[Intel-gfx] hotplug CRT detection

[Adam Jackson]

On Thu, 2010-08-05 at 11:53 +1000, Dave Airlie wrote:
> So I've been reviewing the i915/ironlake CRT detect code and am a bit confused.
> 
> I though on the i945 and above that we had proper CRT hotplug
> detection with an IRQ, that didn't require any polling.
> 
> Now looking at the code when we do a CRT detect, the first thing we do
> is try the hotplug method, however on ILK/SNB this is crap and
> requires powering off the ADPA, which turns off the monitor, if that
> fails we fallback to DDC probing which is pretty non-destructive.
> 
> I get the feeling the whole hotplug detect code is pointless if the
> hardware is already meant to send us an IRQ when something happens,

Yeah, same.  The bit about GM45 needing to do it twice in a row is
probably the only exception.

- ajax
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 198 bytes
Desc: This is a digitally signed message part
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20100805/7062c638/attachment.sig>