Return to addmul_1.s CVS log

Up to [local] / OpenXM_contrib / gmp / mpn / powerpc32

Annotation of OpenXM_contrib/gmp/mpn/powerpc32/addmul_1.s, Revision 1.1.1.1

1.1       maekawa     1: # PowerPC-32 __mpn_addmul_1 -- Multiply a limb vector with a limb and add
                      2: # the result to a second limb vector.
                      3:
                      4: # Copyright (C) 1995 Free Software Foundation, Inc.
                      5:
                      6: # This file is part of the GNU MP Library.
                      7:
                      8: # The GNU MP Library is free software; you can redistribute it and/or modify
                      9: # it under the terms of the GNU Library General Public License as published by
                     10: # the Free Software Foundation; either version 2 of the License, or (at your
                     11: # option) any later version.
                     12:
                     13: # The GNU MP Library is distributed in the hope that it will be useful, but
                     14: # WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
                     15: # or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Library General Public
                     16: # License for more details.
                     17:
                     18: # You should have received a copy of the GNU Library General Public License
                     19: # along with the GNU MP Library; see the file COPYING.LIB.  If not, write to
                     20: # the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
                     21: # MA 02111-1307, USA.
                     22:
                     23:
                     24: # INPUT PARAMETERS
                     25: # res_ptr      r3
                     26: # s1_ptr       r4
                     27: # size         r5
                     28: # s2_limb      r6
                     29:
                     30: # This is a fairly straightforward implementation.  The timing of the PC601
                     31: # is hard to understand, so I will wait to optimize this until I have some
                     32: # hardware to play with.
                     33:
                     34: # The code trivially generalizes to 64 bit limbs for the PC620.
                     35:
                     36:        .toc
                     37:        .csect .__mpn_addmul_1[PR]
                     38:        .align 2
                     39:        .globl __mpn_addmul_1
                     40:        .globl .__mpn_addmul_1
                     41:        .csect __mpn_addmul_1[DS]
                     42: __mpn_addmul_1:
                     43:        .long .__mpn_addmul_1[PR], TOC[tc0], 0
                     44:        .csect .__mpn_addmul_1[PR]
                     45: .__mpn_addmul_1:
                     46:        mtctr   5
                     47:
                     48:        lwz     0,0(4)
                     49:        mullw   7,0,6
                     50:        mulhwu  10,0,6
                     51:        lwz     9,0(3)
                     52:        addc    8,7,9
                     53:        addi    3,3,-4
                     54:        bdz     Lend
                     55:
                     56: Loop:  lwzu    0,4(4)
                     57:        stwu    8,4(3)
                     58:        mullw   8,0,6
                     59:        adde    7,8,10
                     60:        mulhwu  10,0,6
                     61:        lwz     9,4(3)
                     62:        addze   10,10
                     63:        addc    8,7,9
                     64:        bdnz    Loop
                     65:
                     66: Lend:  stw     8,4(3)
                     67:        addze   3,10
                     68:        blr