From: Shiyou Yin
./ffmpeg -i 8_mpeg4_1080p_24fps_12Mbps.avi -f rawvideo -y /dev/null -an
before:376fps
after :433fps
Change-Id: Ic8018562093154887323b508b81d0f489c0d265d
Signed-off-by: Hao Chen
---
libavcodec/hpeldsp.c |2 +
libavcodec/hpeldsp.h
From: gxw
./ffmpeg -i ../1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -y /dev/null -an
before:296
after :308
Change-Id: I748490c1551d0ecfb8b8001625cf7e4e30f28eb4
---
libavcodec/loongarch/Makefile| 1 +
libavcodec/loongarch/videodsp_init.c | 45
libavcodec/video
> On Fri, 2021-12-10 at 13:38 +0800, Wenbin Chen wrote:
> > From: nyanmisaka
> >
> > mfxHDLPair was added to qsv, so modify qsv->opencl map function as well.
> > Now the following commandline works:
> >
> > ffmpeg -v verbose -init_hw_device vaapi=va:/de
> On Wed, 2021-10-27 at 16:57 +0800, Wenbin Chen wrote:
> > Fix: #7706. After commit 5fdcf85bbffe7451c2, vaapi encoder's performance
> > decrease. The reason is that vaRenderPicture() and vaSyncSurface() are
> > called at the same time (vaRenderPicture() always follo
> On 27/10/2021 09:57, Wenbin Chen wrote:
> > Fix: #7706. After commit 5fdcf85bbffe7451c2, vaapi encoder's performance
> > decrease. The reason is that vaRenderPicture() and vaSyncSurface() are
> > called at the same time (vaRenderPicture() always followed by a
> >
> On 10/12/2021 16:05, Wenbin Chen wrote:
> > ffmpeg | branch: master | Wenbin Chen | Tue
> Dec 7 17:05:50 2021 +0800| [f3c9847c2754b7a43eb721c95e356a53085c2491]
> | committer: Lynne
> >
> > libavutil/hwcontext_vaapi: Add a new nv12 format map to support vulkan
>
From: gxw
./ffmpeg -i ../1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -y /dev/null -an
before:296
after :308
---
libavcodec/loongarch/Makefile| 1 +
libavcodec/loongarch/videodsp_init.c | 45
libavcodec/videodsp.c| 2 ++
libavcodec/videodsp.h
@ -0,0 +1,45 @@
+/*
+ * Copyright (c) 2021 Loongson Technology Corporation Limited
+ * Contributed by Hao Chen
+ *
+ * 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
./ffmpeg -i 8_mpeg4_1080p_24fps_12Mbps.avi -f rawvideo -y /dev/null -an
before:376fps
after :552fps
V2: Revised PATCH 1/3 according to the comments.
V3: Resubmit these patches due to miss PATCH v2 1/3.
[PATCH v3 1/3] avcodec: [loongarch] Optimize hpeldsp with LASX.
[PATCH v3 2/3] avcodec: [loonga
From: Shiyou Yin
./ffmpeg -i 8_mpeg4_1080p_24fps_12Mbps.avi -f rawvideo -y /dev/null -an
before:376fps
after :433fps
---
libavcodec/hpeldsp.c |2 +
libavcodec/hpeldsp.h |1 +
libavcodec/loongarch/Makefile |2 +
libavco
@ -0,0 +1,45 @@
+/*
+ * Copyright (c) 2021 Loongson Technology Corporation Limited
+ * Contributed by Hao Chen
+ *
+ * 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
From: gxw
./ffmpeg -i ../1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -y /dev/null -an
before:296
after :308
---
libavcodec/loongarch/Makefile| 1 +
libavcodec/loongarch/videodsp_init.c | 45
libavcodec/videodsp.c| 2 ++
libavcodec/videodsp.h
Add doc for qsv decoder.
Add more option's introduction to qsv encoder.
Signed-off-by: Wenbin Chen
---
doc/decoders.texi | 42
doc/encoders.texi | 253 +-
2 files changed, 292 insertions(+), 3 deletions(-)
diff --git a/doc/decoders.t
. "wait=1" means wait until operation ready. "wait=0" means
query operation's status. If it is ready return 0, if it is still
in progress return EAGAIN.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 47 +--
1 file changed, 40 i
uffer, it will keep previous operation.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 64 ---
libavcodec/vaapi_encode.h | 5 +++
2 files changed, 58 insertions(+), 11 deletions(-)
diff --git a/libavcodec/vaapi_encode.c b/libavcodec/vaapi_encode.c
es) with -async_depth=4 can increase 20%
performance on my environment.
The async increases performance but also introduces frame delay.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 19 ++-
libavcodec/vaapi_encode.h | 12 ++--
2 files changed, 24 insertio
> Wenbin Chen:
> > Fix: #7706. After commit 5fdcf85bbffe7451c2, vaapi encoder's performance
> > decrease. The reason is that vaRenderPicture() and vaSyncBuffer() are
> > called at the same time (vaRenderPicture() always followed by a
> > vaSyncBuffer()). When we
Add doc for qsv decoder.
Add more option's introduction to qsv encoder.
Signed-off-by: Wenbin Chen
---
doc/decoders.texi | 42
doc/encoders.texi | 253 +-
2 files changed, 292 insertions(+), 3 deletions(-)
diff --git a/doc/decoders.t
ces). Now add code to make sure
init_pool_size is only set once. Now the following commandline works:
ffmpeg -hwaccel vaapi -hwaccel_device /dev/dri/renderD128 \
-hwaccel_output_format vaapi -i input.264 \
-vf "hwmap=derive_device=qsv,format=qsv" \
-c:v h264_qsv output.264
Signed-off-by
qsv -hwaccel_output_format qsv -hwaccel_device qs -c:v h264_qsv \
-i input.264 -vf "hwmap=derive_device=opencl,format=opencl,avgblur_opencl, \
hwmap=derive_device=qsv:reverse=1:extra_hw_frames=32,format=qsv" \
-c:v h264_qsv output.264
Signed-off-by: nyanmisaka
Signed-off-by: Wenbin Chen
-
> On Fri, 2022-01-07 at 15:36 +0800, Wenbin Chen wrote:
> > From: nyanmisaka
> >
> > mfxHDLPair was added to qsv, so modify qsv->opencl map function as well.
> > Now the following commandline works:
> >
> > ffmpeg -v verbose -init_hw_device vaapi=va:/de
ces). Now add code to make sure
init_pool_size is only set once. Now the following commandline works:
ffmpeg -hwaccel vaapi -hwaccel_device /dev/dri/renderD128 \
-hwaccel_output_format vaapi -i input.264 \
-vf "hwmap=derive_device=qsv,format=qsv" \
-c:v h264_qsv output.264
Signed-off-by
qsv -hwaccel_output_format qsv -hwaccel_device qs -c:v h264_qsv \
-i input.264 -vf "hwmap=derive_device=opencl,format=opencl,avgblur_opencl, \
hwmap=derive_device=qsv:reverse=1:extra_hw_frames=32,format=qsv" \
-c:v h264_qsv output.264
Signed-off-by: nyanmisaka
Signed-off-by: Wenbin Chen
-
Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
disable deblocking.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 8
libavcodec/qsvenc.h | 3 +++
3 files changed, 17 insertions(+)
diff --git a/doc/encoders.texi b/doc
github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md#preftype
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 36
libavcodec/qsvenc.h | 3 +++
3 files changed, 45 insertions(+)
diff --git a/doc/encoders.t
Add max_frame_size support to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 4 ++--
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index 7cc8be1209..68921fbd40 100644
--- a/doc
Add transform_skip option to hevc_qsv. By enabling this option,
the transform_skip_enabled_flag in PPS will be set to 1.
This option is only supported in the platform newer than ICL.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi| 3 +++
libavcodec/qsvenc.c | 10
> On Wed, 2022-01-12 at 13:28 +0800, Wenbin Chen wrote:
> > Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
> > disable deblocking.
> >
> > Signed-off-by: Wenbin Chen
> > ---
> > doc/encoders.texi | 6 ++
> > libavcodec/q
Add max_frame_size support to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 4 ++--
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index 7cc8be1209..68921fbd40 100644
--- a/doc
Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
disable deblocking.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 8
libavcodec/qsvenc.h | 3 +++
3 files changed, 17 insertions(+)
diff --git a/doc/encoders.texi b/doc
github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md#preftype
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 36
libavcodec/qsvenc.h | 3 +++
3 files changed, 45 insertions(+)
diff --git a/doc/encoders.t
Add transform_skip option to hevc_qsv. By enabling this option,
the transform_skip_enabled_flag in PPS will be set to 1.
This option is supported on the platform equal or newer than CNL.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi| 4
libavcodec/qsvenc.c | 10
Add transform_skip option to hevc_qsv. By enabling this option,
the transform_skip_enabled_flag in PPS will be set to 1.
This option is supported on the platform equal or newer than ICL.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi| 4
libavcodec/qsvenc.c | 10
Add max_frame_size support to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 4 ++--
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index 7cc8be1209..68921fbd40 100644
--- a/doc
Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
disable deblocking.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 8
libavcodec/qsvenc.h | 3 +++
3 files changed, 17 insertions(+)
diff --git a/doc/encoders.texi b/doc
github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md#preftype
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 36
libavcodec/qsvenc.h | 3 +++
3 files changed, 45 insertions(+)
diff --git a/doc/encoders.t
> On Thu, 2022-01-13 at 13:12 +0800, Wenbin Chen wrote:
> > Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
> > disable deblocking.
> >
> > Signed-off-by: Wenbin Chen
> > ---
> > doc/encoders.texi | 6 ++
> > libavcodec/q
> On Thu, 2022-01-13 at 13:12 +0800, Wenbin Chen wrote:
> > Add transform_skip option to hevc_qsv. By enabling this option,
> > the transform_skip_enabled_flag in PPS will be set to 1.
> > This option is supported on the platform equal or newer than CNL.
> >
>
Add max_frame_size support to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 4 ++--
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index 7cc8be1209..68921fbd40 100644
--- a/doc
Add dblk_idc option to 264_qsv and hevc_qsv. Turining on this opion can
disable deblocking.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 7 +++
libavcodec/qsvenc.h | 3 +++
3 files changed, 16 insertions(+)
diff --git a/doc/encoders.texi b/doc
github.com/Intel-Media-SDK/MediaSDK/blob/master/doc/mediasdk-man.md#preftype
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 6 ++
libavcodec/qsvenc.c | 36
libavcodec/qsvenc.h | 3 +++
3 files changed, 45 insertions(+)
diff --git a/doc/encoders.t
Add transform_skip option to hevc_qsv. By enabling this option,
the transform_skip_enabled_flag in PPS will be set to 1.
This option is supported on the platform equal or newer than ICL.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi| 4
libavcodec/qsvenc.c | 13
> On Mon, 27 Nov 2023 02:10:11 +0000 "Chen, Wenbin" intel@ffmpeg.org> wrote:
> > > > From: Niklas Haas
> > > >
> > > > This logic was incongruent with logic used elsewhere, where floating
> > > > point formats are explici
Note: requires oneVPL-intel-gpu version >= 24.1.1 because this version contains
a fix for creating LUT in video memory.
(For details, refer to
https://github.com/oneapi-src/oneVPL-intel-gpu/issues/307)
Chen Yufei (1):
avfilter/vf_vpp_qsv: apply 3D LUT from file.
libavfilter/Makefile
Usage: "vpp_qsv=lut3d_file="
Requires oneVPL, using system memory 3D LUT surface.
Signed-off-by: Chen Yufei
---
libavfilter/Makefile | 8 +-
libavfilter/lut3d.c | 669 +++
libavfilter/lut3d.h | 13 +
libavfilter/vf_lut3d
On Tue, Jan 23, 2024 at 10:00 AM Xiang, Haihao wrote:
>
> On Sa, 2024-01-20 at 23:14 +0800, Chen Yufei wrote:
> > This patch adds support for applying 3D LUT from file using oneVPL VPP.
> >
> > PATCH v1 uses VA-API to create LUT surface. Because oneVPL can't work wit
On Wed, Jan 24, 2024 at 7:39 PM Anton Khirnov wrote:
>
> Quoting Chen Yufei (2024-01-20 16:14:29)
> > Usage: "vpp_qsv=lut3d_file="
>
> Passing file paths to a filter and having the filter load the file is
> not recommended, it is generally preferable to have an
&g
This version of PATCH use `QSV_RUNTIME_VERSION_ATLEAST` to apply 3D LUT when
libvpl runtime API version >= 2.11.
Chen Yufei (1):
avfilter/vf_vpp_qsv: apply 3D LUT from file.
libavfilter/Makefile | 8 +-
libavfilter/lut3d.c | 669 +++
libavfil
Usage: "vpp_qsv=lut3d_file="
Requires oneVPL, using system memory 3D LUT surface.
Signed-off-by: Chen Yufei
---
libavfilter/Makefile | 8 +-
libavfilter/lut3d.c | 669 +++
libavfilter/lut3d.h | 13 +
libavfilter/vf_lut3d
On Sun, Jan 28, 2024 at 10:10 PM Anton Khirnov wrote:
>
> Quoting Zhao Zhili (2024-01-28 14:51:58)
> >
> >
> > > On Jan 28, 2024, at 18:31, Anton Khirnov wrote:
> > >
> > > Quoting Chen Yufei (2024-01-25 17:16:46)
> > >
file name extension to
detect LUT file type and call different parse functions.
If we add another option to specify LUT file type, then vf_lut3d's
command line option would require change.
--
Best regards,
Chen Yufei
___
ffmpeg-devel mailing lis
On Tue, Jan 30, 2024 at 1:07 AM Anton Khirnov wrote:
>
> Quoting Chen Yufei (2024-01-29 04:01:51)
> > On Sun, Jan 28, 2024 at 10:10 PM Anton Khirnov wrote:
> > >
> > > Quoting Zhao Zhili (2024-01-28 14:51:58)
> > > >
> > > >
>
On Tue, Jan 30, 2024 at 10:59 AM Chen Yufei wrote:
>
> On Tue, Jan 30, 2024 at 1:07 AM Anton Khirnov wrote:
> >
> > Quoting Chen Yufei (2024-01-29 04:01:51)
> > > On Sun, Jan 28, 2024 at 10:10 PM Anton Khirnov wrote:
> > > >
> >
From: "Chen,Wenbin"
If the descriptorSetCount is greater than the number of setLayouts,
vkAllocateDescriptorSets will report error. Now fix it.
Now the following command can run on the device that has queue_count greater
than one:
ffmpeg -v verbose -init_hw_device vulkan=vul:0 -filter
From: "Chen,Wenbin"
We should configure VkImageSubresource according to tiling rather than
extension. We use extension to set tiling only when we map from drm.
Normally the output VkImages are not created in this way, and it will
report error when we map these VkImage to drm, so
From: "Chen,Wenbin"
We chould set waitSem and signalSem differently. Current ffmpeg-vulkan
uses the same sem to set waitSem and signalSem and it doesn't work on
latest intel-vulkan-driver. The commit:
a193060221c4df123e26a562949cae5df3e73cde on mesa causes this problem.
This com
From: Bas Nieuwenhuizen
This way we can pass explicit modifiers in. Sometimes the
modifier matters for the number of memory planes that
libva accepts, in particular when dealing with
driver-compressed textures. Furthermore the driver might
not actually be able to determine the implicit modifier
i
From: "Chen,Wenbin"
Vulkan will map nv12 to R8 and GR88, so add this map to vaapi to support
vulkan frame.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vaapi.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c
index
From: "Chen,Wenbin"
Vaapi can import external surface, but all the planes of the external
frames should be in the same drm object. A new flag is introduced and
vulkan can choose to allocate planes in one memory according this flag.
This flag will be enabled when the vulkan device is de
From: "Chen,Wenbin"
According to spec, if we use VkBindImagePlaneMemoryInfo to bind image
we mush create image with disjoint flag.
The offset in subresourcelayout is relative to the base address of
the plane, but the offset in drm is relative to the drm objectis so
I think this offset
From: "Chen,Wenbin"
The vaapi can import external frame, but the planes of the external
frames should be in the same drm object. I add a new function to
allocate vkFrame in one memory and vulkan device will choose a way
to allocate memory according to one_memory flag.
A new variable i
From: Wenbin Chen
On the lastset intel-vulkan-driver the VK_EXT_image_drm_format_modifier
flags is enabled. As what driver log recommand, we need to use
VK_IMAGE_TILING_LINEAR or VK_IMAGE_DRM_FORMAT_MODIFIER_EXT to create
VKImage. Add code to get supported modifier for sw_format and use these
From: "Chen,Wenbin"
Add hwupload and hwdownload support to vulkan when frames are allocated
in one memory
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b
e queue_count greater than 1
>
> Hello,
>
> On Tue, 31 Aug 2021, at 03:43, wenbin.c...@intel.com wrote:
> > From: "Chen,Wenbin"
> > ...
> > Signed-off-by: Wenbin Chen
> > ...
> > email sent by "wenbin.c...@intel.com" (no name)
>
-vf "hwupload=extra_hw_frames=16,scale_vulkan=1920:1080,
hwdownload,format=yuv420p" -f rawvideo output.yuv
Signed-off-by: Wenbin Chen
---
libavfilter/vulkan.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavfilter/vulkan.c b/libavfilter/vulkan.c
index
according to tiling rather than
extension. Now fix it.
Signed-off-by: Wenbin Chen
---
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 94fdad7f06..88db5b8b70 100644
--- a/libavutil
t;hwupload=extra_hw_frames=16,scale_vulkan=1920:1080,
hwdownload,format=yuv420p" -f rawvideo output.yuv
Signed-off-by: Wenbin Chen
---
libavfilter/vf_avgblur_vulkan.c | 4 +--
libavfilter/vf_chromaber_vulkan.c | 4 +--
libavfilter/vf_overlay_vulkan.c | 6 ++--
libavfilter/vf_scale_vu
From: Bas Nieuwenhuizen
This way we can pass explicit modifiers in. Sometimes the
modifier matters for the number of memory planes that
libva accepts, in particular when dealing with
driver-compressed textures. Furthermore the driver might
not actually be able to determine the implicit modifier
i
Vulkan will map nv12 to R8 and GR88, so add this map to vaapi to support
vulkan frame.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vaapi.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c
index 75acc851d6..994b744e4d 100644
this change will not affect current vulkan behaviour.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 12 +++-
1 file changed, 11 insertions(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 9a29267aed..6417f59d4a 100644
--- a
offset of each plane.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 46 +++-
libavutil/hwcontext_vulkan.h | 1 +
2 files changed, 46 insertions(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index
64
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 77 +---
libavutil/hwcontext_vulkan.h | 5 +++
2 files changed, 76 insertions(+), 6 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index 3a639c997b..99b2190dc3 10
Add hwupload and hwdownload support to vulkan when frames are allocated
in one memory
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 10 --
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulkan.c
index
vaapi frame to vulkan I got broken frame, and
setting plane_data->offset to 0 makes command works.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_vulkan.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/libavutil/hwcontext_vulkan.c b/libavutil/hwcontext_vulka
but the offset in drm is relative to the drm objectis so
> > I think this offset should be 0.
> > Also, when I import vaapi frame to vulkan I got broken frame, and
> > setting plane_data->offset to 0 makes command works.
> >
> > Signed-off-by: Wenbin Chen
> &g
esa on intel platform:
> > ffmpeg -v verbose -init_hw_device vulkan=vul:0,linear_images=1 -
> filter_hw_device vul
> > -i input1080p.264 -vf
> "hwupload=extra_hw_frames=16,scale_vulkan=1920:1080,
> > hwdownload,format=yuv420p" -f rawvideo output.yuv
> >
> > Signed
> -Original Message-
> From: Lynne
> Sent: Wednesday, September 1, 2021 7:33 PM
> To: Chen, Wenbin
> Subject: RE: [FFmpeg-devel] [V2 08/10] libavutil/hwcontext_vulkan: fix wrong
> offset of plane
>
> 1 Sept 2021, 03:40 by wenbin.c...@intel.com:
>
> &g
Hi softworkz:
Any updates on this patch?
I want to submit pathes to qsv, but they depends on this patch
Thanks
Wenbin
> -Original Message-
> From: ffmpeg-devel On Behalf Of Soft
> Works
> Sent: Thursday, August 19, 2021 3:52 PM
> To: FFmpeg development discussions and patches de...@ffmp
gt; hwdevice, search for existing device in both directions
>
>
>
> > -Original Message-
> > From: ffmpeg-devel On Behalf Of
> > Chen, Wenbin
> > Sent: Wednesday, 8 September 2021 07:58
> > To: FFmpeg development discussions and patches > de...@ffmpeg
a[3] in vaapi frame.
pair->first is *VASurfaceID while data[3] in vaapi frame is
VASurfaceID. I fix this line of code. Now the command above works.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_qsv.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/libavutil/hwcontext_q
> -Original Message-
> From: ffmpeg-devel On Behalf Of
> James Almer
> Sent: Thursday, September 9, 2021 8:48 PM
> To: ffmpeg-devel@ffmpeg.org
> Subject: Re: [FFmpeg-devel] [PATCH] libavutil/hwcontext_qsv: fix a bug for
> mapping qsv frame to vaapi
>
> On 9/
> On Fri, 2021-09-10 at 02:19 +0000, Chen, Wenbin wrote:
> > > -Original Message-
> > > From: ffmpeg-devel On Behalf Of
> > > James Almer
> > > Sent: Thursday, September 9, 2021 8:48 PM
> > > To: ffmpeg-devel@ffmpeg.org
> > > Su
a[3] in vaapi frame.
pair->first is *VASurfaceID while data[3] in vaapi frame is
VASurfaceID. I fix this line of code. Now the command above works.
Signed-off-by: Wenbin Chen
---
libavutil/hwcontext_qsv.c | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/libavutil/h
Assign pair->first directly to data[3] in vaapi frame.
> pair->first is *VASurfaceID while data[3] in vaapi frame is
> VASurfaceID. I fix this line of code. Now the command above works.
>
> Signed-off-by: Wenbin Chen
> ---
> libavutil/hwcontext_qsv.c | 6 +-
> 1 file
This two variables may be used below with uninitialized value.
Now fix them.
Signed-off-by: Wenbin Chen
---
libavfilter/vf_vpp_qsv.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/libavfilter/vf_vpp_qsv.c b/libavfilter/vf_vpp_qsv.c
index c268b5dba8..22ffe0d7f3 100644
Assign pair->first directly to data[3] in vaapi frame.
> pair->first is *VASurfaceID while data[3] in vaapi frame is
> VASurfaceID. I fix this line of code. Now the command above works.
>
> Signed-off-by: Wenbin Chen
> ---
> libavutil/hwcontext_qsv.c | 6 +-
> 1 file
"qf->frame" ref to input frame but it isn't released. av_frame_unref()
is added before refering qf->frame to new frame to make sure the previous
reference is released.
Signed-off-by: Wenbin Chen
---
libavcodec/qsvenc.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/
put.264
>
> Signed-off-by: nyanmisaka
> Signed-off-by: Wenbin Chen
> ---
> libavutil/hwcontext_opencl.c | 14 +-
> libavutil/hwcontext_qsv.c| 34 ++
> 2 files changed, 47 insertions(+), 1 deletion(-)
>
> diff --gi
evice=qsv,format=qsv" \
> -c:v h264_qsv output.264
>
> Signed-off-by: Wenbin Chen
> ---
> libavcodec/vaapi_decode.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/libavcodec/vaapi_decode.c b/libavcodec/vaapi_decode.c
> index 665af370ed..da0
Add max_slice_size option to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 9 -
2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index e3adbf4325..8966610263 100644
--- a/doc
Add b_strategy option to hevc_qsv. By enabling this option, encoder can
use b frames as reference.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 6 --
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
Add intra refresh support to hevc_qsv as well.
Add an new intra refresh type: "horizontal", and an new param
ref_cycle_dist. This param specify the distance between the
beginnings of the intra-refresh cycles in frames.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi
> On Mon, 2022-01-24 at 10:59 +0800, Wenbin Chen wrote:
> > Add intra refresh support to hevc_qsv as well.
> > Add an new intra refresh type: "horizontal", and an new param
> > ref_cycle_dist. This param specify the distance between the
> > beginnings
Add b_strategy option to hevc_qsv. By enabling this option, encoder can
use b frames as reference.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 6 --
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
Add max_slice_size option to hevc_qsv as well.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi | 3 +++
libavcodec/qsvenc.c | 9 -
2 files changed, 7 insertions(+), 5 deletions(-)
diff --git a/doc/encoders.texi b/doc/encoders.texi
index e3adbf4325..8966610263 100644
--- a/doc
Add intra refresh support to hevc_qsv as well.
Add an new intra refresh type: "horizontal", and an new param
ref_cycle_dist. This param specify the distance between the
beginnings of the intra-refresh cycles in frames.
Signed-off-by: Wenbin Chen
---
doc/encoders.texi
> On Wed, 2022-01-05 at 10:48 +0800, Wenbin Chen wrote:
> > Fix: #7706. After commit 5fdcf85bbffe7451c2, vaapi encoder's performance
> > decrease. The reason is that vaRenderPicture() and vaSyncBuffer() are
> > called at the same time (vaRenderPicture() always foll
. "wait=1" means wait until operation ready. "wait=0" means
query operation's status. If it is ready return 0, if it is still
in progress return EAGAIN.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 47 +--
1 file changed, 40 i
uffer, it will keep previous operation.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 64 ---
libavcodec/vaapi_encode.h | 5 +++
2 files changed, 58 insertions(+), 11 deletions(-)
diff --git a/libavcodec/vaapi_encode.c b/libavcodec/vaapi_encode.c
es) with -async_depth=4 can increase 20%
performance on my environment.
The async increases performance but also introduces frame delay.
Signed-off-by: Wenbin Chen
---
libavcodec/vaapi_encode.c | 16
libavcodec/vaapi_encode.h | 12 ++--
2 files changed, 22 insertio
301 - 400 of 597 matches
Mail list logo
