Return to mul_1.asm CVS log

Up to [local] / OpenXM_contrib / gmp / mpn / sparc32 / v8

Annotation of OpenXM_contrib/gmp/mpn/sparc32/v8/mul_1.asm, Revision 1.1.1.2

1.1       maekawa     1: dnl  SPARC v8 mpn_mul_1 -- Multiply a limb vector with a single limb and
                      2: dnl  store the product in a second limb vector.
                      3:
1.1.1.2 ! ohara       4: dnl  Copyright 1992, 1994, 1995, 2000 Free Software Foundation, Inc.
1.1       maekawa     5:
                      6: dnl  This file is part of the GNU MP Library.
                      7:
                      8: dnl  The GNU MP Library is free software; you can redistribute it and/or modify
                      9: dnl  it under the terms of the GNU Lesser General Public License as published
                     10: dnl  by the Free Software Foundation; either version 2.1 of the License, or (at
                     11: dnl  your option) any later version.
                     12:
                     13: dnl  The GNU MP Library is distributed in the hope that it will be useful, but
                     14: dnl  WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY
                     15: dnl  or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU Lesser General Public
                     16: dnl  License for more details.
                     17:
                     18: dnl  You should have received a copy of the GNU Lesser General Public License
                     19: dnl  along with the GNU MP Library; see the file COPYING.LIB.  If not, write to
                     20: dnl  the Free Software Foundation, Inc., 59 Temple Place - Suite 330, Boston,
                     21: dnl  MA 02111-1307, USA.
                     22:
                     23:
                     24: include(`../config.m4')
                     25:
                     26: C INPUT PARAMETERS
                     27: C res_ptr      o0
                     28: C s1_ptr       o1
                     29: C size         o2
                     30: C s2_limb      o3
                     31:
                     32: ASM_START()
                     33: PROLOGUE(mpn_mul_1)
                     34:        sll     %o2,4,%g1
                     35:        and     %g1,(4-1)<<4,%g1
                     36: ifdef(`PIC',
                     37: `      mov     %o7,%g4         C Save return address register
                     38: 0:     call    1f
                     39:        add     %o7,L(1)-0b,%g3
                     40: 1:     mov     %g4,%o7         C Restore return address register
                     41: ',
                     42: `      sethi   %hi(L(1)),%g3
                     43:        or      %g3,%lo(L(1)),%g3
                     44: ')
                     45:        jmp     %g3+%g1
                     46:        ld      [%o1+0],%o4     C 1
                     47: L(1):
                     48: L(L00):        add     %o0,-4,%o0
                     49:        add     %o1,-4,%o1
                     50:        b       L(loop00)       C 4, 8, 12, ...
                     51:        orcc    %g0,%g0,%g2
                     52: L(L01):        b       L(loop01)       C 1, 5, 9, ...
                     53:        orcc    %g0,%g0,%g2
                     54:        nop
                     55:        nop
                     56: L(L10):        add     %o0,-12,%o0     C 2, 6, 10, ...
                     57:        add     %o1,4,%o1
                     58:        b       L(loop10)
                     59:        orcc    %g0,%g0,%g2
                     60:        nop
                     61: L(L11):        add     %o0,-8,%o0      C 3, 7, 11, ...
                     62:        add     %o1,-8,%o1
                     63:        b       L(loop11)
                     64:        orcc    %g0,%g0,%g2
                     65:
                     66: L(loop):
                     67:        addcc   %g3,%g2,%g3     C 1
                     68:        ld      [%o1+4],%o4     C 2
                     69:        st      %g3,[%o0+0]     C 1
                     70:        rd      %y,%g2          C 1
                     71: L(loop00):
                     72:        umul    %o4,%o3,%g3     C 2
                     73:        addxcc  %g3,%g2,%g3     C 2
                     74:        ld      [%o1+8],%o4     C 3
                     75:        st      %g3,[%o0+4]     C 2
                     76:        rd      %y,%g2          C 2
                     77: L(loop11):
                     78:        umul    %o4,%o3,%g3     C 3
                     79:        addxcc  %g3,%g2,%g3     C 3
                     80:        ld      [%o1+12],%o4    C 4
                     81:        add     %o1,16,%o1
                     82:        st      %g3,[%o0+8]     C 3
                     83:        rd      %y,%g2          C 3
                     84: L(loop10):
                     85:        umul    %o4,%o3,%g3     C 4
                     86:        addxcc  %g3,%g2,%g3     C 4
                     87:        ld      [%o1+0],%o4     C 1
                     88:        st      %g3,[%o0+12]    C 4
                     89:        add     %o0,16,%o0
                     90:        rd      %y,%g2          C 4
                     91:        addx    %g0,%g2,%g2
                     92: L(loop01):
                     93:        addcc   %o2,-4,%o2
                     94:        bg      L(loop)
                     95:        umul    %o4,%o3,%g3     C 1
                     96:
                     97:        addcc   %g3,%g2,%g3     C 4
                     98:        st      %g3,[%o0+0]     C 4
                     99:        rd      %y,%g2          C 4
                    100:
                    101:        retl
                    102:        addx    %g0,%g2,%o0
                    103: EPILOGUE(mpn_mul_1)