FFmpeg

Commit Graph

Author	SHA1	Message	Date
Ronald S. Bultje	715f139c9b	vp9lpf/x86: make filter_16_h work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	8915320db9	vp9lpf/x86: make filter_48/84/88_h work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	725a216481	vp9lpf/x86: make filter_44_h work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	5bfa96c4b3	vp9lpf/x86: make filter_16_v work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	b905e8d2fe	vp9lpf/x86: make filter_48/84_v work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	37637e6590	vp9lpf/x86: make filter_88_v work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	be10834bd9	vp9lpf/x86: make filter_44_v work on 32-bit. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	7c62891efe	vp9lpf/x86: save one register in SIGN_ADD/SUB. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	c6375a83d1	vp9lpf/x86: store unpacked intermediates for filter6/14 on stack. filter16 goes from 508 to 482 (h) or 346 to 314 (v) cycles; filter88 goes from 240 to 238 (h) or 174 to 165 (v) cycles, measured on TOS. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	4ce8ba72f9	vp9lpf/x86: move variable assigned inside macro branch. The value is not used outside the branch. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	e4961035b2	vp9lpf/x86: simplify ABSSUM_CMP by inverting the comparison meaning. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	683da2788e	vp9lpf/x86: remove unused register from ABSSUB_CMP macro. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	6e74e9636b	vp9lpf/x86: slightly simplify 44/48/84/88 h stores. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	6411c328a2	vp9lpf/x86: make cglobal statement more conservative in register allocation. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	a6e288d624	vp9lpf/x86: save one register in loopfilter surface coverage. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Clément Bœsch	0ed21bdc9e	vp9lpf/x86: add ff_vp9_loop_filter_[vh]_44_16_{sse2,ssse3,avx}. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Clément Bœsch	f2e3d706a1	vp9lpf/x86: add ff_vp9_loop_filter_h_{48,84}_16_{sse2,ssse3,avx}(). Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
James Almer	92d47550ea	vp9lpf/x86: add an SSE2 version of vp9_loop_filter_[vh]_88_16 Similar gains as the ssse3 version once again Additional improvements by Clément Bœsch <u@pkh.me>. Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Clément Bœsch	6bea478158	vp9lpf/x86: add ff_vp9_loop_filter_[vh]_88_16_{ssse3,avx}. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
James Almer	1f451eed60	vp9lpf/x86: add ff_vp9_loop_filter_[vh]_16_16_sse2(). Similar gains in performance as the SSSE3 version Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Clément Bœsch	a692724c58	vp9lpf/x86: add x86 SSSE3/AVX SIMD for vp9_loop_filter_[vh]_16_16. Signed-off-by: Anton Khirnov <anton@khirnov.net>	8 years ago
Ronald S. Bultje	a4edaa0270	vp9: add mxext versions of the single-block (w=8,npx=8) h/v loopfilters. Each takes about 0.1% of runtime in my profiles, and they didn't have any SIMD yet so far (we only had simd for npx=16 double-block versions).	8 years ago
Ronald S. Bultje	7ca422bb1b	vp9: add mxext versions of the single-block (w=4,npx=8) h/v loopfilters. Each takes about 0.5% of runtime in my profiles, and they didn't have any SIMD yet so far (we only had simd for npx=16 double-block versions).	8 years ago
Ronald S. Bultje	3aefca68ca	vp9/x86: add myself to copyright holders for loopfilter assembly.	10 years ago
Ronald S. Bultje	afd8c464b7	vp9/x86: make filter_16_h work on 32-bit.	10 years ago
Ronald S. Bultje	b26bc3520f	vp9/x86: make filter_48/84/88_h work on 32-bit.	10 years ago
Ronald S. Bultje	8a1cff1c35	vp9/x86: make filter_44_h work on 32-bit.	10 years ago
Ronald S. Bultje	047088b8c6	vp9/x86: make filter_16_v work on 32-bit.	10 years ago
Ronald S. Bultje	0cc9c23ea1	vp9/x86: make filter_48/84_v work on 32-bit.	10 years ago
Ronald S. Bultje	6433a9133f	vp9/x86: make filter_88_v work on 32-bit.	10 years ago
Ronald S. Bultje	75f8e52089	vp9/x86: make filter_44_v work on 32-bit.	10 years ago
Ronald S. Bultje	7f80c3344c	vp8/x86: save one register in SIGN_ADD/SUB.	10 years ago
Ronald S. Bultje	8ea2194ebb	vp9/x86: store unpacked intermediates for filter6/14 on stack. filter16 goes from 508 to 482 (h) or 346 to 314 (v) cycles; filter88 goes from 240 to 238 (h) or 174 to 165 (v) cycles, measured on TOS.	10 years ago
Ronald S. Bultje	e42409479f	vp8/x86: move variable assigned inside macro branch. The value is not used outside the branch.	10 years ago
Ronald S. Bultje	418c202c63	vp9/x86: simplify ABSSUM_CMP by inverting the comparison meaning.	10 years ago
Ronald S. Bultje	d1c55654e1	vp8/x86: remove unused register from ABSSUB_CMP macro.	10 years ago
Ronald S. Bultje	e59bd08986	vp9/x86: slightly simplify 44/48/84/88 h stores.	10 years ago
Ronald S. Bultje	8132629bd5	vp9/x86: make cglobal statement more conservative in register allocation.	10 years ago
Ronald S. Bultje	c013ca58c5	vp9/x86: save one register in loopfilter surface coverage.	10 years ago
Michael Niedermayer	41d82b85ab	avcodec/x86/vp9lpf: Always include x86util.asm Fixes executable stack Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	10 years ago
Christophe Gisquet	4e128ab0b1	x86: vpx/h264/hevc/mpeg2: share constants Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	10 years ago
James Almer	de417982e8	x86/vp9lpf: use fewer instructions in SPLATB_MIX Signed-off-by: James Almer <jamrial@gmail.com> Signed-off-by: Michael Niedermayer <michaelni@gmx.at>	10 years ago
Clément Bœsch	62d31307c1	avcodec/x86/vp9lpf: add a comment above a bunch of SWAP.	11 years ago
Clément Bœsch	f0d368d758	avcodec/x86/vp9lpf: merge a few movs with other instructions.	11 years ago
Clément Bœsch	010732b73a	vp9/x86: simplify FILTER_INIT. In the 2 FILTER_INIT usages, the source is already preloaded so that extra complexity taken from FILTER_UPDATE is not necessary. Also add forgotten "mask" argument in FILTER_{INIT,UPDATE} comments.	11 years ago
Clément Bœsch	b8d002dc95	vp9/x86: clarify mixed splatb.	11 years ago
Clément Bœsch	669d4f9053	x86/vp9lpf: simplify 2nd transpose in 44/48/88/84. For non-avx optims, this saves 8 movs. before: 1785 decicycles in ff_vp9_loop_filter_h_44_16_ssse3, 524129 runs, 159 skips 3327 decicycles in ff_vp9_loop_filter_h_48_16_ssse3, 262116 runs, 28 skips 2712 decicycles in ff_vp9_loop_filter_h_88_16_ssse3, 4193729 runs, 575 skips 3237 decicycles in ff_vp9_loop_filter_h_84_16_ssse3, 524061 runs, 227 skips after: 1768 decicycles in ff_vp9_loop_filter_h_44_16_ssse3, 524062 runs, 226 skips 3310 decicycles in ff_vp9_loop_filter_h_48_16_ssse3, 262107 runs, 37 skips 2719 decicycles in ff_vp9_loop_filter_h_88_16_ssse3, 4193954 runs, 350 skips 3184 decicycles in ff_vp9_loop_filter_h_84_16_ssse3, 524236 runs, 52 skips	11 years ago
Clément Bœsch	d92a725329	x86/vp9lpf: remove 8 SWAPs in 84/48 transpose.	11 years ago
Clément Bœsch	97dde561de	x86/vp9lpf: remove braindead double pxor.	11 years ago
Clément Bœsch	9a3b05b0a9	x86/vp9lpf: save a few mov in flat8in/hev masks calc.	11 years ago

39 Commits (e6628ac8ffd3124e2d935cab565d9f76d895f43f)