Bugzilla Bug 173702

It's seen a few weeks of testing, all is quiet now, solves a few previously
unsolved bugs in 2.6.19.

I'd also like to request vanilla-sources-2.6.20.6 stable at the same time.  I
know it hasn't been in the tree very long, but it should be fine since
gentoo-sources-2.6.20-r5 is based on the same patchset.

Thanks!

2.6.20 doesnt boot on arm atm ;x

ppc64 done

You directly stabled it for sparc by carrying over -r4 keywords.
Anyway it's fine this time.

on x86:

$ uname -srp
Linux 2.6.20-gentoo-r5 AMD Athlon(tm) XP 2400+

no regressions so far...

Marked ppc stable.

Apart from the wakeup issues: A newer alsa-driver has to go stable at least on
x86, too.  I think about rc2.

(In reply to comment #7)
> Apart from the wakeup issues: A newer alsa-driver has to go stable at least on
> x86, too.  I think about rc2.
> 

Yes Diego said we want -rc3 for 2.6.20.

(In reply to comment #8)
> (In reply to comment #7)
> > Apart from the wakeup issues: A newer alsa-driver has to go stable at least on
> > x86, too.  I think about rc2.
> Yes Diego said we want -rc3 for 2.6.20.

 Ok.  But that still has open issues.  Can anyone verify if they should hold us
back?

(In reply to comment #7)
> Apart from the wakeup issues: A newer alsa-driver has to go stable at least on
> x86, too.  I think about rc2.
> 

Out of interest, why?

(In reply to comment #10)
> (In reply to comment #7)
> > Apart from the wakeup issues: A newer alsa-driver has to go stable at least on
> > x86, too.  I think about rc2.
> > 
> 
> Out of interest, why?

 Why what?  Newer alsa-driver, because current stable fails. rc2 because it has
no open bugs and succeeds on 2.6.20.  rc3 is ok if the issues are non-issues.

(In reply to comment #11)
>  Why what?  Newer alsa-driver, because current stable fails.

Fails in what way?

(In reply to comment #12)
> (In reply to comment #11)
> >  Why what?  Newer alsa-driver, because current stable fails.
> 
> Fails in what way?

 It is not intended to work with .20, according to Ticho, so we need a newer
one. 

2.6.20-r5 doesn't boot half the time for me (amd64). I think its still got some
issues with the MCP51 chipset; see
http://www.gossamer-threads.com/lists/linux/kernel/748576

Please file a bug for your issues.

We decided against switching to 2.6.20 simply because we are too close to the
release and don't want to spend the time validating a completely new kernel. 
Maybe next time.

^ above comment was in reference to 2007.0

arch teams, now that we aren't under pressure in terms of that release, please
hold off on this until -r6, which I'll release later today (hopefully)

(In reply to comment #17)
> ^ above comment was in reference to 2007.0
> 
> arch teams, now that we aren't under pressure in terms of that release, please
> hold off on this until -r6, which I'll release later today (hopefully)
> 

That will of course mean that -r6 will need to be tested for a month or so.

-r6 is out. theres no need for a months testing, the patch acceptance criteria
is very strict. This is entirely an incremental release (and minimal).

OK, then. 2.6.20-r6 looks ok from my side. Needs alsa-driver-1.0.14_rc2 or
later (I have tested _rc2, _rc2-r1 and _rc3).

x86 stable

Looks like Christian forgot to stabilize newer alsa-driver. Did so now,
1.0.14_rc2 is marked stable on x86.

In future please file a dedicated bug (with full info, i.e. the compile error)
when a stable tree package does not compile against a candidate kernel. I do
track these, and would have taken care of ALSA personally if a bug had been
filed earlier.

amd64 stable