arm64: int32_to_float_fmul neon asm
authorJanne Grunau <janne-libav@jannau.net>
Thu, 3 Dec 2015 10:04:29 +0000 (11:04 +0100)
committerJanne Grunau <janne-libav@jannau.net>
Mon, 14 Dec 2015 15:45:02 +0000 (16:45 +0100)
commita0fc780a2093784e8664f88205ee1b215e109cee
tree3d4f4fbf79d02f6eb93fd93d304af9a6e10e0691
parent705f5e5e155f6f280a360af220fc5b30cfcee702
arm64: int32_to_float_fmul neon asm

3% faster dts decoding on a cortex-a57.

                                 cortex-a57   cortex-a53
int32_to_float_fmul_array8_c:    1270.9       4475.6
int32_to_float_fmul_array8_neon:  328.6        569.2
int32_to_float_fmul_scalar_c:     928.5       4119.6
int32_to_float_fmul_scalar_neon:  309.1        524.1
libavcodec/aarch64/Makefile
libavcodec/aarch64/fmtconvert_init.c [new file with mode: 0644]
libavcodec/aarch64/fmtconvert_neon.S [new file with mode: 0644]
libavcodec/fmtconvert.c
libavcodec/fmtconvert.h