[PATCH v3 3/5] drm/i915/scaler: Compute scaling factors for pipe scaler

Nautiyal, Ankit K ankit.k.nautiyal at intel.com
Tue Jan 7 13:54:28 UTC 2025


On 1/6/2025 10:21 PM, Mitul Golani wrote:
> Compute scaling factors and scaler user for pipe scaler if
> particular scaler user is pipe scaler.
>
> --v2:
> - Update typo 'plane scaling' to 'pipe scaling'. [Ankit]
> - Remove FIXME tag which was added to check for pipe scaling
> factors as well. [Ankit]
> - Should be common hscale, vscale instead of local one to
> avoid garbage overwritten.
> - Typo in commit message 'perticular' to 'particular'.
>
> Signed-off-by: Mitul Golani <mitulkumar.ajitkumar.golani at intel.com>
> ---
>   drivers/gpu/drm/i915/display/skl_scaler.c | 34 ++++++++++++++++++-----
>   1 file changed, 27 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/display/skl_scaler.c b/drivers/gpu/drm/i915/display/skl_scaler.c
> index 4f838220f880..294493e5ee53 100644
> --- a/drivers/gpu/drm/i915/display/skl_scaler.c
> +++ b/drivers/gpu/drm/i915/display/skl_scaler.c
> @@ -320,6 +320,7 @@ static int intel_atomic_setup_scaler(struct intel_crtc_state *crtc_state,
>   					&crtc_state->scaler_state;
>   	struct drm_i915_private *dev_priv = to_i915(crtc->base.dev);
>   	u32 mode;
> +	int hscale, vscale;
>   
>   	if (*scaler_id < 0)
>   		*scaler_id = intel_allocate_scaler(scaler_state, crtc);
> @@ -368,15 +369,11 @@ static int intel_atomic_setup_scaler(struct intel_crtc_state *crtc_state,
>   		mode = SKL_PS_SCALER_MODE_DYN;
>   	}
>   
> -	/*
> -	 * FIXME: we should also check the scaler factors for pfit, so
> -	 * this shouldn't be tied directly to planes.
> -	 */
>   	if (plane_state && plane_state->hw.fb) {
>   		const struct drm_framebuffer *fb = plane_state->hw.fb;
>   		const struct drm_rect *src = &plane_state->uapi.src;
>   		const struct drm_rect *dst = &plane_state->uapi.dst;
> -		int hscale, vscale, max_vscale, max_hscale;
> +		int max_vscale, max_hscale;
>   
>   		/*
>   		 * FIXME: When two scalers are needed, but only one of
> @@ -426,11 +423,34 @@ static int intel_atomic_setup_scaler(struct intel_crtc_state *crtc_state,
>   
>   			return -EINVAL;
>   		}
> +	}
> +
> +	if (crtc_state->pch_pfit.enabled) {
> +		struct drm_rect src;
> +
> +		drm_rect_init(&src, 0, 0,
> +			      drm_rect_width(&crtc_state->pipe_src) << 16,
> +			      drm_rect_height(&crtc_state->pipe_src) << 16);
> +
> +		hscale = drm_rect_calc_hscale(&src, &crtc_state->pch_pfit.dst,
> +					      0, INT_MAX);
> +		vscale = drm_rect_calc_vscale(&src, &crtc_state->pch_pfit.dst,
> +					      0, INT_MAX);

We need to do similar to plane scaler case, to check if hscale/vscale 
are less than the max scaling values.

I think it would be better to make a function to get max_hscale and 
vscale for a platform.

With 420 encoding, there is different limit so need to account that too. 
e.g Bspec:50441


Regards,

Ankit



>   
> -		scaler_state->scalers[*scaler_id].hscale = hscale;
> -		scaler_state->scalers[*scaler_id].vscale = vscale;
> +		if (hscale < 0 || vscale < 0) {
> +			drm_dbg_kms(display->drm,
> +				    "Scaler %d doesn't support required pipe scaling\n",
> +				    *scaler_id);
> +			drm_rect_debug_print("src: ", &src, true);
> +			drm_rect_debug_print("dst: ", &crtc_state->pch_pfit.dst, false);
> +
> +			return -EINVAL;
> +		}
>   	}
>   
> +	scaler_state->scalers[*scaler_id].hscale = hscale;
> +	scaler_state->scalers[*scaler_id].vscale = vscale;
> +
>   	drm_dbg_kms(display->drm, "Attached scaler id %u.%u to %s:%d\n",
>   		    crtc->pipe, *scaler_id, name, idx);
>   	scaler_state->scalers[*scaler_id].mode = mode;


More information about the Intel-gfx mailing list