1/* 2 * Copyright (c) 1993,94 Winning Strategies, Inc. 3 * All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by Winning Strategies, Inc. 16 * 4. The name of the author may not be used to endorse or promote products 17 * derived from this software without specific prior written permission 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 20 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 21 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 22 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, 23 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 24 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 25 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 26 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 27 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 28 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 29 */ 30 31#include <machine/asm.h>
| 1/* 2 * Copyright (c) 1993,94 Winning Strategies, Inc. 3 * All rights reserved. 4 * 5 * Redistribution and use in source and binary forms, with or without 6 * modification, are permitted provided that the following conditions 7 * are met: 8 * 1. Redistributions of source code must retain the above copyright 9 * notice, this list of conditions and the following disclaimer. 10 * 2. Redistributions in binary form must reproduce the above copyright 11 * notice, this list of conditions and the following disclaimer in the 12 * documentation and/or other materials provided with the distribution. 13 * 3. All advertising materials mentioning features or use of this software 14 * must display the following acknowledgement: 15 * This product includes software developed by Winning Strategies, Inc. 16 * 4. The name of the author may not be used to endorse or promote products 17 * derived from this software without specific prior written permission 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR 20 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES 21 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. 22 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, 23 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 24 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, 25 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY 26 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 27 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF 28 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 29 */ 30 31#include <machine/asm.h>
|
33 34/* 35 * strncmp(s1, s2, n) 36 * return an integer greater than, equal to, or less than 0, 37 * according as the first n characters of string s1 is greater 38 * than, equal to, or less than the string s2. 39 * 40 * %eax - pointer to s1 41 * %ecx - pointer to s2 42 * %edx - length 43 * 44 * Written by: 45 * J.T. Conklin (jtc@wimsey.com), Winning Strategies, Inc. 46 */ 47 48/* 49 * I've unrolled the loop eight times: large enough to make a 50 * significant difference, and small enough not to totally trash the 51 * cache. 52 * 53 * TODO: change all the jz's back to je for consistency. 54 */ 55 56ENTRY(strncmp) 57 pushl %ebx 58 movl 8(%esp),%eax 59 movl 12(%esp),%ecx 60 movl 16(%esp),%edx 61 testl %edx,%edx 62 jmp L2 /* Jump into the loop! */ 63 64 .align 2,0x90 65L1: incl %eax 66 incl %ecx 67 decl %edx 68L2: jz L4 /* strings are equal */ 69 movb (%eax),%bl 70 testb %bl,%bl 71 jz L3 72 cmpb %bl,(%ecx) 73 jne L3 74 75/* 76 * XXX it might be best to move the next 4 instructions to the end of the 77 * unrolled part of the loop. The unrolled part would then be 78 * movb n(%eax),%bl; testb %bl, %bl; je L3; cmpb n(%ecx); jne L3 79 * or maybe better 80 * movb n(%eax),%bl; cmpb n(%ecx); jne L3; testb %bl,%bl; je return_0 81 * for n = 0, 1, ..., 8. The end of the loop would be 82 * L1: addl $8,%eax; addl $8,%ecx; subl $8,%edx; cmpl $8,%edx; jae Lx 83 * where residual counts of 0 to 7 are handled at Lx. However, this would 84 * be slower for short strings. Cache effects are probably not so 85 * important because we are only handling a byte at a time. 86 */ 87 incl %eax 88 incl %ecx 89 decl %edx 90 jz L4 91 movb (%eax),%bl 92 testb %bl,%bl 93 jz L3 94 cmpb %bl,(%ecx) 95 jne L3 96 97 incl %eax 98 incl %ecx 99 decl %edx 100 jz L4 101 movb (%eax),%bl 102 testb %bl,%bl 103 jz L3 104 cmpb %bl,(%ecx) 105 jne L3 106 107 incl %eax 108 incl %ecx 109 decl %edx 110 jz L4 111 movb (%eax),%bl 112 testb %bl,%bl 113 jz L3 114 cmpb %bl,(%ecx) 115 jne L3 116 117 incl %eax 118 incl %ecx 119 decl %edx 120 jz L4 121 movb (%eax),%bl 122 testb %bl,%bl 123 jz L3 124 cmpb %bl,(%ecx) 125 jne L3 126 127 incl %eax 128 incl %ecx 129 decl %edx 130 jz L4 131 movb (%eax),%bl 132 testb %bl,%bl 133 jz L3 134 cmpb %bl,(%ecx) 135 jne L3 136 137 incl %eax 138 incl %ecx 139 decl %edx 140 jz L4 141 movb (%eax),%bl 142 testb %bl,%bl 143 jz L3 144 cmpb %bl,(%ecx) 145 jne L3 146 147 incl %eax 148 incl %ecx 149 decl %edx 150 jz L4 151 movb (%eax),%bl 152 testb %bl,%bl 153 jz L3 154 cmpb %bl,(%ecx) 155 je L1 156 157 .align 2,0x90 158L3: movzbl (%eax),%eax /* unsigned comparison */ 159 movzbl (%ecx),%ecx 160 subl %ecx,%eax 161 popl %ebx 162 ret 163 .align 2,0x90 164L4: xorl %eax,%eax 165 popl %ebx 166 ret
| 33 34/* 35 * strncmp(s1, s2, n) 36 * return an integer greater than, equal to, or less than 0, 37 * according as the first n characters of string s1 is greater 38 * than, equal to, or less than the string s2. 39 * 40 * %eax - pointer to s1 41 * %ecx - pointer to s2 42 * %edx - length 43 * 44 * Written by: 45 * J.T. Conklin (jtc@wimsey.com), Winning Strategies, Inc. 46 */ 47 48/* 49 * I've unrolled the loop eight times: large enough to make a 50 * significant difference, and small enough not to totally trash the 51 * cache. 52 * 53 * TODO: change all the jz's back to je for consistency. 54 */ 55 56ENTRY(strncmp) 57 pushl %ebx 58 movl 8(%esp),%eax 59 movl 12(%esp),%ecx 60 movl 16(%esp),%edx 61 testl %edx,%edx 62 jmp L2 /* Jump into the loop! */ 63 64 .align 2,0x90 65L1: incl %eax 66 incl %ecx 67 decl %edx 68L2: jz L4 /* strings are equal */ 69 movb (%eax),%bl 70 testb %bl,%bl 71 jz L3 72 cmpb %bl,(%ecx) 73 jne L3 74 75/* 76 * XXX it might be best to move the next 4 instructions to the end of the 77 * unrolled part of the loop. The unrolled part would then be 78 * movb n(%eax),%bl; testb %bl, %bl; je L3; cmpb n(%ecx); jne L3 79 * or maybe better 80 * movb n(%eax),%bl; cmpb n(%ecx); jne L3; testb %bl,%bl; je return_0 81 * for n = 0, 1, ..., 8. The end of the loop would be 82 * L1: addl $8,%eax; addl $8,%ecx; subl $8,%edx; cmpl $8,%edx; jae Lx 83 * where residual counts of 0 to 7 are handled at Lx. However, this would 84 * be slower for short strings. Cache effects are probably not so 85 * important because we are only handling a byte at a time. 86 */ 87 incl %eax 88 incl %ecx 89 decl %edx 90 jz L4 91 movb (%eax),%bl 92 testb %bl,%bl 93 jz L3 94 cmpb %bl,(%ecx) 95 jne L3 96 97 incl %eax 98 incl %ecx 99 decl %edx 100 jz L4 101 movb (%eax),%bl 102 testb %bl,%bl 103 jz L3 104 cmpb %bl,(%ecx) 105 jne L3 106 107 incl %eax 108 incl %ecx 109 decl %edx 110 jz L4 111 movb (%eax),%bl 112 testb %bl,%bl 113 jz L3 114 cmpb %bl,(%ecx) 115 jne L3 116 117 incl %eax 118 incl %ecx 119 decl %edx 120 jz L4 121 movb (%eax),%bl 122 testb %bl,%bl 123 jz L3 124 cmpb %bl,(%ecx) 125 jne L3 126 127 incl %eax 128 incl %ecx 129 decl %edx 130 jz L4 131 movb (%eax),%bl 132 testb %bl,%bl 133 jz L3 134 cmpb %bl,(%ecx) 135 jne L3 136 137 incl %eax 138 incl %ecx 139 decl %edx 140 jz L4 141 movb (%eax),%bl 142 testb %bl,%bl 143 jz L3 144 cmpb %bl,(%ecx) 145 jne L3 146 147 incl %eax 148 incl %ecx 149 decl %edx 150 jz L4 151 movb (%eax),%bl 152 testb %bl,%bl 153 jz L3 154 cmpb %bl,(%ecx) 155 je L1 156 157 .align 2,0x90 158L3: movzbl (%eax),%eax /* unsigned comparison */ 159 movzbl (%ecx),%ecx 160 subl %ecx,%eax 161 popl %ebx 162 ret 163 .align 2,0x90 164L4: xorl %eax,%eax 165 popl %ebx 166 ret
|