This is an automated email from the git hooks/post-receive script.
Git pushed a commit to branch master
in repository ffmpeg.
The following commit(s) were added to refs/heads/master by this push:
new 96d82d90c3 avcodec/cuviddec: honor application-selected software output
96d82d90c3 is described below
commit 96d82d90c3f5b74afda99fc51acaf7c89f39bdf2
Author: Diego de Souza <[email protected]>
AuthorDate: Wed Aug 5 16:37:28 2026 +0200
Commit: Timo Rothenpieler <[email protected]>
CommitDate: Wed Aug 19 22:35:29 2026 +0000
avcodec/cuviddec: honor application-selected software output
The CUARRAY output option made the decoder override a software format
selected by get_format() with CUDA. This broke ordinary CUVID decoding
to software consumers.
Default output_format to auto and only override get_format() when the
user explicitly requests cuda/cuarray or enables zero_copy. Keep the
internal frames context in the corresponding hardware format and reject
opaque/non-opaque transitions during decoder reinitialization.
Signed-off-by: Diego de Souza <[email protected]>
---
libavcodec/cuviddec.c | 78 +++++++++++++++++++++++++++++++++------------------
1 file changed, 50 insertions(+), 28 deletions(-)
diff --git a/libavcodec/cuviddec.c b/libavcodec/cuviddec.c
index 3c8faae3c8..14744cf067 100644
--- a/libavcodec/cuviddec.c
+++ b/libavcodec/cuviddec.c
@@ -152,17 +152,18 @@ static int cuvid_get_requested_hw_format(AVCodecContext
*avctx, CuvidContext *ct
{
enum AVPixelFormat requested = ctx->output_format;
- if (requested == AV_PIX_FMT_NONE)
- requested = AV_PIX_FMT_CUDA;
-
- if (ctx->zero_copy && requested != AV_PIX_FMT_CUARRAY) {
- av_log(avctx, AV_LOG_WARNING,
- "zero_copy requires cuarray output format; "
- "overriding -output_format %s -> cuarray\n",
- av_get_pix_fmt_name(requested) ? av_get_pix_fmt_name(requested)
: "unknown");
+ if (ctx->zero_copy) {
+ if (requested != AV_PIX_FMT_NONE && requested != AV_PIX_FMT_CUARRAY)
+ av_log(avctx, AV_LOG_WARNING,
+ "zero_copy requires cuarray output format; "
+ "overriding -output_format %s -> cuarray\n",
+ av_get_pix_fmt_name(requested) ?
av_get_pix_fmt_name(requested) : "unknown");
requested = AV_PIX_FMT_CUARRAY;
}
+ if (requested == AV_PIX_FMT_NONE)
+ requested = AV_PIX_FMT_CUDA;
+
switch (requested) {
case AV_PIX_FMT_CUDA:
break;
@@ -185,6 +186,26 @@ static int cuvid_get_requested_hw_format(AVCodecContext
*avctx, CuvidContext *ct
return 0;
}
+static int cuvid_output_format_is_forced(const CuvidContext *ctx)
+{
+ return ctx->zero_copy || ctx->output_format != AV_PIX_FMT_NONE;
+}
+
+static enum AVPixelFormat cuvid_apply_forced_output_format(AVCodecContext
*avctx,
+ const CuvidContext
*ctx,
+ enum AVPixelFormat
selected,
+ enum AVPixelFormat
requested)
+{
+ if (cuvid_output_format_is_forced(ctx) && selected != requested) {
+ av_log(avctx, AV_LOG_VERBOSE,
+ "ff_get_format returned %s, overriding to explicitly requested
%s\n",
+ av_get_pix_fmt_name(selected), av_get_pix_fmt_name(requested));
+ selected = requested;
+ }
+
+ return selected;
+}
+
static void cuvid_prepare_format_list(enum AVPixelFormat *pix_fmts,
enum AVPixelFormat hw_format,
enum AVPixelFormat sw_format)
@@ -254,8 +275,9 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
CUVIDDECODECREATEINFO cuinfo;
int surface_fmt;
int chroma_444;
- int old_nb_surfaces, fifo_size_inc, fifo_size_mul = 1;
- enum AVPixelFormat requested_hw_format;
+ int old_nb_surfaces, old_opaque_output = ctx->opaque_output;
+ int fifo_size_inc, fifo_size_mul = 1;
+ enum AVPixelFormat requested_hw_format, internal_hw_format;
int old_width = avctx->width;
int old_height = avctx->height;
@@ -369,13 +391,8 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
return 0;
}
- if (surface_fmt != AV_PIX_FMT_CUDA && surface_fmt != AV_PIX_FMT_CUARRAY) {
- av_log(avctx, AV_LOG_VERBOSE,
- "ff_get_format returned %s, overriding to %s\n",
- av_get_pix_fmt_name(surface_fmt),
- av_get_pix_fmt_name(requested_hw_format));
- surface_fmt = requested_hw_format;
- }
+ surface_fmt = cuvid_apply_forced_output_format(avctx, ctx, surface_fmt,
+ requested_hw_format);
av_log(avctx, AV_LOG_VERBOSE, "Formats: Original: %s | HW: %s | SW: %s\n",
av_get_pix_fmt_name(avctx->pix_fmt),
@@ -384,6 +401,8 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
ctx->opaque_output = (surface_fmt == AV_PIX_FMT_CUARRAY);
avctx->pix_fmt = surface_fmt;
+ internal_hw_format = ctx->opaque_output ? AV_PIX_FMT_CUARRAY
+ : AV_PIX_FMT_CUDA;
if (ctx->opaque_output) {
switch (avctx->sw_pix_fmt) {
@@ -442,6 +461,13 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
avctx->framerate.den = format->frame_rate.denominator;
}
+ if (ctx->cudecoder && ctx->opaque_output != old_opaque_output) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Changing between opaque and non-opaque output is not
supported\n");
+ ctx->internal_error = AVERROR(EINVAL);
+ return 0;
+ }
+
if (ctx->cudecoder
&& avctx->coded_width == format->coded_width
&& avctx->coded_height == format->coded_height
@@ -462,13 +488,13 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
if (hwframe_ctx->pool && (
hwframe_ctx->width < avctx->width ||
hwframe_ctx->height < avctx->height ||
- (hwframe_ctx->format != AV_PIX_FMT_CUDA && hwframe_ctx->format !=
AV_PIX_FMT_CUARRAY) ||
+ hwframe_ctx->format != internal_hw_format ||
hwframe_ctx->sw_format != avctx->sw_pix_fmt)) {
av_log(avctx, AV_LOG_ERROR, "AVHWFramesContext is already initialized
with incompatible parameters\n");
av_log(avctx, AV_LOG_DEBUG, "width: %d <-> %d\n", hwframe_ctx->width,
avctx->width);
av_log(avctx, AV_LOG_DEBUG, "height: %d <-> %d\n",
hwframe_ctx->height, avctx->height);
av_log(avctx, AV_LOG_DEBUG, "format: %s <-> %s\n",
av_get_pix_fmt_name(hwframe_ctx->format),
- av_get_pix_fmt_name(avctx->pix_fmt));
+ av_get_pix_fmt_name(internal_hw_format));
av_log(avctx, AV_LOG_DEBUG, "sw_format: %s <-> %s\n",
av_get_pix_fmt_name(hwframe_ctx->sw_format),
av_get_pix_fmt_name(avctx->sw_pix_fmt));
ctx->internal_error = AVERROR(EINVAL);
@@ -679,7 +705,7 @@ static int CUDAAPI cuvid_handle_video_sequence(void
*opaque, CUVIDEOFORMAT* form
#endif
if (!hwframe_ctx->pool) {
- hwframe_ctx->format = avctx->pix_fmt;
+ hwframe_ctx->format = internal_hw_format;
hwframe_ctx->sw_format = avctx->sw_pix_fmt;
hwframe_ctx->width = avctx->width;
hwframe_ctx->height = avctx->height;
@@ -1433,13 +1459,8 @@ static av_cold int cuvid_decode_init(AVCodecContext
*avctx)
return ret;
}
- if (ret != AV_PIX_FMT_CUDA && ret != AV_PIX_FMT_CUARRAY) {
- av_log(avctx, AV_LOG_VERBOSE,
- "ff_get_format returned %s, overriding to %s\n",
- av_get_pix_fmt_name(ret),
- av_get_pix_fmt_name(requested_hw_format));
- ret = requested_hw_format;
- }
+ ret = cuvid_apply_forced_output_format(avctx, ctx, ret,
+ requested_hw_format);
ctx->opaque_output = (ret == AV_PIX_FMT_CUARRAY);
avctx->pix_fmt = ret;
@@ -1756,7 +1777,8 @@ static const AVOption options[] = {
{ "drop_second_field", "Drop second field when deinterlacing",
OFFSET(drop_second_field), AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VD },
{ "crop", "Crop (top)x(bottom)x(left)x(right)", OFFSET(crop_expr),
AV_OPT_TYPE_STRING, { .str = NULL }, 0, 0, VD },
{ "resize", "Resize (width)x(height)", OFFSET(resize_expr),
AV_OPT_TYPE_STRING, { .str = NULL }, 0, 0, VD },
- { "output_format", "Hardware output format", OFFSET(output_format),
AV_OPT_TYPE_INT, { .i64 = AV_PIX_FMT_CUDA }, 0, INT_MAX, VD, .unit =
"output_format" },
+ { "output_format", "Hardware output format", OFFSET(output_format),
AV_OPT_TYPE_INT, { .i64 = AV_PIX_FMT_NONE }, AV_PIX_FMT_NONE, INT_MAX, VD,
.unit = "output_format" },
+ { "auto", "Honor the application output format selection", 0,
AV_OPT_TYPE_CONST, { .i64 = AV_PIX_FMT_NONE }, 0, 0, VD, .unit =
"output_format" },
{ "cuda", "CUDA pitch-linear output", 0, AV_OPT_TYPE_CONST, {
.i64 = AV_PIX_FMT_CUDA }, 0, 0, VD, .unit = "output_format" },
#ifdef NVDEC_HAVE_OPAQUE_OUTPUT_SUPPORT
{ "cuarray", "CUDA block-linear opaque output", 0, AV_OPT_TYPE_CONST, {
.i64 = AV_PIX_FMT_CUARRAY }, 0, 0, VD, .unit = "output_format" },
--
To stop receiving notification emails like this one, please contact
[email protected].
_______________________________________________
ffmpeg-cvslog mailing list -- [email protected]
To unsubscribe send an email to [email protected]
