Re: [PATCH 2/6] drm/i915: Don't use intel_crtc->new_config in pll calculation code

[Daniel Vetter <daniel@ffwll.ch>]

On Thu, Apr 02, 2015 at 02:47:57PM +0300, Ander Conselvan de Oliveira wrote:
> Move towards atomic by using the atomic state instead.
> 
> Signed-off-by: Ander Conselvan de Oliveira <ander.conselvan.de.oliveira@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_display.c | 15 +++++++++++----
>  1 file changed, 11 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index a6cd8c7..0b7ddee 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -11581,10 +11581,11 @@ intel_modeset_compute_config(struct drm_crtc *crtc,
>  	return intel_atomic_get_crtc_state(state, to_intel_crtc(crtc));;
>  }
>  
> -static int __intel_set_mode_setup_plls(struct drm_device *dev,
> +static int __intel_set_mode_setup_plls(struct drm_atomic_state *state,
>  				       unsigned modeset_pipes,
>  				       unsigned disable_pipes)

Bikeshed: The name of this function is a bit confusion - it's part of the
compute_config phase of modesets still, but not really named like that.
Same with the call to valleyview_modeset_global_pipes right above it.

The problem is also that this is separated from the other compute_config
code. Hence I think we can only apply some polish here once that's all
unified again and merged into the overall atomic_check function.
-Daniel

>  {
> +	struct drm_device *dev = state->dev;
>  	struct drm_i915_private *dev_priv = to_i915(dev);
>  	unsigned clear_pipes = modeset_pipes | disable_pipes;
>  	struct intel_crtc *intel_crtc;
> @@ -11598,9 +11599,15 @@ static int __intel_set_mode_setup_plls(struct drm_device *dev,
>  		goto done;
>  
>  	for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) {
> -		struct intel_crtc_state *state = intel_crtc->new_config;
> +		struct intel_crtc_state *crtc_state =
> +			intel_atomic_get_crtc_state(state, intel_crtc);
> +
> +		/* Modeset pipes should have a new state by now */
> +		if (WARN_ON(IS_ERR(crtc_state)))
> +			continue;
> +
>  		ret = dev_priv->display.crtc_compute_clock(intel_crtc,
> -							   state);
> +							   crtc_state);
>  		if (ret) {
>  			intel_shared_dpll_abort_config(dev_priv);
>  			goto done;
> @@ -11658,7 +11665,7 @@ static int __intel_set_mode(struct drm_crtc *crtc,
>  		prepare_pipes &= ~disable_pipes;
>  	}
>  
> -	ret = __intel_set_mode_setup_plls(dev, modeset_pipes, disable_pipes);
> +	ret = __intel_set_mode_setup_plls(state, modeset_pipes, disable_pipes);
>  	if (ret)
>  		goto done;
>  
> -- 
> 2.1.0
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx