[Intel-gfx] [PATCH 1/2] Revert "drm/i915: Use a dmi quirk to skip a broken SDVO TV output."

[Eric Anholt]

On Thu, 18 Mar 2010 10:07:11 +0800, Zhenyu Wang <zhenyuw at linux.intel.com> wrote:
> On 2010.03.17 13:43:47 -0700, Eric Anholt wrote:
> > On Thu,  4 Mar 2010 16:07:13 +0800, Zhenyu Wang <zhenyuw at linux.intel.com> wrote:
> > > This reverts commit 6070a4a928f8c92b9fae7d6717ebbb05f425d6b2.
> > > 
> > > The quirk for this SDVO device on specific board will break same
> > > device on another OEM's output config. Instead multifunction SDVO
> > > support would fix this.
> > 
> > How would another OEM be impacted by a DMI quirk for IBM?
> 
> I might have made wrong statement of the reason to revert that quirk, sorry.
> What I mean is that it's not the fault of this SDVO chip, the chip has TV and
> VGA capability just fine.
> 
> But this quirk only isolates a very specific IBM system for its no TV output
> config, which is actually a hack for it on old code. You can see from comment
> #61 on fd.o bug #25787, that this quirk still failed on another IBM machine.
> Yeah, this hack won't break other OEM, just become useless with new SDVO
> multifunction code, with which it has been handled properly by creating TV and
> VGA connectors for the chip, and only VGA output could be detected on it.
> 
> How about this one?

Sounds much better.  We can apply it after the fixes to the code.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 197 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20100318/f26247b6/attachment.sig>