| 1 | /* strlen used for beginning of str{n}cat using EVEX 256/512. |
| 2 | Copyright (C) 2011-2024 Free Software Foundation, Inc. |
| 3 | This file is part of the GNU C Library. |
| 4 | |
| 5 | The GNU C Library is free software; you can redistribute it and/or |
| 6 | modify it under the terms of the GNU Lesser General Public |
| 7 | License as published by the Free Software Foundation; either |
| 8 | version 2.1 of the License, or (at your option) any later version. |
| 9 | |
| 10 | The GNU C Library is distributed in the hope that it will be useful, |
| 11 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 12 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 13 | Lesser General Public License for more details. |
| 14 | |
| 15 | You should have received a copy of the GNU Lesser General Public |
| 16 | License along with the GNU C Library; if not, see |
| 17 | <https://www.gnu.org/licenses/>. */ |
| 18 | |
| 19 | |
| 20 | /* NOTE: This file is meant to be included by strcat-evex or |
| 21 | strncat-evex and does not standalone. Before including %rdi |
| 22 | must be saved in %rax. */ |
| 23 | |
| 24 | |
| 25 | /* Simple strlen implementation that ends at |
| 26 | L(strcat_strlen_done). */ |
| 27 | vpxorq %VZERO_128, %VZERO_128, %VZERO_128 |
| 28 | movq %rdi, %r8 |
| 29 | andq $(VEC_SIZE * -1), %r8 |
| 30 | VPCMPEQ (%r8), %VZERO, %k0 |
| 31 | KMOV %k0, %VRCX |
| 32 | #ifdef USE_AS_WCSCPY |
| 33 | subl %r8d, %edi |
| 34 | shrl $2, %edi |
| 35 | #endif |
| 36 | shrx %VRDI, %VRCX, %VRCX |
| 37 | #ifdef USE_AS_WCSCPY |
| 38 | movq %rax, %rdi |
| 39 | #endif |
| 40 | test %VRCX, %VRCX |
| 41 | jnz L(bsf_and_done_v0) |
| 42 | |
| 43 | |
| 44 | VPCMPEQ VEC_SIZE(%r8), %VZERO, %k0 |
| 45 | KMOV %k0, %VRCX |
| 46 | leaq (VEC_SIZE)(%r8), %rdi |
| 47 | test %VRCX, %VRCX |
| 48 | jnz L(bsf_and_done_v0) |
| 49 | |
| 50 | VPCMPEQ (VEC_SIZE * 2)(%r8), %VZERO, %k0 |
| 51 | KMOV %k0, %VRCX |
| 52 | test %VRCX, %VRCX |
| 53 | jnz L(bsf_and_done_v1) |
| 54 | |
| 55 | VPCMPEQ (VEC_SIZE * 3)(%r8), %VZERO, %k0 |
| 56 | KMOV %k0, %VRCX |
| 57 | test %VRCX, %VRCX |
| 58 | jnz L(bsf_and_done_v2) |
| 59 | |
| 60 | VPCMPEQ (VEC_SIZE * 4)(%r8), %VZERO, %k0 |
| 61 | KMOV %k0, %VRCX |
| 62 | test %VRCX, %VRCX |
| 63 | jnz L(bsf_and_done_v3) |
| 64 | |
| 65 | andq $-(VEC_SIZE * 4), %rdi |
| 66 | .p2align 4,, 8 |
| 67 | L(loop_2x_vec): |
| 68 | VMOVA (VEC_SIZE * 4)(%rdi), %VMM(0) |
| 69 | VPMIN (VEC_SIZE * 5)(%rdi), %VMM(0), %VMM(1) |
| 70 | VMOVA (VEC_SIZE * 6)(%rdi), %VMM(2) |
| 71 | VPMIN (VEC_SIZE * 7)(%rdi), %VMM(2), %VMM(3) |
| 72 | VPTESTN %VMM(1), %VMM(1), %k1 |
| 73 | VPTESTN %VMM(3), %VMM(3), %k3 |
| 74 | subq $(VEC_SIZE * -4), %rdi |
| 75 | KORTEST %k1, %k3 |
| 76 | jz L(loop_2x_vec) |
| 77 | |
| 78 | VPTESTN %VMM(0), %VMM(0), %k0 |
| 79 | KMOV %k0, %VRCX |
| 80 | test %VRCX, %VRCX |
| 81 | jnz L(bsf_and_done_v0) |
| 82 | |
| 83 | KMOV %k1, %VRCX |
| 84 | test %VRCX, %VRCX |
| 85 | jnz L(bsf_and_done_v1) |
| 86 | |
| 87 | VPTESTN %VMM(2), %VMM(2), %k0 |
| 88 | KMOV %k0, %VRCX |
| 89 | test %VRCX, %VRCX |
| 90 | jnz L(bsf_and_done_v2) |
| 91 | |
| 92 | KMOV %k3, %VRCX |
| 93 | L(bsf_and_done_v3): |
| 94 | addq $VEC_SIZE, %rdi |
| 95 | L(bsf_and_done_v2): |
| 96 | bsf %VRCX, %VRCX |
| 97 | leaq (VEC_SIZE * 2)(%rdi, %rcx, CHAR_SIZE), %rdi |
| 98 | jmp L(strcat_strlen_done) |
| 99 | |
| 100 | .p2align 4,, 4 |
| 101 | L(bsf_and_done_v1): |
| 102 | addq $VEC_SIZE, %rdi |
| 103 | L(bsf_and_done_v0): |
| 104 | bsf %VRCX, %VRCX |
| 105 | #ifdef USE_AS_WCSCPY |
| 106 | leaq (%rdi, %rcx, CHAR_SIZE), %rdi |
| 107 | #else |
| 108 | addq %rcx, %rdi |
| 109 | #endif |
| 110 | L(strcat_strlen_done): |
| 111 | |