Re: new ata_port_operations for .pmp_{read,write} ?

[Tejun Heo <htejun@gmail.com>]

Mark Lord wrote:
> No, the quickest solution for sata_mv, the one I apparently will now be
> using,
> is to just clone about 250 lines of reset/debouce/probe code from
> libata-core
> and change perhaps five lines of it to work around this issue plus some
> chipset errata.
> 
> What I was thinking of, before, is the SATA specification which details
> use of bits in the SCR to select a PMP port.  Pretty much exactly as
> this chipset does it, except in the SCR rather than a proprieary port.
> 
> But no big deal.  I can clone code and not bother you any more.
> In fact, some of the cloned code was already in sata_mv, and I removed
> it this past week in my local working copy.  I'll just restore that,
> along with another big blob so that we can select pm port where needed.
> 
> What a shame.

The order is somewhat reversed here and I can understand why you're
frustrated but I'm just trying to make things look right in long term,
so feel free to bother me. :-) For temporary solution, I'm okay either
way.  I'll clean things up later when the necessary core changes are made.

* Adding ->pmp_select or ->scr_pmp_rw or whatever callback to work
around the current problem.

* Duplicate code in sata_mv.

Whichever way you go, can you please mark where it's different and why?

Thanks.

-- 
tejun