| 1 | /* PowerPC64 __mpn_addmul_1 -- Multiply a limb vector with a limb and add |
| 2 | the result to a second limb vector. |
| 3 | Copyright (C) 1999-2024 Free Software Foundation, Inc. |
| 4 | This file is part of the GNU C Library. |
| 5 | |
| 6 | The GNU C Library is free software; you can redistribute it and/or |
| 7 | modify it under the terms of the GNU Lesser General Public |
| 8 | License as published by the Free Software Foundation; either |
| 9 | version 2.1 of the License, or (at your option) any later version. |
| 10 | |
| 11 | The GNU C Library is distributed in the hope that it will be useful, |
| 12 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 13 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 14 | Lesser General Public License for more details. |
| 15 | |
| 16 | You should have received a copy of the GNU Lesser General Public |
| 17 | License along with the GNU C Library; if not, see |
| 18 | <https://www.gnu.org/licenses/>. */ |
| 19 | |
| 20 | #include <sysdep.h> |
| 21 | |
| 22 | #ifdef USE_AS_SUBMUL |
| 23 | # define FUNC __mpn_submul_1 |
| 24 | # define ADDSUBC subfe |
| 25 | # define ADDSUB subfc |
| 26 | #else |
| 27 | # define FUNC __mpn_addmul_1 |
| 28 | # define ADDSUBC adde |
| 29 | # define ADDSUB addc |
| 30 | #endif |
| 31 | |
| 32 | #define RP r3 |
| 33 | #define UP r4 |
| 34 | #define N r5 |
| 35 | #define VL r6 |
| 36 | |
| 37 | #define R27SAVE (-40) |
| 38 | #define R28SAVE (-32) |
| 39 | #define R29SAVE (-24) |
| 40 | #define R30SAVE (-16) |
| 41 | #define R31SAVE (-8) |
| 42 | |
| 43 | ENTRY_TOCLESS (FUNC, 5) |
| 44 | std r31, R31SAVE(r1) |
| 45 | rldicl. r0, N, 0, 62 |
| 46 | std r30, R30SAVE(r1) |
| 47 | cmpdi VL, r0, 2 |
| 48 | std r29, R29SAVE(r1) |
| 49 | addi N, N, 3 |
| 50 | std r28, R28SAVE(r1) |
| 51 | srdi N, N, 2 |
| 52 | std r27, R27SAVE(r1) |
| 53 | cfi_offset(r31, R31SAVE) |
| 54 | cfi_offset(r30, R30SAVE) |
| 55 | cfi_offset(r29, R29SAVE) |
| 56 | cfi_offset(r28, R28SAVE) |
| 57 | cfi_offset(r27, R27SAVE) |
| 58 | mtctr N |
| 59 | beq cr0, L(b00) |
| 60 | blt cr6, L(b01) |
| 61 | beq cr6, L(b10) |
| 62 | |
| 63 | L(b11): ld r9, 0(UP) |
| 64 | ld r28, 0(RP) |
| 65 | mulld r0, r9, VL |
| 66 | mulhdu r12, r9, VL |
| 67 | ADDSUB r0, r0, r28 |
| 68 | std r0, 0(RP) |
| 69 | addi RP, RP, 8 |
| 70 | ld r9, 8(UP) |
| 71 | ld r27, 16(UP) |
| 72 | addi UP, UP, 24 |
| 73 | #ifdef USE_AS_SUBMUL |
| 74 | subfe r11, r11, r11 |
| 75 | #endif |
| 76 | b L(bot) |
| 77 | |
| 78 | .align 4 |
| 79 | L(b00): ld r9, 0(UP) |
| 80 | ld r27, 8(UP) |
| 81 | ld r28, 0(RP) |
| 82 | ld r29, 8(RP) |
| 83 | mulld r0, r9, VL |
| 84 | mulhdu N, r9, VL |
| 85 | mulld r7, r27, VL |
| 86 | mulhdu r8, r27, VL |
| 87 | addc r7, r7, N |
| 88 | addze r12, r8 |
| 89 | ADDSUB r0, r0, r28 |
| 90 | std r0, 0(RP) |
| 91 | ADDSUBC r7, r7, r29 |
| 92 | std r7, 8(RP) |
| 93 | addi RP, RP, 16 |
| 94 | ld r9, 16(UP) |
| 95 | ld r27, 24(UP) |
| 96 | addi UP, UP, 32 |
| 97 | #ifdef USE_AS_SUBMUL |
| 98 | subfe r11, r11, r11 |
| 99 | #endif |
| 100 | b L(bot) |
| 101 | |
| 102 | .align 4 |
| 103 | L(b01): bdnz L(gt1) |
| 104 | ld r9, 0(UP) |
| 105 | ld r11, 0(RP) |
| 106 | mulld r0, r9, VL |
| 107 | mulhdu r8, r9, VL |
| 108 | ADDSUB r0, r0, r11 |
| 109 | std r0, 0(RP) |
| 110 | #ifdef USE_AS_SUBMUL |
| 111 | subfe r11, r11, r11 |
| 112 | addic r11, r11, 1 |
| 113 | #endif |
| 114 | addze RP, r8 |
| 115 | blr |
| 116 | |
| 117 | L(gt1): ld r9, 0(UP) |
| 118 | ld r27, 8(UP) |
| 119 | mulld r0, r9, VL |
| 120 | mulhdu N, r9, VL |
| 121 | mulld r7, r27, VL |
| 122 | mulhdu r8, r27, VL |
| 123 | ld r9, 16(UP) |
| 124 | ld r28, 0(RP) |
| 125 | ld r29, 8(RP) |
| 126 | ld r30, 16(RP) |
| 127 | mulld r11, r9, VL |
| 128 | mulhdu r10, r9, VL |
| 129 | addc r7, r7, N |
| 130 | adde r11, r11, r8 |
| 131 | addze r12, r10 |
| 132 | ADDSUB r0, r0, r28 |
| 133 | std r0, 0(RP) |
| 134 | ADDSUBC r7, r7, r29 |
| 135 | std r7, 8(RP) |
| 136 | ADDSUBC r11, r11, r30 |
| 137 | std r11, 16(RP) |
| 138 | addi RP, RP, 24 |
| 139 | ld r9, 24(UP) |
| 140 | ld r27, 32(UP) |
| 141 | addi UP, UP, 40 |
| 142 | #ifdef USE_AS_SUBMUL |
| 143 | subfe r11, r11, r11 |
| 144 | #endif |
| 145 | b L(bot) |
| 146 | |
| 147 | L(b10): addic r0, r0, r0 |
| 148 | li r12, 0 |
| 149 | ld r9, 0(UP) |
| 150 | ld r27, 8(UP) |
| 151 | bdz L(end) |
| 152 | addi UP, UP, 16 |
| 153 | |
| 154 | .align 4 |
| 155 | L(top): mulld r0, r9, VL |
| 156 | mulhdu N, r9, VL |
| 157 | mulld r7, r27, VL |
| 158 | mulhdu r8, r27, VL |
| 159 | ld r9, 0(UP) |
| 160 | ld r28, 0(RP) |
| 161 | ld r27, 8(UP) |
| 162 | ld r29, 8(RP) |
| 163 | adde r0, r0, r12 |
| 164 | adde r7, r7, N |
| 165 | mulld N, r9, VL |
| 166 | mulhdu r10, r9, VL |
| 167 | mulld r11, r27, VL |
| 168 | mulhdu r12, r27, VL |
| 169 | ld r9, 16(UP) |
| 170 | ld r30, 16(RP) |
| 171 | ld r27, 24(UP) |
| 172 | ld r31, 24(RP) |
| 173 | adde N, N, r8 |
| 174 | adde r11, r11, r10 |
| 175 | addze r12, r12 |
| 176 | ADDSUB r0, r0, r28 |
| 177 | std r0, 0(RP) |
| 178 | ADDSUBC r7, r7, r29 |
| 179 | std r7, 8(RP) |
| 180 | ADDSUBC N, N, r30 |
| 181 | std N, 16(RP) |
| 182 | ADDSUBC r11, r11, r31 |
| 183 | std r11, 24(RP) |
| 184 | addi UP, UP, 32 |
| 185 | #ifdef USE_AS_SUBMUL |
| 186 | subfe r11, r11, r11 |
| 187 | #endif |
| 188 | addi RP, RP, 32 |
| 189 | L(bot): |
| 190 | #ifdef USE_AS_SUBMUL |
| 191 | addic r11, r11, 1 |
| 192 | #endif |
| 193 | bdnz L(top) |
| 194 | |
| 195 | L(end): mulld r0, r9, VL |
| 196 | mulhdu N, r9, VL |
| 197 | mulld r7, r27, VL |
| 198 | mulhdu r8, r27, VL |
| 199 | ld r28, 0(RP) |
| 200 | ld r29, 8(RP) |
| 201 | adde r0, r0, r12 |
| 202 | adde r7, r7, N |
| 203 | addze r8, r8 |
| 204 | ADDSUB r0, r0, r28 |
| 205 | std r0, 0(RP) |
| 206 | ADDSUBC r7, r7, r29 |
| 207 | std r7, 8(RP) |
| 208 | #ifdef USE_AS_SUBMUL |
| 209 | subfe r11, r11, r11 |
| 210 | addic r11, r11, 1 |
| 211 | #endif |
| 212 | addze RP, r8 |
| 213 | ld r31, R31SAVE(r1) |
| 214 | ld r30, R30SAVE(r1) |
| 215 | ld r29, R29SAVE(r1) |
| 216 | ld r28, R28SAVE(r1) |
| 217 | ld r27, R27SAVE(r1) |
| 218 | blr |
| 219 | END(FUNC) |
| 220 | |