* Re: [PATCH 4/5] sil24: add ATAPI support
[not found] <20051116080759.GD22807@htj.dyndns.org>
@ 2005-11-16 12:35 ` Jeff Garzik
2005-11-16 13:24 ` Tejun Heo
0 siblings, 1 reply; 2+ messages in thread
From: Jeff Garzik @ 2005-11-16 12:35 UTC (permalink / raw)
To: Tejun Heo; +Cc: linux-ide@vger.kernel.org
Tejun Heo wrote:
> + case ATA_PROT_ATAPI:
> + case ATA_PROT_ATAPI_DMA:
> + case ATA_PROT_ATAPI_NODATA:
> + prb = &cb->atapi.prb;
> + sge = cb->atapi.sge;
> + memset(cb->atapi.cdb, 0, 32);
> + memcpy(cb->atapi.cdb, qc->cdb, ap->cdb_len);
> +
> + if (unlikely((ap->cdb_len == 16) ^ pp->cdb16)) {
> + pp->cdb16 = ap->cdb_len == 16;
> + if (pp->cdb16)
> + writel(PORT_CS_CDB16, port + PORT_CTRL_STAT);
> + else
> + writel(PORT_CS_CDB16, port + PORT_CTRL_CLR);
> + }
NAK:
* you should set PORT_CS_CDB16 from ->dev_config(), not needlessly test
inside the fast path
* thus, no need to cache cdb16 flag. just tell the hardware once, and
then everybody agrees on the correct value.
> @@ -772,6 +802,9 @@ static int sil24_port_start(struct ata_p
>
> ap->private_data = pp;
>
> + /* default to 12byte CDB */
> + writel(PORT_CS_CDB16, port + PORT_CTRL_CLR);
> +
> return 0;
If you set it in ->dev_config(), this should go away.
Jeff
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH 4/5] sil24: add ATAPI support
2005-11-16 12:35 ` [PATCH 4/5] sil24: add ATAPI support Jeff Garzik
@ 2005-11-16 13:24 ` Tejun Heo
0 siblings, 0 replies; 2+ messages in thread
From: Tejun Heo @ 2005-11-16 13:24 UTC (permalink / raw)
To: Jeff Garzik; +Cc: linux-ide@vger.kernel.org
Jeff Garzik wrote:
> Tejun Heo wrote:
>
>> + case ATA_PROT_ATAPI:
>> + case ATA_PROT_ATAPI_DMA:
>> + case ATA_PROT_ATAPI_NODATA:
>> + prb = &cb->atapi.prb;
>> + sge = cb->atapi.sge;
>> + memset(cb->atapi.cdb, 0, 32);
>> + memcpy(cb->atapi.cdb, qc->cdb, ap->cdb_len);
>> +
>> + if (unlikely((ap->cdb_len == 16) ^ pp->cdb16)) {
>> + pp->cdb16 = ap->cdb_len == 16;
>> + if (pp->cdb16)
>> + writel(PORT_CS_CDB16, port + PORT_CTRL_STAT);
>> + else
>> + writel(PORT_CS_CDB16, port + PORT_CTRL_CLR);
>> + }
>
>
> NAK:
> * you should set PORT_CS_CDB16 from ->dev_config(), not needlessly test
> inside the fast path
> * thus, no need to cache cdb16 flag. just tell the hardware once, and
> then everybody agrees on the correct value.
>
Oh.. I didn't know about ->dev_config(). Thanks for pointing out.
>
>> @@ -772,6 +802,9 @@ static int sil24_port_start(struct ata_p
>>
>> ap->private_data = pp;
>>
>> + /* default to 12byte CDB */
>> + writel(PORT_CS_CDB16, port + PORT_CTRL_CLR);
>> +
>> return 0;
>
>
> If you set it in ->dev_config(), this should go away.
Sure.
--
tejun
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2005-11-16 13:24 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <20051116080759.GD22807@htj.dyndns.org>
2005-11-16 12:35 ` [PATCH 4/5] sil24: add ATAPI support Jeff Garzik
2005-11-16 13:24 ` Tejun Heo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).