[FFmpeg-devel] [PATCH] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

This allows some optimizations in driver, such as not having to read back the data if write-only mapping is requested. --- libavutil/hwcontext_vaapi.c | 9 + 1 file changed, 9 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c index 558fed94c6..c3ecd90077

[FFmpeg-devel] [PATCH v2] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

This allows some optimizations in driver, such as not having to read back the data if write-only mapping is requested. --- v2: Fix warning libavutil/hwcontext_vaapi.c | 9 + 1 file changed, 9 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c index 558fe

[FFmpeg-devel] [PATCH v3] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

This allows some optimizations in driver, such as not having to read back the data if write-only mapping is requested. --- v3: Fix another warning libavutil/hwcontext_vaapi.c | 11 +++ 1 file changed, 11 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c

Re: [FFmpeg-devel] [PATCH v3] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

On Fri, Oct 27, 2023 at 7:14 PM Mark Thompson wrote: > > On 27/10/2023 11:00, David Rosca wrote: > > This allows some optimizations in driver, such as not having to read > > back the data if write-only mapping is requested. > > --- > > v3: Fix another warning > &g

[FFmpeg-devel] [PATCH v4] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

This allows some optimizations in driver, such as not having to read back the data if write-only mapping is requested. --- v4: overwrite + note about vaMapBuffer libva fallback libavutil/hwcontext_vaapi.c | 12 1 file changed, 12 insertions(+) diff --git a/libavutil/hwcontext_vaapi.

[FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Set roi_quant_range

--- libavcodec/vaapi_encode_av1.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/libavcodec/vaapi_encode_av1.c b/libavcodec/vaapi_encode_av1.c index a46b882ab9..02a31b894d 100644 --- a/libavcodec/vaapi_encode_av1.c +++ b/libavcodec/vaapi_encode_av1.c @@ -155,6 +155,8 @@ static av_cold int

Re: [FFmpeg-devel] [PATCH v4] lavu/hwcontext_vaapi: Use vaMapBuffer2 for mapping image buffers

On Fri, Nov 24, 2023 at 8:27 AM Xiang, Haihao wrote: > > On Vr, 2023-10-27 at 22:25 +0200, David Rosca wrote: > > This allows some optimizations in driver, such as not having to read > > back the data if write-only mapping is requested. > > --- > > v4: overwrite

[FFmpeg-devel] [PATCH 1/2] lavc/vaapi_decode: Make it possible to send multiple slice params buffers

--- libavcodec/vaapi_av1.c| 2 +- libavcodec/vaapi_decode.c | 3 ++- libavcodec/vaapi_decode.h | 1 + libavcodec/vaapi_h264.c | 2 +- libavcodec/vaapi_hevc.c | 4 ++-- libavcodec/vaapi_mjpeg.c | 2 +- libavcodec/vaapi_mpeg2.c | 2 +- libavcodec/vaapi_mpeg4.c | 2 +- libavcodec/vaapi_vc1

[FFmpeg-devel] [PATCH 2/2] lavc/vaapi_av1: Avoid sending the same slice buffer multiple times

When there are multiple tiles in one slice buffer, use multiple slice params to avoid sending the same slice buffer multiple times and thus increasing the bitstream size the driver will need to upload to hw. --- libavcodec/vaapi_av1.c | 37 +++-- 1 file changed, 23

[FFmpeg-devel] [PATCH] lavc/vaapi_h264: Don't try to merge fields in DPB for non-field pics

This path can be hit when there are missing references while decoding progressive stream and would completely break the DPB contents. --- libavcodec/vaapi_h264.c | 30 -- 1 file changed, 16 insertions(+), 14 deletions(-) diff --git a/libavcodec/vaapi_h264.c b/libavcode

[FFmpeg-devel] [PATCH] lavc/vaapi_h264: Fix merging fields in DPB with missing references

If there are missing references, h264 decode does error concealment by copying previous refs which means there will be duplicated surfaces and this code would try to merge them instead of correctly appending to DPB. Make sure the fields were actually merged before early return. --- libavcodec/vaap

Re: [FFmpeg-devel] [PATCH] lavc/vaapi_h264: Don't try to merge fields in DPB for non-field pics

On Mon, May 6, 2024 at 9:55 PM Mark Thompson wrote: > > On 05/05/2024 17:36, David Rosca wrote: > > This path can be hit when there are missing references while decoding > > progressive stream and would completely break the DPB contents. > > --- > >

[FFmpeg-devel] [PATCH v2 1/2] lavc/vaapi_decode: Make it possible to send multiple slice params buffers

--- v2: No changes libavcodec/vaapi_av1.c| 2 +- libavcodec/vaapi_decode.c | 3 ++- libavcodec/vaapi_decode.h | 1 + libavcodec/vaapi_h264.c | 2 +- libavcodec/vaapi_hevc.c | 4 ++-- libavcodec/vaapi_mjpeg.c | 2 +- libavcodec/vaapi_mpeg2.c | 2 +- libavcodec/vaapi_mpeg4.c | 2 +- liba

[FFmpeg-devel] [PATCH v2 2/2] lavc/vaapi_av1: Avoid sending the same slice buffer multiple times

When there are multiple tiles in one slice buffer, use multiple slice params to avoid sending the same slice buffer multiple times and thus increasing the bitstream size the driver will need to upload to hw. --- v2: Avoid allocations every slice. libavcodec/vaapi_av1.c | 47 ++

Re: [FFmpeg-devel] [PATCH 2/2] lavc/vaapi_av1: Avoid sending the same slice buffer multiple times

On Tue, May 7, 2024 at 10:01 PM Mark Thompson wrote: > > On 28/04/2024 08:26, David Rosca wrote: > > When there are multiple tiles in one slice buffer, use multiple slice > > params to avoid sending the same slice buffer multiple times and thus > > increasing the bitstr

[FFmpeg-devel] [PATCH v2] lavc/vaapi_h264: Fix merging fields in DPB with missing references

If there are missing references, h264 decode does error concealment by copying previous refs which means there will be duplicated surfaces. Check long_ref and frame_idx in addition to surface when looking for the other field to avoid trying to merge with wrong picture. Also allow to merge with mult

Re: [FFmpeg-devel] [PATCH] lavc/vaapi_h264: Fix merging fields in DPB with missing references

On Tue, May 7, 2024 at 10:22 PM Mark Thompson wrote: > > On 07/05/2024 07:00, David Rosca wrote: > > If there are missing references, h264 decode does error concealment > > by copying previous refs which means there will be duplicated surfaces > > and this code would try

[FFmpeg-devel] [PATCH] lavc/vaapi_decode: Reject decoding of frames with no slices

Matches other hwaccels. --- libavcodec/vaapi_decode.c | 5 + 1 file changed, 5 insertions(+) diff --git a/libavcodec/vaapi_decode.c b/libavcodec/vaapi_decode.c index 8e9f647c20..3c4030c073 100644 --- a/libavcodec/vaapi_decode.c +++ b/libavcodec/vaapi_decode.c @@ -157,6 +157,11 @@ int ff_vaapi

[FFmpeg-devel] [PATCH v2] lavc/vaapi_decode: Reject decoding of frames with no slices

Matches other hwaccels. --- v2: AVERROR libavcodec/vaapi_decode.c | 5 + 1 file changed, 5 insertions(+) diff --git a/libavcodec/vaapi_decode.c b/libavcodec/vaapi_decode.c index 8e9f647c20..7c91d50f7b 100644 --- a/libavcodec/vaapi_decode.c +++ b/libavcodec/vaapi_decode.c @@ -157,6 +157,11 @@

[FFmpeg-devel] [PATCH] lavc/vaapi_encode: Add filler_data option

--- libavcodec/vaapi_encode.c | 1 + libavcodec/vaapi_encode.h | 9 - 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/libavcodec/vaapi_encode.c b/libavcodec/vaapi_encode.c index bfca315a7a..f161c76304 100644 --- a/libavcodec/vaapi_encode.c +++ b/libavcodec/vaapi_encode.c @@ -

[FFmpeg-devel] [PATCH v2] lavc/vaapi_encode: Add filler_data option

v2: Add description in encoders.texi --- doc/encoders.texi | 3 +++ libavcodec/vaapi_encode.c | 1 + libavcodec/vaapi_encode.h | 9 - 3 files changed, 12 insertions(+), 1 deletion(-) diff --git a/doc/encoders.texi b/doc/encoders.texi index 25d6b7f09e..f146942aa5 100644 --- a/doc/e

[FFmpeg-devel] [PATCH] lavu/hwcontext_vaapi: Add vaapi_drm_format_map support for x2rgb10

Support for allocating frames with x2rgb10 format was added in c00264f5013, this adds support for importing DMA-BUFs. --- libavutil/hwcontext_vaapi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c index 6c3a227ddd..63544ce476 100644

[FFmpeg-devel] [PATCH v2] lavu/hwcontext_vaapi: Add vaapi_drm_format_map support for x2rgb10

Support for allocating frames with x2rgb10 format was added in c00264f5013, this adds support for importing DMA-BUFs. v2: Fix #ifdef -> #if --- libavutil/hwcontext_vaapi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c index 6c3a22

Re: [FFmpeg-devel] [PATCH] lavu/hwcontext_vaapi: Add vaapi_drm_format_map support for x2rgb10

On Wed, Aug 9, 2023 at 2:35 PM Rémi Denis-Courmont wrote: > > > > Le 9 août 2023 15:02:45 GMT+03:00, David Rosca a écrit : > >Support for allocating frames with x2rgb10 format was added > >in c00264f5013, this adds support for importing DMA-BUFs. > >--- >

[FFmpeg-devel] [PATCH v3] lavu/hwcontext_vaapi: Add vaapi_drm_format_map support for x2rgb10

Support for allocating frames with x2rgb10 format was added in c00264f5013, this adds support for importing DMA-BUFs. --- v2: fix #ifdef -> #if v3: annotate libavutil/hwcontext_vaapi.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c

Re: [FFmpeg-devel] [PATCH v2] lavc/vaapi_encode: Add filler_data option

On Tue, Aug 15, 2023 at 9:25 PM Mark Thompson wrote: > > On 15/08/2023 20:09, Mark Thompson wrote: > > On 15/08/2023 09:02, Xiang, Haihao wrote: > >> On Ma, 2023-08-07 at 09:27 +0200, David Rosca wrote: > >>> v2: Add description in encoders.texi > >>&

[FFmpeg-devel] [PATCH] vaapi_encode_h264: Only set pic_order_cnt_type to 0 with B-frames

--- libavcodec/vaapi_encode_h264.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/libavcodec/vaapi_encode_h264.c b/libavcodec/vaapi_encode_h264.c index dd17be2..d6926c4 100644 --- a/libavcodec/vaapi_encode_h264.c +++ b/libavcodec/vaapi_encode_h264.c @@ -350,7 +350,7 @@ static

Re: [FFmpeg-devel] [PATCH] vaapi_encode_h264: Only set pic_order_cnt_type to 0 with B-frames

On Mon, Jan 9, 2023 at 3:22 AM Xiang, Haihao wrote: > > On Do, 2022-12-29 at 22:20 +0100, David Rosca wrote: > > --- > > libavcodec/vaapi_encode_h264.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/libavcodec/vaapi_encode_h26

[FFmpeg-devel] [PATCH v2] vaapi_encode_h264: Only set pic_order_cnt_type to 0 with B-frames

v2: frame_num steps by 2 --- libavcodec/vaapi_encode_h264.c | 7 +-- 1 file changed, 5 insertions(+), 2 deletions(-) diff --git a/libavcodec/vaapi_encode_h264.c b/libavcodec/vaapi_encode_h264.c index 7a6b54ab6f..8093c47179 100644 --- a/libavcodec/vaapi_encode_h264.c +++ b/libavcodec/vaapi_en

Re: [FFmpeg-devel] [PATCH] vaapi_encode_h264: Only set pic_order_cnt_type to 0 with B-frames

On Mon, Jan 9, 2023 at 10:11 PM Mark Thompson wrote: > > On 09/01/2023 07:37, David Rosca wrote: > > On Mon, Jan 9, 2023 at 3:22 AM Xiang, Haihao wrote: > >> > >> On Do, 2022-12-29 at 22:20 +0100, David Rosca wrote: > >>> --- > >>> libavc

[FFmpeg-devel] [PATCH v3] vaapi_encode_h264: Only set pic_order_cnt_type to 0 with B-frames

v3: pic_order_cnt steps by 2 --- libavcodec/vaapi_encode_h264.c | 10 -- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/libavcodec/vaapi_encode_h264.c b/libavcodec/vaapi_encode_h264.c index f15bcc6..de0951f 100644 --- a/libavcodec/vaapi_encode_h264.c +++ b/libavcodec/vaapi_e

Re: [FFmpeg-devel] [PATCH v1 6/6] lavc/vaapi_encode: Add VAAPI AV1 encoder

On Mon, Jul 10, 2023 at 9:40 AM Fei Wang wrote: > > Signed-off-by: Fei Wang > --- > Changelog |1 + > configure |3 + > doc/encoders.texi | 13 + > libavcodec/Makefile |1 + > libavcodec/allcodecs.c|1 + > li

[FFmpeg-devel] [PATCH] lavc/vaapi_encode: add "quality" option to all encoders

Move "quality" option from h264_vaapi to common options. --- libavcodec/vaapi_encode.c | 3 +++ libavcodec/vaapi_encode.h | 8 +++- libavcodec/vaapi_encode_h264.c | 5 - 3 files changed, 10 insertions(+), 6 deletions(-) diff --git a/libavcodec/vaapi_encode.c b/libavcodec/vaapi_e

[FFmpeg-devel] [PATCH] hw_base_encode: Free pictures on close

Fixes leaking recon surfaces with VAAPI. --- libavcodec/hw_base_encode.c | 5 + 1 file changed, 5 insertions(+) diff --git a/libavcodec/hw_base_encode.c b/libavcodec/hw_base_encode.c index 7b6ec97d3b..912c707a68 100644 --- a/libavcodec/hw_base_encode.c +++ b/libavcodec/hw_base_encode.c @@ -80

[FFmpeg-devel] [PATCH 2/2] lavc/vaapi_encode_h265: Use surface alignment

This is needed to correctly set conformance window crop with Mesa AMD. --- libavcodec/vaapi_encode_h265.c | 12 ++-- 1 file changed, 10 insertions(+), 2 deletions(-) diff --git a/libavcodec/vaapi_encode_h265.c b/libavcodec/vaapi_encode_h265.c index 2283bcc0b4..0f97c9188b 100644 --- a/liba

[FFmpeg-devel] [PATCH 1/2] lavc/vaapi_encode: Query surface alignment

It needs to create temporary config to query surface attribute. --- libavcodec/vaapi_encode.c | 66 +++ libavcodec/vaapi_encode.h | 4 +++ 2 files changed, 70 insertions(+) diff --git a/libavcodec/vaapi_encode.c b/libavcodec/vaapi_encode.c index 16a9a364f0..c8

Re: [FFmpeg-devel] [PATCH v2] hw_base_enc: inject side data to crop output for AV1

On Sat, Sep 14, 2024 at 8:59 AM Lynne via ffmpeg-devel wrote: > > Unlike H264/H265, AV1 contains no fields to crop encoded output > to specific sizes. > AMD's hardware cannot handle encoding of unaligned dimensions for > AV1, hence it codes 1920x1080 as 1920x1088. > > Add side data to crop the out

Re: [FFmpeg-devel] [PATCH 1/2] lavc/vaapi_encode: Query surface alignment

On Tue, Oct 22, 2024 at 5:28 PM David Rosca wrote: > > It needs to create temporary config to query surface attribute. > --- Ping. Thanks, David > libavcodec/vaapi_encode.c | 66 +++ > libavcodec/vaapi_encode.h | 4 +++ > 2 files chan

[FFmpeg-devel] [PATCH] vulkan_encode_h264/5: Fix uninitialized return value in write_extra_headers

--- libavcodec/vulkan_encode_h264.c | 1 + libavcodec/vulkan_encode_h265.c | 1 + 2 files changed, 2 insertions(+) diff --git a/libavcodec/vulkan_encode_h264.c b/libavcodec/vulkan_encode_h264.c index cdc87fb4ca..fdd3cc8a55 100644 --- a/libavcodec/vulkan_encode_h264.c +++ b/libavcodec/vulkan_encod

[FFmpeg-devel] [PATCH] vulkan: Don't try to early reuse exec context with query

It needs to first get the query before the context can be used again. Fixes getting incorrect values from encode query with async_depth > 1. --- libavutil/vulkan.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/libavutil/vulkan.c b/libavutil/vulkan.c index 2c71312d78..8485c5

[FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Always use slot 0 with 1 l0 ref

Don't need to alternate when only one reference is used. Fixes fail after bf9f921ef7 ("avcodec/hw_base_encode: restrict size of next_prev") due to AV1 writer complaining about incorrect ref_order_hint value. --- libavcodec/vaapi_encode_av1.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

Re: [FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Always use slot 0 with 1 l0 ref

On Tue, Mar 4, 2025 at 8:43 AM David Rosca wrote: > > On Tue, Mar 4, 2025 at 2:55 AM Wang, Fei W > wrote: > > > > On Mon, 2025-03-03 at 14:47 +0100, David Rosca wrote: > > > On Wed, Jan 15, 2025 at 10:41 PM David Rosca > > > wrote: > > > > >

[FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Fix ref_order_hint value for second slot

We always use two slots, even when only one L0 reference is supported by the driver. However we still need to set the correct value for the ref_order_hint of the second slot. Fixes bf9f921ef7 ("avcodec/hw_base_encode: restrict size of next_prev") --- libavcodec/vaapi_encode_av1.c | 3 +++ 1 file

[FFmpeg-devel] [PATCH] lavu/hwcontext_vaapi: Add direct map quirk for Intel i965 driver

vaGetImage is currently used for all maps when read access is required because vaDeriveImage is supposed to be very slow on old Intel HW with i965 driver. However, this is not the case with modern drivers and vaDeriveImage is faster also for reading. Add a new quirk for i965 driver that will keep

Re: [FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Always use slot 0 with 1 l0 ref

On Wed, Jan 15, 2025 at 10:41 PM David Rosca wrote: > > Don't need to alternate when only one reference is used. > Fixes fail after bf9f921ef7 ("avcodec/hw_base_encode: restrict size of > next_prev") > due to AV1 writer complaining about incorre

Re: [FFmpeg-devel] [PATCH] lavc/vaapi_encode_av1: Always use slot 0 with 1 l0 ref

On Tue, Mar 4, 2025 at 2:55 AM Wang, Fei W wrote: > > On Mon, 2025-03-03 at 14:47 +0100, David Rosca wrote: > > On Wed, Jan 15, 2025 at 10:41 PM David Rosca > > wrote: > > > > > > Don't need to alternate when only one reference is used. > > > Fi