On Fri, Aug 14, 2015 at 06:28:35PM +0200, Lukas Wunner wrote: > Originally by Seth Forshee <seth.forshee@xxxxxxxxxxxxx>, 2012-10-04: > Some dual graphics machines support muxing the DDC separately from the > display, so make use of this functionality when reading the EDID on the > inactive GPU. Also serialize drm_get_edid() with a mutex to avoid races > on the DDC mux state. > > Modified by Dave Airlie <airlied@xxxxxxxxx>, 2012-12-22: > I can't figure out why I didn't like this, but I rewrote this [...] to > lock/unlock the ddc lines [...]. I think I'd prefer something like that > otherwise the interface got really ugly. > > Modified by Lukas Wunner <lukas@xxxxxxxxx>, 2015-03-27: > Unlock DDC lines if drm_probe_ddc() fails. > > Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=88861 > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=61115 > Tested-by: Pierre Moreau <pierre.morrow@xxxxxxx> > [MBP 5,3 2009 nvidia 9400M + 9600M GT pre-retina] > Tested-by: Paul Hordiienko <pvt.gord@xxxxxxxxx> > [MBP 6,2 2010 intel ILK + nvidia GT216 pre-retina] > Tested-by: William Brown <william@xxxxxxxxxxxxxxxx> > [MBP 8,2 2011 intel SNB + amd turks pre-retina] > Tested-by: Lukas Wunner <lukas@xxxxxxxxx> > [MBP 9,1 2012 intel IVB + nvidia GK107 pre-retina] > Tested-by: Bruno Bierbaumer <bruno@xxxxxxxxxxxxxx> > [MBP 11,3 2013 intel HSW + nvidia GK107 retina -- work in progress] > > Cc: Seth Forshee <seth.forshee@xxxxxxxxxxxxx> > Cc: Dave Airlie <airlied@xxxxxxxxx> > Signed-off-by: Lukas Wunner <lukas@xxxxxxxxx> > --- > drivers/gpu/drm/drm_edid.c | 11 ++++++++++- > 1 file changed, 10 insertions(+), 1 deletion(-) > > diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c > index e6e05bb..cdb2fa1 100644 > --- a/drivers/gpu/drm/drm_edid.c > +++ b/drivers/gpu/drm/drm_edid.c > @@ -32,6 +32,7 @@ > #include <linux/hdmi.h> > #include <linux/i2c.h> > #include <linux/module.h> > +#include <linux/vga_switcheroo.h> > #include <drm/drmP.h> > #include <drm/drm_edid.h> > #include <drm/drm_displayid.h> > @@ -1377,13 +1378,21 @@ struct edid *drm_get_edid(struct drm_connector *connector, > struct i2c_adapter *adapter) > { > struct edid *edid; > + struct pci_dev *pdev = connector->dev->pdev; > > - if (!drm_probe_ddc(adapter)) > + vga_switcheroo_lock_ddc(pdev); > + > + if (!drm_probe_ddc(adapter)) { > + vga_switcheroo_unlock_ddc(pdev); > return NULL; > + } > > edid = drm_do_get_edid(connector, drm_do_probe_ddc_edid, adapter); > if (edid) > drm_get_displayid(connector, edid); > + > + vga_switcheroo_unlock_ddc(pdev); > + > return edid; > } > EXPORT_SYMBOL(drm_get_edid); I think this is backwards and it'd be more explicit (though I suspect slightly more work for this patch) to add a separate helper that does the VGA switcheroo wrapping rather than have this in drm_get_edid() where essentially every driver will go through the motions even if it doesn't remotely support switcheroo. Thierry
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
