Return to com_n.asm CVS log

Up to [local] / OpenXM_contrib / gmp / mpn / x86 / k6 / mmx

Annotation of OpenXM_contrib/gmp/mpn/x86/k6/mmx/com_n.asm, Revision 1.1.1.2

1.1       maekawa     1: dnl  AMD K6-2 mpn_com_n -- mpn bitwise one's complement.
                      2:
1.1.1.2 ! ohara       3: dnl  Copyright 1999, 2000, 2001, 2002 Free Software Foundation, Inc.
1.1       maekawa     4: dnl
                      5: dnl  This file is part of the GNU MP Library.
                      6: dnl
                      7: dnl  The GNU MP Library is free software; you can redistribute it and/or
                      8: dnl  modify it under the terms of the GNU Lesser General Public License as
                      9: dnl  published by the Free Software Foundation; either version 2.1 of the
                     10: dnl  License, or (at your option) any later version.
                     11: dnl
                     12: dnl  The GNU MP Library is distributed in the hope that it will be useful,
                     13: dnl  but WITHOUT ANY WARRANTY; without even the implied warranty of
                     14: dnl  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
                     15: dnl  Lesser General Public License for more details.
                     16: dnl
                     17: dnl  You should have received a copy of the GNU Lesser General Public
                     18: dnl  License along with the GNU MP Library; see the file COPYING.LIB.  If
                     19: dnl  not, write to the Free Software Foundation, Inc., 59 Temple Place -
                     20: dnl  Suite 330, Boston, MA 02111-1307, USA.
                     21:
                     22: include(`../config.m4')
                     23:
1.1.1.2 ! ohara      24: NAILS_SUPPORT(0-31)
        !            25:
        !            26:
        !            27: C    alignment dst/src, A=0mod8 N=4mod8
        !            28: C       A/A   A/N   N/A   N/N
        !            29: C K6-2  1.0   1.18  1.18  1.18  cycles/limb
        !            30: C K6    1.5   1.85  1.75  1.85
        !            31:
1.1       maekawa    32:
                     33: C void mpn_com_n (mp_ptr dst, mp_srcptr src, mp_size_t size);
                     34: C
                     35: C Take the bitwise ones-complement of src,size and write it to dst,size.
                     36:
                     37: defframe(PARAM_SIZE,12)
                     38: defframe(PARAM_SRC, 8)
                     39: defframe(PARAM_DST, 4)
                     40:
1.1.1.2 ! ohara      41:        TEXT
        !            42:        ALIGN(16)
1.1       maekawa    43: PROLOGUE(mpn_com_n)
                     44: deflit(`FRAME',0)
                     45:
                     46:        movl    PARAM_SIZE, %ecx
                     47:        movl    PARAM_SRC, %eax
                     48:        movl    PARAM_DST, %edx
                     49:        shrl    %ecx
                     50:        jnz     L(two_or_more)
                     51:
                     52:        movl    (%eax), %eax
1.1.1.2 ! ohara      53:        notl_or_xorl_GMP_NUMB_MASK(     %eax)
1.1       maekawa    54:        movl    %eax, (%edx)
                     55:        ret
                     56:
                     57:
                     58: L(two_or_more):
1.1.1.2 ! ohara      59:        pushl   %ebx    FRAME_pushl()
        !            60:        pcmpeqd %mm7, %mm7              C all ones
        !            61:
1.1       maekawa    62:        movl    %ecx, %ebx
1.1.1.2 ! ohara      63: ifelse(GMP_NAIL_BITS,0,,
        !            64: `      psrld   $GMP_NAIL_BITS, %mm7')  C clear nails
1.1       maekawa    65:
                     66:
                     67:
1.1.1.2 ! ohara      68:        ALIGN(8)
1.1       maekawa    69: L(top):
                     70:        C eax   src
                     71:        C ebx   floor(size/2)
                     72:        C ecx   counter
                     73:        C edx   dst
1.1.1.2 ! ohara      74:        C
        !            75:        C mm0   scratch
        !            76:        C mm7   mask
1.1       maekawa    77:
                     78:        movq    -8(%eax,%ecx,8), %mm0
                     79:        pxor    %mm7, %mm0
                     80:        movq    %mm0, -8(%edx,%ecx,8)
                     81:        loop    L(top)
                     82:
                     83:
                     84:        jnc     L(no_extra)
                     85:        movl    (%eax,%ebx,8), %eax
1.1.1.2 ! ohara      86:        notl_or_xorl_GMP_NUMB_MASK(     %eax)
1.1       maekawa    87:        movl    %eax, (%edx,%ebx,8)
                     88: L(no_extra):
                     89:
                     90:        popl    %ebx
                     91:        emms_or_femms
                     92:        ret
                     93:
                     94: EPILOGUE()