive
u8's is kinda error-prone.

> --- a/drivers/scsi/aic94xx/aic94xx_task.c
> +++ b/drivers/scsi/aic94xx/aic94xx_task.c
> @@ -373,10 +373,10 @@ static int asd_build_ata_ascb(struct asd_ascb *ascb, struct sas_task *task,
>  
>  	if (unlikely(task->ata_task.device_control_reg_update))
>  		scb->header.opcode = CONTROL_ATA_DEV;
> -	else if (dev->sata_dev.command_set == ATA_COMMAND_SET)
> -		scb->header.opcode = INITIATE_ATA_TASK;
> -	else
> +	else if (dev->sata_dev.class == ATA_DEV_ATAPI)
>  		scb->header.opcode = INITIATE_ATAPI_TASK;
> +	else
> +		scb->header.opcode = INITIATE_ATA_TASK;

Are these changes covered by the patch description?  Looks like the
patch is mixing two separate logical changes.

Thanks.

-- 
tejun