dnl core2 mpn_divexact_byff dnl Copyright 2009 Jason Moxham dnl This file is part of the MPIR Library. dnl The MPIR Library is free software; you can redistribute it and/or modify dnl it under the terms of the GNU Lesser General Public License as published dnl by the Free Software Foundation; either version 2.1 of the License, or (at dnl your option) any later version. dnl The MPIR Library is distributed in the hope that it will be useful, but dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public dnl License for more details. dnl You should have received a copy of the GNU Lesser General Public License dnl along with the MPIR Library; see the file COPYING.LIB. If not, write dnl to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, dnl Boston, MA 02110-1301, USA. include(`../config.m4') C (rdi,rdx)=(rsi,rdx)/0xFFFFFFFFFFFFFFFF C rax="remainder" C where (rsi,rdx) = (rdi,rdx)*(B-1) -rax*B^rdx and 0 <= rax < B-1 B=0xFFFFFFFFFFFFFFFF ASM_START() PROLOGUE(mpn_divexact_byff) C this is good but suffers from alignment slowdown C we dont seem to have much freedom to re-arrange the instructions to avoid C it , I suppose we could detect alignment at the start and have different C routines for different alignments xor %eax,%eax mov %rdx,%rcx and $3,%rcx shr $2,%rdx cmp $0,%rdx C carry flag is clear here jnz loop sbb (%rsi),%rax mov %rax,(%rdi) dec %rcx jz end1 sbb 8(%rsi),%rax mov %rax,8(%rdi) dec %rcx jz end1 sbb 16(%rsi),%rax mov %rax,16(%rdi) dec %rcx end1: sbb $0,%rax ret ALIGN(16) loop: sbb (%rsi),%rax mov %rax,(%rdi) sbb 8(%rsi),%rax mov %rax,8(%rdi) sbb 16(%rsi),%rax mov %rax,16(%rdi) sbb 24(%rsi),%rax mov %rax,24(%rdi) lea 32(%rsi),%rsi dec %rdx lea 32(%rdi),%rdi jnz loop inc %rcx dec %rcx jz end sbb (%rsi),%rax mov %rax,(%rdi) dec %rcx jz end sbb 8(%rsi),%rax mov %rax,8(%rdi) dec %rcx jz end sbb 16(%rsi),%rax mov %rax,16(%rdi) dec %rcx end: sbb $0,%rax ret EPILOGUE()