FFmpeg

Commit Graph

Author	SHA1	Message	Date
xufuji456	bd2f00f665	codec/aarch64/hevc: add transform_luma_neon got 56% speed up (run_count=1000, CPU=Cortex A53) transform_4x4_luma_neon: 45 transform_4x4_luma_c: 103 Signed-off-by: xufuji456 <839789740@qq.com> Signed-off-by: Martin Storsjö <martin@martin.st>	2 years ago
xufuji456	00a062b8d5	codec/aarch64/hevc:add idct_32x32_neon got 73% speed up (run_count=1000, CPU=Cortex A53) idct_32x32_neon: 4826 idct_32x32_c: 18236 idct_32x32_neon: 4824 idct_32x32_c: 18149 idct_32x32_neon: 4937 idct_32x32_c: 18333 Signed-off-by: Martin Storsjö <martin@martin.st>	2 years ago
J. Dekker	37cde570bc	lavc/aarch64: add clip N macro Signed-off-by: J. Dekker <jdek@itanimul.li>	2 years ago
xufuji456	4b4de07721	libavcodec/hevc: add hevc idct4x4 neon of aarch64 Signed-off-by: Martin Storsjö <martin@martin.st>	2 years ago
Martin Storsjö	ec7fa13eb0	aarch64: hevcdsp_idct: Reuse preexisting macros for transposes Signed-off-by: Martin Storsjö <martin@martin.st>	2 years ago
J. Dekker	ce2f47318b	lavc/aarch64: hevc_add_res add 12bit variants hevc_add_res_4x4_12_c: 46.0 hevc_add_res_4x4_12_neon: 18.7 hevc_add_res_8x8_12_c: 194.7 hevc_add_res_8x8_12_neon: 25.2 hevc_add_res_16x16_12_c: 716.0 hevc_add_res_16x16_12_neon: 69.7 hevc_add_res_32x32_12_c: 3820.7 hevc_add_res_32x32_12_neon: 261.0 Signed-off-by: J. Dekker <jdek@itanimul.li>	2 years ago
J. Dekker	aa9eabb7a5	lavc/aarch64: reformat add_res funcs Signed-off-by: J. Dekker <jdek@itanimul.li>	2 years ago
Martin Storsjö	f27e3ccf06	aarch64: hevc_idct: Fix overflows in idct_dc This is marginally slower, but correct for all input values. The previous implementation failed with certain input seeds, e.g. "checkasm --test=hevc_idct 98". Signed-off-by: Martin Storsjö <martin@martin.st>	4 years ago
Josh Dekker	75c2ddfa61	lavc/aarch64: add HEVC idct_dc NEON Signed-off-by: Josh Dekker <josh@itanimul.li>	4 years ago
Reimar Döffinger	00c916ef61	lavc/aarch64: port HEVC add_residual NEON Speedup is fairly small, around 1.5%, but these are fairly simple. Signed-off-by: Josh Dekker <josh@itanimul.li>	4 years ago
Reimar Döffinger	30f80d855b	lavc/aarch64: port HEVC SIMD idct NEON Makes SIMD-optimized 8x8 and 16x16 idcts for 8 and 10 bit depth available on aarch64. For a UHD HDR (10 bit) sample video these were consuming the most time and this optimization reduced overall decode time from 19.4s to 16.4s, approximately 15% speedup. Test sample was the first 300 frames of "LG 4K HDR Demo - New York.ts", running on Apple M1. Signed-off-by: Josh Dekker <josh@itanimul.li>	4 years ago

11 Commits (ae95cd85dc73319a704f1092b7542978ef8869dd)