drm/amd/display: Respect DRM framebuffer info for video surfaces
authorNicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Wed, 13 Mar 2019 16:35:20 +0000 (12:35 -0400)
committerAlex Deucher <alexander.deucher@amd.com>
Thu, 21 Mar 2019 04:39:47 +0000 (23:39 -0500)
[Why]
Incorrect hardcoded assumptions are made regarding luma and chroma
alignment. The actual values set for the DRM framebuffer should be used
when programming the address.

[How]
Respect the given pitch for both luma and chroma planes - it's not like
we can force the alignment to anything else at this point anyway.

Use the FB offset for the chroma planes directly. DRM already
provides this to us so there's no need to calculate it manually.

While we don't actually use the chroma surface size parameters on Raven,
these should have technically been fb->width / 2 and fb->height / 2
since the chroma plane is half size of the luma plane for NV12.

Leave a TODO indicating that those should be set based on the actual
surface format instead since this is only correct for YUV420 formats.

Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c

index 0769b6e..6f9eed9 100644 (file)
@@ -2391,8 +2391,7 @@ fill_plane_tiling_attributes(struct amdgpu_device *adev,
                address->grph.addr.high_part = upper_32_bits(afb->address);
        } else {
                const struct drm_framebuffer *fb = &afb->base;
-               uint64_t awidth = ALIGN(fb->width, 64);
-               uint64_t chroma_addr = afb->address + awidth * fb->height;
+               uint64_t chroma_addr = afb->address + fb->offsets[1];
 
                address->type = PLN_ADDR_TYPE_VIDEO_PROGRESSIVE;
                address->video_progressive.luma_addr.low_part =
@@ -2468,7 +2467,6 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
                                         const struct amdgpu_framebuffer *amdgpu_fb)
 {
        uint64_t tiling_flags;
-       unsigned int awidth;
        const struct drm_framebuffer *fb = &amdgpu_fb->base;
        int ret = 0;
        struct drm_format_name_buf format_name;
@@ -2528,20 +2526,21 @@ static int fill_plane_attributes_from_fb(struct amdgpu_device *adev,
                plane_state->color_space = COLOR_SPACE_SRGB;
 
        } else {
-               awidth = ALIGN(fb->width, 64);
-
                plane_state->plane_size.video.luma_size.x = 0;
                plane_state->plane_size.video.luma_size.y = 0;
-               plane_state->plane_size.video.luma_size.width = awidth;
+               plane_state->plane_size.video.luma_size.width = fb->width;
                plane_state->plane_size.video.luma_size.height = fb->height;
-               /* TODO: unhardcode */
-               plane_state->plane_size.video.luma_pitch = awidth;
+               plane_state->plane_size.video.luma_pitch =
+                       fb->pitches[0] / fb->format->cpp[0];
 
                plane_state->plane_size.video.chroma_size.x = 0;
                plane_state->plane_size.video.chroma_size.y = 0;
-               plane_state->plane_size.video.chroma_size.width = awidth;
-               plane_state->plane_size.video.chroma_size.height = fb->height;
-               plane_state->plane_size.video.chroma_pitch = awidth / 2;
+               /* TODO: set these based on surface format */
+               plane_state->plane_size.video.chroma_size.width = fb->width / 2;
+               plane_state->plane_size.video.chroma_size.height = fb->height / 2;
+
+               plane_state->plane_size.video.chroma_pitch =
+                       fb->pitches[1] / fb->format->cpp[1];
 
                /* TODO: unhardcode */
                plane_state->color_space = COLOR_SPACE_YCBCR709;