1/* Alpha 21064 addmul_1 -- Multiply a limb vector with a limb and add 2 * the result to a second limb vector. 3 * 4 * Copyright (C) 1992, 1994, 1995, 1998, 5 * 2001, 2002 Free Software Foundation, Inc. 6 * 7 * This file is part of Libgcrypt. 8 * 9 * Libgcrypt is free software; you can redistribute it and/or modify 10 * it under the terms of the GNU Lesser General Public License as 11 * published by the Free Software Foundation; either version 2.1 of 12 * the License, or (at your option) any later version. 13 * 14 * Libgcrypt is distributed in the hope that it will be useful, 15 * but WITHOUT ANY WARRANTY; without even the implied warranty of 16 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 17 * GNU Lesser General Public License for more details. 18 * 19 * You should have received a copy of the GNU Lesser General Public 20 * License along with this program; if not, write to the Free Software 21 * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA 22 */ 23 24 25/******************* 26 * mpi_limb_t 27 * _gcry_mpih_addmul_1( mpi_ptr_t res_ptr, (r16) 28 * mpi_ptr_t s1_ptr, (r17) 29 * mpi_size_t s1_size, (r18) 30 * mpi_limb_t s2_limb) (r19) 31 * 32 * This code runs at 42 cycles/limb on EV4 and 18 cycles/limb on EV5. 33 */ 34 35 36 .set noreorder 37 .set noat 38.text 39 .align 3 40 .globl _gcry_mpih_addmul_1 41 .ent _gcry_mpih_addmul_1 2 42_gcry_mpih_addmul_1: 43 .frame $30,0,$26 44 45 ldq $2,0($17) # $2 = s1_limb 46 addq $17,8,$17 # s1_ptr++ 47 subq $18,1,$18 # size-- 48 mulq $2,$19,$3 # $3 = prod_low 49 ldq $5,0($16) # $5 = *res_ptr 50 umulh $2,$19,$0 # $0 = prod_high 51 beq $18,.Lend1 # jump if size was == 1 52 ldq $2,0($17) # $2 = s1_limb 53 addq $17,8,$17 # s1_ptr++ 54 subq $18,1,$18 # size-- 55 addq $5,$3,$3 56 cmpult $3,$5,$4 57 stq $3,0($16) 58 addq $16,8,$16 # res_ptr++ 59 beq $18,.Lend2 # jump if size was == 2 60 61 .align 3 62.Loop: mulq $2,$19,$3 # $3 = prod_low 63 ldq $5,0($16) # $5 = *res_ptr 64 addq $4,$0,$0 # cy_limb = cy_limb + 'cy' 65 subq $18,1,$18 # size-- 66 umulh $2,$19,$4 # $4 = cy_limb 67 ldq $2,0($17) # $2 = s1_limb 68 addq $17,8,$17 # s1_ptr++ 69 addq $3,$0,$3 # $3 = cy_limb + prod_low 70 cmpult $3,$0,$0 # $0 = carry from (cy_limb + prod_low) 71 addq $5,$3,$3 72 cmpult $3,$5,$5 73 stq $3,0($16) 74 addq $16,8,$16 # res_ptr++ 75 addq $5,$0,$0 # combine carries 76 bne $18,.Loop 77 78.Lend2: mulq $2,$19,$3 # $3 = prod_low 79 ldq $5,0($16) # $5 = *res_ptr 80 addq $4,$0,$0 # cy_limb = cy_limb + 'cy' 81 umulh $2,$19,$4 # $4 = cy_limb 82 addq $3,$0,$3 # $3 = cy_limb + prod_low 83 cmpult $3,$0,$0 # $0 = carry from (cy_limb + prod_low) 84 addq $5,$3,$3 85 cmpult $3,$5,$5 86 stq $3,0($16) 87 addq $5,$0,$0 # combine carries 88 addq $4,$0,$0 # cy_limb = prod_high + cy 89 ret $31,($26),1 90.Lend1: addq $5,$3,$3 91 cmpult $3,$5,$5 92 stq $3,0($16) 93 addq $0,$5,$0 94 ret $31,($26),1 95 96 .end _gcry_mpih_addmul_1 97 98