Re: [Intel-gfx] [RFC 1/3] drm/i915: remove !enabled handling from commit primary plane step

[Daniel Vetter <daniel@ffwll.ch>]

On Tue, Sep 09, 2014 at 06:58:47PM +0300, Ville Syrjälä wrote:
> On Tue, Sep 09, 2014 at 11:43:19AM -0300, Gustavo Padovan wrote:
> > From: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
> > 
> > The !crtc->enabled case will now be handled by the !visible code,
> > since the handling is basically the same.
> > 
> > Signed-off-by: Gustavo Padovan <gustavo.padovan@collabora.co.uk>
> > ---
> >  drivers/gpu/drm/i915/intel_display.c | 26 --------------------------
> >  1 file changed, 26 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> > index 5279b99..2ccf7c0 100644
> > --- a/drivers/gpu/drm/i915/intel_display.c
> > +++ b/drivers/gpu/drm/i915/intel_display.c
> > @@ -11837,32 +11837,6 @@ intel_commit_primary_plane(struct drm_plane *plane,
> >  	struct drm_rect *src = &state->src;
> >  	int ret;
> >  
> > -	/*
> > -	 * If the CRTC isn't enabled, we're just pinning the framebuffer,
> > -	 * updating the fb pointer, and returning without touching the
> > -	 * hardware.  This allows us to later do a drmModeSetCrtc with fb=-1 to
> > -	 * turn on the display with all planes setup as desired.
> > -	 */
> > -	if (!crtc->enabled) {
> > -		mutex_lock(&dev->struct_mutex);
> > -
> > -		/*
> > -		 * If we already called setplane while the crtc was disabled,
> > -		 * we may have an fb pinned; unpin it.
> > -		 */
> > -		if (plane->fb)
> > -			intel_unpin_fb_obj(old_obj);
> > -
> > -		i915_gem_track_fb(old_obj, obj,
> > -				  INTEL_FRONTBUFFER_PRIMARY(intel_crtc->pipe));
> > -
> > -		/* Pin and return without programming hardware */
> > -		ret = intel_pin_and_fence_fb_obj(dev, obj, NULL);
> > -		mutex_unlock(&dev->struct_mutex);
> > -
> > -		return ret;
> > -	}
> > -
> >  	intel_crtc_wait_for_pending_flips(crtc);
> 
> Yeah this should work just fine.
> 
> One difference between the code paths is the intel_crtc_wait_for_pending_flips()
> call, but since the crtc isn't enabled there can't be any pending flip.
> 
> The other difference is the pin vs. unpin order, but that shouldn't matter
> unless there's tons of other stuff pinned as well. It's not worth optimizing
> setplane calls on disabled CRTCs too much IMO.
> 
> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Queued for -next, thanks for the patch.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch