[FFmpeg-devel] [PATCH] configure: correct lensfun header function check

2023-12-30 Thread Gyan Doshi 
The function lf_db_new was deprecated in lensfun 09dcd3e7ad in 2017
in favour of lf_db_create. The AVfilter was adjusted in 8b78eb312d
but the configure check wasn't changed.

lensfun removed lf_db_new declaration in lf 35c0017593 so configure
fails to find lensfun.
---
 configure | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/configure b/configure
index cd66e42850..9262c1b650 100755
--- a/configure
+++ b/configure
@@ -6776,7 +6776,7 @@ enabled libjxl&& require_pkg_config libjxl 
"libjxl >= 0.7.0" jxl/dec
  require_pkg_config libjxl_threads "libjxl_threads 
>= 0.7.0" jxl/thread_parallel_runner.h JxlThreadParallelRunner
 enabled libklvanc && require libklvanc libklvanc/vanc.h 
klvanc_context_create -lklvanc
 enabled libkvazaar&& require_pkg_config libkvazaar "kvazaar >= 2.0.0" 
kvazaar.h kvz_api_get
-enabled liblensfun&& require_pkg_config liblensfun lensfun lensfun.h 
lf_db_new
+enabled liblensfun&& require_pkg_config liblensfun lensfun lensfun.h 
lf_db_create
 
 if enabled libmfx && enabled libvpl; then
die "ERROR: can not use libmfx and libvpl together"
-- 
2.39.1

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH] configure: remove Bitrig OS

2023-12-30 Thread Brad Smith 
configure: remove Bitrig OS

Bitrig has been defunct for 7 years.

Signed-off-by: Brad Smith 
---
 configure | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/configure b/configure
index 42786719e9..d96b47ebe5 100755
--- a/configure
+++ b/configure
@@ -5625,7 +5625,7 @@ case $target_os in
 oss_outdev_extralibs="-lossaudio"
 enabled gcc || check_ldflags -Wl,-zmuldefs
 ;;
-openbsd|bitrig)
+openbsd)
 disable symver
 enable section_data_rel_ro
 striptype=""
-- 
2.43.0

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH] configure: update copyright year

2023-12-30 Thread Lynne 
If anyone else wants to do this, it's fine by me.
I did promise to let Paul do it this year, but can't get
in touch with him.

>From 4b85c208eb97f0a72d8286fd07f973d322ece6b5 Mon Sep 17 00:00:00 2001
From: Lynne 
Date: Mon, 1 Jan 2024 00:00:00 +
Subject: [PATCH] configure: update copyright year

---
 configure | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/configure b/configure
index cd66e42850..65b4659b0a 100755
--- a/configure
+++ b/configure
@@ -8045,7 +8045,7 @@ cat > $TMPH <___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH] libavformat/hlsenc.c: Populate OTI using AAC profile in write_codec_attr.

2023-12-30 Thread David Johansen 
On Sat, Dec 30, 2023 at 8:23 AM Romain Beauxis  wrote:

> Le jeu. 28 déc. 2023 à 17:26, David Johansen  a
> écrit :
> >>
> >> I love this change, but it appears that st->codecpar->profile is always
> AV_PROFILE_UNKNOWN when using libfdk_aac as the encoder. Any indications
> where I should look for fix that so this can be used with that encoder?
> >
> >
> > It appears that the issue is that profile doesn't default to what's
> being used so `--profile:a` has to be set explicitly with libfdk_aac and
> then it works. Not sure if that's an issue worth fixing, but if someone
> points me to where it needs to be done, then I'd be glad to take a look at
> fixing it
>
> This feels like a second, separate issue to me?
>
> Maybe we could get these changes in first and then tackle it?
>

But this is technically a breaking change, because it takes commands/uses
that currently work and changes them to no longer include CODECS since the
profile value is unknown by default
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 6/7] lavf/codec2: Silence warnings when either muxer/demuxer is disabled

2023-12-30 Thread Tomas Härdin 

From f64b4a90ea22421db494d87f5904ae8c7294c871 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 22:52:25 +0100
Subject: [PATCH 6/7] lavf/codec2: Silence warnings when either muxer/demuxer
 is disabled

---
 libavformat/codec2.c | 76 +---
 1 file changed, 43 insertions(+), 33 deletions(-)

diff --git a/libavformat/codec2.c b/libavformat/codec2.c
index 7447109752..8ff14bc72a 100644
--- a/libavformat/codec2.c
+++ b/libavformat/codec2.c
@@ -49,6 +49,7 @@ typedef struct {
 int frames_per_packet;
 } Codec2Context;
 
+#if CONFIG_CODEC2_DEMUXER
 static int check_version(uint8_t major, uint8_t minor) {
 //no .c2 files prior to 0.8 or later than 1.X
 if (major == MIN_MAJOR_VERSION && minor < MIN_MINOR_VERSION)
@@ -69,7 +70,9 @@ static int codec2_probe(const AVProbeData *p)
 //32 bits of identification -> low score
 return AVPROBE_SCORE_EXTENSION + 1;
 }
+#endif
 
+#if CONFIG_CODEC2_DEMUXER || CONFIG_CODEC2RAW_DEMUXER
 //Mimics codec2_samples_per_frame()
 static int codec2_mode_frame_size(AVFormatContext *s, int mode)
 {
@@ -161,6 +164,41 @@ static int codec2_read_header_common(AVFormatContext *s, AVStream *st, int heade
 return 0;
 }
 
+static int codec2_read_packet(AVFormatContext *s, AVPacket *pkt)
+{
+Codec2Context *c2 = s->priv_data;
+AVStream *st = s->streams[0];
+int ret, size, n, block_align, frame_size;
+
+block_align = st->codecpar->block_align;
+frame_size  = st->codecpar->frame_size;
+
+if (block_align <= 0 || frame_size <= 0 || c2->frames_per_packet <= 0) {
+return AVERROR(EINVAL);
+}
+
+//try to read desired number of frames, compute n from to actual number of bytes read
+size = c2->frames_per_packet * block_align;
+ret = av_get_packet(s->pb, pkt, size);
+if (ret < 0) {
+return ret;
+}
+
+//only set duration - compute_pkt_fields() and ff_pcm_read_seek() takes care of everything else
+//tested by spamming the seek functionality in ffplay
+n = ret / block_align;
+pkt->duration = n * frame_size;
+
+//un-mark packet as corrupt if size is a multiple of block_align
+//this can happen when frames_per_packet > 1
+if (ret % block_align == 0)
+pkt->flags &= ~AV_PKT_FLAG_CORRUPT;
+
+return ret;
+}
+#endif
+
+#if CONFIG_CODEC2_DEMUXER
 static int codec2_read_header(AVFormatContext *s)
 {
 AVStream *st = avformat_new_stream(s, NULL);
@@ -197,40 +235,9 @@ static int codec2_read_header(AVFormatContext *s)
 
 return codec2_read_header_common(s, st, CODEC2_HEADER_SIZE);
 }
+#endif
 
-static int codec2_read_packet(AVFormatContext *s, AVPacket *pkt)
-{
-Codec2Context *c2 = s->priv_data;
-AVStream *st = s->streams[0];
-int ret, size, n, block_align, frame_size;
-
-block_align = st->codecpar->block_align;
-frame_size  = st->codecpar->frame_size;
-
-if (block_align <= 0 || frame_size <= 0 || c2->frames_per_packet <= 0) {
-return AVERROR(EINVAL);
-}
-
-//try to read desired number of frames, compute n from to actual number of bytes read
-size = c2->frames_per_packet * block_align;
-ret = av_get_packet(s->pb, pkt, size);
-if (ret < 0) {
-return ret;
-}
-
-//only set duration - compute_pkt_fields() and ff_pcm_read_seek() takes care of everything else
-//tested by spamming the seek functionality in ffplay
-n = ret / block_align;
-pkt->duration = n * frame_size;
-
-//un-mark packet as corrupt if size is a multiple of block_align
-//this can happen when frames_per_packet > 1
-if (ret % block_align == 0)
-pkt->flags &= ~AV_PKT_FLAG_CORRUPT;
-
-return ret;
-}
-
+#if CONFIG_CODEC2_MUXER
 static int codec2_write_header(AVFormatContext *s)
 {
 AVStream *st;
@@ -253,7 +260,9 @@ static int codec2_write_header(AVFormatContext *s)
 
 return 0;
 }
+#endif
 
+#if CONFIG_CODEC2RAW_DEMUXER
 static int codec2raw_read_header(AVFormatContext *s)
 {
 Codec2Context *c2 = s->priv_data;
@@ -280,6 +289,7 @@ static int codec2raw_read_header(AVFormatContext *s)
 
 return codec2_read_header_common(s, st, 0);
 }
+#endif
 
 //transcoding report2074.c2 to wav went from 7.391s to 5.322s with -frames_per_packet 1000 compared to default, same sha1sum
 #define FRAMES_PER_PACKET \
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 5/7] lavf/codec2: Multiple of block_align -> not corrupt

2023-12-30 Thread Tomas Härdin 

From 74302cd9f6f213b81e3bbc8a3547b518db41 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 22:50:18 +0100
Subject: [PATCH 5/7] lavf/codec2: Multiple of block_align -> not corrupt

---
 libavformat/codec2.c | 5 +
 1 file changed, 5 insertions(+)

diff --git a/libavformat/codec2.c b/libavformat/codec2.c
index 8299302acc..7447109752 100644
--- a/libavformat/codec2.c
+++ b/libavformat/codec2.c
@@ -223,6 +223,11 @@ static int codec2_read_packet(AVFormatContext *s, AVPacket *pkt)
 n = ret / block_align;
 pkt->duration = n * frame_size;
 
+//un-mark packet as corrupt if size is a multiple of block_align
+//this can happen when frames_per_packet > 1
+if (ret % block_align == 0)
+pkt->flags &= ~AV_PKT_FLAG_CORRUPT;
+
 return ret;
 }
 
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 4/7] lavf/codec2: Allow versions between 0.8 and 1.X

2023-12-30 Thread Tomas Härdin 

From 5a660d91c2f34c4e464212bf3694cdb7bca80f1a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 17:33:18 +0100
Subject: [PATCH 4/7] lavf/codec2: Allow versions between 0.8 and 1.X

---
 libavformat/codec2.c | 36 ++--
 1 file changed, 22 insertions(+), 14 deletions(-)

diff --git a/libavformat/codec2.c b/libavformat/codec2.c
index 78bc339209..8299302acc 100644
--- a/libavformat/codec2.c
+++ b/libavformat/codec2.c
@@ -37,8 +37,11 @@
 
 //the lowest version we should ever run across is 0.8
 //we may run across later versions as the format evolves
-#define EXPECTED_CODEC2_MAJOR_VERSION 0
-#define EXPECTED_CODEC2_MINOR_VERSION 8
+#define MIN_MAJOR_VERSION 0
+#define MIN_MINOR_VERSION 8
+//accept any minor version with major version 1.X
+//as of writing the latest release is 1.2.0
+#define MAX_MAJOR_VERSION 1
 
 typedef struct {
 const AVClass *class;
@@ -46,17 +49,20 @@ typedef struct {
 int frames_per_packet;
 } Codec2Context;
 
+static int check_version(uint8_t major, uint8_t minor) {
+//no .c2 files prior to 0.8 or later than 1.X
+if (major == MIN_MAJOR_VERSION && minor < MIN_MINOR_VERSION)
+return 1;
+if (major > MAX_MAJOR_VERSION)
+return 1;
+return 0;
+}
+
 static int codec2_probe(const AVProbeData *p)
 {
 //must start wih C0 DE C2
-if (AV_RB24(p->buf) != CODEC2_MAGIC) {
-return 0;
-}
-
-//no .c2 files prior to 0.8
-//be strict about major version while we're at it
-if (p->buf[3] != EXPECTED_CODEC2_MAJOR_VERSION ||
-p->buf[4] <  EXPECTED_CODEC2_MINOR_VERSION) {
+if (AV_RB24(p->buf) != CODEC2_MAGIC ||
+check_version(p->buf[3], p->buf[4])) {
 return 0;
 }
 
@@ -158,7 +164,8 @@ static int codec2_read_header_common(AVFormatContext *s, AVStream *st, int heade
 static int codec2_read_header(AVFormatContext *s)
 {
 AVStream *st = avformat_new_stream(s, NULL);
-int ret, version;
+int ret;
+uint8_t major, minor;
 
 if (!st) {
 return AVERROR(ENOMEM);
@@ -179,9 +186,10 @@ static int codec2_read_header(AVFormatContext *s)
 return ret;
 }
 
-version = AV_RB16(st->codecpar->extradata);
-if ((version >> 8) != EXPECTED_CODEC2_MAJOR_VERSION) {
-avpriv_report_missing_feature(s, "Major version %i", version >> 8);
+major = st->codecpar->extradata[0];
+minor = st->codecpar->extradata[1];
+if (check_version(major, minor)) {
+avpriv_report_missing_feature(s, "Version %i.%i", major, minor);
 return AVERROR_PATCHWELCOME;
 }
 
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 3/7] lavf/codec2: Compute duration from filesize

2023-12-30 Thread Tomas Härdin 
Previous version accidentally used CODEC2_EXTRADATA_SIZE not
CODEC2_HEADER_SIZE

/Tomas
From 6fda817b2bc237081e6bf6daeff9087b6543647b Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 16:32:49 +0100
Subject: [PATCH 3/7] lavf/codec2: Compute duration from filesize

---
 libavformat/codec2.c | 13 ++---
 1 file changed, 10 insertions(+), 3 deletions(-)

diff --git a/libavformat/codec2.c b/libavformat/codec2.c
index 1d6ef2a14a..78bc339209 100644
--- a/libavformat/codec2.c
+++ b/libavformat/codec2.c
@@ -124,9 +124,10 @@ static int codec2_mode_bit_rate(AVFormatContext *s, int mode)
 return 8 * 8000 * block_align / frame_size;
 }
 
-static int codec2_read_header_common(AVFormatContext *s, AVStream *st)
+static int codec2_read_header_common(AVFormatContext *s, AVStream *st, int header)
 {
 int mode = codec2_mode_from_extradata(st->codecpar->extradata);
+int64_t sz = avio_size(s->pb);
 
 st->codecpar->codec_type= AVMEDIA_TYPE_AUDIO;
 st->codecpar->codec_id  = AV_CODEC_ID_CODEC2;
@@ -143,6 +144,12 @@ static int codec2_read_header_common(AVFormatContext *s, AVStream *st)
 return AVERROR_INVALIDDATA;
 }
 
+// compute duration from filesize
+if (sz >= header) {
+int64_t frames = (sz - header) / st->codecpar->block_align;
+if (frames <= INT64_MAX / st->codecpar->frame_size)
+st->duration = frames * st->codecpar->frame_size;
+}
 avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
 
 return 0;
@@ -180,7 +187,7 @@ static int codec2_read_header(AVFormatContext *s)
 
 ffformatcontext(s)->data_offset = CODEC2_HEADER_SIZE;
 
-return codec2_read_header_common(s, st);
+return codec2_read_header_common(s, st, CODEC2_HEADER_SIZE);
 }
 
 static int codec2_read_packet(AVFormatContext *s, AVPacket *pkt)
@@ -258,7 +265,7 @@ static int codec2raw_read_header(AVFormatContext *s)
 
 codec2_make_extradata(st->codecpar->extradata, c2->mode);
 
-return codec2_read_header_common(s, st);
+return codec2_read_header_common(s, st, 0);
 }
 
 //transcoding report2074.c2 to wav went from 7.391s to 5.322s with -frames_per_packet 1000 compared to default, same sha1sum
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 2/7] lavc/libcodec2: Report actual bitrate used both when decoding and encoding

2023-12-30 Thread Tomas Härdin 
Without this it looks like the encoder runs at 128 kbit/s

/Tomas
From 5c913a3eaf5cf8c21edc119870b7fdc8251bd900 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Fri, 29 Dec 2023 14:09:51 +0100
Subject: [PATCH 2/7] lavc/libcodec2: Report actual bitrate used both when
 decoding and encoding

---
 libavcodec/libcodec2.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/libavcodec/libcodec2.c b/libavcodec/libcodec2.c
index 950d48685e..2e03236b81 100644
--- a/libavcodec/libcodec2.c
+++ b/libavcodec/libcodec2.c
@@ -72,6 +72,7 @@ static av_cold int libcodec2_init_common(AVCodecContext *avctx, int mode)
 }
 
 codec2_set_natural_or_gray(c2->codec, 1);
+avctx->bit_rate = avctx->sample_rate / avctx->frame_size * avctx->block_align * 8;
 
 return 0;
 
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 1/7] lavc/codec2utils: Use actual libcodec2 version

2023-12-30 Thread Tomas Härdin 
Should compile without libcodec2 installed now

/Tomas
From 098a3b7e78dc0b1bbdfce051de21526f0a886b4d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 17:32:21 +0100
Subject: [PATCH 1/7] lavc/codec2utils: Use actual libcodec2 version

---
 libavcodec/codec2utils.h | 17 ++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/libavcodec/codec2utils.h b/libavcodec/codec2utils.h
index 6812ae895c..e1314b05b6 100644
--- a/libavcodec/codec2utils.h
+++ b/libavcodec/codec2utils.h
@@ -23,6 +23,10 @@
 #define AVCODEC_CODEC2UTILS_H
 
 #include 
+#include "config.h"
+#if CONFIG_LIBCODEC2
+#include 
+#endif
 
 //Highest mode we're willing to use.
 //Don't want to let users accidentally produce files that can't be decoded in the future.
@@ -49,9 +53,16 @@
 
 //Used in codec2raw demuxer and libcodec2 encoder
 static inline void codec2_make_extradata(uint8_t *ptr, int mode) {
-//version 0.8 as of 2017-12-23 (r3386)
-ptr[0] = 0; //major
-ptr[1] = 8; //minor
+#if CONFIG_LIBCODEC2
+ptr[0] = CODEC2_VERSION_MAJOR;
+ptr[1] = CODEC2_VERSION_MINOR;
+#else
+// this codepath is only hit when demuxing raw codec2 with libcodec2 disabled,
+// and only relevant when remuxing from raw codec2 to .c2,
+// because version information is ignored except when writing the .c2 header
+ptr[0] = 1; // version 1.2.0 is the latest we know of
+ptr[1] = 2;
+#endif
 ptr[2] = mode;  //mode
 ptr[3] = 0; //flags
 }
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 6/6] Add myself as APC maintainer

2023-12-30 Thread Tomas Härdin 

From 65f4215a519e457d6a3d6c943d14ebd739083eed Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 16:45:33 +0100
Subject: [PATCH 6/6] Add myself as APC maintainer

---
 MAINTAINERS | 1 +
 1 file changed, 1 insertion(+)

diff --git a/MAINTAINERS b/MAINTAINERS
index 9c4683db19..fa98e043c5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -391,6 +391,7 @@ Muxers/Demuxers:
   aiffenc.c Baptiste Coudurier, Matthieu Bouron
   alp.c Zane van Iperen
   amvenc.c  Zane van Iperen
+  apc*  Tomas Härdin
   apm.c Zane van Iperen
   apngdec.c Benoit Fouet
   argo_asf.cZane van Iperen
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 5/6] lavc: Add ADPCM IMA CRYO APC encoder

2023-12-30 Thread Tomas Härdin 

From 701eec2a7e742fa0fcc375a6c3784ac012100d0a Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 14:32:20 +0100
Subject: [PATCH 5/6] lavc: Add ADPCM IMA CRYO APC encoder

No trellis quantization yet
---
 Changelog  |  1 +
 doc/general_contents.texi  |  2 +-
 libavcodec/adpcmenc.c  | 33 +
 libavcodec/allcodecs.c |  1 +
 libavcodec/version.h   |  2 +-
 tests/fate/acodec.mak  |  2 ++
 tests/ref/acodec/adpcm-ima_apc |  4 
 7 files changed, 43 insertions(+), 2 deletions(-)
 create mode 100644 tests/ref/acodec/adpcm-ima_apc

diff --git a/Changelog b/Changelog
index 31c9e25be6..586ccd1a9c 100644
--- a/Changelog
+++ b/Changelog
@@ -14,6 +14,7 @@ version :
 - D3D12VA hardware accelerated H264, HEVC, VP9, AV1, MPEG-2 and VC1 decoding
 - tiltandshift filter
 - CRYO APC muxer
+- ADPCM IMA APC encoder
 
 version 6.1:
 - libaribcaption decoder
diff --git a/doc/general_contents.texi b/doc/general_contents.texi
index df31edd060..9f450dc794 100644
--- a/doc/general_contents.texi
+++ b/doc/general_contents.texi
@@ -1188,7 +1188,7 @@ following image formats are supported:
 @item ADPCM IMA Acorn Replay @tab @tab  X
 @item ADPCM IMA AMV  @tab  X  @tab  X
 @tab Used in AMV files
-@item ADPCM IMA APC  @tab @tab  X
+@item ADPCM IMA APC  @tab  X  @tab  X
 @tab Codec used in games by Cryo Interactive
 @item ADPCM IMA Cunning Developments  @tab @tab  X
 @item ADPCM IMA Electronic Arts EACS  @tab @tab  X
diff --git a/libavcodec/adpcmenc.c b/libavcodec/adpcmenc.c
index 7f18fc2daf..7abd71541d 100644
--- a/libavcodec/adpcmenc.c
+++ b/libavcodec/adpcmenc.c
@@ -74,6 +74,7 @@ typedef struct ADPCMEncodeContext {
 TrellisNode *node_buf;
 TrellisNode **nodep_buf;
 uint8_t *trellis_hash;
+int extradata_updated;
 } ADPCMEncodeContext;
 
 #define FREEZE_INTERVAL 128
@@ -157,6 +158,15 @@ static av_cold int adpcm_encode_init(AVCodecContext *avctx)
 bytestream_put_le16(&extradata, ff_adpcm_AdaptCoeff2[i] * 4);
 }
 ) /* End of CASE */
+CASE(ADPCM_IMA_APC,
+if (avctx->trellis) {
+av_log(avctx, AV_LOG_ERROR, "trellis encoding not implemented for CRYO APC\n");
+return AVERROR_PATCHWELCOME;
+}
+//extradata will be output in adpcm_encode_frame()
+avctx->frame_size  = s->block_size * 2 / channels;
+avctx->block_align = s->block_size;
+) /* End of CASE */
 CASE(ADPCM_YAMAHA,
 avctx->frame_size  = s->block_size * 2 / channels;
 avctx->block_align = s->block_size;
@@ -622,6 +632,28 @@ static int adpcm_encode_frame(AVCodecContext *avctx, AVPacket *avpkt,
 dst = avpkt->data;
 
 switch(avctx->codec->id) {
+CASE(ADPCM_IMA_APC,
+//initialize predictors using initial samples
+if (!c->extradata_updated) {
+uint8_t *side_data = av_packet_new_side_data(
+avpkt, AV_PKT_DATA_NEW_EXTRADATA, 8);
+
+if (!side_data) {
+return AVERROR(ENOMEM);
+}
+
+for (int ch = 0; ch < channels; ch++) {
+c->status[ch].prev_sample = samples[ch];
+bytestream_put_le32(&side_data, c->status[ch].prev_sample);
+}
+c->extradata_updated = 1;
+}
+for (int i = 0; i < frame->nb_samples*channels/2; i++) {
+uint8_t l = adpcm_ima_compress_sample(&c->status[0],  samples[2*i+0]);
+uint8_t r = adpcm_ima_compress_sample(&c->status[st], samples[2*i+1]);
+*dst++ = (l<<4) | r;
+}
+) /* End of CASE */
 CASE(ADPCM_IMA_WAV,
 int blocks = (frame->nb_samples - 1) / 8;
 
@@ -1027,6 +1059,7 @@ ADPCM_ENCODER(ADPCM_IMA_QT,  adpcm_ima_qt,  sample_fmts_p, 0,
 ADPCM_ENCODER(ADPCM_IMA_SSI, adpcm_ima_ssi, sample_fmts,   AV_CODEC_CAP_SMALL_LAST_FRAME, "ADPCM IMA Simon & Schuster Interactive")
 ADPCM_ENCODER(ADPCM_IMA_WAV, adpcm_ima_wav, sample_fmts_p, 0, "ADPCM IMA WAV")
 ADPCM_ENCODER(ADPCM_IMA_WS,  adpcm_ima_ws,  sample_fmts,   AV_CODEC_CAP_SMALL_LAST_FRAME, "ADPCM IMA Westwood")
+ADPCM_ENCODER(ADPCM_IMA_APC, adpcm_ima_apc, sample_fmts,   0, "ADPCM IMA CRYO APC")
 ADPCM_ENCODER(ADPCM_MS,  adpcm_ms,  sample_fmts,   0, "ADPCM Microsoft")
 ADPCM_ENCODER(ADPCM_SWF, adpcm_swf, sample_fmts,   0, "ADPCM Shockwave Flash")
 ADPCM_ENCODER(ADPCM_YAMAHA,  adpcm_yamaha,  sample_fmts,   0, "ADPCM Yamaha")
diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
index b0f004e15c..34662a88f3 100644
--- a/libavcodec/allcodecs.c
+++ b/libavcodec/allcodecs.c
@@ -669,6 +669,7 @@ extern const FFCodec ff_adpcm_ima_amv_encoder;
 extern const FFCodec ff_adpcm_ima_alp_decoder;
 extern const FFCodec ff_adpcm_ima_alp_encoder;
 extern const F

[FFmpeg-devel] [PATCH v2 4/6] lavf: Add CRYO APC muxer

2023-12-30 Thread Tomas Härdin 

From 1fab30dea9528478ccbe78dc68a3933acb8115b8 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 14:32:10 +0100
Subject: [PATCH 4/6] lavf: Add CRYO APC muxer

---
 Changelog |   1 +
 doc/general_contents.texi |   2 +-
 libavformat/Makefile  |   1 +
 libavformat/allformats.c  |   1 +
 libavformat/apcenc.c  | 129 ++
 libavformat/version.h |   2 +-
 6 files changed, 134 insertions(+), 2 deletions(-)
 create mode 100644 libavformat/apcenc.c

diff --git a/Changelog b/Changelog
index a638c03250..31c9e25be6 100644
--- a/Changelog
+++ b/Changelog
@@ -13,6 +13,7 @@ version :
 - IAMF raw demuxer and muxer
 - D3D12VA hardware accelerated H264, HEVC, VP9, AV1, MPEG-2 and VC1 decoding
 - tiltandshift filter
+- CRYO APC muxer
 
 version 6.1:
 - libaribcaption decoder
diff --git a/doc/general_contents.texi b/doc/general_contents.texi
index ec7516dce8..df31edd060 100644
--- a/doc/general_contents.texi
+++ b/doc/general_contents.texi
@@ -464,7 +464,7 @@ library:
 @item CRC testing format@tab X @tab
 @item Creative Voice@tab X @tab X
 @tab Created for the Sound Blaster Pro.
-@item CRYO APC  @tab   @tab X
+@item CRYO APC  @tab X @tab X
 @tab Audio format used in some games by CRYO Interactive Entertainment.
 @item D-Cinema audio@tab X @tab X
 @item Deluxe Paint Animation@tab   @tab X
diff --git a/libavformat/Makefile b/libavformat/Makefile
index 581e378d95..f605f4ae69 100644
--- a/libavformat/Makefile
+++ b/libavformat/Makefile
@@ -105,6 +105,7 @@ OBJS-$(CONFIG_AMV_MUXER) += amvenc.o
 OBJS-$(CONFIG_ANM_DEMUXER)   += anm.o
 OBJS-$(CONFIG_APAC_DEMUXER)  += apac.o rawdec.o
 OBJS-$(CONFIG_APC_DEMUXER)   += apc.o
+OBJS-$(CONFIG_APC_MUXER) += apcenc.o
 OBJS-$(CONFIG_APE_DEMUXER)   += ape.o apetag.o img2.o
 OBJS-$(CONFIG_APM_DEMUXER)   += apm.o
 OBJS-$(CONFIG_APM_MUXER) += apm.o rawenc.o
diff --git a/libavformat/allformats.c b/libavformat/allformats.c
index ce6be5f04d..18a33d7448 100644
--- a/libavformat/allformats.c
+++ b/libavformat/allformats.c
@@ -60,6 +60,7 @@ extern const FFOutputFormat ff_amv_muxer;
 extern const AVInputFormat  ff_anm_demuxer;
 extern const AVInputFormat  ff_apac_demuxer;
 extern const AVInputFormat  ff_apc_demuxer;
+extern const FFOutputFormat ff_apc_muxer;
 extern const AVInputFormat  ff_ape_demuxer;
 extern const AVInputFormat  ff_apm_demuxer;
 extern const FFOutputFormat ff_apm_muxer;
diff --git a/libavformat/apcenc.c b/libavformat/apcenc.c
new file mode 100644
index 00..7c7f3a1f26
--- /dev/null
+++ b/libavformat/apcenc.c
@@ -0,0 +1,129 @@
+/*
+ * CRYO APC audio format muxer
+ * Copyright (c) 2023 Tomas Härdin 
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "avio_internal.h"
+#include "internal.h"
+#include "mux.h"
+
+#define APC_HEADER_SIZE (8*4)
+
+static int apc_write_header(AVFormatContext *s)
+{
+AVCodecParameters *par;
+AVStream *st;
+
+if (s->nb_streams != 1) {
+av_log(s, AV_LOG_ERROR, "Must have exactly one stream\n");
+return AVERROR(EINVAL);
+}
+
+st = s->streams[0];
+par = st->codecpar;
+
+if (par->ch_layout.nb_channels <= 0 || par->ch_layout.nb_channels > 2) {
+av_log(s, AV_LOG_ERROR, "Must be mono or stereo\n");
+return AVERROR(EINVAL);
+}
+
+avpriv_set_pts_info(st, 64, 1, par->sample_rate);
+// delay writing the actual header until we get extradata from the encoder
+return 0;
+}
+
+static int apc_write_header_delayed(AVFormatContext *s, AVPacket *pkt)
+{
+AVIOContext *pb = s->pb;
+AVCodecParameters *par = s->streams[0]->codecpar;
+size_t new_extradata_size = 0, extradata_size;
+const uint8_t *new_extradata = av_packet_get_side_data(pkt, AV_PKT_DATA_NEW_EXTRADATA, &new_extradata_size), *extradata;
+
+// prefer updated extradata from the encoder
+if (new_extradata) {
+extradata_size = new_extradata_size;
+extradata = new_extradata;
+} else {
+extradata_size = par->extradata_size;
+extradata = par->extradata;
+}
+
+if (extradata_size

[FFmpeg-devel] [PATCH v2 3/6] Un-mark IMA APC as intra-only, only mark first packet as keyframe

2023-12-30 Thread Tomas Härdin 

From f652daf2e9bad345c0e98ab058a5e07c3d45d13e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 16:30:35 +0100
Subject: [PATCH 3/6] Un-mark IMA APC as intra-only, only mark first packet as
 keyframe

Packets must be decoded in order.
There is no way to seek and get a bitexact decode.
---
 libavcodec/codec_desc.c | 2 +-
 libavformat/apc.c   | 5 +
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/libavcodec/codec_desc.c b/libavcodec/codec_desc.c
index 033344304c..45ef4c5091 100644
--- a/libavcodec/codec_desc.c
+++ b/libavcodec/codec_desc.c
@@ -2431,7 +2431,7 @@ static const AVCodecDescriptor codec_descriptors[] = {
 .type  = AVMEDIA_TYPE_AUDIO,
 .name  = "adpcm_ima_apc",
 .long_name = NULL_IF_CONFIG_SMALL("ADPCM IMA CRYO APC"),
-.props = AV_CODEC_PROP_INTRA_ONLY | AV_CODEC_PROP_LOSSY,
+.props = AV_CODEC_PROP_LOSSY,
 },
 {
 .id= AV_CODEC_ID_ADPCM_VIMA,
diff --git a/libavformat/apc.c b/libavformat/apc.c
index 7765c53088..d1891954a8 100644
--- a/libavformat/apc.c
+++ b/libavformat/apc.c
@@ -75,8 +75,13 @@ static int apc_read_header(AVFormatContext *s)
 
 static int apc_read_packet(AVFormatContext *s, AVPacket *pkt)
 {
+int first = avio_tell(s->pb) == 32;
 if (av_get_packet(s->pb, pkt, MAX_READ_SIZE) <= 0)
 return AVERROR(EIO);
+// each IMA APC packet depends on the one before for bitexact decode
+// extradata is used to initialize the decoder
+if (first)
+pkt->flags |= AV_PKT_FLAG_KEY;
 pkt->stream_index = 0;
 return 0;
 }
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 2/6] lavf/apc: Read duration from file

2023-12-30 Thread Tomas Härdin 

From 2ef5c89c3e3173b26690bad0279fd484a30a6268 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 14:31:59 +0100
Subject: [PATCH 2/6] lavf/apc: Read duration from file

---
 libavformat/apc.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/libavformat/apc.c b/libavformat/apc.c
index b8b18c966c..7765c53088 100644
--- a/libavformat/apc.c
+++ b/libavformat/apc.c
@@ -24,6 +24,7 @@
 #include "libavutil/channel_layout.h"
 #include "avformat.h"
 #include "demux.h"
+#include "internal.h"
 
 static int apc_probe(const AVProbeData *p)
 {
@@ -51,8 +52,9 @@ static int apc_read_header(AVFormatContext *s)
 st->codecpar->codec_type = AVMEDIA_TYPE_AUDIO;
 st->codecpar->codec_id = AV_CODEC_ID_ADPCM_IMA_APC;
 
-avio_rl32(pb); /* number of samples */
+st->duration = avio_rl32(pb); /* number of samples */
 st->codecpar->sample_rate = avio_rl32(pb);
+avpriv_set_pts_info(st, 64, 1, st->codecpar->sample_rate);
 
 /* initial predictor values for adpcm decoder */
 if ((ret = ff_get_extradata(s, st->codecpar, pb, 2 * 4)) < 0)
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH v2 1/6] doc/general_contents.texi: Add missing ADPCM IMA APC entry

2023-12-30 Thread Tomas Härdin 
Addressed some comments and reordered some patches

/Tomas
From 90954f54d82bd0ceda0400ceb0e92e1affecf3db Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Tue, 26 Dec 2023 15:03:51 +0100
Subject: [PATCH 1/6] doc/general_contents.texi: Add missing ADPCM IMA APC
 entry

---
 doc/general_contents.texi | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/doc/general_contents.texi b/doc/general_contents.texi
index 8b48fed060..ec7516dce8 100644
--- a/doc/general_contents.texi
+++ b/doc/general_contents.texi
@@ -1188,6 +1188,8 @@ following image formats are supported:
 @item ADPCM IMA Acorn Replay @tab @tab  X
 @item ADPCM IMA AMV  @tab  X  @tab  X
 @tab Used in AMV files
+@item ADPCM IMA APC  @tab @tab  X
+@tab Codec used in games by Cryo Interactive
 @item ADPCM IMA Cunning Developments  @tab @tab  X
 @item ADPCM IMA Electronic Arts EACS  @tab @tab  X
 @item ADPCM IMA Electronic Arts SEAD  @tab @tab  X
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 1/4] avfilter/avf_showwaves: Check history_nb_samples

2023-12-30 Thread Michael Niedermayer 
On Sat, Dec 30, 2023 at 03:49:02AM +0100, Michael Niedermayer wrote:
> Fixes: out of array access
> Fixes: tickets/10756/poc18ffmpeg
> 
> Discovered by Zeng Yunxiang
> Signed-off-by: Michael Niedermayer 
> ---
>  libavfilter/avf_showwaves.c | 2 ++
>  1 file changed, 2 insertions(+)

will apply patchset, so the fixes can go in 6.1.1

thx

[...]
-- 
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The real ebay dictionary, page 2
"100% positive feedback" - "All either got their money back or didnt complain"
"Best seller ever, very honest" - "Seller refunded buyer after failed scam"


signature.asc
Description: PGP signature
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 3/6] Add CRYO APC muxer

2023-12-30 Thread Tomas Härdin 
fre 2023-12-29 klockan 18:33 +0100 skrev Michael Niedermayer:
> On Thu, Dec 28, 2023 at 11:31:05AM +0100, Tomas Härdin wrote:
> > tis 2023-12-26 klockan 23:30 +0100 skrev Michael Niedermayer:
> > > Hi
> > > 
> > > On Tue, Dec 26, 2023 at 04:52:47PM +0100, Tomas Härdin wrote:
> > > > 
> > > 
> > > [...]
> > > > +
> > > > +static int apc_write_header(AVFormatContext *s)
> > > > +{
> > > > +    AVIOContext *pb = s->pb;
> > > > +    AVCodecParameters *par;
> > > > +    AVStream *st;
> > > > +
> > > > +    if (s->nb_streams != 1) {
> > > > +    av_log(s, AV_LOG_ERROR, "Must have exactly one
> > > > stream\n");
> > > > +    return AVERROR(EINVAL);
> > > > +    }
> > > > +
> > > > +    st = s->streams[0];
> > > > +    par = st->codecpar;
> > > > +
> > > > +    if (par->ch_layout.nb_channels <= 0 || par-
> > > > > ch_layout.nb_channels > 2) {
> > > > +    av_log(s, AV_LOG_ERROR, "Must be mono or stereo\n");
> > > > +    return AVERROR(EINVAL);
> > > > +    }
> > > > +
> > > > +    if (par->extradata_size != 0 && par->extradata_size != 8)
> > > > {
> > > > +    av_log(s, AV_LOG_ERROR,
> > > > +    "Must have exactly 0 or 8 bytes of extradata, got
> > > > %i\n",
> > > > +    par->extradata_size);
> > > > +    return AVERROR(EINVAL);
> > > > +    }
> > > > +
> > > > +    ffio_wfourcc(pb, "CRYO");
> > > > +    ffio_wfourcc(pb, "_APC");
> > > > +    ffio_wfourcc(pb, "1.20");
> > > 
> > > > +    avio_wl32(pb, 0); // number or samples
> > > 
> > > please add to the comment "updated in apc_write_trailer()"
> > 
> > Sure
> > 
> > 
> > > > +static int apc_write_packet(AVFormatContext *s, AVPacket *pkt)
> > > > +{
> > > > +    size_t extradata_size = 0;
> > > 
> > > > +    const uint8_t *extradata = av_packet_get_side_data(
> > > > +    pkt, AV_PKT_DATA_NEW_EXTRADATA, &extradata_size);
> > > > +
> > > > +    if (extradata_size == 8) {
> > > > +    // we got predictors from encoder
> > > > +    // try to seek back end write them
> > > > +    int64_t pos = avio_tell(s->pb), err;
> > > > +    if ((err = avio_seek(s->pb, 20, SEEK_SET)) >= 0) {
> > > > +    avio_write(s->pb, extradata, extradata_size);
> > > > +    avio_seek(s->pb, pos, SEEK_SET);
> > > > +    } else {
> > > > +    av_log(s, AV_LOG_ERROR, "Got predictors from
> > > > encoder
> > > > but couldn't seek back to write them\n");
> > > > +    // fail since we should always be able to do this
> > > > within the avio cache
> > > > +    // unless the encoder gave us predictors way too
> > > > late
> > > > for some reason
> > > > +    return err;
> > > > +    }
> > > > +    }
> > > 
> > > I think the encoder should buffer data or use 2 passes
> > > if it needs future data. seeking back in the muxer is a bit odd.
> > > 
> > > if it becomes available always in teh first packet then maybe the
> > > whole header could be written in the first packet
> > 
> > Good idea. IIRC the issue was that extradata is either given
> > immediately (remuxing) or after a single-packet delay (encoding),
> > hence
> > the check for *new* metadata
> > 
> > > I would suggest, you add APC support to nut.
> > > Thats a good test to ensure the packet and extradata is set at
> > > the
> > > right time
> > > for a generic muxer
> > 
> > How would I go about doing that?
> 
> nut mainly needs to be able to map the ffmpeg API CODEC_ID to some
> external id
> so the codec id needs to be in one of the table nut scans for it

I doubt this codec has an official RIFF TwoCC. I see some other IMA
codecs in  riff.c though. I suspect this would require coordinating
with Microsoft, which I feel shouldn't hold up this patchset

/Tomas
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

2023-12-30 Thread flow gg 
I mistook it, seeing the vector length as the length of the vector register
..
I have modified it in this reply.



Rémi Denis-Courmont  于2023年12月30日周六 20:15写道：

>
>
> Le 29 décembre 2023 12:57:20 GMT+01:00, flow gg  a
> écrit :
> >C908
> >ssd_int8_vs_int16_c: 207.7
> >ssd_int8_vs_int16_rvv_i32: 28.0
>
> At a quick glance, it won't work if the input length is not a multiple of
> the vector length.
>
> Also do you really need to extend accumulators to 32 bits?
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
>
From ffa495739e163a402bd580ddbdd54a42e1ef6712 Mon Sep 17 00:00:00 2001
From: sunyuechi 
Date: Fri, 29 Dec 2023 13:27:31 +0800
Subject: [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

C908
ssd_int8_vs_int16_c: 207.7
ssd_int8_vs_int16_rvv_i32: 17.7
---
 libavcodec/riscv/Makefile  |  2 ++
 libavcodec/riscv/svqenc_init.c | 41 +++
 libavcodec/riscv/svqenc_rvv.S  | 44 ++
 libavcodec/svq1enc.c   |  2 ++
 libavcodec/svq1encdsp.h|  1 +
 5 files changed, 90 insertions(+)
 create mode 100644 libavcodec/riscv/svqenc_init.c
 create mode 100644 libavcodec/riscv/svqenc_rvv.S

diff --git a/libavcodec/riscv/Makefile b/libavcodec/riscv/Makefile
index 7f253bba12..4e14c3d094 100644
--- a/libavcodec/riscv/Makefile
+++ b/libavcodec/riscv/Makefile
@@ -46,6 +46,8 @@ RVV-OBJS-$(CONFIG_OPUS_DECODER) += riscv/opusdsp_rvv.o
 OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_init.o
 RV-OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_rvi.o
 RVV-OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_rvv.o
+OBJS-$(CONFIG_SVQ1_ENCODER) += riscv/svqenc_init.o
+RVV-OBJS-$(CONFIG_SVQ1_ENCODER) += riscv/svqenc_rvv.o
 OBJS-$(CONFIG_TAK_DECODER) += riscv/takdsp_init.o
 RVV-OBJS-$(CONFIG_TAK_DECODER) += riscv/takdsp_rvv.o
 OBJS-$(CONFIG_UTVIDEO_DECODER) += riscv/utvideodsp_init.o
diff --git a/libavcodec/riscv/svqenc_init.c b/libavcodec/riscv/svqenc_init.c
new file mode 100644
index 00..f4c398960c
--- /dev/null
+++ b/libavcodec/riscv/svqenc_init.c
@@ -0,0 +1,41 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "config.h"
+
+#include "libavutil/attributes.h"
+#include "libavutil/cpu.h"
+#include "libavcodec/svq1encdsp.h"
+
+int ff_ssd_int8_vs_int16_rvv(const int8_t *pix1, const int16_t *pix2,
+  intptr_t size);
+
+av_cold void ff_svq1enc_init_riscv(SVQ1EncDSPContext *c)
+{
+#if HAVE_RVV
+int flags = av_get_cpu_flags();
+
+if (flags & AV_CPU_FLAG_RVV_I32) {
+if (flags & AV_CPU_FLAG_RVB_ADDR) {
+c->ssd_int8_vs_int16 = ff_ssd_int8_vs_int16_rvv;
+}
+}
+#endif
+}
diff --git a/libavcodec/riscv/svqenc_rvv.S b/libavcodec/riscv/svqenc_rvv.S
new file mode 100644
index 00..985bab81a0
--- /dev/null
+++ b/libavcodec/riscv/svqenc_rvv.S
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "libavutil/riscv/asm.S"
+
+func ff_ssd_int8_vs_int16_rvv, zve32x
+vsetvli t0, zero, e32, m8, ta, ma
+vmv.v.x v24, zero
+1:
+vsetvli  t0, a2, e8, m2, tu, ma
+vle8.v   v0, (a0)
+sub  a2, a2, t0
+vset

Re: [FFmpeg-devel] [PATCH 1/6] lavc/codec2utils: Use actual libcodec2 version

2023-12-30 Thread Tomas Härdin 
fre 2023-12-29 klockan 00:30 +0100 skrev Michael Niedermayer:
> On Thu, Dec 28, 2023 at 10:21:23PM +0100, Tomas Härdin wrote:
> > tor 2023-12-28 klockan 21:29 +0100 skrev Michael Niedermayer:
> > > On Thu, Dec 28, 2023 at 07:43:09PM +0100, Tomas Härdin wrote:
> > > > libcodec2 has reached 1.X so it's about time the bindings
> > > > receive
> > > > some
> > > > attention. This patchset also includes some much-needed tests.
> > > > Where do
> > > > I submit files for inclusion in FATE again? They are very small
> > > > as
> > > > is
> > > > to be expected from a codec that runs at 700-3200 bit/s
> > > > 
> > > > /Tomas
> > > 
> > > >  codec2utils.h |    6 +++---
> > > >  1 file changed, 3 insertions(+), 3 deletions(-)
> > > > a79a98f8ffc27ca6d79c9d373957fa0bdc985ef2  0001-lavc-
> > > > codec2utils-
> > > > Use-actual-libcodec2-version.patch
> > > > From 7205e741aaadc354b403010c97f9cd803eec612d Mon Sep 17
> > > > 00:00:00
> > > > 2001
> > > > From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
> > > > Date: Wed, 27 Dec 2023 17:32:21 +0100
> > > > Subject: [PATCH 1/6] lavc/codec2utils: Use actual libcodec2
> > > > version
> > > 
> > > This breaks build here
> > > 
> > > CC  libavformat/codec2.o
> > > In file included from libavformat/codec2.c:24:0:
> > > ./libavcodec/codec2utils.h:26:10: fatal error: codec2/version.h:
> > > No
> > > such file or directory
> > >  #include 
> > >   ^~
> > 
> > What version of libcodec2 are you using? I don't think we can rely
> > on
> > versions that don't expose version information, which only comes
> > about
> > because I explicitly told them to please expose it in a header. I
> > don't
> > think we can even probe version at configure time otherwise.
> > 
> > libcodec2-dev on Debian bookworm works for me
> 
> i think theres no libcodec2* on the box and this header is used
> outside
> checks for libcodec2* ...

Updated patch attached

/Tomas
From 098a3b7e78dc0b1bbdfce051de21526f0a886b4d Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Tomas=20H=C3=A4rdin?= 
Date: Wed, 27 Dec 2023 17:32:21 +0100
Subject: [PATCH 1/7] lavc/codec2utils: Use actual libcodec2 version

---
 libavcodec/codec2utils.h | 17 ++---
 1 file changed, 14 insertions(+), 3 deletions(-)

diff --git a/libavcodec/codec2utils.h b/libavcodec/codec2utils.h
index 6812ae895c..e1314b05b6 100644
--- a/libavcodec/codec2utils.h
+++ b/libavcodec/codec2utils.h
@@ -23,6 +23,10 @@
 #define AVCODEC_CODEC2UTILS_H
 
 #include 
+#include "config.h"
+#if CONFIG_LIBCODEC2
+#include 
+#endif
 
 //Highest mode we're willing to use.
 //Don't want to let users accidentally produce files that can't be decoded in the future.
@@ -49,9 +53,16 @@
 
 //Used in codec2raw demuxer and libcodec2 encoder
 static inline void codec2_make_extradata(uint8_t *ptr, int mode) {
-//version 0.8 as of 2017-12-23 (r3386)
-ptr[0] = 0; //major
-ptr[1] = 8; //minor
+#if CONFIG_LIBCODEC2
+ptr[0] = CODEC2_VERSION_MAJOR;
+ptr[1] = CODEC2_VERSION_MINOR;
+#else
+// this codepath is only hit when demuxing raw codec2 with libcodec2 disabled,
+// and only relevant when remuxing from raw codec2 to .c2,
+// because version information is ignored except when writing the .c2 header
+ptr[0] = 1; // version 1.2.0 is the latest we know of
+ptr[1] = 2;
+#endif
 ptr[2] = mode;  //mode
 ptr[3] = 0; //flags
 }
-- 
2.39.2

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

2023-12-30 Thread Rémi Denis-Courmont 


Le 30 décembre 2023 15:00:53 GMT+01:00, flow gg  a écrit :
>> At a quick glance, it won't work if the input length is not a multiple of
>the vector length.
>
>Why?

You're not handling tails as far as I see.

> I tried 1024, 32*3, 32*7 and all passed the test.

They're all multiples of the vector length.

>> Also do you really need to extend accumulators to 32 bits?
>
>It won't overflow after the test is changed, so it's not needed anymore.
>I have modified it in this reply.
>
>Rémi Denis-Courmont  于2023年12月30日周六 20:15写道：
>
>>
>>
>> Le 29 décembre 2023 12:57:20 GMT+01:00, flow gg  a
>> écrit :
>> >C908
>> >ssd_int8_vs_int16_c: 207.7
>> >ssd_int8_vs_int16_rvv_i32: 28.0
>>
>> At a quick glance, it won't work if the input length is not a multiple of
>> the vector length.
>>
>> Also do you really need to extend accumulators to 32 bits?
>> ___
>> ffmpeg-devel mailing list
>> ffmpeg-devel@ffmpeg.org
>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>> To unsubscribe, visit link above, or email
>> ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
>>
>___
>ffmpeg-devel mailing list
>ffmpeg-devel@ffmpeg.org
>https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
>To unsubscribe, visit link above, or email
>ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH] libavformat/hlsenc.c: Populate OTI using AAC profile in write_codec_attr.

2023-12-30 Thread Romain Beauxis 
Le jeu. 28 déc. 2023 à 17:26, David Johansen  a écrit :
>>
>> I love this change, but it appears that st->codecpar->profile is always 
>> AV_PROFILE_UNKNOWN when using libfdk_aac as the encoder. Any indications 
>> where I should look for fix that so this can be used with that encoder?
>
>
> It appears that the issue is that profile doesn't default to what's being 
> used so `--profile:a` has to be set explicitly with libfdk_aac and then it 
> works. Not sure if that's an issue worth fixing, but if someone points me to 
> where it needs to be done, then I'd be glad to take a look at fixing it

This feels like a second, separate issue to me?

Maybe we could get these changes in first and then tackle it?

-- Romain
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH] fftool/ffplay: avoid same name in local variable

2023-12-30 Thread Zhao Zhili 
> -Original Message-
> From: ffmpeg-devel  On Behalf Of xufuji456 
> via ffmpeg-devel
> Sent: 2023年12月30日 16:41
> To: ffmpeg-devel@ffmpeg.org
> Cc: xufuji456 <839789...@qq.com>
> Subject: [FFmpeg-devel] [PATCH] fftool/ffplay: avoid same name in local 
> variable
> 
> There is a warning in XCode:"Declaration shadows a local variable"
> 
> Signed-off-by: xufuji456 <839789...@qq.com>
> ---
>  fftools/ffplay.c | 10 +-
>  1 file changed, 5 insertions(+), 5 deletions(-)
> 
> diff --git a/fftools/ffplay.c b/fftools/ffplay.c
> index ea5ff31393..17861e60be 100644
> --- a/fftools/ffplay.c
> +++ b/fftools/ffplay.c
> @@ -1930,11 +1930,11 @@ static int configure_video_filters(AVFilterGraph 
> *graph, VideoState *is, const c
>  if (sd)
>  displaymatrix = (int32_t *)sd->data;
>  if (!displaymatrix) {
> -const AVPacketSideData *sd = 
> av_packet_side_data_get(is->video_st->codecpar->coded_side_data,
> - 
> is->video_st->codecpar->nb_coded_side_data,
> - 
> AV_PKT_DATA_DISPLAYMATRIX);
> -if (sd)
> -displaymatrix = (int32_t *)sd->data;
> +const AVPacketSideData *psd = 
> av_packet_side_data_get(is->video_st->codecpar->coded_side_data,
> +  
> is->video_st->codecpar->nb_coded_side_data,
> +  
> AV_PKT_DATA_DISPLAYMATRIX);
> +if (psd)
> +displaymatrix = (int32_t *)psd->data;
>  }
>  theta = get_rotation(displaymatrix);

LGTM. The commit message could be reworded like "fix a variable shadowing".

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

2023-12-30 Thread flow gg 
Thank you, I learned this and updated it in this reply.

James Almer  于2023年12月30日周六 22:46写道：

> On 12/30/2023 10:59 AM, flow gg wrote:
> > Okay, it has been modified in this reply.
>
> > From d62f363e3aad534c7ead5f3015029b3e7cbbff46 Mon Sep 17 00:00:00 2001
> > From: sunyuechi 
> > Date: Fri, 29 Dec 2023 13:09:21 +0800
> > Subject: [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test
> >
> > ---
> >  tests/checkasm/Makefile   |  1 +
> >  tests/checkasm/checkasm.c |  3 ++
> >  tests/checkasm/checkasm.h |  1 +
> >  tests/checkasm/svq1enc.c  | 68 +++
> >  tests/fate/checkasm.mak   |  1 +
> >  5 files changed, 74 insertions(+)
> >  create mode 100644 tests/checkasm/svq1enc.c
> >
> > diff --git a/tests/checkasm/Makefile b/tests/checkasm/Makefile
> > index 47328b59b9..f507e3 100644
> > --- a/tests/checkasm/Makefile
> > +++ b/tests/checkasm/Makefile
> > @@ -34,6 +34,7 @@ AVCODECOBJS-$(CONFIG_JPEG2000_DECODER)  +=
> jpeg2000dsp.o
> >  AVCODECOBJS-$(CONFIG_OPUS_DECODER)  += opusdsp.o
> >  AVCODECOBJS-$(CONFIG_PIXBLOCKDSP)   += pixblockdsp.o
> >  AVCODECOBJS-$(CONFIG_HEVC_DECODER)  += hevc_add_res.o
> hevc_deblock.o hevc_idct.o hevc_sao.o hevc_pel.o
> > +AVCODECOBJS-$(CONFIG_SVQ1_ENCODER)  += svq1enc.o
> >  AVCODECOBJS-$(CONFIG_TAK_DECODER)   += takdsp.o
> >  AVCODECOBJS-$(CONFIG_UTVIDEO_DECODER)   += utvideodsp.o
> >  AVCODECOBJS-$(CONFIG_V210_DECODER)  += v210dec.o
> > diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
> > index 57613af5f7..ff07d8b18d 100644
> > --- a/tests/checkasm/checkasm.c
> > +++ b/tests/checkasm/checkasm.c
> > @@ -162,6 +162,9 @@ static const struct {
> >  #if CONFIG_PIXBLOCKDSP
> >  { "pixblockdsp", checkasm_check_pixblockdsp },
> >  #endif
> > +#if CONFIG_SVQ1_ENCODER
> > +{ "svq1enc", checkasm_check_svq1enc },
> > +#endif
> >  #if CONFIG_TAK_DECODER
> >  { "takdsp", checkasm_check_takdsp },
> >  #endif
> > diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
> > index d7974772d3..c1f11cfa36 100644
> > --- a/tests/checkasm/checkasm.h
> > +++ b/tests/checkasm/checkasm.h
> > @@ -82,6 +82,7 @@ void checkasm_check_nlmeans(void);
> >  void checkasm_check_opusdsp(void);
> >  void checkasm_check_pixblockdsp(void);
> >  void checkasm_check_sbrdsp(void);
> > +void checkasm_check_svq1enc(void);
> >  void checkasm_check_synth_filter(void);
> >  void checkasm_check_sw_gbrp(void);
> >  void checkasm_check_sw_rgb(void);
> > diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
> > new file mode 100644
> > index 00..065dd798ca
> > --- /dev/null
> > +++ b/tests/checkasm/svq1enc.c
> > @@ -0,0 +1,68 @@
> > +/*
> > + * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences
> (ISCAS).
> > + *
> > + * This file is part of FFmpeg.
> > + *
> > + * FFmpeg is free software; you can redistribute it and/or modify
> > + * it under the terms of the GNU General Public License as published by
> > + * the Free Software Foundation; either version 2 of the License, or
> > + * (at your option) any later version.
> > + *
> > + * FFmpeg is distributed in the hope that it will be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
> > + * GNU General Public License for more details.
> > + *
> > + * You should have received a copy of the GNU General Public License
> along
> > + * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
> > + * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
> > + */
> > +
> > +#include "libavutil/mem.h"
> > +#include "libavutil/mem_internal.h"
> > +
> > +#include "libavcodec/svq1encdsp.h"
> > +
> > +#include "checkasm.h"
> > +
> > +#define BUF_SIZE 1024
> > +
> > +#define randomize(buf, len) \
> > +do { \
> > +for (int i = 0; i < len; i++) \
> > +buf[i] = ((rnd() % 65281) - 32641); \
> > +} while (0)
> > +
> > +static void test_ssd_int8_vs_int16(SVQ1EncDSPContext *s) {
> > +declare_func(int, const int8_t *pix1, const int16_t *pix2, intptr_t
> size);
> > +
> > +int r1, r2;
> > +
> > +if (check_func(s->ssd_int8_vs_int16, "ssd_int8_vs_int16")) {
> > +LOCAL_ALIGNED_32(int8_t, p1, [BUF_SIZE]);
>
> This one should be LOCAL_ALIGNED_4(). See libavcodec/svq1_cb.h
>
> > +LOCAL_ALIGNED_32(int16_t, p2, [BUF_SIZE]);
>
> And this one LOCAL_ALIGNED_16(). See encoded_block_levels[] in
> libavcodec/svq1enc.c
>
> > +
> > +randomize(p1, BUF_SIZE);
> > +randomize(p2, BUF_SIZE);
> > +
> > +r1 = call_ref(p1, p2, BUF_SIZE);
> > +r2 = call_new(p1, p2, BUF_SIZE);
> > +
> > +if (r1 != r2) {
> > +fail();
> > +}
> > +
> > +bench_new(p1, p2, BUF_SIZE);
> > +}
> > +
> > +report("ssd_int8_vs_int16");
> > +
> > +}
> > +
> > +void checkasm_check_svq1enc(void)
> > +{
> > +SVQ1EncDSPContext s = { 0 };
> >

Re: [FFmpeg-devel] [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

2023-12-30 Thread James Almer 

On 12/30/2023 10:59 AM, flow gg wrote:

Okay, it has been modified in this reply.



From d62f363e3aad534c7ead5f3015029b3e7cbbff46 Mon Sep 17 00:00:00 2001
From: sunyuechi 
Date: Fri, 29 Dec 2023 13:09:21 +0800
Subject: [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

---
 tests/checkasm/Makefile   |  1 +
 tests/checkasm/checkasm.c |  3 ++
 tests/checkasm/checkasm.h |  1 +
 tests/checkasm/svq1enc.c  | 68 +++
 tests/fate/checkasm.mak   |  1 +
 5 files changed, 74 insertions(+)
 create mode 100644 tests/checkasm/svq1enc.c

diff --git a/tests/checkasm/Makefile b/tests/checkasm/Makefile
index 47328b59b9..f507e3 100644
--- a/tests/checkasm/Makefile
+++ b/tests/checkasm/Makefile
@@ -34,6 +34,7 @@ AVCODECOBJS-$(CONFIG_JPEG2000_DECODER)  += jpeg2000dsp.o
 AVCODECOBJS-$(CONFIG_OPUS_DECODER)  += opusdsp.o
 AVCODECOBJS-$(CONFIG_PIXBLOCKDSP)   += pixblockdsp.o
 AVCODECOBJS-$(CONFIG_HEVC_DECODER)  += hevc_add_res.o hevc_deblock.o 
hevc_idct.o hevc_sao.o hevc_pel.o
+AVCODECOBJS-$(CONFIG_SVQ1_ENCODER)  += svq1enc.o
 AVCODECOBJS-$(CONFIG_TAK_DECODER)   += takdsp.o
 AVCODECOBJS-$(CONFIG_UTVIDEO_DECODER)   += utvideodsp.o
 AVCODECOBJS-$(CONFIG_V210_DECODER)  += v210dec.o
diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
index 57613af5f7..ff07d8b18d 100644
--- a/tests/checkasm/checkasm.c
+++ b/tests/checkasm/checkasm.c
@@ -162,6 +162,9 @@ static const struct {
 #if CONFIG_PIXBLOCKDSP
 { "pixblockdsp", checkasm_check_pixblockdsp },
 #endif
+#if CONFIG_SVQ1_ENCODER
+{ "svq1enc", checkasm_check_svq1enc },
+#endif
 #if CONFIG_TAK_DECODER
 { "takdsp", checkasm_check_takdsp },
 #endif
diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
index d7974772d3..c1f11cfa36 100644
--- a/tests/checkasm/checkasm.h
+++ b/tests/checkasm/checkasm.h
@@ -82,6 +82,7 @@ void checkasm_check_nlmeans(void);
 void checkasm_check_opusdsp(void);
 void checkasm_check_pixblockdsp(void);
 void checkasm_check_sbrdsp(void);
+void checkasm_check_svq1enc(void);
 void checkasm_check_synth_filter(void);
 void checkasm_check_sw_gbrp(void);
 void checkasm_check_sw_rgb(void);
diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
new file mode 100644
index 00..065dd798ca
--- /dev/null
+++ b/tests/checkasm/svq1enc.c
@@ -0,0 +1,68 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include "libavutil/mem.h"
+#include "libavutil/mem_internal.h"
+
+#include "libavcodec/svq1encdsp.h"
+
+#include "checkasm.h"
+
+#define BUF_SIZE 1024
+
+#define randomize(buf, len) \
+do { \
+for (int i = 0; i < len; i++) \
+buf[i] = ((rnd() % 65281) - 32641); \
+} while (0)
+
+static void test_ssd_int8_vs_int16(SVQ1EncDSPContext *s) {
+declare_func(int, const int8_t *pix1, const int16_t *pix2, intptr_t size);
+
+int r1, r2;
+
+if (check_func(s->ssd_int8_vs_int16, "ssd_int8_vs_int16")) {
+LOCAL_ALIGNED_32(int8_t, p1, [BUF_SIZE]);


This one should be LOCAL_ALIGNED_4(). See libavcodec/svq1_cb.h


+LOCAL_ALIGNED_32(int16_t, p2, [BUF_SIZE]);


And this one LOCAL_ALIGNED_16(). See encoded_block_levels[] in 
libavcodec/svq1enc.c



+
+randomize(p1, BUF_SIZE);
+randomize(p2, BUF_SIZE);
+
+r1 = call_ref(p1, p2, BUF_SIZE);
+r2 = call_new(p1, p2, BUF_SIZE);
+
+if (r1 != r2) {
+fail();
+}
+
+bench_new(p1, p2, BUF_SIZE);
+}
+
+report("ssd_int8_vs_int16");
+
+}
+
+void checkasm_check_svq1enc(void)
+{
+SVQ1EncDSPContext s = { 0 };
+ff_svq1enc_init(&s);
+
+test_ssd_int8_vs_int16(&s);
+}
diff --git a/tests/fate/checkasm.mak b/tests/fate/checkasm.mak
index 9e04b1b3cb..3d775549ee 100644
--- a/tests/fate/checkasm.mak
+++ b/tests/fate/checkasm.mak
@@ -34,6 +34,7 @@ FATE_CHECKASM = fate-checkasm-aacencdsp   
  \
 fate-checkasm-opusdsp   \
 fate-checkasm-pixblockdsp   \
 fate-checkasm-sbrdsp\
+fate-checkasm-svq1enc

Re: [FFmpeg-devel] [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

2023-12-30 Thread flow gg 
flow gg  于2023年12月30日周六 22:00写道：

> > At a quick glance, it won't work if the input length is not a multiple
> of the vector length.
>
> Why? I tried 1024, 32*3, 32*7 and all passed the test.
>
> > Also do you really need to extend accumulators to 32 bits?
>
> It won't overflow after the test is changed, so it's not needed anymore.
> I have modified it in this reply.
>
> Rémi Denis-Courmont  于2023年12月30日周六 20:15写道：
>
>>
>>
>> Le 29 décembre 2023 12:57:20 GMT+01:00, flow gg  a
>> écrit :
>> >C908
>> >ssd_int8_vs_int16_c: 207.7
>> >ssd_int8_vs_int16_rvv_i32: 28.0
>>
>> At a quick glance, it won't work if the input length is not a multiple of
>> the vector length.
>>
>> Also do you really need to extend accumulators to 32 bits?
>> ___
>> ffmpeg-devel mailing list
>> ffmpeg-devel@ffmpeg.org
>> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>>
>> To unsubscribe, visit link above, or email
>> ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
>>
>
From 7bb35b54a4e578848660b1437b96333dd740cbd3 Mon Sep 17 00:00:00 2001
From: sunyuechi 
Date: Fri, 29 Dec 2023 13:27:31 +0800
Subject: [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

C908
ssd_int8_vs_int16_c: 207.7
ssd_int8_vs_int16_rvv_i32: 17.7
---
 libavcodec/riscv/Makefile  |  2 ++
 libavcodec/riscv/svqenc_init.c | 41 +++
 libavcodec/riscv/svqenc_rvv.S  | 44 ++
 libavcodec/svq1enc.c   |  2 ++
 libavcodec/svq1encdsp.h|  1 +
 5 files changed, 90 insertions(+)
 create mode 100644 libavcodec/riscv/svqenc_init.c
 create mode 100644 libavcodec/riscv/svqenc_rvv.S

diff --git a/libavcodec/riscv/Makefile b/libavcodec/riscv/Makefile
index 7f253bba12..4e14c3d094 100644
--- a/libavcodec/riscv/Makefile
+++ b/libavcodec/riscv/Makefile
@@ -46,6 +46,8 @@ RVV-OBJS-$(CONFIG_OPUS_DECODER) += riscv/opusdsp_rvv.o
 OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_init.o
 RV-OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_rvi.o
 RVV-OBJS-$(CONFIG_PIXBLOCKDSP) += riscv/pixblockdsp_rvv.o
+OBJS-$(CONFIG_SVQ1_ENCODER) += riscv/svqenc_init.o
+RVV-OBJS-$(CONFIG_SVQ1_ENCODER) += riscv/svqenc_rvv.o
 OBJS-$(CONFIG_TAK_DECODER) += riscv/takdsp_init.o
 RVV-OBJS-$(CONFIG_TAK_DECODER) += riscv/takdsp_rvv.o
 OBJS-$(CONFIG_UTVIDEO_DECODER) += riscv/utvideodsp_init.o
diff --git a/libavcodec/riscv/svqenc_init.c b/libavcodec/riscv/svqenc_init.c
new file mode 100644
index 00..f4c398960c
--- /dev/null
+++ b/libavcodec/riscv/svqenc_init.c
@@ -0,0 +1,41 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+#include "config.h"
+
+#include "libavutil/attributes.h"
+#include "libavutil/cpu.h"
+#include "libavcodec/svq1encdsp.h"
+
+int ff_ssd_int8_vs_int16_rvv(const int8_t *pix1, const int16_t *pix2,
+  intptr_t size);
+
+av_cold void ff_svq1enc_init_riscv(SVQ1EncDSPContext *c)
+{
+#if HAVE_RVV
+int flags = av_get_cpu_flags();
+
+if (flags & AV_CPU_FLAG_RVV_I32) {
+if (flags & AV_CPU_FLAG_RVB_ADDR) {
+c->ssd_int8_vs_int16 = ff_ssd_int8_vs_int16_rvv;
+}
+}
+#endif
+}
diff --git a/libavcodec/riscv/svqenc_rvv.S b/libavcodec/riscv/svqenc_rvv.S
new file mode 100644
index 00..c52913c2e7
--- /dev/null
+++ b/libavcodec/riscv/svqenc_rvv.S
@@ -0,0 +1,44 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
+ *
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */

Re: [FFmpeg-devel] [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

2023-12-30 Thread flow gg 
> At a quick glance, it won't work if the input length is not a multiple of
the vector length.

Why? I tried 1024, 32*3, 32*7 and all passed the test.

> Also do you really need to extend accumulators to 32 bits?

It won't overflow after the test is changed, so it's not needed anymore.
I have modified it in this reply.

Rémi Denis-Courmont  于2023年12月30日周六 20:15写道：

>
>
> Le 29 décembre 2023 12:57:20 GMT+01:00, flow gg  a
> écrit :
> >C908
> >ssd_int8_vs_int16_c: 207.7
> >ssd_int8_vs_int16_rvv_i32: 28.0
>
> At a quick glance, it won't work if the input length is not a multiple of
> the vector length.
>
> Also do you really need to extend accumulators to 32 bits?
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
>
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

2023-12-30 Thread flow gg 
Okay, it has been modified in this reply.

Martin Storsjö  于2023年12月29日周五 22:35写道：

> On Fri, 29 Dec 2023, James Almer wrote:
>
> > On 12/29/2023 9:16 AM, Martin Storsjö wrote:
> >> On Fri, 29 Dec 2023, flow gg wrote:
> >>
> >>> Tests on x86 might fail, possibly due to a 16-bit sub overflow
> >>
> >> If this only happens in checkasm but not in real life use, it means
> > that
> >> the checkasm test input is out of range, and we should try to mimic the
> >> real input data.
> >>
> >> If the issue is off-by-one due to rounding, we could allow it by making
> >> a comparison per element, and tolerate a certain diff (maybe only on
> > x86).
> >>
> >> But if it is an actual overflow on valid input data, the x86 asm
> >> definitely needs to be fixed, by someone... Worst case, we might want
> > to
> >> disable that checkasm test on x86 temporarily.
> >
> > The function subtracts an int16 from an int8, whose result may not fit
> > on an int16. The x86 asm expects it to do, hence the failures.
> >
> > I don't know if such cases exist in real world files, so like you said,
> > if they don't then the test should generate in-range input. But if they
> > do, i'll fix the x86 asm (sadly, this means sse4 will be required
> > instead of sse2).
>
> I think it's probably best to limit the checkasm inputs that don't
> overflow, for now.
>
> // Martin
> ___
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
>
From d62f363e3aad534c7ead5f3015029b3e7cbbff46 Mon Sep 17 00:00:00 2001
From: sunyuechi 
Date: Fri, 29 Dec 2023 13:09:21 +0800
Subject: [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

---
 tests/checkasm/Makefile   |  1 +
 tests/checkasm/checkasm.c |  3 ++
 tests/checkasm/checkasm.h |  1 +
 tests/checkasm/svq1enc.c  | 68 +++
 tests/fate/checkasm.mak   |  1 +
 5 files changed, 74 insertions(+)
 create mode 100644 tests/checkasm/svq1enc.c

diff --git a/tests/checkasm/Makefile b/tests/checkasm/Makefile
index 47328b59b9..f507e3 100644
--- a/tests/checkasm/Makefile
+++ b/tests/checkasm/Makefile
@@ -34,6 +34,7 @@ AVCODECOBJS-$(CONFIG_JPEG2000_DECODER)  += jpeg2000dsp.o
 AVCODECOBJS-$(CONFIG_OPUS_DECODER)  += opusdsp.o
 AVCODECOBJS-$(CONFIG_PIXBLOCKDSP)   += pixblockdsp.o
 AVCODECOBJS-$(CONFIG_HEVC_DECODER)  += hevc_add_res.o hevc_deblock.o hevc_idct.o hevc_sao.o hevc_pel.o
+AVCODECOBJS-$(CONFIG_SVQ1_ENCODER)  += svq1enc.o
 AVCODECOBJS-$(CONFIG_TAK_DECODER)   += takdsp.o
 AVCODECOBJS-$(CONFIG_UTVIDEO_DECODER)   += utvideodsp.o
 AVCODECOBJS-$(CONFIG_V210_DECODER)  += v210dec.o
diff --git a/tests/checkasm/checkasm.c b/tests/checkasm/checkasm.c
index 57613af5f7..ff07d8b18d 100644
--- a/tests/checkasm/checkasm.c
+++ b/tests/checkasm/checkasm.c
@@ -162,6 +162,9 @@ static const struct {
 #if CONFIG_PIXBLOCKDSP
 { "pixblockdsp", checkasm_check_pixblockdsp },
 #endif
+#if CONFIG_SVQ1_ENCODER
+{ "svq1enc", checkasm_check_svq1enc },
+#endif
 #if CONFIG_TAK_DECODER
 { "takdsp", checkasm_check_takdsp },
 #endif
diff --git a/tests/checkasm/checkasm.h b/tests/checkasm/checkasm.h
index d7974772d3..c1f11cfa36 100644
--- a/tests/checkasm/checkasm.h
+++ b/tests/checkasm/checkasm.h
@@ -82,6 +82,7 @@ void checkasm_check_nlmeans(void);
 void checkasm_check_opusdsp(void);
 void checkasm_check_pixblockdsp(void);
 void checkasm_check_sbrdsp(void);
+void checkasm_check_svq1enc(void);
 void checkasm_check_synth_filter(void);
 void checkasm_check_sw_gbrp(void);
 void checkasm_check_sw_rgb(void);
diff --git a/tests/checkasm/svq1enc.c b/tests/checkasm/svq1enc.c
new file mode 100644
index 00..065dd798ca
--- /dev/null
+++ b/tests/checkasm/svq1enc.c
@@ -0,0 +1,68 @@
+/*
+ * Copyright (c) 2023 Institue of Software Chinese Academy of Sciences (ISCAS).
+ *
+ * This file is part of FFmpeg.
+ *
+ * FFmpeg is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with FFmpeg; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
+ */
+
+#include "libavutil/mem.h"
+#include "libavutil/mem_internal.h"
+
+#include "libavcodec/svq1encdsp.h"
+
+#include "checkasm.h"
+
+#define BUF_SIZE 1024
+
+#define randomize(buf, len) \
+do { \
+for (int i = 0; i < len; i++) \
+

Re: [FFmpeg-devel] [PATCH 3/3] lavc/svq1enc: R-V V ssd_int8_vs_int16

2023-12-30 Thread Rémi Denis-Courmont 


Le 29 décembre 2023 12:57:20 GMT+01:00, flow gg  a écrit :
>C908
>ssd_int8_vs_int16_c: 207.7
>ssd_int8_vs_int16_rvv_i32: 28.0

At a quick glance, it won't work if the input length is not a multiple of the 
vector length.

Also do you really need to extend accumulators to 32 bits?
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

Re: [FFmpeg-devel] [PATCH 2/3] checkasm/svqenc: add ssd_int8_vs_int16 test

2023-12-30 Thread Rémi Denis-Courmont 


Le 29 décembre 2023 12:57:01 GMT+01:00, flow gg  a écrit :
>Tests on x86 might fail, possibly due to a 16-bit sub overflow

I don't know anything about the SVQ encoder. Still, especially for an encoder, 
overflows are probably not expected. So then it is as Martin wrote.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".

[FFmpeg-devel] [PATCH] fftool/ffplay: avoid same name in local variable

2023-12-30 Thread xufuji456 via ffmpeg-devel 
There is a warning in XCode:"Declaration shadows a local variable"

Signed-off-by: xufuji456 <839789...@qq.com>
---
 fftools/ffplay.c | 10 +-
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/fftools/ffplay.c b/fftools/ffplay.c
index ea5ff31393..17861e60be 100644
--- a/fftools/ffplay.c
+++ b/fftools/ffplay.c
@@ -1930,11 +1930,11 @@ static int configure_video_filters(AVFilterGraph 
*graph, VideoState *is, const c
 if (sd)
 displaymatrix = (int32_t *)sd->data;
 if (!displaymatrix) {
-const AVPacketSideData *sd = 
av_packet_side_data_get(is->video_st->codecpar->coded_side_data,
- 
is->video_st->codecpar->nb_coded_side_data,
- 
AV_PKT_DATA_DISPLAYMATRIX);
-if (sd)
-displaymatrix = (int32_t *)sd->data;
+const AVPacketSideData *psd = 
av_packet_side_data_get(is->video_st->codecpar->coded_side_data,
+  
is->video_st->codecpar->nb_coded_side_data,
+  
AV_PKT_DATA_DISPLAYMATRIX);
+if (psd)
+displaymatrix = (int32_t *)psd->data;
 }
 theta = get_rotation(displaymatrix);
 
-- 
2.39.3 (Apple Git-145)

___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel

To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".