At Thu, 13 Sep 2012 16:18:08 +0200, Daniel Vetter wrote:

On Thu, Sep 13, 2012 at 4:00 PM, Takashi Iwai tiwai@suse.de wrote:

...

But, this makes me wonder why this patch can't go into 3.6. It's a clear bug and the fix is trivial.

Daniel, isn't it better to pick the fix (for 3.6) to drm-intel-fixes, then merge back to drm-intel-next?

Well, on a quick read-through it didn't sound serious (just a bit of noise in dmesg and a few spurious reconfigs),

For the graphics side, yes. But for the sound driver side, you'll miss the jack-detection unplug event without this fix.

and we're rather late in the -rc cycle.

Yes, but a bug is a bug is a bug :) And the fix is two-liners. If there were no conflict, would you mind it? I don't think so.

Applying this on top of -fixes will cause a conflict, so didn't looked like I should bother.

Yeah, that's bad.

Also, I _want_ the patch for -next so that git blame shows the commit and not some random conflicting merge.

With parallel development, you can't always get straight lines, as you know well...

Takashi