yuezonghe | 824eb0c | 2024-06-27 02:32:26 -0700 | [diff] [blame^] | 1 | /* Optimized version of the standard memchr() function. |
| 2 | This file is part of the GNU C Library. |
| 3 | Copyright (C) 2000, 2001, 2003 Free Software Foundation, Inc. |
| 4 | Contributed by Dan Pop <Dan.Pop@cern.ch>. |
| 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, write to the Free |
| 18 | Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA |
| 19 | 02111-1307 USA. */ |
| 20 | |
| 21 | /* Return: the address of the first occurence of chr in str or NULL |
| 22 | |
| 23 | Inputs: |
| 24 | in0: str |
| 25 | in1: chr |
| 26 | in2: byte count |
| 27 | |
| 28 | This implementation assumes little endian mode. For big endian mode, |
| 29 | the instruction czx1.r should be replaced by czx1.l. |
| 30 | |
| 31 | The algorithm is fairly straightforward: search byte by byte until we |
| 32 | we get to a word aligned address, then search word by word as much as |
| 33 | possible; the remaining few bytes are searched one at a time. |
| 34 | |
| 35 | The word by word search is performed by xor-ing the word with a word |
| 36 | containing chr in every byte. If there is a hit, the result will |
| 37 | contain a zero byte in the corresponding position. The presence and |
| 38 | position of that zero byte is detected with a czx instruction. |
| 39 | |
| 40 | All the loops in this function could have had the internal branch removed |
| 41 | if br.ctop and br.cloop could be predicated :-(. */ |
| 42 | |
| 43 | #include "sysdep.h" |
| 44 | #undef ret |
| 45 | |
| 46 | #define saved_pr r15 |
| 47 | #define saved_lc r16 |
| 48 | #define chr r17 |
| 49 | #define len r18 |
| 50 | #define pos0 r20 |
| 51 | #define val r21 |
| 52 | #define tmp r24 |
| 53 | #define chrx8 r25 |
| 54 | #define loopcnt r30 |
| 55 | |
| 56 | #define str in0 |
| 57 | |
| 58 | ENTRY(__memchr) |
| 59 | .prologue |
| 60 | alloc r2 = ar.pfs, 3, 0, 29, 32 |
| 61 | #include "softpipe.h" |
| 62 | .rotr value[MEMLAT+1], addr[MEMLAT+3], aux[2], poschr[2] |
| 63 | .rotp p[MEMLAT+3] |
| 64 | .save ar.lc, saved_lc |
| 65 | mov saved_lc = ar.lc /* save the loop counter */ |
| 66 | .save pr, saved_pr |
| 67 | mov saved_pr = pr /* save the predicates */ |
| 68 | .body |
| 69 | mov ret0 = str |
| 70 | and tmp = 7, str /* tmp = str % 8 */ |
| 71 | cmp.ne p7, p0 = r0, r0 /* clear p7 */ |
| 72 | extr.u chr = in1, 0, 8 /* chr = (unsigned char) in1 */ |
| 73 | mov len = in2 |
| 74 | cmp.gtu p6, p0 = 16, in2 /* use a simple loop for short */ |
| 75 | (p6) br.cond.spnt .srchfew ;; /* searches */ |
| 76 | sub loopcnt = 8, tmp /* loopcnt = 8 - tmp */ |
| 77 | cmp.eq p6, p0 = tmp, r0 |
| 78 | (p6) br.cond.sptk .str_aligned;; |
| 79 | sub len = len, loopcnt |
| 80 | adds loopcnt = -1, loopcnt;; |
| 81 | mov ar.lc = loopcnt |
| 82 | .l1: |
| 83 | ld1 val = [ret0], 1 |
| 84 | ;; |
| 85 | cmp.eq p6, p0 = val, chr |
| 86 | (p6) br.cond.spnt .foundit |
| 87 | br.cloop.sptk .l1 ;; |
| 88 | .str_aligned: |
| 89 | cmp.ne p6, p0 = r0, r0 /* clear p6 */ |
| 90 | shr.u loopcnt = len, 3 /* loopcnt = len / 8 */ |
| 91 | and len = 7, len ;; /* remaining len = len & 7 */ |
| 92 | adds loopcnt = -1, loopcnt |
| 93 | mov ar.ec = MEMLAT + 3 |
| 94 | mux1 chrx8 = chr, @brcst ;; /* get a word full of chr */ |
| 95 | mov ar.lc = loopcnt |
| 96 | mov pr.rot = 1 << 16 ;; |
| 97 | .l2: |
| 98 | (p[0]) mov addr[0] = ret0 |
| 99 | (p[0]) ld8 value[0] = [ret0], 8 |
| 100 | (p[MEMLAT]) xor aux[0] = value[MEMLAT], chrx8 |
| 101 | (p[MEMLAT+1]) czx1.r poschr[0] = aux[1] |
| 102 | (p[MEMLAT+2]) cmp.ne p7, p0 = 8, poschr[1] |
| 103 | (p7) br.cond.dpnt .foundit |
| 104 | br.ctop.dptk .l2 |
| 105 | .srchfew: |
| 106 | adds loopcnt = -1, len |
| 107 | cmp.eq p6, p0 = len, r0 |
| 108 | (p6) br.cond.spnt .notfound ;; |
| 109 | mov ar.lc = loopcnt |
| 110 | .l3: |
| 111 | ld1 val = [ret0], 1 |
| 112 | ;; |
| 113 | cmp.eq p6, p0 = val, chr |
| 114 | (p6) br.cond.dpnt .foundit |
| 115 | br.cloop.sptk .l3 ;; |
| 116 | .notfound: |
| 117 | cmp.ne p6, p0 = r0, r0 /* clear p6 (p7 was already 0 when we got here) */ |
| 118 | mov ret0 = r0 ;; /* return NULL */ |
| 119 | .foundit: |
| 120 | .pred.rel "mutex" p6, p7 |
| 121 | (p6) adds ret0 = -1, ret0 /* if we got here from l1 or l3 */ |
| 122 | (p7) add ret0 = addr[MEMLAT+2], poschr[1] /* if we got here from l2 */ |
| 123 | mov pr = saved_pr, -1 |
| 124 | mov ar.lc = saved_lc |
| 125 | br.ret.sptk.many b0 |
| 126 | |
| 127 | END(__memchr) |
| 128 | |
| 129 | weak_alias(__memchr, memchr) |
| 130 | weak_alias(__memchr, __ubp_memchr) |
| 131 | libc_hidden_def(memchr) |