Re: [PATCH 2.6.22] Add LED trigger to libata core

[Tejun Heo <htejun@gmail.com>]

Alan Cox wrote:
> On Mon, 19 Mar 2007 13:42:37 +0900
> Tejun Heo <htejun@gmail.com> wrote:
> 
>> Tony Vroon wrote:
>>> This duplicates the IDE core LED trigger in the libata core.
>>> I plan to use this by allowing PMU LED control on G5 towers. My test platform 
>>> is a PowerMac 7,3 (Dual G5 2.0GHz, June 2004) with a K2 (sata_svw) controller.
>> I think this fits better in libata-core.c::ata_qc_issue().  Can you move
>> it to there?
> 
> Gak. I'd rather it stayed out of ata_qc_issue() which is a critical path
> for performance.

The original place is in the critical path too.  It's just at the outer 
function which eventually calls ata_qc_issue() (the mapping is almost 
one to one).

> Our command issu is already too heavy and not all
> controllers have queueing to absorb that. How many controllers actually
> need this hook and can we not have ata_qc_issue_with_led() helpers for
> them ?

Our issue path is somewhat expensive due to SCSI -> ATA translation but 
I don't think it really matters on any modern cpu.  It can definitely 
hurt on embedded tho.  :-(

-- 
tejun