Commit 920fffcc8912 ("drm/mediatek: update cursors by using async atomic
update") added support to async updates of cursors by using the new atomic
interface for that. Unfortunately, introduced two issues. The first one is
that since then, the drm_atomic_helper_async_commit triggers a WARNING due
current fb is not the new fb. The second one, is that we get a black screen
connecting the external display on Elm device and another WARNING due vblank
wait timed out.
Swap the fb in mtk_plane_atomic_async_update to fix both issues.
Fixes: 920fffcc8912 ("drm/mediatek: update cursors by using async atomic update")
Signed-off-by: Enric Balletbo i Serra <[email protected]>
---
drivers/gpu/drm/mediatek/mtk_drm_plane.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
index 914cc7619cd7..7eb10115e72a 100644
--- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c
+++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
@@ -116,6 +116,7 @@ static void mtk_plane_atomic_async_update(struct drm_plane *plane,
plane->state->src_h = new_state->src_h;
plane->state->src_w = new_state->src_w;
state->pending.async_dirty = true;
+ swap(plane->state->fb, new_state->fb);
mtk_drm_crtc_async_update(new_state->crtc, plane, new_state);
}
--
2.25.0
Hi,
On 13/2/20 13:01, Enric Balletbo i Serra wrote:
> Commit 920fffcc8912 ("drm/mediatek: update cursors by using async atomic
> update") added support to async updates of cursors by using the new atomic
> interface for that. Unfortunately, introduced two issues. The first one is
> that since then, the drm_atomic_helper_async_commit triggers a WARNING due
> current fb is not the new fb. The second one, is that we get a black screen
> connecting the external display on Elm device and another WARNING due vblank
> wait timed out.
>
> Swap the fb in mtk_plane_atomic_async_update to fix both issues.
>
> Fixes: 920fffcc8912 ("drm/mediatek: update cursors by using async atomic update")
> Signed-off-by: Enric Balletbo i Serra <[email protected]>
> ---
I just noticed this, which should fix the problem too, so you can ignore this patch.
https://patchwork.kernel.org/patch/11379571/
>
> drivers/gpu/drm/mediatek/mtk_drm_plane.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/gpu/drm/mediatek/mtk_drm_plane.c b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> index 914cc7619cd7..7eb10115e72a 100644
> --- a/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> +++ b/drivers/gpu/drm/mediatek/mtk_drm_plane.c
> @@ -116,6 +116,7 @@ static void mtk_plane_atomic_async_update(struct drm_plane *plane,
> plane->state->src_h = new_state->src_h;
> plane->state->src_w = new_state->src_w;
> state->pending.async_dirty = true;
> + swap(plane->state->fb, new_state->fb);
>
> mtk_drm_crtc_async_update(new_state->crtc, plane, new_state);
> }
>