arch/c6x/lib/remi.S

   1 ;;  Copyright 2010  Free Software Foundation, Inc.
   2 ;;  Contributed by Bernd Schmidt <bernds@codesourcery.com>.
   3 ;;
   4 ;; This program is free software; you can redistribute it and/or modify
   5 ;; it under the terms of the GNU General Public License as published by
   6 ;; the Free Software Foundation; either version 2 of the License, or
   7 ;; (at your option) any later version.
   8 ;;
   9 ;; This program is distributed in the hope that it will be useful,
  10 ;; but WITHOUT ANY WARRANTY; without even the implied warranty of
  11 ;; MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  12 ;; GNU General Public License for more details.
  13 ;;
  14 ;; You should have received a copy of the GNU General Public License
  15 ;; along with this program; if not, write to the Free Software
  16 ;; Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
  17
  18 #include <linux/linkage.h>
  19
  20         ;; ABI considerations for the divide functions
  21         ;; The following registers are call-used:
  22         ;; __c6xabi_divi A0,A1,A2,A4,A6,B0,B1,B2,B4,B5
  23         ;; __c6xabi_divu A0,A1,A2,A4,A6,B0,B1,B2,B4
  24         ;; __c6xabi_remi A1,A2,A4,A5,A6,B0,B1,B2,B4
  25         ;; __c6xabi_remu A1,A4,A5,A7,B0,B1,B2,B4
  26         ;;
  27         ;; In our implementation, divu and remu are leaf functions,
  28         ;; while both divi and remi call into divu.
  29         ;; A0 is not clobbered by any of the functions.
  30         ;; divu does not clobber B2 either, which is taken advantage of
  31         ;; in remi.
  32         ;; divi uses B5 to hold the original return address during
  33         ;; the call to divu.
  34         ;; remi uses B2 and A5 to hold the input values during the
  35         ;; call to divu.  It stores B3 in on the stack.
  36
  37         .text
  38
  39 ENTRY(__c6xabi_remi)
  40         stw     .d2t2   B3, *B15--[2]
  41 ||      cmpgt   .l1     0, A4, A1
  42 ||      cmpgt   .l2     0, B4, B2
  43 ||      mv      .s1     A4, A5
  44 ||      call    .s2     __c6xabi_divu
  45
  46    [A1] neg     .l1     A4, A4
  47 || [B2] neg     .l2     B4, B4
  48 ||      xor     .s2x    B2, A1, B0
  49 ||      mv      .d2     B4, B2
  50
  51    [B0] addkpc  .s2     _divu_ret_1, B3, 1
  52   [!B0] addkpc  .s2     _divu_ret_2, B3, 1
  53         nop     2
  54 _divu_ret_1:
  55         neg     .l1     A4, A4
  56 _divu_ret_2:
  57         ldw     .d2t2   *++B15[2], B3
  58
  59         mpy32   .m1x    A4, B2, A6
  60         nop             3
  61         ret     .s2     B3
  62         sub     .l1     A5, A6, A4
  63         nop     4
  64 ENDPROC(__c6xabi_remi)