glibc/sysdeps/x86_64/fpu/multiarch/Makefile
H.J. Lu 5c7adbd8ed x86-64: Add log2f with FMA
For workload-spec2017.wrf, on Skylake, it improves performance by:

                           Before            After     Improvement
reciprocal-throughput      16.5937          14.0789       17%
latency                    41.7755          35.3586       18%

	* sysdeps/x86_64/fpu/multiarch/Makefile (libm-sysdep_routines):
	Add e_log2f-fma.
	(CFLAGS-e_log2f-fma.c): New.
	* sysdeps/x86_64/fpu/multiarch/e_log2f-fma.c: New file.
	* sysdeps/x86_64/fpu/multiarch/e_log2f.c: Likewise.
2017-10-22 08:06:58 -07:00

143 lines
5.1 KiB
Makefile

ifeq ($(subdir),math)
libm-sysdep_routines += s_floor-c s_ceil-c s_floorf-c s_ceilf-c \
s_rint-c s_rintf-c s_nearbyint-c s_nearbyintf-c \
s_trunc-c s_truncf-c
libm-sysdep_routines += s_ceil-sse4_1 s_ceilf-sse4_1 s_floor-sse4_1 \
s_floorf-sse4_1 s_nearbyint-sse4_1 \
s_nearbyintf-sse4_1 s_rint-sse4_1 s_rintf-sse4_1 \
s_trunc-sse4_1 s_truncf-sse4_1
libm-sysdep_routines += e_exp-fma e_log-fma e_pow-fma s_atan-fma \
e_asin-fma e_atan2-fma s_sin-fma s_tan-fma \
mplog-fma mpa-fma slowexp-fma slowpow-fma \
sincos32-fma doasin-fma dosincos-fma \
halfulp-fma mpexp-fma \
mpatan2-fma mpatan-fma mpsqrt-fma mptan-fma
CFLAGS-doasin-fma.c = -mfma -mavx2
CFLAGS-dosincos-fma.c = -mfma -mavx2
CFLAGS-e_asin-fma.c = -mfma -mavx2
CFLAGS-e_atan2-fma.c = -mfma -mavx2
CFLAGS-e_exp-fma.c = -mfma -mavx2
CFLAGS-e_log-fma.c = -mfma -mavx2
CFLAGS-e_pow-fma.c = -mfma -mavx2 $(config-cflags-nofma)
CFLAGS-halfulp-fma.c = -mfma -mavx2
CFLAGS-mpa-fma.c = -mfma -mavx2
CFLAGS-mpatan-fma.c = -mfma -mavx2
CFLAGS-mpatan2-fma.c = -mfma -mavx2
CFLAGS-mpexp-fma.c = -mfma -mavx2
CFLAGS-mplog-fma.c = -mfma -mavx2
CFLAGS-mpsqrt-fma.c = -mfma -mavx2
CFLAGS-mptan-fma.c = -mfma -mavx2
CFLAGS-s_atan-fma.c = -mfma -mavx2
CFLAGS-sincos32-fma.c = -mfma -mavx2
CFLAGS-slowexp-fma.c = -mfma -mavx2
CFLAGS-slowpow-fma.c = -mfma -mavx2
CFLAGS-s_sin-fma.c = -mfma -mavx2
CFLAGS-s_tan-fma.c = -mfma -mavx2
libm-sysdep_routines += e_exp2f-fma e_expf-fma e_log2f-fma e_logf-fma
CFLAGS-e_exp2f-fma.c = -mfma -mavx2
CFLAGS-e_expf-fma.c = -mfma -mavx2
CFLAGS-e_log2f-fma.c = -mfma -mavx2
CFLAGS-e_logf-fma.c = -mfma -mavx2
libm-sysdep_routines += e_exp-fma4 e_log-fma4 e_pow-fma4 s_atan-fma4 \
e_asin-fma4 e_atan2-fma4 s_sin-fma4 s_tan-fma4 \
mplog-fma4 mpa-fma4 slowexp-fma4 slowpow-fma4 \
sincos32-fma4 doasin-fma4 dosincos-fma4 \
halfulp-fma4 mpexp-fma4 \
mpatan2-fma4 mpatan-fma4 mpsqrt-fma4 mptan-fma4
CFLAGS-doasin-fma4.c = -mfma4
CFLAGS-dosincos-fma4.c = -mfma4
CFLAGS-e_asin-fma4.c = -mfma4
CFLAGS-e_atan2-fma4.c = -mfma4
CFLAGS-e_exp-fma4.c = -mfma4
CFLAGS-e_log-fma4.c = -mfma4
CFLAGS-e_pow-fma4.c = -mfma4 $(config-cflags-nofma)
CFLAGS-halfulp-fma4.c = -mfma4
CFLAGS-mpa-fma4.c = -mfma4
CFLAGS-mpatan-fma4.c = -mfma4
CFLAGS-mpatan2-fma4.c = -mfma4
CFLAGS-mpexp-fma4.c = -mfma4
CFLAGS-mplog-fma4.c = -mfma4
CFLAGS-mpsqrt-fma4.c = -mfma4
CFLAGS-mptan-fma4.c = -mfma4
CFLAGS-s_atan-fma4.c = -mfma4
CFLAGS-sincos32-fma4.c = -mfma4
CFLAGS-slowexp-fma4.c = -mfma4
CFLAGS-slowpow-fma4.c = -mfma4
CFLAGS-s_sin-fma4.c = -mfma4
CFLAGS-s_tan-fma4.c = -mfma4
libm-sysdep_routines += e_exp-avx e_log-avx s_atan-avx \
e_atan2-avx s_sin-avx s_tan-avx \
mplog-avx mpa-avx slowexp-avx \
mpexp-avx
CFLAGS-e_atan2-avx.c = -msse2avx -DSSE2AVX
CFLAGS-e_exp-avx.c = -msse2avx -DSSE2AVX
CFLAGS-e_log-avx.c = -msse2avx -DSSE2AVX
CFLAGS-mpa-avx.c = -msse2avx -DSSE2AVX
CFLAGS-mpexp-avx.c = -msse2avx -DSSE2AVX
CFLAGS-mplog-avx.c = -msse2avx -DSSE2AVX
CFLAGS-s_atan-avx.c = -msse2avx -DSSE2AVX
CFLAGS-s_sin-avx.c = -msse2avx -DSSE2AVX
CFLAGS-slowexp-avx.c = -msse2avx -DSSE2AVX
CFLAGS-s_tan-avx.c = -msse2avx -DSSE2AVX
endif
ifeq ($(subdir),mathvec)
libmvec-sysdep_routines += svml_d_cos2_core_sse4 svml_d_cos4_core_avx2 \
svml_d_cos8_core_avx512 svml_d_sin2_core_sse4 \
svml_d_sin4_core_avx2 svml_d_sin8_core_avx512 \
svml_d_log2_core_sse4 svml_d_log4_core_avx2 \
svml_d_log8_core_avx512 svml_d_sincos2_core_sse4 \
svml_d_sincos4_core_avx2 svml_d_sincos8_core_avx512 \
svml_s_cosf4_core_sse4 svml_s_cosf8_core_avx2 \
svml_s_cosf16_core_avx512 svml_s_sinf4_core_sse4 \
svml_s_sinf8_core_avx2 svml_s_sinf16_core_avx512 \
svml_s_logf4_core_sse4 svml_s_logf8_core_avx2 \
svml_s_logf16_core_avx512 svml_d_exp2_core_sse4 \
svml_d_exp4_core_avx2 svml_d_exp8_core_avx512 \
svml_s_expf4_core_sse4 svml_s_expf8_core_avx2 \
svml_s_expf16_core_avx512 svml_d_pow2_core_sse4 \
svml_d_pow4_core_avx2 svml_d_pow8_core_avx512 \
svml_s_powf4_core_sse4 svml_s_powf8_core_avx2 \
svml_s_powf16_core_avx512 svml_s_sincosf4_core_sse4 \
svml_s_sincosf8_core_avx2 \
svml_s_sincosf16_core_avx512 \
svml_d_cos2_core-sse2 svml_d_cos4_core-sse \
svml_d_cos8_core-avx2 svml_d_exp2_core-sse2 \
svml_d_exp4_core-sse svml_d_exp8_core-avx2 \
svml_d_log2_core-sse2 svml_d_log4_core-sse \
svml_d_log8_core-avx2 svml_d_pow2_core-sse2 \
svml_d_pow4_core-sse svml_d_pow8_core-avx2 \
svml_d_sin2_core-sse2 svml_d_sin4_core-sse \
svml_d_sin8_core-avx2 \
svml_d_sincos2_core-sse2 \
svml_d_sincos4_core-sse \
svml_d_sincos8_core-avx2 \
svml_s_cosf16_core-avx2 \
svml_s_cosf4_core-sse2 \
svml_s_cosf8_core-sse \
svml_s_expf16_core-avx2 \
svml_s_expf4_core-sse2 \
svml_s_expf8_core-sse \
svml_s_logf16_core-avx2 \
svml_s_logf4_core-sse2 \
svml_s_logf8_core-sse \
svml_s_powf16_core-avx2 \
svml_s_powf4_core-sse2 \
svml_s_powf8_core-sse \
svml_s_sincosf16_core-avx2 \
svml_s_sincosf4_core-sse2 \
svml_s_sincosf8_core-sse \
svml_s_sinf16_core-avx2 \
svml_s_sinf4_core-sse2 \
svml_s_sinf8_core-sse
endif