libgcc/config/rl78/mulsi3.S

   1 ;   Copyright (C) 2011 Free Software Foundation, Inc.
   2 ;   Contributed by Red Hat.
   3 ;
   4 ; This file is free software; you can redistribute it and/or modify it
   5 ; under the terms of the GNU General Public License as published by the
   6 ; Free Software Foundation; either version 3, or (at your option) any
   7 ; later version.
   8 ;
   9 ; This file is distributed in the hope that it will be useful, but
  10 ; WITHOUT ANY WARRANTY; without even the implied warranty of
  11 ; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  12 ; General Public License for more details.
  13 ;
  14 ; Under Section 7 of GPL version 3, you are granted additional
  15 ; permissions described in the GCC Runtime Library Exception, version
  16 ; 3.1, as published by the Free Software Foundation.
  17 ;
  18 ; You should have received a copy of the GNU General Public License and
  19 ; a copy of the GCC Runtime Library Exception along with this program;
  20 ; see the files COPYING3 and COPYING.RUNTIME respectively.  If not, see
  21 ; <http://www.gnu.org/licenses/>.
  22
  23 ;; 32x32=32 multiply
  24
  25 ; real
  26 ; GAS defines r0..r7 as aliases for real registers; we want the saddr
  27 ; forms here.
  28 r_0     =       0xffef8
  29 r_1     =       0xffef9
  30 r_2     =       0xffefa
  31 r_3     =       0xffefb
  32 r_4     =       0xffefc
  33 r_5     =       0xffefd
  34 r_6     =       0xffefe
  35 r_7     =       0xffeff
  36 ; clobberable
  37 r8      =       0xffef0
  38 r9      =       0xffef1
  39 r10     =       0xffef2
  40 r11     =       0xffef3
  41 r12     =       0xffef4
  42 r13     =       0xffef5
  43 r14     =       0xffef6
  44 r15     =       0xffef7
  45 ; preserved
  46 r16     =       0xffee8
  47 r17     =       0xffee9
  48 r18     =       0xffeea
  49 r19     =       0xffeeb
  50 r20     =       0xffeec
  51 r21     =       0xffeed
  52 r22     =       0xffeee
  53 r23     =       0xffeef
  54
  55
  56 ;----------------------------------------------------------------------
  57
  58 ; Register use:
  59 ;       RB0     RB1     RB2
  60 ; AX    op2L    res32L  res32H
  61 ; BC    op2H    (resH)  op1
  62 ; DE    count   (resL-tmp)
  63 ; HL    [sp+4]
  64
  65         .text
  66         nop
  67         .global ___mulsi3               ; (USI a, USI b)
  68 ___mulsi3:
  69         ;; A is at [sp+4]
  70         ;; B is at [sp+8]
  71         ;; result is in R8..R11
  72
  73         movw    ax, sp
  74         addw    ax, #4
  75         movw    hl, ax
  76
  77         sel     rb2
  78         push    ax
  79         push    bc
  80         sel     rb0
  81
  82         clrw    ax
  83         movw    r8, ax
  84         movw    r16, ax
  85
  86         movw    ax, [hl+6]
  87         cmpw    ax, #0
  88         bz      $1f
  89         cmpw    ax, #0xffff
  90         bnz     $2f
  91         movw    ax, [hl]
  92         sel     rb1
  93         subw    ax, r_0
  94         sel     rb0
  95         br      $1f
  96 2:
  97         movw    bc, ax
  98         movw    ax, [hl]
  99         cmpw    ax, #0
 100         skz
 101         call    !.Lmul_hi
 102 1:
 103
 104         movw    ax, [hl+2]
 105         cmpw    ax, #0
 106         bz      $1f
 107         cmpw    ax, #0xffff
 108         bnz     $2f
 109         movw    ax, [hl+4]
 110         sel     rb1
 111         subw    ax, r_0
 112         sel     rb0
 113         br      $1f
 114 2:
 115         movw    bc, ax
 116         movw    ax, [hl+4]
 117         cmpw    ax, #0
 118         skz
 119         call    !.Lmul_hi
 120 1:
 121
 122         movw    ax, r8
 123         movw    r16, ax
 124         clrw    ax
 125         movw    r8, ax
 126
 127         ;; now do R16:R8 += op1L * op2L
 128
 129         ;; op1 is in AX.0 (needs to shrw)
 130         ;; op2 is in BC.2 and BC.1 (bc can shlw/rolcw)
 131         ;; res is in AX.2 and AX.1 (needs to addw)
 132
 133         movw    ax, [hl]
 134         movw    r10, ax         ; BC.1
 135         movw    ax, [hl+4]
 136
 137         cmpw    ax, r10
 138         bc      $.Lmul_hisi_top
 139         movw    bc, r10
 140         movw    r10, ax
 141         movw    ax, bc
 142
 143
 144 .Lmul_hisi_top:
 145         movw    bc, #0
 146
 147 .Lmul_hisi_loop:
 148         shrw    ax, 1
 149         bnc     $.Lmul_hisi_no_add
 150         sel     rb1
 151         addw    ax, bc
 152         sel     rb2
 153         sknc
 154         incw    ax
 155         addw    ax, r_2
 156 .Lmul_hisi_no_add:
 157         sel     rb1
 158         shlw    bc, 1
 159         sel     rb0
 160         rolwc   bc, 1
 161         cmpw    ax, #0
 162         bz      $.Lmul_hisi_done
 163
 164         shrw    ax, 1
 165         bnc     $.Lmul_hisi_no_add2
 166         sel     rb1
 167         addw    ax, bc
 168         sel     rb2
 169         sknc
 170         incw    ax
 171         addw    ax, r_2
 172 .Lmul_hisi_no_add2:
 173         sel     rb1
 174         shlw    bc, 1
 175         sel     rb0
 176         rolwc   bc, 1
 177         cmpw    ax, #0
 178         bnz     $.Lmul_hisi_loop
 179
 180 .Lmul_hisi_done:
 181
 182         movw    ax, r16
 183         movw    r10, ax
 184
 185         sel     rb2
 186         pop     bc
 187         pop     ax
 188         sel     rb0
 189
 190         ret
 191
 192 ;----------------------------------------------------------------------
 193
 194         ;; R8 += AX * BC
 195 .Lmul_hi:
 196         cmpw    ax, bc
 197         skc
 198         xchw    ax, bc
 199         br      $.Lmul_hi_loop
 200
 201 .Lmul_hi_top:
 202         sel     rb1
 203         addw    ax, r_2
 204         sel     rb0
 205 .Lmul_hi_no_add:
 206         shlw    bc, 1
 207 .Lmul_hi_loop:
 208         shrw    ax, 1
 209         bc      $.Lmul_hi_top
 210         cmpw    ax, #0
 211         bz      $.Lmul_hi_done
 212
 213         shlw    bc, 1
 214         shrw    ax, 1
 215         bc      $.Lmul_hi_top
 216         cmpw    ax, #0
 217         bnz     $.Lmul_hi_no_add
 218
 219 .Lmul_hi_done:
 220         ret
 221
 222 ;----------------------------------------------------------------------
 223
 224         .global ___mulhi3
 225 ___mulhi3:
 226         sel     rb1
 227         clrw    ax
 228         sel     rb0
 229         movw    ax, sp
 230         addw    ax, #4
 231         movw    hl, ax
 232         movw    ax, [hl+2]
 233         movw    bc, ax
 234         movw    ax, [hl]
 235         br      $.Lmul_hi