Re: [PATCH 3/4] drm/i915: dont save/restore VGA state for kms

[Daniel Vetter <daniel@ffwll.ch>]

On Fri, Jan 25, 2013 at 09:09:58PM +0200, Ville Syrjälä wrote:
> On Fri, Jan 25, 2013 at 05:53:21PM +0100, Daniel Vetter wrote:
> > The only thing we really care about that it is off. To do so, reuse
> > the recently created i915_redisable_vga function, which is already
> > used to put obnoxious firmware into check on lid reopening.
> > 
> > Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
> > ---
> >  drivers/gpu/drm/i915/i915_drv.h      |  1 +
> >  drivers/gpu/drm/i915/i915_suspend.c  | 48 +++++++++++++++++++-----------------
> >  drivers/gpu/drm/i915/intel_display.c |  2 +-
> >  3 files changed, 28 insertions(+), 23 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> > index 953060c..0ce7c8a 100644
> > --- a/drivers/gpu/drm/i915/i915_drv.h
> > +++ b/drivers/gpu/drm/i915/i915_drv.h
> > @@ -1788,6 +1788,7 @@ extern void intel_modeset_cleanup(struct drm_device *dev);
> >  extern int intel_modeset_vga_set_state(struct drm_device *dev, bool state);
> >  extern void intel_modeset_setup_hw_state(struct drm_device *dev,
> >  					 bool force_restore);
> > +extern void i915_redisable_vga(struct drm_device *dev);
> >  extern bool intel_fbc_enabled(struct drm_device *dev);
> >  extern void intel_disable_fbc(struct drm_device *dev);
> >  extern bool ironlake_set_drps(struct drm_device *dev, u8 val);
> > diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
> > index 056bd12..edcbfaf 100644
> > --- a/drivers/gpu/drm/i915/i915_suspend.c
> > +++ b/drivers/gpu/drm/i915/i915_suspend.c
> > @@ -69,6 +69,15 @@ static void i915_save_vga(struct drm_device *dev)
> >  	int i;
> >  	u16 cr_index, cr_data, st01;
> >  
> > +	/* VGA state */
> > +	dev_priv->regfile.saveVGA0 = I915_READ(VGA0);
> > +	dev_priv->regfile.saveVGA1 = I915_READ(VGA1);
> > +	dev_priv->regfile.saveVGA_PD = I915_READ(VGA_PD);
> 
> BTW these three don't seem to exist on PCH platforms at all. I guess
> gen <= 4 would be the most appropriate check since VLV doesn't have
> them either, but in this code it doesn't matter much if the check is
> for non-PCH instead. I'll include a patch for this in my 
> IS_DISPLAYREG() series.

We refuse to load the driver if modesetting isn't enabled and we're on
gen6+. And ums on ilk was only ever supported by rhel. So Imo we can just
punt on this as not a problem, now that the legacy vga state is only
saved/restored in ums mode.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch