1 /* SPDX-License-Identifier: GPL-2.0 */ !! 1 /* 2 /* linux/arch/sparc/lib/memset.S: Sparc optimi !! 2 * This file is subject to the terms and conditions of the GNU General Public 3 * Copyright (C) 1991,1996 Free Software Found !! 3 * License. See the file "COPYING" in the main directory of this archive 4 * Copyright (C) 1996,1997 Jakub Jelinek (jj@s !! 4 * for more details. 5 * Copyright (C) 1996 David S. Miller (davem@c << 6 * 5 * 7 * Calls to memset returns initial %o0. Calls !! 6 * Copyright (C) 1998 by Ralf Baechle 8 * number of bytes not yet set if exception oc << 9 * clear_user. << 10 */ 7 */ >> 8 #include <asm/asm.h> >> 9 #include <asm/offset.h> >> 10 #include <asm/regdef.h> >> 11 >> 12 #define EX(insn,reg,addr,handler) \ >> 13 9: insn reg, addr; \ >> 14 .section __ex_table,"a"; \ >> 15 PTR 9b, handler; \ >> 16 .previous >> 17 >> 18 #define F_FILL64(dst, offset, val, fixup) \ >> 19 EX(sw, val, (offset + 0x00)(dst), fixup); \ >> 20 EX(sw, val, (offset + 0x04)(dst), fixup); \ >> 21 EX(sw, val, (offset + 0x08)(dst), fixup); \ >> 22 EX(sw, val, (offset + 0x0c)(dst), fixup); \ >> 23 EX(sw, val, (offset + 0x10)(dst), fixup); \ >> 24 EX(sw, val, (offset + 0x14)(dst), fixup); \ >> 25 EX(sw, val, (offset + 0x18)(dst), fixup); \ >> 26 EX(sw, val, (offset + 0x1c)(dst), fixup); \ >> 27 EX(sw, val, (offset + 0x20)(dst), fixup); \ >> 28 EX(sw, val, (offset + 0x24)(dst), fixup); \ >> 29 EX(sw, val, (offset + 0x28)(dst), fixup); \ >> 30 EX(sw, val, (offset + 0x2c)(dst), fixup); \ >> 31 EX(sw, val, (offset + 0x30)(dst), fixup); \ >> 32 EX(sw, val, (offset + 0x34)(dst), fixup); \ >> 33 EX(sw, val, (offset + 0x38)(dst), fixup); \ >> 34 EX(sw, val, (offset + 0x3c)(dst), fixup) 11 35 12 #include <linux/export.h> !! 36 /* 13 #include <asm/ptrace.h> !! 37 * memset(void *s, int c, size_t n) 14 !! 38 * 15 /* Work around cpp -rob */ !! 39 * a0: start of area to clear 16 #define ALLOC #alloc !! 40 * a1: char to fill with 17 #define EXECINSTR #execinstr !! 41 * a2: size of area to clear 18 #define EX(x,y,a,b) !! 42 */ 19 98: x,y; !! 43 .set noreorder 20 .section .fixup,ALLOC,EXECINSTR; !! 44 .align 5 21 .align 4; !! 45 LEAF(memset) 22 99: retl; !! 46 beqz a1, 1f 23 a, b, %o0; !! 47 move v0, a0 /* result */ 24 .section __ex_table,ALLOC; !! 48 25 .align 4; !! 49 andi a1, 0xff /* spread fillword */ 26 .word 98b, 99b; !! 50 sll t1, a1, 8 27 .text; !! 51 or a1, t1 28 .align 4 !! 52 sll t1, a1, 16 29 !! 53 or a1, t1 30 #define STORE(source, base, offset, n) << 31 98: std source, [base + offset + n]; << 32 .section .fixup,ALLOC,EXECINSTR; << 33 .align 4; << 34 99: ba 30f; << 35 sub %o3, n - offset, %o3; << 36 .section __ex_table,ALLOC; << 37 .align 4; << 38 .word 98b, 99b; << 39 .text; << 40 .align 4; << 41 << 42 #define STORE_LAST(source, base, offset, n) << 43 EX(std source, [base - offset - n], << 44 add %o1, offset + n); << 45 << 46 /* Please don't change these macros, unless yo << 47 * in the .fixup section below as well. << 48 * Store 64 bytes at (BASE + OFFSET) using val << 49 #define ZERO_BIG_BLOCK(base, offset, source) << 50 STORE(source, base, offset, 0x00); << 51 STORE(source, base, offset, 0x08); << 52 STORE(source, base, offset, 0x10); << 53 STORE(source, base, offset, 0x18); << 54 STORE(source, base, offset, 0x20); << 55 STORE(source, base, offset, 0x28); << 56 STORE(source, base, offset, 0x30); << 57 STORE(source, base, offset, 0x38); << 58 << 59 #define ZERO_LAST_BLOCKS(base, offset, source) << 60 STORE_LAST(source, base, offset, 0x38) << 61 STORE_LAST(source, base, offset, 0x30) << 62 STORE_LAST(source, base, offset, 0x28) << 63 STORE_LAST(source, base, offset, 0x20) << 64 STORE_LAST(source, base, offset, 0x18) << 65 STORE_LAST(source, base, offset, 0x10) << 66 STORE_LAST(source, base, offset, 0x08) << 67 STORE_LAST(source, base, offset, 0x00) << 68 << 69 .text << 70 .align 4 << 71 << 72 .globl __bzero_begin << 73 __bzero_begin: << 74 << 75 .globl __bzero << 76 .type __bzero,#function << 77 .globl memset << 78 EXPORT_SYMBOL(__bzero) << 79 EXPORT_SYMBOL(memset) << 80 memset: << 81 mov %o0, %g1 << 82 mov 1, %g4 << 83 and %o1, 0xff, %g3 << 84 sll %g3, 8, %g2 << 85 or %g3, %g2, %g3 << 86 sll %g3, 16, %g2 << 87 or %g3, %g2, %g3 << 88 b 1f << 89 mov %o2, %o1 << 90 3: << 91 cmp %o2, 3 << 92 be 2f << 93 EX(stb %g3, [%o0], sub %o1, 0) << 94 << 95 cmp %o2, 2 << 96 be 2f << 97 EX(stb %g3, [%o0 + 0x01], sub %o1, 1) << 98 << 99 EX(stb %g3, [%o0 + 0x02], sub %o1, 2) << 100 2: << 101 sub %o2, 4, %o2 << 102 add %o1, %o2, %o1 << 103 b 4f << 104 sub %o0, %o2, %o0 << 105 << 106 __bzero: << 107 clr %g4 << 108 mov %g0, %g3 << 109 1: 54 1: 110 cmp %o1, 7 << 111 bleu 7f << 112 andcc %o0, 3, %o2 << 113 << 114 bne 3b << 115 4: << 116 andcc %o0, 4, %g0 << 117 << 118 be 2f << 119 mov %g3, %g2 << 120 << 121 EX(st %g3, [%o0], sub %o1, 0) << 122 sub %o1, 4, %o1 << 123 add %o0, 4, %o0 << 124 2: << 125 andcc %o1, 0xffffff80, %o3 ! Now << 126 be 9f << 127 andcc %o1, 0x78, %o2 << 128 10: << 129 ZERO_BIG_BLOCK(%o0, 0x00, %g2) << 130 subcc %o3, 128, %o3 << 131 ZERO_BIG_BLOCK(%o0, 0x40, %g2) << 132 bne 10b << 133 add %o0, 128, %o0 << 134 << 135 orcc %o2, %g0, %g0 << 136 9: << 137 be 13f << 138 andcc %o1, 7, %o1 << 139 << 140 srl %o2, 1, %o3 << 141 set 13f, %o4 << 142 sub %o4, %o3, %o4 << 143 jmp %o4 << 144 add %o0, %o2, %o0 << 145 << 146 ZERO_LAST_BLOCKS(%o0, 0x48, %g2) << 147 ZERO_LAST_BLOCKS(%o0, 0x08, %g2) << 148 13: << 149 be 8f << 150 andcc %o1, 4, %g0 << 151 55 152 be 1f !! 56 EXPORT(__bzero) 153 andcc %o1, 2, %g0 !! 57 sltiu t0, a2, 4 /* very small region? */ >> 58 bnez t0, small_memset >> 59 andi t0, a0, 3 /* aligned? */ >> 60 >> 61 beqz t0, 1f >> 62 subu t0, 4 /* alignment in bytes */ >> 63 >> 64 #ifdef __MIPSEB__ >> 65 EX(swl, a1, (a0), first_fixup) /* make word aligned */ >> 66 #endif >> 67 #ifdef __MIPSEL__ >> 68 EX(swr, a1, (a0), first_fixup) /* make word aligned */ >> 69 #endif >> 70 subu a0, t0 /* word align ptr */ >> 71 addu a2, t0 /* correct size */ >> 72 >> 73 1: ori t1, a2, 0x3f /* # of full blocks */ >> 74 xori t1, 0x3f >> 75 beqz t1, memset_partial /* no block to fill */ >> 76 andi t0, a2, 0x3c >> 77 >> 78 addu t1, a0 /* end address */ >> 79 .set reorder >> 80 1: addiu a0, 64 >> 81 F_FILL64(a0, -64, a1, fwd_fixup) >> 82 bne t1, a0, 1b >> 83 .set noreorder >> 84 >> 85 memset_partial: >> 86 PTR_LA t1, 2f /* where to start */ >> 87 subu t1, t0 >> 88 jr t1 >> 89 addu a0, t0 /* dest ptr */ >> 90 >> 91 .set push >> 92 .set noreorder >> 93 .set nomacro >> 94 F_FILL64(a0, -64, a1, partial_fixup) /* ... but first do wrds ... */ >> 95 2: .set pop >> 96 andi a2, 3 /* 0 <= n <= 3 to go */ >> 97 >> 98 beqz a2, 1f >> 99 addu a0, a2 /* What's left */ >> 100 #ifdef __MIPSEB__ >> 101 EX(swr, a1, -1(a0), last_fixup) >> 102 #endif >> 103 #ifdef __MIPSEL__ >> 104 EX(swl, a1, -1(a0), last_fixup) >> 105 #endif >> 106 1: jr ra >> 107 move a2, zero >> 108 >> 109 small_memset: >> 110 beqz a2, 2f >> 111 addu t1, a0, a2 >> 112 >> 113 1: addiu a0, 1 /* fill bytewise */ >> 114 bne t1, a0, 1b >> 115 sb a1, -1(a0) >> 116 >> 117 2: jr ra /* done */ >> 118 move a2, zero >> 119 END(memset) 154 120 155 EX(st %g3, [%o0], and %o1, 7) !! 121 first_fixup: 156 add %o0, 4, %o0 !! 122 jr ra 157 1: << 158 be 1f << 159 andcc %o1, 1, %g0 << 160 << 161 EX(sth %g3, [%o0], and %o1, 3) << 162 add %o0, 2, %o0 << 163 1: << 164 bne,a 8f << 165 EX(stb %g3, [%o0], and %o1, 1) << 166 8: << 167 b 0f << 168 nop << 169 7: << 170 be 13b << 171 orcc %o1, 0, %g0 << 172 << 173 be 0f << 174 8: << 175 add %o0, 1, %o0 << 176 subcc %o1, 1, %o1 << 177 bne 8b << 178 EX(stb %g3, [%o0 - 1], add %o1, 1) << 179 0: << 180 andcc %g4, 1, %g0 << 181 be 5f << 182 nop 123 nop 183 retl << 184 mov %g1, %o0 << 185 5: << 186 retl << 187 clr %o0 << 188 << 189 .section .fixup,#alloc,#execinstr << 190 .align 4 << 191 30: << 192 and %o1, 0x7f, %o1 << 193 retl << 194 add %o3, %o1, %o0 << 195 124 196 .globl __bzero_end !! 125 fwd_fixup: 197 __bzero_end: !! 126 lw t0, THREAD_BUADDR($28) >> 127 andi a2, 0x3f >> 128 addu a2, t1 >> 129 jr ra >> 130 subu a2, t0 >> 131 >> 132 partial_fixup: >> 133 lw t0, THREAD_BUADDR($28) >> 134 andi a2, 3 >> 135 addu a2, t1 >> 136 jr ra >> 137 subu a2, t0 >> 138 >> 139 last_fixup: >> 140 jr ra >> 141 andi v1, a2, 3
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.