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
|
#include "texobj.h"
#include "intel_context.h"
#include "intel_mipmap_tree.h"
#include "intel_tex.h"
#define FILE_DEBUG_FLAG DEBUG_TEXTURE
static GLboolean
intelIsTextureResident(GLcontext * ctx, struct gl_texture_object *texObj)
{
#if 0
struct intel_context *intel = intel_context(ctx);
struct intel_texture_object *intelObj = intel_texture_object(texObj);
return
intelObj->mt &&
intelObj->mt->region &&
intel_is_region_resident(intel, intelObj->mt->region);
#endif
return 1;
}
static struct gl_texture_image *
intelNewTextureImage(GLcontext * ctx)
{
DBG("%s\n", __FUNCTION__);
(void) ctx;
return (struct gl_texture_image *) CALLOC_STRUCT(intel_texture_image);
}
static struct gl_texture_object *
intelNewTextureObject(GLcontext * ctx, GLuint name, GLenum target)
{
struct intel_texture_object *obj = CALLOC_STRUCT(intel_texture_object);
DBG("%s\n", __FUNCTION__);
_mesa_initialize_texture_object(&obj->base, name, target);
return &obj->base;
}
static void
intelDeleteTextureObject(GLcontext *ctx,
struct gl_texture_object *texObj)
{
struct intel_context *intel = intel_context(ctx);
struct intel_texture_object *intelObj = intel_texture_object(texObj);
if (intelObj->mt)
intel_miptree_release(intel, &intelObj->mt);
_mesa_delete_texture_object(ctx, texObj);
}
static void
intelFreeTextureImageData(GLcontext * ctx, struct gl_texture_image *texImage)
{
struct intel_context *intel = intel_context(ctx);
struct intel_texture_image *intelImage = intel_texture_image(texImage);
DBG("%s\n", __FUNCTION__);
if (intelImage->mt) {
intel_miptree_release(intel, &intelImage->mt);
}
if (texImage->Data) {
free(texImage->Data);
texImage->Data = NULL;
}
}
/* ================================================================
* From linux kernel i386 header files, copes with odd sizes better
* than COPY_DWORDS would:
* XXX Put this in src/mesa/main/imports.h ???
*/
#if defined(i386) || defined(__i386__)
static INLINE void *
__memcpy(void *to, const void *from, size_t n)
{
int d0, d1, d2;
__asm__ __volatile__("rep ; movsl\n\t"
"testb $2,%b4\n\t"
"je 1f\n\t"
"movsw\n"
"1:\ttestb $1,%b4\n\t"
"je 2f\n\t"
"movsb\n" "2:":"=&c"(d0), "=&D"(d1), "=&S"(d2)
:"0"(n / 4), "q"(n), "1"((long) to), "2"((long) from)
:"memory");
return (to);
}
#else
#define __memcpy(a,b,c) memcpy(a,b,c)
#endif
/* The system memcpy (at least on ubuntu 5.10) has problems copying
* to agp (writecombined) memory from a source which isn't 64-byte
* aligned - there is a 4x performance falloff.
*
* The x86 __memcpy is immune to this but is slightly slower
* (10%-ish) than the system memcpy.
*
* The sse_memcpy seems to have a slight cliff at 64/32 bytes, but
* isn't much faster than x86_memcpy for agp copies.
*
* TODO: switch dynamically.
*/
static void *
do_memcpy(void *dest, const void *src, size_t n)
{
if ((((unsigned) src) & 63) || (((unsigned) dest) & 63)) {
return __memcpy(dest, src, n);
}
else
return memcpy(dest, src, n);
}
#if DO_DEBUG
#ifndef __x86_64__
static unsigned
fastrdtsc(void)
{
unsigned eax;
__asm__ volatile ("\t"
"pushl %%ebx\n\t"
"cpuid\n\t" ".byte 0x0f, 0x31\n\t"
"popl %%ebx\n":"=a" (eax)
:"0"(0)
:"ecx", "edx", "cc");
return eax;
}
#else
static unsigned
fastrdtsc(void)
{
unsigned eax;
__asm__ volatile ("\t" "cpuid\n\t" ".byte 0x0f, 0x31\n\t":"=a" (eax)
:"0"(0)
:"ecx", "edx", "ebx", "cc");
return eax;
}
#endif
static unsigned
time_diff(unsigned t, unsigned t2)
{
return ((t < t2) ? t2 - t : 0xFFFFFFFFU - (t - t2 - 1));
}
static void *
timed_memcpy(void *dest, const void *src, size_t n)
{
void *ret;
unsigned t1, t2;
double rate;
if ((((unsigned) src) & 63) || (((unsigned) dest) & 63))
_mesa_printf("Warning - non-aligned texture copy!\n");
t1 = fastrdtsc();
ret = do_memcpy(dest, src, n);
t2 = fastrdtsc();
rate = time_diff(t1, t2);
rate /= (double) n;
_mesa_printf("timed_memcpy: %u %u --> %f clocks/byte\n", t1, t2, rate);
return ret;
}
#endif /* DO_DEBUG */
void
intelInitTextureFuncs(struct dd_function_table *functions)
{
functions->ChooseTextureFormat = intelChooseTextureFormat;
functions->TexImage1D = intelTexImage1D;
functions->TexImage2D = intelTexImage2D;
functions->TexImage3D = intelTexImage3D;
functions->TexSubImage1D = intelTexSubImage1D;
functions->TexSubImage2D = intelTexSubImage2D;
functions->TexSubImage3D = intelTexSubImage3D;
functions->CopyTexImage1D = intelCopyTexImage1D;
functions->CopyTexImage2D = intelCopyTexImage2D;
functions->CopyTexSubImage1D = intelCopyTexSubImage1D;
functions->CopyTexSubImage2D = intelCopyTexSubImage2D;
functions->GetTexImage = intelGetTexImage;
/* compressed texture functions */
functions->CompressedTexImage2D = intelCompressedTexImage2D;
functions->GetCompressedTexImage = intelGetCompressedTexImage;
functions->NewTextureObject = intelNewTextureObject;
functions->NewTextureImage = intelNewTextureImage;
functions->DeleteTexture = intelDeleteTextureObject;
functions->FreeTexImageData = intelFreeTextureImageData;
functions->UpdateTexturePalette = 0;
functions->IsTextureResident = intelIsTextureResident;
#if DO_DEBUG
if (INTEL_DEBUG & DEBUG_BUFMGR)
functions->TextureMemCpy = timed_memcpy;
else
#endif
functions->TextureMemCpy = do_memcpy;
}
|
