Le 10/09/2024 à 21:44, Nicolas Dufresne a écrit :
Hi,
Le mardi 10 septembre 2024 à 14:10 +0000, Benjamin Gaignard a écrit :
Always get new destination buffer for reference frame because nothing
garanty the one set previously is still valid or unused.
Mind documenting here which tests got fixed with this change ?
Only one from chromium test suite:
https://chromium.googlesource.com/chromium/src/media/+/refs/heads/main/test/data/test-25fps.av1.ivf
Fluster AV1 score remains unchanged.
Fixes: 727a400686a2 ("media: verisilicon: Add Rockchip AV1 decoder")
Signed-off-by: Benjamin Gaignard <benjamin.gaignard@xxxxxxxxxxxxx>
---
.../media/platform/verisilicon/rockchip_vpu981_hw_av1_dec.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/media/platform/verisilicon/rockchip_vpu981_hw_av1_dec.c b/drivers/media/platform/verisilicon/rockchip_vpu981_hw_av1_dec.c
index 372dfcd0fcd9..2b9a1047479c 100644
--- a/drivers/media/platform/verisilicon/rockchip_vpu981_hw_av1_dec.c
+++ b/drivers/media/platform/verisilicon/rockchip_vpu981_hw_av1_dec.c
@@ -161,8 +161,7 @@ static int rockchip_vpu981_av1_dec_frame_ref(struct hantro_ctx *ctx,
av1_dec->frame_refs[i].timestamp = timestamp;
av1_dec->frame_refs[i].frame_type = frame->frame_type;
av1_dec->frame_refs[i].order_hint = frame->order_hint;
- if (!av1_dec->frame_refs[i].vb2_ref)
- av1_dec->frame_refs[i].vb2_ref = hantro_get_dst_buf(ctx);
+ av1_dec->frame_refs[i].vb2_ref = hantro_get_dst_buf(ctx);
Good catch, would still be nice to improve the commit message.
Reviewed-by: Nicolas Dufresne <nicolas.dufresne@xxxxxxxxxxxxx>
for (j = 0; j < V4L2_AV1_TOTAL_REFS_PER_FRAME; j++)
av1_dec->frame_refs[i].order_hints[j] = frame->order_hints[j];
