[PATCH] Revert "drm/radeon: handle PCIe root ports with addressing limitations"

Alex Deucher alexdeucher at gmail.com
Wed Sep 16 13:08:39 UTC 2020


On Wed, Sep 16, 2020 at 2:32 AM Greg KH <gregkh at linuxfoundation.org> wrote:
>
> On Tue, Sep 15, 2020 at 02:46:07PM -0400, Alex Deucher wrote:
> > This change breaks tons of systems.
>
> Very vague :(

Screen corruption making the system unusable.

>
> This commit has also been merged for over a year, why the sudden
> problem now?
>

It was noticed by several people closer to when the change went in as
well.  If you notice, most of the bugs date back quite a while.  We
looked into it a bit at the time but couldn't determine the problem.
It only seems to affect really old chips (like 15-20 years old) which
makes it hard to reproduce if you don't have an old system.  There
were a couple of threads at the time, but nothing was resolved.  I was
able to find one of them:
https://lkml.org/lkml/2019/12/14/263

There were several new bugs filed which brought the issue back to my
attention recently.

> > This reverts commit 33b3ad3788aba846fc8b9a065fe2685a0b64f713.
>
> You mean "33b3ad3788ab ("drm/radeon: handle PCIe root ports with
> addressing limitations")"?
>
> That's the proper way to reference commits in changelogs please.  It's
> even documented that way...

When you revert a patch with git, that is what it does.  Maybe we
should fix git to change the formatting.

>
> >
> > Bug: https://bugzilla.kernel.org/show_bug.cgi?id=206973
> > Bug: https://bugzilla.kernel.org/show_bug.cgi?id=206697
> > Bug: https://bugzilla.kernel.org/show_bug.cgi?id=207763
> > Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1140
> > Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1287
> > Signed-off-by: Alex Deucher <alexander.deucher at amd.com>
> > Cc: stable at vger.kernel.org
> > Cc: Christoph Hellwig <hch at lst.de>
> > Cc: christian.koenig at amd.com
>
> Fixes: 33b3ad3788ab ("drm/radeon: handle PCIe root ports with addressing limitations")
>

Sure, I can add that, but it doesn't really fix it, it reverts it.
But point taken, it does fix the commit by removing it.

Alex

> as well?
>
> thanks,
>
> greg k-h


More information about the dri-devel mailing list