Re: [PATCH 1/2] drm/atomic: Pass the full state to CRTC atomic_check

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Wed, Oct 28, 2020 at 01:32:21PM +0100, Maxime Ripard wrote:
> The current atomic helpers have either their object state being passed as
> an argument or the full atomic state.
> 
> The former is the pattern that was done at first, before switching to the
> latter for new hooks or when it was needed.
> 
> Let's start convert all the remaining helpers to provide a consistent
> interface, starting with the CRTC's atomic_check.
> 
> The conversion was done using the coccinelle script below,
> built tested on all the drivers and actually tested on vc4.
> 
> virtual report

?

<snip>
> @ depends on crtc_atomic_func @
> identifier crtc_atomic_func.func;
> expression E;
> type T;
> @@
> 
>  int func(...)
>  {
> 	...
> -	T state = E;
> +	T crtc_state = E;
>  	<+...
> -	state
> +	crtc_state
>  	...+>

>  }
> 
> @ depends on crtc_atomic_func @
> identifier crtc_atomic_func.func;
> type T;
> @@
> 
>  int func(...)
>  {
>  	...
> -	T state;
> +	T crtc_state;
>  	<+...
> -	state
> +	crtc_state
>  	...+>
>  }

These two seem a bit fuzzy. AFAICS 'state' could be any
kind of state given the constrainsts. Though I guess
the fact that this is the crtc .atomic_check() it's most
likely to either the crtc state or the whole atomic state.

Not sure what would be the best way to tighten this up.
Maybe a regex thing on the assignment? But I'm not sure
you can even do that on an expression.

Anyways, doesn't look like this went wrong anywhere, so
seems good enough for a onetime job.

<snip>
> diff --git a/drivers/gpu/drm/mxsfb/mxsfb_kms.c b/drivers/gpu/drm/mxsfb/mxsfb_kms.c
> index 956f631997f2..b0757f84a979 100644
> --- a/drivers/gpu/drm/mxsfb/mxsfb_kms.c
> +++ b/drivers/gpu/drm/mxsfb/mxsfb_kms.c
> @@ -269,17 +269,19 @@ static void mxsfb_crtc_mode_set_nofb(struct mxsfb_drm_private *mxsfb)
>  }
>  
>  static int mxsfb_crtc_atomic_check(struct drm_crtc *crtc,
> -				   struct drm_crtc_state *state)
> +				   struct drm_atomic_state *state)
>  {
> -	bool has_primary = state->plane_mask &
> +	struct drm_crtc_state *crtc_state = drm_atomic_get_new_crtc_state(state,
> +									  crtc);
> +	bool has_primary = crtc_state->plane_mask &
>  			   drm_plane_mask(crtc->primary);
>  
>  	/* The primary plane has to be enabled when the CRTC is active. */
> -	if (state->active && !has_primary)
> +	if (crtc_state->active && !has_primary)
>  		return -EINVAL;
>  
>  	/* TODO: Is this needed ? */
> -	return drm_atomic_add_affected_planes(state->state, crtc);
> +	return drm_atomic_add_affected_planes(crtc_state->state, crtc);

Could also s/crtc_state->state/state/ in various places.

But that could done as a followup as well.

Didn't spot any mistakes:
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

-- 
Ville Syrjälä
Intel
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel