1dnl PPC-64 mpn_divrem_2 -- Divide an mpn number by a normalized 2-limb number. 2 3dnl Copyright 2007, 2008 Free Software Foundation, Inc. 4 5dnl This file is part of the GNU MP Library. 6dnl 7dnl The GNU MP Library is free software; you can redistribute it and/or modify 8dnl it under the terms of either: 9dnl 10dnl * the GNU Lesser General Public License as published by the Free 11dnl Software Foundation; either version 3 of the License, or (at your 12dnl option) any later version. 13dnl 14dnl or 15dnl 16dnl * the GNU General Public License as published by the Free Software 17dnl Foundation; either version 2 of the License, or (at your option) any 18dnl later version. 19dnl 20dnl or both in parallel, as here. 21dnl 22dnl The GNU MP Library is distributed in the hope that it will be useful, but 23dnl WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY 24dnl or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License 25dnl for more details. 26dnl 27dnl You should have received copies of the GNU General Public License and the 28dnl GNU Lesser General Public License along with the GNU MP Library. If not, 29dnl see https://www.gnu.org/licenses/. 30 31include(`../config.m4') 32 33C cycles/limb 34C norm frac 35C POWER3/PPC630 36C POWER4/PPC970 ? ? 37C POWER5 37 ? 38C POWER6 62 ? 39C POWER6 30.5 ? 40 41C INPUT PARAMETERS 42C qp = r3 43C fn = r4 44C up = r5 45C un = r6 46C dp = r7 47 48 49ifdef(`DARWIN',,` 50define(`r2',`r31')') C FIXME! 51 52ASM_START() 53 54EXTERN_FUNC(mpn_invert_limb) 55 56PROLOGUE(mpn_divrem_2,toc) 57 mflr r0 58 std r23, -72(r1) 59 std r24, -64(r1) 60 std r25, -56(r1) 61 std r26, -48(r1) 62 std r27, -40(r1) 63 std r28, -32(r1) 64 std r29, -24(r1) 65 std r30, -16(r1) 66 std r31, -8(r1) 67 std r0, 16(r1) 68 stdu r1, -192(r1) 69 mr r24, r3 70 mr r25, r4 71 sldi r0, r6, 3 72 add r26, r5, r0 73 addi r26, r26, -24 74 ld r30, 8(r7) 75 ld r28, 0(r7) 76 ld r29, 16(r26) 77 ld r31, 8(r26) 78 79ifelse(0,1,` 80 li r23, 0 81 cmpld cr7, r29, r30 82 blt cr7, L(8) 83 bgt cr7, L(9) 84 cmpld cr0, r31, r28 85 blt cr0, L(8) 86L(9): subfc r31, r28, r31 87 subfe r29, r30, r29 88 li r23, 1 89',` 90 li r23, 0 91 cmpld cr7, r29, r30 92 blt cr7, L(8) 93 mfcr r0 94 rlwinm r0, r0, 30, 1 95 subfc r9, r28, r31 96 addze. r0, r0 97 nop 98 beq cr0, L(8) 99 subfc r31, r28, r31 100 subfe r29, r30, r29 101 li r23, 1 102') 103 104L(8): 105 add r27, r25, r6 106 addic. r27, r27, -3 107 blt cr0, L(18) 108 mr r3, r30 109 CALL( mpn_invert_limb) 110 mulld r10, r3, r30 111 mulhdu r0, r3, r28 112 addc r8, r10, r28 113 subfe r11, r1, r1 114 addc r10, r8, r0 115 addze. r11, r11 116 blt cr0, L(91) 117L(40): 118 subfc r10, r30, r10 119 addme. r11, r11 120 addi r3, r3, -1 121 bge cr0, L(40) 122L(91): 123 addi r5, r27, 1 124 mtctr r5 125 sldi r0, r27, 3 126 add r24, r24, r0 127 ALIGN(16) 128L(loop): 129 mulhdu r8, r29, r3 130 mulld r6, r29, r3 131 addc r6, r6, r31 132 adde r8, r8, r29 133 cmpd cr7, r27, r25 134 mulld r0, r30, r8 135 mulhdu r11, r28, r8 136 mulld r10, r28, r8 137 subf r31, r0, r31 138 li r7, 0 139 blt cr7, L(60) 140 ld r7, 0(r26) 141 addi r26, r26, -8 142 nop 143L(60): subfc r7, r28, r7 144 subfe r31, r30, r31 145 subfc r7, r10, r7 146 subfe r4, r11, r31 147 subfc r9, r6, r4 148 subfe r9, r1, r1 149 andc r6, r28, r9 150 andc r0, r30, r9 151 addc r31, r7, r6 152 adde r29, r4, r0 153 subf r8, r9, r8 154 cmpld cr7, r29, r30 155 bge- cr7, L(fix) 156L(bck): std r8, 0(r24) 157 addi r24, r24, -8 158 addi r27, r27, -1 159 bdnz L(loop) 160L(18): 161 std r31, 8(r26) 162 std r29, 16(r26) 163 mr r3, r23 164 addi r1, r1, 192 165 ld r0, 16(r1) 166 mtlr r0 167 ld r23, -72(r1) 168 ld r24, -64(r1) 169 ld r25, -56(r1) 170 ld r26, -48(r1) 171 ld r27, -40(r1) 172 ld r28, -32(r1) 173 ld r29, -24(r1) 174 ld r30, -16(r1) 175 ld r31, -8(r1) 176 blr 177L(fix): 178 mfcr r0 179 rlwinm r0, r0, 30, 1 180 subfc r9, r28, r31 181 addze. r0, r0 182 beq cr0, L(bck) 183 subfc r31, r28, r31 184 subfe r29, r30, r29 185 addi r8, r8, 1 186 b L(bck) 187EPILOGUE() 188