These consts are only used in the switch(planar) case located in
avisynth_create_stream_video and nowhere else in the demuxer,
so move them into that function directly.
Signed-off-by: Stephen Hutchinson
---
libavformat/avisynth.c | 22 +++---
1 file changed, 11 insertions(+), 11
As part of this, the mutexes are no longer necessary, and
avisynth_read_close needs to check that avs->avs_library.library
still exists before it attempts to call avisynth_context_destroy
and dlclose.
Signed-off-by: Stephen Hutchinson
---
libavformat/avisynth.c | 133
Am 28.07.24 um 15:15 schrieb Ramiro Polla:
+ void *vslibrary = NULL;
+#ifdef _WIN32
+ const HKEY hkeys[] = {HKEY_CURRENT_USER, HKEY_LOCAL_MACHINE};
+ LONG r;
+ WCHAR vss_path[512];
+ DWORD buf_size = sizeof(vss_path) - 2;
+ char *vss_path_utf8;
+ int i;
+
+ for (i = 0; i <
Am 28.07.24 um 15:09 schrieb Ramiro Polla:
if (st->codecpar->format == AV_PIX_FMT_NONE) {
- av_log(s, AV_LOG_ERROR, "Unsupported VS pixel format %s\n",
info->format->name);
+ if(vs->vsapi->getVideoFormatName(&info->format, vsfmt))
+ av_log(s, AV_LOG_ERROR, "Unsupport
On 7/14/24 1:00 AM, Marth64 wrote:
The description advertises fast as "Default fast search", but
this has not been the default for a long time (current default
is twoloop).
Signed-off-by: Marth64
---
libavcodec/aacenc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/liba
On 2024/07/29 1:30, TADANO Tokumei wrote:
On 2024/07/28 18:26, Michael Niedermayer wrote:
On Sun, Jul 28, 2024 at 01:42:09AM +0900, TADANO Tokumei wrote:
On 2024/07/27 13:30, TADANO Tokumei wrote:
Add an OpenCL filter for filtering GoPro Max native .360 files
into standard equirectangular or
Likely a tighter check can be done
Fixes: signed integer overflow: 3305606804154370442 * 8 cannot be represented
in type 'long'
Fixes:
70449/clusterfuzz-testcase-minimized-ffmpeg_dem_WEBM_DASH_MANIFEST_fuzzer-4771166007918592
Found-by: continuous fuzzing process
https://github.com/google/oss-f
This basically ignores the overflow without undefined behavior, alternatively
we could detect and error out
Fixes: signed integer overflow: 6310596683470275584 + 7660622966157213696
cannot be represented in type 'long'
Fixes:
70433/clusterfuzz-testcase-minimized-ffmpeg_IO_DEMUXER_fuzzer-5483347
On 7/19/24 12:04 PM, Leo Izen wrote:
The PNG specification[1] says that sBIT entries must be at most the bit
depth specified in IHDR, unless the PNG is indexed-color, in which case
sBIT must be between 1 and 8. We should not reject valid sBITs on PNGs
with indexed color.
[1]: https://www.w3.org/
T-Head C908 (cycles):
rv34_idct_dc_add_c: 113.2
rv34_idct_dc_add_rvv_i32: 48.5 (before)
rv34_idct_dc_add_rvv_i32: 39.5 (after)
---
libavcodec/riscv/rv34dsp_rvv.S | 19 +++
1 file changed, 11 insertions(+), 8 deletions(-)
diff --git a/libavcodec/riscv/rv34dsp_rvv.S b/libavcode
---
libavcodec/riscv/Makefile| 1 +
libavcodec/riscv/bswapdsp_init.c | 11 ++-
libavcodec/riscv/bswapdsp_rvvb.S | 52
3 files changed, 63 insertions(+), 1 deletion(-)
create mode 100644 libavcodec/riscv/bswapdsp_rvvb.S
diff --git a/libavcodec/riscv/M
The implementation of ff_read_time() for RISC-V uses rdtime which has
precision on existing hardware too low (!) for benchmarking purposes.
Deleting this implementation falls back on clock_gettime() which was
added as the default ff_read_time() implementation in 33e4cc9.
Below are metrics gathere
On 2024/07/28 18:26, Michael Niedermayer wrote:
On Sun, Jul 28, 2024 at 01:42:09AM +0900, TADANO Tokumei wrote:
On 2024/07/27 13:30, TADANO Tokumei wrote:
Add an OpenCL filter for filtering GoPro Max native .360 files
into standard equirectangular or youtube equiangular cubemap (eac)
projectio
On 7/28/24 9:38 AM, Ramiro Polla wrote:
@@ -1134,6 +1089,7 @@ static av_cold int avisynth_read_close(AVFormatContext *s)
return AVERROR_UNKNOWN;
avisynth_context_destroy(s->priv_data);
+dlclose(avs_library.library);
Maybe it's best to wrap this around an if (avs_library.lib
On 28/07/2024 15:06, Tong Wu wrote:
Tong Wu:
Subject: [FFmpeg-devel][PATCH 2/2] lavc/d3d12va_encode: trim header
alignment at output
It is d3d12va's requirement that the FrameStartOffset must be aligned as per
hardware limitation. However, we could trim this alignment at output to reduce
coded
Ping, thank you
___
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".
I have found a whitespace error here and need to restore 2 lines of logic,
but will wait a few days for comment before putting up v2.
___
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, v
Hi Sean_McG, good day,
First of all, there is a typo in the commit message. It should say
"at the mercy of dvdnav_time_search()"
> I think avpriv_report_missing_feature() might be more appropriate here.
But there is no missing feature, the implementation is working great
with dvdnav_time_search()
> On Jun 12, 2024, at 21:42, Matthieu Bouron wrote:
>
> ---
> configure | 14 ++
> libavcodec/Makefile | 7 +
> libavcodec/allcodecs.c| 7 +
> libavcodec/mediacodecdec.c| 215 ++-
> libavcodec/mediacodecdec_common.c | 3
Hi Marth64,
On Sun, Jul 28, 2024 at 3:35 AM Marth64 wrote:
>
> Player applications can now enjoy seeking while playing back
> a title. Accuracy is at the mercy of what libdvdnav offers,
> which is currently dvdnav_jump_to_sector_by_time().
>
> Signed-off-by: Marth64
> ---
> libavformat/dvdvideo
Is it possible to use this without using the portals API and without
systemd? That would be much appreciated if so since the portal is not very
flexible. As for systemd it would be great to be able to use this on
non-systemd platforms.
___
ffmpeg-devel ma
On Fri, Jul 19, 2024 at 7:51 PM Stephen Hutchinson wrote:
>
> The atexit() handler in the avisynth demuxer was added because
> there was a conflict in AvxSynth that arose due to their use
> of C++ global objects, particularly in relation to having
> added a logging function relying on log4cpp.
>
>
On 2024-07-23 16:59, Stefan Oltmanns via ffmpeg-devel wrote:
This is the second part for loading the library at runtime, changes
compared to previous patch revisions:
-No atexit anymore
-No global states anymore
-Moved the registry read for Windows from a separate function inside the
function
On 2024-07-23 16:51, Stefan Oltmanns via ffmpeg-devel wrote:
this is revised patch, this is the first part that just updates to the
API v4 of VapourSynth.
Changes in API v4:
-All functions previously in header are now part of the "vssapi" object
-Renames of different types and functions
-YCoCg
Tong Wu:
>Subject: [FFmpeg-devel][PATCH 2/2] lavc/d3d12va_encode: trim header
>alignment at output
>
>It is d3d12va's requirement that the FrameStartOffset must be aligned as per
>hardware limitation. However, we could trim this alignment at output to reduce
>coded size. A aligned_header_size is ad
On Sun, 28 Jul 2024, Rémi Denis-Courmont wrote:
Le 28 juillet 2024 07:37:51 GMT+03:00, Brad Smith
a écrit :
On 2024-07-26 7:56 a.m., Rémi Denis-Courmont wrote:
Le 26 juillet 2024 13:58:34 GMT+03:00, Brad Smith
a écrit :
aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBS
From: Niklas Haas
In particular, validate that the chosen compression level is compatible
with the chosen profile.
---
libavcodec/dovi_rpu.h| 1 +
libavcodec/dovi_rpuenc.c | 29 -
2 files changed, 29 insertions(+), 1 deletion(-)
diff --git a/libavcodec/dovi_rpu.
From: Niklas Haas
Slightly re-organize the logic around extension blocks in order to allow
expanding the state tracking in a following commit.
---
libavcodec/dovi_rpu.c| 1 -
libavcodec/dovi_rpu.h| 9 +++--
libavcodec/dovi_rpudec.c | 35 ++-
libavcod
From: Niklas Haas
Keyframes must reset the metadata compression state, so we need to
also signal this at rpu generation time.
Default to uncompressed, because encoders cannot generally know if
a given frame will be a keyframe before they finish encoding, but also
cannot retroactively attach the
From: Niklas Haas
More flexible version of ff_dovi_configure() which does not require an
AVCodecContext. Usable, for example, inside a bitstream filter.
---
libavcodec/dovi_rpu.h| 16 +++-
libavcodec/dovi_rpuenc.c | 85 +++-
2 files changed, 72 inserti
From: Niklas Haas
Provides direct access to the AVDOVIMetadata without having to attach it
to a frame.
---
libavcodec/dovi_rpu.h| 9 +
libavcodec/dovi_rpudec.c | 40 +++-
2 files changed, 36 insertions(+), 13 deletions(-)
diff --git a/libavcodec/
From: Niklas Haas
Will be used to control compression, encapsulation etc.
---
libavcodec/dovi_rpu.h| 2 +-
libavcodec/dovi_rpuenc.c | 2 +-
libavcodec/libaomenc.c | 2 +-
libavcodec/libsvtav1.c | 2 +-
libavcodec/libx265.c | 3 ++-
5 files changed, 6 insertions(+), 5 deletions(-)
di
From: Niklas Haas
And move the choice of desired container to `flags`. This is needed to
handle differing API requirements (e.g. libx265 requires the NAL RBSP,
but CBS BSF requires the unescaped bytes).
---
libavcodec/dovi_rpu.h| 16 ++--
libavcodec/dovi_rpuenc.c | 22 ++-
From: Niklas Haas
While this is technically a spec violation, the result is still
decodable (and will look perfectly fine to clients ignoring Dolby Vision
metadata). It will also only happen in garbage in, garbage out scenarios.
---
libavcodec/libsvtav1.c | 4 ++--
1 file changed, 2 insertions(+
From: Niklas Haas
This implements limited metadata compression. To be a bit more lenient,
we try and re-order the static extension blocks when testing for an
exact match.
For sanity, and to avoid producing bitstreams we couldn't ourselves
decode, we don't accept partial matches - if some extensi
From: Niklas Haas
See previous commit for justification
---
libavcodec/libx265.c | 8 +---
1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/libavcodec/libx265.c b/libavcodec/libx265.c
index 718bd21b20..325c4fdd07 100644
--- a/libavcodec/libx265.c
+++ b/libavcodec/libx265.c
@@ -7
From: Niklas Haas
In the absence of an RPU header, we can consult the colorspace tags to
make a more informed guess about whether we're looking at profile 5 or
profile 8.
---
libavcodec/dovi_rpuenc.c | 15 ++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/libavcodec/d
From: Niklas Haas
Better safe than sorry.
---
libavcodec/dovi_rpudec.c | 4
1 file changed, 4 insertions(+)
diff --git a/libavcodec/dovi_rpudec.c b/libavcodec/dovi_rpudec.c
index 9f295d4fe4..878950d66d 100644
--- a/libavcodec/dovi_rpudec.c
+++ b/libavcodec/dovi_rpudec.c
@@ -482,6 +482,10 @
From: Niklas Haas
---
libavcodec/dovi_rpudec.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/dovi_rpudec.c b/libavcodec/dovi_rpudec.c
index 6ef7a88ffd..3465665961 100644
--- a/libavcodec/dovi_rpudec.c
+++ b/libavcodec/dovi_rpudec.c
@@ -289,7 +289,7 @@ static int
From: Niklas Haas
Some DM types do not fill the whole struct, so just clear it entirely
before going filling the decoded values.
---
libavcodec/dovi_rpudec.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavcodec/dovi_rpudec.c b/libavcodec/dovi_rpudec.c
index 91c9e41926..9f295d4fe4 10064
From: Niklas Haas
This implements the limited DM metadata compression scheme described in
chapter 9 of the dolby vision bitstream specification.
The spec is a bit unclear about how to handle the presence of static
metadata inside compressed frames; in that it doesn't explicitly forbid
an encoder
From: Niklas Haas
Static and dynamic extension blocks are handled differently by metadata
compression, so we need to separate the extension block array into two.
---
libavcodec/dovi_rpu.h| 20 ++--
libavcodec/dovi_rpudec.c | 31 +--
libavcodec/dovi
From: Niklas Haas
This can be used to strip dovi metadata, or enable/disable dovi
metadata compression. Possibly more use cases in the future.
---
configure | 1 +
doc/bitstream_filters.texi | 23 +++
libavcodec/bitstream_filters.c | 1 +
libavcodec/bsf/Makefile
From: Niklas Haas
Limited mode can only ever maintain a single VDR RPU reference, and
furthermore requires vdr_rpu_id == 0. So in practice, it will only ever
use VDR RPU slot 0. All remaining slots get flushed in this case, to
avoid leaking partial state.
---
libavcodec/dovi_rpuenc.c | 26 ++
From: Niklas Haas
---
libavutil/dovi_meta.h | 28 +++-
1 file changed, 15 insertions(+), 13 deletions(-)
diff --git a/libavutil/dovi_meta.h b/libavutil/dovi_meta.h
index c942d0e133..5e8a1e43d7 100644
--- a/libavutil/dovi_meta.h
+++ b/libavutil/dovi_meta.h
@@ -301,26 +301
From: Niklas Haas
This struct itself contains vdr_rpu_id, so we can never match it except
in the case of i == vdr_rpu_id. So just directly use this ID.
---
libavcodec/dovi_rpuenc.c | 11 +--
1 file changed, 1 insertion(+), 10 deletions(-)
diff --git a/libavcodec/dovi_rpuenc.c b/libavcod
From: Niklas Haas
Add some error checking. I've limited it to AV_EF_CAREFUL and
AV_EF_COMPLIANT for now, because we can technically decode such RPUs
just fine.
---
libavcodec/dovi_rpudec.c | 15 +++
1 file changed, 15 insertions(+)
diff --git a/libavcodec/dovi_rpudec.c b/libavcodec/
From: Niklas Haas
As the comment implies, DOVIContext.ext_blocks should also reflect the
current state after ff_dovi_rpu_generate().
Fluff for now, but will be needed once we start implementing metadata
compression for extension blocks as well.
---
libavcodec/dovi_rpuenc.c | 12
1
On Sun, Jul 28, 2024 at 01:42:09AM +0900, TADANO Tokumei wrote:
>
> On 2024/07/27 13:30, TADANO Tokumei wrote:
> > Add an OpenCL filter for filtering GoPro Max native .360 files
> > into standard equirectangular or youtube equiangular cubemap (eac)
> > projection.
> >
> > The .360 file contains s
Signed-off-by: Marth64
---
doc/demuxers.texi | 11 +--
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/doc/demuxers.texi b/doc/demuxers.texi
index 04293c4813..d67be0312e 100644
--- a/doc/demuxers.texi
+++ b/doc/demuxers.texi
@@ -384,7 +384,7 @@ Default is 0, the first langua
Signed-off-by: Marth64
---
libavformat/dvdvideodec.c | 13 ++---
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/libavformat/dvdvideodec.c b/libavformat/dvdvideodec.c
index afc7836038..04c5cbea8e 100644
--- a/libavformat/dvdvideodec.c
+++ b/libavformat/dvdvideodec.c
@@ -21,7
Signed-off-by: Marth64
---
libavformat/dvdvideodec.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/libavformat/dvdvideodec.c b/libavformat/dvdvideodec.c
index a908acfa7f..afc7836038 100644
--- a/libavformat/dvdvideodec.c
+++ b/libavformat/dvdvideodec.c
@@ -44,8 +44,6 @@
#include "libavuti
Menus can have chapter markers by way of cells, and empty menus
can be detected with existing functionality. Menu selection
also is in need of usability improvement.
Implement chapter markers for menus, let the trim function detect
empty menus, remove the unnecessary ceremony around the -pg
option
The methods are effectively the same but reference different
public enums, so roll them up into a macro.
Signed-off-by: Marth64
---
libavformat/dvdvideodec.c | 64 +++
1 file changed, 24 insertions(+), 40 deletions(-)
diff --git a/libavformat/dvdvideodec.c b/
Player applications can now enjoy seeking while playing back
a title. Accuracy is at the mercy of what libdvdnav offers,
which is currently dvdnav_jump_to_sector_by_time().
Signed-off-by: Marth64
---
libavformat/dvdvideodec.c | 93 ++-
1 file changed, 82 inser
DVDs naturally consist of segmented MPEG-PS blobs within a VOB
(i.e. VOBs are not linear). NAV packs set the segment boundaries.
When switching between segments, discontinuities occur and thus
the subdemuxer needs to be reset. The current approach to manage
this is by invoking ff_read_frame_flush()
This patch set starts with a critical bug fix to resolve packet drops,
then follows on with seeking support and menu chapter marker support.
In addition, some cleanup occurs and documentation/logging is improved.
Seeking can be tested with mpv,
```
/mpv "av://dvdvideo:$IN"
```
Compare/view on Git
Le 28 juillet 2024 07:37:51 GMT+03:00, Brad Smith
a écrit :
>On 2024-07-26 7:56 a.m., Rémi Denis-Courmont wrote:
>>
>> Le 26 juillet 2024 13:58:34 GMT+03:00, Brad Smith
>> a écrit :
>>> aarch64: Implement support for elf_aux_info(3) on FreeBSD and OpenBSD
>>>
>>> FreeBSD 12.0+, OpenBSD -cur
58 matches
Mail list logo
