Chiebot-Mirror
/
FFmpeg
mirror of https://github.com/FFmpeg/FFmpeg.git
8
0
Fork 0
Code Issues Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
108652 Commits
37 Branches
400 Tags
465 MiB
Tag: Branch: Tree: d2fd0ea1d7
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'd2fd0ea1d7'
${ noResults }
FFmpeg/libavutil/aarch64/Makefile

7 lines
361 B
Raw Normal View History Unescape

aarch64: float_dsp NEON assembler Ported from arm NEON and added vector_dmul_scalar. Functions between 1.5 and 5 times faster than the C implementations using Apple's clang-503.0.19 on A7.
11 years ago
OBJS += aarch64/cpu.o \
aarch64/float_dsp_init.o \
lavu/tx: implement aarch64 NEON SIMD FFT The fastest fast Fourier transform in not just the west, but the world, now for the most popular toy ISA. On a high level, it follows the design of the AVX2 version closely, with the exception that the input is slightly less permuted as we don't have to do lane switching with the input on double 4pt and 8pt. On a low level, the lack of subadd/addsub instructions REALLY penalizes any attempt at writing an FFT. That single register matters a lot, and reloading it simply takes unacceptably long. In x86 land, vendors would've noticed developers need this. In ARM land, you get a badly designed complex multiplication instruction we cannot use, that's not present on 95% of devices. Because only compilers matter, right? Future optimization options are very few, perhaps better register management to use more ld1/st1s. All timings below are in cycles: A53: Length | C | New (lavu) | Old (lavc) | FFTW ------ |-------------|-------------|-------------|----- 4 | 842 | 420 | 1210 | 1460 8 | 1538 | 1020 | 1850 | 2520 16 | 3717 | 1900 | 3700 | 3990 32 | 9156 | 4070 | 8289 | 8860 64 | 21160 | 9931 | 18600 | 19625 128 | 49180 | 23278 | 41922 | 41922 256 | 112073 | 53876 | 93202 | 101092 512 | 252864 | 122884 | 205897 | 207868 1024 | 560512 | 278322 | 458071 | 453053 2048 | 1295402 | 775835 | 1038205 | 1020265 4096 | 3281263 | 2021221 | 2409718 | 2577554 8192 | 8577845 | 4780526 | 5673041 | 6802722 Apple M1 New - Total for len 512 reps 2097152 = 1.459141 s Old - Total for len 512 reps 2097152 = 2.251344 s FFTW - Total for len 512 reps 2097152 = 1.868429 s New - Total for len 1024 reps 4194304 = 6.490080 s Old - Total for len 1024 reps 4194304 = 9.604949 s FFTW - Total for len 1024 reps 4194304 = 7.889281 s New - Total for len 16384 reps 262144 = 10.374001 s Old - Total for len 16384 reps 262144 = 15.266713 s FFTW - Total for len 16384 reps 262144 = 12.341745 s New - Total for len 65536 reps 8192 = 1.769812 s Old - Total for len 65536 reps 8192 = 4.209413 s FFTW - Total for len 65536 reps 8192 = 3.012365 s New - Total for len 131072 reps 4096 = 1.942836 s Old - Segfaults FFTW - Total for len 131072 reps 4096 = 3.713713 s Thanks to wbs for some simplifications, assembler fixes and a review and to jannau for giving it a look.
3 years ago
aarch64/tx_float_init.o \
aarch64: float_dsp NEON assembler Ported from arm NEON and added vector_dmul_scalar. Functions between 1.5 and 5 times faster than the C implementations using Apple's clang-503.0.19 on A7.
11 years ago
lavu/tx: implement aarch64 NEON SIMD FFT The fastest fast Fourier transform in not just the west, but the world, now for the most popular toy ISA. On a high level, it follows the design of the AVX2 version closely, with the exception that the input is slightly less permuted as we don't have to do lane switching with the input on double 4pt and 8pt. On a low level, the lack of subadd/addsub instructions REALLY penalizes any attempt at writing an FFT. That single register matters a lot, and reloading it simply takes unacceptably long. In x86 land, vendors would've noticed developers need this. In ARM land, you get a badly designed complex multiplication instruction we cannot use, that's not present on 95% of devices. Because only compilers matter, right? Future optimization options are very few, perhaps better register management to use more ld1/st1s. All timings below are in cycles: A53: Length | C | New (lavu) | Old (lavc) | FFTW ------ |-------------|-------------|-------------|----- 4 | 842 | 420 | 1210 | 1460 8 | 1538 | 1020 | 1850 | 2520 16 | 3717 | 1900 | 3700 | 3990 32 | 9156 | 4070 | 8289 | 8860 64 | 21160 | 9931 | 18600 | 19625 128 | 49180 | 23278 | 41922 | 41922 256 | 112073 | 53876 | 93202 | 101092 512 | 252864 | 122884 | 205897 | 207868 1024 | 560512 | 278322 | 458071 | 453053 2048 | 1295402 | 775835 | 1038205 | 1020265 4096 | 3281263 | 2021221 | 2409718 | 2577554 8192 | 8577845 | 4780526 | 5673041 | 6802722 Apple M1 New - Total for len 512 reps 2097152 = 1.459141 s Old - Total for len 512 reps 2097152 = 2.251344 s FFTW - Total for len 512 reps 2097152 = 1.868429 s New - Total for len 1024 reps 4194304 = 6.490080 s Old - Total for len 1024 reps 4194304 = 9.604949 s FFTW - Total for len 1024 reps 4194304 = 7.889281 s New - Total for len 16384 reps 262144 = 10.374001 s Old - Total for len 16384 reps 262144 = 15.266713 s FFTW - Total for len 16384 reps 262144 = 12.341745 s New - Total for len 65536 reps 8192 = 1.769812 s Old - Total for len 65536 reps 8192 = 4.209413 s FFTW - Total for len 65536 reps 8192 = 3.012365 s New - Total for len 131072 reps 4096 = 1.942836 s Old - Segfaults FFTW - Total for len 131072 reps 4096 = 3.713713 s Thanks to wbs for some simplifications, assembler fixes and a review and to jannau for giving it a look.
3 years ago
NEON-OBJS += aarch64/float_dsp_neon.o \
aarch64/tx_float_neon.o \