diff options
Diffstat (limited to 'sysdeps/i386/i586/rshift.S')
| -rw-r--r-- | sysdeps/i386/i586/rshift.S | 82 |
1 files changed, 43 insertions, 39 deletions
diff --git a/sysdeps/i386/i586/rshift.S b/sysdeps/i386/i586/rshift.S index 1860daf4d3..14ffbb8454 100644 --- a/sysdeps/i386/i586/rshift.S +++ b/sysdeps/i386/i586/rshift.S @@ -1,23 +1,21 @@ /* Pentium optimized __mpn_rshift -- + Copyright (C) 1992, 1994, 1995, 1996, 1997 Free Software Foundation, Inc. + This file is part of the GNU MP Library. -Copyright (C) 1992, 1994, 1995, 1996 Free Software Foundation, Inc. + The GNU MP Library is free software; you can redistribute it and/or modify + it under the terms of the GNU Library General Public License as published by + the Free Software Foundation; either version 2 of the License, or (at your + option) any later version. -This file is part of the GNU MP Library. + The GNU MP 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 Library General Public + License for more details. -The GNU MP Library is free software; you can redistribute it and/or modify -it under the terms of the GNU Library General Public License as published by -the Free Software Foundation; either version 2 of the License, or (at your -option) any later version. - -The GNU MP 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 Library General Public -License for more details. - -You should have received a copy of the GNU Library General Public License -along with the GNU MP Library; see the file COPYING.LIB. If not, write to -the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, -MA 02111-1307, USA. */ + You should have received a copy of the GNU Library General Public License + along with the GNU MP Library; see the file COPYING.LIB. If not, write to + the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston, + MA 02111-1307, USA. */ /* INPUT PARAMETERS @@ -46,15 +44,15 @@ C_SYMBOL_NAME(__mpn_rshift:) /* We can use faster code for shift-by-1 under certain conditions. */ cmp $1,%ecx - jne Lnormal + jne L(normal) leal 4(%edi),%eax cmpl %esi,%eax - jnc Lspecial /* jump if res_ptr + 1 >= s_ptr */ + jnc L(special) /* jump if res_ptr + 1 >= s_ptr */ leal (%edi,%ebp,4),%eax cmpl %eax,%esi - jnc Lspecial /* jump if s_ptr >= res_ptr + size */ + jnc L(special) /* jump if s_ptr >= res_ptr + size */ -Lnormal: +L(normal): movl (%esi),%edx addl $4,%esi xorl %eax,%eax @@ -64,12 +62,12 @@ Lnormal: decl %ebp pushl %ebp shrl $3,%ebp - jz Lend + jz L(end) movl (%edi),%eax /* fetch destination cache line */ ALIGN (2) -Loop: movl 28(%edi),%eax /* fetch destination cache line */ +L(oop): movl 28(%edi),%eax /* fetch destination cache line */ movl %edx,%ebx movl (%esi),%eax @@ -103,21 +101,23 @@ Loop: movl 28(%edi),%eax /* fetch destination cache line */ addl $32,%esi addl $32,%edi decl %ebp - jnz Loop + jnz L(oop) -Lend: popl %ebp +L(end): popl %ebp andl $7,%ebp - jz Lend2 -Loop2: movl (%esi),%eax + jz L(end2) +L(oop2): + movl (%esi),%eax shrdl %cl,%eax,%edx /* compute result limb */ movl %edx,(%edi) movl %eax,%edx addl $4,%esi addl $4,%edi decl %ebp - jnz Loop2 + jnz L(oop2) -Lend2: shrl %cl,%edx /* compute most significant limb */ +L(end2): + shrl %cl,%edx /* compute most significant limb */ movl %edx,(%edi) /* store it */ popl %eax /* pop carry limb */ @@ -133,7 +133,7 @@ Lend2: shrl %cl,%edx /* compute most significant limb */ function is documented to work for overlapping source and destination. */ -Lspecial: +L(special): leal -4(%edi,%ebp,4),%edi leal -4(%esi,%ebp,4),%esi @@ -147,12 +147,13 @@ Lspecial: shrl $1,%edx incl %ebp decl %ebp - jz LLend + jz L(Lend) movl (%edi),%eax /* fetch destination cache line */ ALIGN (2) -LLoop: movl -28(%edi),%eax /* fetch destination cache line */ +L(Loop): + movl -28(%edi),%eax /* fetch destination cache line */ movl %edx,%ebx movl (%esi),%eax @@ -186,14 +187,16 @@ LLoop: movl -28(%edi),%eax /* fetch destination cache line */ leal -32(%esi),%esi /* use leal not to clobber carry */ leal -32(%edi),%edi decl %ebp - jnz LLoop + jnz L(Loop) -LLend: popl %ebp +L(Lend): + popl %ebp sbbl %eax,%eax /* save carry in %eax */ andl $7,%ebp - jz LLend2 + jz L(Lend2) addl %eax,%eax /* restore carry from eax */ -LLoop2: movl %edx,%ebx +L(Loop2): + movl %edx,%ebx movl (%esi),%edx rcrl $1,%edx movl %ebx,(%edi) @@ -201,11 +204,12 @@ LLoop2: movl %edx,%ebx leal -4(%esi),%esi /* use leal not to clobber carry */ leal -4(%edi),%edi decl %ebp - jnz LLoop2 + jnz L(Loop2) - jmp LL1 -LLend2: addl %eax,%eax /* restore carry from eax */ -LL1: movl %edx,(%edi) /* store last limb */ + jmp L(L1) +L(Lend2): + addl %eax,%eax /* restore carry from eax */ +L(L1): movl %edx,(%edi) /* store last limb */ movl $0,%eax rcrl $1,%eax |