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
|
/*
* Simple C functions to supplement the C library
*
* Copyright (c) 2006 Fabrice Bellard
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to deal
* in the Software without restriction, including without limitation the rights
* to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
* copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
* THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
* OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
* THE SOFTWARE.
*/
#include "qemu/osdep.h"
#include "qemu-common.h"
#include "qemu/cutils.h"
#include "qemu/bswap.h"
/* vector definitions */
extern void link_error(void);
#define ACCEL_BUFFER_ZERO(NAME, SIZE, VECTYPE, NONZERO) \
static bool NAME(const void *buf, size_t len) \
{ \
const void *end = buf + len; \
do { \
const VECTYPE *p = buf; \
VECTYPE t; \
if (SIZE == sizeof(VECTYPE) * 4) { \
t = (p[0] | p[1]) | (p[2] | p[3]); \
} else if (SIZE == sizeof(VECTYPE) * 8) { \
t = p[0] | p[1]; \
t |= p[2] | p[3]; \
t |= p[4] | p[5]; \
t |= p[6] | p[7]; \
} else { \
link_error(); \
} \
if (unlikely(NONZERO(t))) { \
return false; \
} \
buf += SIZE; \
} while (buf < end); \
return true; \
}
static bool
buffer_zero_int(const void *buf, size_t len)
{
if (unlikely(len < 8)) {
/* For a very small buffer, simply accumulate all the bytes. */
const unsigned char *p = buf;
const unsigned char *e = buf + len;
unsigned char t = 0;
do {
t |= *p++;
} while (p < e);
return t == 0;
} else {
/* Otherwise, use the unaligned memory access functions to
handle the beginning and end of the buffer, with a couple
of loops handling the middle aligned section. */
uint64_t t = ldq_he_p(buf);
const uint64_t *p = (uint64_t *)(((uintptr_t)buf + 8) & -8);
const uint64_t *e = (uint64_t *)(((uintptr_t)buf + len) & -8);
for (; p + 8 <= e; p += 8) {
__builtin_prefetch(p + 8);
if (t) {
return false;
}
t = p[0] | p[1] | p[2] | p[3] | p[4] | p[5] | p[6] | p[7];
}
while (p < e) {
t |= *p++;
}
t |= ldq_he_p(buf + len - 8);
return t == 0;
}
}
#if defined(__ALTIVEC__)
#include <altivec.h>
/* The altivec.h header says we're allowed to undef these for
* C++ compatibility. Here we don't care about C++, but we
* undef them anyway to avoid namespace pollution.
* altivec.h may redefine the bool macro as vector type.
* Reset it to POSIX semantics.
*/
#undef vector
#undef pixel
#undef bool
#define bool _Bool
#define DO_NONZERO(X) vec_any_ne(X, (__vector unsigned char){ 0 })
ACCEL_BUFFER_ZERO(buffer_zero_ppc, 128, __vector unsigned char, DO_NONZERO)
static bool select_accel_fn(const void *buf, size_t len)
{
uintptr_t ibuf = (uintptr_t)buf;
if (len % 128 == 0 && ibuf % sizeof(__vector unsigned char) == 0) {
return buffer_zero_ppc(buf, len);
}
return buffer_zero_int(buf, len);
}
#elif defined(CONFIG_AVX2_OPT) || (defined(CONFIG_CPUID_H) && defined(__SSE2__))
#include <cpuid.h>
/* Do not use push_options pragmas unnecessarily, because clang
* does not support them.
*/
#ifndef __SSE2__
#pragma GCC push_options
#pragma GCC target("sse2")
#endif
#include <emmintrin.h>
#define SSE2_NONZERO(X) \
(_mm_movemask_epi8(_mm_cmpeq_epi8((X), _mm_setzero_si128())) != 0xFFFF)
ACCEL_BUFFER_ZERO(buffer_zero_sse2, 64, __m128i, SSE2_NONZERO)
#ifndef __SSE2__
#pragma GCC pop_options
#endif
#ifdef CONFIG_AVX2_OPT
#pragma GCC push_options
#pragma GCC target("avx2")
#include <immintrin.h>
#define AVX2_NONZERO(X) !_mm256_testz_si256((X), (X))
ACCEL_BUFFER_ZERO(buffer_zero_avx2, 128, __m256i, AVX2_NONZERO)
#pragma GCC pop_options
#endif
#define CACHE_AVX2 2
#define CACHE_AVX1 4
#define CACHE_SSE4 8
#define CACHE_SSE2 16
static unsigned cpuid_cache;
static void __attribute__((constructor)) init_cpuid_cache(void)
{
int max = __get_cpuid_max(0, NULL);
int a, b, c, d;
unsigned cache = 0;
if (max >= 1) {
__cpuid(1, a, b, c, d);
if (d & bit_SSE2) {
cache |= CACHE_SSE2;
}
#ifdef CONFIG_AVX2_OPT
if (c & bit_SSE4_1) {
cache |= CACHE_SSE4;
}
/* We must check that AVX is not just available, but usable. */
if ((c & bit_OSXSAVE) && (c & bit_AVX)) {
__asm("xgetbv" : "=a"(a), "=d"(d) : "c"(0));
if ((a & 6) == 6) {
cache |= CACHE_AVX1;
if (max >= 7) {
__cpuid_count(7, 0, a, b, c, d);
if (b & bit_AVX2) {
cache |= CACHE_AVX2;
}
}
}
}
#endif
}
cpuid_cache = cache;
}
static bool select_accel_fn(const void *buf, size_t len)
{
uintptr_t ibuf = (uintptr_t)buf;
#ifdef CONFIG_AVX2_OPT
if (len % 128 == 0 && ibuf % 32 == 0 && (cpuid_cache & CACHE_AVX2)) {
return buffer_zero_avx2(buf, len);
}
#endif
if (len % 64 == 0 && ibuf % 16 == 0 && (cpuid_cache & CACHE_SSE2)) {
return buffer_zero_sse2(buf, len);
}
return buffer_zero_int(buf, len);
}
#else
#define select_accel_fn buffer_zero_int
#endif
/*
* Checks if a buffer is all zeroes
*/
bool buffer_is_zero(const void *buf, size_t len)
{
if (unlikely(len == 0)) {
return true;
}
/* Use an optimized zero check if possible. Note that this also
includes a check for an unrolled loop over 64-bit integers. */
return select_accel_fn(buf, len);
}
|
