Return to submul_1.s CVS log

Up to [local] / OpenXM / src / kan96xx / gmp-2.0.2-ssh-2 / mpn / sparc64

Annotation of OpenXM/src/kan96xx/gmp-2.0.2-ssh-2/mpn/sparc64/submul_1.s, Revision 1.1.1.1

1.1       takayama    1: ! SPARC v9 __mpn_submul_1 -- Multiply a limb vector with a single limb and
                      2: ! subtract the product from a second limb vector.
                      3:
                      4: ! Copyright (C) 1996 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      o0
                     26: ! s1_ptr       o1
                     27: ! size         o2
                     28: ! s2_limb      o3
                     29:
                     30: .section       ".text"
                     31:        .align 4
                     32:        .global __mpn_submul_1
                     33:        .type    __mpn_submul_1,#function
                     34:        .proc   016
                     35: __mpn_submul_1:
                     36:        !#PROLOGUE#     0
                     37:        save    %sp,-160,%sp
                     38:        !#PROLOGUE#     1
                     39:        sub     %g0,%i2,%o7
                     40:        sllx    %o7,3,%g5
                     41:        sub     %i1,%g5,%o3
                     42:        sub     %i0,%g5,%o4
                     43:        mov     0,%o0                   ! zero cy_limb
                     44:
                     45:        srl     %i3,0,%o1               ! extract low 32 bits of s2_limb
                     46:        srlx    %i3,32,%i3              ! extract high 32 bits of s2_limb
                     47:        mov     1,%o2
                     48:        sllx    %o2,32,%o2              ! o2 = 0x100000000
                     49:
                     50:        !   hi   !
                     51:              !  mid-1 !
                     52:              !  mid-2 !
                     53:                 !   lo   !
                     54: .Loop:
                     55:        sllx    %o7,3,%g1
                     56:        ldx     [%o3+%g1],%g5
                     57:        srl     %g5,0,%i0               ! zero hi bits
                     58:        srlx    %g5,32,%g5
                     59:        mulx    %o1,%i0,%i4             ! lo product
                     60:        mulx    %i3,%i0,%i1             ! mid-1 product
                     61:        mulx    %o1,%g5,%l2             ! mid-2 product
                     62:        mulx    %i3,%g5,%i5             ! hi product
                     63:        srlx    %i4,32,%i0              ! extract high 32 bits of lo product...
                     64:        add     %i1,%i0,%i1             ! ...and add it to the mid-1 product
                     65:        addcc   %i1,%l2,%i1             ! add mid products
                     66:        mov     0,%l0                   ! we need the carry from that add...
                     67:        movcs   %xcc,%o2,%l0            ! ...compute it and...
                     68:        add     %i5,%l0,%i5             ! ...add to bit 32 of the hi product
                     69:        sllx    %i1,32,%i0              ! align low bits of mid product
                     70:        srl     %i4,0,%g5               ! zero high 32 bits of lo product
                     71:        add     %i0,%g5,%i0             ! combine into low 64 bits of result
                     72:        srlx    %i1,32,%i1              ! extract high bits of mid product...
                     73:        add     %i5,%i1,%i1             ! ...and add them to the high result
                     74:        addcc   %i0,%o0,%i0             ! add cy_limb to low 64 bits of result
                     75:        mov     0,%g5
                     76:        movcs   %xcc,1,%g5
                     77:        add     %o7,1,%o7
                     78:        ldx     [%o4+%g1],%l1
                     79:        subcc   %l1,%i0,%i0
                     80:        movcs   %xcc,1,%g5
                     81:        stx     %i0,[%o4+%g1]
                     82:        brnz    %o7,.Loop
                     83:        add     %i1,%g5,%o0             ! compute new cy_limb
                     84:
                     85:        mov     %o0,%i0
                     86:        ret
                     87:        restore
                     88: .LLfe1:
                     89:        .size  __mpn_submul_1,.LLfe1-__mpn_submul_1