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
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
|
/**************************************************************************
*
* Copyright 2010 VMware, Inc. All Rights Reserved.
*
* 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, sub license, 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 (including the
* next paragraph) 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 NON-INFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS AND/OR ITS SUPPLIERS 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.
*
**************************************************************************/
/**
* Code to convert images from tiled to linear and back.
* XXX there are quite a few assumptions about color and z/stencil being
* 32bpp.
*/
#include "util/u_format.h"
#include "lp_tile_soa.h"
#include "lp_tile_image.h"
#define BYTES_PER_TILE (TILE_SIZE * TILE_SIZE * 4)
/**
* Untile a 4x4 block of 32-bit words (all contiguous) to linear layout
* at dst, with dst_stride words between rows.
*/
static void
untile_4_4_uint32(const uint32_t *src, uint32_t *dst, unsigned dst_stride)
{
uint32_t *d0 = dst;
uint32_t *d1 = d0 + dst_stride;
uint32_t *d2 = d1 + dst_stride;
uint32_t *d3 = d2 + dst_stride;
d0[0] = src[0]; d0[1] = src[1]; d0[2] = src[4]; d0[3] = src[5];
d1[0] = src[2]; d1[1] = src[3]; d1[2] = src[6]; d1[3] = src[7];
d2[0] = src[8]; d2[1] = src[9]; d2[2] = src[12]; d2[3] = src[13];
d3[0] = src[10]; d3[1] = src[11]; d3[2] = src[14]; d3[3] = src[15];
}
/**
* Untile a 4x4 block of 16-bit words (all contiguous) to linear layout
* at dst, with dst_stride words between rows.
*/
static void
untile_4_4_uint16(const uint16_t *src, uint16_t *dst, unsigned dst_stride)
{
uint16_t *d0 = dst;
uint16_t *d1 = d0 + dst_stride;
uint16_t *d2 = d1 + dst_stride;
uint16_t *d3 = d2 + dst_stride;
d0[0] = src[0]; d0[1] = src[1]; d0[2] = src[4]; d0[3] = src[5];
d1[0] = src[2]; d1[1] = src[3]; d1[2] = src[6]; d1[3] = src[7];
d2[0] = src[8]; d2[1] = src[9]; d2[2] = src[12]; d2[3] = src[13];
d3[0] = src[10]; d3[1] = src[11]; d3[2] = src[14]; d3[3] = src[15];
}
/**
* Convert a 4x4 rect of 32-bit words from a linear layout into tiled
* layout (in which all 16 words are contiguous).
*/
static void
tile_4_4_uint32(const uint32_t *src, uint32_t *dst, unsigned src_stride)
{
const uint32_t *s0 = src;
const uint32_t *s1 = s0 + src_stride;
const uint32_t *s2 = s1 + src_stride;
const uint32_t *s3 = s2 + src_stride;
dst[0] = s0[0]; dst[1] = s0[1]; dst[4] = s0[2]; dst[5] = s0[3];
dst[2] = s1[0]; dst[3] = s1[1]; dst[6] = s1[2]; dst[7] = s1[3];
dst[8] = s2[0]; dst[9] = s2[1]; dst[12] = s2[2]; dst[13] = s2[3];
dst[10] = s3[0]; dst[11] = s3[1]; dst[14] = s3[2]; dst[15] = s3[3];
}
/**
* Convert a 4x4 rect of 16-bit words from a linear layout into tiled
* layout (in which all 16 words are contiguous).
*/
static void
tile_4_4_uint16(const uint16_t *src, uint16_t *dst, unsigned src_stride)
{
const uint16_t *s0 = src;
const uint16_t *s1 = s0 + src_stride;
const uint16_t *s2 = s1 + src_stride;
const uint16_t *s3 = s2 + src_stride;
dst[0] = s0[0]; dst[1] = s0[1]; dst[4] = s0[2]; dst[5] = s0[3];
dst[2] = s1[0]; dst[3] = s1[1]; dst[6] = s1[2]; dst[7] = s1[3];
dst[8] = s2[0]; dst[9] = s2[1]; dst[12] = s2[2]; dst[13] = s2[3];
dst[10] = s3[0]; dst[11] = s3[1]; dst[14] = s3[2]; dst[15] = s3[3];
}
/**
* Convert a tiled image into a linear image.
* \param dst_stride dest row stride in bytes
*/
void
lp_tiled_to_linear(const void *src, void *dst,
unsigned x, unsigned y,
unsigned width, unsigned height,
enum pipe_format format,
unsigned dst_stride,
unsigned tiles_per_row)
{
assert(x % TILE_SIZE == 0);
assert(y % TILE_SIZE == 0);
/*assert(width % TILE_SIZE == 0);
assert(height % TILE_SIZE == 0);*/
/* Note that Z/stencil surfaces use a different tiling size than
* color surfaces.
*/
if (util_format_is_depth_or_stencil(format)) {
const uint bpp = util_format_get_blocksize(format);
const uint tile_w = TILE_VECTOR_WIDTH, tile_h = TILE_VECTOR_HEIGHT;
dst_stride /= bpp; /* convert from bytes to words */
if (bpp == 4) {
const uint32_t *src32 = (const uint32_t *) src;
uint32_t *dst32 = (uint32_t *) dst;
uint i, j;
for (j = 0; j < height; j += tile_h) {
for (i = 0; i < width; i += tile_w) {
/* compute offsets in 32-bit words */
uint ii = i + x, jj = j + y;
uint src_offset = (jj / tile_h * tiles_per_row + ii / tile_w)
* (tile_w * tile_h);
uint dst_offset = jj * dst_stride + ii;
untile_4_4_uint32(src32 + src_offset,
dst32 + dst_offset,
dst_stride);
}
}
}
else {
const uint16_t *src16 = (const uint16_t *) src;
uint16_t *dst16 = (uint16_t *) dst;
uint i, j;
assert(bpp == 2);
for (j = 0; j < height; j += tile_h) {
for (i = 0; i < width; i += tile_w) {
/* compute offsets in 16-bit words */
uint ii = i + x, jj = j + y;
uint src_offset = (jj / tile_h * tiles_per_row + ii / tile_w)
* (tile_w * tile_h);
uint dst_offset = jj * dst_stride + ii;
untile_4_4_uint16(src16 + src_offset,
dst16 + dst_offset,
dst_stride);
}
}
}
}
else {
/* color image */
const uint bpp = 4;
const uint tile_w = TILE_SIZE, tile_h = TILE_SIZE;
const uint bytes_per_tile = tile_w * tile_h * bpp;
uint i, j;
for (j = 0; j < height; j += tile_h) {
for (i = 0; i < width; i += tile_w) {
uint ii = i + x, jj = j + y;
uint tile_offset = ((jj / tile_h) * tiles_per_row + ii / tile_w);
uint byte_offset = tile_offset * bytes_per_tile;
const uint8_t *src_tile = (uint8_t *) src + byte_offset;
lp_tile_write_4ub(format,
src_tile,
dst, dst_stride,
ii, jj, tile_w, tile_h);
}
}
}
}
/**
* Convert a linear image into a tiled image.
* \param src_stride source row stride in bytes
*/
void
lp_linear_to_tiled(const void *src, void *dst,
unsigned x, unsigned y,
unsigned width, unsigned height,
enum pipe_format format,
unsigned src_stride,
unsigned tiles_per_row)
{
assert(x % TILE_SIZE == 0);
assert(y % TILE_SIZE == 0);
/*
assert(width % TILE_SIZE == 0);
assert(height % TILE_SIZE == 0);
*/
if (util_format_is_depth_or_stencil(format)) {
const uint bpp = util_format_get_blocksize(format);
const uint tile_w = TILE_VECTOR_WIDTH, tile_h = TILE_VECTOR_HEIGHT;
src_stride /= bpp; /* convert from bytes to words */
if (bpp == 4) {
const uint32_t *src32 = (const uint32_t *) src;
uint32_t *dst32 = (uint32_t *) dst;
uint i, j;
for (j = 0; j < height; j += tile_h) {
for (i = 0; i < width; i += tile_w) {
/* compute offsets in 32-bit words */
uint ii = i + x, jj = j + y;
uint src_offset = jj * src_stride + ii;
uint dst_offset = (jj / tile_h * tiles_per_row + ii / tile_w)
* (tile_w * tile_h);
tile_4_4_uint32(src32 + src_offset,
dst32 + dst_offset,
src_stride);
}
}
}
else {
const uint16_t *src16 = (const uint16_t *) src;
uint16_t *dst16 = (uint16_t *) dst;
uint i, j;
assert(bpp == 2);
for (j = 0; j < height; j += tile_h) {
for (i = 0; i < width; i += tile_w) {
/* compute offsets in 16-bit words */
uint ii = i + x, jj = j + y;
uint src_offset = jj * src_stride + ii;
uint dst_offset = (jj / tile_h * tiles_per_row + ii / tile_w)
* (tile_w * tile_h);
tile_4_4_uint16(src16 + src_offset,
dst16 + dst_offset,
src_stride);
}
}
}
}
else {
const uint bpp = 4;
const uint tile_w = TILE_SIZE, tile_h = TILE_SIZE;
const uint bytes_per_tile = tile_w * tile_h * bpp;
uint i, j;
for (j = 0; j < height; j += TILE_SIZE) {
for (i = 0; i < width; i += TILE_SIZE) {
uint ii = i + x, jj = j + y;
uint tile_offset = ((jj / tile_h) * tiles_per_row + ii / tile_w);
uint byte_offset = tile_offset * bytes_per_tile;
uint8_t *dst_tile = (uint8_t *) dst + byte_offset;
lp_tile_read_4ub(format,
dst_tile,
src, src_stride,
ii, jj, tile_w, tile_h);
}
}
}
}
/**
* For testing only.
*/
void
test_tiled_linear_conversion(void *data,
enum pipe_format format,
unsigned width, unsigned height,
unsigned stride)
{
/* size in tiles */
unsigned wt = (width + TILE_SIZE - 1) / TILE_SIZE;
unsigned ht = (height + TILE_SIZE - 1) / TILE_SIZE;
uint8_t *tiled = malloc(wt * ht * TILE_SIZE * TILE_SIZE * 4);
/*unsigned tiled_stride = wt * TILE_SIZE * TILE_SIZE * 4;*/
lp_linear_to_tiled(data, tiled, 0, 0, width, height, format,
stride, wt);
lp_tiled_to_linear(tiled, data, 0, 0, width, height, format,
stride, wt);
free(tiled);
}
|