Empty MessageFrom 89743ce8da188800536c78b46fc3ee43e1538fb5 Mon Sep 17 00:00:00 2001
From: asivery
Date: Wed, 3 Jul 2024 16:25:05 +0200
Subject: [PATCH 2/2] fate/atrac1: add test for 4-track file
Signed-off-by: asivery
---
tests/fate/atrac.mak | 4
1 file changed, 4 insertions(+)
diff --gi
dt
wrote:
> asivery via ffmpeg-devel:
>
> > I'm sending the patch again, so that it is correctly rebased on the current
> > master.
> > Here are the two sample files required by the FATE test:
> > https://0x0.st/Xaw2.aea/boxboy333_house_mus
The issue with the old mechanism is that we had to introduce new
API each time we needed a new queue family, and all the queue families
were functionally fixed to a given purpose.
Nvidia's GPUs are able to handle video encoding and compute on the
same queue, which results in a speedup when pre-pro
---
libavutil/vulkan.c | 68 ++
libavutil/vulkan.h | 2 +-
2 files changed, 21 insertions(+), 49 deletions(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index e0208c5a7c..d98e863711 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@
The issue with the old mechanism is that we had to introduce new
API each time we needed a new queue family, and all the queue families
were functionally fixed to a given purpose.
Nvidia's GPUs are able to handle video encoding and compute on the
same queue, which results in a speedup when pre-pro
---
libavutil/vulkan.c | 68 ++
libavutil/vulkan.h | 2 +-
2 files changed, 21 insertions(+), 49 deletions(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index e0208c5a7c..d98e863711 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@
On 09/07/2024 08:57, Anton Khirnov wrote:
Quoting Lynne via ffmpeg-devel (2024-07-09 03:07:12)
@@ -151,6 +162,17 @@ typedef struct AVVulkanDeviceContext {
* Similar to lock_queue(), unlocks a queue. Must only be called after
locking.
*/
void (*unlock_queue)(struct
The issue with the old mechanism is that we had to introduce new
API each time we needed a new queue family, and all the queue families
were functionally fixed to a given purpose.
Nvidia's GPUs are able to handle video encoding and compute on the
same queue, which results in a speedup when pre-pro
On 10/07/2024 10:18, Anton Khirnov wrote:
Quoting Lynne via ffmpeg-devel (2024-07-10 01:56:57)
On 09/07/2024 08:57, Anton Khirnov wrote:
Quoting Lynne via ffmpeg-devel (2024-07-09 03:07:12)
@@ -151,6 +162,17 @@ typedef struct AVVulkanDeviceContext {
* Similar to lock_queue(), unlocks
The search of the current DirectShow device list has been customized so
that audio devices are always found even if no video device is connected.
Signed-off-by: Jens Frederich
---
libavdevice/dshow.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavdevice/dshow.c b/libavd
This allows us to support different video ops on different queues,
as well as any other arbitrary queues we need.
---
libavutil/hwcontext_vulkan.c | 262 ++-
1 file changed, 167 insertions(+), 95 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hw
Lets us implement FPS conversion.
---
libavutil/hwcontext_vulkan.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 5baf68660a..19338cc069 100644
--- a/libavutil/hwcontext_vulkan.c
+++ b/libavutil/hwcontext_vulkan.c
@@ -1177,6 +1
s fix in 5.1 release branch, is this possible?
Jens
On Mon, Jul 15, 2024 at 12:51 AM patches via ffmpeg-devel
wrote:
>
> The search of the current DirectShow device list has been customized
> so that audio devices are always found even if no video device is connected.
>
> Signed
The code predates ff_vk_get_pooled_buffer().
---
libavcodec/vulkan_decode.c | 47 +++---
libavcodec/vulkan_decode.h | 2 +
libavcodec/vulkan_video.c | 82 --
libavcodec/vulkan_video.h | 7
4 files changed, 34 insertions(+), 104 deletions
This commit was long overdue. The old transfer dubiously tried to
merge as much code as possible, and had very little in the way
of optimizations, apart from basic host-mapping.
The new code uses buffer pools for any temporary bufflers, and
handles falling back to buffer-based uploads if host-mapp
On 19/07/2024 15:45, Tong Wu wrote:
From: Tong Wu
Sent: 2024年7月8日 23:13
To: [email protected]
Cc: Tong Wu
Subject: [FFmpeg-devel][PATCH 1/2] lavc/hw_base_encode: correct the
timestamp when input_order = decode_delay
Fixed the command line: ffmpeg -hwaccel vaapi -pix_fmt nv12 -s:v
width
On 20/07/2024 01:42, Michael Niedermayer wrote:
On Thu, May 30, 2024 at 04:37:08AM +0200, Lynne via ffmpeg-devel wrote:
This commit adds a decoder for the frequency-domain part of USAC.
Changes over version 4:
- Actually reset entropy decoding upon configuration.
- Support for LFE channels
The __builtin_available function does not do compile time check
for the availablity of the CVBufferCopyAttachments function
which will fail the build. Check the availability during configure.
Signed-off-by: Gnattu OC
---
configure | 2 ++
1 file changed, 2 insertions(+)
diff --git a/configure b
The __builtin_available function does not do compile time check
for the availablity of the CVBufferCopyAttachments function
which will fail the build. Check the availability during configure.
Signed-off-by: Gnattu OC
---
configure | 2 ++
libavutil/hwcontext_videotoolbo
On 22/07/2024 00:53, James Almer wrote:
Signed-off-by: James Almer
---
libavcodec/h2645_sei.c | 37 +
libavcodec/h2645_sei.h | 5 +
libavcodec/itut35.h| 2 ++
3 files changed, 44 insertions(+)
diff --git a/libavcodec/h2645_sei.c b/libavcodec/h2
The issue is that ac->output_elements is populated from
ac->che, which may be freed, leaving dangling pointers in this
list.
Should fix clusterfuzz.
---
libavcodec/aac/aacdec.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavcodec/aac/aacdec.c b/libavcodec/aac/aacdec.c
index ea2ba84a80..
vices if
no video device is present
[You don't often get email from [email protected]. Learn why this is
important at https://aka.ms/LearnAboutSenderIdentification ]
LGTM
On Mon, Jul 15, 2024 at 12:51 AM patches via ffmpeg-devel
wrote:
>
> The search of the current DirectShow
-Original Message-
From: Roger Pack
Sent: Monday, July 22, 2024 5:52 PM
To: patches
Cc: FFmpeg development discussions and patches
Subject: Re: [FFmpeg-devel] [PATCH] avdevice/dshow: Don't skip audio devices if
no video device is present
On Wed, Jul 17, 2024 at 1:43 AM patches wrot
-Original Message-
From: Roger Pack
Sent: Monday, July 22, 2024 5:52 PM
To: patches
Cc: FFmpeg development discussions and patches
Subject: Re: [FFmpeg-devel] [PATCH] avdevice/dshow: Don't skip audio devices if
no video device is present
On Wed, Jul 17, 2024 at 1:43 AM patches wrote
08:21:31AM +, patches via ffmpeg-devel wrote:
>
> -Original Message-
> From: Roger Pack
> Sent: Monday, July 22, 2024 5:52 PM
> To: patches
> Cc: FFmpeg development discussions and patches
>
> Subject: Re: [FFmpeg-devel] [PATCH] avdevice/dshow: Don't skip a
Its not feasible to add an AVClass in the main context, as
it would waste space, as the main context is recursive, and
every bit of assembly would need to be changed.
While its true that on paper av_log has access to the main
context, that functionality is not used as no options are
available for
On 26/07/2024 10:22, Andreas Rheinhardt wrote:
Lynne via ffmpeg-devel:
Its not feasible to add an AVClass in the main context, as
it would waste space, as the main context is recursive, and
every bit of assembly would need to be changed.
While its true that on paper av_log has access to the
On 26/07/2024 09:47, Anton Khirnov wrote:
Quoting Lynne via ffmpeg-devel (2024-07-26 08:42:11)
Its not feasible to add an AVClass in the main context, as
it would waste space, as the main context is recursive, and
every bit of assembly would need to be changed.
While its true that on paper
On 26/07/2024 11:03, Andreas Rheinhardt wrote:
Lynne via ffmpeg-devel:
On 26/07/2024 10:22, Andreas Rheinhardt wrote:
Lynne via ffmpeg-devel:
Its not feasible to add an AVClass in the main context, as
it would waste space, as the main context is recursive, and
every bit of assembly would need
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
On 31/07/2024 16:35, James Almer wrote:
Fixes relevant warnings when compiling with --enable-hardcoded-tables
Signed-off-by: James Almer
---
libavcodec/aacps_tablegen_template.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavcodec/aacps_tablegen_template.c
b/libavcodec/aacps_tab
On 31/07/2024 21:54, Michael Niedermayer wrote:
Fixes: out of array access
Fixes:
70734/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_LATM_fuzzer-4741427068731392
Found-by: continuous fuzzing process
https://github.com/google/oss-fuzz/tree/master/projects/ffmpeg
Signed-off-by: Michael
From: Gnattu OC
Allocate a dedicated attachments dictionary instead of trying to get
one from the pixel buffer. The attachments got from the pixel buffer
confuses the CVImageBufferCreateColorSpaceFromAttachments method and
will make it to output a wrong colorspace that causes problem like
#10884.
The color trc key was assigned a color primaries value which causes
the resulting colorspace is always SDR.
Fixes #10884.
Signed-off-by: Gnattu OC
---
libavutil/hwcontext_videotoolbox.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_videotoolbox.c
b/lib
On 06/08/2024 11:06, Jonas Karlman wrote:
Add a hwdevice type for V4L2 Request API with transfer_data_from support
for AV_PIX_FMT_DRM_PRIME, based on AV_HWDEVICE_TYPE_DRM.
AVV4L2RequestDeviceContext.media_fd can be set by the application or a
media device path can be supplied when hwdevice is cr
The issue with the old mechanism is that we had to introduce new
API each time we needed a new queue family, and all the queue families
were functionally fixed to a given purpose.
Nvidia's GPUs are able to handle video encoding and compute on the
same queue, which results in a speedup when pre-pro
---
libavutil/vulkan.c | 68 ++
libavutil/vulkan.h | 2 +-
2 files changed, 21 insertions(+), 49 deletions(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index 11c17ee6f3..cec8354ba6 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@
This allows us to support different video ops on different queues,
as well as any other arbitrary queues we need.
---
libavutil/hwcontext_vulkan.c | 262 ++-
1 file changed, 167 insertions(+), 95 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hw
Lets us implement FPS conversion.
---
libavutil/hwcontext_vulkan.c | 18 +++---
libavutil/vulkan.c | 6 +-
libavutil/vulkan.h | 1 +
libavutil/vulkan_functions.h | 8
libavutil/vulkan_loader.h| 1 +
5 files changed, 30 insertions(+), 4 deletions
The code predates ff_vk_get_pooled_buffer().
---
libavcodec/vulkan_decode.c | 41 ---
libavcodec/vulkan_decode.h | 2 +
libavcodec/vulkan_video.c | 82 --
libavcodec/vulkan_video.h | 15 ---
4 files changed, 28 insertions(+), 112 deletions
---
libavutil/hwcontext_vulkan.c | 13 -
1 file changed, 13 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index c81fc95af2..e2ef599a0d 100644
--- a/libavutil/hwcontext_vulkan.c
+++ b/libavutil/hwcontext_vulkan.c
@@ -67,19 +67,6 @@
#define CHECK
This commit was long overdue. The old transfer dubiously tried to
merge as much code as possible, and had very little in the way
of optimizations, apart from basic host-mapping.
The new code uses buffer pools for any temporary bufflers, and
handles falling back to buffer-based uploads if host-mapp
The struct data seem to get corrupted otherwise.
Possibly a validation layer or libvulkan issue.
---
libavutil/hwcontext_vulkan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 443862be3b..07c2a970a8 100644
---
---
libavutil/hwcontext_vulkan.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 07c2a970a8..1e46ee2cee 100644
--- a/libavutil/hwcontext_vulkan.c
+++ b/libavutil/hwcontext_vulkan.c
@@ -1399,6 +1399,8 @@ static int
vulkan_devi
The issue is that shaderc_result_get_num_errors may sometime
return 0 even when shaderc_result_get_compilation_status returns
a non-zero error code.
Since we use the result from the former, override the status
if it returned 0.
---
libavfilter/vulkan_shaderc.c | 3 +++
1 file changed, 3 insertions
---
libavfilter/vulkan_shaderc.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavfilter/vulkan_shaderc.c b/libavfilter/vulkan_shaderc.c
index 455e81767e..9e8a3d17ac 100644
--- a/libavfilter/vulkan_shaderc.c
+++ b/libavfilter/vulkan_shaderc.c
@@ -51,6 +51,7 @@ static int shdc_shader_compil
We'd like to use it eventually, and its already covered by
the minimum version of the headers we require.
---
libavutil/hwcontext_vulkan.c | 14 --
libavutil/vulkan_functions.h | 7 ++-
libavutil/vulkan_loader.h| 1 +
3 files changed, 19 insertions(+), 3 deletions(-)
diff --
This would've let to a segfault if custom allocators were used.
---
libavutil/vulkan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index df7758cc1e..7b45e43a89 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@ -855,7 +855,7
---
libavutil/hwcontext_vulkan.c | 10 ++
libavutil/vulkan_functions.h | 7 +++
libavutil/vulkan_loader.h| 3 +++
3 files changed, 20 insertions(+)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index df7f42d995..9576e2393a 100644
--- a/libavutil/hwcont
---
libavutil/vulkan.c | 26 --
1 file changed, 16 insertions(+), 10 deletions(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index 7b45e43a89..13344b7aed 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@ -281,6 +281,15 @@ int ff_vk_exec_pool_init(FFVulk
In vtenc_populate_extradata, supported_props should always be released
to avoid memory leak.
Regression from cd2f8a22e94700c68b1de7968df11e8bebfd315b
Signed-off-by: gnattu
---
libavcodec/videotoolboxenc.c | 13 ++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/libavco
Hardware frames with RGB colorspace will not have a YCbCrMatrixKey.
Currently, it will spam the console with warning if rgb frame is
uploaded.
Signed-off-by: Gnattu OC
---
libavutil/hwcontext_videotoolbox.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_v
---
doc/APIchanges | 9 +
libavutil/hwcontext_vulkan.c | 8
libavutil/hwcontext_vulkan.h | 12
libavutil/version.h | 3 ++-
4 files changed, 31 insertions(+), 1 deletion(-)
diff --git a/doc/APIchanges b/doc/APIchanges
index 046828ded1..173f3
On 13/08/2024 16:03, J. Dekker wrote:
Port dav1d's checkasm output format to FFmpeg's checkasm, includes
relative speedups and aligns results.
Signed-off-by: J. Dekker
---
tests/checkasm/checkasm.c | 53 +++
1 file changed, 48 insertions(+), 5 deletions(-)
The issue is that VulkanContext mostly always used the AVClass *
from its structure, which we don't set in decode.
---
libavcodec/vulkan_decode.c| 2 ++
libavfilter/vf_avgblur_vulkan.c | 2 ++
libavfilter/vf_blend_vulkan.c | 2 ++
libavfilter/vf_bwdif_vulkan.c | 4 +--
libavf
In 680d969a305c0927480573a1b455024088b51aeb, the new API was
used to find a queue family for dispatch, but the found queue
family was not used for decoding, just for dispatching.
---
libavcodec/vulkan_decode.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavcodec/vulkan_de
The validation layer option only supported GPU-assisted validation.
This is mutually exclusive with shader debug printfs, so we need to
differentiate between the two.
This also fixes issues with user-given layers, and leaks in case of
errors.
---
libavutil/hwcontext_vulkan.c | 215 +++
It was added to Vulkan 1.1 a long time ago.
Validation layer will warn if this is enabled.
---
libavutil/hwcontext_vulkan.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 506629141e..2c958b86bb 100644
--- a/libavutil/hwcontext_v
The issue is that enabling features requires that the device
extension is supported. The extensions bitfield was set later,
so it was always 0, leading to no features being added.
---
libavutil/hwcontext_vulkan.c | 73 +++-
1 file changed, 38 insertions(+), 35 delet
On 14/08/2024 14:41, Michael Niedermayer wrote:
Hi all
Are there any upcoming LTS releases that want to/could include FFmpeg 7.1 ?
If so please reply here and list the date before which we would have to
finish the 7.1 release so it can be included with no problems
Otherwise, are there any prefe
We require it for encoding.
---
libavutil/hwcontext_vulkan.c | 12 +++-
libavutil/vulkan_functions.h | 1 +
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 3e562716d8..a82355e8bf 100644
--- a/libavutil/hwcont
Vulkan encoding was designed in a very... consolidated way.
You had to know the exact codec and profile that the image was going to
eventually be encoded as at... image creation time. Unfortunately, as good
as our code is, glimpsing into the exact future isn't what its capable of.
video_maintenanc
On 14/08/2024 12:18, Andreas Rheinhardt wrote:
Lynne via ffmpeg-devel:
The issue is that VulkanContext mostly always used the AVClass *
from its structure, which we don't set in decode.
---
libavcodec/vulkan_decode.c| 2 ++
libavfilter/vf_avgblur_vulkan.c | 2 ++
libavf
The color range should be set to match the input when creating
the VideoToolbox context. Otherwise, the new context will default
to limited range, creates inconsistencies with full range inputs.
Signed-off-by: Gnattu OC
---
libavfilter/vf_scale_vt.c | 1 +
libavfilter/vf_transpose_vt.c
This commit cleans up and refactors the mess of private state upon
private state that used to be.
Now, FFHWBaseEncodePicture is fully initialized upon call-time,
and, most importantly, this lets APIs which require initialization
data for frames (VkImageViews) to initialize this for both the
input
The sentence was added as an advice, given the maturity of the xHE-AAC
implementation, but didn't deserve its own paragraph, nor its subject
deserved being left a referential footnote in the stead of article.
---
src/index | 3 ---
1 file changed, 3 deletions(-)
diff --git a/src/index b/src/index
On 19/08/2024 17:07, Tong Wu wrote:
Lynne:
Subject: [PATCH] hw_base_encode: refactor picture allocation/freeing
This commit cleans up and refactors the mess of private state upon private state
that used to be.
Now, FFHWBaseEncodePicture is fully initialized upon call-time, and, most
importantl
This commit cleans up and refactors the mess of private state upon
private state that used to be.
Now, FFHWBaseEncodePicture is fully initialized upon call-time,
and, most importantly, this lets APIs which require initialization
data for frames (VkImageViews) to initialize this for both the
input
On 16/08/2024 12:48, J. Dekker wrote:
Lynne via ffmpeg-devel writes:
On 13/08/2024 16:03, J. Dekker wrote:
Port dav1d's checkasm output format to FFmpeg's checkasm, includes
relative speedups and aligns results.
Signed-off-by: J. Dekker
---
tests/checkasm/check
On 20/08/2024 14:13, [email protected] wrote:
More generally, which is the recommanded way to cut a video with a specific
starting point and specific length, with minimal re-encoding?
Millions of hours of CPU-time are probably wasted to reencode
already-perfectly-encoded content, just for cutting ;)
On 20/08/2024 22:33, Michael Niedermayer wrote:
On Tue, Aug 20, 2024 at 02:36:53PM +0200, Lynne via ffmpeg-devel wrote:
On 20/08/2024 14:13, [email protected] wrote:
More generally, which is the recommanded way to cut a video with a specific
starting point and specific length, with minimal re
On 29/08/2024 11:34, Anton Khirnov wrote:
---
libavcodec/Makefile | 10 ++-
libavcodec/aarch64/opusdsp_init.c | 2 +-
libavcodec/opus/Makefile | 30 +++
libavcodec/{opus_celt.c => opus/celt.c} | 6 ++--
l
---
libavcodec/vulkan_video.c | 49 +++
libavcodec/vulkan_video.h | 8 +++
2 files changed, 57 insertions(+)
diff --git a/libavcodec/vulkan_video.c b/libavcodec/vulkan_video.c
index b9a0ed5022..a676f0332a 100644
--- a/libavcodec/vulkan_video.c
+++ b/libavc
---
libavcodec/Makefile | 2 +-
libavcodec/hw_base_encode_h264.c | 265 +++
libavcodec/hw_base_encode_h264.h | 53 +++
libavcodec/vaapi_encode_h264.c | 264 +++---
4 files changed, 343 insertions(+), 241 deletions(-)
create
Vulkan requires this, as it needs to initialize state upfront.
---
libavcodec/d3d12va_encode.c | 6 --
libavcodec/hw_base_encode.c | 6 ++
libavcodec/vaapi_encode.c | 5 -
3 files changed, 6 insertions(+), 11 deletions(-)
diff --git a/libavcodec/d3d12va_encode.c b/libavcodec/d3d12va
This function sets the class correctly, and calls functions
that all users have to call anyway.
---
v1 had stray changes.
libavutil/vulkan.c | 57 +-
libavutil/vulkan.h | 12 +-
2 files changed, 67 insertions(+), 2 deletions(-)
diff --git a/lib
This caused images to be created without a storage usage, which broke
at least lavapipe.
---
libavfilter/vulkan_filter.c | 11 ++-
1 file changed, 10 insertions(+), 1 deletion(-)
diff --git a/libavfilter/vulkan_filter.c b/libavfilter/vulkan_filter.c
index 64e9b8768a..c31d42b91a 100644
---
---
libavutil/vulkan.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index bb8e7ae786..fe5affa72a 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@ -444,7 +444,7 @@ VkResult ff_vk_exec_get_query(FFVulkanContext *s,
FFVkExecCo
Vulkan supports some stupidly odd hardware, that unfortunately,
most modern GPUs happen to be.
The DPB images for encoders may be required to be preallocated
all at once, and rather than be individual frames, be layers of
a single frame.
As the hw_base_encode code is written with the thought that
---
libavcodec/Makefile | 2 +-
libavcodec/hw_base_encode_h264.c | 265 +++
libavcodec/hw_base_encode_h264.h | 53 +++
libavcodec/vaapi_encode_h264.c | 262 +++---
4 files changed, 341 insertions(+), 241 deletions(-)
create
The issue is that we ask for storage images by default if
available, but because that is gated by the format supporting
storage images, and the check for the format supporting storage
images is gated by the usage, this resulted in a catch-22.
---
libavutil/hwcontext_vulkan.c | 3 ++-
1 file change
Shared between decoders and encoders.
---
libavcodec/vulkan_av1.c| 3 +-
libavcodec/vulkan_decode.c | 137 +
libavcodec/vulkan_decode.h | 9 ---
libavcodec/vulkan_h264.c | 3 +-
libavcodec/vulkan_hevc.c | 3 +-
libavcodec/vulkan_video.c | 80 ++
Both of these are fundamentally incompatible with video decoding.
---
libavutil/hwcontext_vulkan.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 763fb9d120..b214791738 100644
--- a/libavutil/hwcontext_vulkan.c
+++ b/libavuti
This function sets the class correctly, and calls functions
that all users have to call anyway.
---
libavutil/vulkan.c | 62 +++---
libavutil/vulkan.h | 12 -
2 files changed, 69 insertions(+), 5 deletions(-)
diff --git a/libavutil/vulkan.c b/libavu
---
libavutil/vulkan.c | 5 +
1 file changed, 5 insertions(+)
diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c
index fe5affa72a..c2a3b5cfa3 100644
--- a/libavutil/vulkan.c
+++ b/libavutil/vulkan.c
@@ -446,6 +446,11 @@ VkResult ff_vk_exec_get_query(FFVulkanContext *s,
FFVkExecContext *e,
The old query code never worked properly, and did some hideous
heuristics to read the status bit, and work that into a return
code.
This is all best left to callers to do, which simplifies
our code a lot.
This also fixes minor validation errors regarding calling queries
which are not in their acti
Vulkan filters don't need images which can be transferred into.
---
libavutil/hwcontext_vulkan.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index e4f155db6e..383e75397e 100644
--- a/libavutil/hwcontext_vul
This solves the issue of an av_log function being called
with a context with invalid class.
Most of the patch was written by Anton Khirnov.
---
libavcodec/vulkan_decode.c | 13 +++--
1 file changed, 3 insertions(+), 10 deletions(-)
diff --git a/libavcodec/vulkan_decode.c b/libavcodec/vul
On 02/09/2024 16:32, Michael Niedermayer wrote:
My ping on a patch on 21 Jul has no reaction
Signed-off-by: Michael Niedermayer
---
MAINTAINERS | 1 +
1 file changed, 1 insertion(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index a513bec72a9..d73536370a0 100644
--- a/MAINTAINERS
+++ b/MAINTAIN
On 01/08/2024 19:07, Michael Niedermayer wrote:
On Thu, Aug 01, 2024 at 05:11:18PM +0200, Lynne via ffmpeg-devel wrote:
On 31/07/2024 21:54, Michael Niedermayer wrote:
Fixes: out of array access
Fixes:
70734/clusterfuzz-testcase-minimized-ffmpeg_AV_CODEC_ID_AAC_LATM_fuzzer-4741427068731392
On 04/09/2024 16:09, Tong Wu wrote:
Lynne:
Subject: [FFmpeg-devel] [PATCH 02/10] hw_base_encode: move VAAPI SPS/PPS
constructors to a shared file
---
libavcodec/Makefile | 2 +-
libavcodec/hw_base_encode_h264.c | 265
+++ libavcodec/hw_base_encode_h264.
On 04/09/2024 16:36, Tong Wu wrote:
Lynne:
Subject: [FFmpeg-devel] [PATCH 03/10] hw_base_encode: allocate DPB image
upfront
Vulkan requires this, as it needs to initialize state upfront.
---
libavcodec/d3d12va_encode.c | 6 --
libavcodec/hw_base_encode.c | 6 ++
libavcodec/vaapi_encode.c
On 05/09/2024 07:16, Cameron Gutman wrote:
Signed-off-by: Cameron Gutman
---
libavcodec/amfenc.c | 34 +-
1 file changed, 33 insertions(+), 1 deletion(-)
diff --git a/libavcodec/amfenc.c b/libavcodec/amfenc.c
index 41eaef9758..6f2b211d14 100644
--- a/libavcode
On 06/09/2024 11:48, Tong Wu wrote:
Lynne:
To: Tong Wu ; FFmpeg development discussions and
patches
Subject: Re: [FFmpeg-devel] [PATCH 02/10] hw_base_encode: move VAAPI
SPS/PPS constructors to a shared file
On 04/09/2024 16:09, Tong Wu wrote:
Lynne:
Subject: [FFmpeg-devel] [PATCH 02/10] hw_b
This solves
ffmpeg -help bsf=trace_headers =>
Supported codecs: av1 h264 hevc vvc mjpeg mpeg2video vp8 vp9
---
fftools/opt_common.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/fftools/opt_common.c b/fftools/opt_common.c
index f44fc4c97c..021ed75272 100644
--- a/fftools/opt_common.c
+++ b/
This commit adds the common Vulkan video encoding framework.
It makes full use of the asynchronous features of our new common
hardware encoding code, and of Vulkan.
The code is able to handle anything from H264 to AV1 and MJPEG.
---
configure | 2 +
libavcodec/Makefile|
This commit adds the first Vulkan hardware encoder.
Currently, P, and **B**-frames are supported. This marks the
first implementation to support both.
The encoder has feature-parity with VAAPI.
---
configure |1 +
libavcodec/Makefile |3 +
libavcodec/all
On 09/09/2024 11:57, Anton Khirnov wrote:
Quoting Lynne via ffmpeg-devel (2024-09-09 10:34:21)
This solves
ffmpeg -help bsf=trace_headers =>
Supported codecs: av1 h264 hevc vvc mjpeg mpeg2video vp8 vp9
---
fftools/opt_common.c | 1 +
1 file changed, 1 insertion(+)
Looks ok
Pus
This commit adds the common Vulkan video encoding framework.
It makes full use of the asynchronous features of our new common
hardware encoding code, and of Vulkan.
The code is able to handle anything from H264 to AV1 and MJPEG.
---
configure | 2 +
libavcodec/Makefile|
This commit adds the first Vulkan hardware encoder.
Currently, P, and **B**-frames are supported. This marks the
first implementation to support both.
The encoder has feature-parity with VAAPI.
---
configure |1 +
libavcodec/Makefile |3 +
libavcodec/all
201 - 300 of 3424 matches
Mail list logo
