ffmpeg | branch: master | Rémi Denis-Courmont | Sun Sep 27
13:49:09 2015 +0300| [b10b6ac7a902f28e09e37a29c392e2f0c19e9526] | committer:
Anton Khirnov
vdpau: deprecate av_vdpau_get_profile()
This function can intrinsically not deal with codec profile fallback
(for H.264 Constrained Baseline
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Oct 28
21:21:51 2015 +0200| [d35d0c723e3c8fc8cde76bf677f67928f5e179a8] | committer:
Vittorio Giovara
vdpau: remove dysfunctional H.263 support
The VDPAU API never explicitly supported H.263 in the first place.
Signed-off-by: Vittorio
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 26
21:24:12 2014 +0200| [26ab504ad8d2b23535c9a0ad43bf1fd0e6aa0893] | committer:
Anton Khirnov
vdpau/h264: request MAIN rather than BASELINE VDPAU profile for CBP
The H.264 Constrained Baseline Profile (CBP) is a subset of both the
Main
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Oct 26
21:33:58 2014 +0200| [ce91b2eae6ea52fc1a7003566d26db20ca62d745] | committer:
Anton Khirnov
vdpau: return MAIN instead of BASELINE for H.264 CBP
This is the same as the previous change, but for applications using the
old API (such as
ffmpeg | branch: release/2.2 | Rémi Denis-Courmont | Thu Sep
25 11:59:57 2014 +0300| [c7caed88a03567e8777a606f4bd42f093c6b302c] | committer:
Luca Barbato
h264: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: release/2.2 | Rémi Denis-Courmont | Thu Sep
25 11:59:58 2014 +0300| [0989a120f1dec400c54fcb54670cb84bba36d99b] | committer:
Luca Barbato
mpeg12: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Oct 29
19:07:20 2014 +0200| [4ff670d99bebd97429322719089363d83143477d] | committer:
Luca Barbato
hwaccel: Deinitialize hardware acceleration early enough
The application will destroy the underlying hardware handles when
get_format() gets
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Dec 13
18:42:19 2014 +0200| [8b51bcfed75721e0d20832e5a90e1d585318c9d9] | committer:
Anton Khirnov
vdpau: revector macro to reduce line span
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Dec 13
18:42:20 2014 +0200| [8502c1e9ff9c1dbb6e467630c048d098f4064021] | committer:
Anton Khirnov
vdpau: add mapping for H.264 Extended profile
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Dec 13
18:42:21 2014 +0200| [559fa0d41b5cf8f7e1ba89d1ac70e74712eb2816] | committer:
Anton Khirnov
vdpau: add mapping for H.264 Constrained Baseline profile and fallback
Old VDPAU drivers do not support this newly defined profile, so falling
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:06 2014 +0200| [4cfbeef31d4e6096c0596359d212f5d99a7ba4b5] | committer:
Anton Khirnov
h264: factor hwaccel pixel formats list
This is to avoid proliferation of similar tables in following changes.
Signed-off-by: Rémi Denis
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:07 2014 +0200| [57b6704ecd0f56d6a3092e448687cfd837bb0ac1] | committer:
Anton Khirnov
avcodec: add AVCodecContext.sw_pix_fmt
This carries the pixel format that would be used if it were not for
hardware acceleration. This is equal
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:08 2014 +0200| [6c99c92a42add7f6a462114d5a4a53c93c551058] | committer:
Anton Khirnov
libavcodec: add AV_HWACCEL_ALLOW_HIGH_DEPTH flag
This can be used by the application to signal its ability to cope with
video surface of types
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:09 2014 +0200| [c220a60f92dde9c7c118fc4deddff5c1f617cda9] | committer:
Anton Khirnov
vdpau: add helper for surface chroma type and size
Since the VDPAU pixel format does not distinguish between different
VDPAU video surface
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:10 2014 +0200| [ebd5320afd42d4315851f3e0ca7f5d4a6300eb68] | committer:
Anton Khirnov
vdpau: add support for 4:2:2 and 4:4:4 chroma sampling
Signed-off-by: Rémi Denis-Courmont
Signed-off-by: Anton Khirnov
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:11 2014 +0200| [1f9237f2ac46dfbed1bfa1f4f0f1314c2a1d62ec] | committer:
Anton Khirnov
avconv_vdpau: allocate video surface of VDPAU-specified size
Signed-off-by: Rémi Denis-Courmont
Signed-off-by: Anton Khirnov
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Dec 19
19:15:05 2014 +0200| [737d35e33408263c04d7730f5487eed0d04938ba] | committer:
Anton Khirnov
vdpau: add support for the H.264 High 4:4:4 Predictive profile
Signed-off-by: Rémi Denis-Courmont
Signed-off-by: Anton Khirnov
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jan 26
21:17:31 2015 +0200| [60392480181f24ebf3ab48d8ac3614705de90152] | committer:
Luca Barbato
mem: fix pointer pointer aliasing violations
This uses explicit memory copying to read and write pointer to pointers
of arbitrary object types
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Aug 20
23:06:07 2014 +0300| [6ee1cb5740e7490151db7dcec7e20ceaf8a2fe1f] | committer:
Martin Storsjö
libavformat: use MSG_NOSIGNAL when applicable
If the remote end of a connection oriented socket hangs up, generating
an EPIPE error is
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Sep 16
22:17:47 2014 +0300| [1c80c9d7ef809180042257200c7b5f6b81d0b0e2] | committer:
Luca Barbato
hwaccel: Call ->get_format again if hwaccel init fails
This allows the application to fall back on another hwaccel or,
more likely, softw
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Sep 25
11:59:57 2014 +0300| [ce42316321145c435029dc288c2b5f1d48519dde] | committer:
Luca Barbato
h264: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Sep 25
11:59:58 2014 +0300| [4e03b2097ca59843ed65424a572b7acd5345f088] | committer:
Luca Barbato
mpeg12: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: release/2.4 | Rémi Denis-Courmont | Thu Sep
25 11:59:58 2014 +0300| [ce0972ecddef3eee3c03e393108e5728821e3e37] | committer:
Luca Barbato
mpeg12: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: release/2.4 | Rémi Denis-Courmont | Thu Sep
25 11:59:57 2014 +0300| [09c848855a7bf11a9ab4a69971cced260b17817d] | committer:
Luca Barbato
h264: Always invoke the get_format() callback
Signed-off-by: Luca Barbato
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:02 2014 +0300| [319424d25c53f82b87187ce03ed984d438f1bee6] | committer:
Anton Khirnov
vdpau: add helper for VDPAU to libav error codes conversion
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:03 2014 +0300| [fcc1022611f79c2f3aa2f392a5ce14c74be9c1d7] | committer:
Anton Khirnov
vdpau: factor out common end-of-frame handling
Also add error handling.
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:04 2014 +0300| [ce083282f0a8b7d63c4047c30b7bac498f9806dd] | committer:
Anton Khirnov
vdpau: common support for managing the VdpDecoder in avcodec
Using the not so new init and uninit callbacks, avcodec can now take
care of
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:05 2014 +0300| [89ac99ba5f2dc9f69ad3bc294753930eb0b3e4a4] | committer:
Anton Khirnov
vdpau: pass codec-specific parameters from hwaccel
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:06 2014 +0300| [502cde409ca5ee97ef70c2cdede88b9101746ff6] | committer:
Anton Khirnov
vdpau: force reinitialization when output resolution changes
This is necessary to recreate the decoder with the correct parameters,
as not all
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:07 2014 +0300| [e3e158e81f0666b8fe66be9ce1cad63a535920e0] | committer:
Anton Khirnov
vdpau: add av_vdpau_bind_context()
This function provides an explicit VDPAU device and VDPAU driver to
libavcodec, so that the application is
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:08 2014 +0300| [bef067f88c74190cdf7e76d12f02e12e069974aa] | committer:
Anton Khirnov
vdpau: check video surface and decoder capabilities
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Oct 4
16:55:09 2014 +0300| [8de1d67967a9f9e22c66cb0c1e518ae4f30d07dd] | committer:
Anton Khirnov
avconv_vdpau: update to new VDPAU interface
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Oct 6
17:41:00 2014 +0300| [153fadc390d05aa47e5e2c56290401898fe41a23] | committer:
Anton Khirnov
ff_get_format: fix infinite loop
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Oct 8
20:37:00 2014 +0300| [1f29e5d7a2b0950f3b6820896e97e2c02e6a10a9] | committer:
Janne Grunau
h263dec: call get_format after setting resolution and profile
Bug-Id: 541
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Oct 8
20:37:01 2014 +0300| [c80a816142699dea9cf9fa66689a7838a487ed7e] | committer:
Janne Grunau
h263dec: call get_format() on resolution changes
Fail safe if the pixel format changes.
> http://git.videolan.org/gitweb.cgi/ffmpeg.git
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Oct 14
12:59:52 2014 +0300| [e09a947c1a856d9d34199a13ee378fe46be53d3a] | committer:
Anton Khirnov
doc: fix typing mistake
Signed-off-by: Anton Khirnov
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Oct 14
12:59:53 2014 +0300| [c1724623ce0433c6a9ee72133b1fd4db75ec7193] | committer:
Anton Khirnov
vdpau: have av_vdpau_bind_context() fail on unsupported flag
Currently, no flags are supported.
Signed-off-by: Anton Khirnov
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Oct 14
12:59:54 2014 +0300| [d565fef1b83b6c5f8afb32229260b79f67c68109] | committer:
Anton Khirnov
vdpau: add AV_HWACCEL_FLAG_IGNORE_LEVEL to skip the codec level check
Decoding acceleration may work even if the codec level is higher than
the
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 1
18:55:44 2024 +0300| [30797e4ff6c8c537471c386cd019a6a48a721f01] | committer:
Rémi Denis-Courmont
lavc/vp8dsp: add R-V V vp7_idct_dc_add
This just computes the direct coefficient and hands over to code shared
with VP8. Accordingly the
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 1
21:32:56 2024 +0300| [4e120fbbbd087c3acbad6ce2e8c7b1262a5c8632] | committer:
Rémi Denis-Courmont
lavc/vp8dsp: add R-V V vp7_idct_dc_add4y
As with idct_dc_add, most of the code is shared with, and replaces, the
previous VP8 function
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 2
12:03:33 2024 +0300| [121fb846b97db5afd3a24b0bde2be151400104be] | committer:
Rémi Denis-Courmont
lavc/vp7dsp: add R-V V vp7_idct_dc_add4uv
This is almost the same story as vp7_idct_add4y. We just have to use
strided loads of 2 64-bit
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 2
12:13:25 2024 +0300| [225de53c9d446ddf1cc3ece6e99c06c8cce9b78f] | committer:
Rémi Denis-Courmont
lavc/vp8dsp: rework R-V V idct_dc_add4y
DCT-related FFmpeg functions often add an unsigned 8-bit sample to a
signed 16-bit coefficient
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 2
13:23:24 2024 +0300| [0415bb74c81ab0019e48bd2989ddf48d66918e9e] | committer:
Rémi Denis-Courmont
lavc/vp8dsp: remove no longer used macros
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 6
22:21:30 2024 +0300| [c3dbbb316ee44b1685b6437756a23a545fb31fb1] | committer:
Rémi Denis-Courmont
lavc/flacdsp: fix sign extension in R-V V wasted33
We need to use either VWCVT.X.X.V or VSEXT.VF2. The later is preferable
to avoid
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 1
10:55:51 2024 +0300| [8d117024fe05dcf0f45693b47118e4c26e0940af] | committer:
Rémi Denis-Courmont
checkasm: disable unaligned access emulation
The OS may silently fix (emulate) unaligned hardware access exceptions.
This is extremely
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 2
19:54:48 2024 +0300| [04397a29deccf72575a7b02ed56eb37bef2934cc] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: R-V V vc1_inv_trans_8x8
T-Head C908 (cycles):
vc1dsp.vc1_inv_trans_8x8_c: 871.7
vc1dsp.vc1_inv_trans_8x8_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jun 3
19:11:33 2024 +0300| [a169f3bca5937cf71d61de1fb85b522bd69c6cfc] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: R-V V vc1_inv_trans_8x4
T-Head C908 (cycles):
vc1dsp.vc1_inv_trans_8x4_c: 626.2
vc1dsp.vc1_inv_trans_8x4_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jun 4
19:32:18 2024 +0300| [6ffa639c8aa6046178d9e04256fb8961c31dc734] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: R-V V vc1_inv_trans_4x8
T-Head C908 (cycles):
vc1dsp.vc1_inv_trans_4x8_c: 653.2
vc1dsp.vc1_inv_trans_4x8_rvv_i32: 234.0
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jun 4
21:33:53 2024 +0300| [3152c684cb59e929ed572039ad9cdb3dde273dc3] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: R-V V vc1_inv_trans_4x4
T-Head C908 (cycles):
vc1dsp.vc1_inv_trans_4x4_c: 310.7
vc1dsp.vc1_inv_trans_4x4_rvv_i32: 120.0
We
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jun 4
22:57:33 2024 +0300| [79dfdac4dbc7d6cb5316292247b26b39962dd471] | committer:
Rémi Denis-Courmont
sws/input: R-V V rgb24ToY & bgr24ToY
T-Head C908:
rgb24_to_y_8_c:2.0
rgb24_to_y_8_rvv_i32: 2.7
rgb24_to_y_1
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jun 5
19:32:31 2024 +0300| [e0f4d185f1ea12e9a7bfa078a4224687102751e7] | committer:
Rémi Denis-Courmont
sws/input: R-V V rgb24ToUV_half and bgr24ToUV_half
T-Head C908:
rgb24_to_uv_half_4_c: 2.0
rgb24_to_uv_half_4_rvv_i32: 3.5
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jun 5
18:36:42 2024 +0300| [3ef5867e4b6d7966a061c99f62da274e7a5386b7] | committer:
Rémi Denis-Courmont
sws/input: R-V V rgb24ToUV and bgr24ToUV
T-Head C908:
rgb24_to_uv_8_c:2.7
rgb24_to_uv_8_rvv_i32: 3.2
rgb24_to_uv_128_c
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jun 5
21:55:22 2024 +0300| [658439934b255215548269116481e2d48da9ee3b] | committer:
Rémi Denis-Courmont
lavc/vp8dsp: R-V V vp8_idct_add
T-Head C908 (cycles):
vp8_idct_add_c: 312.2
vp8_idct_add_rvv_i32: 117.0
> http://git.videolan.
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 8
15:04:46 2024 +0300| [daac101e61697cdc39e6e01d77ff6201efc70845] | committer:
Rémi Denis-Courmont
lavc/aacencdsp: fix rounding in R-V V quantize_bands
We need to round toward zero here.
> http://git.videolan.org/gitweb.cgi/ffmpeg.
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 6
17:49:21 2024 +0300| [fcb106a17b702e19ff1373418019e64557da] | committer:
Rémi Denis-Courmont
sws/input: R-V V 32-bit RGB to Y
T-Head C908:
abgr_to_y_8_c:2.5
abgr_to_y_8_rvv_i32: 2.2
abgr_to_y_128_c: 37.0
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 6
18:12:50 2024 +0300| [e2f069905e7664e37e0ad54ddd3ffbfcdbb268c0] | committer:
Rémi Denis-Courmont
sws/input: R-V V 32-bit RGB to UV
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=e2f069905e7664e37e0ad54ddd3ffbfcdbb26
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 6
21:15:08 2024 +0300| [7a3369398f8b7869bec56fd2b96f22d5cc967e30] | committer:
Rémi Denis-Courmont
sws/input: R-V V 32-bit RGB to halved UV
T-Head C908:
abgr_to_uv_half_8_c:2.2
abgr_to_uv_half_8_rvv_i32: 3.5
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 7
18:36:25 2024 +0300| [417957ec5eb948709c04ce081b7758ba8796b10b] | committer:
Rémi Denis-Courmont
sws/range_convert: R-V V to/from JPEG
C908 X60
chrRangeFromJpeg_8_c: 2.72.5
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jun 10
20:29:56 2024 +0300| [b6f37ffba71fa26b6176eb964cadcb442a115a54] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: match C block layout in inv_trans_4x8_rvv
Although checkasm does not verify this, the decoder requires that the
transform
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jun 10
20:29:56 2024 +0300| [6c05069e680f6b9055ac14bc19c663914b554fdb] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: match C block layout in inv_trans_4x4_rvv
Although checkasm does not verify this, the decoder requires that the
transform
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 8
09:33:49 2024 +0300| [18adaf9fe558587cb1b707c647af83015b69da48] | committer:
Rémi Denis-Courmont
checkasm/lls: adjust buffer sizes and alignments
var must be padded.
param has `order + 1`, not `order` elements and is *not* over
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 8
10:57:15 2024 +0300| [378d1b06c350e09ac604566130b39126fd858478] | committer:
Rémi Denis-Courmont
riscv: probe for Zbb extension at load time
Due to hysterical raisins, most RISC-V Linux distributions target a
RV64GC baseline excluding
ffmpeg | branch: master | Rémi Denis-Courmont | Thu May 30
22:04:10 2024 +0300| [324899b7483529c336f399022c63721df14663ef] | committer:
Rémi Denis-Courmont
lavu/riscv: use Zbb REV8 at run-time
This adds runtime support to use Zbb REV8 for 32- and 64-bit byte-wise
swaps. The result is about
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 8
12:02:27 2024 +0300| [98db1409102c5443491c2738aa601531f7f9b938] | committer:
Rémi Denis-Courmont
lavu/riscv: use Zbb CPOP/CPOPW at run-time
Zbb staticZbb dynamic I baseline
popcount 1.336129286 3.469067758
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jun 8
13:27:10 2024 +0300| [f6d0a41c8ca67672f8e4b3c6e16cb3abd9f379bf] | committer:
Rémi Denis-Courmont
lavu/riscv: use Zbb CLZ/CTZ/CLZW/CTZW at run-time
Zbb staticZbb dynamic I baseline
clz 0.668032642 1.336072283
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 7
19:43:09 2024 +0300| [424ac848396b634d87ed148f39c22988bb17440f] | committer:
Rémi Denis-Courmont
lavu/arm: remove GCC 4.6- stuff
Since the C11 support is required, those GCC versions can no longer be
supported anyhow. (Clang pretends
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 7
19:43:09 2024 +0300| [0231097d1bd6803a4724465797fcaddbe1fcc92d] | committer:
Rémi Denis-Courmont
lavu/x86: remove GCC 4.4- stuff
Since the C11 support is required, those GCC versions can no longer be
supported anyhow. (Clang pretends
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 7
21:08:21 2024 +0300| [c5f69719bcb407072cc0a3c7b852a38c48b9e561] | committer:
Rémi Denis-Courmont
lavu/bswap: remove some inline assembler
C code or compiler built-ins are preferable over inline assembler for
byte-swaps as it allows
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 7
21:17:43 2024 +0300| [a5d2d267424a0c0568d703362dc89795734f3d54] | committer:
Rémi Denis-Courmont
sh4: remove architecture
Support for SuperH was dropped over a decade ago. There no longer is any
architecture-specific code to be found
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 9
14:43:34 2024 +0300| [4819aeebf4d0cd497913cd45b88ec3eb02da0bd8] | committer:
Rémi Denis-Courmont
avr32: remove explicit support
The vendor has long since switched to Arm, with the last product
reaching their official end-of-life over
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 9
15:00:56 2024 +0300| [4ef97c0d7de7825d31c96cb1cad17e0125b9] | committer:
Rémi Denis-Courmont
configure: remove stray bfin architecture
This seems to have been omitted in
880e2aa23645ed9871c66ee1cbd00f93c72d2d73.
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jun 11
19:46:13 2024 +0300| [0d748eec8d52715f2949bd03519d39f2bf5baac4] | committer:
Rémi Denis-Courmont
configure: detect 64-bit generic platforms
Currently, any unrecognised platform is treated as 32-bit. This should
detect *most* 64-bit
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jun 26
21:44:23 2024 +0300| [d5e603ddc07211d3c2376d592ec161113c55e036] | committer:
Rémi Denis-Courmont
lavu/lls: remove useless VSETVL
This changes neither VL nor VTYPE, so it can safely be removed.
> http://git.videolan.org/gitweb.
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 27
21:21:29 2024 +0300| [349c49fd1bc5ae2d81ea8d93e0b31823fe8b2c59] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: fix overflow in R-V V inv_trans_8
The last set of additions/subtractions can break the 16-bit limit, and
require 17 bits of
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jun 27
22:00:05 2024 +0300| [b818dff8d8e16f0831a379a5cdc15e813474a841] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: fix potential overflow in R-V V inv_trans_4
Judging by the coefficients, the last round of add/sub can overflow
to 17 bits
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jun 11
17:48:00 2024 +0300| [2c900d4c11199561673cbffadeebb5cff62a7d8c] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: factor R-V V inv_trans_8 code
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 30
14:15:33 2024 +0300| [78e1565f847588cb3d34d1a920902fb253d9c03c] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: fuse multiply-adds in R-V V inv_trans_4
T-Head C908 (cycles):before after
vc1dsp.vc1_inv_trans_4x4_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 30
14:15:33 2024 +0300| [4a2de380b7c27f4542af03fdf91d913fabe091fc] | committer:
Rémi Denis-Courmont
lavc/vc1dsp: fuse multiply-adds in R-V V inv_trans_8
T-Head C908 (cycles) before after
vc1dsp.vc1_inv_trans_4x8_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jun 28
21:56:12 2024 +0300| [5a6e333fc7ac514255bcd6b424924d92ef558bf0] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit luma loop filter
T-Head C908 (cycles):
h264_h_loop_filter_luma_8bpp_c: 297.5
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jun 30
11:24:43 2024 +0300| [e2af5904f0fa86dbb2b7755ab579f54d14523e8e] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit MBAFF loop filter
Performance is (unfortunately) the same as with non-MBAFF, since the
hardware under test does
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jul 1
23:41:37 2024 +0300| [e0eff64ed1e60d14391d55a91732ec612abf3f64] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_idct8_add4
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jul 1
23:41:37 2024 +0300| [30475c95ba50d40cf7605cb382bfd7852c825deb] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_idct_add16
While this *tends* to be faster than plain C, the performance numbers
are all over the place
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jul 1
23:41:37 2024 +0300| [d1f0c1fbf8db8dffd514c706905a2a8b8e986cb4] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_idct_add16intra
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jul 2
22:03:07 2024 +0300| [f447189b0c8067edf54a16c8c6d5513b2de77276] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_idct_add
T-Head C908 (cycles):
h264_idct4_add_8bpp_c: 271.5
h264_idct4_add_8bpp_rvv_i32: 91.5
> h
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jul 3
19:49:54 2024 +0300| [f9d1230224c0012c1731634910e65a0ff734d1ef] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_idct8_add
T-Head C908 (cycles):
h264_idct8_add_8bpp_c: 1072.0
h264_idct8_add_8bpp_rvv_i32: 318.5
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jul 4
21:38:48 2024 +0300| [3606e592ea239ebaf2e30570bf99abeefd52b7e0] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_weight_pixels
There are two implementations here:
- a generic scalable one processing two columns at a
ffmpeg | branch: master | Rémi Denis-Courmont | Fri Jul 5
23:03:28 2024 +0300| [f1ed351d3b3c247ec372cdb3dc8b112e570d5dec] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V 8-bit h264_biweight_pixels
T-Head C908:
h264_biweight2_8_c:58.0
h264_biweight2_8_rvv_i32: 11.2
ffmpeg | branch: master | Rémi Denis-Courmont | Wed Jul 10
23:20:54 2024 +0300| [d28a7e8eb7eb2cb2efcc9c8e9d506cdddcae0a26] | committer:
Rémi Denis-Courmont
lavc/h264dsp: avoid \+ expansion
This seems to be unsupported by LLVM-as.
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jul 2
22:03:07 2024 +0300| [4e0e872881ff57f1953517067cf82d7ea925b432] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V high-depth h264_idct_add
T-Head C908 (cycles):
h264_idct4_add_9bpp_c:248.2
h264_idct4_add_9bpp_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jul 7
14:39:11 2024 +0300| [a194131cb6adb6d5f2156e785f1bf859d9449def] | committer:
Rémi Denis-Courmont
lavc/h264dsp: remove MMI 8-bit chroma DC dequant
The function is exactly identical to the C reference, only with the
constant propagated
ffmpeg | branch: master | Rémi Denis-Courmont | Sun Jul 7
14:39:11 2024 +0300| [8b3d997bede0e065f19bcaa96f99dfa9978fd181] | committer:
Rémi Denis-Courmont
lavc/h264dsp: remove MMI 8-bit 4:2:2 chroma DC dequant
The function is exactly identical to the C reference, only with the
constant
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jul 11
22:01:25 2024 +0300| [c654e37254efb6de5d3f1355dd7936979dd0dca3] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V high-depth h264_idct8_add
Unlike the 8-bit version, we need two iterations to process this within
128-bit vectors
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jul 13
15:43:27 2024 +0300| [7744c08240808c8517a9c088b465c15235a34c86] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V add_pixels4 and 8-bit add_pixels8
T-Head C908 (cycles):
h264_add_pixels4_8bpp_c:93.5
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jul 13
16:01:14 2024 +0300| [3002310b707071ec7aa671ab7428feed5ee9740c] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V high-depth add_pixels8
T-Head C908 (cycles);
h264_add_pixels8_9bpp_c:270.5
h264_add_pixels8_9bpp_rvv_i32
ffmpeg | branch: master | Rémi Denis-Courmont | Mon Jul 15
21:19:39 2024 +0300| [483fd732abc3eda80681811250cd4d2937ade265] | committer:
Rémi Denis-Courmont
lavc/h264dsp: R-V V high-depth idct_add{,intra}16, idct8_add4
As with 8-bit, this tends to be faster, but results are all over the
place
ffmpeg | branch: master | Rémi Denis-Courmont | Sat Jul 13
20:28:34 2024 +0300| [c4c811b3d92721dd14452a02d092ca220c9f6b8b] | committer:
Rémi Denis-Courmont
checkasm/h264dsp: test TX bypass
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c4c811b3d92721dd14452a02d092ca220c9f6
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jul 16
22:45:14 2024 +0300| [d15169c51fd5bc6ea768c83977315954bf583e57] | committer:
Rémi Denis-Courmont
lavc/h264dsp: factor some mostly identical R-V V code
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jul 18
21:24:35 2024 +0300| [0a5b5bae89ecf87a8fa79ae2f133ed926927b98b] | committer:
Rémi Denis-Courmont
lavc/h264dsp: correct VL and LMUL in idct_dc_add
T-Head C908 (cycles):
h264_idct4_dc_add_8bpp_c:94.7
ffmpeg | branch: master | Rémi Denis-Courmont | Thu Jul 18
20:53:22 2024 +0300| [245f76ad74519c0f2a2f82d74041eb48c72d7eba] | committer:
Rémi Denis-Courmont
lavc/h264dsp: reuse the R-V V IDCT DC add functions
This reuses the DC bypass functions from the multiple IDCT functions, to
leverage
ffmpeg | branch: master | Rémi Denis-Courmont | Tue Jul 16
22:59:34 2024 +0300| [9135dffd177d457a8a1781b9e6c6d400648165cb] | committer:
Rémi Denis-Courmont
lavc/h264dsp: reduce spills in R-V V idct_add16
> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commi
1 - 100 of 435 matches
Mail list logo
