libgcc/config/rl78/mulsi3.S

   1 ;   Copyright (C) 2011-2017 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 #include "vregs.h"
  26
  27 ;----------------------------------------------------------------------
  28
  29 ; Register use:
  30 ;       RB0     RB1     RB2
  31 ; AX    op2L    res32L  res32H
  32 ; BC    op2H    (resH)  op1
  33 ; DE    count   (resL-tmp)
  34 ; HL    [sp+4]
  35
  36 ; Register use (G10):
  37 ;
  38 ; AX        op2L
  39 ; BC        op2H
  40 ; DE        count
  41 ; HL        [sp+4]
  42 ; r8/r9     res32L
  43 ; r10/r11   (resH)
  44 ; r12/r13   (resL-tmp)
  45 ; r16/r17   res32H
  46 ; r18/r19   op1
  47
  48 START_FUNC ___mulsi3
  49         ;; A is at [sp+4]
  50         ;; B is at [sp+8]
  51         ;; result is in R8..R11
  52
  53 #ifdef __RL78_G10__
  54         movw    ax, r16
  55         push    ax
  56         movw    ax, r18
  57         push    ax
  58 #else
  59         sel     rb2
  60         push    ax
  61         push    bc
  62         sel     rb0
  63 #endif
  64         clrw    ax
  65         movw    r8, ax
  66         movw    r16, ax
  67
  68         movw    ax, [sp+14]
  69         cmpw    ax, #0
  70         bz      $1f
  71         cmpw    ax, #0xffff
  72         bnz     $2f
  73         movw    ax, [sp+8]
  74 #ifdef __RL78_G10__
  75         push    bc
  76         movw    bc, r8
  77         xchw    ax, bc
  78         subw    ax, bc
  79         movw    r8, ax
  80         movw    ax, bc
  81         pop     bc
  82 #else
  83         sel     rb1
  84         subw    ax, r_0
  85         sel     rb0
  86 #endif
  87         br      $1f
  88 2:
  89         movw    bc, ax
  90         movw    ax, [sp+8]
  91         cmpw    ax, #0
  92         skz
  93         call    !.Lmul_hi
  94 1:
  95
  96         movw    ax, [sp+10]
  97         cmpw    ax, #0
  98         bz      $1f
  99         cmpw    ax, #0xffff
 100         bnz     $2f
 101         movw    ax, [sp+12]
 102 #ifdef __RL78_G10__
 103         push    bc
 104         movw    bc, r8
 105         xchw    ax, bc
 106         subw    ax, bc
 107         movw    r8, ax
 108         movw    ax, bc
 109         pop     bc
 110 #else
 111         sel     rb1
 112         subw    ax, r_0
 113         sel     rb0
 114 #endif
 115         br      $1f
 116 2:
 117         movw    bc, ax
 118         movw    ax, [sp+12]
 119         cmpw    ax, #0
 120         skz
 121         call    !.Lmul_hi
 122 1:
 123
 124         movw    ax, r8
 125         movw    r16, ax
 126         clrw    ax
 127         movw    r8, ax
 128
 129         ;; now do R16:R8 += op1L * op2L
 130
 131         ;; op1 is in AX.0 (needs to shrw)
 132         ;; op2 is in BC.2 and BC.1 (bc can shlw/rolcw)
 133         ;; res is in AX.2 and AX.1 (needs to addw)
 134
 135         movw    ax, [sp+8]
 136         movw    r10, ax         ; BC.1
 137         movw    ax, [sp+12]
 138
 139         cmpw    ax, r10
 140         bc      $.Lmul_hisi_top
 141         movw    bc, r10
 142         movw    r10, ax
 143         movw    ax, bc
 144
 145 .Lmul_hisi_top:
 146         movw    bc, #0
 147
 148 .Lmul_hisi_loop:
 149         shrw    ax, 1
 150 #ifdef __RL78_G10__
 151         push    ax
 152         bnc     $.Lmul_hisi_no_add_g10
 153         movw    ax, r8
 154         addw    ax, r10
 155         movw    r8, ax
 156         sknc
 157         incw    r16
 158         movw    ax, r16
 159         addw    ax, r_2
 160         movw    r16, ax
 161 .Lmul_hisi_no_add_g10:
 162         movw    ax, r10
 163         shlw    ax, 1
 164         movw    r10, ax
 165         pop     ax
 166 #else
 167         bnc     $.Lmul_hisi_no_add
 168         sel     rb1
 169         addw    ax, bc
 170         sel     rb2
 171         sknc
 172         incw    ax
 173         addw    ax, r_2
 174 .Lmul_hisi_no_add:
 175         sel     rb1
 176         shlw    bc, 1
 177         sel     rb0
 178 #endif
 179         rolwc   bc, 1
 180         cmpw    ax, #0
 181         bz      $.Lmul_hisi_done
 182
 183         shrw    ax, 1
 184 #ifdef __RL78_G10__
 185         push    ax
 186         bnc     $.Lmul_hisi_no_add2_g10
 187         movw    ax, r8
 188         addw    ax, r10
 189         movw    r8, ax
 190         movw    ax, r16
 191         sknc
 192         incw    ax
 193         addw    ax, r_2
 194         movw    r16, ax
 195 .Lmul_hisi_no_add2_g10:
 196         movw    ax, r10
 197         shlw    ax, 1
 198         movw    r10, ax
 199         pop     ax
 200 #else
 201         bnc     $.Lmul_hisi_no_add2
 202         sel     rb1
 203         addw    ax, bc
 204         sel     rb2
 205         sknc
 206         incw    ax
 207         addw    ax, r_2
 208 .Lmul_hisi_no_add2:
 209         sel     rb1
 210         shlw    bc, 1
 211         sel     rb0
 212 #endif
 213         rolwc   bc, 1
 214         cmpw    ax, #0
 215         bnz     $.Lmul_hisi_loop
 216
 217 .Lmul_hisi_done:
 218
 219         movw    ax, r16
 220         movw    r10, ax
 221
 222 #ifdef __RL78_G10__
 223         pop     ax
 224         movw    r18, ax
 225         pop     ax
 226         movw    r16, ax
 227 #else
 228         sel     rb2
 229         pop     bc
 230         pop     ax
 231         sel     rb0
 232 #endif
 233
 234         ret
 235 END_FUNC ___mulsi3
 236
 237 ;----------------------------------------------------------------------
 238
 239 START_FUNC ___mulhi3
 240         movw    r8, #0
 241         movw    ax, [sp+6]
 242         movw    bc, ax
 243         movw    ax, [sp+4]
 244
 245         ;; R8 += AX * BC
 246 .Lmul_hi:
 247         cmpw    ax, bc
 248         skc
 249         xchw    ax, bc
 250         br      $.Lmul_hi_loop
 251
 252 .Lmul_hi_top:
 253 #ifdef __RL78_G10__
 254         push    ax
 255         movw    ax, r8
 256         addw    ax, r_2
 257         movw    r8, ax
 258         pop     ax
 259 #else
 260         sel     rb1
 261         addw    ax, r_2
 262         sel     rb0
 263 #endif
 264
 265 .Lmul_hi_no_add:
 266         shlw    bc, 1
 267 .Lmul_hi_loop:
 268         shrw    ax, 1
 269         bc      $.Lmul_hi_top
 270         cmpw    ax, #0
 271         bz      $.Lmul_hi_done
 272
 273         shlw    bc, 1
 274         shrw    ax, 1
 275         bc      $.Lmul_hi_top
 276         cmpw    ax, #0
 277         bnz     $.Lmul_hi_no_add
 278
 279 .Lmul_hi_done:
 280         ret
 281 END_FUNC ___mulhi3
 282
 283 ;;; --------------------------------------
 284 #ifdef __RL78_G10__
 285         START_FUNC ___mulqi3
 286
 287         mov     a, [sp+4]
 288         mov     r9, a
 289         mov     a, [sp+6]
 290         mov     r10, a
 291         mov     a, #9
 292         mov     r11, a
 293         clrb    a
 294         mov     r8, a
 295 .L2:
 296         cmp0    r10
 297         skz
 298         dec     r11
 299         sknz
 300         ret
 301         mov     a, r10
 302         and     a, #1
 303         mov     r12, a
 304         cmp0    r12
 305         sknz
 306         br      !!.L3
 307         mov     a, r9
 308         mov     l, a
 309         mov     a, r8
 310         add     a, l
 311         mov     r8, a
 312 .L3:
 313         mov     a, r9
 314         add     a, a
 315         mov     r9, a
 316         mov     a, r10
 317         shr     a, 1
 318         mov     r10, a
 319         br      !!.L2
 320
 321         END_FUNC   ___mulqi3
 322 #endif
 323