[RFC PATCH] Re: KMS:RV635:Radeon 3650HD graphics driver broken on a laptop when connected to a docking station and external display

[Pasi Kärkkäinen]

On Mon, Jun 21, 2010 at 09:18:01PM +0300, Pasi Kärkkäinen wrote:
> > >>
> > >> I am not sure your patch is right, my guess is that devices field of
> > >> radeon connector structure btw the HDMI & DVI connector are different
> > >> and thus that drm_detect_hdmi_monitor is not call. I expect it's normal
> > >> for the device field to be different (my understanding being that HDMI
> > >> will have one more bit set than the DVI connector). Bottom line is
> > >> i think we want to discard non hdmi connector just in case we face
> > >> shared connector.
> > >>
> > >> Alex would have to comment on the device field test, maybe your atombios
> > >> is broken. Does it works if instead of removing
> > >> if (connector == list_connector)
> > >>        continue;
> > >> You remove :
> > >> if (radeon_connector->devices == list_radeon_connector->devices)
> > >>
> > >> Also can you print the radeon_connector->devices of all your connector
> > >> so we can compare them.
> > >
> > > The code was for systems with shared ddc lines and shared encoders,
> > > but it looks like your system may have a shared ddc line and (sort of)
> > > non-shared encoders.  I suspect the DVI port is using UNIPHY A or A+B
> > > (for dual link) and HDMI is using UNIPHY link B.  It's probably
> > > actually using router objects (gpio toggle to switch gpio routing
> > > between physical ports), however support for that is not implemented
> > > yet which is why we have the current code.  Can you send me a copy of
> > > your vbios so I can verify?
> > >
> > > To access your vbios (as root):
> > > cd /sys/bus/pci/device/<pci bus id>
> > > echo 1 > rom
> > > cat rom > /tmp/vbios.rom
> > > echo 0 > rom
> > 
> > Does the attached patch fix the issue?
> > 
> 
> Thanks for the patch. I'll try it when I'm back where the dock is.. will take a couple of days.
> 

Sorry it took a while.. I tried your patch now. 
It seems to fix the problem! 

When you send it upstream please also CC it for 2.6.33/2.6.34 stable branches?
Just hoping to get the fix into Fedora 13..

-- Pasi

> 
> 
> > From 557b452536c9390105539a264d342d963d71b087 Mon Sep 17 00:00:00 2001
> > From: Alex Deucher <alexdeucher at gmail.com>
> > Date: Mon, 21 Jun 2010 12:07:52 -0400
> > Subject: [PATCH] drm/radeon/kms: fix shared ddc handling
> > MIME-Version: 1.0
> > Content-Type: text/plain; charset=UTF-8
> > Content-Transfer-Encoding: 8bit
> > 
> > Connectors with a shared ddc line can be connected to different
> > encoders.
> > 
> > Reported by Pasi Kärkkäinen <pasik at iki.fi> on dri-devel
> > 
> > Signed-off-by: Alex Deucher <alexdeucher at gmail.com>
> > ---
> >  drivers/gpu/drm/radeon/radeon_connectors.c |    4 +++-
> >  1 files changed, 3 insertions(+), 1 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
> > index 0c7ccc6..f58f8bd 100644
> > --- a/drivers/gpu/drm/radeon/radeon_connectors.c
> > +++ b/drivers/gpu/drm/radeon/radeon_connectors.c
> > @@ -785,7 +785,9 @@ static enum drm_connector_status radeon_dvi_detect(struct drm_connector *connect
> >  					if (connector == list_connector)
> >  						continue;
> >  					list_radeon_connector = to_radeon_connector(list_connector);
> > -					if (radeon_connector->devices == list_radeon_connector->devices) {
> > +					if (list_radeon_connector->shared_ddc &&
> > +					    (list_radeon_connector->ddc_bus->rec.i2c_id ==
> > +					     radeon_connector->ddc_bus->rec.i2c_id)) {
> >  						if (drm_detect_hdmi_monitor(radeon_connector->edid)) {
> >  							if (connector->connector_type == DRM_MODE_CONNECTOR_DVID) {
> >  								kfree(radeon_connector->edid);
> > -- 
> > 1.7.0.1
> > 
>