Comme=
nt # 111
on bug 91880<=
/a>
from John Bridgman
(In reply to Jan Ziak from comment #109)
> My question would be: Why isn't the _k patch alr=
eady in linux-git?
>=20
> I tested the patch on my R9 390, so the Hawaii-specific part of the pa=
tch
> works on at least one machine in the world outside of freedesktop.org.
>=20
> Or is there a reason to delay patch submission to linux-git?
>=20
> _k firmware files are already available in Gentoo Linux for example.
The patch is queued up for 4.8. It's the usual chicken-and-egg problem... i=
f we
can't get enough users testing a patch early in a kernel cycle then it ends=
up
having to wait for the next merge window.=20
(In reply to Chris Waters from com=
ment #110)
> (In reply to John Bridgman from comment #108)
> > Chris, are you using the -k firmware files and associated kernel =
patches,
> > updated initrd if needed, etc... ?
>=20
> I have not for two reasons.
>=20
> 1. As Christoffer mentions, the _k firmware only seems good for the 39=
0X
I think that is part of the "few different issues" point. The -k =
microcode may
not be the only change required but AFAIK it is one part of the solution.=20
In other words if the patch and new ucode don't fix the problem that's still
useful information, and it doesn't mean the new ucode isn't worth running.=
=20
>=20
> 2. I'm unfamiliar with the process of doing such and the steps for doi=
ng it
> are spread across 100+ comments with no real sign of what steps are the
> right ones.
Yep, that's a fair point. I was just trying to make sure we were collecting
good data. Thanks.