diff options
| author | Sunil K Pandey <skpgkp2@gmail.com> | 2021-12-29 08:53:16 -0800 |
|---|---|---|
| committer | Sunil K Pandey <skpgkp2@gmail.com> | 2021-12-29 11:37:42 -0800 |
| commit | ef7ea9c1327be11c54462cb539ea8854979a21f3 (patch) | |
| tree | ddf578d007fe3fe3d15e42f63f82cc432123c47e | |
| parent | 8b726453d560eef536c9ff730f5f5be05b4ad6a7 (diff) | |
| download | glibc-ef7ea9c1327be11c54462cb539ea8854979a21f3.tar glibc-ef7ea9c1327be11c54462cb539ea8854979a21f3.tar.gz glibc-ef7ea9c1327be11c54462cb539ea8854979a21f3.tar.bz2 glibc-ef7ea9c1327be11c54462cb539ea8854979a21f3.zip | |
x86-64: Add vector cosh/coshf implementation to libmvec
Implement vectorized cosh/coshf containing SSE, AVX, AVX2 and
AVX512 versions for libmvec as per vector ABI. It also contains
accuracy and ABI tests for vector cosh/coshf with regenerated ulps.
Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
50 files changed, 2637 insertions, 1 deletions
diff --git a/bits/libm-simd-decl-stubs.h b/bits/libm-simd-decl-stubs.h index bc18621f17..35c6ac57a8 100644 --- a/bits/libm-simd-decl-stubs.h +++ b/bits/libm-simd-decl-stubs.h @@ -164,4 +164,15 @@ #define __DECL_SIMD_exp10f32x #define __DECL_SIMD_exp10f64x #define __DECL_SIMD_exp10f128x + +#define __DECL_SIMD_cosh +#define __DECL_SIMD_coshf +#define __DECL_SIMD_coshl +#define __DECL_SIMD_coshf16 +#define __DECL_SIMD_coshf32 +#define __DECL_SIMD_coshf64 +#define __DECL_SIMD_coshf128 +#define __DECL_SIMD_coshf32x +#define __DECL_SIMD_coshf64x +#define __DECL_SIMD_coshf128x #endif diff --git a/math/bits/mathcalls.h b/math/bits/mathcalls.h index 870778457f..60a314f69e 100644 --- a/math/bits/mathcalls.h +++ b/math/bits/mathcalls.h @@ -68,7 +68,7 @@ __MATHCALL (tan,, (_Mdouble_ __x)); /* Hyperbolic functions. */ /* Hyperbolic cosine of X. */ -__MATHCALL (cosh,, (_Mdouble_ __x)); +__MATHCALL_VEC (cosh,, (_Mdouble_ __x)); /* Hyperbolic sine of X. */ __MATHCALL (sinh,, (_Mdouble_ __x)); /* Hyperbolic tangent of X. */ diff --git a/sysdeps/unix/sysv/linux/x86_64/libmvec.abilist b/sysdeps/unix/sysv/linux/x86_64/libmvec.abilist index b3c1f59593..4907680143 100644 --- a/sysdeps/unix/sysv/linux/x86_64/libmvec.abilist +++ b/sysdeps/unix/sysv/linux/x86_64/libmvec.abilist @@ -49,48 +49,56 @@ GLIBC_2.22 _ZGVeN8vvv_sincos F GLIBC_2.35 _ZGVbN2v_acos F GLIBC_2.35 _ZGVbN2v_asin F GLIBC_2.35 _ZGVbN2v_atan F +GLIBC_2.35 _ZGVbN2v_cosh F GLIBC_2.35 _ZGVbN2v_exp10 F GLIBC_2.35 _ZGVbN2v_exp2 F GLIBC_2.35 _ZGVbN2vv_hypot F GLIBC_2.35 _ZGVbN4v_acosf F GLIBC_2.35 _ZGVbN4v_asinf F GLIBC_2.35 _ZGVbN4v_atanf F +GLIBC_2.35 _ZGVbN4v_coshf F GLIBC_2.35 _ZGVbN4v_exp10f F GLIBC_2.35 _ZGVbN4v_exp2f F GLIBC_2.35 _ZGVbN4vv_hypotf F GLIBC_2.35 _ZGVcN4v_acos F GLIBC_2.35 _ZGVcN4v_asin F GLIBC_2.35 _ZGVcN4v_atan F +GLIBC_2.35 _ZGVcN4v_cosh F GLIBC_2.35 _ZGVcN4v_exp10 F GLIBC_2.35 _ZGVcN4v_exp2 F GLIBC_2.35 _ZGVcN4vv_hypot F GLIBC_2.35 _ZGVcN8v_acosf F GLIBC_2.35 _ZGVcN8v_asinf F GLIBC_2.35 _ZGVcN8v_atanf F +GLIBC_2.35 _ZGVcN8v_coshf F GLIBC_2.35 _ZGVcN8v_exp10f F GLIBC_2.35 _ZGVcN8v_exp2f F GLIBC_2.35 _ZGVcN8vv_hypotf F GLIBC_2.35 _ZGVdN4v_acos F GLIBC_2.35 _ZGVdN4v_asin F GLIBC_2.35 _ZGVdN4v_atan F +GLIBC_2.35 _ZGVdN4v_cosh F GLIBC_2.35 _ZGVdN4v_exp10 F GLIBC_2.35 _ZGVdN4v_exp2 F GLIBC_2.35 _ZGVdN4vv_hypot F GLIBC_2.35 _ZGVdN8v_acosf F GLIBC_2.35 _ZGVdN8v_asinf F GLIBC_2.35 _ZGVdN8v_atanf F +GLIBC_2.35 _ZGVdN8v_coshf F GLIBC_2.35 _ZGVdN8v_exp10f F GLIBC_2.35 _ZGVdN8v_exp2f F GLIBC_2.35 _ZGVdN8vv_hypotf F GLIBC_2.35 _ZGVeN16v_acosf F GLIBC_2.35 _ZGVeN16v_asinf F GLIBC_2.35 _ZGVeN16v_atanf F +GLIBC_2.35 _ZGVeN16v_coshf F GLIBC_2.35 _ZGVeN16v_exp10f F GLIBC_2.35 _ZGVeN16v_exp2f F GLIBC_2.35 _ZGVeN16vv_hypotf F GLIBC_2.35 _ZGVeN8v_acos F GLIBC_2.35 _ZGVeN8v_asin F GLIBC_2.35 _ZGVeN8v_atan F +GLIBC_2.35 _ZGVeN8v_cosh F GLIBC_2.35 _ZGVeN8v_exp10 F GLIBC_2.35 _ZGVeN8v_exp2 F GLIBC_2.35 _ZGVeN8vv_hypot F diff --git a/sysdeps/x86/fpu/bits/math-vector.h b/sysdeps/x86/fpu/bits/math-vector.h index f3f9c2e092..708e81b3d0 100644 --- a/sysdeps/x86/fpu/bits/math-vector.h +++ b/sysdeps/x86/fpu/bits/math-vector.h @@ -82,6 +82,10 @@ # define __DECL_SIMD_exp10 __DECL_SIMD_x86_64 # undef __DECL_SIMD_exp10f # define __DECL_SIMD_exp10f __DECL_SIMD_x86_64 +# undef __DECL_SIMD_cosh +# define __DECL_SIMD_cosh __DECL_SIMD_x86_64 +# undef __DECL_SIMD_coshf +# define __DECL_SIMD_coshf __DECL_SIMD_x86_64 # endif #endif diff --git a/sysdeps/x86/fpu/finclude/math-vector-fortran.h b/sysdeps/x86/fpu/finclude/math-vector-fortran.h index c033abbedc..81d0238ebf 100644 --- a/sysdeps/x86/fpu/finclude/math-vector-fortran.h +++ b/sysdeps/x86/fpu/finclude/math-vector-fortran.h @@ -40,6 +40,8 @@ !GCC$ builtin (exp2f) attributes simd (notinbranch) if('x86_64') !GCC$ builtin (exp10) attributes simd (notinbranch) if('x86_64') !GCC$ builtin (exp10f) attributes simd (notinbranch) if('x86_64') +!GCC$ builtin (cosh) attributes simd (notinbranch) if('x86_64') +!GCC$ builtin (coshf) attributes simd (notinbranch) if('x86_64') !GCC$ builtin (cos) attributes simd (notinbranch) if('x32') !GCC$ builtin (cosf) attributes simd (notinbranch) if('x32') @@ -65,3 +67,5 @@ !GCC$ builtin (exp2f) attributes simd (notinbranch) if('x32') !GCC$ builtin (exp10) attributes simd (notinbranch) if('x32') !GCC$ builtin (exp10f) attributes simd (notinbranch) if('x32') +!GCC$ builtin (cosh) attributes simd (notinbranch) if('x32') +!GCC$ builtin (coshf) attributes simd (notinbranch) if('x32') diff --git a/sysdeps/x86_64/fpu/Makeconfig b/sysdeps/x86_64/fpu/Makeconfig index fd0a9da439..5bc2df134f 100644 --- a/sysdeps/x86_64/fpu/Makeconfig +++ b/sysdeps/x86_64/fpu/Makeconfig @@ -26,6 +26,7 @@ libmvec-funcs = \ asin \ atan \ cos \ + cosh \ exp \ exp10 \ exp2 \ diff --git a/sysdeps/x86_64/fpu/Versions b/sysdeps/x86_64/fpu/Versions index f29cfa4cbf..53346d16a2 100644 --- a/sysdeps/x86_64/fpu/Versions +++ b/sysdeps/x86_64/fpu/Versions @@ -17,12 +17,14 @@ libmvec { _ZGVbN2v_acos; _ZGVcN4v_acos; _ZGVdN4v_acos; _ZGVeN8v_acos; _ZGVbN2v_asin; _ZGVcN4v_asin; _ZGVdN4v_asin; _ZGVeN8v_asin; _ZGVbN2v_atan; _ZGVcN4v_atan; _ZGVdN4v_atan; _ZGVeN8v_atan; + _ZGVbN2v_cosh; _ZGVcN4v_cosh; _ZGVdN4v_cosh; _ZGVeN8v_cosh; _ZGVbN2v_exp10; _ZGVcN4v_exp10; _ZGVdN4v_exp10; _ZGVeN8v_exp10; _ZGVbN2v_exp2; _ZGVcN4v_exp2; _ZGVdN4v_exp2; _ZGVeN8v_exp2; _ZGVbN2vv_hypot; _ZGVcN4vv_hypot; _ZGVdN4vv_hypot; _ZGVeN8vv_hypot; _ZGVbN4v_acosf; _ZGVcN8v_acosf; _ZGVdN8v_acosf; _ZGVeN16v_acosf; _ZGVbN4v_asinf; _ZGVcN8v_asinf; _ZGVdN8v_asinf; _ZGVeN16v_asinf; _ZGVbN4v_atanf; _ZGVcN8v_atanf; _ZGVdN8v_atanf; _ZGVeN16v_atanf; + _ZGVbN4v_coshf; _ZGVcN8v_coshf; _ZGVdN8v_coshf; _ZGVeN16v_coshf; _ZGVbN4v_exp10f; _ZGVcN8v_exp10f; _ZGVdN8v_exp10f; _ZGVeN16v_exp10f; _ZGVbN4v_exp2f; _ZGVcN8v_exp2f; _ZGVdN8v_exp2f; _ZGVeN16v_exp2f; _ZGVbN4vv_hypotf; _ZGVcN8vv_hypotf; _ZGVdN8vv_hypotf; _ZGVeN16vv_hypotf; diff --git a/sysdeps/x86_64/fpu/libm-test-ulps b/sysdeps/x86_64/fpu/libm-test-ulps index 45f2e4bb53..ac70f15208 100644 --- a/sysdeps/x86_64/fpu/libm-test-ulps +++ b/sysdeps/x86_64/fpu/libm-test-ulps @@ -891,6 +891,26 @@ float: 2 float128: 3 ldouble: 3 +Function: "cosh_vlen16": +float: 2 + +Function: "cosh_vlen2": +double: 2 + +Function: "cosh_vlen4": +double: 2 +float: 2 + +Function: "cosh_vlen4_avx2": +double: 2 + +Function: "cosh_vlen8": +double: 2 +float: 2 + +Function: "cosh_vlen8_avx2": +float: 2 + Function: Real part of "cpow": double: 2 float: 5 diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core-sse2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core-sse2.S new file mode 100644 index 0000000000..bfe4e3d0f0 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core-sse2.S @@ -0,0 +1,20 @@ +/* SSE2 version of vectorized cosh, vector length is 2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVbN2v_cosh _ZGVbN2v_cosh_sse2 +#include "../svml_d_cosh2_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core.c b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core.c new file mode 100644 index 0000000000..99561fea47 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core.c @@ -0,0 +1,27 @@ +/* Multiple versions of vectorized cosh, vector length is 2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVbN2v_cosh +#include "ifunc-mathvec-sse4_1.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVbN2v_cosh, __GI__ZGVbN2v_cosh, __redirect__ZGVbN2v_cosh) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core_sse4.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core_sse4.S new file mode 100644 index 0000000000..150bfae7e1 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh2_core_sse4.S @@ -0,0 +1,396 @@ +/* Function cosh vectorized with SSE4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_dcosh_data_internal + */ +#define _dbT 0 +#define _dbInvLn2 2064 +#define _dbLn2hi 2080 +#define _dbLn2lo 2096 +#define _dbShifter 2112 +#define _iIndexMask 2128 +#define _dPC2 2144 +#define _dPC3 2160 +#define _dPC4 2176 +#define _iMaxIndex 2192 +#define _lExpMask 2208 +#define _dSign 2224 +#define _iDomainRange 2240 + +#include <sysdep.h> + + .text + .section .text.sse4,"ax",@progbits +ENTRY(_ZGVbN2v_cosh_sse4) + subq $72, %rsp + cfi_def_cfa_offset(80) + movaps %xmm0, %xmm4 + movups _dSign+__svml_dcosh_data_internal(%rip), %xmm2 + lea _dbT+__svml_dcosh_data_internal(%rip), %r8 + +/* Abs argument */ + movaps %xmm2, %xmm5 + +/* dXSign=0x001000000000 */ + psrlq $11, %xmm2 + +/* + * Load argument + * dM = x*2^K/log(2) + RShifter + */ + movups _dbInvLn2+__svml_dcosh_data_internal(%rip), %xmm3 + andnps %xmm4, %xmm5 + mulpd %xmm5, %xmm3 + movups _dbShifter+__svml_dcosh_data_internal(%rip), %xmm1 + addpd %xmm1, %xmm3 + +/* + * R + * dN = dM - RShifter + */ + movaps %xmm3, %xmm15 + subpd %xmm1, %xmm15 + +/* dR = dX - dN*Log2_hi/2^K */ + movups _dbLn2hi+__svml_dcosh_data_internal(%rip), %xmm14 + mulpd %xmm15, %xmm14 + +/* dR = (dX - dN*Log2_hi/2^K) - dN*Log2_lo/2^K */ + movups _dbLn2lo+__svml_dcosh_data_internal(%rip), %xmm1 + mulpd %xmm15, %xmm1 + +/* + * Check for overflow\underflow + * + */ + pshufd $221, %xmm5, %xmm7 + subpd %xmm14, %xmm5 + movq _iIndexMask+__svml_dcosh_data_internal(%rip), %xmm8 + +/* Index and lookup */ + pshufd $136, %xmm3, %xmm9 + +/* + * G1,G2,G3: dTdif,dTn * 2^N,2^(-N) + * NB: copied from sinh_la - to be optimized!!!!! + */ + psllq $44, %xmm3 + +/* + * trick + * 256=-iIndex + */ + movq _iMaxIndex+__svml_dcosh_data_internal(%rip), %xmm12 + pand %xmm8, %xmm9 + subpd %xmm1, %xmm5 + psubd %xmm9, %xmm12 + +/* iIndex*=3 */ + movdqa %xmm9, %xmm10 + +/* iDomainRange*=3 */ + pslld $3, %xmm12 + pslld $3, %xmm10 + movd %xmm12, %esi + pshufd $1, %xmm12, %xmm13 + movq _iDomainRange+__svml_dcosh_data_internal(%rip), %xmm6 + movd %xmm13, %edi + pcmpgtd %xmm6, %xmm7 + movmskps %xmm7, %eax + +/* dR2 = dR^2 */ + movaps %xmm5, %xmm7 + +/* lM now is an EXP(2^N) */ + pand _lExpMask+__svml_dcosh_data_internal(%rip), %xmm3 + pshufd $1, %xmm10, %xmm11 + movslq %esi, %rsi + mulpd %xmm5, %xmm7 + movd %xmm10, %edx + movsd (%r8,%rsi), %xmm6 + movd %xmm11, %ecx + movslq %edi, %rdi + movslq %edx, %rdx + movslq %ecx, %rcx + movhpd (%r8,%rdi), %xmm6 + +/* */ + psubq %xmm3, %xmm6 + +/* lX- = EXP(1/2) */ + psubq %xmm2, %xmm6 + +/* + * sinh(r) = r +r*r^2*a3 .... + * dSinh_r = r^2*a3 + */ + movups _dPC3+__svml_dcosh_data_internal(%rip), %xmm2 + mulpd %xmm7, %xmm2 + +/* dSinh_r = r + r*r^2*a3 */ + mulpd %xmm5, %xmm2 + movsd (%r8,%rdx), %xmm0 + movhpd (%r8,%rcx), %xmm0 + paddq %xmm3, %xmm0 + addpd %xmm2, %xmm5 + +/* dTn = dTn*2^N - dTn*2^-N */ + movaps %xmm0, %xmm3 + subpd %xmm6, %xmm3 + +/* dTp = dTn*2^N + dTn*2^-N */ + addpd %xmm6, %xmm0 + mulpd %xmm5, %xmm3 + +/* poly(r) = dTp + dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + movups _dPC4+__svml_dcosh_data_internal(%rip), %xmm5 + mulpd %xmm7, %xmm5 + addpd _dPC2+__svml_dcosh_data_internal(%rip), %xmm5 + mulpd %xmm5, %xmm7 + +/* dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + mulpd %xmm0, %xmm7 + addpd %xmm7, %xmm3 + +/* _VRES1 = dTp + dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + addpd %xmm3, %xmm0 + andl $3, %eax + +/* Ret H */ + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx rbp r12 r13 r14 r15 eax xmm0 xmm4 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + addq $72, %rsp + cfi_def_cfa_offset(8) + ret + cfi_def_cfa_offset(80) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + movups %xmm4, 32(%rsp) + movups %xmm0, 48(%rsp) + # LOE rbx rbp r12 r13 r14 r15 eax xmm0 + + xorl %edx, %edx + movq %r12, 16(%rsp) + cfi_offset(12, -64) + movl %edx, %r12d + movq %r13, 8(%rsp) + cfi_offset(13, -72) + movl %eax, %r13d + movq %r14, (%rsp) + cfi_offset(14, -80) + # LOE rbx rbp r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx rbp r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $2, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx rbp r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + movups 48(%rsp), %xmm0 + +/* Go to exit */ + jmp L(EXIT) + cfi_offset(12, -64) + cfi_offset(13, -72) + cfi_offset(14, -80) + # LOE rbx rbp r12 r13 r14 r15 xmm0 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movsd 32(%rsp,%r14,8), %xmm0 + call cosh@PLT + # LOE rbx rbp r14 r15 r12d r13d xmm0 + + movsd %xmm0, 48(%rsp,%r14,8) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx rbp r15 r12d r13d +END(_ZGVbN2v_cosh_sse4) + + .section .rodata, "a" + .align 16 + +#ifdef __svml_dcosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(16)) VUINT32 _dbT[(1 + (1<<8))][2]; //dTpj ONLY! + __declspec(align(16)) VUINT32 _dbInvLn2[2][2]; + __declspec(align(16)) VUINT32 _dbLn2hi[2][2]; + __declspec(align(16)) VUINT32 _dbLn2lo[2][2]; + __declspec(align(16)) VUINT32 _dbShifter[2][2]; + __declspec(align(16)) VUINT32 _iIndexMask[4][1]; //(1<<K)1- + __declspec(align(16)) VUINT32 _dPC2[2][2]; + __declspec(align(16)) VUINT32 _dPC3[2][2]; + __declspec(align(16)) VUINT32 _dPC4[2][2]; + __declspec(align(16)) VUINT32 _iMaxIndex[4][1]; //(1<<K) + __declspec(align(16)) VUINT32 _lExpMask[2][2]; + __declspec(align(16)) VUINT32 _dSign[2][2]; //0x8000000000000000 + __declspec(align(16)) VUINT32 _iDomainRange[4][1]; +} __svml_dcosh_data_internal; +#endif +__svml_dcosh_data_internal: + /*== _dbT ==*/ + .quad 0x3fe0000000000000, 0x3fe00b1afa5abcbf, 0x3fe0163da9fb3335, 0x3fe02168143b0281 + .quad 0x3fe02c9a3e778061, 0x3fe037d42e11bbcc, 0x3fe04315e86e7f85, 0x3fe04e5f72f654b1 + .quad 0x3fe059b0d3158574, 0x3fe0650a0e3c1f89, 0x3fe0706b29ddf6de, 0x3fe07bd42b72a836 + .quad 0x3fe0874518759bc8, 0x3fe092bdf66607e0, 0x3fe09e3ecac6f383, 0x3fe0a9c79b1f3919 + .quad 0x3fe0b5586cf9890f, 0x3fe0c0f145e46c85, 0x3fe0cc922b7247f7, 0x3fe0d83b23395dec + .quad 0x3fe0e3ec32d3d1a2, 0x3fe0efa55fdfa9c5, 0x3fe0fb66affed31b, 0x3fe1073028d7233e + .quad 0x3fe11301d0125b51, 0x3fe11edbab5e2ab6, 0x3fe12abdc06c31cc, 0x3fe136a814f204ab + .quad 0x3fe1429aaea92de0, 0x3fe14e95934f312e, 0x3fe15a98c8a58e51, 0x3fe166a45471c3c2 + .quad 0x3fe172b83c7d517b, 0x3fe17ed48695bbc0, 0x3fe18af9388c8dea, 0x3fe1972658375d2f + .quad 0x3fe1a35beb6fcb75, 0x3fe1af99f8138a1c, 0x3fe1bbe084045cd4, 0x3fe1c82f95281c6b + .quad 0x3fe1d4873168b9aa, 0x3fe1e0e75eb44027, 0x3fe1ed5022fcd91d, 0x3fe1f9c18438ce4d + .quad 0x3fe2063b88628cd6, 0x3fe212be3578a819, 0x3fe21f49917ddc96, 0x3fe22bdda27912d1 + .quad 0x3fe2387a6e756238, 0x3fe2451ffb82140a, 0x3fe251ce4fb2a63f, 0x3fe25e85711ece75 + .quad 0x3fe26b4565e27cdd, 0x3fe2780e341ddf29, 0x3fe284dfe1f56381, 0x3fe291ba7591bb70 + .quad 0x3fe29e9df51fdee1, 0x3fe2ab8a66d10f13, 0x3fe2b87fd0dad990, 0x3fe2c57e39771b2f + .quad 0x3fe2d285a6e4030b, 0x3fe2df961f641589, 0x3fe2ecafa93e2f56, 0x3fe2f9d24abd886b + .quad 0x3fe306fe0a31b715, 0x3fe31432edeeb2fd, 0x3fe32170fc4cd831, 0x3fe32eb83ba8ea32 + .quad 0x3fe33c08b26416ff, 0x3fe3496266e3fa2d, 0x3fe356c55f929ff1, 0x3fe36431a2de883b + .quad 0x3fe371a7373aa9cb, 0x3fe37f26231e754a, 0x3fe38cae6d05d866, 0x3fe39a401b7140ef + .quad 0x3fe3a7db34e59ff7, 0x3fe3b57fbfec6cf4, 0x3fe3c32dc313a8e5, 0x3fe3d0e544ede173 + .quad 0x3fe3dea64c123422, 0x3fe3ec70df1c5175, 0x3fe3fa4504ac801c, 0x3fe40822c367a024 + .quad 0x3fe4160a21f72e2a, 0x3fe423fb2709468a, 0x3fe431f5d950a897, 0x3fe43ffa3f84b9d4 + .quad 0x3fe44e086061892d, 0x3fe45c2042a7d232, 0x3fe46a41ed1d0057, 0x3fe4786d668b3237 + .quad 0x3fe486a2b5c13cd0, 0x3fe494e1e192aed2, 0x3fe4a32af0d7d3de, 0x3fe4b17dea6db7d7 + .quad 0x3fe4bfdad5362a27, 0x3fe4ce41b817c114, 0x3fe4dcb299fddd0d, 0x3fe4eb2d81d8abff + .quad 0x3fe4f9b2769d2ca7, 0x3fe508417f4531ee, 0x3fe516daa2cf6642, 0x3fe5257de83f4eef + .quad 0x3fe5342b569d4f82, 0x3fe542e2f4f6ad27, 0x3fe551a4ca5d920f, 0x3fe56070dde910d2 + .quad 0x3fe56f4736b527da, 0x3fe57e27dbe2c4cf, 0x3fe58d12d497c7fd, 0x3fe59c0827ff07cc + .quad 0x3fe5ab07dd485429, 0x3fe5ba11fba87a03, 0x3fe5c9268a5946b7, 0x3fe5d84590998b93 + .quad 0x3fe5e76f15ad2148, 0x3fe5f6a320dceb71, 0x3fe605e1b976dc09, 0x3fe6152ae6cdf6f4 + .quad 0x3fe6247eb03a5585, 0x3fe633dd1d1929fd, 0x3fe6434634ccc320, 0x3fe652b9febc8fb7 + .quad 0x3fe6623882552225, 0x3fe671c1c70833f6, 0x3fe68155d44ca973, 0x3fe690f4b19e9538 + .quad 0x3fe6a09e667f3bcd, 0x3fe6b052fa75173e, 0x3fe6c012750bdabf, 0x3fe6cfdcddd47645 + .quad 0x3fe6dfb23c651a2f, 0x3fe6ef9298593ae5, 0x3fe6ff7df9519484, 0x3fe70f7466f42e87 + .quad 0x3fe71f75e8ec5f74, 0x3fe72f8286ead08a, 0x3fe73f9a48a58174, 0x3fe74fbd35d7cbfd + .quad 0x3fe75feb564267c9, 0x3fe77024b1ab6e09, 0x3fe780694fde5d3f, 0x3fe790b938ac1cf6 + .quad 0x3fe7a11473eb0187, 0x3fe7b17b0976cfdb, 0x3fe7c1ed0130c132, 0x3fe7d26a62ff86f0 + .quad 0x3fe7e2f336cf4e62, 0x3fe7f3878491c491, 0x3fe80427543e1a12, 0x3fe814d2add106d9 + .quad 0x3fe82589994cce13, 0x3fe8364c1eb941f7, 0x3fe8471a4623c7ad, 0x3fe857f4179f5b21 + .quad 0x3fe868d99b4492ed, 0x3fe879cad931a436, 0x3fe88ac7d98a6699, 0x3fe89bd0a478580f + .quad 0x3fe8ace5422aa0db, 0x3fe8be05bad61778, 0x3fe8cf3216b5448c, 0x3fe8e06a5e0866d9 + .quad 0x3fe8f1ae99157736, 0x3fe902fed0282c8a, 0x3fe9145b0b91ffc6, 0x3fe925c353aa2fe2 + .quad 0x3fe93737b0cdc5e5, 0x3fe948b82b5f98e5, 0x3fe95a44cbc8520f, 0x3fe96bdd9a7670b3 + .quad 0x3fe97d829fde4e50, 0x3fe98f33e47a22a2, 0x3fe9a0f170ca07ba, 0x3fe9b2bb4d53fe0d + .quad 0x3fe9c49182a3f090, 0x3fe9d674194bb8d5, 0x3fe9e86319e32323, 0x3fe9fa5e8d07f29e + .quad 0x3fea0c667b5de565, 0x3fea1e7aed8eb8bb, 0x3fea309bec4a2d33, 0x3fea42c980460ad8 + .quad 0x3fea5503b23e255d, 0x3fea674a8af46052, 0x3fea799e1330b358, 0x3fea8bfe53c12e59 + .quad 0x3fea9e6b5579fdbf, 0x3feab0e521356eba, 0x3feac36bbfd3f37a, 0x3fead5ff3a3c2774 + .quad 0x3feae89f995ad3ad, 0x3feafb4ce622f2ff, 0x3feb0e07298db666, 0x3feb20ce6c9a8952 + .quad 0x3feb33a2b84f15fb, 0x3feb468415b749b1, 0x3feb59728de5593a, 0x3feb6c6e29f1c52a + .quad 0x3feb7f76f2fb5e47, 0x3feb928cf22749e4, 0x3feba5b030a1064a, 0x3febb8e0b79a6f1f + .quad 0x3febcc1e904bc1d2, 0x3febdf69c3f3a207, 0x3febf2c25bd71e09, 0x3fec06286141b33d + .quad 0x3fec199bdd85529c, 0x3fec2d1cd9fa652c, 0x3fec40ab5fffd07a, 0x3fec544778fafb22 + .quad 0x3fec67f12e57d14b, 0x3fec7ba88988c933, 0x3fec8f6d9406e7b5, 0x3feca3405751c4db + .quad 0x3fecb720dcef9069, 0x3feccb0f2e6d1675, 0x3fecdf0b555dc3fa, 0x3fecf3155b5bab74 + .quad 0x3fed072d4a07897c, 0x3fed1b532b08c968, 0x3fed2f87080d89f2, 0x3fed43c8eacaa1d6 + .quad 0x3fed5818dcfba487, 0x3fed6c76e862e6d3, 0x3fed80e316c98398, 0x3fed955d71ff6075 + .quad 0x3feda9e603db3285, 0x3fedbe7cd63a8315, 0x3fedd321f301b460, 0x3fede7d5641c0658 + .quad 0x3fedfc97337b9b5f, 0x3fee11676b197d17, 0x3fee264614f5a129, 0x3fee3b333b16ee12 + .quad 0x3fee502ee78b3ff6, 0x3fee653924676d76, 0x3fee7a51fbc74c83, 0x3fee8f7977cdb740 + .quad 0x3feea4afa2a490da, 0x3feeb9f4867cca6e, 0x3feecf482d8e67f1, 0x3feee4aaa2188510 + .quad 0x3feefa1bee615a27, 0x3fef0f9c1cb6412a, 0x3fef252b376bba97, 0x3fef3ac948dd7274 + .quad 0x3fef50765b6e4540, 0x3fef6632798844f8, 0x3fef7bfdad9cbe14, 0x3fef91d802243c89 + .quad 0x3fefa7c1819e90d8, 0x3fefbdba3692d514, 0x3fefd3c22b8f71f1, 0x3fefe9d96b2a23d9 + .quad 0x3ff0000000000000 + .align 16 + .quad 0x3ff71547652b82fe, 0x3ff71547652b82fe /* _dbInvLn2 = 1/log(2) */ + .align 16 + .quad 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000 /* _dbLn2hi = log(2) hi*/ + .align 16 + .quad 0xBDAC610CA86C3899, 0xBDAC610CA86C3899 /* _dbLn2lo = log(2) lo*/ + .align 16 + .quad 0x42B8000000000000, 0x42B8000000000000 /* _dbShifter */ + .align 16 + .long 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF /* _iIndexMask */ + .align 16 + .quad 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD /* _dPC2 */ + .align 16 + .quad 0x3FC5555570813E14, 0x3FC5555570813E14 /* _dPC3 */ + .align 16 + .quad 0x3FA55555CF16D299, 0x3FA55555CF16D299 /* _dPC4 */ + .align 16 + .long 0x00000100, 0x00000100, 0x00000100, 0x00000100 /* _iMaxIndex */ + .align 16 + .quad 0x7ff0000000000000, 0x7ff0000000000000 /* _lExpMask */ + .align 16 + .quad 0x8000000000000000, 0x8000000000000000 /* _dSign*/ + .align 16 + .long 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99 /* _iDomainRange 0x40861d9ac12a3e85 =(1021*2^K-0.5)*log(2)/2^K -needed for quick exp*/ + .align 16 + .type __svml_dcosh_data_internal,@object + .size __svml_dcosh_data_internal,.-__svml_dcosh_data_internal diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core-sse.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core-sse.S new file mode 100644 index 0000000000..4410d34583 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core-sse.S @@ -0,0 +1,20 @@ +/* SSE version of vectorized cosh, vector length is 4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVdN4v_cosh _ZGVdN4v_cosh_sse_wrapper +#include "../svml_d_cosh4_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core.c b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core.c new file mode 100644 index 0000000000..c4f59206a9 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core.c @@ -0,0 +1,27 @@ +/* Multiple versions of vectorized cosh, vector length is 4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVdN4v_cosh +#include "ifunc-mathvec-avx2.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVdN4v_cosh, __GI__ZGVdN4v_cosh, __redirect__ZGVdN4v_cosh) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S new file mode 100644 index 0000000000..2d86a02923 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh4_core_avx2.S @@ -0,0 +1,412 @@ +/* Function cosh vectorized with AVX2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_dcosh_data_internal + */ +#define _dbT 0 +#define _dbInvLn2 2080 +#define _dbLn2hi 2112 +#define _dbLn2lo 2144 +#define _dbShifter 2176 +#define _iIndexMask 2208 +#define _dPC2 2240 +#define _dPC3 2272 +#define _dPC4 2304 +#define _iMaxIndex 2336 +#define _lExpMask 2368 +#define _dSign 2400 +#define _iDomainRange 2432 + +#include <sysdep.h> + + .text + .section .text.avx2,"ax",@progbits +ENTRY(_ZGVdN4v_cosh_avx2) + pushq %rbp + cfi_def_cfa_offset(16) + movq %rsp, %rbp + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + andq $-32, %rsp + subq $96, %rsp + lea _dbT+__svml_dcosh_data_internal(%rip), %rax + vmovupd _dSign+__svml_dcosh_data_internal(%rip), %ymm8 + vmovupd _dbShifter+__svml_dcosh_data_internal(%rip), %ymm6 + +/* + * Load argument + * dM = x*2^K/log(2) + RShifter + */ + vmovupd _dbInvLn2+__svml_dcosh_data_internal(%rip), %ymm3 + +/* + * trick + * 256=-iIndex + */ + vmovups _iMaxIndex+__svml_dcosh_data_internal(%rip), %xmm14 + +/* dXSign=0x001000000000 */ + vpsrlq $11, %ymm8, %ymm5 + vmovapd %ymm0, %ymm7 + +/* Abs argument */ + vandnpd %ymm7, %ymm8, %ymm4 + vfmadd213pd %ymm6, %ymm4, %ymm3 + +/* Index and lookup */ + vextractf128 $1, %ymm3, %xmm12 + vshufps $136, %xmm12, %xmm3, %xmm13 + vpand _iIndexMask+__svml_dcosh_data_internal(%rip), %xmm13, %xmm15 + vpsubd %xmm15, %xmm14, %xmm0 + +/* iDomainRange*=3 */ + vpslld $3, %xmm0, %xmm2 + vmovd %xmm2, %r9d + vpextrd $2, %xmm2, %r11d + movslq %r9d, %r9 + vpextrd $1, %xmm2, %r10d + movslq %r11d, %r11 + movslq %r10d, %r10 + vmovsd (%rax,%r9), %xmm12 + +/* + * Check for overflow\underflow + * + */ + vextractf128 $1, %ymm4, %xmm9 + vmovsd (%rax,%r11), %xmm14 + vmovhpd (%rax,%r10), %xmm12, %xmm13 + vshufps $221, %xmm9, %xmm4, %xmm10 + +/* iIndex*=3 */ + vpslld $3, %xmm15, %xmm9 + +/* + * R + * dN = dM - RShifter + */ + vsubpd %ymm6, %ymm3, %ymm15 + vmovd %xmm9, %ecx + vpcmpgtd _iDomainRange+__svml_dcosh_data_internal(%rip), %xmm10, %xmm11 + vmovupd _dbLn2hi+__svml_dcosh_data_internal(%rip), %ymm6 + +/* + * G1,G2,G3: dTdif,dTn * 2^N,2^(-N) + * NB: copied from sinh_la - to be optimized!!!!! + */ + vpsllq $44, %ymm3, %ymm3 + vmovmskps %xmm11, %edx + +/* dR = dX - dN*Log2_hi/2^K */ + vfnmadd231pd %ymm6, %ymm15, %ymm4 + +/* lM now is an EXP(2^N) */ + vpand _lExpMask+__svml_dcosh_data_internal(%rip), %ymm3, %ymm3 + +/* dR = (dX - dN*Log2_hi/2^K) - dN*Log2_lo/2^K */ + vfnmadd231pd _dbLn2lo+__svml_dcosh_data_internal(%rip), %ymm15, %ymm4 + movslq %ecx, %rcx + vpextrd $2, %xmm9, %edi + vpextrd $1, %xmm9, %esi + movslq %edi, %rdi + vmovsd (%rax,%rcx), %xmm1 + vpextrd $3, %xmm9, %r8d + vpextrd $3, %xmm2, %ecx + movslq %esi, %rsi + movslq %r8d, %r8 + movslq %ecx, %rcx + +/* dR2 = dR^2 */ + vmulpd %ymm4, %ymm4, %ymm0 + vmovsd (%rax,%rdi), %xmm10 + vmovhpd (%rax,%rsi), %xmm1, %xmm8 + vmovhpd (%rax,%r8), %xmm10, %xmm11 + vmovhpd (%rax,%rcx), %xmm14, %xmm2 + vinsertf128 $1, %xmm11, %ymm8, %ymm1 + vinsertf128 $1, %xmm2, %ymm13, %ymm2 + vpaddq %ymm3, %ymm1, %ymm6 + +/* */ + vpsubq %ymm3, %ymm2, %ymm1 + +/* + * sinh(r) = r +r*r^2*a3 .... + * dSinh_r = r^2*a3 + */ + vmulpd _dPC3+__svml_dcosh_data_internal(%rip), %ymm0, %ymm2 + +/* lX- = EXP(1/2) */ + vpsubq %ymm5, %ymm1, %ymm5 + +/* dSinh_r = r + r*r^2*a3 */ + vfmadd213pd %ymm4, %ymm4, %ymm2 + +/* poly(r) = dTp + dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + vmovupd _dPC4+__svml_dcosh_data_internal(%rip), %ymm4 + +/* dTn = dTn*2^N - dTn*2^-N */ + vsubpd %ymm5, %ymm6, %ymm1 + +/* dTp = dTn*2^N + dTn*2^-N */ + vaddpd %ymm5, %ymm6, %ymm3 + vfmadd213pd _dPC2+__svml_dcosh_data_internal(%rip), %ymm0, %ymm4 + vmulpd %ymm2, %ymm1, %ymm1 + vmulpd %ymm4, %ymm0, %ymm0 + +/* dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + vfmadd213pd %ymm1, %ymm3, %ymm0 + +/* _VRES1 = dTp + dTn*sinh(dR)+dTp*dR2*(a2 +a4*dR2) */ + vaddpd %ymm0, %ymm3, %ymm0 + +/* Ret H */ + testl %edx, %edx + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx r12 r13 r14 r15 edx ymm0 ymm7 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + movq %rbp, %rsp + popq %rbp + cfi_def_cfa(7, 8) + cfi_restore(6) + ret + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + vmovupd %ymm7, 32(%rsp) + vmovupd %ymm0, 64(%rsp) + # LOE rbx r12 r13 r14 r15 edx ymm0 + + xorl %eax, %eax + # LOE rbx r12 r13 r14 r15 eax edx + + vzeroupper + movq %r12, 16(%rsp) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22 + movl %eax, %r12d + movq %r13, 8(%rsp) + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22 + movl %edx, %r13d + movq %r14, (%rsp) + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $4, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + vmovupd 64(%rsp), %ymm0 + +/* Go to exit */ + jmp L(EXIT) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r12 r13 r14 r15 ymm0 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movsd 32(%rsp,%r14,8), %xmm0 + call cosh@PLT + # LOE rbx r14 r15 r12d r13d xmm0 + + movsd %xmm0, 64(%rsp,%r14,8) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx r15 r12d r13d +END(_ZGVdN4v_cosh_avx2) + + .section .rodata, "a" + .align 32 + +#ifdef __svml_dcosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(32)) VUINT32 _dbT[(1 + (1<<8))][2]; //dTpj ONLY! + __declspec(align(32)) VUINT32 _dbInvLn2[4][2]; + __declspec(align(32)) VUINT32 _dbLn2hi[4][2]; + __declspec(align(32)) VUINT32 _dbLn2lo[4][2]; + __declspec(align(32)) VUINT32 _dbShifter[4][2]; + __declspec(align(32)) VUINT32 _iIndexMask[8][1]; //(1<<K)1- + __declspec(align(32)) VUINT32 _dPC2[4][2]; + __declspec(align(32)) VUINT32 _dPC3[4][2]; + __declspec(align(32)) VUINT32 _dPC4[4][2]; + __declspec(align(32)) VUINT32 _iMaxIndex[8][1]; //(1<<K) + __declspec(align(32)) VUINT32 _lExpMask[4][2]; + __declspec(align(32)) VUINT32 _dSign[4][2]; //0x8000000000000000 + __declspec(align(32)) VUINT32 _iDomainRange[8][1]; +} __svml_dcosh_data_internal; +#endif +__svml_dcosh_data_internal: + /*== _dbT ==*/ + .quad 0x3fe0000000000000, 0x3fe00b1afa5abcbf, 0x3fe0163da9fb3335, 0x3fe02168143b0281 + .quad 0x3fe02c9a3e778061, 0x3fe037d42e11bbcc, 0x3fe04315e86e7f85, 0x3fe04e5f72f654b1 + .quad 0x3fe059b0d3158574, 0x3fe0650a0e3c1f89, 0x3fe0706b29ddf6de, 0x3fe07bd42b72a836 + .quad 0x3fe0874518759bc8, 0x3fe092bdf66607e0, 0x3fe09e3ecac6f383, 0x3fe0a9c79b1f3919 + .quad 0x3fe0b5586cf9890f, 0x3fe0c0f145e46c85, 0x3fe0cc922b7247f7, 0x3fe0d83b23395dec + .quad 0x3fe0e3ec32d3d1a2, 0x3fe0efa55fdfa9c5, 0x3fe0fb66affed31b, 0x3fe1073028d7233e + .quad 0x3fe11301d0125b51, 0x3fe11edbab5e2ab6, 0x3fe12abdc06c31cc, 0x3fe136a814f204ab + .quad 0x3fe1429aaea92de0, 0x3fe14e95934f312e, 0x3fe15a98c8a58e51, 0x3fe166a45471c3c2 + .quad 0x3fe172b83c7d517b, 0x3fe17ed48695bbc0, 0x3fe18af9388c8dea, 0x3fe1972658375d2f + .quad 0x3fe1a35beb6fcb75, 0x3fe1af99f8138a1c, 0x3fe1bbe084045cd4, 0x3fe1c82f95281c6b + .quad 0x3fe1d4873168b9aa, 0x3fe1e0e75eb44027, 0x3fe1ed5022fcd91d, 0x3fe1f9c18438ce4d + .quad 0x3fe2063b88628cd6, 0x3fe212be3578a819, 0x3fe21f49917ddc96, 0x3fe22bdda27912d1 + .quad 0x3fe2387a6e756238, 0x3fe2451ffb82140a, 0x3fe251ce4fb2a63f, 0x3fe25e85711ece75 + .quad 0x3fe26b4565e27cdd, 0x3fe2780e341ddf29, 0x3fe284dfe1f56381, 0x3fe291ba7591bb70 + .quad 0x3fe29e9df51fdee1, 0x3fe2ab8a66d10f13, 0x3fe2b87fd0dad990, 0x3fe2c57e39771b2f + .quad 0x3fe2d285a6e4030b, 0x3fe2df961f641589, 0x3fe2ecafa93e2f56, 0x3fe2f9d24abd886b + .quad 0x3fe306fe0a31b715, 0x3fe31432edeeb2fd, 0x3fe32170fc4cd831, 0x3fe32eb83ba8ea32 + .quad 0x3fe33c08b26416ff, 0x3fe3496266e3fa2d, 0x3fe356c55f929ff1, 0x3fe36431a2de883b + .quad 0x3fe371a7373aa9cb, 0x3fe37f26231e754a, 0x3fe38cae6d05d866, 0x3fe39a401b7140ef + .quad 0x3fe3a7db34e59ff7, 0x3fe3b57fbfec6cf4, 0x3fe3c32dc313a8e5, 0x3fe3d0e544ede173 + .quad 0x3fe3dea64c123422, 0x3fe3ec70df1c5175, 0x3fe3fa4504ac801c, 0x3fe40822c367a024 + .quad 0x3fe4160a21f72e2a, 0x3fe423fb2709468a, 0x3fe431f5d950a897, 0x3fe43ffa3f84b9d4 + .quad 0x3fe44e086061892d, 0x3fe45c2042a7d232, 0x3fe46a41ed1d0057, 0x3fe4786d668b3237 + .quad 0x3fe486a2b5c13cd0, 0x3fe494e1e192aed2, 0x3fe4a32af0d7d3de, 0x3fe4b17dea6db7d7 + .quad 0x3fe4bfdad5362a27, 0x3fe4ce41b817c114, 0x3fe4dcb299fddd0d, 0x3fe4eb2d81d8abff + .quad 0x3fe4f9b2769d2ca7, 0x3fe508417f4531ee, 0x3fe516daa2cf6642, 0x3fe5257de83f4eef + .quad 0x3fe5342b569d4f82, 0x3fe542e2f4f6ad27, 0x3fe551a4ca5d920f, 0x3fe56070dde910d2 + .quad 0x3fe56f4736b527da, 0x3fe57e27dbe2c4cf, 0x3fe58d12d497c7fd, 0x3fe59c0827ff07cc + .quad 0x3fe5ab07dd485429, 0x3fe5ba11fba87a03, 0x3fe5c9268a5946b7, 0x3fe5d84590998b93 + .quad 0x3fe5e76f15ad2148, 0x3fe5f6a320dceb71, 0x3fe605e1b976dc09, 0x3fe6152ae6cdf6f4 + .quad 0x3fe6247eb03a5585, 0x3fe633dd1d1929fd, 0x3fe6434634ccc320, 0x3fe652b9febc8fb7 + .quad 0x3fe6623882552225, 0x3fe671c1c70833f6, 0x3fe68155d44ca973, 0x3fe690f4b19e9538 + .quad 0x3fe6a09e667f3bcd, 0x3fe6b052fa75173e, 0x3fe6c012750bdabf, 0x3fe6cfdcddd47645 + .quad 0x3fe6dfb23c651a2f, 0x3fe6ef9298593ae5, 0x3fe6ff7df9519484, 0x3fe70f7466f42e87 + .quad 0x3fe71f75e8ec5f74, 0x3fe72f8286ead08a, 0x3fe73f9a48a58174, 0x3fe74fbd35d7cbfd + .quad 0x3fe75feb564267c9, 0x3fe77024b1ab6e09, 0x3fe780694fde5d3f, 0x3fe790b938ac1cf6 + .quad 0x3fe7a11473eb0187, 0x3fe7b17b0976cfdb, 0x3fe7c1ed0130c132, 0x3fe7d26a62ff86f0 + .quad 0x3fe7e2f336cf4e62, 0x3fe7f3878491c491, 0x3fe80427543e1a12, 0x3fe814d2add106d9 + .quad 0x3fe82589994cce13, 0x3fe8364c1eb941f7, 0x3fe8471a4623c7ad, 0x3fe857f4179f5b21 + .quad 0x3fe868d99b4492ed, 0x3fe879cad931a436, 0x3fe88ac7d98a6699, 0x3fe89bd0a478580f + .quad 0x3fe8ace5422aa0db, 0x3fe8be05bad61778, 0x3fe8cf3216b5448c, 0x3fe8e06a5e0866d9 + .quad 0x3fe8f1ae99157736, 0x3fe902fed0282c8a, 0x3fe9145b0b91ffc6, 0x3fe925c353aa2fe2 + .quad 0x3fe93737b0cdc5e5, 0x3fe948b82b5f98e5, 0x3fe95a44cbc8520f, 0x3fe96bdd9a7670b3 + .quad 0x3fe97d829fde4e50, 0x3fe98f33e47a22a2, 0x3fe9a0f170ca07ba, 0x3fe9b2bb4d53fe0d + .quad 0x3fe9c49182a3f090, 0x3fe9d674194bb8d5, 0x3fe9e86319e32323, 0x3fe9fa5e8d07f29e + .quad 0x3fea0c667b5de565, 0x3fea1e7aed8eb8bb, 0x3fea309bec4a2d33, 0x3fea42c980460ad8 + .quad 0x3fea5503b23e255d, 0x3fea674a8af46052, 0x3fea799e1330b358, 0x3fea8bfe53c12e59 + .quad 0x3fea9e6b5579fdbf, 0x3feab0e521356eba, 0x3feac36bbfd3f37a, 0x3fead5ff3a3c2774 + .quad 0x3feae89f995ad3ad, 0x3feafb4ce622f2ff, 0x3feb0e07298db666, 0x3feb20ce6c9a8952 + .quad 0x3feb33a2b84f15fb, 0x3feb468415b749b1, 0x3feb59728de5593a, 0x3feb6c6e29f1c52a + .quad 0x3feb7f76f2fb5e47, 0x3feb928cf22749e4, 0x3feba5b030a1064a, 0x3febb8e0b79a6f1f + .quad 0x3febcc1e904bc1d2, 0x3febdf69c3f3a207, 0x3febf2c25bd71e09, 0x3fec06286141b33d + .quad 0x3fec199bdd85529c, 0x3fec2d1cd9fa652c, 0x3fec40ab5fffd07a, 0x3fec544778fafb22 + .quad 0x3fec67f12e57d14b, 0x3fec7ba88988c933, 0x3fec8f6d9406e7b5, 0x3feca3405751c4db + .quad 0x3fecb720dcef9069, 0x3feccb0f2e6d1675, 0x3fecdf0b555dc3fa, 0x3fecf3155b5bab74 + .quad 0x3fed072d4a07897c, 0x3fed1b532b08c968, 0x3fed2f87080d89f2, 0x3fed43c8eacaa1d6 + .quad 0x3fed5818dcfba487, 0x3fed6c76e862e6d3, 0x3fed80e316c98398, 0x3fed955d71ff6075 + .quad 0x3feda9e603db3285, 0x3fedbe7cd63a8315, 0x3fedd321f301b460, 0x3fede7d5641c0658 + .quad 0x3fedfc97337b9b5f, 0x3fee11676b197d17, 0x3fee264614f5a129, 0x3fee3b333b16ee12 + .quad 0x3fee502ee78b3ff6, 0x3fee653924676d76, 0x3fee7a51fbc74c83, 0x3fee8f7977cdb740 + .quad 0x3feea4afa2a490da, 0x3feeb9f4867cca6e, 0x3feecf482d8e67f1, 0x3feee4aaa2188510 + .quad 0x3feefa1bee615a27, 0x3fef0f9c1cb6412a, 0x3fef252b376bba97, 0x3fef3ac948dd7274 + .quad 0x3fef50765b6e4540, 0x3fef6632798844f8, 0x3fef7bfdad9cbe14, 0x3fef91d802243c89 + .quad 0x3fefa7c1819e90d8, 0x3fefbdba3692d514, 0x3fefd3c22b8f71f1, 0x3fefe9d96b2a23d9 + .quad 0x3ff0000000000000 + .align 32 + .quad 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe /* _dbInvLn2 = 1/log(2) */ + .align 32 + .quad 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000 /* _dbLn2hi = log(2) hi*/ + .align 32 + .quad 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899 /* _dbLn2lo = log(2) lo*/ + .align 32 + .quad 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000 /* _dbShifter */ + .align 32 + .long 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF, 0x000000FF /* _iIndexMask */ + .align 32 + .quad 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD /* _dPC2 */ + .align 32 + .quad 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14 /* _dPC3 */ + .align 32 + .quad 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299 /* _dPC4 */ + .align 32 + .long 0x00000100, 0x00000100, 0x00000100, 0x00000100, 0x00000100, 0x00000100, 0x00000100, 0x00000100 /* _iMaxIndex */ + .align 32 + .quad 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000 /* _lExpMask */ + .align 32 + .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000 /* _dSign*/ + .align 32 + .long 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99 /* _iDomainRange 0x40861d9ac12a3e85 =(1021*2^K-0.5)*log(2)/2^K -needed for quick exp*/ + .align 32 + .type __svml_dcosh_data_internal,@object + .size __svml_dcosh_data_internal,.-__svml_dcosh_data_internal diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core-avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core-avx2.S new file mode 100644 index 0000000000..8b385cc297 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core-avx2.S @@ -0,0 +1,20 @@ +/* AVX2 version of vectorized cosh, vector length is 8. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVeN8v_cosh _ZGVeN8v_cosh_avx2_wrapper +#include "../svml_d_cosh8_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core.c b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core.c new file mode 100644 index 0000000000..576b3186d5 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core.c @@ -0,0 +1,27 @@ +/* Multiple versions of vectorized cosh, vector length is 8. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVeN8v_cosh +#include "ifunc-mathvec-avx512-skx.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVeN8v_cosh, __GI__ZGVeN8v_cosh, __redirect__ZGVeN8v_cosh) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S new file mode 100644 index 0000000000..53040cef9a --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_d_cosh8_core_avx512.S @@ -0,0 +1,323 @@ +/* Function cosh vectorized with AVX-512. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_dcosh_data_internal + */ +#define _dTp_h 0 +#define _dTn_h 128 +#define _dbShifter_UISA 256 +#define _dPC2_UISA 320 +#define _dPC3_UISA 384 +#define _dPC4_UISA 448 +#define _dPC5_UISA 512 +#define _dPC6_UISA 576 +#define _dPC7_UISA 640 +#define _dbInvLn2 704 +#define _dbLn2hi 768 +#define _dbLn2lo 832 +#define _dbShifter 896 +#define _dPC2 960 +#define _dPC3 1024 +#define _dPC4 1088 +#define _lExpMask 1152 +#define _dSign 1216 +#define _iDomainRange 1280 + +#include <sysdep.h> + + .text + .section .text.evex512,"ax",@progbits +ENTRY(_ZGVeN8v_cosh_skx) + pushq %rbp + cfi_def_cfa_offset(16) + movq %rsp, %rbp + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + andq $-64, %rsp + subq $192, %rsp + vmovups _dSign+__svml_dcosh_data_internal(%rip), %zmm11 + vmovups _dbShifter_UISA+__svml_dcosh_data_internal(%rip), %zmm15 + +/* + * Load argument + * dM = x*2^K/log(2) + RShifter + */ + vmovups _dbInvLn2+__svml_dcosh_data_internal(%rip), %zmm4 + vmovups _dbLn2hi+__svml_dcosh_data_internal(%rip), %zmm2 + vmovups _dbLn2lo+__svml_dcosh_data_internal(%rip), %zmm3 + vmovups _dPC7_UISA+__svml_dcosh_data_internal(%rip), %zmm8 + vmovups _dPC6_UISA+__svml_dcosh_data_internal(%rip), %zmm9 + vmovups _dPC2_UISA+__svml_dcosh_data_internal(%rip), %zmm7 + vmovups _dPC3_UISA+__svml_dcosh_data_internal(%rip), %zmm6 + vmovaps %zmm0, %zmm10 + +/* Abs argument */ + vandnpd %zmm10, %zmm11, %zmm5 + +/* Index and lookup */ + vmovups __svml_dcosh_data_internal(%rip), %zmm11 + vmovups _dTn_h+__svml_dcosh_data_internal(%rip), %zmm0 + vfmadd213pd {rn-sae}, %zmm15, %zmm5, %zmm4 + +/* + * Check for overflow\underflow + * + */ + vpsrlq $32, %zmm5, %zmm12 + +/* dN = dM - RShifter */ + vsubpd {rn-sae}, %zmm15, %zmm4, %zmm1 + vpmovqd %zmm12, %ymm13 + vpermt2pd _dTn_h+64+__svml_dcosh_data_internal(%rip), %zmm4, %zmm0 + vpermt2pd _dTp_h+64+__svml_dcosh_data_internal(%rip), %zmm4, %zmm11 + +/* dR = dX - dN*Log2_hi/2^K */ + vfnmadd231pd {rn-sae}, %zmm2, %zmm1, %zmm5 + +/* + * poly(r) = Gmjp(1 + a2*r^2 + a4*r^4) + Gmjn*(r+ a3*r^3 +a5*r^5) = + * = Gmjp_h +Gmjp_l+ Gmjp*r^2*(a2 + a4*r^2) + Gmjn*(r+ r^3*(a3 +a5*r^2) + */ + vmovups _dPC5_UISA+__svml_dcosh_data_internal(%rip), %zmm12 + vpsllq $48, %zmm4, %zmm2 + +/* dR = dX - dN*Log2_hi/2^K */ + vfnmadd231pd {rn-sae}, %zmm3, %zmm1, %zmm5 + vmulpd {rn-sae}, %zmm5, %zmm5, %zmm1 + vfmadd231pd {rn-sae}, %zmm1, %zmm8, %zmm12 + vmovups _dPC4_UISA+__svml_dcosh_data_internal(%rip), %zmm8 + vfmadd213pd {rn-sae}, %zmm6, %zmm1, %zmm12 + vfmadd231pd {rn-sae}, %zmm1, %zmm9, %zmm8 + vfmadd213pd {rn-sae}, %zmm7, %zmm1, %zmm8 + vpcmpgtd _iDomainRange+__svml_dcosh_data_internal(%rip), %ymm13, %ymm14 + vmovmskps %ymm14, %edx + +/* dOut=r^2*(a2 + a4*r^2) */ + vmulpd {rn-sae}, %zmm1, %zmm8, %zmm6 + +/* lM now is an EXP(2^N) */ + vpandq _lExpMask+__svml_dcosh_data_internal(%rip), %zmm2, %zmm3 + vpaddq %zmm3, %zmm11, %zmm4 + vpsubq %zmm3, %zmm0, %zmm0 + vsubpd {rn-sae}, %zmm0, %zmm4, %zmm14 + vaddpd {rn-sae}, %zmm0, %zmm4, %zmm13 + +/* dM=r^2*(a3 +a5*r^2) */ + vmulpd {rn-sae}, %zmm1, %zmm12, %zmm0 + vfmadd213pd {rn-sae}, %zmm13, %zmm13, %zmm6 + +/* dM= r + r^3*(a3 +a5*r^2) */ + vfmadd213pd {rn-sae}, %zmm5, %zmm5, %zmm0 + vfmadd213pd {rn-sae}, %zmm6, %zmm14, %zmm0 + testl %edx, %edx + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx r12 r13 r14 r15 edx zmm0 zmm10 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + movq %rbp, %rsp + popq %rbp + cfi_def_cfa(7, 8) + cfi_restore(6) + ret + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + vmovups %zmm10, 64(%rsp) + vmovups %zmm0, 128(%rsp) + # LOE rbx r12 r13 r14 r15 edx zmm0 + + xorl %eax, %eax + # LOE rbx r12 r13 r14 r15 eax edx + + vzeroupper + movq %r12, 16(%rsp) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22 + movl %eax, %r12d + movq %r13, 8(%rsp) + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22 + movl %edx, %r13d + movq %r14, (%rsp) + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $8, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + vmovups 128(%rsp), %zmm0 + +/* Go to exit */ + jmp L(EXIT) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r12 r13 r14 r15 zmm0 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movsd 64(%rsp,%r14,8), %xmm0 + call cosh@PLT + # LOE rbx r14 r15 r12d r13d xmm0 + + movsd %xmm0, 128(%rsp,%r14,8) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx r15 r12d r13d +END(_ZGVeN8v_cosh_skx) + + .section .rodata, "a" + .align 64 + +#ifdef __svml_dcosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(64)) VUINT32 _dTp_h[(1<<4)][2]; + __declspec(align(64)) VUINT32 _dTn_h[(1<<4)][2]; + __declspec(align(64)) VUINT32 _dbShifter_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC2_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC3_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC4_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC5_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC6_UISA[8][2]; + __declspec(align(64)) VUINT32 _dPC7_UISA[8][2]; + __declspec(align(64)) VUINT32 _dbInvLn2[8][2]; + __declspec(align(64)) VUINT32 _dbLn2hi[8][2]; + __declspec(align(64)) VUINT32 _dbLn2lo[8][2]; + __declspec(align(64)) VUINT32 _dbShifter[8][2]; + __declspec(align(64)) VUINT32 _dPC2[8][2]; + __declspec(align(64)) VUINT32 _dPC3[8][2]; + __declspec(align(64)) VUINT32 _dPC4[8][2]; + __declspec(align(64)) VUINT32 _lExpMask[8][2]; + __declspec(align(64)) VUINT32 _dSign[8][2]; //0x8000000000000000 + __declspec(align(64)) VUINT32 _iDomainRange[16][1]; +} __svml_dcosh_data_internal; +#endif +__svml_dcosh_data_internal: + /*== _dTp_h ==*/ + .quad 0x3fe0000000000000, 0x3fe0b5586cf9890f, 0x3fe172b83c7d517b, 0x3fe2387a6e756238 + .quad 0x3fe306fe0a31b715, 0x3fe3dea64c123422, 0x3fe4bfdad5362a27, 0x3fe5ab07dd485429 + .quad 0x3fe6a09e667f3bcd, 0x3fe7a11473eb0187, 0x3fe8ace5422aa0db, 0x3fe9c49182a3f090 + .quad 0x3feae89f995ad3ad, 0x3fec199bdd85529c, 0x3fed5818dcfba487, 0x3feea4afa2a490da + /*== dTn_h ==*/ + .align 64 + .quad 0x3fe0000000000000, 0x3fdea4afa2a490da, 0x3fdd5818dcfba487, 0x3fdc199bdd85529c + .quad 0x3fdae89f995ad3ad, 0x3fd9c49182a3f090, 0x3fd8ace5422aa0db, 0x3fd7a11473eb0187 + .quad 0x3fd6a09e667f3bcd, 0x3fd5ab07dd485429, 0x3fd4bfdad5362a27, 0x3fd3dea64c123422 + .quad 0x3fd306fe0a31b715, 0x3fd2387a6e756238, 0x3fd172b83c7d517b, 0x3fd0b5586cf9890f + .align 64 + .quad 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000, 0x42F8000000000000 /* _dbShifter_UISA */ + .align 64 + .quad 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004, 0x3fe0000000000004 /* _dPC2_UISA */ + .align 64 + .quad 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543, 0x3fc5555555555543 /* _dPC3_UISA */ + .align 64 + .quad 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37, 0x3fa5555555484f37 /* _dPC4_UISA */ + .align 64 + .quad 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c, 0x3f81111111286a0c /* _dPC5_UISA */ + .align 64 + .quad 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116, 0x3f56c183da08f116 /* _dPC6_UISA */ + .align 64 + .quad 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da, 0x3f2a018d76da03da /* _dPC7_UISA */ + /*== _dbT ==*/ + .align 64 + .quad 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe, 0x3ff71547652b82fe /* _dbInvLn2 = 1/log(2) */ + .align 64 + .quad 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000, 0x3FE62E42FEFC0000 /* _dbLn2hi = log(2) hi*/ + .align 64 + .quad 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899, 0xBDAC610CA86C3899 /* _dbLn2lo = log(2) lo*/ + .align 64 + .quad 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000, 0x42B8000000000000 /* _dbShifter */ + .align 64 + .quad 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD, 0x3FDFFFFFFFFFFDBD /* _dPC2 */ + .align 64 + .quad 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14, 0x3FC5555570813E14 /* _dPC3 */ + .align 64 + .quad 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299, 0x3FA55555CF16D299 /* _dPC4 */ + .align 64 + .quad 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000, 0x7ff0000000000000 /* _lExpMask */ + .align 64 + .quad 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000, 0x8000000000000000 /* _dSign*/ + .align 64 + .long 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99, 0x40861d99 /* _iDomainRange 0x40861d9ac12a3e85 =(1021*2^K-0.5)*log(2)/2^K -needed for quick exp*/ + .align 64 + .type __svml_dcosh_data_internal,@object + .size __svml_dcosh_data_internal,.-__svml_dcosh_data_internal diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core-avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core-avx2.S new file mode 100644 index 0000000000..456d8a129f --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core-avx2.S @@ -0,0 +1,20 @@ +/* AVX2 version of vectorized coshf. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVeN16v_coshf _ZGVeN16v_coshf_avx2_wrapper +#include "../svml_s_coshf16_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core.c b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core.c new file mode 100644 index 0000000000..34c008871a --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core.c @@ -0,0 +1,28 @@ +/* Multiple versions of vectorized coshf, vector length is 16. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVeN16v_coshf +#include "ifunc-mathvec-avx512-skx.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVeN16v_coshf, __GI__ZGVeN16v_coshf, + __redirect__ZGVeN16v_coshf) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S new file mode 100644 index 0000000000..276e3cfe4d --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf16_core_avx512.S @@ -0,0 +1,321 @@ +/* Function coshf vectorized with AVX-512. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_scosh_data_internal + */ +#define _sExp_tbl_PH 0 +#define _sExp_tbl_NH 128 +#define _sShifter_UISA 256 +#define _iDomainRange_UISA 320 +#define _sPC1_UISA 384 +#define _sPC2_UISA 448 +#define _sPC3_UISA 512 +#define _sInvLn2 576 +#define _sLn2hi 640 +#define _sLn2lo 704 +#define _sSign 768 +#define _iExpMask 832 +#define _sShifter 896 +#define _iDomainRange 960 +#define _sPC1 1024 +#define _sPC2 1088 +#define _sPC3 1152 + +#include <sysdep.h> + + .text + .section .text.exex512,"ax",@progbits +ENTRY(_ZGVeN16v_coshf_skx) + pushq %rbp + cfi_def_cfa_offset(16) + movq %rsp, %rbp + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + andq $-64, %rsp + subq $192, %rsp + vmovups _sSign+__svml_scosh_data_internal(%rip), %zmm4 + vmovups _sShifter_UISA+__svml_scosh_data_internal(%rip), %zmm6 + +/* + * Load argument + * dM = x/log(2) + RShifter + */ + vmovups _sInvLn2+__svml_scosh_data_internal(%rip), %zmm10 + vmovups _sLn2hi+__svml_scosh_data_internal(%rip), %zmm7 + vmovups _sLn2lo+__svml_scosh_data_internal(%rip), %zmm9 + +/* */ + vmovups _sPC3_UISA+__svml_scosh_data_internal(%rip), %zmm2 + +/* x^2 */ + vmovups _sPC2_UISA+__svml_scosh_data_internal(%rip), %zmm3 + +/* G1,G2 2^N,2^(-N) */ + vmovups __svml_scosh_data_internal(%rip), %zmm12 + vmovups _sExp_tbl_NH+__svml_scosh_data_internal(%rip), %zmm13 + +/* + * Implementation + * Abs argument + */ + vandnps %zmm0, %zmm4, %zmm1 + +/* Check for overflow\underflow */ + vpternlogd $255, %zmm5, %zmm5, %zmm5 + vfmadd213ps {rn-sae}, %zmm6, %zmm1, %zmm10 + vpcmpd $1, _iDomainRange_UISA+__svml_scosh_data_internal(%rip), %zmm1, %k1 + +/* iM now is an EXP(2^N) */ + vpslld $18, %zmm10, %zmm11 + +/* + * R + * sN = sM - RShifter + */ + vsubps {rn-sae}, %zmm6, %zmm10, %zmm8 + vpermt2ps _sExp_tbl_PH+64+__svml_scosh_data_internal(%rip), %zmm10, %zmm12 + vpermt2ps _sExp_tbl_NH+64+__svml_scosh_data_internal(%rip), %zmm10, %zmm13 + vpandnd %zmm1, %zmm1, %zmm5{%k1} + +/* sR = sX - sN*Log2_hi */ + vfnmadd231ps {rn-sae}, %zmm7, %zmm8, %zmm1 + vptestmd %zmm5, %zmm5, %k0 + +/* sR = (sX - sN*Log2_hi) - sN*Log2_lo */ + vfnmadd231ps {rn-sae}, %zmm9, %zmm8, %zmm1 + kmovw %k0, %edx + vmulps {rn-sae}, %zmm1, %zmm1, %zmm4 + vmulps {rn-sae}, %zmm4, %zmm2, %zmm2 + +/* sSinh_r = r + r*(r^2*(a3)) */ + vfmadd213ps {rn-sae}, %zmm1, %zmm1, %zmm2 + +/* sOut = r^2*(a2) */ + vmulps {rn-sae}, %zmm4, %zmm3, %zmm1 + vpandd _iExpMask+__svml_scosh_data_internal(%rip), %zmm11, %zmm14 + vpaddd %zmm14, %zmm12, %zmm15 + vpsubd %zmm14, %zmm13, %zmm10 + +/* sG2 = 2^N*Th + 2^(-N)*T_h */ + vaddps {rn-sae}, %zmm10, %zmm15, %zmm5 + +/* sG1 = 2^N*Th - 2^(-N)*T_h */ + vsubps {rn-sae}, %zmm10, %zmm15, %zmm6 + +/* res = sG1*(r + r*(r^2*(a3))) + sG2*(1+r^2*(a2)) */ + vfmadd213ps {rn-sae}, %zmm5, %zmm5, %zmm1 + vfmadd213ps {rn-sae}, %zmm1, %zmm2, %zmm6 + testl %edx, %edx + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx r12 r13 r14 r15 edx zmm0 zmm6 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + vmovaps %zmm6, %zmm0 + movq %rbp, %rsp + popq %rbp + cfi_def_cfa(7, 8) + cfi_restore(6) + ret + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + vmovups %zmm0, 64(%rsp) + vmovups %zmm6, 128(%rsp) + # LOE rbx r12 r13 r14 r15 edx zmm6 + + xorl %eax, %eax + # LOE rbx r12 r13 r14 r15 eax edx + + vzeroupper + movq %r12, 16(%rsp) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22 + movl %eax, %r12d + movq %r13, 8(%rsp) + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22 + movl %edx, %r13d + movq %r14, (%rsp) + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $16, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + vmovups 128(%rsp), %zmm6 + +/* Go to exit */ + jmp L(EXIT) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -176; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x50, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -184; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x48, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -64; DW_OP_and; DW_OP_const4s: -192; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xc0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0x40, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r12 r13 r14 r15 zmm6 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movss 64(%rsp,%r14,4), %xmm0 + call coshf@PLT + # LOE rbx r14 r15 r12d r13d xmm0 + + movss %xmm0, 128(%rsp,%r14,4) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx r15 r12d r13d +END(_ZGVeN16v_coshf_skx) + + .section .rodata, "a" + .align 64 + +#ifdef __svml_scosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(64)) VUINT32 _sExp_tbl_PH[32][1]; + __declspec(align(64)) VUINT32 _sExp_tbl_NH[32][1]; + __declspec(align(64)) VUINT32 _sShifter_UISA[16][1]; + __declspec(align(64)) VUINT32 _iDomainRange_UISA[16][1]; + __declspec(align(64)) VUINT32 _sPC1_UISA[16][1]; + __declspec(align(64)) VUINT32 _sPC2_UISA[16][1]; + __declspec(align(64)) VUINT32 _sPC3_UISA[16][1]; + __declspec(align(64)) VUINT32 _sInvLn2[16][1]; + __declspec(align(64)) VUINT32 _sLn2hi[16][1]; + __declspec(align(64)) VUINT32 _sLn2lo[16][1]; + __declspec(align(64)) VUINT32 _sSign[16][1]; + __declspec(align(64)) VUINT32 _iExpMask[16][1]; + __declspec(align(64)) VUINT32 _sShifter[16][1]; + __declspec(align(64)) VUINT32 _iDomainRange[16][1]; + __declspec(align(64)) VUINT32 _sPC1[16][1]; + __declspec(align(64)) VUINT32 _sPC2[16][1]; + __declspec(align(64)) VUINT32 _sPC3[16][1]; +} __svml_scosh_data_internal; +#endif +__svml_scosh_data_internal: + /* _sExp_tbl_PH 2^(i/32-1), i=0..31 */ + .long 0x3f000000, 0x3f02cd87, 0x3f05aac3, 0x3f08980f + .long 0x3f0b95c2, 0x3f0ea43a, 0x3f11c3d3, 0x3f14f4f0 + .long 0x3f1837f0, 0x3f1b8d3a, 0x3f1ef532, 0x3f227043 + .long 0x3f25fed7, 0x3f29a15b, 0x3f2d583f, 0x3f3123f6 + .long 0x3f3504f3, 0x3f38fbaf, 0x3f3d08a4, 0x3f412c4d + .long 0x3f45672a, 0x3f49b9be, 0x3f4e248c, 0x3f52a81e + .long 0x3f5744fd, 0x3f5bfbb8, 0x3f60ccdf, 0x3f65b907 + .long 0x3f6ac0c7, 0x3f6fe4ba, 0x3f75257d, 0x3f7a83b3 + /* _sExp_tbl_NH 2^(-i/32-1), i=0..31 */ + .align 64 + .long 0x3f000000, 0x3efa83b3, 0x3ef5257d, 0x3eefe4ba + .long 0x3eeac0c7, 0x3ee5b907, 0x3ee0ccdf, 0x3edbfbb8 + .long 0x3ed744fd, 0x3ed2a81e, 0x3ece248c, 0x3ec9b9be + .long 0x3ec5672a, 0x3ec12c4d, 0x3ebd08a4, 0x3eb8fbaf + .long 0x3eb504f3, 0x3eb123f6, 0x3ead583f, 0x3ea9a15b + .long 0x3ea5fed7, 0x3ea27043, 0x3e9ef532, 0x3e9b8d3a + .long 0x3e9837f0, 0x3e94f4f0, 0x3e91c3d3, 0x3e8ea43a + .long 0x3e8b95c2, 0x3e88980f, 0x3e85aac3, 0x3e82cd87 + .align 64 + .long 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000, 0x48c00000 /* 1.5*2^18 _sShifter_UISA */ + .align 64 + .long 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E /* _iDomainRange_UISA */ + .align 64 + .long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 /* _sPC1_UISA=1 */ + .align 64 + .long 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f, 0x3f00010f /* _sPC2_UISA */ + .align 64 + .long 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd, 0x3e2aaacd /* _sPC3_UISA */ + .align 64 + .long 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B /* _sInvLn2 */ //k=0 + .align 64 + .long 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000 /* _sLn2hi */ + .align 64 + .long 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4 /* _sLn2lo */ + .align 64 + .long 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000 /* _sSign */ + .align 64 + .long 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000, 0x7f800000 /* _iExpMask */ + .align 64 + .long 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000 /* _sShifter */ + .align 64 + .long 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E /* _iDomainRange */ + .align 64 + .long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 /* _sPC1=1 */ + .align 64 + .long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _sPC2 */ + .align 64 + .long 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57 /* _sPC3 */ + .align 64 + .type __svml_scosh_data_internal,@object + .size __svml_scosh_data_internal,.-__svml_scosh_data_internal diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core-sse2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core-sse2.S new file mode 100644 index 0000000000..c719dc7d6a --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core-sse2.S @@ -0,0 +1,20 @@ +/* SSE2 version of vectorized coshf, vector length is 4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVbN4v_coshf _ZGVbN4v_coshf_sse2 +#include "../svml_s_coshf4_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core.c b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core.c new file mode 100644 index 0000000000..c2dfcd44f8 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core.c @@ -0,0 +1,28 @@ +/* Multiple versions of vectorized coshf, vector length is 4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVbN4v_coshf +#include "ifunc-mathvec-sse4_1.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVbN4v_coshf, __GI__ZGVbN4v_coshf, + __redirect__ZGVbN4v_coshf) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core_sse4.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core_sse4.S new file mode 100644 index 0000000000..506f6a4bd9 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf4_core_sse4.S @@ -0,0 +1,305 @@ +/* Function coshf vectorized with SSE4. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_scosh_data_internal + */ +#define _sInvLn2 0 +#define _sLn2hi 16 +#define _sLn2lo 32 +#define _sSign 48 +#define _sShifter 64 +#define _iDomainRange 80 +#define _sPC1 96 +#define _sPC2 112 +#define _sPC3 128 +#define _sPC4 144 +#define _sPC5 160 +#define _sPC6 176 +#define _iHalf 192 + +#include <sysdep.h> + + .text + .section .text.sse4,"ax",@progbits +ENTRY(_ZGVbN4v_coshf_sse4) + subq $72, %rsp + cfi_def_cfa_offset(80) + +/* + * Implementation + * Abs argument + */ + movups _sSign+__svml_scosh_data_internal(%rip), %xmm1 + +/* + * Load argument + * dM = x/log(2) + RShifter + */ + movups _sInvLn2+__svml_scosh_data_internal(%rip), %xmm9 + andnps %xmm0, %xmm1 + mulps %xmm1, %xmm9 + +/* Check for overflow\underflow */ + movaps %xmm1, %xmm3 + movups _sShifter+__svml_scosh_data_internal(%rip), %xmm4 + movups _sLn2hi+__svml_scosh_data_internal(%rip), %xmm5 + addps %xmm4, %xmm9 + +/* + * R + * sN = sM - RShifter + */ + movaps %xmm9, %xmm6 + +/* + * G1,G2 2^N,2^(-N) + * iM now is an EXP(2^N) + */ + pslld $23, %xmm9 + movups _sLn2lo+__svml_scosh_data_internal(%rip), %xmm7 + subps %xmm4, %xmm6 + +/* sR = sX - sN*Log2_hi */ + mulps %xmm6, %xmm5 + +/* sR = (sX - sN*Log2_hi) - sN*Log2_lo */ + mulps %xmm6, %xmm7 + movdqu _iDomainRange+__svml_scosh_data_internal(%rip), %xmm2 + pcmpgtd %xmm2, %xmm3 + pcmpeqd %xmm1, %xmm2 + +/* + * sinh(r) = r*((a1=1)+r^2*(a3+r^2*(a5+{v1 r^2*a7})))) = r + r*(r^2*(a3+r^2*(a5+r^2*a7))) .... + * sSinh_r = (a3+r^2*a5) + */ + movups _sPC5+__svml_scosh_data_internal(%rip), %xmm10 + por %xmm2, %xmm3 + +/* + * sinh(X) = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) + * sOut = (a4 +a6*sR2) + */ + movups _sPC6+__svml_scosh_data_internal(%rip), %xmm11 + subps %xmm5, %xmm1 + movmskps %xmm3, %edx + movdqu _iHalf+__svml_scosh_data_internal(%rip), %xmm8 + subps %xmm7, %xmm1 + +/* sR2 = sR^2,shaffled */ + movaps %xmm1, %xmm13 + movdqa %xmm8, %xmm2 + mulps %xmm1, %xmm13 + paddd %xmm9, %xmm2 + mulps %xmm13, %xmm10 + psubd %xmm9, %xmm8 + mulps %xmm13, %xmm11 + addps _sPC3+__svml_scosh_data_internal(%rip), %xmm10 + addps _sPC4+__svml_scosh_data_internal(%rip), %xmm11 + +/* sSinh_r = r^2*(a3+r^2*a5) */ + mulps %xmm13, %xmm10 + +/* sOut = a2+sR2*(a4+a6*sR2) */ + mulps %xmm13, %xmm11 + +/* sSinh_r = r + r*(r^2*(a3+r^2*a5)) */ + mulps %xmm1, %xmm10 + addps _sPC2+__svml_scosh_data_internal(%rip), %xmm11 + addps %xmm10, %xmm1 + +/* sOut = sR2*(a2+sR2*(a4+a6*sR2) */ + mulps %xmm11, %xmm13 + +/* sG1 = 2^(N-1)-2^(-N-1) */ + movdqa %xmm2, %xmm12 + +/* sG2 = 2^(N-1)+2^(-N-1) */ + addps %xmm8, %xmm2 + subps %xmm8, %xmm12 + +/* sOut = sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + mulps %xmm2, %xmm13 + +/* sOut = sG1*sinh(dR)+sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + mulps %xmm1, %xmm12 + addps %xmm12, %xmm13 + +/* sOut = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + addps %xmm13, %xmm2 + +/* Ret H */ + testl %edx, %edx + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx rbp r12 r13 r14 r15 edx xmm0 xmm2 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + movaps %xmm2, %xmm0 + addq $72, %rsp + cfi_def_cfa_offset(8) + ret + cfi_def_cfa_offset(80) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + movups %xmm0, 32(%rsp) + movups %xmm2, 48(%rsp) + # LOE rbx rbp r12 r13 r14 r15 edx + + xorl %eax, %eax + movq %r12, 16(%rsp) + cfi_offset(12, -64) + movl %eax, %r12d + movq %r13, 8(%rsp) + cfi_offset(13, -72) + movl %edx, %r13d + movq %r14, (%rsp) + cfi_offset(14, -80) + # LOE rbx rbp r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx rbp r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $4, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx rbp r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + movups 48(%rsp), %xmm2 + +/* Go to exit */ + jmp L(EXIT) + cfi_offset(12, -64) + cfi_offset(13, -72) + cfi_offset(14, -80) + # LOE rbx rbp r12 r13 r14 r15 xmm2 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movss 32(%rsp,%r14,4), %xmm0 + call coshf@PLT + # LOE rbx rbp r14 r15 r12d r13d xmm0 + + movss %xmm0, 48(%rsp,%r14,4) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx rbp r15 r12d r13d +END(_ZGVbN4v_coshf_sse4) + + .section .rodata, "a" + .align 16 + +#ifdef __svml_scosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(16)) VUINT32 _sInvLn2[4][1]; + __declspec(align(16)) VUINT32 _sLn2hi[4][1]; + __declspec(align(16)) VUINT32 _sLn2lo[4][1]; + __declspec(align(16)) VUINT32 _sSign[4][1]; + __declspec(align(16)) VUINT32 _sShifter[4][1]; + __declspec(align(16)) VUINT32 _iDomainRange[4][1]; + __declspec(align(16)) VUINT32 _sPC1[4][1]; + __declspec(align(16)) VUINT32 _sPC2[4][1]; + __declspec(align(16)) VUINT32 _sPC3[4][1]; + __declspec(align(16)) VUINT32 _sPC4[4][1]; + __declspec(align(16)) VUINT32 _sPC5[4][1]; + __declspec(align(16)) VUINT32 _sPC6[4][1]; + __declspec(align(16)) VUINT32 _iHalf[4][1]; +} __svml_scosh_data_internal; +#endif +__svml_scosh_data_internal: + .long 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B /* _sInvLn2 */ //k=0 + .align 16 + .long 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000 /* _sLn2hi */ + .align 16 + .long 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4 /* _sLn2lo */ + .align 16 + .long 0x80000000, 0x80000000, 0x80000000, 0x80000000 /* _sSign */ + .align 16 + .long 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000 /* _sShifter */ + .align 16 + .long 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E /* _iDomainRange */ + .align 16 + .long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 /* _sPC1=1 */ + .align 16 + .long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _sPC2 */ + .align 16 + .long 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57 /* _sPC3 */ + .align 16 + .long 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72 /* _sPC4 */ + .align 16 + .long 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461 /* _sPC5 */ + .align 16 + .long 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3 /* _sPC6 */ + // Integer constants + .align 16 + .long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _iHalf*/ + .align 16 + .type __svml_scosh_data_internal,@object + .size __svml_scosh_data_internal,.-__svml_scosh_data_internal diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core-sse.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core-sse.S new file mode 100644 index 0000000000..c27229e1fa --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core-sse.S @@ -0,0 +1,20 @@ +/* SSE version of vectorized coshf, vector length is 8. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define _ZGVdN8v_coshf _ZGVdN8v_coshf_sse_wrapper +#include "../svml_s_coshf8_core.S" diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core.c b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core.c new file mode 100644 index 0000000000..e82818b2c9 --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core.c @@ -0,0 +1,28 @@ +/* Multiple versions of vectorized coshf, vector length is 8. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#define SYMBOL_NAME _ZGVdN8v_coshf +#include "ifunc-mathvec-avx2.h" + +libc_ifunc_redirected (REDIRECT_NAME, SYMBOL_NAME, IFUNC_SELECTOR ()); + +#ifdef SHARED +__hidden_ver1 (_ZGVdN8v_coshf, __GI__ZGVdN8v_coshf, + __redirect__ZGVdN8v_coshf) + __attribute__ ((visibility ("hidden"))); +#endif diff --git a/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S new file mode 100644 index 0000000000..9149061e7e --- /dev/null +++ b/sysdeps/x86_64/fpu/multiarch/svml_s_coshf8_core_avx2.S @@ -0,0 +1,308 @@ +/* Function coshf vectorized with AVX2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + https://www.gnu.org/licenses/. */ + +/* + * ALGORITHM DESCRIPTION: + * + * Compute cosh(x) as (exp(x)+exp(-x))/2, + * where exp is calculated as + * exp(M*ln2 + ln2*(j/2^k) + r) = 2^M * 2^(j/2^k) * exp(r) + * + * Special cases: + * + * cosh(NaN) = quiet NaN, and raise invalid exception + * cosh(INF) = that INF + * cosh(0) = 1 + * cosh(x) overflows for big x and returns MAXLOG+log(2) + * + */ + +/* Offsets for data table __svml_scosh_data_internal + */ +#define _sInvLn2 0 +#define _sLn2hi 32 +#define _sLn2lo 64 +#define _sSign 96 +#define _sShifter 128 +#define _iDomainRange 160 +#define _sPC1 192 +#define _sPC2 224 +#define _sPC3 256 +#define _sPC4 288 +#define _sPC5 320 +#define _sPC6 352 +#define _iHalf 384 + +#include <sysdep.h> + + .text + .section .text.avx2,"ax",@progbits +ENTRY(_ZGVdN8v_coshf_avx2) + pushq %rbp + cfi_def_cfa_offset(16) + movq %rsp, %rbp + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + andq $-32, %rsp + subq $96, %rsp + vmovups _sSign+__svml_scosh_data_internal(%rip), %ymm2 + vmovups _sShifter+__svml_scosh_data_internal(%rip), %ymm7 + +/* + * Load argument + * dM = x/log(2) + RShifter + */ + vmovups _sInvLn2+__svml_scosh_data_internal(%rip), %ymm10 + vmovups _sLn2hi+__svml_scosh_data_internal(%rip), %ymm8 + vmovups _iDomainRange+__svml_scosh_data_internal(%rip), %ymm3 + +/* + * sinh(r) = r*((a1=1)+r^2*(a3+r^2*(a5+{v1 r^2*a7})))) = r + r*(r^2*(a3+r^2*(a5+r^2*a7))) .... + * sSinh_r = (a3+r^2*a5) + */ + vmovups _sPC5+__svml_scosh_data_internal(%rip), %ymm15 + vmovups _iHalf+__svml_scosh_data_internal(%rip), %ymm11 + vmovaps %ymm0, %ymm1 + +/* + * Implementation + * Abs argument + */ + vandnps %ymm1, %ymm2, %ymm0 + vfmadd213ps %ymm7, %ymm0, %ymm10 + +/* + * R + * sN = sM - RShifter + */ + vsubps %ymm7, %ymm10, %ymm9 + +/* + * G1,G2 2^N,2^(-N) + * iM now is an EXP(2^N) + */ + vpslld $23, %ymm10, %ymm12 + +/* Check for overflow\underflow */ + vpcmpgtd %ymm3, %ymm0, %ymm4 + vpcmpeqd %ymm3, %ymm0, %ymm5 + +/* sR = sX - sN*Log2_hi */ + vfnmadd231ps %ymm8, %ymm9, %ymm0 + vpaddd %ymm12, %ymm11, %ymm13 + vpsubd %ymm12, %ymm11, %ymm14 + vpor %ymm5, %ymm4, %ymm6 + +/* sR = (sX - sN*Log2_hi) - sN*Log2_lo */ + vfnmadd231ps _sLn2lo+__svml_scosh_data_internal(%rip), %ymm9, %ymm0 + +/* sG1 = 2^(N-1)-2^(-N-1) */ + vsubps %ymm14, %ymm13, %ymm4 + +/* sG2 = 2^(N-1)+2^(-N-1) */ + vaddps %ymm14, %ymm13, %ymm3 + +/* sR2 = sR^2,shaffled */ + vmulps %ymm0, %ymm0, %ymm2 + vfmadd213ps _sPC3+__svml_scosh_data_internal(%rip), %ymm2, %ymm15 + +/* sSinh_r = r^2*(a3+r^2*a5) */ + vmulps %ymm15, %ymm2, %ymm13 + +/* sSinh_r = r + r*(r^2*(a3+r^2*a5)) */ + vfmadd213ps %ymm0, %ymm0, %ymm13 + +/* + * sinh(X) = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) + * sOut = (a4 +a6*sR2) + */ + vmovups _sPC6+__svml_scosh_data_internal(%rip), %ymm0 + vfmadd213ps _sPC4+__svml_scosh_data_internal(%rip), %ymm2, %ymm0 + +/* sOut = a2+sR2*(a4+a6*sR2) */ + vfmadd213ps _sPC2+__svml_scosh_data_internal(%rip), %ymm2, %ymm0 + +/* sOut = sR2*(a2+sR2*(a4+a6*sR2) */ + vmulps %ymm0, %ymm2, %ymm15 + +/* sOut = sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + vmulps %ymm15, %ymm3, %ymm14 + +/* sOut = sG1*sinh(dR)+sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + vfmadd213ps %ymm14, %ymm13, %ymm4 + vmovmskps %ymm6, %edx + +/* sOut = sG2 + sG1*sinh(dR) + sG2*sR2*(a2+sR2*(a4+a6*sR2) */ + vaddps %ymm4, %ymm3, %ymm0 + +/* Ret H */ + testl %edx, %edx + +/* Go to special inputs processing branch */ + jne L(SPECIAL_VALUES_BRANCH) + # LOE rbx r12 r13 r14 r15 edx ymm0 ymm1 + +/* Restore registers + * and exit the function + */ + +L(EXIT): + movq %rbp, %rsp + popq %rbp + cfi_def_cfa(7, 8) + cfi_restore(6) + ret + cfi_def_cfa(6, 16) + cfi_offset(6, -16) + +/* Branch to process + * special inputs + */ + +L(SPECIAL_VALUES_BRANCH): + vmovups %ymm1, 32(%rsp) + vmovups %ymm0, 64(%rsp) + # LOE rbx r12 r13 r14 r15 edx ymm0 + + xorl %eax, %eax + # LOE rbx r12 r13 r14 r15 eax edx + + vzeroupper + movq %r12, 16(%rsp) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22 + movl %eax, %r12d + movq %r13, 8(%rsp) + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22 + movl %edx, %r13d + movq %r14, (%rsp) + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r15 r12d r13d + +/* Range mask + * bits check + */ + +L(RANGEMASK_CHECK): + btl %r12d, %r13d + +/* Call scalar math function */ + jc L(SCALAR_MATH_CALL) + # LOE rbx r15 r12d r13d + +/* Special inputs + * processing loop + */ + +L(SPECIAL_VALUES_LOOP): + incl %r12d + cmpl $8, %r12d + +/* Check bits in range mask */ + jl L(RANGEMASK_CHECK) + # LOE rbx r15 r12d r13d + + movq 16(%rsp), %r12 + cfi_restore(12) + movq 8(%rsp), %r13 + cfi_restore(13) + movq (%rsp), %r14 + cfi_restore(14) + vmovups 64(%rsp), %ymm0 + +/* Go to exit */ + jmp L(EXIT) + /* DW_CFA_expression: r12 (r12) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -80; DW_OP_plus) */ + .cfi_escape 0x10, 0x0c, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xb0, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r13 (r13) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -88; DW_OP_plus) */ + .cfi_escape 0x10, 0x0d, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa8, 0xff, 0xff, 0xff, 0x22 + /* DW_CFA_expression: r14 (r14) (DW_OP_lit8; DW_OP_minus; DW_OP_const4s: -32; DW_OP_and; DW_OP_const4s: -96; DW_OP_plus) */ + .cfi_escape 0x10, 0x0e, 0x0e, 0x38, 0x1c, 0x0d, 0xe0, 0xff, 0xff, 0xff, 0x1a, 0x0d, 0xa0, 0xff, 0xff, 0xff, 0x22 + # LOE rbx r12 r13 r14 r15 ymm0 + +/* Scalar math fucntion call + * to process special input + */ + +L(SCALAR_MATH_CALL): + movl %r12d, %r14d + movss 32(%rsp,%r14,4), %xmm0 + call coshf@PLT + # LOE rbx r14 r15 r12d r13d xmm0 + + movss %xmm0, 64(%rsp,%r14,4) + +/* Process special inputs in loop */ + jmp L(SPECIAL_VALUES_LOOP) + # LOE rbx r15 r12d r13d +END(_ZGVdN8v_coshf_avx2) + + .section .rodata, "a" + .align 32 + +#ifdef __svml_scosh_data_internal_typedef +typedef unsigned int VUINT32; +typedef struct +{ + __declspec(align(32)) VUINT32 _sInvLn2[8][1]; + __declspec(align(32)) VUINT32 _sLn2hi[8][1]; + __declspec(align(32)) VUINT32 _sLn2lo[8][1]; + __declspec(align(32)) VUINT32 _sSign[8][1]; + __declspec(align(32)) VUINT32 _sShifter[8][1]; + __declspec(align(32)) VUINT32 _iDomainRange[8][1]; + __declspec(align(32)) VUINT32 _sPC1[8][1]; + __declspec(align(32)) VUINT32 _sPC2[8][1]; + __declspec(align(32)) VUINT32 _sPC3[8][1]; + __declspec(align(32)) VUINT32 _sPC4[8][1]; + __declspec(align(32)) VUINT32 _sPC5[8][1]; + __declspec(align(32)) VUINT32 _sPC6[8][1]; + __declspec(align(32)) VUINT32 _iHalf[8][1]; +} __svml_scosh_data_internal; +#endif +__svml_scosh_data_internal: + .long 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B, 0x3FB8AA3B /* _sInvLn2 */ //k=0 + .align 32 + .long 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000, 0x3F317000 /* _sLn2hi */ + .align 32 + .long 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4, 0x3805fdf4 /* _sLn2lo */ + .align 32 + .long 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000, 0x80000000 /* _sSign */ + .align 32 + .long 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000, 0x4b400000 /* _sShifter */ + .align 32 + .long 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E, 0x42AEAC4E /* _iDomainRange */ + .align 32 + .long 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000, 0x3F800000 /* _sPC1=1 */ + .align 32 + .long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _sPC2 */ + .align 32 + .long 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57, 0x3e2aaa57 /* _sPC3 */ + .align 32 + .long 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72, 0x3d2aaa72 /* _sPC4 */ + .align 32 + .long 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461, 0x3c091461 /* _sPC5 */ + .align 32 + .long 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3, 0x3ab6a8a3 /* _sPC6 */ + // Integer constants + .align 32 + .long 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000, 0x3f000000 /* _iHalf*/ + .align 32 + .type __svml_scosh_data_internal,@object + .size __svml_scosh_data_internal,.-__svml_scosh_data_internal diff --git a/sysdeps/x86_64/fpu/svml_d_cosh2_core.S b/sysdeps/x86_64/fpu/svml_d_cosh2_core.S new file mode 100644 index 0000000000..f95952cfe5 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_d_cosh2_core.S @@ -0,0 +1,29 @@ +/* Function cosh vectorized with SSE2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_d_wrapper_impl.h" + + .text +ENTRY (_ZGVbN2v_cosh) +WRAPPER_IMPL_SSE2 cosh +END (_ZGVbN2v_cosh) + +#ifndef USE_MULTIARCH + libmvec_hidden_def (_ZGVbN2v_cosh) +#endif diff --git a/sysdeps/x86_64/fpu/svml_d_cosh4_core.S b/sysdeps/x86_64/fpu/svml_d_cosh4_core.S new file mode 100644 index 0000000000..cc24d0fb6b --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_d_cosh4_core.S @@ -0,0 +1,29 @@ +/* Function cosh vectorized with AVX2, wrapper version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_d_wrapper_impl.h" + + .text +ENTRY (_ZGVdN4v_cosh) +WRAPPER_IMPL_AVX _ZGVbN2v_cosh +END (_ZGVdN4v_cosh) + +#ifndef USE_MULTIARCH + libmvec_hidden_def (_ZGVdN4v_cosh) +#endif diff --git a/sysdeps/x86_64/fpu/svml_d_cosh4_core_avx.S b/sysdeps/x86_64/fpu/svml_d_cosh4_core_avx.S new file mode 100644 index 0000000000..4323f5e308 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_d_cosh4_core_avx.S @@ -0,0 +1,25 @@ +/* Function cosh vectorized in AVX ISA as wrapper to SSE4 ISA version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_d_wrapper_impl.h" + + .text +ENTRY (_ZGVcN4v_cosh) +WRAPPER_IMPL_AVX _ZGVbN2v_cosh +END (_ZGVcN4v_cosh) diff --git a/sysdeps/x86_64/fpu/svml_d_cosh8_core.S b/sysdeps/x86_64/fpu/svml_d_cosh8_core.S new file mode 100644 index 0000000000..90ee1ca125 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_d_cosh8_core.S @@ -0,0 +1,25 @@ +/* Function cosh vectorized with AVX-512, wrapper to AVX2. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_d_wrapper_impl.h" + + .text +ENTRY (_ZGVeN8v_cosh) +WRAPPER_IMPL_AVX512 _ZGVdN4v_cosh +END (_ZGVeN8v_cosh) diff --git a/sysdeps/x86_64/fpu/svml_s_coshf16_core.S b/sysdeps/x86_64/fpu/svml_s_coshf16_core.S new file mode 100644 index 0000000000..fe243b8b94 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_s_coshf16_core.S @@ -0,0 +1,25 @@ +/* Function coshf vectorized with AVX-512. Wrapper to AVX2 version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_s_wrapper_impl.h" + + .text +ENTRY (_ZGVeN16v_coshf) +WRAPPER_IMPL_AVX512 _ZGVdN8v_coshf +END (_ZGVeN16v_coshf) diff --git a/sysdeps/x86_64/fpu/svml_s_coshf4_core.S b/sysdeps/x86_64/fpu/svml_s_coshf4_core.S new file mode 100644 index 0000000000..b55ede6e38 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_s_coshf4_core.S @@ -0,0 +1,29 @@ +/* Function coshf vectorized with SSE2, wrapper version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_s_wrapper_impl.h" + + .text +ENTRY (_ZGVbN4v_coshf) +WRAPPER_IMPL_SSE2 coshf +END (_ZGVbN4v_coshf) + +#ifndef USE_MULTIARCH + libmvec_hidden_def (_ZGVbN4v_coshf) +#endif diff --git a/sysdeps/x86_64/fpu/svml_s_coshf8_core.S b/sysdeps/x86_64/fpu/svml_s_coshf8_core.S new file mode 100644 index 0000000000..3ea02d0f19 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_s_coshf8_core.S @@ -0,0 +1,29 @@ +/* Function coshf vectorized with AVX2, wrapper version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_s_wrapper_impl.h" + + .text +ENTRY (_ZGVdN8v_coshf) +WRAPPER_IMPL_AVX _ZGVbN4v_coshf +END (_ZGVdN8v_coshf) + +#ifndef USE_MULTIARCH + libmvec_hidden_def (_ZGVdN8v_coshf) +#endif diff --git a/sysdeps/x86_64/fpu/svml_s_coshf8_core_avx.S b/sysdeps/x86_64/fpu/svml_s_coshf8_core_avx.S new file mode 100644 index 0000000000..9b3002f7c9 --- /dev/null +++ b/sysdeps/x86_64/fpu/svml_s_coshf8_core_avx.S @@ -0,0 +1,25 @@ +/* Function coshf vectorized in AVX ISA as wrapper to SSE4 ISA version. + Copyright (C) 2021 Free Software Foundation, Inc. + This file is part of the GNU C Library. + + The GNU C Library is free software; you can redistribute it and/or + modify it under the terms of the GNU Lesser General Public + License as published by the Free Software Foundation; either + version 2.1 of the License, or (at your option) any later version. + + The GNU C Library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + Lesser General Public License for more details. + + You should have received a copy of the GNU Lesser General Public + License along with the GNU C Library; if not, see + <https://www.gnu.org/licenses/>. */ + +#include <sysdep.h> +#include "svml_s_wrapper_impl.h" + + .text +ENTRY (_ZGVcN8v_coshf) +WRAPPER_IMPL_AVX _ZGVbN4v_coshf +END (_ZGVcN8v_coshf) diff --git a/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx.c b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx.c new file mode 100644 index 0000000000..1dd311a562 --- /dev/null +++ b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx.c @@ -0,0 +1 @@ +#include "test-double-libmvec-cosh.c" diff --git a/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx2.c b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx2.c new file mode 100644 index 0000000000..1dd311a562 --- /dev/null +++ b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx2.c @@ -0,0 +1 @@ +#include "test-double-libmvec-cosh.c" diff --git a/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx512f.c b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx512f.c new file mode 100644 index 0000000000..1dd311a562 --- /dev/null +++ b/sysdeps/x86_64/fpu/test-double-libmvec-cosh-avx512f.c @@ -0,0 +1 @@ +#include "test-double-libmvec-cosh.c" diff --git a/sysdeps/x86_64/fpu/test-double-libmvec-cosh.c b/sysdeps/x86_64/fpu/test-double-libmvec-cosh.c new file mode 100644 index 0000000000..cf49ec5d87 --- /dev/null +++ b/sysdeps/x86_64/fpu/test-double-libmvec-cosh.c @@ -0,0 +1,3 @@ +#define LIBMVEC_TYPE double +#define LIBMVEC_FUNC cosh +#include "test-vector-abi-arg1.h" diff --git a/sysdeps/x86_64/fpu/test-double-vlen2-wrappers.c b/sysdeps/x86_64/fpu/test-double-vlen2-wrappers.c index 256e8f07c9..68c449e04a 100644 --- a/sysdeps/x86_64/fpu/test-double-vlen2-wrappers.c +++ b/sysdeps/x86_64/fpu/test-double-vlen2-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asin), _ZGVbN2v_asin) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypot), _ZGVbN2vv_hypot) VECTOR_WRAPPER (WRAPPER_NAME (exp2), _ZGVbN2v_exp2) VECTOR_WRAPPER (WRAPPER_NAME (exp10), _ZGVbN2v_exp10) +VECTOR_WRAPPER (WRAPPER_NAME (cosh), _ZGVbN2v_cosh) #define VEC_INT_TYPE __m128i diff --git a/sysdeps/x86_64/fpu/test-double-vlen4-avx2-wrappers.c b/sysdeps/x86_64/fpu/test-double-vlen4-avx2-wrappers.c index 9de1dab2c2..df67306373 100644 --- a/sysdeps/x86_64/fpu/test-double-vlen4-avx2-wrappers.c +++ b/sysdeps/x86_64/fpu/test-double-vlen4-avx2-wrappers.c @@ -36,6 +36,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asin), _ZGVdN4v_asin) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypot), _ZGVdN4vv_hypot) VECTOR_WRAPPER (WRAPPER_NAME (exp2), _ZGVdN4v_exp2) VECTOR_WRAPPER (WRAPPER_NAME (exp10), _ZGVdN4v_exp10) +VECTOR_WRAPPER (WRAPPER_NAME (cosh), _ZGVdN4v_cosh) #ifndef __ILP32__ # define VEC_INT_TYPE __m256i diff --git a/sysdeps/x86_64/fpu/test-double-vlen4-wrappers.c b/sysdeps/x86_64/fpu/test-double-vlen4-wrappers.c index 43865ab099..1a6731098f 100644 --- a/sysdeps/x86_64/fpu/test-double-vlen4-wrappers.c +++ b/sysdeps/x86_64/fpu/test-double-vlen4-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asin), _ZGVcN4v_asin) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypot), _ZGVcN4vv_hypot) VECTOR_WRAPPER (WRAPPER_NAME (exp2), _ZGVcN4v_exp2) VECTOR_WRAPPER (WRAPPER_NAME (exp10), _ZGVcN4v_exp10) +VECTOR_WRAPPER (WRAPPER_NAME (cosh), _ZGVcN4v_cosh) #define VEC_INT_TYPE __m128i diff --git a/sysdeps/x86_64/fpu/test-double-vlen8-wrappers.c b/sysdeps/x86_64/fpu/test-double-vlen8-wrappers.c index 5dbdacf617..4cdfa918e8 100644 --- a/sysdeps/x86_64/fpu/test-double-vlen8-wrappers.c +++ b/sysdeps/x86_64/fpu/test-double-vlen8-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asin), _ZGVeN8v_asin) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypot), _ZGVeN8vv_hypot) VECTOR_WRAPPER (WRAPPER_NAME (exp2), _ZGVeN8v_exp2) VECTOR_WRAPPER (WRAPPER_NAME (exp10), _ZGVeN8v_exp10) +VECTOR_WRAPPER (WRAPPER_NAME (cosh), _ZGVeN8v_cosh) #ifndef __ILP32__ # define VEC_INT_TYPE __m512i diff --git a/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx.c b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx.c new file mode 100644 index 0000000000..905dc3ca4a --- /dev/null +++ b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx.c @@ -0,0 +1 @@ +#include "test-float-libmvec-coshf.c" diff --git a/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx2.c b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx2.c new file mode 100644 index 0000000000..905dc3ca4a --- /dev/null +++ b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx2.c @@ -0,0 +1 @@ +#include "test-float-libmvec-coshf.c" diff --git a/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx512f.c b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx512f.c new file mode 100644 index 0000000000..905dc3ca4a --- /dev/null +++ b/sysdeps/x86_64/fpu/test-float-libmvec-coshf-avx512f.c @@ -0,0 +1 @@ +#include "test-float-libmvec-coshf.c" diff --git a/sysdeps/x86_64/fpu/test-float-libmvec-coshf.c b/sysdeps/x86_64/fpu/test-float-libmvec-coshf.c new file mode 100644 index 0000000000..94b899076b --- /dev/null +++ b/sysdeps/x86_64/fpu/test-float-libmvec-coshf.c @@ -0,0 +1,3 @@ +#define LIBMVEC_TYPE float +#define LIBMVEC_FUNC coshf +#include "test-vector-abi-arg1.h" diff --git a/sysdeps/x86_64/fpu/test-float-vlen16-wrappers.c b/sysdeps/x86_64/fpu/test-float-vlen16-wrappers.c index c159c8f583..47a9862233 100644 --- a/sysdeps/x86_64/fpu/test-float-vlen16-wrappers.c +++ b/sysdeps/x86_64/fpu/test-float-vlen16-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asinf), _ZGVeN16v_asinf) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypotf), _ZGVeN16vv_hypotf) VECTOR_WRAPPER (WRAPPER_NAME (exp2f), _ZGVeN16v_exp2f) VECTOR_WRAPPER (WRAPPER_NAME (exp10f), _ZGVeN16v_exp10f) +VECTOR_WRAPPER (WRAPPER_NAME (coshf), _ZGVeN16v_coshf) #define VEC_INT_TYPE __m512i diff --git a/sysdeps/x86_64/fpu/test-float-vlen4-wrappers.c b/sysdeps/x86_64/fpu/test-float-vlen4-wrappers.c index c745ef744a..e7c5410e7b 100644 --- a/sysdeps/x86_64/fpu/test-float-vlen4-wrappers.c +++ b/sysdeps/x86_64/fpu/test-float-vlen4-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asinf), _ZGVbN4v_asinf) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypotf), _ZGVbN4vv_hypotf) VECTOR_WRAPPER (WRAPPER_NAME (exp2f), _ZGVbN4v_exp2f) VECTOR_WRAPPER (WRAPPER_NAME (exp10f), _ZGVbN4v_exp10f) +VECTOR_WRAPPER (WRAPPER_NAME (coshf), _ZGVbN4v_coshf) #define VEC_INT_TYPE __m128i diff --git a/sysdeps/x86_64/fpu/test-float-vlen8-avx2-wrappers.c b/sysdeps/x86_64/fpu/test-float-vlen8-avx2-wrappers.c index c9226cf4dc..b8e9d48cd6 100644 --- a/sysdeps/x86_64/fpu/test-float-vlen8-avx2-wrappers.c +++ b/sysdeps/x86_64/fpu/test-float-vlen8-avx2-wrappers.c @@ -36,6 +36,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asinf), _ZGVdN8v_asinf) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypotf), _ZGVdN8vv_hypotf) VECTOR_WRAPPER (WRAPPER_NAME (exp2f), _ZGVdN8v_exp2f) VECTOR_WRAPPER (WRAPPER_NAME (exp10f), _ZGVdN8v_exp10f) +VECTOR_WRAPPER (WRAPPER_NAME (coshf), _ZGVdN8v_coshf) /* Redefinition of wrapper to be compatible with _ZGVdN8vvv_sincosf. */ #undef VECTOR_WRAPPER_fFF diff --git a/sysdeps/x86_64/fpu/test-float-vlen8-wrappers.c b/sysdeps/x86_64/fpu/test-float-vlen8-wrappers.c index 92970c5ace..328c827b27 100644 --- a/sysdeps/x86_64/fpu/test-float-vlen8-wrappers.c +++ b/sysdeps/x86_64/fpu/test-float-vlen8-wrappers.c @@ -33,6 +33,7 @@ VECTOR_WRAPPER (WRAPPER_NAME (asinf), _ZGVcN8v_asinf) VECTOR_WRAPPER_ff (WRAPPER_NAME (hypotf), _ZGVcN8vv_hypotf) VECTOR_WRAPPER (WRAPPER_NAME (exp2f), _ZGVcN8v_exp2f) VECTOR_WRAPPER (WRAPPER_NAME (exp10f), _ZGVcN8v_exp10f) +VECTOR_WRAPPER (WRAPPER_NAME (coshf), _ZGVcN8v_coshf) #define VEC_INT_TYPE __m128i |