Re: [Intel-gfx] [PATCH 4/4] drm/i915: Don't allocate extra ddb during async flip for DG2

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

On Tue, Jan 18, 2022 at 12:48:39PM +0200, Stanislav Lisovskiy wrote:
> In terms of async flip optimization we don't to allocate
> extra ddb space, so lets skip it.
> 
> v2: - Extracted min ddb async flip check to separate function
>       (Ville Syrjälä)
>     - Used this function to prevent false positive WARN
>       to be triggered(Ville Syrjälä)
> 
> Signed-off-by: Stanislav Lisovskiy <stanislav.lisovskiy@intel.com>
> ---
>  drivers/gpu/drm/i915/intel_pm.c | 34 ++++++++++++++++++++++++++-------
>  1 file changed, 27 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
> index 5d350ddc447f..4922c9108f08 100644
> --- a/drivers/gpu/drm/i915/intel_pm.c
> +++ b/drivers/gpu/drm/i915/intel_pm.c
> @@ -5118,6 +5118,12 @@ static bool icl_need_wm1_wa(struct drm_i915_private *i915,
>  	       (IS_DISPLAY_VER(i915, 12, 13) && plane_id == PLANE_CURSOR);
>  }
>  
> +static bool dg2_need_min_ddb(struct drm_i915_private *i915,
> +			     struct intel_crtc_state *crtc_state)
> +{
> +	return IS_DG2(i915) && crtc_state->uapi.async_flip;

Why are we using IS_DG2 here vs. DISPLAY>=13 for the wm0 only decision?

> +}
> +
>  static int
>  skl_allocate_plane_ddb(struct intel_atomic_state *state,
>  		       struct intel_crtc *crtc)
> @@ -5226,9 +5232,15 @@ skl_allocate_plane_ddb(struct intel_atomic_state *state,
>  			break;
>  
>  		rate = crtc_state->plane_data_rate[plane_id];
> -		extra = min_t(u16, alloc_size,
> -			      DIV64_U64_ROUND_UP(alloc_size * rate,
> -						 total_data_rate));
> +
> +		if (dg2_need_min_ddb(dev_priv, crtc_state)) {
> +			extra = 0;
> +		} else {
> +			extra = min_t(u16, alloc_size,
> +				      DIV64_U64_ROUND_UP(alloc_size * rate,
> +							 total_data_rate));
> +		}

Hmm. I wonder if we should just set rate=0 instead. That would
let the other planes pick up the now unused extra ddb space. Would 
also avoid having to skip the WARN since we'd still allocate the 
full ddb.

> +
>  		total[plane_id] = wm->wm[level].min_ddb_alloc + extra;
>  		alloc_size -= extra;
>  		total_data_rate -= rate;
> @@ -5237,14 +5249,22 @@ skl_allocate_plane_ddb(struct intel_atomic_state *state,
>  			break;
>  
>  		rate = crtc_state->uv_plane_data_rate[plane_id];
> -		extra = min_t(u16, alloc_size,
> -			      DIV64_U64_ROUND_UP(alloc_size * rate,
> -						 total_data_rate));
> +
> +		if (dg2_need_min_ddb(dev_priv, crtc_state)) {
> +			extra = 0;
> +		} else {
> +			extra = min_t(u16, alloc_size,
> +				      DIV64_U64_ROUND_UP(alloc_size * rate,
> +							 total_data_rate));
> +		}
> +
>  		uv_total[plane_id] = wm->uv_wm[level].min_ddb_alloc + extra;
>  		alloc_size -= extra;
>  		total_data_rate -= rate;
>  	}
> -	drm_WARN_ON(&dev_priv->drm, alloc_size != 0 || total_data_rate != 0);
> +
> +	if (!dg2_need_min_ddb(dev_priv, crtc_state))
> +		drm_WARN_ON(&dev_priv->drm, alloc_size != 0 || total_data_rate != 0);
>  
>  	/* Set the actual DDB start/end points for each plane */
>  	start = alloc->start;
> -- 
> 2.24.1.485.gad05a3d8e5

-- 
Ville Syrjälä
Intel