blob: d648d9d1dc1f387f9a29d08f336aa54b2b3f606f [file] [log] [blame]
#
# (C) 2008-2009 Advanced Micro Devices, Inc. All Rights Reserved.
#
# This file is part of libacml_mv.
#
# libacml_mv 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.
#
# libacml_mv 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 libacml_mv. If not, see
# <http://www.gnu.org/licenses/>.
#
#
# v4hfrcpal.s
#
# Helper routines for testing the x4 double and x8 single vector
# math functions.
#
# Prototype:
#
# void v4frcpa(__m128d x1, __m128d x2, double * ya);
#
# Computes 4 frcpa values simultaneously and returns them
# in the v4a array.
# Assumes that ya is 16 byte aligned.
#
#
#ifdef __ELF__
.section .note.GNU-stack,"",@progbits
#endif
# xmm0 - __m128d x1
# xmm1 - __m128d x2
# rdi - double *ya
.extern __vrd4_frcpa
.text
.align 16
.p2align 4,,15
.globl v4frcpa
.type v4frcpa,@function
v4frcpa:
push %rdi
call __vrd4_frcpa@PLT
pop %rdi
movdqa %xmm0,(%rdi)
movdqa %xmm1,16(%rdi)
ret