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
|
/*
* Cayman shaders
*
* Copyright (C) 2011 Advanced Micro Devices, Inc.
*
* 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
* BRIAN PAUL 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.
*/
/*
* Shader macros
*/
#ifndef __SHADER_H__
#define __SHADER_H__
#include "radeon.h"
/* Oder of instructions: All CF, All ALU, All Tex/Vtx fetches */
// CF insts
// addr
#define ADDR(x) (x)
// jumptable
#define JUMPTABLE_SEL(x) (x)
// pc
#define POP_COUNT(x) (x)
// const
#define CF_CONST(x) (x)
// cond
#define COND(x) (x) // SQ_COND_*
// count
#define I_COUNT(x) ((x) ? ((x) - 1) : 0)
// vpm
#define VALID_PIXEL_MODE(x) (x)
// cf inst
#define CF_INST(x) (x) // SQ_CF_INST_*
// wqm
#define WHOLE_QUAD_MODE(x) (x)
// barrier
#define BARRIER(x) (x)
//kb0
#define KCACHE_BANK0(x) (x)
//kb1
#define KCACHE_BANK1(x) (x)
// km0/1
#define KCACHE_MODE0(x) (x)
#define KCACHE_MODE1(x) (x) // SQ_CF_KCACHE_*
//
#define KCACHE_ADDR0(x) (x)
#define KCACHE_ADDR1(x) (x)
#define ALT_CONST(x) (x)
#define ARRAY_BASE(x) (x)
// export pixel
#define CF_PIXEL_MRT0 0
#define CF_PIXEL_MRT1 1
#define CF_PIXEL_MRT2 2
#define CF_PIXEL_MRT3 3
#define CF_PIXEL_MRT4 4
#define CF_PIXEL_MRT5 5
#define CF_PIXEL_MRT6 6
#define CF_PIXEL_MRT7 7
// computed Z
#define CF_COMPUTED_Z 61
// export pos
#define CF_POS0 60
#define CF_POS1 61
#define CF_POS2 62
#define CF_POS3 63
// export param
// 0...31
#define TYPE(x) (x) // SQ_EXPORT_*
#define RW_GPR(x) (x)
#define RW_REL(x) (x)
#define ABSOLUTE 0
#define RELATIVE 1
#define INDEX_GPR(x) (x)
#define ELEM_SIZE(x) (x ? (x - 1) : 0)
#define BURST_COUNT(x) (x ? (x - 1) : 0)
#define MARK(x) (x)
// swiz
#define SRC_SEL_X(x) (x) // SQ_SEL_* each
#define SRC_SEL_Y(x) (x)
#define SRC_SEL_Z(x) (x)
#define SRC_SEL_W(x) (x)
#define CF_DWORD0(addr, jmptbl) ((addr) | ((jmptbl) << 24))
#define CF_DWORD1(pc, cf_const, cond, count, vpm, cf_inst, b) \
(((pc) << 0) | ((cf_const) << 3) | ((cond) << 8) | ((count) << 10) | \
((vpm) << 20) | ((cf_inst) << 22) | ((b) << 31))
#define CF_ALU_DWORD0(addr, kb0, kb1, km0) (((addr) << 0) | ((kb0) << 22) | ((kb1) << 26) | ((km0) << 30))
#define CF_ALU_DWORD1(km1, kcache_addr0, kcache_addr1, count, alt_const, cf_inst, wqm, b) \
(((km1) << 0) | ((kcache_addr0) << 2) | ((kcache_addr1) << 10) | \
((count) << 18) | ((alt_const) << 25) | ((cf_inst) << 26) | ((wqm) << 30) | ((b) << 31))
#define CF_ALLOC_IMP_EXP_DWORD0(array_base, type, rw_gpr, rr, index_gpr, es) \
(((array_base) << 0) | ((type) << 13) | ((rw_gpr) << 15) | ((rr) << 22) | \
((index_gpr) << 23) | ((es) << 30))
#define CF_ALLOC_IMP_EXP_DWORD1_SWIZ(sel_x, sel_y, sel_z, sel_w, bc, vpm, cf_inst, m, b) \
(((sel_x) << 0) | ((sel_y) << 3) | ((sel_z) << 6) | ((sel_w) << 9) | \
((bc) << 16) | ((vpm) << 20) | ((cf_inst) << 22) | ((m) << 30) | ((b) << 31))
// ALU clause insts
#define SRC0_SEL(x) (x)
#define SRC1_SEL(x) (x)
#define SRC2_SEL(x) (x)
// src[0-2]_sel
// 0-127 GPR
// 128-159 kcache constants bank 0
// 160-191 kcache constants bank 1
// 192-255 inline const values
// 256-287 kcache constants bank 2
// 288-319 kcache constants bank 3
// 219-255 special SQ_ALU_SRC_* (0, 1, etc.)
// 488-520 src param space
#define ALU_SRC_GPR_BASE 0
#define ALU_SRC_KCACHE0_BASE 128
#define ALU_SRC_KCACHE1_BASE 160
#define ALU_SRC_INLINE_K_BASE 192
#define ALU_SRC_KCACHE2_BASE 256
#define ALU_SRC_KCACHE3_BASE 288
#define ALU_SRC_PARAM_BASE 448
#define SRC0_REL(x) (x)
#define SRC1_REL(x) (x)
#define SRC2_REL(x) (x)
// elem
#define SRC0_ELEM(x) (x)
#define SRC1_ELEM(x) (x)
#define SRC2_ELEM(x) (x)
#define ELEM_X 0
#define ELEM_Y 1
#define ELEM_Z 2
#define ELEM_W 3
// neg
#define SRC0_NEG(x) (x)
#define SRC1_NEG(x) (x)
#define SRC2_NEG(x) (x)
// im
#define INDEX_MODE(x) (x) // SQ_INDEX_*
// ps
#define PRED_SEL(x) (x) // SQ_PRED_SEL_*
// last
#define LAST(x) (x)
// abs
#define SRC0_ABS(x) (x)
#define SRC1_ABS(x) (x)
// uem
#define UPDATE_EXECUTE_MASK(x) (x)
// up
#define UPDATE_PRED(x) (x)
// wm
#define WRITE_MASK(x) (x)
// omod
#define OMOD(x) (x) // SQ_ALU_OMOD_*
// alu inst
#define ALU_INST(x) (x) // SQ_ALU_INST_*
//bs
#define BANK_SWIZZLE(x) (x) // SQ_ALU_VEC_*
#define DST_GPR(x) (x)
#define DST_REL(x) (x)
#define DST_ELEM(x) (x)
#define CLAMP(x) (x)
#define ALU_DWORD0(src0_sel, s0r, s0e, s0n, src1_sel, s1r, s1e, s1n, im, ps, last) \
(((src0_sel) << 0) | ((s0r) << 9) | ((s0e) << 10) | ((s0n) << 12) | \
((src1_sel) << 13) | ((s1r) << 22) | ((s1e) << 23) | ((s1n) << 25) | \
((im) << 26) | ((ps) << 29) | ((last) << 31))
#define ALU_DWORD1_OP2(s0a, s1a, uem, up, wm, omod, alu_inst, bs, dst_gpr, dr, de, clamp) \
(((s0a) << 0) | ((s1a) << 1) | ((uem) << 2) | ((up) << 3) | ((wm) << 4) | \
((omod) << 5) | ((alu_inst) << 7) | ((bs) << 18) | ((dst_gpr) << 21) | \
((dr) << 28) | ((de) << 29) | ((clamp) << 31))
#define ALU_DWORD1_OP3(src2_sel, s2r, s2e, s2n, alu_inst, bs, dst_gpr, dr, de, clamp) \
(((src2_sel) << 0) | ((s2r) << 9) | ((s2e) << 10) | ((s2n) << 12) | \
((alu_inst) << 13) | ((bs) << 18) | ((dst_gpr) << 21) | ((dr) << 28) | \
((de) << 29) | ((clamp) << 31))
// VTX clause insts
// vxt insts
#define VTX_INST(x) (x) // SQ_VTX_INST_*
// fetch type
#define FETCH_TYPE(x) (x) // SQ_VTX_FETCH_*
#define FETCH_WHOLE_QUAD(x) (x)
#define BUFFER_ID(x) (x)
#define SRC_GPR(x) (x)
#define SRC_REL(x) (x)
#define STRUCTURED_READ(x) (x)
#define LDS_REQ(x) (x)
#define COALESCED_READ(x) (x)
#define DST_SEL_X(x) (x)
#define DST_SEL_Y(x) (x)
#define DST_SEL_Z(x) (x)
#define DST_SEL_W(x) (x)
#define USE_CONST_FIELDS(x) (x)
#define DATA_FORMAT(x) (x)
// num format
#define NUM_FORMAT_ALL(x) (x) // SQ_NUM_FORMAT_*
// format comp
#define FORMAT_COMP_ALL(x) (x) // SQ_FORMAT_COMP_*
// sma
#define SRF_MODE_ALL(x) (x)
#define SRF_MODE_ZERO_CLAMP_MINUS_ONE 0
#define SRF_MODE_NO_ZERO 1
#define OFFSET(x) (x)
// endian swap
#define ENDIAN_SWAP(x) (x) // SQ_ENDIAN_*
#define CONST_BUF_NO_STRIDE(x) (x)
#define BUFFER_INDEX_MODE(x) (x)
#define VTX_DWORD0(vtx_inst, ft, fwq, buffer_id, src_gpr, sr, ssx, ssy, str, ldsr, cr) \
(((vtx_inst) << 0) | ((ft) << 5) | ((fwq) << 7) | ((buffer_id) << 8) | \
((src_gpr) << 16) | ((sr) << 23) | ((ssx) << 24) | ((ssy) << 26) | \
((str) << 28) | ((ldsr) << 30) | ((cr) << 31))
#define VTX_DWORD1_GPR(dst_gpr, dr, dsx, dsy, dsz, dsw, ucf, data_format, nfa, fca, sma) \
(((dst_gpr) << 0) | ((dr) << 7) | ((dsx) << 9) | ((dsy) << 12) | ((dsz) << 15) | ((dsw) << 18) | \
((ucf) << 21) | ((data_format) << 22) | ((nfa) << 28) | ((fca) << 30) | ((sma) << 31))
#define VTX_DWORD2(offset, es, cbns, alt_const, bim) \
(((offset) << 0) | ((es) << 16) | ((cbns) << 18) | ((alt_const) << 20) | ((bim) << 21))
#define VTX_DWORD_PAD 0x00000000
// TEX clause insts
// tex insts
#define TEX_INST(x) (x) // SQ_TEX_INST_*
#define INST_MOD(x) (x)
#define FETCH_WHOLE_QUAD(x) (x)
#define RESOURCE_ID(x) (x)
#define RESOURCE_INDEX_MODE(x) (x)
#define SAMPLER_INDEX_MODE(x) (x)
#define LOD_BIAS(x) (x)
//ct
#define COORD_TYPE_X(x) (x)
#define COORD_TYPE_Y(x) (x)
#define COORD_TYPE_Z(x) (x)
#define COORD_TYPE_W(x) (x)
#define TEX_UNNORMALIZED 0
#define TEX_NORMALIZED 1
#define OFFSET_X(x) (((int)(x) * 2) & 0x1f) /* 4:1-bits 2's-complement fixed-point: [-8.0..7.5] */
#define OFFSET_Y(x) (((int)(x) * 2) & 0x1f)
#define OFFSET_Z(x) (((int)(x) * 2) & 0x1f)
#define SAMPLER_ID(x) (x)
#define TEX_DWORD0(tex_inst, im, fwq, resource_id, src_gpr, sr, ac, rim, sim) \
(((tex_inst) << 0) | ((im) << 5) | ((fwq) << 7) | ((resource_id) << 8) | \
((src_gpr) << 16) | ((sr) << 23) | ((ac) << 24) | ((rim) << 25) | ((sim) << 27))
#define TEX_DWORD1(dst_gpr, dr, dsx, dsy, dsz, dsw, lod_bias, ctx, cty, ctz, ctw) \
(((dst_gpr) << 0) | ((dr) << 7) | ((dsx) << 9) | ((dsy) << 12) | ((dsz) << 15) | ((dsw) << 18) | \
((lod_bias) << 21) | ((ctx) << 28) | ((cty) << 29) | ((ctz) << 30) | ((ctw) << 31))
#define TEX_DWORD2(offset_x, offset_y, offset_z, sampler_id, ssx, ssy, ssz, ssw) \
(((offset_x) << 0) | ((offset_y) << 5) | ((offset_z) << 10) | ((sampler_id) << 15) | \
((ssx) << 20) | ((ssy) << 23) | ((ssz) << 26) | ((ssw) << 29))
#define TEX_DWORD_PAD 0x00000000
#endif
|
