Re: [PATCHSET 06/11] prep for hotplug support, take 2

[Tejun Heo <htejun@gmail.com>]

Jeff Garzik wrote:
> At this point, I run out of steam.
> 
> After the minor changes are committed, I could pull <= patchset #5 into 
> #upstream immediately.

Great.  Once I get your responses and sorts out AHCI EH, I'll make new 
git branch you can pull from.  I can't wait to push those to #upstream. 
  Managing 120 stacked patches was a nightmare.  If I change something 
early in the series, even a simple/silly stuff, it causes a lot of 
rejects in later patches.

> The upcoming ata_link patchset looks quite nice; that is definitely the 
> direction I want to go.

Great.

> I've already given hotplug plenty of thought (i.e. general design 
> principles I like), so reviewing HP-related patchset should not take 
> very long.  I'll need to refresh my PM-related design thoughts a bit 
> before diving into that patchset, so that might take several days to get 
> to.

I'm still working on power applied together case because ahci is failing 
the first reset in a lot of cases.  Waiting for several secs before 
hotplug reset solved the problem for sata_sil and sata_sil24 but ahci's 
problem seems to be a different one - it doesn't seem to update TF for 
the first FIS after hotplug reset.  However, those changes are minor and 
wouldn't affect reviewing process, I think.

Thanks for reviewing and good night. :)

-- 
tejun