[FFmpeg-devel] [PATCH 4/5] avcodec/mjpegdec: Only use receive_frame for SMVJPEG
Andreas Rheinhardt
andreas.rheinhardt at outlook.com
Wed Dec 7 21:02:49 EET 2022
Only one codec using mjpegdec.c actually creates multiple
frames from a single packet, namely SMVJPEG. The other can
use the ordinary decode callback just fine. This e.g. has
the advantage of confining the special SP5X/AMV code to sp5xdec.c.
This reverts most of commit e9a2a8777317d91af658f774c68442ac4aa726ec;
of course it is not a simple revert: Way too much has changed;
furthermore, outright reverting the sp5xdec.c changes would readd
a stack packet to sp5x_decode_frame() which is not desired.
In order to avoid this without modifying the given AVPacket,
a variant of ff_mjpeg_decode_frame() with explicit buf and size
parameters has been added.
Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt at outlook.com>
---
Weird that these receive_frame decoders have FF_CODEC_CAP_SETS_PKT_DTS
set, despite this flag only being used for decode_frame decoders.
libavcodec/jpeglsdec.c | 5 ++-
libavcodec/mjpegdec.c | 80 ++++++++++++++++++------------------------
libavcodec/mjpegdec.h | 11 +++---
libavcodec/sp5xdec.c | 32 ++++++++---------
4 files changed, 59 insertions(+), 69 deletions(-)
diff --git a/libavcodec/jpeglsdec.c b/libavcodec/jpeglsdec.c
index 2e6d018ea6..ec163b8964 100644
--- a/libavcodec/jpeglsdec.c
+++ b/libavcodec/jpeglsdec.c
@@ -558,8 +558,7 @@ const FFCodec ff_jpegls_decoder = {
.priv_data_size = sizeof(MJpegDecodeContext),
.init = ff_mjpeg_decode_init,
.close = ff_mjpeg_decode_end,
- FF_CODEC_RECEIVE_FRAME_CB(ff_mjpeg_receive_frame),
+ FF_CODEC_DECODE_CB(ff_mjpeg_decode_frame),
.p.capabilities = AV_CODEC_CAP_DR1,
- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP |
- FF_CODEC_CAP_SETS_PKT_DTS,
+ .caps_internal = FF_CODEC_CAP_INIT_CLEANUP,
};
diff --git a/libavcodec/mjpegdec.c b/libavcodec/mjpegdec.c
index 2abc42a082..aa1a6b5208 100644
--- a/libavcodec/mjpegdec.c
+++ b/libavcodec/mjpegdec.c
@@ -131,8 +131,6 @@ av_cold int ff_mjpeg_decode_init(AVCodecContext *avctx)
s->picture_ptr = s->picture;
}
- s->pkt = avctx->internal->in_pkt;
-
s->avctx = avctx;
ff_blockdsp_init(&s->bdsp);
ff_hpeldsp_init(&s->hdsp, avctx->flags);
@@ -2366,31 +2364,9 @@ static void smv_process_frame(AVCodecContext *avctx, AVFrame *frame)
av_frame_unref(s->smv_frame);
}
-static int mjpeg_get_packet(AVCodecContext *avctx)
-{
- MJpegDecodeContext *s = avctx->priv_data;
- int ret;
-
- av_packet_unref(s->pkt);
- ret = ff_decode_get_packet(avctx, s->pkt);
- if (ret < 0)
- return ret;
-
-#if CONFIG_SP5X_DECODER || CONFIG_AMV_DECODER
- if (avctx->codec_id == AV_CODEC_ID_SP5X ||
- avctx->codec_id == AV_CODEC_ID_AMV) {
- ret = ff_sp5x_process_packet(avctx, s->pkt);
- if (ret < 0)
- return ret;
- }
-#endif
-
- s->buf_size = s->pkt->size;
-
- return 0;
-}
-
-int ff_mjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame)
+int ff_mjpeg_decode_frame_from_buf(AVCodecContext *avctx, AVFrame *frame,
+ int *got_frame, const AVPacket *avpkt,
+ const uint8_t *buf, const int buf_size)
{
MJpegDecodeContext *s = avctx->priv_data;
const uint8_t *buf_end, *buf_ptr;
@@ -2405,6 +2381,8 @@ int ff_mjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame)
s->force_pal8 = 0;
+ s->buf_size = buf_size;
+
av_dict_free(&s->exif_metadata);
av_freep(&s->stereo3d);
s->adobe_transform = -1;
@@ -2412,12 +2390,9 @@ int ff_mjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame)
if (s->iccnum != 0)
reset_icc_profile(s);
- ret = mjpeg_get_packet(avctx);
- if (ret < 0)
- return ret;
redo_for_pal8:
- buf_ptr = s->pkt->data;
- buf_end = s->pkt->data + s->pkt->size;
+ buf_ptr = buf;
+ buf_end = buf + buf_size;
while (buf_ptr < buf_end) {
/* find start next marker */
start_code = ff_mjpeg_find_marker(s, &buf_ptr, buf_end,
@@ -2429,7 +2404,7 @@ redo_for_pal8:
} else if (unescaped_buf_size > INT_MAX / 8) {
av_log(avctx, AV_LOG_ERROR,
"MJPEG packet 0x%x too big (%d/%d), corrupt data?\n",
- start_code, unescaped_buf_size, s->pkt->size);
+ start_code, unescaped_buf_size, buf_size);
return AVERROR_INVALIDDATA;
}
av_log(avctx, AV_LOG_DEBUG, "marker=%x avail_size_in_buf=%"PTRDIFF_SPECIFIER"\n",
@@ -2568,7 +2543,6 @@ eoi_parser:
}
if (avctx->skip_frame == AVDISCARD_ALL) {
s->got_picture = 0;
- ret = AVERROR(EAGAIN);
goto the_end_no_picture;
}
if (s->avctx->hwaccel) {
@@ -2580,10 +2554,9 @@ eoi_parser:
}
if ((ret = av_frame_ref(frame, s->picture_ptr)) < 0)
return ret;
+ *got_frame = 1;
s->got_picture = 0;
- frame->pkt_dts = s->pkt->dts;
-
if (!s->lossless && avctx->debug & FF_DEBUG_QP) {
int qp = FFMAX3(s->qscale[0],
s->qscale[1],
@@ -2909,15 +2882,20 @@ the_end:
frame->crop_top = frame->height - avctx->height;
}
- ret = 0;
-
the_end_no_picture:
av_log(avctx, AV_LOG_DEBUG, "decode frame unused %"PTRDIFF_SPECIFIER" bytes\n",
buf_end - buf_ptr);
+ return buf_ptr - buf;
+}
- return ret;
+int ff_mjpeg_decode_frame(AVCodecContext *avctx, AVFrame *frame, int *got_frame,
+ AVPacket *avpkt)
+{
+ return ff_mjpeg_decode_frame_from_buf(avctx, frame, got_frame,
+ avpkt, avpkt->data, avpkt->size);
}
+
/* mxpeg may call the following function (with a blank MJpegDecodeContext)
* even without having called ff_mjpeg_decode_init(). */
av_cold int ff_mjpeg_decode_end(AVCodecContext *avctx)
@@ -2993,7 +2971,7 @@ const FFCodec ff_mjpeg_decoder = {
.priv_data_size = sizeof(MJpegDecodeContext),
.init = ff_mjpeg_decode_init,
.close = ff_mjpeg_decode_end,
- FF_CODEC_RECEIVE_FRAME_CB(ff_mjpeg_receive_frame),
+ FF_CODEC_DECODE_CB(ff_mjpeg_decode_frame),
.flush = decode_flush,
.p.capabilities = AV_CODEC_CAP_DR1,
.p.max_lowres = 3,
@@ -3001,7 +2979,6 @@ const FFCodec ff_mjpeg_decoder = {
.p.profiles = NULL_IF_CONFIG_SMALL(ff_mjpeg_profiles),
.caps_internal = FF_CODEC_CAP_INIT_CLEANUP |
FF_CODEC_CAP_SKIP_FRAME_FILL_PARAM |
- FF_CODEC_CAP_SETS_PKT_DTS |
FF_CODEC_CAP_ICC_PROFILES,
.hw_configs = (const AVCodecHWConfigInternal *const []) {
#if CONFIG_MJPEG_NVDEC_HWACCEL
@@ -3023,12 +3000,11 @@ const FFCodec ff_thp_decoder = {
.priv_data_size = sizeof(MJpegDecodeContext),
.init = ff_mjpeg_decode_init,
.close = ff_mjpeg_decode_end,
- FF_CODEC_RECEIVE_FRAME_CB(ff_mjpeg_receive_frame),
+ FF_CODEC_DECODE_CB(ff_mjpeg_decode_frame),
.flush = decode_flush,
.p.capabilities = AV_CODEC_CAP_DR1,
.p.max_lowres = 3,
- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP |
- FF_CODEC_CAP_SETS_PKT_DTS,
+ .caps_internal = FF_CODEC_CAP_INIT_CLEANUP,
};
#endif
@@ -3036,6 +3012,9 @@ const FFCodec ff_thp_decoder = {
static int smvjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame)
{
MJpegDecodeContext *s = avctx->priv_data;
+ AVPacket *const pkt = avctx->internal->in_pkt;
+ int64_t pkt_dts;
+ int got_frame = 0;
int ret;
if (s->smv_next_frame > 0) {
@@ -3048,10 +3027,21 @@ static int smvjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame)
return 0;
}
- ret = ff_mjpeg_receive_frame(avctx, frame);
+ ret = ff_decode_get_packet(avctx, pkt);
+ if (ret < 0)
+ return ret;
+
+ ret = ff_mjpeg_decode_frame(avctx, frame, &got_frame, pkt);
+ pkt_dts = pkt->dts;
+ av_packet_unref(pkt);
if (ret < 0)
return ret;
+ if (!got_frame)
+ return AVERROR(EAGAIN);
+
+ frame->pkt_dts = pkt_dts;
+
av_assert0(frame->buf[0]);
av_frame_unref(s->smv_frame);
ret = av_frame_ref(s->smv_frame, frame);
diff --git a/libavcodec/mjpegdec.h b/libavcodec/mjpegdec.h
index 2cb218902c..13c524d597 100644
--- a/libavcodec/mjpegdec.h
+++ b/libavcodec/mjpegdec.h
@@ -57,8 +57,6 @@ typedef struct MJpegDecodeContext {
GetBitContext gb;
int buf_size;
- AVPacket *pkt;
-
int start_code; /* current start code */
int buffer_size;
uint8_t *buffer;
@@ -173,7 +171,12 @@ int ff_mjpeg_build_vlc(VLC *vlc, const uint8_t *bits_table,
const uint8_t *val_table, int is_ac, void *logctx);
int ff_mjpeg_decode_init(AVCodecContext *avctx);
int ff_mjpeg_decode_end(AVCodecContext *avctx);
-int ff_mjpeg_receive_frame(AVCodecContext *avctx, AVFrame *frame);
+int ff_mjpeg_decode_frame(AVCodecContext *avctx,
+ AVFrame *frame, int *got_frame,
+ AVPacket *avpkt);
+int ff_mjpeg_decode_frame_from_buf(AVCodecContext *avctx,
+ AVFrame *frame, int *got_frame,
+ const AVPacket *avpkt, const uint8_t *buf, int buf_size);
int ff_mjpeg_decode_dqt(MJpegDecodeContext *s);
int ff_mjpeg_decode_dht(MJpegDecodeContext *s);
int ff_mjpeg_decode_sof(MJpegDecodeContext *s);
@@ -184,6 +187,4 @@ int ff_mjpeg_find_marker(MJpegDecodeContext *s,
const uint8_t **buf_ptr, const uint8_t *buf_end,
const uint8_t **unescaped_buf_ptr, int *unescaped_buf_size);
-int ff_sp5x_process_packet(AVCodecContext *avctx, AVPacket *avpkt);
-
#endif /* AVCODEC_MJPEGDEC_H */
diff --git a/libavcodec/sp5xdec.c b/libavcodec/sp5xdec.c
index 394448c5a9..dfed725500 100644
--- a/libavcodec/sp5xdec.c
+++ b/libavcodec/sp5xdec.c
@@ -32,21 +32,23 @@
#include "mjpegdec.h"
#include "sp5x.h"
-int ff_sp5x_process_packet(AVCodecContext *avctx, AVPacket *avpkt)
+
+static int sp5x_decode_frame(AVCodecContext *avctx,
+ AVFrame *frame, int *got_frame,
+ AVPacket *avpkt)
{
const uint8_t *buf = avpkt->data;
int buf_size = avpkt->size;
- AVBufferRef *buf_recoded;
uint8_t *recoded;
int i = 0, j = 0;
+ int ret;
if (!avctx->width || !avctx->height)
return -1;
- buf_recoded = av_buffer_allocz(buf_size + 1024);
- if (!buf_recoded)
+ recoded = av_mallocz(buf_size + 1024);
+ if (!recoded)
return -1;
- recoded = buf_recoded->data;
/* SOI */
recoded[j++] = 0xFF;
@@ -83,12 +85,12 @@ int ff_sp5x_process_packet(AVCodecContext *avctx, AVPacket *avpkt)
recoded[j++] = 0xFF;
recoded[j++] = 0xD9;
- av_buffer_unref(&avpkt->buf);
- avpkt->buf = buf_recoded;
- avpkt->data = recoded;
- avpkt->size = j;
+ ret = ff_mjpeg_decode_frame_from_buf(avctx, frame, got_frame,
+ avpkt, recoded, j);
+
+ av_free(recoded);
- return 0;
+ return ret < 0 ? ret : avpkt->size;
}
#if CONFIG_SP5X_DECODER
@@ -100,11 +102,10 @@ const FFCodec ff_sp5x_decoder = {
.priv_data_size = sizeof(MJpegDecodeContext),
.init = ff_mjpeg_decode_init,
.close = ff_mjpeg_decode_end,
- FF_CODEC_RECEIVE_FRAME_CB(ff_mjpeg_receive_frame),
+ FF_CODEC_DECODE_CB(sp5x_decode_frame),
.p.capabilities = AV_CODEC_CAP_DR1,
.p.max_lowres = 3,
- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP |
- FF_CODEC_CAP_SETS_PKT_DTS,
+ .caps_internal = FF_CODEC_CAP_INIT_CLEANUP,
};
#endif
#if CONFIG_AMV_DECODER
@@ -116,10 +117,9 @@ const FFCodec ff_amv_decoder = {
.priv_data_size = sizeof(MJpegDecodeContext),
.init = ff_mjpeg_decode_init,
.close = ff_mjpeg_decode_end,
- FF_CODEC_RECEIVE_FRAME_CB(ff_mjpeg_receive_frame),
+ FF_CODEC_DECODE_CB(sp5x_decode_frame),
.p.max_lowres = 3,
.p.capabilities = AV_CODEC_CAP_DR1,
- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP |
- FF_CODEC_CAP_SETS_PKT_DTS,
+ .caps_internal = FF_CODEC_CAP_INIT_CLEANUP,
};
#endif
--
2.34.1
More information about the ffmpeg-devel
mailing list