ffmpeg | branch: master | gxw | Wed Dec 29 18:18:22
2021 +0800| [8ca7d474c19f856b28fa588e040b23b1a58368ed] | committer: Michael
Niedermayer
avcodec: [loongarch] Optimize prefetch with loongarch.
./ffmpeg -i ../1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -y /dev/null -an
before:296
after :308
ffmpeg | branch: master | gxw | Wed Dec 15 11:51:07
2021 +0800| [3f294ec8799390af0bfc6916c312e5b1e0035c10] | committer: Michael
Niedermayer
avcodec: [loongarch] Optimize h264dsp with LASX.
./ffmpeg -i ../1_h264_1080p_30fps_3Mbps.mp4 -f rawvideo -y /dev/null -an
before:225
after :282
Change
ffmpeg | branch: master | gxw | Mon Apr 12 23:37:04
2021 +0800| [464d28c0708f23fb35271265f8a5c73fbb51db44] | committer: Michael
Niedermayer
avcodec/mips: Refine ff_h264_h_lpf_luma_inter_msa
Using mask to avoid judgment, H264 4K decoding speed
improved about 0.1fps tested on 3A4000
Signed-off
ffmpeg | branch: master | gxw | Mon Apr 12 23:37:03
2021 +0800| [6458c6bdb4931e1903b5b39df82f32e673e15f12] | committer: Michael
Niedermayer
avcodec/mips: Optimize function ff_h264_loop_filter_strength_msa.
Speed of decoding H264 1080P: 5.05x ==> 5.13x
Signed-off-by: Shiyou Yin
Signed-
ffmpeg | branch: master | gxw | Thu Sep 3 14:29:51
2020 +0800| [a4f7b09536c1d4b643992c03fa7105d2229da28e] | committer: Michael
Niedermayer
avcodec/mips: [loongson] Fixed mmi optimization
Test case fate-checkasm-h264pred failed in latest community code.
This patch fixed the bug.
Signed-off
ffmpeg | branch: master | gxw | Mon Oct 21 15:56:47
2019 +0800| [648b422e171d5eab18f6c6fd346e4050d717b936] | committer: Michael
Niedermayer
avcodec/mips: msa optimizations for vc1dsp
Performance of WMV3 decoding has speed up from 3.66x to 5.23x tested on 3A4000.
Reviewed-by: Shiyou Yin
ffmpeg | branch: master | gxw | Sat Oct 12 10:48:19
2019 +0800| [21d19f49b78e76acd17ccbf7e632c570f9cc642e] | committer: Michael
Niedermayer
avcodec/mips: Fixed four warnings in vc1dsp
Change the stride argument to ptrdiff_t in the following functions:
ff_put_no_rnd_vc1_chroma_mc8_mmi
ffmpeg | branch: master | gxw | Tue Aug 6 19:11:16
2019 +0800| [92fc0bfa54d82e3bb18138f9f0ebc3a7322b64ce] | committer: Michael
Niedermayer
avutil/mips: refactor msa SLDI_Bn_0 and SLDI_Bn macros.
Changing details as following:
1. The previous order of parameters are irregular and difficult
ffmpeg | branch: master | gxw | Wed Aug 7 17:52:00
2019 +0800| [a3e572d96fd1dd6291f6b28e173db858c08ff8d8] | committer: Michael
Niedermayer
avutil/mips: refine msa macros CLIP_*.
Changing details as following:
1. Remove the local variable 'out_m' in 'CLIP_SH' and store the result in
source
ffmpeg | branch: release/3.4 | gxw | Mon Dec 24
14:07:44 2018 +0800| [4656ad1e962512db26ef40e8c280b942468107ae] | committer:
Michael Niedermayer
avcodec/mips: Fix failed case: hevc-conformance-AMP_A_Samsung_* when enable msa
The AV_INPUT_BUFFER_PADDING_SIZE has been increased to 64
ffmpeg | branch: release/4.0 | gxw | Mon Dec 24
14:07:44 2018 +0800| [4dbfbcef16703ed44c1c1605827cb27945a3c897] | committer:
Michael Niedermayer
avcodec/mips: Fix failed case: hevc-conformance-AMP_A_Samsung_* when enable msa
The AV_INPUT_BUFFER_PADDING_SIZE has been increased to 64
ffmpeg | branch: master | gxw | Mon Feb 25 18:13:51
2019 +0800| [4571c7c05d8488cbc2ae0b337751f132f9670fec] | committer: Michael
Niedermayer
avcodec/mips: [loongson] mmi optimizations for VP9 put and avg functions
VP9 decoding speed improved about 60.5%(from 38fps to 61fps, tested on loongson
ffmpeg | branch: master | gxw | Tue Feb 12 18:55:48
2019 +0800| [1466dc140d49d70ce07e03a5aebb21725166045a] | committer: Michael
Niedermayer
avcodec/mips: [loongson] optimize theora decoding with mmi.
Optimize theora decoding with mmi in functions:
1. ff_vp3_idct_add_mmi
2. ff_vp3_idct_put_mmi
ffmpeg | branch: release/4.1 | gxw | Mon Dec 24
14:07:44 2018 +0800| [929b5519d8e9fb0bada84a708c7b43bdb447574d] | committer:
Michael Niedermayer
avcodec/mips: Fix failed case: hevc-conformance-AMP_A_Samsung_* when enable msa
The AV_INPUT_BUFFER_PADDING_SIZE has been increased to 64
ffmpeg | branch: release/3.2 | gxw | Mon Dec 24
14:07:44 2018 +0800| [0f9cb4e726fc4771a61a60731c8c0482a72daf23] | committer:
Michael Niedermayer
avcodec/mips: Fix failed case: hevc-conformance-AMP_A_Samsung_* when enable msa
The AV_INPUT_BUFFER_PADDING_SIZE has been increased to 64
ffmpeg | branch: master | gxw | Thu Dec 27 14:16:22
2018 +0800| [d86f698e386431e589c0085ecff959a2baa4b8ef] | committer: Michael
Niedermayer
avcodec/mips: [loongson] optimize theora decoding in vp3dsp.
Optimize theora decoding with msa in functions:
1. ff_vp3_idct_add_msa
2
ffmpeg | branch: master | gxw | Mon Dec 24 14:07:44
2018 +0800| [f652c7a45c60427db0a89fae665e63b546af6ebb] | committer: Michael
Niedermayer
avcodec/mips: Fix failed case: hevc-conformance-AMP_A_Samsung_* when enable msa
The AV_INPUT_BUFFER_PADDING_SIZE has been increased to 64, but the value
ffmpeg | branch: master | gxw | Wed Sep 5 18:31:06
2018 +0800| [090647da84f975c7ffb163436040cc8aecf46a9c] | committer: Michael
Niedermayer
avcodec/mips: [loongson] optimize vp8 decoding in vp8dsp.
Optimize vp8 loop filter with mmi, four functions optimized:
1. ff_vp8_h_loop_filter8uv_mmi.
2
18 matches
Mail list logo
