aco: Fix integer overflows when emitting parallel copies during RA
[mesa.git] / src / amd / vulkan / radv_cs.h
1 /*
2 * Copyright © 2016 Red Hat.
3 * Copyright © 2016 Bas Nieuwenhuizen
4 *
5 * Permission is hereby granted, free of charge, to any person obtaining a
6 * copy of this software and associated documentation files (the "Software"),
7 * to deal in the Software without restriction, including without limitation
8 * the rights to use, copy, modify, merge, publish, distribute, sublicense,
9 * and/or sell copies of the Software, and to permit persons to whom the
10 * Software is furnished to do so, subject to the following conditions:
11 *
12 * The above copyright notice and this permission notice (including the next
13 * paragraph) shall be included in all copies or substantial portions of the
14 * Software.
15 *
16 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
17 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
18 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
19 * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
20 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
21 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
22 * IN THE SOFTWARE.
23 */
24
25 #ifndef RADV_CS_H
26 #define RADV_CS_H
27
28 #include <string.h>
29 #include <stdint.h>
30 #include <assert.h>
31 #include "radv_private.h"
32 #include "sid.h"
33
34 static inline unsigned radeon_check_space(struct radeon_winsys *ws,
35 struct radeon_cmdbuf *cs,
36 unsigned needed)
37 {
38 if (cs->max_dw - cs->cdw < needed)
39 ws->cs_grow(cs, needed);
40 return cs->cdw + needed;
41 }
42
43 static inline void radeon_set_config_reg_seq(struct radeon_cmdbuf *cs, unsigned reg, unsigned num)
44 {
45 assert(reg >= SI_CONFIG_REG_OFFSET && reg < SI_CONFIG_REG_END);
46 assert(cs->cdw + 2 + num <= cs->max_dw);
47 assert(num);
48 radeon_emit(cs, PKT3(PKT3_SET_CONFIG_REG, num, 0));
49 radeon_emit(cs, (reg - SI_CONFIG_REG_OFFSET) >> 2);
50 }
51
52 static inline void radeon_set_config_reg(struct radeon_cmdbuf *cs, unsigned reg, unsigned value)
53 {
54 radeon_set_config_reg_seq(cs, reg, 1);
55 radeon_emit(cs, value);
56 }
57
58 static inline void radeon_set_context_reg_seq(struct radeon_cmdbuf *cs, unsigned reg, unsigned num)
59 {
60 assert(reg >= SI_CONTEXT_REG_OFFSET && reg < SI_CONTEXT_REG_END);
61 assert(cs->cdw + 2 + num <= cs->max_dw);
62 assert(num);
63 radeon_emit(cs, PKT3(PKT3_SET_CONTEXT_REG, num, 0));
64 radeon_emit(cs, (reg - SI_CONTEXT_REG_OFFSET) >> 2);
65 }
66
67 static inline void radeon_set_context_reg(struct radeon_cmdbuf *cs, unsigned reg, unsigned value)
68 {
69 radeon_set_context_reg_seq(cs, reg, 1);
70 radeon_emit(cs, value);
71 }
72
73
74 static inline void radeon_set_context_reg_idx(struct radeon_cmdbuf *cs,
75 unsigned reg, unsigned idx,
76 unsigned value)
77 {
78 assert(reg >= SI_CONTEXT_REG_OFFSET && reg < SI_CONTEXT_REG_END);
79 assert(cs->cdw + 3 <= cs->max_dw);
80 radeon_emit(cs, PKT3(PKT3_SET_CONTEXT_REG, 1, 0));
81 radeon_emit(cs, (reg - SI_CONTEXT_REG_OFFSET) >> 2 | (idx << 28));
82 radeon_emit(cs, value);
83 }
84
85 static inline void radeon_set_context_reg_rmw(struct radeon_cmdbuf *cs,
86 unsigned reg, unsigned value,
87 unsigned mask)
88 {
89 assert(reg >= SI_CONTEXT_REG_OFFSET && reg < SI_CONTEXT_REG_END);
90 assert(cs->cdw + 4 <= cs->max_dw);
91 radeon_emit(cs, PKT3(PKT3_CONTEXT_REG_RMW, 2, 0));
92 radeon_emit(cs, (reg - SI_CONTEXT_REG_OFFSET) >> 2);
93 radeon_emit(cs, mask);
94 radeon_emit(cs, value);
95 }
96
97 static inline void radeon_set_sh_reg_seq(struct radeon_cmdbuf *cs, unsigned reg, unsigned num)
98 {
99 assert(reg >= SI_SH_REG_OFFSET && reg < SI_SH_REG_END);
100 assert(cs->cdw + 2 + num <= cs->max_dw);
101 assert(num);
102 radeon_emit(cs, PKT3(PKT3_SET_SH_REG, num, 0));
103 radeon_emit(cs, (reg - SI_SH_REG_OFFSET) >> 2);
104 }
105
106 static inline void radeon_set_sh_reg(struct radeon_cmdbuf *cs, unsigned reg, unsigned value)
107 {
108 radeon_set_sh_reg_seq(cs, reg, 1);
109 radeon_emit(cs, value);
110 }
111
112 static inline void radeon_set_sh_reg_idx(const struct radv_physical_device *pdevice,
113 struct radeon_cmdbuf *cs,
114 unsigned reg, unsigned idx,
115 unsigned value)
116 {
117 assert(reg >= SI_SH_REG_OFFSET && reg < SI_SH_REG_END);
118 assert(cs->cdw + 3 <= cs->max_dw);
119 assert(idx);
120
121 unsigned opcode = PKT3_SET_SH_REG_INDEX;
122 if (pdevice->rad_info.chip_class < GFX10)
123 opcode = PKT3_SET_SH_REG;
124
125 radeon_emit(cs, PKT3(opcode, 1, 0));
126 radeon_emit(cs, (reg - SI_SH_REG_OFFSET) >> 2 | (idx << 28));
127 radeon_emit(cs, value);
128 }
129
130 static inline void radeon_set_uconfig_reg_seq(struct radeon_cmdbuf *cs, unsigned reg, unsigned num)
131 {
132 assert(reg >= CIK_UCONFIG_REG_OFFSET && reg < CIK_UCONFIG_REG_END);
133 assert(cs->cdw + 2 + num <= cs->max_dw);
134 assert(num);
135 radeon_emit(cs, PKT3(PKT3_SET_UCONFIG_REG, num, 0));
136 radeon_emit(cs, (reg - CIK_UCONFIG_REG_OFFSET) >> 2);
137 }
138
139 static inline void radeon_set_uconfig_reg(struct radeon_cmdbuf *cs, unsigned reg, unsigned value)
140 {
141 radeon_set_uconfig_reg_seq(cs, reg, 1);
142 radeon_emit(cs, value);
143 }
144
145 static inline void radeon_set_uconfig_reg_idx(const struct radv_physical_device *pdevice,
146 struct radeon_cmdbuf *cs,
147 unsigned reg, unsigned idx,
148 unsigned value)
149 {
150 assert(reg >= CIK_UCONFIG_REG_OFFSET && reg < CIK_UCONFIG_REG_END);
151 assert(cs->cdw + 3 <= cs->max_dw);
152 assert(idx);
153
154 unsigned opcode = PKT3_SET_UCONFIG_REG_INDEX;
155 if (pdevice->rad_info.chip_class < GFX9 ||
156 (pdevice->rad_info.chip_class == GFX9 && pdevice->rad_info.me_fw_version < 26))
157 opcode = PKT3_SET_UCONFIG_REG;
158
159 radeon_emit(cs, PKT3(opcode, 1, 0));
160 radeon_emit(cs, (reg - CIK_UCONFIG_REG_OFFSET) >> 2 | (idx << 28));
161 radeon_emit(cs, value);
162 }
163
164 static inline void radeon_set_privileged_config_reg(struct radeon_cmdbuf *cs,
165 unsigned reg,
166 unsigned value)
167 {
168 assert(reg < CIK_UCONFIG_REG_OFFSET);
169 assert(cs->cdw + 6 <= cs->max_dw);
170
171 radeon_emit(cs, PKT3(PKT3_COPY_DATA, 4, 0));
172 radeon_emit(cs, COPY_DATA_SRC_SEL(COPY_DATA_IMM) |
173 COPY_DATA_DST_SEL(COPY_DATA_PERF));
174 radeon_emit(cs, value);
175 radeon_emit(cs, 0); /* unused */
176 radeon_emit(cs, reg >> 2);
177 radeon_emit(cs, 0); /* unused */
178 }
179
180 #endif /* RADV_CS_H */