Re: [PATCH 5/6] drm/vc4: Make sure that planes aren't scaled.

From: Daniel Vetter
Date: Fri Oct 23 2015 - 07:25:18 EST


On Fri, Oct 23, 2015 at 11:32:35AM +0100, Eric Anholt wrote:
> We would scan out the memory around them if an upscale was attempted,
> and would just scan out incorrectly for downscaling.
>
> Signed-off-by: Eric Anholt <eric@xxxxxxxxxx>
> ---
>
> It looks like, while modetest only wants to set scaling on overlay
> planes, one could do so on our primary/cursor planes if the fd was in
> universal planes mode.
>
> drivers/gpu/drm/vc4/vc4_plane.c | 10 ++++++++++
> 1 file changed, 10 insertions(+)
>
> diff --git a/drivers/gpu/drm/vc4/vc4_plane.c b/drivers/gpu/drm/vc4/vc4_plane.c
> index 2db5092..887f3ca 100644
> --- a/drivers/gpu/drm/vc4/vc4_plane.c
> +++ b/drivers/gpu/drm/vc4/vc4_plane.c
> @@ -157,6 +157,16 @@ static int vc4_plane_mode_set(struct drm_plane *plane,
> int crtc_w = state->crtc_w;
> int crtc_h = state->crtc_h;
>
> + if (state->crtc_w << 16 != state->src_w ||
> + state->crtc_h << 16 != state->src_h) {
> + /* We don't support scaling yet, which involves
> + * allocating the LBM memory for scaling temporary
> + * storage, and putting filter kernels in the HVS
> + * context.
> + */
> + return -EINVAL;
> + }

Just aside: There's drm_plane_helper_check_update for the common cases.
Unfortunately it suffers a bit from a bikeshed since it wants struct
drm_rect but atomic states don't give you that.
-Daniel

> +
> if (crtc_x < 0) {
> offset += drm_format_plane_cpp(fb->pixel_format, 0) * -crtc_x;
> crtc_w += crtc_x;
> --
> 2.6.1
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@xxxxxxxxxxxxxxxxxxxxx
> http://lists.freedesktop.org/mailman/listinfo/dri-devel

--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/