aboutsummaryrefslogtreecommitdiff
path: root/src/liblzma/check/crc64_x86.S
blob: 47f608181ea81a276dacf988cd2cbcf8c9a96da0 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
/* SPDX-License-Identifier: 0BSD */

/*
 * Speed-optimized CRC64 using slicing-by-four algorithm
 *
 * This uses only i386 instructions, but it is optimized for i686 and later
 * (including e.g. Pentium II/III/IV, Athlon XP, and Core 2).
 *
 * Authors: Igor Pavlov (original CRC32 assembly code)
 *          Lasse Collin (CRC64 adaptation of the modified CRC32 code)
 *
 * This code needs lzma_crc64_table, which can be created using the
 * following C code:

uint64_t lzma_crc64_table[4][256];

void
init_table(void)
{
	// ECMA-182
	static const uint64_t poly64 = UINT64_C(0xC96C5795D7870F42);

	for (size_t s = 0; s < 4; ++s) {
		for (size_t b = 0; b < 256; ++b) {
			uint64_t r = s == 0 ? b : lzma_crc64_table[s - 1][b];

			for (size_t i = 0; i < 8; ++i) {
				if (r & 1)
					r = (r >> 1) ^ poly64;
				else
					r >>= 1;
			}

			lzma_crc64_table[s][b] = r;
		}
	}
}

 * The prototype of the CRC64 function:
 * extern uint64_t lzma_crc64(const uint8_t *buf, size_t size, uint64_t crc);
 */

/* When Intel CET is enabled, include <cet.h> in assembly code to mark
   Intel CET support.  */
#ifdef __CET__
# include <cet.h>
#else
# define _CET_ENDBR
#endif

/*
 * On some systems, the functions need to be prefixed. The prefix is
 * usually an underscore.
 */
#ifndef __USER_LABEL_PREFIX__
#	define __USER_LABEL_PREFIX__
#endif
#define MAKE_SYM_CAT(prefix, sym) prefix ## sym
#define MAKE_SYM(prefix, sym) MAKE_SYM_CAT(prefix, sym)
#define LZMA_CRC64 MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc64)
#define LZMA_CRC64_TABLE MAKE_SYM(__USER_LABEL_PREFIX__, lzma_crc64_table)

/*
 * Solaris assembler doesn't have .p2align, and Darwin uses .align
 * differently than GNU/Linux and Solaris.
 */
#if defined(__APPLE__) || defined(__MSDOS__)
#	define ALIGN(pow2, abs) .align pow2
#else
#	define ALIGN(pow2, abs) .align abs
#endif

	.text
	.globl	LZMA_CRC64

#if !defined(__APPLE__) && !defined(_WIN32) && !defined(__CYGWIN__) \
		&& !defined(__MSDOS__)
	.type	LZMA_CRC64, @function
#endif

	ALIGN(4, 16)
LZMA_CRC64:
	_CET_ENDBR
	/*
	 * Register usage:
	 * %eax crc LSB
	 * %edx crc MSB
	 * %esi buf
	 * %edi size or buf + size
	 * %ebx lzma_crc64_table
	 * %ebp Table index
	 * %ecx Temporary
	 */
	pushl	%ebx
	pushl	%esi
	pushl	%edi
	pushl	%ebp
	movl	0x14(%esp), %esi /* buf */
	movl	0x18(%esp), %edi /* size */
	movl	0x1C(%esp), %eax /* crc LSB */
	movl	0x20(%esp), %edx /* crc MSB */

	/*
	 * Store the address of lzma_crc64_table to %ebx. This is needed to
	 * get position-independent code (PIC).
	 *
	 * The PIC macro is defined by libtool, while __PIC__ is defined
	 * by GCC but only on some systems. Testing for both makes it simpler
	 * to test this code without libtool, and keeps the code working also
	 * when built with libtool but using something else than GCC.
	 *
	 * I understood that libtool may define PIC on Windows even though
	 * the code in Windows DLLs is not PIC in sense that it is in ELF
	 * binaries, so we need a separate check to always use the non-PIC
	 * code on Windows.
	 */
#if (!defined(PIC) && !defined(__PIC__)) \
		|| (defined(_WIN32) || defined(__CYGWIN__))
	/* Not PIC */
	movl	$ LZMA_CRC64_TABLE, %ebx
#elif defined(__APPLE__)
	/* Mach-O */
	call	.L_get_pc
.L_pic:
	leal	.L_lzma_crc64_table$non_lazy_ptr-.L_pic(%ebx), %ebx
	movl	(%ebx), %ebx
#else
	/* ELF */
	call	.L_get_pc
	addl	$_GLOBAL_OFFSET_TABLE_, %ebx
	movl	LZMA_CRC64_TABLE@GOT(%ebx), %ebx
#endif

	/* Complement the initial value. */
	notl	%eax
	notl	%edx

.L_align:
	/*
	 * Check if there is enough input to use slicing-by-four.
	 * We need eight bytes, because the loop pre-reads four bytes.
	 */
	cmpl	$8, %edi
	jb	.L_rest

	/* Check if we have reached alignment of four bytes. */
	testl	$3, %esi
	jz	.L_slice

	/* Calculate CRC of the next input byte. */
	movzbl	(%esi), %ebp
	incl	%esi
	movzbl	%al, %ecx
	xorl	%ecx, %ebp
	shrdl	$8, %edx, %eax
	xorl	(%ebx, %ebp, 8), %eax
	shrl	$8, %edx
	xorl	4(%ebx, %ebp, 8), %edx
	decl	%edi
	jmp	.L_align

.L_slice:
	/*
	 * If we get here, there's at least eight bytes of aligned input
	 * available. Make %edi multiple of four bytes. Store the possible
	 * remainder over the "size" variable in the argument stack.
	 */
	movl	%edi, 0x18(%esp)
	andl	$-4, %edi
	subl	%edi, 0x18(%esp)

	/*
	 * Let %edi be buf + size - 4 while running the main loop. This way
	 * we can compare for equality to determine when exit the loop.
	 */
	addl	%esi, %edi
	subl	$4, %edi

	/* Read in the first four aligned bytes. */
	movl	(%esi), %ecx

.L_loop:
	xorl	%eax, %ecx
	movzbl	%cl, %ebp
	movl	0x1800(%ebx, %ebp, 8), %eax
	xorl	%edx, %eax
	movl	0x1804(%ebx, %ebp, 8), %edx
	movzbl	%ch, %ebp
	xorl	0x1000(%ebx, %ebp, 8), %eax
	xorl	0x1004(%ebx, %ebp, 8), %edx
	shrl	$16, %ecx
	movzbl	%cl, %ebp
	xorl	0x0800(%ebx, %ebp, 8), %eax
	xorl	0x0804(%ebx, %ebp, 8), %edx
	movzbl	%ch, %ebp
	addl	$4, %esi
	xorl	(%ebx, %ebp, 8), %eax
	xorl	4(%ebx, %ebp, 8), %edx

	/* Check for end of aligned input. */
	cmpl	%edi, %esi

	/*
	 * Copy the next input byte to %ecx. It is slightly faster to
	 * read it here than at the top of the loop.
	 */
	movl	(%esi), %ecx
	jb	.L_loop

	/*
	 * Process the remaining four bytes, which we have already
	 * copied to %ecx.
	 */
	xorl	%eax, %ecx
	movzbl	%cl, %ebp
	movl	0x1800(%ebx, %ebp, 8), %eax
	xorl	%edx, %eax
	movl	0x1804(%ebx, %ebp, 8), %edx
	movzbl	%ch, %ebp
	xorl	0x1000(%ebx, %ebp, 8), %eax
	xorl	0x1004(%ebx, %ebp, 8), %edx
	shrl	$16, %ecx
	movzbl	%cl, %ebp
	xorl	0x0800(%ebx, %ebp, 8), %eax
	xorl	0x0804(%ebx, %ebp, 8), %edx
	movzbl	%ch, %ebp
	addl	$4, %esi
	xorl	(%ebx, %ebp, 8), %eax
	xorl	4(%ebx, %ebp, 8), %edx

	/* Copy the number of remaining bytes to %edi. */
	movl	0x18(%esp), %edi

.L_rest:
	/* Check for end of input. */
	testl	%edi, %edi
	jz	.L_return

	/* Calculate CRC of the next input byte. */
	movzbl	(%esi), %ebp
	incl	%esi
	movzbl	%al, %ecx
	xorl	%ecx, %ebp
	shrdl	$8, %edx, %eax
	xorl	(%ebx, %ebp, 8), %eax
	shrl	$8, %edx
	xorl	4(%ebx, %ebp, 8), %edx
	decl	%edi
	jmp	.L_rest

.L_return:
	/* Complement the final value. */
	notl	%eax
	notl	%edx

	popl	%ebp
	popl	%edi
	popl	%esi
	popl	%ebx
	ret

#if defined(PIC) || defined(__PIC__)
	ALIGN(4, 16)
.L_get_pc:
	movl	(%esp), %ebx
	ret
#endif

#if defined(__APPLE__) && (defined(PIC) || defined(__PIC__))
	/* Mach-O PIC */
	.section __IMPORT,__pointers,non_lazy_symbol_pointers
.L_lzma_crc64_table$non_lazy_ptr:
	.indirect_symbol LZMA_CRC64_TABLE
	.long 0

#elif defined(_WIN32) || defined(__CYGWIN__)
#	ifdef DLL_EXPORT
	/* This is equivalent of __declspec(dllexport). */
	.section .drectve
	.ascii " -export:lzma_crc64"
#	endif

#elif !defined(__MSDOS__)
	/* ELF */
	.size	LZMA_CRC64, .-LZMA_CRC64
#endif

/*
 * This is needed to support non-executable stack. It's ugly to
 * use __FreeBSD__ and __linux__ here, but I don't know a way to detect when
 * we are using GNU assembler.
 */
#if defined(__ELF__) && (defined(__FreeBSD__) || defined(__linux__))
	.section	.note.GNU-stack,"",@progbits
#endif