[Intel-gfx] [PATCH 08/19] drm/i915: Don't use encoder->new_crtc in intel_modeset_pipe_config()

[Ander Conselvan De Oliveira]

On Thu, 2015-03-19 at 00:44 +0000, Konduru, Chandra wrote:
> > -----Original Message-----
> > From: Conselvan De Oliveira, Ander
> > Sent: Friday, March 13, 2015 2:49 AM
> > To: intel-gfx at lists.freedesktop.org
> > Cc: Konduru, Chandra; Conselvan De Oliveira, Ander
> > Subject: [PATCH 08/19] drm/i915: Don't use encoder->new_crtc in
> > intel_modeset_pipe_config()
> > 
> > Move towards atomic by using the legacy modeset's drm_atomic_state instead.
> > 
> > v2: Move call to drm_atomic_add_affected_connectors() to
> >     intel_modeset_compute_config(). (Daniel)
> > 
> > Signed-off-by: Ander Conselvan de Oliveira
> > <ander.conselvan.de.oliveira at intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_display.c | 17 +++++++++++++++--
> >  1 file changed, 15 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_display.c
> > b/drivers/gpu/drm/i915/intel_display.c
> > index 6465f6d..1609628 100644
> > --- a/drivers/gpu/drm/i915/intel_display.c
> > +++ b/drivers/gpu/drm/i915/intel_display.c
> > @@ -10435,8 +10435,11 @@ intel_modeset_pipe_config(struct drm_crtc *crtc,
> > {
> >  	struct drm_device *dev = crtc->dev;
> >  	struct intel_encoder *encoder;
> > +	struct intel_connector *connector;
> > +	struct drm_connector_state *connector_state;
> >  	struct intel_crtc_state *pipe_config;
> >  	int plane_bpp, ret = -EINVAL;
> > +	int i;
> >  	bool retry = true;
> > 
> >  	if (!check_encoder_cloning(to_intel_crtc(crtc))) { @@ -10510,11
> > +10513,17 @@ encoder_retry:
> >  	 * adjust it according to limitations or connector properties, and also
> >  	 * a chance to reject the mode entirely.
> >  	 */
> > -	for_each_intel_encoder(dev, encoder) {
> > +	for (i = 0; i < state->num_connector; i++) {
> > +		connector = to_intel_connector(state->connectors[i]);
> > +		if (!connector)
> > +			continue;
> > 
> > -		if (&encoder->new_crtc->base != crtc)
> > +		connector_state = state->connector_states[i];
> > +		if (connector_state->crtc != crtc)
> >  			continue;
> > 
> > +		encoder = to_intel_encoder(connector_state->best_encoder);
> > +
> >  		if (!(encoder->compute_config(encoder, pipe_config))) {
> >  			DRM_DEBUG_KMS("Encoder config failure\n");
> >  			goto fail;
> > @@ -11238,6 +11247,10 @@ intel_modeset_compute_config(struct drm_crtc
> > *crtc,
> >  	struct intel_crtc *intel_crtc;
> >  	int ret = 0;
> > 
> > +	ret = drm_atomic_add_affected_connectors(state, crtc);
> 
> If the current transaction is started by DRM core, above operation 
> will be added by core, right?
> And when we move to full atomic_modeset, then above needs
> to be deleted?

That call is done in drm_atomic_helper_check_modeset(), so after the
conversion, we won't need this call there. But calling it twice doesn't
have any ill-effects, so we can leave it there until this is all sorted
out.

Ander


> 
> > +	if (ret)
> > +		return ERR_PTR(ret);
> > +
> >  	intel_modeset_affected_pipes(crtc, modeset_pipes,
> >  				     prepare_pipes, disable_pipes);
> > 
> > --
> > 2.1.0
>