Re: [PATCH v3 3/3] libata: Add support for queued DSM TRIM

[Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>]

Hello.

On 08/09/2013 08:49 AM, Marc C wrote:

> From: Marc Carino <marc.ceeeee@gmail.com>

> Some new SSDs support the queued version of the DSM TRIM command.
> Let the driver use the new command if supported.

> Signed-off-by: Marc Carino <marc.ceeeee@gmail.com>
> ---
>   drivers/ata/libata-scsi.c | 27 +++++++++++++++++++++------
>   1 file changed, 21 insertions(+), 6 deletions(-)

> diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
> index 83c0890..1605ffc 100644
> --- a/drivers/ata/libata-scsi.c
> +++ b/drivers/ata/libata-scsi.c
> @@ -3098,12 +3098,27 @@ static unsigned int ata_scsi_write_same_xlat(struct ata_queued_cmd *qc)
>   	buf = page_address(sg_page(scsi_sglist(scmd)));
>   	size = ata_set_lba_range_entries(buf, 512, block, n_block);
>
> -	tf->protocol = ATA_PROT_DMA;
> -	tf->hob_feature = 0;
> -	tf->feature = ATA_DSM_TRIM;
> -	tf->hob_nsect = (size / 512) >> 8;
> -	tf->nsect = size / 512;
> -	tf->command = ATA_CMD_DSM;
> +	if (ata_ncq_enabled(dev) &&
> +	    (dev->ncq_send_recv_cmds[ATA_LOG_NCQ_SEND_RECV_DSM_OFFSET] &
> +	     ATA_LOG_NCQ_SEND_RECV_DSM_TRIM)) {
> +		/* Newer devices support queued TRIM commands */
> +		tf->protocol = ATA_PROT_NCQ;
> +		tf->command = ATA_CMD_FPDMA_SEND;
> +		tf->hob_nsect = ATA_SUBCMD_FPDMA_SEND_DSM & 0x1f;
> +		tf->nsect = qc->tag << 3;
> +		tf->hob_feature = (size / 512) >> 8;
> +		tf->feature = size / 512;
> +
> +		qc->auxiliary = 1;
> +	} else {
> +		tf->protocol = ATA_PROT_DMA;
> +		tf->hob_feature = 0;
> +		tf->feature = ATA_DSM_TRIM;
> +		tf->hob_nsect = (size / 512) >> 8;
> +		tf->nsect = size / 512;
> +		tf->command = ATA_CMD_DSM;
> +	}
> +

    The remaining issue I see with this is that we don't check whether the 
controller/driver is capable of transferring the new commands to device, i.e. 
that it's FIS-based.

WBR, Sergei