Support Partitioned subgroup builtin functions with extended types
Also refactor the generation of those builtin functions programmatically
combine names and types.
diff --git a/Test/baseResults/spv.subgroupExtendedTypesPartitioned.comp.out b/Test/baseResults/spv.subgroupExtendedTypesPartitioned.comp.out
new file mode 100644
index 0000000..f384609
--- /dev/null
+++ b/Test/baseResults/spv.subgroupExtendedTypesPartitioned.comp.out
@@ -0,0 +1,1836 @@
+spv.subgroupExtendedTypesPartitioned.comp
+Validation failed
+// Module Version 10300
+// Generated by (magic number): 80007
+// Id's are bound by 1558
+
+ Capability Shader
+ Capability Float16
+ Capability Int64
+ Capability Int16
+ Capability Int8
+ Capability GroupNonUniform
+ Capability StorageUniformBufferBlock16
+ Capability CapabilityStorageBuffer8BitAccess
+ Capability GroupNonUniformPartitionedNV
+ Extension "SPV_KHR_8bit_storage"
+ Extension "SPV_NV_shader_subgroup_partitioned"
+ 1: ExtInstImport "GLSL.std.450"
+ MemoryModel Logical GLSL450
+ EntryPoint GLCompute 4 "main" 10 12
+ ExecutionMode 4 LocalSize 8 1 1
+ Source GLSL 450
+ SourceExtension "GL_EXT_shader_explicit_arithmetic_types_float16"
+ SourceExtension "GL_EXT_shader_explicit_arithmetic_types_int16"
+ SourceExtension "GL_EXT_shader_explicit_arithmetic_types_int64"
+ SourceExtension "GL_EXT_shader_explicit_arithmetic_types_int8"
+ SourceExtension "GL_EXT_shader_subgroup_extended_types_float16"
+ SourceExtension "GL_EXT_shader_subgroup_extended_types_int16"
+ SourceExtension "GL_EXT_shader_subgroup_extended_types_int64"
+ SourceExtension "GL_EXT_shader_subgroup_extended_types_int8"
+ SourceExtension "GL_KHR_shader_subgroup_basic"
+ SourceExtension "GL_NV_shader_subgroup_partitioned"
+ Name 4 "main"
+ Name 8 "invocation"
+ Name 10 "gl_SubgroupInvocationID"
+ Name 12 "gl_SubgroupSize"
+ Name 19 "ballot"
+ Name 34 "Buffers"
+ MemberName 34(Buffers) 0 "i8"
+ MemberName 34(Buffers) 1 "u8"
+ MemberName 34(Buffers) 2 "i16"
+ MemberName 34(Buffers) 3 "u16"
+ MemberName 34(Buffers) 4 "i64"
+ MemberName 34(Buffers) 5 "u64"
+ MemberName 34(Buffers) 6 "f16"
+ Name 37 "data"
+ Decorate 10(gl_SubgroupInvocationID) RelaxedPrecision
+ Decorate 10(gl_SubgroupInvocationID) BuiltIn SubgroupLocalInvocationId
+ Decorate 11 RelaxedPrecision
+ Decorate 12(gl_SubgroupSize) RelaxedPrecision
+ Decorate 12(gl_SubgroupSize) BuiltIn SubgroupSize
+ Decorate 13 RelaxedPrecision
+ Decorate 14 RelaxedPrecision
+ Decorate 16 RelaxedPrecision
+ MemberDecorate 34(Buffers) 0 Offset 0
+ MemberDecorate 34(Buffers) 1 Offset 4
+ MemberDecorate 34(Buffers) 2 Offset 8
+ MemberDecorate 34(Buffers) 3 Offset 16
+ MemberDecorate 34(Buffers) 4 Offset 32
+ MemberDecorate 34(Buffers) 5 Offset 64
+ MemberDecorate 34(Buffers) 6 Offset 96
+ Decorate 34(Buffers) Block
+ Decorate 37(data) DescriptorSet 0
+ Decorate 37(data) Binding 0
+ Decorate 1557 BuiltIn WorkgroupSize
+ 2: TypeVoid
+ 3: TypeFunction 2
+ 6: TypeInt 32 0
+ 7: TypePointer Function 6(int)
+ 9: TypePointer Input 6(int)
+10(gl_SubgroupInvocationID): 9(ptr) Variable Input
+12(gl_SubgroupSize): 9(ptr) Variable Input
+ 15: 6(int) Constant 4
+ 17: TypeVector 6(int) 4
+ 18: TypePointer Function 17(ivec4)
+ 20: TypeInt 8 1
+ 21: TypeVector 20(int8_t) 4
+ 22: TypeInt 8 0
+ 23: TypeVector 22(int8_t) 4
+ 24: TypeInt 16 1
+ 25: TypeVector 24(int16_t) 4
+ 26: TypeInt 16 0
+ 27: TypeVector 26(int16_t) 4
+ 28: TypeInt 64 1
+ 29: TypeVector 28(int64_t) 4
+ 30: TypeInt 64 0
+ 31: TypeVector 30(int64_t) 4
+ 32: TypeFloat 16
+ 33: TypeVector 32(float16_t) 4
+ 34(Buffers): TypeStruct 21(i8vec4) 23(i8vec4) 25(i16vec4) 27(i16vec4) 29(i64vec4) 31(i64vec4) 33(f16vec4)
+ 35: TypeArray 34(Buffers) 15
+ 36: TypePointer StorageBuffer 35
+ 37(data): 36(ptr) Variable StorageBuffer
+ 38: TypeInt 32 1
+ 39: 38(int) Constant 0
+ 40: 6(int) Constant 0
+ 41: TypePointer StorageBuffer 20(int8_t)
+ 45: 38(int) Constant 1
+ 46: TypeVector 20(int8_t) 2
+ 47: TypePointer StorageBuffer 21(i8vec4)
+ 52: 38(int) Constant 2
+ 53: TypeVector 20(int8_t) 3
+ 58: 38(int) Constant 3
+ 62: TypePointer StorageBuffer 22(int8_t)
+ 66: TypeVector 22(int8_t) 2
+ 67: TypePointer StorageBuffer 23(i8vec4)
+ 72: TypeVector 22(int8_t) 3
+ 80: TypePointer StorageBuffer 24(int16_t)
+ 84: TypeVector 24(int16_t) 2
+ 85: TypePointer StorageBuffer 25(i16vec4)
+ 90: TypeVector 24(int16_t) 3
+ 98: TypePointer StorageBuffer 26(int16_t)
+ 102: TypeVector 26(int16_t) 2
+ 103: TypePointer StorageBuffer 27(i16vec4)
+ 108: TypeVector 26(int16_t) 3
+ 116: 38(int) Constant 4
+ 117: TypePointer StorageBuffer 28(int64_t)
+ 121: TypeVector 28(int64_t) 2
+ 122: TypePointer StorageBuffer 29(i64vec4)
+ 127: TypeVector 28(int64_t) 3
+ 135: 38(int) Constant 5
+ 136: TypePointer StorageBuffer 30(int64_t)
+ 140: TypeVector 30(int64_t) 2
+ 141: TypePointer StorageBuffer 31(i64vec4)
+ 146: TypeVector 30(int64_t) 3
+ 154: 38(int) Constant 6
+ 155: TypePointer StorageBuffer 32(float16_t)
+ 159: TypeVector 32(float16_t) 2
+ 160: TypePointer StorageBuffer 33(f16vec4)
+ 165: TypeVector 32(float16_t) 3
+ 177: 6(int) Constant 3
+ 1554: TypeVector 6(int) 3
+ 1555: 6(int) Constant 8
+ 1556: 6(int) Constant 1
+ 1557: 1554(ivec3) ConstantComposite 1555 1556 1556
+ 4(main): 2 Function None 3
+ 5: Label
+ 8(invocation): 7(ptr) Variable Function
+ 19(ballot): 18(ptr) Variable Function
+ 11: 6(int) Load 10(gl_SubgroupInvocationID)
+ 13: 6(int) Load 12(gl_SubgroupSize)
+ 14: 6(int) IAdd 11 13
+ 16: 6(int) UMod 14 15
+ Store 8(invocation) 16
+ 42: 41(ptr) AccessChain 37(data) 39 39 40
+ 43: 20(int8_t) Load 42
+ 44: 17(ivec4) GroupNonUniformPartitionNV 43
+ Store 19(ballot) 44
+ 48: 47(ptr) AccessChain 37(data) 45 39
+ 49: 21(i8vec4) Load 48
+ 50: 46(i8vec2) VectorShuffle 49 49 0 1
+ 51: 17(ivec4) GroupNonUniformPartitionNV 50
+ Store 19(ballot) 51
+ 54: 47(ptr) AccessChain 37(data) 52 39
+ 55: 21(i8vec4) Load 54
+ 56: 53(i8vec3) VectorShuffle 55 55 0 1 2
+ 57: 17(ivec4) GroupNonUniformPartitionNV 56
+ Store 19(ballot) 57
+ 59: 47(ptr) AccessChain 37(data) 58 39
+ 60: 21(i8vec4) Load 59
+ 61: 17(ivec4) GroupNonUniformPartitionNV 60
+ Store 19(ballot) 61
+ 63: 62(ptr) AccessChain 37(data) 39 45 40
+ 64: 22(int8_t) Load 63
+ 65: 17(ivec4) GroupNonUniformPartitionNV 64
+ Store 19(ballot) 65
+ 68: 67(ptr) AccessChain 37(data) 45 45
+ 69: 23(i8vec4) Load 68
+ 70: 66(i8vec2) VectorShuffle 69 69 0 1
+ 71: 17(ivec4) GroupNonUniformPartitionNV 70
+ Store 19(ballot) 71
+ 73: 67(ptr) AccessChain 37(data) 52 45
+ 74: 23(i8vec4) Load 73
+ 75: 72(i8vec3) VectorShuffle 74 74 0 1 2
+ 76: 17(ivec4) GroupNonUniformPartitionNV 75
+ Store 19(ballot) 76
+ 77: 67(ptr) AccessChain 37(data) 58 45
+ 78: 23(i8vec4) Load 77
+ 79: 17(ivec4) GroupNonUniformPartitionNV 78
+ Store 19(ballot) 79
+ 81: 80(ptr) AccessChain 37(data) 39 52 40
+ 82: 24(int16_t) Load 81
+ 83: 17(ivec4) GroupNonUniformPartitionNV 82
+ Store 19(ballot) 83
+ 86: 85(ptr) AccessChain 37(data) 45 52
+ 87: 25(i16vec4) Load 86
+ 88: 84(i16vec2) VectorShuffle 87 87 0 1
+ 89: 17(ivec4) GroupNonUniformPartitionNV 88
+ Store 19(ballot) 89
+ 91: 85(ptr) AccessChain 37(data) 52 52
+ 92: 25(i16vec4) Load 91
+ 93: 90(i16vec3) VectorShuffle 92 92 0 1 2
+ 94: 17(ivec4) GroupNonUniformPartitionNV 93
+ Store 19(ballot) 94
+ 95: 85(ptr) AccessChain 37(data) 58 52
+ 96: 25(i16vec4) Load 95
+ 97: 17(ivec4) GroupNonUniformPartitionNV 96
+ Store 19(ballot) 97
+ 99: 98(ptr) AccessChain 37(data) 39 58 40
+ 100: 26(int16_t) Load 99
+ 101: 17(ivec4) GroupNonUniformPartitionNV 100
+ Store 19(ballot) 101
+ 104: 103(ptr) AccessChain 37(data) 45 58
+ 105: 27(i16vec4) Load 104
+ 106:102(i16vec2) VectorShuffle 105 105 0 1
+ 107: 17(ivec4) GroupNonUniformPartitionNV 106
+ Store 19(ballot) 107
+ 109: 103(ptr) AccessChain 37(data) 52 58
+ 110: 27(i16vec4) Load 109
+ 111:108(i16vec3) VectorShuffle 110 110 0 1 2
+ 112: 17(ivec4) GroupNonUniformPartitionNV 111
+ Store 19(ballot) 112
+ 113: 103(ptr) AccessChain 37(data) 58 58
+ 114: 27(i16vec4) Load 113
+ 115: 17(ivec4) GroupNonUniformPartitionNV 114
+ Store 19(ballot) 115
+ 118: 117(ptr) AccessChain 37(data) 39 116 40
+ 119: 28(int64_t) Load 118
+ 120: 17(ivec4) GroupNonUniformPartitionNV 119
+ Store 19(ballot) 120
+ 123: 122(ptr) AccessChain 37(data) 45 116
+ 124: 29(i64vec4) Load 123
+ 125:121(i64vec2) VectorShuffle 124 124 0 1
+ 126: 17(ivec4) GroupNonUniformPartitionNV 125
+ Store 19(ballot) 126
+ 128: 122(ptr) AccessChain 37(data) 52 116
+ 129: 29(i64vec4) Load 128
+ 130:127(i64vec3) VectorShuffle 129 129 0 1 2
+ 131: 17(ivec4) GroupNonUniformPartitionNV 130
+ Store 19(ballot) 131
+ 132: 122(ptr) AccessChain 37(data) 58 116
+ 133: 29(i64vec4) Load 132
+ 134: 17(ivec4) GroupNonUniformPartitionNV 133
+ Store 19(ballot) 134
+ 137: 136(ptr) AccessChain 37(data) 39 135 40
+ 138: 30(int64_t) Load 137
+ 139: 17(ivec4) GroupNonUniformPartitionNV 138
+ Store 19(ballot) 139
+ 142: 141(ptr) AccessChain 37(data) 45 135
+ 143: 31(i64vec4) Load 142
+ 144:140(i64vec2) VectorShuffle 143 143 0 1
+ 145: 17(ivec4) GroupNonUniformPartitionNV 144
+ Store 19(ballot) 145
+ 147: 141(ptr) AccessChain 37(data) 52 135
+ 148: 31(i64vec4) Load 147
+ 149:146(i64vec3) VectorShuffle 148 148 0 1 2
+ 150: 17(ivec4) GroupNonUniformPartitionNV 149
+ Store 19(ballot) 150
+ 151: 141(ptr) AccessChain 37(data) 58 135
+ 152: 31(i64vec4) Load 151
+ 153: 17(ivec4) GroupNonUniformPartitionNV 152
+ Store 19(ballot) 153
+ 156: 155(ptr) AccessChain 37(data) 39 154 40
+ 157:32(float16_t) Load 156
+ 158: 17(ivec4) GroupNonUniformPartitionNV 157
+ Store 19(ballot) 158
+ 161: 160(ptr) AccessChain 37(data) 45 154
+ 162: 33(f16vec4) Load 161
+ 163:159(f16vec2) VectorShuffle 162 162 0 1
+ 164: 17(ivec4) GroupNonUniformPartitionNV 163
+ Store 19(ballot) 164
+ 166: 160(ptr) AccessChain 37(data) 52 154
+ 167: 33(f16vec4) Load 166
+ 168:165(f16vec3) VectorShuffle 167 167 0 1 2
+ 169: 17(ivec4) GroupNonUniformPartitionNV 168
+ Store 19(ballot) 169
+ 170: 160(ptr) AccessChain 37(data) 58 154
+ 171: 33(f16vec4) Load 170
+ 172: 17(ivec4) GroupNonUniformPartitionNV 171
+ Store 19(ballot) 172
+ 173: 6(int) Load 8(invocation)
+ 174: 41(ptr) AccessChain 37(data) 39 39 40
+ 175: 20(int8_t) Load 174
+ 176: 17(ivec4) Load 19(ballot)
+ 178: 20(int8_t) GroupNonUniformIAdd 177 PartitionedReduceNV 175 176
+ 179: 41(ptr) AccessChain 37(data) 173 39 40
+ Store 179 178
+ 180: 6(int) Load 8(invocation)
+ 181: 47(ptr) AccessChain 37(data) 45 39
+ 182: 21(i8vec4) Load 181
+ 183: 46(i8vec2) VectorShuffle 182 182 0 1
+ 184: 17(ivec4) Load 19(ballot)
+ 185: 46(i8vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 183 184
+ 186: 47(ptr) AccessChain 37(data) 180 39
+ 187: 21(i8vec4) Load 186
+ 188: 21(i8vec4) VectorShuffle 187 185 4 5 2 3
+ Store 186 188
+ 189: 6(int) Load 8(invocation)
+ 190: 47(ptr) AccessChain 37(data) 52 39
+ 191: 21(i8vec4) Load 190
+ 192: 53(i8vec3) VectorShuffle 191 191 0 1 2
+ 193: 17(ivec4) Load 19(ballot)
+ 194: 53(i8vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 192 193
+ 195: 47(ptr) AccessChain 37(data) 189 39
+ 196: 21(i8vec4) Load 195
+ 197: 21(i8vec4) VectorShuffle 196 194 4 5 6 3
+ Store 195 197
+ 198: 6(int) Load 8(invocation)
+ 199: 47(ptr) AccessChain 37(data) 58 39
+ 200: 21(i8vec4) Load 199
+ 201: 17(ivec4) Load 19(ballot)
+ 202: 21(i8vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 200 201
+ 203: 47(ptr) AccessChain 37(data) 198 39
+ Store 203 202
+ 204: 6(int) Load 8(invocation)
+ 205: 41(ptr) AccessChain 37(data) 39 39 40
+ 206: 20(int8_t) Load 205
+ 207: 17(ivec4) Load 19(ballot)
+ 208: 20(int8_t) GroupNonUniformIMul 177 PartitionedReduceNV 206 207
+ 209: 41(ptr) AccessChain 37(data) 204 39 40
+ Store 209 208
+ 210: 6(int) Load 8(invocation)
+ 211: 47(ptr) AccessChain 37(data) 45 39
+ 212: 21(i8vec4) Load 211
+ 213: 46(i8vec2) VectorShuffle 212 212 0 1
+ 214: 17(ivec4) Load 19(ballot)
+ 215: 46(i8vec2) GroupNonUniformIMul 177 PartitionedReduceNV 213 214
+ 216: 47(ptr) AccessChain 37(data) 210 39
+ 217: 21(i8vec4) Load 216
+ 218: 21(i8vec4) VectorShuffle 217 215 4 5 2 3
+ Store 216 218
+ 219: 6(int) Load 8(invocation)
+ 220: 47(ptr) AccessChain 37(data) 52 39
+ 221: 21(i8vec4) Load 220
+ 222: 53(i8vec3) VectorShuffle 221 221 0 1 2
+ 223: 17(ivec4) Load 19(ballot)
+ 224: 53(i8vec3) GroupNonUniformIMul 177 PartitionedReduceNV 222 223
+ 225: 47(ptr) AccessChain 37(data) 219 39
+ 226: 21(i8vec4) Load 225
+ 227: 21(i8vec4) VectorShuffle 226 224 4 5 6 3
+ Store 225 227
+ 228: 6(int) Load 8(invocation)
+ 229: 47(ptr) AccessChain 37(data) 58 39
+ 230: 21(i8vec4) Load 229
+ 231: 17(ivec4) Load 19(ballot)
+ 232: 21(i8vec4) GroupNonUniformIMul 177 PartitionedReduceNV 230 231
+ 233: 47(ptr) AccessChain 37(data) 228 39
+ Store 233 232
+ 234: 6(int) Load 8(invocation)
+ 235: 41(ptr) AccessChain 37(data) 39 39 40
+ 236: 20(int8_t) Load 235
+ 237: 17(ivec4) Load 19(ballot)
+ 238: 20(int8_t) GroupNonUniformSMin 177 PartitionedReduceNV 236 237
+ 239: 41(ptr) AccessChain 37(data) 234 39 40
+ Store 239 238
+ 240: 6(int) Load 8(invocation)
+ 241: 47(ptr) AccessChain 37(data) 45 39
+ 242: 21(i8vec4) Load 241
+ 243: 46(i8vec2) VectorShuffle 242 242 0 1
+ 244: 17(ivec4) Load 19(ballot)
+ 245: 46(i8vec2) GroupNonUniformSMin 177 PartitionedReduceNV 243 244
+ 246: 47(ptr) AccessChain 37(data) 240 39
+ 247: 21(i8vec4) Load 246
+ 248: 21(i8vec4) VectorShuffle 247 245 4 5 2 3
+ Store 246 248
+ 249: 6(int) Load 8(invocation)
+ 250: 47(ptr) AccessChain 37(data) 52 39
+ 251: 21(i8vec4) Load 250
+ 252: 53(i8vec3) VectorShuffle 251 251 0 1 2
+ 253: 17(ivec4) Load 19(ballot)
+ 254: 53(i8vec3) GroupNonUniformSMin 177 PartitionedReduceNV 252 253
+ 255: 47(ptr) AccessChain 37(data) 249 39
+ 256: 21(i8vec4) Load 255
+ 257: 21(i8vec4) VectorShuffle 256 254 4 5 6 3
+ Store 255 257
+ 258: 6(int) Load 8(invocation)
+ 259: 47(ptr) AccessChain 37(data) 58 39
+ 260: 21(i8vec4) Load 259
+ 261: 17(ivec4) Load 19(ballot)
+ 262: 21(i8vec4) GroupNonUniformSMin 177 PartitionedReduceNV 260 261
+ 263: 47(ptr) AccessChain 37(data) 258 39
+ Store 263 262
+ 264: 6(int) Load 8(invocation)
+ 265: 41(ptr) AccessChain 37(data) 39 39 40
+ 266: 20(int8_t) Load 265
+ 267: 17(ivec4) Load 19(ballot)
+ 268: 20(int8_t) GroupNonUniformSMax 177 PartitionedReduceNV 266 267
+ 269: 41(ptr) AccessChain 37(data) 264 39 40
+ Store 269 268
+ 270: 6(int) Load 8(invocation)
+ 271: 47(ptr) AccessChain 37(data) 45 39
+ 272: 21(i8vec4) Load 271
+ 273: 46(i8vec2) VectorShuffle 272 272 0 1
+ 274: 17(ivec4) Load 19(ballot)
+ 275: 46(i8vec2) GroupNonUniformSMax 177 PartitionedReduceNV 273 274
+ 276: 47(ptr) AccessChain 37(data) 270 39
+ 277: 21(i8vec4) Load 276
+ 278: 21(i8vec4) VectorShuffle 277 275 4 5 2 3
+ Store 276 278
+ 279: 6(int) Load 8(invocation)
+ 280: 47(ptr) AccessChain 37(data) 52 39
+ 281: 21(i8vec4) Load 280
+ 282: 53(i8vec3) VectorShuffle 281 281 0 1 2
+ 283: 17(ivec4) Load 19(ballot)
+ 284: 53(i8vec3) GroupNonUniformSMax 177 PartitionedReduceNV 282 283
+ 285: 47(ptr) AccessChain 37(data) 279 39
+ 286: 21(i8vec4) Load 285
+ 287: 21(i8vec4) VectorShuffle 286 284 4 5 6 3
+ Store 285 287
+ 288: 6(int) Load 8(invocation)
+ 289: 47(ptr) AccessChain 37(data) 58 39
+ 290: 21(i8vec4) Load 289
+ 291: 17(ivec4) Load 19(ballot)
+ 292: 21(i8vec4) GroupNonUniformSMax 177 PartitionedReduceNV 290 291
+ 293: 47(ptr) AccessChain 37(data) 288 39
+ Store 293 292
+ 294: 6(int) Load 8(invocation)
+ 295: 41(ptr) AccessChain 37(data) 39 39 40
+ 296: 20(int8_t) Load 295
+ 297: 17(ivec4) Load 19(ballot)
+ 298: 20(int8_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 296 297
+ 299: 41(ptr) AccessChain 37(data) 294 39 40
+ Store 299 298
+ 300: 6(int) Load 8(invocation)
+ 301: 47(ptr) AccessChain 37(data) 45 39
+ 302: 21(i8vec4) Load 301
+ 303: 46(i8vec2) VectorShuffle 302 302 0 1
+ 304: 17(ivec4) Load 19(ballot)
+ 305: 46(i8vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 303 304
+ 306: 47(ptr) AccessChain 37(data) 300 39
+ 307: 21(i8vec4) Load 306
+ 308: 21(i8vec4) VectorShuffle 307 305 4 5 2 3
+ Store 306 308
+ 309: 6(int) Load 8(invocation)
+ 310: 47(ptr) AccessChain 37(data) 52 39
+ 311: 21(i8vec4) Load 310
+ 312: 53(i8vec3) VectorShuffle 311 311 0 1 2
+ 313: 17(ivec4) Load 19(ballot)
+ 314: 53(i8vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 312 313
+ 315: 47(ptr) AccessChain 37(data) 309 39
+ 316: 21(i8vec4) Load 315
+ 317: 21(i8vec4) VectorShuffle 316 314 4 5 6 3
+ Store 315 317
+ 318: 6(int) Load 8(invocation)
+ 319: 47(ptr) AccessChain 37(data) 58 39
+ 320: 21(i8vec4) Load 319
+ 321: 17(ivec4) Load 19(ballot)
+ 322: 21(i8vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 320 321
+ 323: 47(ptr) AccessChain 37(data) 318 39
+ Store 323 322
+ 324: 6(int) Load 8(invocation)
+ 325: 41(ptr) AccessChain 37(data) 39 39 40
+ 326: 20(int8_t) Load 325
+ 327: 17(ivec4) Load 19(ballot)
+ 328: 20(int8_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 326 327
+ 329: 41(ptr) AccessChain 37(data) 324 39 40
+ Store 329 328
+ 330: 6(int) Load 8(invocation)
+ 331: 47(ptr) AccessChain 37(data) 45 39
+ 332: 21(i8vec4) Load 331
+ 333: 46(i8vec2) VectorShuffle 332 332 0 1
+ 334: 17(ivec4) Load 19(ballot)
+ 335: 46(i8vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 333 334
+ 336: 47(ptr) AccessChain 37(data) 330 39
+ 337: 21(i8vec4) Load 336
+ 338: 21(i8vec4) VectorShuffle 337 335 4 5 2 3
+ Store 336 338
+ 339: 6(int) Load 8(invocation)
+ 340: 47(ptr) AccessChain 37(data) 52 39
+ 341: 21(i8vec4) Load 340
+ 342: 53(i8vec3) VectorShuffle 341 341 0 1 2
+ 343: 17(ivec4) Load 19(ballot)
+ 344: 53(i8vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 342 343
+ 345: 47(ptr) AccessChain 37(data) 339 39
+ 346: 21(i8vec4) Load 345
+ 347: 21(i8vec4) VectorShuffle 346 344 4 5 6 3
+ Store 345 347
+ 348: 6(int) Load 8(invocation)
+ 349: 47(ptr) AccessChain 37(data) 58 39
+ 350: 21(i8vec4) Load 349
+ 351: 17(ivec4) Load 19(ballot)
+ 352: 21(i8vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 350 351
+ 353: 47(ptr) AccessChain 37(data) 348 39
+ Store 353 352
+ 354: 6(int) Load 8(invocation)
+ 355: 41(ptr) AccessChain 37(data) 39 39 40
+ 356: 20(int8_t) Load 355
+ 357: 17(ivec4) Load 19(ballot)
+ 358: 20(int8_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 356 357
+ 359: 41(ptr) AccessChain 37(data) 354 39 40
+ Store 359 358
+ 360: 6(int) Load 8(invocation)
+ 361: 47(ptr) AccessChain 37(data) 45 39
+ 362: 21(i8vec4) Load 361
+ 363: 46(i8vec2) VectorShuffle 362 362 0 1
+ 364: 17(ivec4) Load 19(ballot)
+ 365: 46(i8vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 363 364
+ 366: 47(ptr) AccessChain 37(data) 360 39
+ 367: 21(i8vec4) Load 366
+ 368: 21(i8vec4) VectorShuffle 367 365 4 5 2 3
+ Store 366 368
+ 369: 6(int) Load 8(invocation)
+ 370: 47(ptr) AccessChain 37(data) 52 39
+ 371: 21(i8vec4) Load 370
+ 372: 53(i8vec3) VectorShuffle 371 371 0 1 2
+ 373: 17(ivec4) Load 19(ballot)
+ 374: 53(i8vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 372 373
+ 375: 47(ptr) AccessChain 37(data) 369 39
+ 376: 21(i8vec4) Load 375
+ 377: 21(i8vec4) VectorShuffle 376 374 4 5 6 3
+ Store 375 377
+ 378: 6(int) Load 8(invocation)
+ 379: 47(ptr) AccessChain 37(data) 58 39
+ 380: 21(i8vec4) Load 379
+ 381: 17(ivec4) Load 19(ballot)
+ 382: 21(i8vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 380 381
+ 383: 47(ptr) AccessChain 37(data) 378 39
+ Store 383 382
+ 384: 6(int) Load 8(invocation)
+ 385: 62(ptr) AccessChain 37(data) 39 45 40
+ 386: 22(int8_t) Load 385
+ 387: 17(ivec4) Load 19(ballot)
+ 388: 22(int8_t) GroupNonUniformIAdd 177 PartitionedReduceNV 386 387
+ 389: 62(ptr) AccessChain 37(data) 384 45 40
+ Store 389 388
+ 390: 6(int) Load 8(invocation)
+ 391: 67(ptr) AccessChain 37(data) 45 45
+ 392: 23(i8vec4) Load 391
+ 393: 66(i8vec2) VectorShuffle 392 392 0 1
+ 394: 17(ivec4) Load 19(ballot)
+ 395: 66(i8vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 393 394
+ 396: 67(ptr) AccessChain 37(data) 390 45
+ 397: 23(i8vec4) Load 396
+ 398: 23(i8vec4) VectorShuffle 397 395 4 5 2 3
+ Store 396 398
+ 399: 6(int) Load 8(invocation)
+ 400: 67(ptr) AccessChain 37(data) 52 45
+ 401: 23(i8vec4) Load 400
+ 402: 72(i8vec3) VectorShuffle 401 401 0 1 2
+ 403: 17(ivec4) Load 19(ballot)
+ 404: 72(i8vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 402 403
+ 405: 67(ptr) AccessChain 37(data) 399 45
+ 406: 23(i8vec4) Load 405
+ 407: 23(i8vec4) VectorShuffle 406 404 4 5 6 3
+ Store 405 407
+ 408: 6(int) Load 8(invocation)
+ 409: 67(ptr) AccessChain 37(data) 58 45
+ 410: 23(i8vec4) Load 409
+ 411: 17(ivec4) Load 19(ballot)
+ 412: 23(i8vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 410 411
+ 413: 67(ptr) AccessChain 37(data) 408 45
+ Store 413 412
+ 414: 6(int) Load 8(invocation)
+ 415: 62(ptr) AccessChain 37(data) 39 45 40
+ 416: 22(int8_t) Load 415
+ 417: 17(ivec4) Load 19(ballot)
+ 418: 22(int8_t) GroupNonUniformIMul 177 PartitionedReduceNV 416 417
+ 419: 62(ptr) AccessChain 37(data) 414 45 40
+ Store 419 418
+ 420: 6(int) Load 8(invocation)
+ 421: 67(ptr) AccessChain 37(data) 45 45
+ 422: 23(i8vec4) Load 421
+ 423: 66(i8vec2) VectorShuffle 422 422 0 1
+ 424: 17(ivec4) Load 19(ballot)
+ 425: 66(i8vec2) GroupNonUniformIMul 177 PartitionedReduceNV 423 424
+ 426: 67(ptr) AccessChain 37(data) 420 45
+ 427: 23(i8vec4) Load 426
+ 428: 23(i8vec4) VectorShuffle 427 425 4 5 2 3
+ Store 426 428
+ 429: 6(int) Load 8(invocation)
+ 430: 67(ptr) AccessChain 37(data) 52 45
+ 431: 23(i8vec4) Load 430
+ 432: 72(i8vec3) VectorShuffle 431 431 0 1 2
+ 433: 17(ivec4) Load 19(ballot)
+ 434: 72(i8vec3) GroupNonUniformIMul 177 PartitionedReduceNV 432 433
+ 435: 67(ptr) AccessChain 37(data) 429 45
+ 436: 23(i8vec4) Load 435
+ 437: 23(i8vec4) VectorShuffle 436 434 4 5 6 3
+ Store 435 437
+ 438: 6(int) Load 8(invocation)
+ 439: 67(ptr) AccessChain 37(data) 58 45
+ 440: 23(i8vec4) Load 439
+ 441: 17(ivec4) Load 19(ballot)
+ 442: 23(i8vec4) GroupNonUniformIMul 177 PartitionedReduceNV 440 441
+ 443: 67(ptr) AccessChain 37(data) 438 45
+ Store 443 442
+ 444: 6(int) Load 8(invocation)
+ 445: 62(ptr) AccessChain 37(data) 39 45 40
+ 446: 22(int8_t) Load 445
+ 447: 17(ivec4) Load 19(ballot)
+ 448: 22(int8_t) GroupNonUniformSMin 177 PartitionedReduceNV 446 447
+ 449: 62(ptr) AccessChain 37(data) 444 45 40
+ Store 449 448
+ 450: 6(int) Load 8(invocation)
+ 451: 67(ptr) AccessChain 37(data) 45 45
+ 452: 23(i8vec4) Load 451
+ 453: 66(i8vec2) VectorShuffle 452 452 0 1
+ 454: 17(ivec4) Load 19(ballot)
+ 455: 66(i8vec2) GroupNonUniformSMin 177 PartitionedReduceNV 453 454
+ 456: 67(ptr) AccessChain 37(data) 450 45
+ 457: 23(i8vec4) Load 456
+ 458: 23(i8vec4) VectorShuffle 457 455 4 5 2 3
+ Store 456 458
+ 459: 6(int) Load 8(invocation)
+ 460: 67(ptr) AccessChain 37(data) 52 45
+ 461: 23(i8vec4) Load 460
+ 462: 72(i8vec3) VectorShuffle 461 461 0 1 2
+ 463: 17(ivec4) Load 19(ballot)
+ 464: 72(i8vec3) GroupNonUniformSMin 177 PartitionedReduceNV 462 463
+ 465: 67(ptr) AccessChain 37(data) 459 45
+ 466: 23(i8vec4) Load 465
+ 467: 23(i8vec4) VectorShuffle 466 464 4 5 6 3
+ Store 465 467
+ 468: 6(int) Load 8(invocation)
+ 469: 67(ptr) AccessChain 37(data) 58 45
+ 470: 23(i8vec4) Load 469
+ 471: 17(ivec4) Load 19(ballot)
+ 472: 23(i8vec4) GroupNonUniformSMin 177 PartitionedReduceNV 470 471
+ 473: 67(ptr) AccessChain 37(data) 468 45
+ Store 473 472
+ 474: 6(int) Load 8(invocation)
+ 475: 62(ptr) AccessChain 37(data) 39 45 40
+ 476: 22(int8_t) Load 475
+ 477: 17(ivec4) Load 19(ballot)
+ 478: 22(int8_t) GroupNonUniformSMax 177 PartitionedReduceNV 476 477
+ 479: 62(ptr) AccessChain 37(data) 474 45 40
+ Store 479 478
+ 480: 6(int) Load 8(invocation)
+ 481: 67(ptr) AccessChain 37(data) 45 45
+ 482: 23(i8vec4) Load 481
+ 483: 66(i8vec2) VectorShuffle 482 482 0 1
+ 484: 17(ivec4) Load 19(ballot)
+ 485: 66(i8vec2) GroupNonUniformSMax 177 PartitionedReduceNV 483 484
+ 486: 67(ptr) AccessChain 37(data) 480 45
+ 487: 23(i8vec4) Load 486
+ 488: 23(i8vec4) VectorShuffle 487 485 4 5 2 3
+ Store 486 488
+ 489: 6(int) Load 8(invocation)
+ 490: 67(ptr) AccessChain 37(data) 52 45
+ 491: 23(i8vec4) Load 490
+ 492: 72(i8vec3) VectorShuffle 491 491 0 1 2
+ 493: 17(ivec4) Load 19(ballot)
+ 494: 72(i8vec3) GroupNonUniformSMax 177 PartitionedReduceNV 492 493
+ 495: 67(ptr) AccessChain 37(data) 489 45
+ 496: 23(i8vec4) Load 495
+ 497: 23(i8vec4) VectorShuffle 496 494 4 5 6 3
+ Store 495 497
+ 498: 6(int) Load 8(invocation)
+ 499: 67(ptr) AccessChain 37(data) 58 45
+ 500: 23(i8vec4) Load 499
+ 501: 17(ivec4) Load 19(ballot)
+ 502: 23(i8vec4) GroupNonUniformSMax 177 PartitionedReduceNV 500 501
+ 503: 67(ptr) AccessChain 37(data) 498 45
+ Store 503 502
+ 504: 6(int) Load 8(invocation)
+ 505: 62(ptr) AccessChain 37(data) 39 45 40
+ 506: 22(int8_t) Load 505
+ 507: 17(ivec4) Load 19(ballot)
+ 508: 22(int8_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 506 507
+ 509: 62(ptr) AccessChain 37(data) 504 45 40
+ Store 509 508
+ 510: 6(int) Load 8(invocation)
+ 511: 67(ptr) AccessChain 37(data) 45 45
+ 512: 23(i8vec4) Load 511
+ 513: 66(i8vec2) VectorShuffle 512 512 0 1
+ 514: 17(ivec4) Load 19(ballot)
+ 515: 66(i8vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 513 514
+ 516: 67(ptr) AccessChain 37(data) 510 45
+ 517: 23(i8vec4) Load 516
+ 518: 23(i8vec4) VectorShuffle 517 515 4 5 2 3
+ Store 516 518
+ 519: 6(int) Load 8(invocation)
+ 520: 67(ptr) AccessChain 37(data) 52 45
+ 521: 23(i8vec4) Load 520
+ 522: 72(i8vec3) VectorShuffle 521 521 0 1 2
+ 523: 17(ivec4) Load 19(ballot)
+ 524: 72(i8vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 522 523
+ 525: 67(ptr) AccessChain 37(data) 519 45
+ 526: 23(i8vec4) Load 525
+ 527: 23(i8vec4) VectorShuffle 526 524 4 5 6 3
+ Store 525 527
+ 528: 6(int) Load 8(invocation)
+ 529: 67(ptr) AccessChain 37(data) 58 45
+ 530: 23(i8vec4) Load 529
+ 531: 17(ivec4) Load 19(ballot)
+ 532: 23(i8vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 530 531
+ 533: 67(ptr) AccessChain 37(data) 528 45
+ Store 533 532
+ 534: 6(int) Load 8(invocation)
+ 535: 62(ptr) AccessChain 37(data) 39 45 40
+ 536: 22(int8_t) Load 535
+ 537: 17(ivec4) Load 19(ballot)
+ 538: 22(int8_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 536 537
+ 539: 62(ptr) AccessChain 37(data) 534 45 40
+ Store 539 538
+ 540: 6(int) Load 8(invocation)
+ 541: 67(ptr) AccessChain 37(data) 45 45
+ 542: 23(i8vec4) Load 541
+ 543: 66(i8vec2) VectorShuffle 542 542 0 1
+ 544: 17(ivec4) Load 19(ballot)
+ 545: 66(i8vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 543 544
+ 546: 67(ptr) AccessChain 37(data) 540 45
+ 547: 23(i8vec4) Load 546
+ 548: 23(i8vec4) VectorShuffle 547 545 4 5 2 3
+ Store 546 548
+ 549: 6(int) Load 8(invocation)
+ 550: 67(ptr) AccessChain 37(data) 52 45
+ 551: 23(i8vec4) Load 550
+ 552: 72(i8vec3) VectorShuffle 551 551 0 1 2
+ 553: 17(ivec4) Load 19(ballot)
+ 554: 72(i8vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 552 553
+ 555: 67(ptr) AccessChain 37(data) 549 45
+ 556: 23(i8vec4) Load 555
+ 557: 23(i8vec4) VectorShuffle 556 554 4 5 6 3
+ Store 555 557
+ 558: 6(int) Load 8(invocation)
+ 559: 67(ptr) AccessChain 37(data) 58 45
+ 560: 23(i8vec4) Load 559
+ 561: 17(ivec4) Load 19(ballot)
+ 562: 23(i8vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 560 561
+ 563: 67(ptr) AccessChain 37(data) 558 45
+ Store 563 562
+ 564: 6(int) Load 8(invocation)
+ 565: 62(ptr) AccessChain 37(data) 39 45 40
+ 566: 22(int8_t) Load 565
+ 567: 17(ivec4) Load 19(ballot)
+ 568: 22(int8_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 566 567
+ 569: 62(ptr) AccessChain 37(data) 564 45 40
+ Store 569 568
+ 570: 6(int) Load 8(invocation)
+ 571: 67(ptr) AccessChain 37(data) 45 45
+ 572: 23(i8vec4) Load 571
+ 573: 66(i8vec2) VectorShuffle 572 572 0 1
+ 574: 17(ivec4) Load 19(ballot)
+ 575: 66(i8vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 573 574
+ 576: 67(ptr) AccessChain 37(data) 570 45
+ 577: 23(i8vec4) Load 576
+ 578: 23(i8vec4) VectorShuffle 577 575 4 5 2 3
+ Store 576 578
+ 579: 6(int) Load 8(invocation)
+ 580: 67(ptr) AccessChain 37(data) 52 45
+ 581: 23(i8vec4) Load 580
+ 582: 72(i8vec3) VectorShuffle 581 581 0 1 2
+ 583: 17(ivec4) Load 19(ballot)
+ 584: 72(i8vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 582 583
+ 585: 67(ptr) AccessChain 37(data) 579 45
+ 586: 23(i8vec4) Load 585
+ 587: 23(i8vec4) VectorShuffle 586 584 4 5 6 3
+ Store 585 587
+ 588: 6(int) Load 8(invocation)
+ 589: 67(ptr) AccessChain 37(data) 58 45
+ 590: 23(i8vec4) Load 589
+ 591: 17(ivec4) Load 19(ballot)
+ 592: 23(i8vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 590 591
+ 593: 67(ptr) AccessChain 37(data) 588 45
+ Store 593 592
+ 594: 6(int) Load 8(invocation)
+ 595: 80(ptr) AccessChain 37(data) 39 52 40
+ 596: 24(int16_t) Load 595
+ 597: 17(ivec4) Load 19(ballot)
+ 598: 24(int16_t) GroupNonUniformIAdd 177 PartitionedReduceNV 596 597
+ 599: 80(ptr) AccessChain 37(data) 594 52 40
+ Store 599 598
+ 600: 6(int) Load 8(invocation)
+ 601: 85(ptr) AccessChain 37(data) 45 52
+ 602: 25(i16vec4) Load 601
+ 603: 84(i16vec2) VectorShuffle 602 602 0 1
+ 604: 17(ivec4) Load 19(ballot)
+ 605: 84(i16vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 603 604
+ 606: 85(ptr) AccessChain 37(data) 600 52
+ 607: 25(i16vec4) Load 606
+ 608: 25(i16vec4) VectorShuffle 607 605 4 5 2 3
+ Store 606 608
+ 609: 6(int) Load 8(invocation)
+ 610: 85(ptr) AccessChain 37(data) 52 52
+ 611: 25(i16vec4) Load 610
+ 612: 90(i16vec3) VectorShuffle 611 611 0 1 2
+ 613: 17(ivec4) Load 19(ballot)
+ 614: 90(i16vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 612 613
+ 615: 85(ptr) AccessChain 37(data) 609 52
+ 616: 25(i16vec4) Load 615
+ 617: 25(i16vec4) VectorShuffle 616 614 4 5 6 3
+ Store 615 617
+ 618: 6(int) Load 8(invocation)
+ 619: 85(ptr) AccessChain 37(data) 58 52
+ 620: 25(i16vec4) Load 619
+ 621: 17(ivec4) Load 19(ballot)
+ 622: 25(i16vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 620 621
+ 623: 85(ptr) AccessChain 37(data) 618 52
+ Store 623 622
+ 624: 6(int) Load 8(invocation)
+ 625: 80(ptr) AccessChain 37(data) 39 52 40
+ 626: 24(int16_t) Load 625
+ 627: 17(ivec4) Load 19(ballot)
+ 628: 24(int16_t) GroupNonUniformIMul 177 PartitionedReduceNV 626 627
+ 629: 80(ptr) AccessChain 37(data) 624 52 40
+ Store 629 628
+ 630: 6(int) Load 8(invocation)
+ 631: 85(ptr) AccessChain 37(data) 45 52
+ 632: 25(i16vec4) Load 631
+ 633: 84(i16vec2) VectorShuffle 632 632 0 1
+ 634: 17(ivec4) Load 19(ballot)
+ 635: 84(i16vec2) GroupNonUniformIMul 177 PartitionedReduceNV 633 634
+ 636: 85(ptr) AccessChain 37(data) 630 52
+ 637: 25(i16vec4) Load 636
+ 638: 25(i16vec4) VectorShuffle 637 635 4 5 2 3
+ Store 636 638
+ 639: 6(int) Load 8(invocation)
+ 640: 85(ptr) AccessChain 37(data) 52 52
+ 641: 25(i16vec4) Load 640
+ 642: 90(i16vec3) VectorShuffle 641 641 0 1 2
+ 643: 17(ivec4) Load 19(ballot)
+ 644: 90(i16vec3) GroupNonUniformIMul 177 PartitionedReduceNV 642 643
+ 645: 85(ptr) AccessChain 37(data) 639 52
+ 646: 25(i16vec4) Load 645
+ 647: 25(i16vec4) VectorShuffle 646 644 4 5 6 3
+ Store 645 647
+ 648: 6(int) Load 8(invocation)
+ 649: 85(ptr) AccessChain 37(data) 58 52
+ 650: 25(i16vec4) Load 649
+ 651: 17(ivec4) Load 19(ballot)
+ 652: 25(i16vec4) GroupNonUniformIMul 177 PartitionedReduceNV 650 651
+ 653: 85(ptr) AccessChain 37(data) 648 52
+ Store 653 652
+ 654: 6(int) Load 8(invocation)
+ 655: 80(ptr) AccessChain 37(data) 39 52 40
+ 656: 24(int16_t) Load 655
+ 657: 17(ivec4) Load 19(ballot)
+ 658: 24(int16_t) GroupNonUniformSMin 177 PartitionedReduceNV 656 657
+ 659: 80(ptr) AccessChain 37(data) 654 52 40
+ Store 659 658
+ 660: 6(int) Load 8(invocation)
+ 661: 85(ptr) AccessChain 37(data) 45 52
+ 662: 25(i16vec4) Load 661
+ 663: 84(i16vec2) VectorShuffle 662 662 0 1
+ 664: 17(ivec4) Load 19(ballot)
+ 665: 84(i16vec2) GroupNonUniformSMin 177 PartitionedReduceNV 663 664
+ 666: 85(ptr) AccessChain 37(data) 660 52
+ 667: 25(i16vec4) Load 666
+ 668: 25(i16vec4) VectorShuffle 667 665 4 5 2 3
+ Store 666 668
+ 669: 6(int) Load 8(invocation)
+ 670: 85(ptr) AccessChain 37(data) 52 52
+ 671: 25(i16vec4) Load 670
+ 672: 90(i16vec3) VectorShuffle 671 671 0 1 2
+ 673: 17(ivec4) Load 19(ballot)
+ 674: 90(i16vec3) GroupNonUniformSMin 177 PartitionedReduceNV 672 673
+ 675: 85(ptr) AccessChain 37(data) 669 52
+ 676: 25(i16vec4) Load 675
+ 677: 25(i16vec4) VectorShuffle 676 674 4 5 6 3
+ Store 675 677
+ 678: 6(int) Load 8(invocation)
+ 679: 85(ptr) AccessChain 37(data) 58 52
+ 680: 25(i16vec4) Load 679
+ 681: 17(ivec4) Load 19(ballot)
+ 682: 25(i16vec4) GroupNonUniformSMin 177 PartitionedReduceNV 680 681
+ 683: 85(ptr) AccessChain 37(data) 678 52
+ Store 683 682
+ 684: 6(int) Load 8(invocation)
+ 685: 80(ptr) AccessChain 37(data) 39 52 40
+ 686: 24(int16_t) Load 685
+ 687: 17(ivec4) Load 19(ballot)
+ 688: 24(int16_t) GroupNonUniformSMax 177 PartitionedReduceNV 686 687
+ 689: 80(ptr) AccessChain 37(data) 684 52 40
+ Store 689 688
+ 690: 6(int) Load 8(invocation)
+ 691: 85(ptr) AccessChain 37(data) 45 52
+ 692: 25(i16vec4) Load 691
+ 693: 84(i16vec2) VectorShuffle 692 692 0 1
+ 694: 17(ivec4) Load 19(ballot)
+ 695: 84(i16vec2) GroupNonUniformSMax 177 PartitionedReduceNV 693 694
+ 696: 85(ptr) AccessChain 37(data) 690 52
+ 697: 25(i16vec4) Load 696
+ 698: 25(i16vec4) VectorShuffle 697 695 4 5 2 3
+ Store 696 698
+ 699: 6(int) Load 8(invocation)
+ 700: 85(ptr) AccessChain 37(data) 52 52
+ 701: 25(i16vec4) Load 700
+ 702: 90(i16vec3) VectorShuffle 701 701 0 1 2
+ 703: 17(ivec4) Load 19(ballot)
+ 704: 90(i16vec3) GroupNonUniformSMax 177 PartitionedReduceNV 702 703
+ 705: 85(ptr) AccessChain 37(data) 699 52
+ 706: 25(i16vec4) Load 705
+ 707: 25(i16vec4) VectorShuffle 706 704 4 5 6 3
+ Store 705 707
+ 708: 6(int) Load 8(invocation)
+ 709: 85(ptr) AccessChain 37(data) 58 52
+ 710: 25(i16vec4) Load 709
+ 711: 17(ivec4) Load 19(ballot)
+ 712: 25(i16vec4) GroupNonUniformSMax 177 PartitionedReduceNV 710 711
+ 713: 85(ptr) AccessChain 37(data) 708 52
+ Store 713 712
+ 714: 6(int) Load 8(invocation)
+ 715: 80(ptr) AccessChain 37(data) 39 52 40
+ 716: 24(int16_t) Load 715
+ 717: 17(ivec4) Load 19(ballot)
+ 718: 24(int16_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 716 717
+ 719: 80(ptr) AccessChain 37(data) 714 52 40
+ Store 719 718
+ 720: 6(int) Load 8(invocation)
+ 721: 85(ptr) AccessChain 37(data) 45 52
+ 722: 25(i16vec4) Load 721
+ 723: 84(i16vec2) VectorShuffle 722 722 0 1
+ 724: 17(ivec4) Load 19(ballot)
+ 725: 84(i16vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 723 724
+ 726: 85(ptr) AccessChain 37(data) 720 52
+ 727: 25(i16vec4) Load 726
+ 728: 25(i16vec4) VectorShuffle 727 725 4 5 2 3
+ Store 726 728
+ 729: 6(int) Load 8(invocation)
+ 730: 85(ptr) AccessChain 37(data) 52 52
+ 731: 25(i16vec4) Load 730
+ 732: 90(i16vec3) VectorShuffle 731 731 0 1 2
+ 733: 17(ivec4) Load 19(ballot)
+ 734: 90(i16vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 732 733
+ 735: 85(ptr) AccessChain 37(data) 729 52
+ 736: 25(i16vec4) Load 735
+ 737: 25(i16vec4) VectorShuffle 736 734 4 5 6 3
+ Store 735 737
+ 738: 6(int) Load 8(invocation)
+ 739: 85(ptr) AccessChain 37(data) 58 52
+ 740: 25(i16vec4) Load 739
+ 741: 17(ivec4) Load 19(ballot)
+ 742: 25(i16vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 740 741
+ 743: 85(ptr) AccessChain 37(data) 738 52
+ Store 743 742
+ 744: 6(int) Load 8(invocation)
+ 745: 80(ptr) AccessChain 37(data) 39 52 40
+ 746: 24(int16_t) Load 745
+ 747: 17(ivec4) Load 19(ballot)
+ 748: 24(int16_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 746 747
+ 749: 80(ptr) AccessChain 37(data) 744 52 40
+ Store 749 748
+ 750: 6(int) Load 8(invocation)
+ 751: 85(ptr) AccessChain 37(data) 45 52
+ 752: 25(i16vec4) Load 751
+ 753: 84(i16vec2) VectorShuffle 752 752 0 1
+ 754: 17(ivec4) Load 19(ballot)
+ 755: 84(i16vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 753 754
+ 756: 85(ptr) AccessChain 37(data) 750 52
+ 757: 25(i16vec4) Load 756
+ 758: 25(i16vec4) VectorShuffle 757 755 4 5 2 3
+ Store 756 758
+ 759: 6(int) Load 8(invocation)
+ 760: 85(ptr) AccessChain 37(data) 52 52
+ 761: 25(i16vec4) Load 760
+ 762: 90(i16vec3) VectorShuffle 761 761 0 1 2
+ 763: 17(ivec4) Load 19(ballot)
+ 764: 90(i16vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 762 763
+ 765: 85(ptr) AccessChain 37(data) 759 52
+ 766: 25(i16vec4) Load 765
+ 767: 25(i16vec4) VectorShuffle 766 764 4 5 6 3
+ Store 765 767
+ 768: 6(int) Load 8(invocation)
+ 769: 85(ptr) AccessChain 37(data) 58 52
+ 770: 25(i16vec4) Load 769
+ 771: 17(ivec4) Load 19(ballot)
+ 772: 25(i16vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 770 771
+ 773: 85(ptr) AccessChain 37(data) 768 52
+ Store 773 772
+ 774: 6(int) Load 8(invocation)
+ 775: 80(ptr) AccessChain 37(data) 39 52 40
+ 776: 24(int16_t) Load 775
+ 777: 17(ivec4) Load 19(ballot)
+ 778: 24(int16_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 776 777
+ 779: 80(ptr) AccessChain 37(data) 774 52 40
+ Store 779 778
+ 780: 6(int) Load 8(invocation)
+ 781: 85(ptr) AccessChain 37(data) 45 52
+ 782: 25(i16vec4) Load 781
+ 783: 84(i16vec2) VectorShuffle 782 782 0 1
+ 784: 17(ivec4) Load 19(ballot)
+ 785: 84(i16vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 783 784
+ 786: 85(ptr) AccessChain 37(data) 780 52
+ 787: 25(i16vec4) Load 786
+ 788: 25(i16vec4) VectorShuffle 787 785 4 5 2 3
+ Store 786 788
+ 789: 6(int) Load 8(invocation)
+ 790: 85(ptr) AccessChain 37(data) 52 52
+ 791: 25(i16vec4) Load 790
+ 792: 90(i16vec3) VectorShuffle 791 791 0 1 2
+ 793: 17(ivec4) Load 19(ballot)
+ 794: 90(i16vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 792 793
+ 795: 85(ptr) AccessChain 37(data) 789 52
+ 796: 25(i16vec4) Load 795
+ 797: 25(i16vec4) VectorShuffle 796 794 4 5 6 3
+ Store 795 797
+ 798: 6(int) Load 8(invocation)
+ 799: 85(ptr) AccessChain 37(data) 58 52
+ 800: 25(i16vec4) Load 799
+ 801: 17(ivec4) Load 19(ballot)
+ 802: 25(i16vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 800 801
+ 803: 85(ptr) AccessChain 37(data) 798 52
+ Store 803 802
+ 804: 6(int) Load 8(invocation)
+ 805: 98(ptr) AccessChain 37(data) 39 58 40
+ 806: 26(int16_t) Load 805
+ 807: 17(ivec4) Load 19(ballot)
+ 808: 26(int16_t) GroupNonUniformIAdd 177 PartitionedReduceNV 806 807
+ 809: 98(ptr) AccessChain 37(data) 804 58 40
+ Store 809 808
+ 810: 6(int) Load 8(invocation)
+ 811: 103(ptr) AccessChain 37(data) 45 58
+ 812: 27(i16vec4) Load 811
+ 813:102(i16vec2) VectorShuffle 812 812 0 1
+ 814: 17(ivec4) Load 19(ballot)
+ 815:102(i16vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 813 814
+ 816: 103(ptr) AccessChain 37(data) 810 58
+ 817: 27(i16vec4) Load 816
+ 818: 27(i16vec4) VectorShuffle 817 815 4 5 2 3
+ Store 816 818
+ 819: 6(int) Load 8(invocation)
+ 820: 103(ptr) AccessChain 37(data) 52 58
+ 821: 27(i16vec4) Load 820
+ 822:108(i16vec3) VectorShuffle 821 821 0 1 2
+ 823: 17(ivec4) Load 19(ballot)
+ 824:108(i16vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 822 823
+ 825: 103(ptr) AccessChain 37(data) 819 58
+ 826: 27(i16vec4) Load 825
+ 827: 27(i16vec4) VectorShuffle 826 824 4 5 6 3
+ Store 825 827
+ 828: 6(int) Load 8(invocation)
+ 829: 103(ptr) AccessChain 37(data) 58 58
+ 830: 27(i16vec4) Load 829
+ 831: 17(ivec4) Load 19(ballot)
+ 832: 27(i16vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 830 831
+ 833: 103(ptr) AccessChain 37(data) 828 58
+ Store 833 832
+ 834: 6(int) Load 8(invocation)
+ 835: 98(ptr) AccessChain 37(data) 39 58 40
+ 836: 26(int16_t) Load 835
+ 837: 17(ivec4) Load 19(ballot)
+ 838: 26(int16_t) GroupNonUniformIMul 177 PartitionedReduceNV 836 837
+ 839: 98(ptr) AccessChain 37(data) 834 58 40
+ Store 839 838
+ 840: 6(int) Load 8(invocation)
+ 841: 103(ptr) AccessChain 37(data) 45 58
+ 842: 27(i16vec4) Load 841
+ 843:102(i16vec2) VectorShuffle 842 842 0 1
+ 844: 17(ivec4) Load 19(ballot)
+ 845:102(i16vec2) GroupNonUniformIMul 177 PartitionedReduceNV 843 844
+ 846: 103(ptr) AccessChain 37(data) 840 58
+ 847: 27(i16vec4) Load 846
+ 848: 27(i16vec4) VectorShuffle 847 845 4 5 2 3
+ Store 846 848
+ 849: 6(int) Load 8(invocation)
+ 850: 103(ptr) AccessChain 37(data) 52 58
+ 851: 27(i16vec4) Load 850
+ 852:108(i16vec3) VectorShuffle 851 851 0 1 2
+ 853: 17(ivec4) Load 19(ballot)
+ 854:108(i16vec3) GroupNonUniformIMul 177 PartitionedReduceNV 852 853
+ 855: 103(ptr) AccessChain 37(data) 849 58
+ 856: 27(i16vec4) Load 855
+ 857: 27(i16vec4) VectorShuffle 856 854 4 5 6 3
+ Store 855 857
+ 858: 6(int) Load 8(invocation)
+ 859: 103(ptr) AccessChain 37(data) 58 58
+ 860: 27(i16vec4) Load 859
+ 861: 17(ivec4) Load 19(ballot)
+ 862: 27(i16vec4) GroupNonUniformIMul 177 PartitionedReduceNV 860 861
+ 863: 103(ptr) AccessChain 37(data) 858 58
+ Store 863 862
+ 864: 6(int) Load 8(invocation)
+ 865: 98(ptr) AccessChain 37(data) 39 58 40
+ 866: 26(int16_t) Load 865
+ 867: 17(ivec4) Load 19(ballot)
+ 868: 26(int16_t) GroupNonUniformSMin 177 PartitionedReduceNV 866 867
+ 869: 98(ptr) AccessChain 37(data) 864 58 40
+ Store 869 868
+ 870: 6(int) Load 8(invocation)
+ 871: 103(ptr) AccessChain 37(data) 45 58
+ 872: 27(i16vec4) Load 871
+ 873:102(i16vec2) VectorShuffle 872 872 0 1
+ 874: 17(ivec4) Load 19(ballot)
+ 875:102(i16vec2) GroupNonUniformSMin 177 PartitionedReduceNV 873 874
+ 876: 103(ptr) AccessChain 37(data) 870 58
+ 877: 27(i16vec4) Load 876
+ 878: 27(i16vec4) VectorShuffle 877 875 4 5 2 3
+ Store 876 878
+ 879: 6(int) Load 8(invocation)
+ 880: 103(ptr) AccessChain 37(data) 52 58
+ 881: 27(i16vec4) Load 880
+ 882:108(i16vec3) VectorShuffle 881 881 0 1 2
+ 883: 17(ivec4) Load 19(ballot)
+ 884:108(i16vec3) GroupNonUniformSMin 177 PartitionedReduceNV 882 883
+ 885: 103(ptr) AccessChain 37(data) 879 58
+ 886: 27(i16vec4) Load 885
+ 887: 27(i16vec4) VectorShuffle 886 884 4 5 6 3
+ Store 885 887
+ 888: 6(int) Load 8(invocation)
+ 889: 103(ptr) AccessChain 37(data) 58 58
+ 890: 27(i16vec4) Load 889
+ 891: 17(ivec4) Load 19(ballot)
+ 892: 27(i16vec4) GroupNonUniformSMin 177 PartitionedReduceNV 890 891
+ 893: 103(ptr) AccessChain 37(data) 888 58
+ Store 893 892
+ 894: 6(int) Load 8(invocation)
+ 895: 98(ptr) AccessChain 37(data) 39 58 40
+ 896: 26(int16_t) Load 895
+ 897: 17(ivec4) Load 19(ballot)
+ 898: 26(int16_t) GroupNonUniformSMax 177 PartitionedReduceNV 896 897
+ 899: 98(ptr) AccessChain 37(data) 894 58 40
+ Store 899 898
+ 900: 6(int) Load 8(invocation)
+ 901: 103(ptr) AccessChain 37(data) 45 58
+ 902: 27(i16vec4) Load 901
+ 903:102(i16vec2) VectorShuffle 902 902 0 1
+ 904: 17(ivec4) Load 19(ballot)
+ 905:102(i16vec2) GroupNonUniformSMax 177 PartitionedReduceNV 903 904
+ 906: 103(ptr) AccessChain 37(data) 900 58
+ 907: 27(i16vec4) Load 906
+ 908: 27(i16vec4) VectorShuffle 907 905 4 5 2 3
+ Store 906 908
+ 909: 6(int) Load 8(invocation)
+ 910: 103(ptr) AccessChain 37(data) 52 58
+ 911: 27(i16vec4) Load 910
+ 912:108(i16vec3) VectorShuffle 911 911 0 1 2
+ 913: 17(ivec4) Load 19(ballot)
+ 914:108(i16vec3) GroupNonUniformSMax 177 PartitionedReduceNV 912 913
+ 915: 103(ptr) AccessChain 37(data) 909 58
+ 916: 27(i16vec4) Load 915
+ 917: 27(i16vec4) VectorShuffle 916 914 4 5 6 3
+ Store 915 917
+ 918: 6(int) Load 8(invocation)
+ 919: 103(ptr) AccessChain 37(data) 58 58
+ 920: 27(i16vec4) Load 919
+ 921: 17(ivec4) Load 19(ballot)
+ 922: 27(i16vec4) GroupNonUniformSMax 177 PartitionedReduceNV 920 921
+ 923: 103(ptr) AccessChain 37(data) 918 58
+ Store 923 922
+ 924: 6(int) Load 8(invocation)
+ 925: 98(ptr) AccessChain 37(data) 39 58 40
+ 926: 26(int16_t) Load 925
+ 927: 17(ivec4) Load 19(ballot)
+ 928: 26(int16_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 926 927
+ 929: 98(ptr) AccessChain 37(data) 924 58 40
+ Store 929 928
+ 930: 6(int) Load 8(invocation)
+ 931: 103(ptr) AccessChain 37(data) 45 58
+ 932: 27(i16vec4) Load 931
+ 933:102(i16vec2) VectorShuffle 932 932 0 1
+ 934: 17(ivec4) Load 19(ballot)
+ 935:102(i16vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 933 934
+ 936: 103(ptr) AccessChain 37(data) 930 58
+ 937: 27(i16vec4) Load 936
+ 938: 27(i16vec4) VectorShuffle 937 935 4 5 2 3
+ Store 936 938
+ 939: 6(int) Load 8(invocation)
+ 940: 103(ptr) AccessChain 37(data) 52 58
+ 941: 27(i16vec4) Load 940
+ 942:108(i16vec3) VectorShuffle 941 941 0 1 2
+ 943: 17(ivec4) Load 19(ballot)
+ 944:108(i16vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 942 943
+ 945: 103(ptr) AccessChain 37(data) 939 58
+ 946: 27(i16vec4) Load 945
+ 947: 27(i16vec4) VectorShuffle 946 944 4 5 6 3
+ Store 945 947
+ 948: 6(int) Load 8(invocation)
+ 949: 103(ptr) AccessChain 37(data) 58 58
+ 950: 27(i16vec4) Load 949
+ 951: 17(ivec4) Load 19(ballot)
+ 952: 27(i16vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 950 951
+ 953: 103(ptr) AccessChain 37(data) 948 58
+ Store 953 952
+ 954: 6(int) Load 8(invocation)
+ 955: 98(ptr) AccessChain 37(data) 39 58 40
+ 956: 26(int16_t) Load 955
+ 957: 17(ivec4) Load 19(ballot)
+ 958: 26(int16_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 956 957
+ 959: 98(ptr) AccessChain 37(data) 954 58 40
+ Store 959 958
+ 960: 6(int) Load 8(invocation)
+ 961: 103(ptr) AccessChain 37(data) 45 58
+ 962: 27(i16vec4) Load 961
+ 963:102(i16vec2) VectorShuffle 962 962 0 1
+ 964: 17(ivec4) Load 19(ballot)
+ 965:102(i16vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 963 964
+ 966: 103(ptr) AccessChain 37(data) 960 58
+ 967: 27(i16vec4) Load 966
+ 968: 27(i16vec4) VectorShuffle 967 965 4 5 2 3
+ Store 966 968
+ 969: 6(int) Load 8(invocation)
+ 970: 103(ptr) AccessChain 37(data) 52 58
+ 971: 27(i16vec4) Load 970
+ 972:108(i16vec3) VectorShuffle 971 971 0 1 2
+ 973: 17(ivec4) Load 19(ballot)
+ 974:108(i16vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 972 973
+ 975: 103(ptr) AccessChain 37(data) 969 58
+ 976: 27(i16vec4) Load 975
+ 977: 27(i16vec4) VectorShuffle 976 974 4 5 6 3
+ Store 975 977
+ 978: 6(int) Load 8(invocation)
+ 979: 103(ptr) AccessChain 37(data) 58 58
+ 980: 27(i16vec4) Load 979
+ 981: 17(ivec4) Load 19(ballot)
+ 982: 27(i16vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 980 981
+ 983: 103(ptr) AccessChain 37(data) 978 58
+ Store 983 982
+ 984: 6(int) Load 8(invocation)
+ 985: 98(ptr) AccessChain 37(data) 39 58 40
+ 986: 26(int16_t) Load 985
+ 987: 17(ivec4) Load 19(ballot)
+ 988: 26(int16_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 986 987
+ 989: 98(ptr) AccessChain 37(data) 984 58 40
+ Store 989 988
+ 990: 6(int) Load 8(invocation)
+ 991: 103(ptr) AccessChain 37(data) 45 58
+ 992: 27(i16vec4) Load 991
+ 993:102(i16vec2) VectorShuffle 992 992 0 1
+ 994: 17(ivec4) Load 19(ballot)
+ 995:102(i16vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 993 994
+ 996: 103(ptr) AccessChain 37(data) 990 58
+ 997: 27(i16vec4) Load 996
+ 998: 27(i16vec4) VectorShuffle 997 995 4 5 2 3
+ Store 996 998
+ 999: 6(int) Load 8(invocation)
+ 1000: 103(ptr) AccessChain 37(data) 52 58
+ 1001: 27(i16vec4) Load 1000
+ 1002:108(i16vec3) VectorShuffle 1001 1001 0 1 2
+ 1003: 17(ivec4) Load 19(ballot)
+ 1004:108(i16vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1002 1003
+ 1005: 103(ptr) AccessChain 37(data) 999 58
+ 1006: 27(i16vec4) Load 1005
+ 1007: 27(i16vec4) VectorShuffle 1006 1004 4 5 6 3
+ Store 1005 1007
+ 1008: 6(int) Load 8(invocation)
+ 1009: 103(ptr) AccessChain 37(data) 58 58
+ 1010: 27(i16vec4) Load 1009
+ 1011: 17(ivec4) Load 19(ballot)
+ 1012: 27(i16vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1010 1011
+ 1013: 103(ptr) AccessChain 37(data) 1008 58
+ Store 1013 1012
+ 1014: 6(int) Load 8(invocation)
+ 1015: 117(ptr) AccessChain 37(data) 39 116 40
+ 1016: 28(int64_t) Load 1015
+ 1017: 17(ivec4) Load 19(ballot)
+ 1018: 28(int64_t) GroupNonUniformIAdd 177 PartitionedReduceNV 1016 1017
+ 1019: 117(ptr) AccessChain 37(data) 1014 116 40
+ Store 1019 1018
+ 1020: 6(int) Load 8(invocation)
+ 1021: 122(ptr) AccessChain 37(data) 45 116
+ 1022: 29(i64vec4) Load 1021
+ 1023:121(i64vec2) VectorShuffle 1022 1022 0 1
+ 1024: 17(ivec4) Load 19(ballot)
+ 1025:121(i64vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 1023 1024
+ 1026: 122(ptr) AccessChain 37(data) 1020 116
+ 1027: 29(i64vec4) Load 1026
+ 1028: 29(i64vec4) VectorShuffle 1027 1025 4 5 2 3
+ Store 1026 1028
+ 1029: 6(int) Load 8(invocation)
+ 1030: 122(ptr) AccessChain 37(data) 52 116
+ 1031: 29(i64vec4) Load 1030
+ 1032:127(i64vec3) VectorShuffle 1031 1031 0 1 2
+ 1033: 17(ivec4) Load 19(ballot)
+ 1034:127(i64vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 1032 1033
+ 1035: 122(ptr) AccessChain 37(data) 1029 116
+ 1036: 29(i64vec4) Load 1035
+ 1037: 29(i64vec4) VectorShuffle 1036 1034 4 5 6 3
+ Store 1035 1037
+ 1038: 6(int) Load 8(invocation)
+ 1039: 122(ptr) AccessChain 37(data) 58 116
+ 1040: 29(i64vec4) Load 1039
+ 1041: 17(ivec4) Load 19(ballot)
+ 1042: 29(i64vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 1040 1041
+ 1043: 122(ptr) AccessChain 37(data) 1038 116
+ Store 1043 1042
+ 1044: 6(int) Load 8(invocation)
+ 1045: 117(ptr) AccessChain 37(data) 39 116 40
+ 1046: 28(int64_t) Load 1045
+ 1047: 17(ivec4) Load 19(ballot)
+ 1048: 28(int64_t) GroupNonUniformIMul 177 PartitionedReduceNV 1046 1047
+ 1049: 117(ptr) AccessChain 37(data) 1044 116 40
+ Store 1049 1048
+ 1050: 6(int) Load 8(invocation)
+ 1051: 122(ptr) AccessChain 37(data) 45 116
+ 1052: 29(i64vec4) Load 1051
+ 1053:121(i64vec2) VectorShuffle 1052 1052 0 1
+ 1054: 17(ivec4) Load 19(ballot)
+ 1055:121(i64vec2) GroupNonUniformIMul 177 PartitionedReduceNV 1053 1054
+ 1056: 122(ptr) AccessChain 37(data) 1050 116
+ 1057: 29(i64vec4) Load 1056
+ 1058: 29(i64vec4) VectorShuffle 1057 1055 4 5 2 3
+ Store 1056 1058
+ 1059: 6(int) Load 8(invocation)
+ 1060: 122(ptr) AccessChain 37(data) 52 116
+ 1061: 29(i64vec4) Load 1060
+ 1062:127(i64vec3) VectorShuffle 1061 1061 0 1 2
+ 1063: 17(ivec4) Load 19(ballot)
+ 1064:127(i64vec3) GroupNonUniformIMul 177 PartitionedReduceNV 1062 1063
+ 1065: 122(ptr) AccessChain 37(data) 1059 116
+ 1066: 29(i64vec4) Load 1065
+ 1067: 29(i64vec4) VectorShuffle 1066 1064 4 5 6 3
+ Store 1065 1067
+ 1068: 6(int) Load 8(invocation)
+ 1069: 122(ptr) AccessChain 37(data) 58 116
+ 1070: 29(i64vec4) Load 1069
+ 1071: 17(ivec4) Load 19(ballot)
+ 1072: 29(i64vec4) GroupNonUniformIMul 177 PartitionedReduceNV 1070 1071
+ 1073: 122(ptr) AccessChain 37(data) 1068 116
+ Store 1073 1072
+ 1074: 6(int) Load 8(invocation)
+ 1075: 117(ptr) AccessChain 37(data) 39 116 40
+ 1076: 28(int64_t) Load 1075
+ 1077: 17(ivec4) Load 19(ballot)
+ 1078: 28(int64_t) GroupNonUniformSMin 177 PartitionedReduceNV 1076 1077
+ 1079: 117(ptr) AccessChain 37(data) 1074 116 40
+ Store 1079 1078
+ 1080: 6(int) Load 8(invocation)
+ 1081: 122(ptr) AccessChain 37(data) 45 116
+ 1082: 29(i64vec4) Load 1081
+ 1083:121(i64vec2) VectorShuffle 1082 1082 0 1
+ 1084: 17(ivec4) Load 19(ballot)
+ 1085:121(i64vec2) GroupNonUniformSMin 177 PartitionedReduceNV 1083 1084
+ 1086: 122(ptr) AccessChain 37(data) 1080 116
+ 1087: 29(i64vec4) Load 1086
+ 1088: 29(i64vec4) VectorShuffle 1087 1085 4 5 2 3
+ Store 1086 1088
+ 1089: 6(int) Load 8(invocation)
+ 1090: 122(ptr) AccessChain 37(data) 52 116
+ 1091: 29(i64vec4) Load 1090
+ 1092:127(i64vec3) VectorShuffle 1091 1091 0 1 2
+ 1093: 17(ivec4) Load 19(ballot)
+ 1094:127(i64vec3) GroupNonUniformSMin 177 PartitionedReduceNV 1092 1093
+ 1095: 122(ptr) AccessChain 37(data) 1089 116
+ 1096: 29(i64vec4) Load 1095
+ 1097: 29(i64vec4) VectorShuffle 1096 1094 4 5 6 3
+ Store 1095 1097
+ 1098: 6(int) Load 8(invocation)
+ 1099: 122(ptr) AccessChain 37(data) 58 116
+ 1100: 29(i64vec4) Load 1099
+ 1101: 17(ivec4) Load 19(ballot)
+ 1102: 29(i64vec4) GroupNonUniformSMin 177 PartitionedReduceNV 1100 1101
+ 1103: 122(ptr) AccessChain 37(data) 1098 116
+ Store 1103 1102
+ 1104: 6(int) Load 8(invocation)
+ 1105: 117(ptr) AccessChain 37(data) 39 116 40
+ 1106: 28(int64_t) Load 1105
+ 1107: 17(ivec4) Load 19(ballot)
+ 1108: 28(int64_t) GroupNonUniformSMax 177 PartitionedReduceNV 1106 1107
+ 1109: 117(ptr) AccessChain 37(data) 1104 116 40
+ Store 1109 1108
+ 1110: 6(int) Load 8(invocation)
+ 1111: 122(ptr) AccessChain 37(data) 45 116
+ 1112: 29(i64vec4) Load 1111
+ 1113:121(i64vec2) VectorShuffle 1112 1112 0 1
+ 1114: 17(ivec4) Load 19(ballot)
+ 1115:121(i64vec2) GroupNonUniformSMax 177 PartitionedReduceNV 1113 1114
+ 1116: 122(ptr) AccessChain 37(data) 1110 116
+ 1117: 29(i64vec4) Load 1116
+ 1118: 29(i64vec4) VectorShuffle 1117 1115 4 5 2 3
+ Store 1116 1118
+ 1119: 6(int) Load 8(invocation)
+ 1120: 122(ptr) AccessChain 37(data) 52 116
+ 1121: 29(i64vec4) Load 1120
+ 1122:127(i64vec3) VectorShuffle 1121 1121 0 1 2
+ 1123: 17(ivec4) Load 19(ballot)
+ 1124:127(i64vec3) GroupNonUniformSMax 177 PartitionedReduceNV 1122 1123
+ 1125: 122(ptr) AccessChain 37(data) 1119 116
+ 1126: 29(i64vec4) Load 1125
+ 1127: 29(i64vec4) VectorShuffle 1126 1124 4 5 6 3
+ Store 1125 1127
+ 1128: 6(int) Load 8(invocation)
+ 1129: 122(ptr) AccessChain 37(data) 58 116
+ 1130: 29(i64vec4) Load 1129
+ 1131: 17(ivec4) Load 19(ballot)
+ 1132: 29(i64vec4) GroupNonUniformSMax 177 PartitionedReduceNV 1130 1131
+ 1133: 122(ptr) AccessChain 37(data) 1128 116
+ Store 1133 1132
+ 1134: 6(int) Load 8(invocation)
+ 1135: 117(ptr) AccessChain 37(data) 39 116 40
+ 1136: 28(int64_t) Load 1135
+ 1137: 17(ivec4) Load 19(ballot)
+ 1138: 28(int64_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1136 1137
+ 1139: 117(ptr) AccessChain 37(data) 1134 116 40
+ Store 1139 1138
+ 1140: 6(int) Load 8(invocation)
+ 1141: 122(ptr) AccessChain 37(data) 45 116
+ 1142: 29(i64vec4) Load 1141
+ 1143:121(i64vec2) VectorShuffle 1142 1142 0 1
+ 1144: 17(ivec4) Load 19(ballot)
+ 1145:121(i64vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1143 1144
+ 1146: 122(ptr) AccessChain 37(data) 1140 116
+ 1147: 29(i64vec4) Load 1146
+ 1148: 29(i64vec4) VectorShuffle 1147 1145 4 5 2 3
+ Store 1146 1148
+ 1149: 6(int) Load 8(invocation)
+ 1150: 122(ptr) AccessChain 37(data) 52 116
+ 1151: 29(i64vec4) Load 1150
+ 1152:127(i64vec3) VectorShuffle 1151 1151 0 1 2
+ 1153: 17(ivec4) Load 19(ballot)
+ 1154:127(i64vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1152 1153
+ 1155: 122(ptr) AccessChain 37(data) 1149 116
+ 1156: 29(i64vec4) Load 1155
+ 1157: 29(i64vec4) VectorShuffle 1156 1154 4 5 6 3
+ Store 1155 1157
+ 1158: 6(int) Load 8(invocation)
+ 1159: 122(ptr) AccessChain 37(data) 58 116
+ 1160: 29(i64vec4) Load 1159
+ 1161: 17(ivec4) Load 19(ballot)
+ 1162: 29(i64vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1160 1161
+ 1163: 122(ptr) AccessChain 37(data) 1158 116
+ Store 1163 1162
+ 1164: 6(int) Load 8(invocation)
+ 1165: 117(ptr) AccessChain 37(data) 39 116 40
+ 1166: 28(int64_t) Load 1165
+ 1167: 17(ivec4) Load 19(ballot)
+ 1168: 28(int64_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1166 1167
+ 1169: 117(ptr) AccessChain 37(data) 1164 116 40
+ Store 1169 1168
+ 1170: 6(int) Load 8(invocation)
+ 1171: 122(ptr) AccessChain 37(data) 45 116
+ 1172: 29(i64vec4) Load 1171
+ 1173:121(i64vec2) VectorShuffle 1172 1172 0 1
+ 1174: 17(ivec4) Load 19(ballot)
+ 1175:121(i64vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1173 1174
+ 1176: 122(ptr) AccessChain 37(data) 1170 116
+ 1177: 29(i64vec4) Load 1176
+ 1178: 29(i64vec4) VectorShuffle 1177 1175 4 5 2 3
+ Store 1176 1178
+ 1179: 6(int) Load 8(invocation)
+ 1180: 122(ptr) AccessChain 37(data) 52 116
+ 1181: 29(i64vec4) Load 1180
+ 1182:127(i64vec3) VectorShuffle 1181 1181 0 1 2
+ 1183: 17(ivec4) Load 19(ballot)
+ 1184:127(i64vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1182 1183
+ 1185: 122(ptr) AccessChain 37(data) 1179 116
+ 1186: 29(i64vec4) Load 1185
+ 1187: 29(i64vec4) VectorShuffle 1186 1184 4 5 6 3
+ Store 1185 1187
+ 1188: 6(int) Load 8(invocation)
+ 1189: 122(ptr) AccessChain 37(data) 58 116
+ 1190: 29(i64vec4) Load 1189
+ 1191: 17(ivec4) Load 19(ballot)
+ 1192: 29(i64vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1190 1191
+ 1193: 122(ptr) AccessChain 37(data) 1188 116
+ Store 1193 1192
+ 1194: 6(int) Load 8(invocation)
+ 1195: 117(ptr) AccessChain 37(data) 39 116 40
+ 1196: 28(int64_t) Load 1195
+ 1197: 17(ivec4) Load 19(ballot)
+ 1198: 28(int64_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1196 1197
+ 1199: 117(ptr) AccessChain 37(data) 1194 116 40
+ Store 1199 1198
+ 1200: 6(int) Load 8(invocation)
+ 1201: 122(ptr) AccessChain 37(data) 45 116
+ 1202: 29(i64vec4) Load 1201
+ 1203:121(i64vec2) VectorShuffle 1202 1202 0 1
+ 1204: 17(ivec4) Load 19(ballot)
+ 1205:121(i64vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1203 1204
+ 1206: 122(ptr) AccessChain 37(data) 1200 116
+ 1207: 29(i64vec4) Load 1206
+ 1208: 29(i64vec4) VectorShuffle 1207 1205 4 5 2 3
+ Store 1206 1208
+ 1209: 6(int) Load 8(invocation)
+ 1210: 122(ptr) AccessChain 37(data) 52 116
+ 1211: 29(i64vec4) Load 1210
+ 1212:127(i64vec3) VectorShuffle 1211 1211 0 1 2
+ 1213: 17(ivec4) Load 19(ballot)
+ 1214:127(i64vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1212 1213
+ 1215: 122(ptr) AccessChain 37(data) 1209 116
+ 1216: 29(i64vec4) Load 1215
+ 1217: 29(i64vec4) VectorShuffle 1216 1214 4 5 6 3
+ Store 1215 1217
+ 1218: 6(int) Load 8(invocation)
+ 1219: 122(ptr) AccessChain 37(data) 58 116
+ 1220: 29(i64vec4) Load 1219
+ 1221: 17(ivec4) Load 19(ballot)
+ 1222: 29(i64vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1220 1221
+ 1223: 122(ptr) AccessChain 37(data) 1218 116
+ Store 1223 1222
+ 1224: 6(int) Load 8(invocation)
+ 1225: 136(ptr) AccessChain 37(data) 39 135 40
+ 1226: 30(int64_t) Load 1225
+ 1227: 17(ivec4) Load 19(ballot)
+ 1228: 30(int64_t) GroupNonUniformIAdd 177 PartitionedReduceNV 1226 1227
+ 1229: 136(ptr) AccessChain 37(data) 1224 135 40
+ Store 1229 1228
+ 1230: 6(int) Load 8(invocation)
+ 1231: 141(ptr) AccessChain 37(data) 45 135
+ 1232: 31(i64vec4) Load 1231
+ 1233:140(i64vec2) VectorShuffle 1232 1232 0 1
+ 1234: 17(ivec4) Load 19(ballot)
+ 1235:140(i64vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 1233 1234
+ 1236: 141(ptr) AccessChain 37(data) 1230 135
+ 1237: 31(i64vec4) Load 1236
+ 1238: 31(i64vec4) VectorShuffle 1237 1235 4 5 2 3
+ Store 1236 1238
+ 1239: 6(int) Load 8(invocation)
+ 1240: 141(ptr) AccessChain 37(data) 52 135
+ 1241: 31(i64vec4) Load 1240
+ 1242:146(i64vec3) VectorShuffle 1241 1241 0 1 2
+ 1243: 17(ivec4) Load 19(ballot)
+ 1244:146(i64vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 1242 1243
+ 1245: 141(ptr) AccessChain 37(data) 1239 135
+ 1246: 31(i64vec4) Load 1245
+ 1247: 31(i64vec4) VectorShuffle 1246 1244 4 5 6 3
+ Store 1245 1247
+ 1248: 6(int) Load 8(invocation)
+ 1249: 141(ptr) AccessChain 37(data) 58 135
+ 1250: 31(i64vec4) Load 1249
+ 1251: 17(ivec4) Load 19(ballot)
+ 1252: 31(i64vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 1250 1251
+ 1253: 141(ptr) AccessChain 37(data) 1248 135
+ Store 1253 1252
+ 1254: 6(int) Load 8(invocation)
+ 1255: 136(ptr) AccessChain 37(data) 39 135 40
+ 1256: 30(int64_t) Load 1255
+ 1257: 17(ivec4) Load 19(ballot)
+ 1258: 30(int64_t) GroupNonUniformIMul 177 PartitionedReduceNV 1256 1257
+ 1259: 136(ptr) AccessChain 37(data) 1254 135 40
+ Store 1259 1258
+ 1260: 6(int) Load 8(invocation)
+ 1261: 141(ptr) AccessChain 37(data) 45 135
+ 1262: 31(i64vec4) Load 1261
+ 1263:140(i64vec2) VectorShuffle 1262 1262 0 1
+ 1264: 17(ivec4) Load 19(ballot)
+ 1265:140(i64vec2) GroupNonUniformIMul 177 PartitionedReduceNV 1263 1264
+ 1266: 141(ptr) AccessChain 37(data) 1260 135
+ 1267: 31(i64vec4) Load 1266
+ 1268: 31(i64vec4) VectorShuffle 1267 1265 4 5 2 3
+ Store 1266 1268
+ 1269: 6(int) Load 8(invocation)
+ 1270: 141(ptr) AccessChain 37(data) 52 135
+ 1271: 31(i64vec4) Load 1270
+ 1272:146(i64vec3) VectorShuffle 1271 1271 0 1 2
+ 1273: 17(ivec4) Load 19(ballot)
+ 1274:146(i64vec3) GroupNonUniformIMul 177 PartitionedReduceNV 1272 1273
+ 1275: 141(ptr) AccessChain 37(data) 1269 135
+ 1276: 31(i64vec4) Load 1275
+ 1277: 31(i64vec4) VectorShuffle 1276 1274 4 5 6 3
+ Store 1275 1277
+ 1278: 6(int) Load 8(invocation)
+ 1279: 141(ptr) AccessChain 37(data) 58 135
+ 1280: 31(i64vec4) Load 1279
+ 1281: 17(ivec4) Load 19(ballot)
+ 1282: 31(i64vec4) GroupNonUniformIMul 177 PartitionedReduceNV 1280 1281
+ 1283: 141(ptr) AccessChain 37(data) 1278 135
+ Store 1283 1282
+ 1284: 6(int) Load 8(invocation)
+ 1285: 136(ptr) AccessChain 37(data) 39 135 40
+ 1286: 30(int64_t) Load 1285
+ 1287: 17(ivec4) Load 19(ballot)
+ 1288: 30(int64_t) GroupNonUniformUMin 177 PartitionedReduceNV 1286 1287
+ 1289: 136(ptr) AccessChain 37(data) 1284 135 40
+ Store 1289 1288
+ 1290: 6(int) Load 8(invocation)
+ 1291: 141(ptr) AccessChain 37(data) 45 135
+ 1292: 31(i64vec4) Load 1291
+ 1293:140(i64vec2) VectorShuffle 1292 1292 0 1
+ 1294: 17(ivec4) Load 19(ballot)
+ 1295:140(i64vec2) GroupNonUniformUMin 177 PartitionedReduceNV 1293 1294
+ 1296: 141(ptr) AccessChain 37(data) 1290 135
+ 1297: 31(i64vec4) Load 1296
+ 1298: 31(i64vec4) VectorShuffle 1297 1295 4 5 2 3
+ Store 1296 1298
+ 1299: 6(int) Load 8(invocation)
+ 1300: 141(ptr) AccessChain 37(data) 52 135
+ 1301: 31(i64vec4) Load 1300
+ 1302:146(i64vec3) VectorShuffle 1301 1301 0 1 2
+ 1303: 17(ivec4) Load 19(ballot)
+ 1304:146(i64vec3) GroupNonUniformUMin 177 PartitionedReduceNV 1302 1303
+ 1305: 141(ptr) AccessChain 37(data) 1299 135
+ 1306: 31(i64vec4) Load 1305
+ 1307: 31(i64vec4) VectorShuffle 1306 1304 4 5 6 3
+ Store 1305 1307
+ 1308: 6(int) Load 8(invocation)
+ 1309: 141(ptr) AccessChain 37(data) 58 135
+ 1310: 31(i64vec4) Load 1309
+ 1311: 17(ivec4) Load 19(ballot)
+ 1312: 31(i64vec4) GroupNonUniformUMin 177 PartitionedReduceNV 1310 1311
+ 1313: 141(ptr) AccessChain 37(data) 1308 135
+ Store 1313 1312
+ 1314: 6(int) Load 8(invocation)
+ 1315: 136(ptr) AccessChain 37(data) 39 135 40
+ 1316: 30(int64_t) Load 1315
+ 1317: 17(ivec4) Load 19(ballot)
+ 1318: 30(int64_t) GroupNonUniformUMax 177 PartitionedReduceNV 1316 1317
+ 1319: 136(ptr) AccessChain 37(data) 1314 135 40
+ Store 1319 1318
+ 1320: 6(int) Load 8(invocation)
+ 1321: 141(ptr) AccessChain 37(data) 45 135
+ 1322: 31(i64vec4) Load 1321
+ 1323:140(i64vec2) VectorShuffle 1322 1322 0 1
+ 1324: 17(ivec4) Load 19(ballot)
+ 1325:140(i64vec2) GroupNonUniformUMax 177 PartitionedReduceNV 1323 1324
+ 1326: 141(ptr) AccessChain 37(data) 1320 135
+ 1327: 31(i64vec4) Load 1326
+ 1328: 31(i64vec4) VectorShuffle 1327 1325 4 5 2 3
+ Store 1326 1328
+ 1329: 6(int) Load 8(invocation)
+ 1330: 141(ptr) AccessChain 37(data) 52 135
+ 1331: 31(i64vec4) Load 1330
+ 1332:146(i64vec3) VectorShuffle 1331 1331 0 1 2
+ 1333: 17(ivec4) Load 19(ballot)
+ 1334:146(i64vec3) GroupNonUniformUMax 177 PartitionedReduceNV 1332 1333
+ 1335: 141(ptr) AccessChain 37(data) 1329 135
+ 1336: 31(i64vec4) Load 1335
+ 1337: 31(i64vec4) VectorShuffle 1336 1334 4 5 6 3
+ Store 1335 1337
+ 1338: 6(int) Load 8(invocation)
+ 1339: 141(ptr) AccessChain 37(data) 58 135
+ 1340: 31(i64vec4) Load 1339
+ 1341: 17(ivec4) Load 19(ballot)
+ 1342: 31(i64vec4) GroupNonUniformUMax 177 PartitionedReduceNV 1340 1341
+ 1343: 141(ptr) AccessChain 37(data) 1338 135
+ Store 1343 1342
+ 1344: 6(int) Load 8(invocation)
+ 1345: 136(ptr) AccessChain 37(data) 39 135 40
+ 1346: 30(int64_t) Load 1345
+ 1347: 17(ivec4) Load 19(ballot)
+ 1348: 30(int64_t) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1346 1347
+ 1349: 136(ptr) AccessChain 37(data) 1344 135 40
+ Store 1349 1348
+ 1350: 6(int) Load 8(invocation)
+ 1351: 141(ptr) AccessChain 37(data) 45 135
+ 1352: 31(i64vec4) Load 1351
+ 1353:140(i64vec2) VectorShuffle 1352 1352 0 1
+ 1354: 17(ivec4) Load 19(ballot)
+ 1355:140(i64vec2) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1353 1354
+ 1356: 141(ptr) AccessChain 37(data) 1350 135
+ 1357: 31(i64vec4) Load 1356
+ 1358: 31(i64vec4) VectorShuffle 1357 1355 4 5 2 3
+ Store 1356 1358
+ 1359: 6(int) Load 8(invocation)
+ 1360: 141(ptr) AccessChain 37(data) 52 135
+ 1361: 31(i64vec4) Load 1360
+ 1362:146(i64vec3) VectorShuffle 1361 1361 0 1 2
+ 1363: 17(ivec4) Load 19(ballot)
+ 1364:146(i64vec3) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1362 1363
+ 1365: 141(ptr) AccessChain 37(data) 1359 135
+ 1366: 31(i64vec4) Load 1365
+ 1367: 31(i64vec4) VectorShuffle 1366 1364 4 5 6 3
+ Store 1365 1367
+ 1368: 6(int) Load 8(invocation)
+ 1369: 141(ptr) AccessChain 37(data) 58 135
+ 1370: 31(i64vec4) Load 1369
+ 1371: 17(ivec4) Load 19(ballot)
+ 1372: 31(i64vec4) GroupNonUniformBitwiseAnd 177 PartitionedReduceNV 1370 1371
+ 1373: 141(ptr) AccessChain 37(data) 1368 135
+ Store 1373 1372
+ 1374: 6(int) Load 8(invocation)
+ 1375: 136(ptr) AccessChain 37(data) 39 135 40
+ 1376: 30(int64_t) Load 1375
+ 1377: 17(ivec4) Load 19(ballot)
+ 1378: 30(int64_t) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1376 1377
+ 1379: 136(ptr) AccessChain 37(data) 1374 135 40
+ Store 1379 1378
+ 1380: 6(int) Load 8(invocation)
+ 1381: 141(ptr) AccessChain 37(data) 45 135
+ 1382: 31(i64vec4) Load 1381
+ 1383:140(i64vec2) VectorShuffle 1382 1382 0 1
+ 1384: 17(ivec4) Load 19(ballot)
+ 1385:140(i64vec2) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1383 1384
+ 1386: 141(ptr) AccessChain 37(data) 1380 135
+ 1387: 31(i64vec4) Load 1386
+ 1388: 31(i64vec4) VectorShuffle 1387 1385 4 5 2 3
+ Store 1386 1388
+ 1389: 6(int) Load 8(invocation)
+ 1390: 141(ptr) AccessChain 37(data) 52 135
+ 1391: 31(i64vec4) Load 1390
+ 1392:146(i64vec3) VectorShuffle 1391 1391 0 1 2
+ 1393: 17(ivec4) Load 19(ballot)
+ 1394:146(i64vec3) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1392 1393
+ 1395: 141(ptr) AccessChain 37(data) 1389 135
+ 1396: 31(i64vec4) Load 1395
+ 1397: 31(i64vec4) VectorShuffle 1396 1394 4 5 6 3
+ Store 1395 1397
+ 1398: 6(int) Load 8(invocation)
+ 1399: 141(ptr) AccessChain 37(data) 58 135
+ 1400: 31(i64vec4) Load 1399
+ 1401: 17(ivec4) Load 19(ballot)
+ 1402: 31(i64vec4) GroupNonUniformBitwiseOr 177 PartitionedReduceNV 1400 1401
+ 1403: 141(ptr) AccessChain 37(data) 1398 135
+ Store 1403 1402
+ 1404: 6(int) Load 8(invocation)
+ 1405: 136(ptr) AccessChain 37(data) 39 135 40
+ 1406: 30(int64_t) Load 1405
+ 1407: 17(ivec4) Load 19(ballot)
+ 1408: 30(int64_t) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1406 1407
+ 1409: 136(ptr) AccessChain 37(data) 1404 135 40
+ Store 1409 1408
+ 1410: 6(int) Load 8(invocation)
+ 1411: 141(ptr) AccessChain 37(data) 45 135
+ 1412: 31(i64vec4) Load 1411
+ 1413:140(i64vec2) VectorShuffle 1412 1412 0 1
+ 1414: 17(ivec4) Load 19(ballot)
+ 1415:140(i64vec2) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1413 1414
+ 1416: 141(ptr) AccessChain 37(data) 1410 135
+ 1417: 31(i64vec4) Load 1416
+ 1418: 31(i64vec4) VectorShuffle 1417 1415 4 5 2 3
+ Store 1416 1418
+ 1419: 6(int) Load 8(invocation)
+ 1420: 141(ptr) AccessChain 37(data) 52 135
+ 1421: 31(i64vec4) Load 1420
+ 1422:146(i64vec3) VectorShuffle 1421 1421 0 1 2
+ 1423: 17(ivec4) Load 19(ballot)
+ 1424:146(i64vec3) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1422 1423
+ 1425: 141(ptr) AccessChain 37(data) 1419 135
+ 1426: 31(i64vec4) Load 1425
+ 1427: 31(i64vec4) VectorShuffle 1426 1424 4 5 6 3
+ Store 1425 1427
+ 1428: 6(int) Load 8(invocation)
+ 1429: 141(ptr) AccessChain 37(data) 58 135
+ 1430: 31(i64vec4) Load 1429
+ 1431: 17(ivec4) Load 19(ballot)
+ 1432: 31(i64vec4) GroupNonUniformBitwiseXor 177 PartitionedReduceNV 1430 1431
+ 1433: 141(ptr) AccessChain 37(data) 1428 135
+ Store 1433 1432
+ 1434: 6(int) Load 8(invocation)
+ 1435: 155(ptr) AccessChain 37(data) 39 154 40
+ 1436:32(float16_t) Load 1435
+ 1437: 17(ivec4) Load 19(ballot)
+ 1438:32(float16_t) GroupNonUniformIAdd 177 PartitionedReduceNV 1436 1437
+ 1439: 155(ptr) AccessChain 37(data) 1434 154 40
+ Store 1439 1438
+ 1440: 6(int) Load 8(invocation)
+ 1441: 160(ptr) AccessChain 37(data) 45 154
+ 1442: 33(f16vec4) Load 1441
+ 1443:159(f16vec2) VectorShuffle 1442 1442 0 1
+ 1444: 17(ivec4) Load 19(ballot)
+ 1445:159(f16vec2) GroupNonUniformIAdd 177 PartitionedReduceNV 1443 1444
+ 1446: 160(ptr) AccessChain 37(data) 1440 154
+ 1447: 33(f16vec4) Load 1446
+ 1448: 33(f16vec4) VectorShuffle 1447 1445 4 5 2 3
+ Store 1446 1448
+ 1449: 6(int) Load 8(invocation)
+ 1450: 160(ptr) AccessChain 37(data) 52 154
+ 1451: 33(f16vec4) Load 1450
+ 1452:165(f16vec3) VectorShuffle 1451 1451 0 1 2
+ 1453: 17(ivec4) Load 19(ballot)
+ 1454:165(f16vec3) GroupNonUniformIAdd 177 PartitionedReduceNV 1452 1453
+ 1455: 160(ptr) AccessChain 37(data) 1449 154
+ 1456: 33(f16vec4) Load 1455
+ 1457: 33(f16vec4) VectorShuffle 1456 1454 4 5 6 3
+ Store 1455 1457
+ 1458: 6(int) Load 8(invocation)
+ 1459: 160(ptr) AccessChain 37(data) 58 154
+ 1460: 33(f16vec4) Load 1459
+ 1461: 17(ivec4) Load 19(ballot)
+ 1462: 33(f16vec4) GroupNonUniformIAdd 177 PartitionedReduceNV 1460 1461
+ 1463: 160(ptr) AccessChain 37(data) 1458 154
+ Store 1463 1462
+ 1464: 6(int) Load 8(invocation)
+ 1465: 155(ptr) AccessChain 37(data) 39 154 40
+ 1466:32(float16_t) Load 1465
+ 1467: 17(ivec4) Load 19(ballot)
+ 1468:32(float16_t) GroupNonUniformIMul 177 PartitionedReduceNV 1466 1467
+ 1469: 155(ptr) AccessChain 37(data) 1464 154 40
+ Store 1469 1468
+ 1470: 6(int) Load 8(invocation)
+ 1471: 160(ptr) AccessChain 37(data) 45 154
+ 1472: 33(f16vec4) Load 1471
+ 1473:159(f16vec2) VectorShuffle 1472 1472 0 1
+ 1474: 17(ivec4) Load 19(ballot)
+ 1475:159(f16vec2) GroupNonUniformIMul 177 PartitionedReduceNV 1473 1474
+ 1476: 160(ptr) AccessChain 37(data) 1470 154
+ 1477: 33(f16vec4) Load 1476
+ 1478: 33(f16vec4) VectorShuffle 1477 1475 4 5 2 3
+ Store 1476 1478
+ 1479: 6(int) Load 8(invocation)
+ 1480: 160(ptr) AccessChain 37(data) 52 154
+ 1481: 33(f16vec4) Load 1480
+ 1482:165(f16vec3) VectorShuffle 1481 1481 0 1 2
+ 1483: 17(ivec4) Load 19(ballot)
+ 1484:165(f16vec3) GroupNonUniformIMul 177 PartitionedReduceNV 1482 1483
+ 1485: 160(ptr) AccessChain 37(data) 1479 154
+ 1486: 33(f16vec4) Load 1485
+ 1487: 33(f16vec4) VectorShuffle 1486 1484 4 5 6 3
+ Store 1485 1487
+ 1488: 6(int) Load 8(invocation)
+ 1489: 160(ptr) AccessChain 37(data) 58 154
+ 1490: 33(f16vec4) Load 1489
+ 1491: 17(ivec4) Load 19(ballot)
+ 1492: 33(f16vec4) GroupNonUniformIMul 177 PartitionedReduceNV 1490 1491
+ 1493: 160(ptr) AccessChain 37(data) 1488 154
+ Store 1493 1492
+ 1494: 6(int) Load 8(invocation)
+ 1495: 155(ptr) AccessChain 37(data) 39 154 40
+ 1496:32(float16_t) Load 1495
+ 1497: 17(ivec4) Load 19(ballot)
+ 1498:32(float16_t) GroupNonUniformSMin 177 PartitionedReduceNV 1496 1497
+ 1499: 155(ptr) AccessChain 37(data) 1494 154 40
+ Store 1499 1498
+ 1500: 6(int) Load 8(invocation)
+ 1501: 160(ptr) AccessChain 37(data) 45 154
+ 1502: 33(f16vec4) Load 1501
+ 1503:159(f16vec2) VectorShuffle 1502 1502 0 1
+ 1504: 17(ivec4) Load 19(ballot)
+ 1505:159(f16vec2) GroupNonUniformSMin 177 PartitionedReduceNV 1503 1504
+ 1506: 160(ptr) AccessChain 37(data) 1500 154
+ 1507: 33(f16vec4) Load 1506
+ 1508: 33(f16vec4) VectorShuffle 1507 1505 4 5 2 3
+ Store 1506 1508
+ 1509: 6(int) Load 8(invocation)
+ 1510: 160(ptr) AccessChain 37(data) 52 154
+ 1511: 33(f16vec4) Load 1510
+ 1512:165(f16vec3) VectorShuffle 1511 1511 0 1 2
+ 1513: 17(ivec4) Load 19(ballot)
+ 1514:165(f16vec3) GroupNonUniformSMin 177 PartitionedReduceNV 1512 1513
+ 1515: 160(ptr) AccessChain 37(data) 1509 154
+ 1516: 33(f16vec4) Load 1515
+ 1517: 33(f16vec4) VectorShuffle 1516 1514 4 5 6 3
+ Store 1515 1517
+ 1518: 6(int) Load 8(invocation)
+ 1519: 160(ptr) AccessChain 37(data) 58 154
+ 1520: 33(f16vec4) Load 1519
+ 1521: 17(ivec4) Load 19(ballot)
+ 1522: 33(f16vec4) GroupNonUniformSMin 177 PartitionedReduceNV 1520 1521
+ 1523: 160(ptr) AccessChain 37(data) 1518 154
+ Store 1523 1522
+ 1524: 6(int) Load 8(invocation)
+ 1525: 155(ptr) AccessChain 37(data) 39 154 40
+ 1526:32(float16_t) Load 1525
+ 1527: 17(ivec4) Load 19(ballot)
+ 1528:32(float16_t) GroupNonUniformSMax 177 PartitionedReduceNV 1526 1527
+ 1529: 155(ptr) AccessChain 37(data) 1524 154 40
+ Store 1529 1528
+ 1530: 6(int) Load 8(invocation)
+ 1531: 160(ptr) AccessChain 37(data) 45 154
+ 1532: 33(f16vec4) Load 1531
+ 1533:159(f16vec2) VectorShuffle 1532 1532 0 1
+ 1534: 17(ivec4) Load 19(ballot)
+ 1535:159(f16vec2) GroupNonUniformSMax 177 PartitionedReduceNV 1533 1534
+ 1536: 160(ptr) AccessChain 37(data) 1530 154
+ 1537: 33(f16vec4) Load 1536
+ 1538: 33(f16vec4) VectorShuffle 1537 1535 4 5 2 3
+ Store 1536 1538
+ 1539: 6(int) Load 8(invocation)
+ 1540: 160(ptr) AccessChain 37(data) 52 154
+ 1541: 33(f16vec4) Load 1540
+ 1542:165(f16vec3) VectorShuffle 1541 1541 0 1 2
+ 1543: 17(ivec4) Load 19(ballot)
+ 1544:165(f16vec3) GroupNonUniformSMax 177 PartitionedReduceNV 1542 1543
+ 1545: 160(ptr) AccessChain 37(data) 1539 154
+ 1546: 33(f16vec4) Load 1545
+ 1547: 33(f16vec4) VectorShuffle 1546 1544 4 5 6 3
+ Store 1545 1547
+ 1548: 6(int) Load 8(invocation)
+ 1549: 160(ptr) AccessChain 37(data) 58 154
+ 1550: 33(f16vec4) Load 1549
+ 1551: 17(ivec4) Load 19(ballot)
+ 1552: 33(f16vec4) GroupNonUniformSMax 177 PartitionedReduceNV 1550 1551
+ 1553: 160(ptr) AccessChain 37(data) 1548 154
+ Store 1553 1552
+ Return
+ FunctionEnd
diff --git a/Test/baseResults/spv.subgroupExtendedTypesPartitionedNeg.comp.out b/Test/baseResults/spv.subgroupExtendedTypesPartitionedNeg.comp.out
new file mode 100644
index 0000000..c029617
--- /dev/null
+++ b/Test/baseResults/spv.subgroupExtendedTypesPartitionedNeg.comp.out
@@ -0,0 +1,217 @@
+spv.subgroupExtendedTypesPartitionedNeg.comp
+ERROR: 0:27: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:28: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:29: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:30: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:32: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:33: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:34: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:35: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:37: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:38: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:39: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:40: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:42: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:43: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:44: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:45: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:47: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:48: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:49: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:50: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:52: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:53: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:54: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:55: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:57: ' temp float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:58: ' temp 2-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:59: ' temp 3-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:60: 'layout( column_major std430) buffer 4-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:62: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:63: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:64: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:65: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:67: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:68: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:69: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:70: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:72: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:73: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:74: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:75: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:77: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:78: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:79: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:80: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:82: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:83: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:84: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:85: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:87: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:88: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:89: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:90: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:92: ' temp int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:93: ' temp 2-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:94: ' temp 3-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:95: 'layout( column_major std430) buffer 4-component vector of int8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:97: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:98: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:99: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:100: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:102: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:103: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:104: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:105: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:107: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:108: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:109: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:110: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:112: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:113: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:114: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:115: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:117: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:118: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:119: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:120: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:122: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:123: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:124: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:125: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:127: ' temp uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:128: ' temp 2-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:129: ' temp 3-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:130: 'layout( column_major std430) buffer 4-component vector of uint8_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int8
+ERROR: 0:132: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:133: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:134: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:135: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:137: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:138: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:139: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:140: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:142: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:143: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:144: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:145: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:147: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:148: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:149: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:150: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:152: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:153: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:154: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:155: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:157: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:158: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:159: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:160: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:162: ' temp int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:163: ' temp 2-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:164: ' temp 3-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:165: 'layout( column_major std430) buffer 4-component vector of int16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:167: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:168: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:169: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:170: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:172: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:173: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:174: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:175: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:177: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:178: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:179: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:180: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:182: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:183: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:184: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:185: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:187: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:188: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:189: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:190: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:192: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:193: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:194: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:195: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:197: ' temp uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:198: ' temp 2-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:199: ' temp 3-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:200: 'layout( column_major std430) buffer 4-component vector of uint16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int16
+ERROR: 0:202: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:203: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:204: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:205: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:207: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:208: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:209: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:210: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:212: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:213: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:214: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:215: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:217: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:218: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:219: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:220: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:222: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:223: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:224: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:225: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:227: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:228: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:229: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:230: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:232: ' temp int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:233: ' temp 2-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:234: ' temp 3-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:235: 'layout( column_major std430) buffer 4-component vector of int64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:237: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:238: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:239: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:240: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:242: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:243: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:244: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:245: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:247: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:248: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:249: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:250: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:252: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:253: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:254: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:255: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:257: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:258: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:259: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:260: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:262: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:263: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:264: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:265: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:267: ' temp uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:268: ' temp 2-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:269: ' temp 3-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:270: 'layout( column_major std430) buffer 4-component vector of uint64_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_int64
+ERROR: 0:272: ' temp highp float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:273: ' temp highp 2-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:274: ' temp highp 3-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:275: 'layout( column_major std430) buffer highp 4-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:277: ' temp highp float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:278: ' temp highp 2-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:279: ' temp highp 3-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:280: 'layout( column_major std430) buffer highp 4-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:282: ' temp highp float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:283: ' temp highp 2-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:284: ' temp highp 3-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:285: 'layout( column_major std430) buffer highp 4-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:287: ' temp highp float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:288: ' temp highp 2-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:289: ' temp highp 3-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 0:290: 'layout( column_major std430) buffer highp 4-component vector of float16_t' : required extension not requested: GL_EXT_shader_subgroup_extended_types_float16
+ERROR: 212 compilation errors. No code generated.
+
+
+SPIR-V is not generated for failed compile or link
diff --git a/Test/spv.subgroupExtendedTypesPartitioned.comp b/Test/spv.subgroupExtendedTypesPartitioned.comp
new file mode 100644
index 0000000..382a5b3
--- /dev/null
+++ b/Test/spv.subgroupExtendedTypesPartitioned.comp
@@ -0,0 +1,291 @@
+#version 450
+
+#extension GL_NV_shader_subgroup_partitioned: enable
+#extension GL_EXT_shader_subgroup_extended_types_int8: enable
+#extension GL_EXT_shader_subgroup_extended_types_int16: enable
+#extension GL_EXT_shader_subgroup_extended_types_int64: enable
+#extension GL_EXT_shader_subgroup_extended_types_float16: enable
+
+layout (local_size_x = 8) in;
+
+layout(binding = 0) buffer Buffers
+{
+ i8vec4 i8;
+ u8vec4 u8;
+ i16vec4 i16;
+ u16vec4 u16;
+ i64vec4 i64;
+ u64vec4 u64;
+ f16vec4 f16;
+} data[4];
+
+void main()
+{
+ uint invocation = (gl_SubgroupInvocationID + gl_SubgroupSize) % 4;
+
+ uvec4 ballot;
+ ballot = subgroupPartitionNV(data[0].i8.x);
+ ballot = subgroupPartitionNV(data[1].i8.xy);
+ ballot = subgroupPartitionNV(data[2].i8.xyz);
+ ballot = subgroupPartitionNV(data[3].i8);
+
+ ballot = subgroupPartitionNV(data[0].u8.x);
+ ballot = subgroupPartitionNV(data[1].u8.xy);
+ ballot = subgroupPartitionNV(data[2].u8.xyz);
+ ballot = subgroupPartitionNV(data[3].u8);
+
+ ballot = subgroupPartitionNV(data[0].i16.x);
+ ballot = subgroupPartitionNV(data[1].i16.xy);
+ ballot = subgroupPartitionNV(data[2].i16.xyz);
+ ballot = subgroupPartitionNV(data[3].i16);
+
+ ballot = subgroupPartitionNV(data[0].u16.x);
+ ballot = subgroupPartitionNV(data[1].u16.xy);
+ ballot = subgroupPartitionNV(data[2].u16.xyz);
+ ballot = subgroupPartitionNV(data[3].u16);
+
+ ballot = subgroupPartitionNV(data[0].i64.x);
+ ballot = subgroupPartitionNV(data[1].i64.xy);
+ ballot = subgroupPartitionNV(data[2].i64.xyz);
+ ballot = subgroupPartitionNV(data[3].i64);
+
+ ballot = subgroupPartitionNV(data[0].u64.x);
+ ballot = subgroupPartitionNV(data[1].u64.xy);
+ ballot = subgroupPartitionNV(data[2].u64.xyz);
+ ballot = subgroupPartitionNV(data[3].u64);
+
+ ballot = subgroupPartitionNV(data[0].f16.x);
+ ballot = subgroupPartitionNV(data[1].f16.xy);
+ ballot = subgroupPartitionNV(data[2].f16.xyz);
+ ballot = subgroupPartitionNV(data[3].f16);
+
+ data[invocation].i8.x = subgroupPartitionedAddNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedAddNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedAddNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedAddNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMulNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMulNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMulNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMulNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMinNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMinNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMinNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMinNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMaxNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMaxNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMaxNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMaxNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedAndNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedAndNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedAndNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedAndNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedOrNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedOrNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedOrNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedOrNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedXorNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedXorNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedXorNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedXorNV(data[3].i8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedAddNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedAddNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedAddNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedAddNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMulNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMulNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMulNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMulNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMinNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMinNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMinNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMinNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMaxNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMaxNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMaxNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMaxNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedAndNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedAndNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedAndNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedAndNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedOrNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedOrNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedOrNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedOrNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedXorNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedXorNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedXorNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedXorNV(data[3].u8, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedAddNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedAddNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedAddNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedAddNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMulNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMulNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMulNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMulNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMinNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMinNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMinNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMinNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMaxNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMaxNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMaxNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMaxNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedAndNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedAndNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedAndNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedAndNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedOrNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedOrNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedOrNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedOrNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedXorNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedXorNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedXorNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedXorNV(data[3].i16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedAddNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedAddNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedAddNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedAddNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMulNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMulNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMulNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMulNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMinNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMinNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMinNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMinNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMaxNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMaxNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMaxNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMaxNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedAndNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedAndNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedAndNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedAndNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedOrNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedOrNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedOrNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedOrNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedXorNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedXorNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedXorNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedXorNV(data[3].u16, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedAddNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedAddNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedAddNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedAddNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMulNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMulNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMulNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMulNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMinNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMinNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMinNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMinNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMaxNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMaxNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMaxNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMaxNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedAndNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedAndNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedAndNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedAndNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedOrNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedOrNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedOrNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedOrNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedXorNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedXorNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedXorNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedXorNV(data[3].i64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedAddNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedAddNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedAddNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedAddNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMulNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMulNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMulNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMulNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMinNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMinNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMinNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMinNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMaxNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMaxNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMaxNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMaxNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedAndNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedAndNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedAndNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedAndNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedOrNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedOrNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedOrNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedOrNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedXorNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedXorNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedXorNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedXorNV(data[3].u64, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedAddNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedAddNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedAddNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedAddNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMulNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMulNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMulNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMulNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMinNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMinNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMinNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMinNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMaxNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMaxNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMaxNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMaxNV(data[3].f16, ballot);
+}
diff --git a/Test/spv.subgroupExtendedTypesPartitionedNeg.comp b/Test/spv.subgroupExtendedTypesPartitionedNeg.comp
new file mode 100644
index 0000000..ad94e35
--- /dev/null
+++ b/Test/spv.subgroupExtendedTypesPartitionedNeg.comp
@@ -0,0 +1,291 @@
+#version 450
+
+#extension GL_NV_shader_subgroup_partitioned: enable
+#extension GL_EXT_shader_explicit_arithmetic_types_int8: enable
+#extension GL_EXT_shader_explicit_arithmetic_types_int16: enable
+#extension GL_EXT_shader_explicit_arithmetic_types_int64: enable
+#extension GL_EXT_shader_explicit_arithmetic_types_float16: enable
+
+layout (local_size_x = 8) in;
+
+layout(binding = 0) buffer Buffers
+{
+ i8vec4 i8;
+ u8vec4 u8;
+ i16vec4 i16;
+ u16vec4 u16;
+ i64vec4 i64;
+ u64vec4 u64;
+ f16vec4 f16;
+} data[4];
+
+void main()
+{
+ uint invocation = (gl_SubgroupInvocationID + gl_SubgroupSize) % 4;
+
+ uvec4 ballot;
+ ballot = subgroupPartitionNV(data[0].i8.x);
+ ballot = subgroupPartitionNV(data[1].i8.xy);
+ ballot = subgroupPartitionNV(data[2].i8.xyz);
+ ballot = subgroupPartitionNV(data[3].i8);
+
+ ballot = subgroupPartitionNV(data[0].u8.x);
+ ballot = subgroupPartitionNV(data[1].u8.xy);
+ ballot = subgroupPartitionNV(data[2].u8.xyz);
+ ballot = subgroupPartitionNV(data[3].u8);
+
+ ballot = subgroupPartitionNV(data[0].i16.x);
+ ballot = subgroupPartitionNV(data[1].i16.xy);
+ ballot = subgroupPartitionNV(data[2].i16.xyz);
+ ballot = subgroupPartitionNV(data[3].i16);
+
+ ballot = subgroupPartitionNV(data[0].u16.x);
+ ballot = subgroupPartitionNV(data[1].u16.xy);
+ ballot = subgroupPartitionNV(data[2].u16.xyz);
+ ballot = subgroupPartitionNV(data[3].u16);
+
+ ballot = subgroupPartitionNV(data[0].i64.x);
+ ballot = subgroupPartitionNV(data[1].i64.xy);
+ ballot = subgroupPartitionNV(data[2].i64.xyz);
+ ballot = subgroupPartitionNV(data[3].i64);
+
+ ballot = subgroupPartitionNV(data[0].u64.x);
+ ballot = subgroupPartitionNV(data[1].u64.xy);
+ ballot = subgroupPartitionNV(data[2].u64.xyz);
+ ballot = subgroupPartitionNV(data[3].u64);
+
+ ballot = subgroupPartitionNV(data[0].f16.x);
+ ballot = subgroupPartitionNV(data[1].f16.xy);
+ ballot = subgroupPartitionNV(data[2].f16.xyz);
+ ballot = subgroupPartitionNV(data[3].f16);
+
+ data[invocation].i8.x = subgroupPartitionedAddNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedAddNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedAddNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedAddNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMulNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMulNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMulNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMulNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMinNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMinNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMinNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMinNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedMaxNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedMaxNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedMaxNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedMaxNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedAndNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedAndNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedAndNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedAndNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedOrNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedOrNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedOrNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedOrNV(data[3].i8, ballot);
+
+ data[invocation].i8.x = subgroupPartitionedXorNV(data[0].i8.x, ballot);
+ data[invocation].i8.xy = subgroupPartitionedXorNV(data[1].i8.xy, ballot);
+ data[invocation].i8.xyz = subgroupPartitionedXorNV(data[2].i8.xyz, ballot);
+ data[invocation].i8 = subgroupPartitionedXorNV(data[3].i8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedAddNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedAddNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedAddNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedAddNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMulNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMulNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMulNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMulNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMinNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMinNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMinNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMinNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedMaxNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedMaxNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedMaxNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedMaxNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedAndNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedAndNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedAndNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedAndNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedOrNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedOrNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedOrNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedOrNV(data[3].u8, ballot);
+
+ data[invocation].u8.x = subgroupPartitionedXorNV(data[0].u8.x, ballot);
+ data[invocation].u8.xy = subgroupPartitionedXorNV(data[1].u8.xy, ballot);
+ data[invocation].u8.xyz = subgroupPartitionedXorNV(data[2].u8.xyz, ballot);
+ data[invocation].u8 = subgroupPartitionedXorNV(data[3].u8, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedAddNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedAddNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedAddNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedAddNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMulNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMulNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMulNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMulNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMinNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMinNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMinNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMinNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedMaxNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedMaxNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedMaxNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedMaxNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedAndNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedAndNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedAndNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedAndNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedOrNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedOrNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedOrNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedOrNV(data[3].i16, ballot);
+
+ data[invocation].i16.x = subgroupPartitionedXorNV(data[0].i16.x, ballot);
+ data[invocation].i16.xy = subgroupPartitionedXorNV(data[1].i16.xy, ballot);
+ data[invocation].i16.xyz = subgroupPartitionedXorNV(data[2].i16.xyz, ballot);
+ data[invocation].i16 = subgroupPartitionedXorNV(data[3].i16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedAddNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedAddNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedAddNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedAddNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMulNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMulNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMulNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMulNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMinNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMinNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMinNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMinNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedMaxNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedMaxNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedMaxNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedMaxNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedAndNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedAndNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedAndNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedAndNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedOrNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedOrNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedOrNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedOrNV(data[3].u16, ballot);
+
+ data[invocation].u16.x = subgroupPartitionedXorNV(data[0].u16.x, ballot);
+ data[invocation].u16.xy = subgroupPartitionedXorNV(data[1].u16.xy, ballot);
+ data[invocation].u16.xyz = subgroupPartitionedXorNV(data[2].u16.xyz, ballot);
+ data[invocation].u16 = subgroupPartitionedXorNV(data[3].u16, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedAddNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedAddNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedAddNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedAddNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMulNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMulNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMulNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMulNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMinNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMinNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMinNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMinNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedMaxNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedMaxNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedMaxNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedMaxNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedAndNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedAndNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedAndNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedAndNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedOrNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedOrNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedOrNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedOrNV(data[3].i64, ballot);
+
+ data[invocation].i64.x = subgroupPartitionedXorNV(data[0].i64.x, ballot);
+ data[invocation].i64.xy = subgroupPartitionedXorNV(data[1].i64.xy, ballot);
+ data[invocation].i64.xyz = subgroupPartitionedXorNV(data[2].i64.xyz, ballot);
+ data[invocation].i64 = subgroupPartitionedXorNV(data[3].i64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedAddNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedAddNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedAddNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedAddNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMulNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMulNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMulNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMulNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMinNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMinNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMinNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMinNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedMaxNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedMaxNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedMaxNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedMaxNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedAndNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedAndNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedAndNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedAndNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedOrNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedOrNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedOrNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedOrNV(data[3].u64, ballot);
+
+ data[invocation].u64.x = subgroupPartitionedXorNV(data[0].u64.x, ballot);
+ data[invocation].u64.xy = subgroupPartitionedXorNV(data[1].u64.xy, ballot);
+ data[invocation].u64.xyz = subgroupPartitionedXorNV(data[2].u64.xyz, ballot);
+ data[invocation].u64 = subgroupPartitionedXorNV(data[3].u64, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedAddNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedAddNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedAddNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedAddNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMulNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMulNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMulNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMulNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMinNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMinNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMinNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMinNV(data[3].f16, ballot);
+
+ data[invocation].f16.x = subgroupPartitionedMaxNV(data[0].f16.x, ballot);
+ data[invocation].f16.xy = subgroupPartitionedMaxNV(data[1].f16.xy, ballot);
+ data[invocation].f16.xyz = subgroupPartitionedMaxNV(data[2].f16.xyz, ballot);
+ data[invocation].f16 = subgroupPartitionedMaxNV(data[3].f16, ballot);
+}
diff --git a/glslang/MachineIndependent/Initialize.cpp b/glslang/MachineIndependent/Initialize.cpp
index 6fe57ca..a49a560 100644
--- a/glslang/MachineIndependent/Initialize.cpp
+++ b/glslang/MachineIndependent/Initialize.cpp
@@ -1884,58 +1884,6 @@
"bool subgroupAll(bool);\n"
"bool subgroupAny(bool);\n"
-
- "bool subgroupAllEqual(float);\n"
- "bool subgroupAllEqual(vec2);\n"
- "bool subgroupAllEqual(vec3);\n"
- "bool subgroupAllEqual(vec4);\n"
- "bool subgroupAllEqual(int);\n"
- "bool subgroupAllEqual(ivec2);\n"
- "bool subgroupAllEqual(ivec3);\n"
- "bool subgroupAllEqual(ivec4);\n"
- "bool subgroupAllEqual(uint);\n"
- "bool subgroupAllEqual(uvec2);\n"
- "bool subgroupAllEqual(uvec3);\n"
- "bool subgroupAllEqual(uvec4);\n"
- "bool subgroupAllEqual(bool);\n"
- "bool subgroupAllEqual(bvec2);\n"
- "bool subgroupAllEqual(bvec3);\n"
- "bool subgroupAllEqual(bvec4);\n"
-
- "float subgroupBroadcast(float, uint);\n"
- "vec2 subgroupBroadcast(vec2, uint);\n"
- "vec3 subgroupBroadcast(vec3, uint);\n"
- "vec4 subgroupBroadcast(vec4, uint);\n"
- "int subgroupBroadcast(int, uint);\n"
- "ivec2 subgroupBroadcast(ivec2, uint);\n"
- "ivec3 subgroupBroadcast(ivec3, uint);\n"
- "ivec4 subgroupBroadcast(ivec4, uint);\n"
- "uint subgroupBroadcast(uint, uint);\n"
- "uvec2 subgroupBroadcast(uvec2, uint);\n"
- "uvec3 subgroupBroadcast(uvec3, uint);\n"
- "uvec4 subgroupBroadcast(uvec4, uint);\n"
- "bool subgroupBroadcast(bool, uint);\n"
- "bvec2 subgroupBroadcast(bvec2, uint);\n"
- "bvec3 subgroupBroadcast(bvec3, uint);\n"
- "bvec4 subgroupBroadcast(bvec4, uint);\n"
-
- "float subgroupBroadcastFirst(float);\n"
- "vec2 subgroupBroadcastFirst(vec2);\n"
- "vec3 subgroupBroadcastFirst(vec3);\n"
- "vec4 subgroupBroadcastFirst(vec4);\n"
- "int subgroupBroadcastFirst(int);\n"
- "ivec2 subgroupBroadcastFirst(ivec2);\n"
- "ivec3 subgroupBroadcastFirst(ivec3);\n"
- "ivec4 subgroupBroadcastFirst(ivec4);\n"
- "uint subgroupBroadcastFirst(uint);\n"
- "uvec2 subgroupBroadcastFirst(uvec2);\n"
- "uvec3 subgroupBroadcastFirst(uvec3);\n"
- "uvec4 subgroupBroadcastFirst(uvec4);\n"
- "bool subgroupBroadcastFirst(bool);\n"
- "bvec2 subgroupBroadcastFirst(bvec2);\n"
- "bvec3 subgroupBroadcastFirst(bvec3);\n"
- "bvec4 subgroupBroadcastFirst(bvec4);\n"
-
"uvec4 subgroupBallot(bool);\n"
"bool subgroupInverseBallot(uvec4);\n"
"bool subgroupBallotBitExtract(uvec4, uint);\n"
@@ -1945,2196 +1893,132 @@
"uint subgroupBallotFindLSB(uvec4);\n"
"uint subgroupBallotFindMSB(uvec4);\n"
- "float subgroupShuffle(float, uint);\n"
- "vec2 subgroupShuffle(vec2, uint);\n"
- "vec3 subgroupShuffle(vec3, uint);\n"
- "vec4 subgroupShuffle(vec4, uint);\n"
- "int subgroupShuffle(int, uint);\n"
- "ivec2 subgroupShuffle(ivec2, uint);\n"
- "ivec3 subgroupShuffle(ivec3, uint);\n"
- "ivec4 subgroupShuffle(ivec4, uint);\n"
- "uint subgroupShuffle(uint, uint);\n"
- "uvec2 subgroupShuffle(uvec2, uint);\n"
- "uvec3 subgroupShuffle(uvec3, uint);\n"
- "uvec4 subgroupShuffle(uvec4, uint);\n"
- "bool subgroupShuffle(bool, uint);\n"
- "bvec2 subgroupShuffle(bvec2, uint);\n"
- "bvec3 subgroupShuffle(bvec3, uint);\n"
- "bvec4 subgroupShuffle(bvec4, uint);\n"
+ );
- "float subgroupShuffleXor(float, uint);\n"
- "vec2 subgroupShuffleXor(vec2, uint);\n"
- "vec3 subgroupShuffleXor(vec3, uint);\n"
- "vec4 subgroupShuffleXor(vec4, uint);\n"
- "int subgroupShuffleXor(int, uint);\n"
- "ivec2 subgroupShuffleXor(ivec2, uint);\n"
- "ivec3 subgroupShuffleXor(ivec3, uint);\n"
- "ivec4 subgroupShuffleXor(ivec4, uint);\n"
- "uint subgroupShuffleXor(uint, uint);\n"
- "uvec2 subgroupShuffleXor(uvec2, uint);\n"
- "uvec3 subgroupShuffleXor(uvec3, uint);\n"
- "uvec4 subgroupShuffleXor(uvec4, uint);\n"
- "bool subgroupShuffleXor(bool, uint);\n"
- "bvec2 subgroupShuffleXor(bvec2, uint);\n"
- "bvec3 subgroupShuffleXor(bvec3, uint);\n"
- "bvec4 subgroupShuffleXor(bvec4, uint);\n"
-
- "float subgroupShuffleUp(float, uint delta);\n"
- "vec2 subgroupShuffleUp(vec2, uint delta);\n"
- "vec3 subgroupShuffleUp(vec3, uint delta);\n"
- "vec4 subgroupShuffleUp(vec4, uint delta);\n"
- "int subgroupShuffleUp(int, uint delta);\n"
- "ivec2 subgroupShuffleUp(ivec2, uint delta);\n"
- "ivec3 subgroupShuffleUp(ivec3, uint delta);\n"
- "ivec4 subgroupShuffleUp(ivec4, uint delta);\n"
- "uint subgroupShuffleUp(uint, uint delta);\n"
- "uvec2 subgroupShuffleUp(uvec2, uint delta);\n"
- "uvec3 subgroupShuffleUp(uvec3, uint delta);\n"
- "uvec4 subgroupShuffleUp(uvec4, uint delta);\n"
- "bool subgroupShuffleUp(bool, uint delta);\n"
- "bvec2 subgroupShuffleUp(bvec2, uint delta);\n"
- "bvec3 subgroupShuffleUp(bvec3, uint delta);\n"
- "bvec4 subgroupShuffleUp(bvec4, uint delta);\n"
-
- "float subgroupShuffleDown(float, uint delta);\n"
- "vec2 subgroupShuffleDown(vec2, uint delta);\n"
- "vec3 subgroupShuffleDown(vec3, uint delta);\n"
- "vec4 subgroupShuffleDown(vec4, uint delta);\n"
- "int subgroupShuffleDown(int, uint delta);\n"
- "ivec2 subgroupShuffleDown(ivec2, uint delta);\n"
- "ivec3 subgroupShuffleDown(ivec3, uint delta);\n"
- "ivec4 subgroupShuffleDown(ivec4, uint delta);\n"
- "uint subgroupShuffleDown(uint, uint delta);\n"
- "uvec2 subgroupShuffleDown(uvec2, uint delta);\n"
- "uvec3 subgroupShuffleDown(uvec3, uint delta);\n"
- "uvec4 subgroupShuffleDown(uvec4, uint delta);\n"
- "bool subgroupShuffleDown(bool, uint delta);\n"
- "bvec2 subgroupShuffleDown(bvec2, uint delta);\n"
- "bvec3 subgroupShuffleDown(bvec3, uint delta);\n"
- "bvec4 subgroupShuffleDown(bvec4, uint delta);\n"
-
- "float subgroupAdd(float);\n"
- "vec2 subgroupAdd(vec2);\n"
- "vec3 subgroupAdd(vec3);\n"
- "vec4 subgroupAdd(vec4);\n"
- "int subgroupAdd(int);\n"
- "ivec2 subgroupAdd(ivec2);\n"
- "ivec3 subgroupAdd(ivec3);\n"
- "ivec4 subgroupAdd(ivec4);\n"
- "uint subgroupAdd(uint);\n"
- "uvec2 subgroupAdd(uvec2);\n"
- "uvec3 subgroupAdd(uvec3);\n"
- "uvec4 subgroupAdd(uvec4);\n"
-
- "float subgroupMul(float);\n"
- "vec2 subgroupMul(vec2);\n"
- "vec3 subgroupMul(vec3);\n"
- "vec4 subgroupMul(vec4);\n"
- "int subgroupMul(int);\n"
- "ivec2 subgroupMul(ivec2);\n"
- "ivec3 subgroupMul(ivec3);\n"
- "ivec4 subgroupMul(ivec4);\n"
- "uint subgroupMul(uint);\n"
- "uvec2 subgroupMul(uvec2);\n"
- "uvec3 subgroupMul(uvec3);\n"
- "uvec4 subgroupMul(uvec4);\n"
-
- "float subgroupMin(float);\n"
- "vec2 subgroupMin(vec2);\n"
- "vec3 subgroupMin(vec3);\n"
- "vec4 subgroupMin(vec4);\n"
- "int subgroupMin(int);\n"
- "ivec2 subgroupMin(ivec2);\n"
- "ivec3 subgroupMin(ivec3);\n"
- "ivec4 subgroupMin(ivec4);\n"
- "uint subgroupMin(uint);\n"
- "uvec2 subgroupMin(uvec2);\n"
- "uvec3 subgroupMin(uvec3);\n"
- "uvec4 subgroupMin(uvec4);\n"
-
- "float subgroupMax(float);\n"
- "vec2 subgroupMax(vec2);\n"
- "vec3 subgroupMax(vec3);\n"
- "vec4 subgroupMax(vec4);\n"
- "int subgroupMax(int);\n"
- "ivec2 subgroupMax(ivec2);\n"
- "ivec3 subgroupMax(ivec3);\n"
- "ivec4 subgroupMax(ivec4);\n"
- "uint subgroupMax(uint);\n"
- "uvec2 subgroupMax(uvec2);\n"
- "uvec3 subgroupMax(uvec3);\n"
- "uvec4 subgroupMax(uvec4);\n"
-
- "int subgroupAnd(int);\n"
- "ivec2 subgroupAnd(ivec2);\n"
- "ivec3 subgroupAnd(ivec3);\n"
- "ivec4 subgroupAnd(ivec4);\n"
- "uint subgroupAnd(uint);\n"
- "uvec2 subgroupAnd(uvec2);\n"
- "uvec3 subgroupAnd(uvec3);\n"
- "uvec4 subgroupAnd(uvec4);\n"
- "bool subgroupAnd(bool);\n"
- "bvec2 subgroupAnd(bvec2);\n"
- "bvec3 subgroupAnd(bvec3);\n"
- "bvec4 subgroupAnd(bvec4);\n"
-
- "int subgroupOr(int);\n"
- "ivec2 subgroupOr(ivec2);\n"
- "ivec3 subgroupOr(ivec3);\n"
- "ivec4 subgroupOr(ivec4);\n"
- "uint subgroupOr(uint);\n"
- "uvec2 subgroupOr(uvec2);\n"
- "uvec3 subgroupOr(uvec3);\n"
- "uvec4 subgroupOr(uvec4);\n"
- "bool subgroupOr(bool);\n"
- "bvec2 subgroupOr(bvec2);\n"
- "bvec3 subgroupOr(bvec3);\n"
- "bvec4 subgroupOr(bvec4);\n"
-
- "int subgroupXor(int);\n"
- "ivec2 subgroupXor(ivec2);\n"
- "ivec3 subgroupXor(ivec3);\n"
- "ivec4 subgroupXor(ivec4);\n"
- "uint subgroupXor(uint);\n"
- "uvec2 subgroupXor(uvec2);\n"
- "uvec3 subgroupXor(uvec3);\n"
- "uvec4 subgroupXor(uvec4);\n"
- "bool subgroupXor(bool);\n"
- "bvec2 subgroupXor(bvec2);\n"
- "bvec3 subgroupXor(bvec3);\n"
- "bvec4 subgroupXor(bvec4);\n"
-
- "float subgroupInclusiveAdd(float);\n"
- "vec2 subgroupInclusiveAdd(vec2);\n"
- "vec3 subgroupInclusiveAdd(vec3);\n"
- "vec4 subgroupInclusiveAdd(vec4);\n"
- "int subgroupInclusiveAdd(int);\n"
- "ivec2 subgroupInclusiveAdd(ivec2);\n"
- "ivec3 subgroupInclusiveAdd(ivec3);\n"
- "ivec4 subgroupInclusiveAdd(ivec4);\n"
- "uint subgroupInclusiveAdd(uint);\n"
- "uvec2 subgroupInclusiveAdd(uvec2);\n"
- "uvec3 subgroupInclusiveAdd(uvec3);\n"
- "uvec4 subgroupInclusiveAdd(uvec4);\n"
-
- "float subgroupInclusiveMul(float);\n"
- "vec2 subgroupInclusiveMul(vec2);\n"
- "vec3 subgroupInclusiveMul(vec3);\n"
- "vec4 subgroupInclusiveMul(vec4);\n"
- "int subgroupInclusiveMul(int);\n"
- "ivec2 subgroupInclusiveMul(ivec2);\n"
- "ivec3 subgroupInclusiveMul(ivec3);\n"
- "ivec4 subgroupInclusiveMul(ivec4);\n"
- "uint subgroupInclusiveMul(uint);\n"
- "uvec2 subgroupInclusiveMul(uvec2);\n"
- "uvec3 subgroupInclusiveMul(uvec3);\n"
- "uvec4 subgroupInclusiveMul(uvec4);\n"
-
- "float subgroupInclusiveMin(float);\n"
- "vec2 subgroupInclusiveMin(vec2);\n"
- "vec3 subgroupInclusiveMin(vec3);\n"
- "vec4 subgroupInclusiveMin(vec4);\n"
- "int subgroupInclusiveMin(int);\n"
- "ivec2 subgroupInclusiveMin(ivec2);\n"
- "ivec3 subgroupInclusiveMin(ivec3);\n"
- "ivec4 subgroupInclusiveMin(ivec4);\n"
- "uint subgroupInclusiveMin(uint);\n"
- "uvec2 subgroupInclusiveMin(uvec2);\n"
- "uvec3 subgroupInclusiveMin(uvec3);\n"
- "uvec4 subgroupInclusiveMin(uvec4);\n"
-
- "float subgroupInclusiveMax(float);\n"
- "vec2 subgroupInclusiveMax(vec2);\n"
- "vec3 subgroupInclusiveMax(vec3);\n"
- "vec4 subgroupInclusiveMax(vec4);\n"
- "int subgroupInclusiveMax(int);\n"
- "ivec2 subgroupInclusiveMax(ivec2);\n"
- "ivec3 subgroupInclusiveMax(ivec3);\n"
- "ivec4 subgroupInclusiveMax(ivec4);\n"
- "uint subgroupInclusiveMax(uint);\n"
- "uvec2 subgroupInclusiveMax(uvec2);\n"
- "uvec3 subgroupInclusiveMax(uvec3);\n"
- "uvec4 subgroupInclusiveMax(uvec4);\n"
-
- "int subgroupInclusiveAnd(int);\n"
- "ivec2 subgroupInclusiveAnd(ivec2);\n"
- "ivec3 subgroupInclusiveAnd(ivec3);\n"
- "ivec4 subgroupInclusiveAnd(ivec4);\n"
- "uint subgroupInclusiveAnd(uint);\n"
- "uvec2 subgroupInclusiveAnd(uvec2);\n"
- "uvec3 subgroupInclusiveAnd(uvec3);\n"
- "uvec4 subgroupInclusiveAnd(uvec4);\n"
- "bool subgroupInclusiveAnd(bool);\n"
- "bvec2 subgroupInclusiveAnd(bvec2);\n"
- "bvec3 subgroupInclusiveAnd(bvec3);\n"
- "bvec4 subgroupInclusiveAnd(bvec4);\n"
-
- "int subgroupInclusiveOr(int);\n"
- "ivec2 subgroupInclusiveOr(ivec2);\n"
- "ivec3 subgroupInclusiveOr(ivec3);\n"
- "ivec4 subgroupInclusiveOr(ivec4);\n"
- "uint subgroupInclusiveOr(uint);\n"
- "uvec2 subgroupInclusiveOr(uvec2);\n"
- "uvec3 subgroupInclusiveOr(uvec3);\n"
- "uvec4 subgroupInclusiveOr(uvec4);\n"
- "bool subgroupInclusiveOr(bool);\n"
- "bvec2 subgroupInclusiveOr(bvec2);\n"
- "bvec3 subgroupInclusiveOr(bvec3);\n"
- "bvec4 subgroupInclusiveOr(bvec4);\n"
-
- "int subgroupInclusiveXor(int);\n"
- "ivec2 subgroupInclusiveXor(ivec2);\n"
- "ivec3 subgroupInclusiveXor(ivec3);\n"
- "ivec4 subgroupInclusiveXor(ivec4);\n"
- "uint subgroupInclusiveXor(uint);\n"
- "uvec2 subgroupInclusiveXor(uvec2);\n"
- "uvec3 subgroupInclusiveXor(uvec3);\n"
- "uvec4 subgroupInclusiveXor(uvec4);\n"
- "bool subgroupInclusiveXor(bool);\n"
- "bvec2 subgroupInclusiveXor(bvec2);\n"
- "bvec3 subgroupInclusiveXor(bvec3);\n"
- "bvec4 subgroupInclusiveXor(bvec4);\n"
-
- "float subgroupExclusiveAdd(float);\n"
- "vec2 subgroupExclusiveAdd(vec2);\n"
- "vec3 subgroupExclusiveAdd(vec3);\n"
- "vec4 subgroupExclusiveAdd(vec4);\n"
- "int subgroupExclusiveAdd(int);\n"
- "ivec2 subgroupExclusiveAdd(ivec2);\n"
- "ivec3 subgroupExclusiveAdd(ivec3);\n"
- "ivec4 subgroupExclusiveAdd(ivec4);\n"
- "uint subgroupExclusiveAdd(uint);\n"
- "uvec2 subgroupExclusiveAdd(uvec2);\n"
- "uvec3 subgroupExclusiveAdd(uvec3);\n"
- "uvec4 subgroupExclusiveAdd(uvec4);\n"
-
- "float subgroupExclusiveMul(float);\n"
- "vec2 subgroupExclusiveMul(vec2);\n"
- "vec3 subgroupExclusiveMul(vec3);\n"
- "vec4 subgroupExclusiveMul(vec4);\n"
- "int subgroupExclusiveMul(int);\n"
- "ivec2 subgroupExclusiveMul(ivec2);\n"
- "ivec3 subgroupExclusiveMul(ivec3);\n"
- "ivec4 subgroupExclusiveMul(ivec4);\n"
- "uint subgroupExclusiveMul(uint);\n"
- "uvec2 subgroupExclusiveMul(uvec2);\n"
- "uvec3 subgroupExclusiveMul(uvec3);\n"
- "uvec4 subgroupExclusiveMul(uvec4);\n"
-
- "float subgroupExclusiveMin(float);\n"
- "vec2 subgroupExclusiveMin(vec2);\n"
- "vec3 subgroupExclusiveMin(vec3);\n"
- "vec4 subgroupExclusiveMin(vec4);\n"
- "int subgroupExclusiveMin(int);\n"
- "ivec2 subgroupExclusiveMin(ivec2);\n"
- "ivec3 subgroupExclusiveMin(ivec3);\n"
- "ivec4 subgroupExclusiveMin(ivec4);\n"
- "uint subgroupExclusiveMin(uint);\n"
- "uvec2 subgroupExclusiveMin(uvec2);\n"
- "uvec3 subgroupExclusiveMin(uvec3);\n"
- "uvec4 subgroupExclusiveMin(uvec4);\n"
-
- "float subgroupExclusiveMax(float);\n"
- "vec2 subgroupExclusiveMax(vec2);\n"
- "vec3 subgroupExclusiveMax(vec3);\n"
- "vec4 subgroupExclusiveMax(vec4);\n"
- "int subgroupExclusiveMax(int);\n"
- "ivec2 subgroupExclusiveMax(ivec2);\n"
- "ivec3 subgroupExclusiveMax(ivec3);\n"
- "ivec4 subgroupExclusiveMax(ivec4);\n"
- "uint subgroupExclusiveMax(uint);\n"
- "uvec2 subgroupExclusiveMax(uvec2);\n"
- "uvec3 subgroupExclusiveMax(uvec3);\n"
- "uvec4 subgroupExclusiveMax(uvec4);\n"
-
- "int subgroupExclusiveAnd(int);\n"
- "ivec2 subgroupExclusiveAnd(ivec2);\n"
- "ivec3 subgroupExclusiveAnd(ivec3);\n"
- "ivec4 subgroupExclusiveAnd(ivec4);\n"
- "uint subgroupExclusiveAnd(uint);\n"
- "uvec2 subgroupExclusiveAnd(uvec2);\n"
- "uvec3 subgroupExclusiveAnd(uvec3);\n"
- "uvec4 subgroupExclusiveAnd(uvec4);\n"
- "bool subgroupExclusiveAnd(bool);\n"
- "bvec2 subgroupExclusiveAnd(bvec2);\n"
- "bvec3 subgroupExclusiveAnd(bvec3);\n"
- "bvec4 subgroupExclusiveAnd(bvec4);\n"
-
- "int subgroupExclusiveOr(int);\n"
- "ivec2 subgroupExclusiveOr(ivec2);\n"
- "ivec3 subgroupExclusiveOr(ivec3);\n"
- "ivec4 subgroupExclusiveOr(ivec4);\n"
- "uint subgroupExclusiveOr(uint);\n"
- "uvec2 subgroupExclusiveOr(uvec2);\n"
- "uvec3 subgroupExclusiveOr(uvec3);\n"
- "uvec4 subgroupExclusiveOr(uvec4);\n"
- "bool subgroupExclusiveOr(bool);\n"
- "bvec2 subgroupExclusiveOr(bvec2);\n"
- "bvec3 subgroupExclusiveOr(bvec3);\n"
- "bvec4 subgroupExclusiveOr(bvec4);\n"
-
- "int subgroupExclusiveXor(int);\n"
- "ivec2 subgroupExclusiveXor(ivec2);\n"
- "ivec3 subgroupExclusiveXor(ivec3);\n"
- "ivec4 subgroupExclusiveXor(ivec4);\n"
- "uint subgroupExclusiveXor(uint);\n"
- "uvec2 subgroupExclusiveXor(uvec2);\n"
- "uvec3 subgroupExclusiveXor(uvec3);\n"
- "uvec4 subgroupExclusiveXor(uvec4);\n"
- "bool subgroupExclusiveXor(bool);\n"
- "bvec2 subgroupExclusiveXor(bvec2);\n"
- "bvec3 subgroupExclusiveXor(bvec3);\n"
- "bvec4 subgroupExclusiveXor(bvec4);\n"
-
- "float subgroupClusteredAdd(float, uint);\n"
- "vec2 subgroupClusteredAdd(vec2, uint);\n"
- "vec3 subgroupClusteredAdd(vec3, uint);\n"
- "vec4 subgroupClusteredAdd(vec4, uint);\n"
- "int subgroupClusteredAdd(int, uint);\n"
- "ivec2 subgroupClusteredAdd(ivec2, uint);\n"
- "ivec3 subgroupClusteredAdd(ivec3, uint);\n"
- "ivec4 subgroupClusteredAdd(ivec4, uint);\n"
- "uint subgroupClusteredAdd(uint, uint);\n"
- "uvec2 subgroupClusteredAdd(uvec2, uint);\n"
- "uvec3 subgroupClusteredAdd(uvec3, uint);\n"
- "uvec4 subgroupClusteredAdd(uvec4, uint);\n"
-
- "float subgroupClusteredMul(float, uint);\n"
- "vec2 subgroupClusteredMul(vec2, uint);\n"
- "vec3 subgroupClusteredMul(vec3, uint);\n"
- "vec4 subgroupClusteredMul(vec4, uint);\n"
- "int subgroupClusteredMul(int, uint);\n"
- "ivec2 subgroupClusteredMul(ivec2, uint);\n"
- "ivec3 subgroupClusteredMul(ivec3, uint);\n"
- "ivec4 subgroupClusteredMul(ivec4, uint);\n"
- "uint subgroupClusteredMul(uint, uint);\n"
- "uvec2 subgroupClusteredMul(uvec2, uint);\n"
- "uvec3 subgroupClusteredMul(uvec3, uint);\n"
- "uvec4 subgroupClusteredMul(uvec4, uint);\n"
-
- "float subgroupClusteredMin(float, uint);\n"
- "vec2 subgroupClusteredMin(vec2, uint);\n"
- "vec3 subgroupClusteredMin(vec3, uint);\n"
- "vec4 subgroupClusteredMin(vec4, uint);\n"
- "int subgroupClusteredMin(int, uint);\n"
- "ivec2 subgroupClusteredMin(ivec2, uint);\n"
- "ivec3 subgroupClusteredMin(ivec3, uint);\n"
- "ivec4 subgroupClusteredMin(ivec4, uint);\n"
- "uint subgroupClusteredMin(uint, uint);\n"
- "uvec2 subgroupClusteredMin(uvec2, uint);\n"
- "uvec3 subgroupClusteredMin(uvec3, uint);\n"
- "uvec4 subgroupClusteredMin(uvec4, uint);\n"
-
- "float subgroupClusteredMax(float, uint);\n"
- "vec2 subgroupClusteredMax(vec2, uint);\n"
- "vec3 subgroupClusteredMax(vec3, uint);\n"
- "vec4 subgroupClusteredMax(vec4, uint);\n"
- "int subgroupClusteredMax(int, uint);\n"
- "ivec2 subgroupClusteredMax(ivec2, uint);\n"
- "ivec3 subgroupClusteredMax(ivec3, uint);\n"
- "ivec4 subgroupClusteredMax(ivec4, uint);\n"
- "uint subgroupClusteredMax(uint, uint);\n"
- "uvec2 subgroupClusteredMax(uvec2, uint);\n"
- "uvec3 subgroupClusteredMax(uvec3, uint);\n"
- "uvec4 subgroupClusteredMax(uvec4, uint);\n"
-
- "int subgroupClusteredAnd(int, uint);\n"
- "ivec2 subgroupClusteredAnd(ivec2, uint);\n"
- "ivec3 subgroupClusteredAnd(ivec3, uint);\n"
- "ivec4 subgroupClusteredAnd(ivec4, uint);\n"
- "uint subgroupClusteredAnd(uint, uint);\n"
- "uvec2 subgroupClusteredAnd(uvec2, uint);\n"
- "uvec3 subgroupClusteredAnd(uvec3, uint);\n"
- "uvec4 subgroupClusteredAnd(uvec4, uint);\n"
- "bool subgroupClusteredAnd(bool, uint);\n"
- "bvec2 subgroupClusteredAnd(bvec2, uint);\n"
- "bvec3 subgroupClusteredAnd(bvec3, uint);\n"
- "bvec4 subgroupClusteredAnd(bvec4, uint);\n"
-
- "int subgroupClusteredOr(int, uint);\n"
- "ivec2 subgroupClusteredOr(ivec2, uint);\n"
- "ivec3 subgroupClusteredOr(ivec3, uint);\n"
- "ivec4 subgroupClusteredOr(ivec4, uint);\n"
- "uint subgroupClusteredOr(uint, uint);\n"
- "uvec2 subgroupClusteredOr(uvec2, uint);\n"
- "uvec3 subgroupClusteredOr(uvec3, uint);\n"
- "uvec4 subgroupClusteredOr(uvec4, uint);\n"
- "bool subgroupClusteredOr(bool, uint);\n"
- "bvec2 subgroupClusteredOr(bvec2, uint);\n"
- "bvec3 subgroupClusteredOr(bvec3, uint);\n"
- "bvec4 subgroupClusteredOr(bvec4, uint);\n"
-
- "int subgroupClusteredXor(int, uint);\n"
- "ivec2 subgroupClusteredXor(ivec2, uint);\n"
- "ivec3 subgroupClusteredXor(ivec3, uint);\n"
- "ivec4 subgroupClusteredXor(ivec4, uint);\n"
- "uint subgroupClusteredXor(uint, uint);\n"
- "uvec2 subgroupClusteredXor(uvec2, uint);\n"
- "uvec3 subgroupClusteredXor(uvec3, uint);\n"
- "uvec4 subgroupClusteredXor(uvec4, uint);\n"
- "bool subgroupClusteredXor(bool, uint);\n"
- "bvec2 subgroupClusteredXor(bvec2, uint);\n"
- "bvec3 subgroupClusteredXor(bvec3, uint);\n"
- "bvec4 subgroupClusteredXor(bvec4, uint);\n"
-
- "float subgroupQuadBroadcast(float, uint);\n"
- "vec2 subgroupQuadBroadcast(vec2, uint);\n"
- "vec3 subgroupQuadBroadcast(vec3, uint);\n"
- "vec4 subgroupQuadBroadcast(vec4, uint);\n"
- "int subgroupQuadBroadcast(int, uint);\n"
- "ivec2 subgroupQuadBroadcast(ivec2, uint);\n"
- "ivec3 subgroupQuadBroadcast(ivec3, uint);\n"
- "ivec4 subgroupQuadBroadcast(ivec4, uint);\n"
- "uint subgroupQuadBroadcast(uint, uint);\n"
- "uvec2 subgroupQuadBroadcast(uvec2, uint);\n"
- "uvec3 subgroupQuadBroadcast(uvec3, uint);\n"
- "uvec4 subgroupQuadBroadcast(uvec4, uint);\n"
- "bool subgroupQuadBroadcast(bool, uint);\n"
- "bvec2 subgroupQuadBroadcast(bvec2, uint);\n"
- "bvec3 subgroupQuadBroadcast(bvec3, uint);\n"
- "bvec4 subgroupQuadBroadcast(bvec4, uint);\n"
-
- "float subgroupQuadSwapHorizontal(float);\n"
- "vec2 subgroupQuadSwapHorizontal(vec2);\n"
- "vec3 subgroupQuadSwapHorizontal(vec3);\n"
- "vec4 subgroupQuadSwapHorizontal(vec4);\n"
- "int subgroupQuadSwapHorizontal(int);\n"
- "ivec2 subgroupQuadSwapHorizontal(ivec2);\n"
- "ivec3 subgroupQuadSwapHorizontal(ivec3);\n"
- "ivec4 subgroupQuadSwapHorizontal(ivec4);\n"
- "uint subgroupQuadSwapHorizontal(uint);\n"
- "uvec2 subgroupQuadSwapHorizontal(uvec2);\n"
- "uvec3 subgroupQuadSwapHorizontal(uvec3);\n"
- "uvec4 subgroupQuadSwapHorizontal(uvec4);\n"
- "bool subgroupQuadSwapHorizontal(bool);\n"
- "bvec2 subgroupQuadSwapHorizontal(bvec2);\n"
- "bvec3 subgroupQuadSwapHorizontal(bvec3);\n"
- "bvec4 subgroupQuadSwapHorizontal(bvec4);\n"
-
- "float subgroupQuadSwapVertical(float);\n"
- "vec2 subgroupQuadSwapVertical(vec2);\n"
- "vec3 subgroupQuadSwapVertical(vec3);\n"
- "vec4 subgroupQuadSwapVertical(vec4);\n"
- "int subgroupQuadSwapVertical(int);\n"
- "ivec2 subgroupQuadSwapVertical(ivec2);\n"
- "ivec3 subgroupQuadSwapVertical(ivec3);\n"
- "ivec4 subgroupQuadSwapVertical(ivec4);\n"
- "uint subgroupQuadSwapVertical(uint);\n"
- "uvec2 subgroupQuadSwapVertical(uvec2);\n"
- "uvec3 subgroupQuadSwapVertical(uvec3);\n"
- "uvec4 subgroupQuadSwapVertical(uvec4);\n"
- "bool subgroupQuadSwapVertical(bool);\n"
- "bvec2 subgroupQuadSwapVertical(bvec2);\n"
- "bvec3 subgroupQuadSwapVertical(bvec3);\n"
- "bvec4 subgroupQuadSwapVertical(bvec4);\n"
-
- "float subgroupQuadSwapDiagonal(float);\n"
- "vec2 subgroupQuadSwapDiagonal(vec2);\n"
- "vec3 subgroupQuadSwapDiagonal(vec3);\n"
- "vec4 subgroupQuadSwapDiagonal(vec4);\n"
- "int subgroupQuadSwapDiagonal(int);\n"
- "ivec2 subgroupQuadSwapDiagonal(ivec2);\n"
- "ivec3 subgroupQuadSwapDiagonal(ivec3);\n"
- "ivec4 subgroupQuadSwapDiagonal(ivec4);\n"
- "uint subgroupQuadSwapDiagonal(uint);\n"
- "uvec2 subgroupQuadSwapDiagonal(uvec2);\n"
- "uvec3 subgroupQuadSwapDiagonal(uvec3);\n"
- "uvec4 subgroupQuadSwapDiagonal(uvec4);\n"
- "bool subgroupQuadSwapDiagonal(bool);\n"
- "bvec2 subgroupQuadSwapDiagonal(bvec2);\n"
- "bvec3 subgroupQuadSwapDiagonal(bvec3);\n"
- "bvec4 subgroupQuadSwapDiagonal(bvec4);\n");
-
- // GL_EXT_shader_subgroup_extended_types
- commonBuiltins.append(
- "bool subgroupAllEqual(int8_t);\n"
- "bool subgroupAllEqual(i8vec2);\n"
- "bool subgroupAllEqual(i8vec3);\n"
- "bool subgroupAllEqual(i8vec4);\n"
- "bool subgroupAllEqual(uint8_t);\n"
- "bool subgroupAllEqual(u8vec2);\n"
- "bool subgroupAllEqual(u8vec3);\n"
- "bool subgroupAllEqual(u8vec4);\n"
-
- "int8_t subgroupBroadcast(int8_t, uint);\n"
- "i8vec2 subgroupBroadcast(i8vec2, uint);\n"
- "i8vec3 subgroupBroadcast(i8vec3, uint);\n"
- "i8vec4 subgroupBroadcast(i8vec4, uint);\n"
- "uint8_t subgroupBroadcast(uint8_t, uint);\n"
- "u8vec2 subgroupBroadcast(u8vec2, uint);\n"
- "u8vec3 subgroupBroadcast(u8vec3, uint);\n"
- "u8vec4 subgroupBroadcast(u8vec4, uint);\n"
-
- "int8_t subgroupBroadcastFirst(int8_t);\n"
- "i8vec2 subgroupBroadcastFirst(i8vec2);\n"
- "i8vec3 subgroupBroadcastFirst(i8vec3);\n"
- "i8vec4 subgroupBroadcastFirst(i8vec4);\n"
- "uint8_t subgroupBroadcastFirst(uint8_t);\n"
- "u8vec2 subgroupBroadcastFirst(u8vec2);\n"
- "u8vec3 subgroupBroadcastFirst(u8vec3);\n"
- "u8vec4 subgroupBroadcastFirst(u8vec4);\n"
-
- "int8_t subgroupShuffle(int8_t, uint);\n"
- "i8vec2 subgroupShuffle(i8vec2, uint);\n"
- "i8vec3 subgroupShuffle(i8vec3, uint);\n"
- "i8vec4 subgroupShuffle(i8vec4, uint);\n"
- "uint8_t subgroupShuffle(uint8_t, uint);\n"
- "u8vec2 subgroupShuffle(u8vec2, uint);\n"
- "u8vec3 subgroupShuffle(u8vec3, uint);\n"
- "u8vec4 subgroupShuffle(u8vec4, uint);\n"
-
- "int8_t subgroupShuffleXor(int8_t, uint);\n"
- "i8vec2 subgroupShuffleXor(i8vec2, uint);\n"
- "i8vec3 subgroupShuffleXor(i8vec3, uint);\n"
- "i8vec4 subgroupShuffleXor(i8vec4, uint);\n"
- "uint8_t subgroupShuffleXor(uint8_t, uint);\n"
- "u8vec2 subgroupShuffleXor(u8vec2, uint);\n"
- "u8vec3 subgroupShuffleXor(u8vec3, uint);\n"
- "u8vec4 subgroupShuffleXor(u8vec4, uint);\n"
-
- "int8_t subgroupShuffleUp(int8_t, uint delta);\n"
- "i8vec2 subgroupShuffleUp(i8vec2, uint delta);\n"
- "i8vec3 subgroupShuffleUp(i8vec3, uint delta);\n"
- "i8vec4 subgroupShuffleUp(i8vec4, uint delta);\n"
- "uint8_t subgroupShuffleUp(uint8_t, uint delta);\n"
- "u8vec2 subgroupShuffleUp(u8vec2, uint delta);\n"
- "u8vec3 subgroupShuffleUp(u8vec3, uint delta);\n"
- "u8vec4 subgroupShuffleUp(u8vec4, uint delta);\n"
-
- "int8_t subgroupShuffleDown(int8_t, uint delta);\n"
- "i8vec2 subgroupShuffleDown(i8vec2, uint delta);\n"
- "i8vec3 subgroupShuffleDown(i8vec3, uint delta);\n"
- "i8vec4 subgroupShuffleDown(i8vec4, uint delta);\n"
- "uint8_t subgroupShuffleDown(uint8_t, uint delta);\n"
- "u8vec2 subgroupShuffleDown(u8vec2, uint delta);\n"
- "u8vec3 subgroupShuffleDown(u8vec3, uint delta);\n"
- "u8vec4 subgroupShuffleDown(u8vec4, uint delta);\n"
-
- "int8_t subgroupAdd(int8_t);\n"
- "i8vec2 subgroupAdd(i8vec2);\n"
- "i8vec3 subgroupAdd(i8vec3);\n"
- "i8vec4 subgroupAdd(i8vec4);\n"
- "uint8_t subgroupAdd(uint8_t);\n"
- "u8vec2 subgroupAdd(u8vec2);\n"
- "u8vec3 subgroupAdd(u8vec3);\n"
- "u8vec4 subgroupAdd(u8vec4);\n"
-
- "int8_t subgroupMul(int8_t);\n"
- "i8vec2 subgroupMul(i8vec2);\n"
- "i8vec3 subgroupMul(i8vec3);\n"
- "i8vec4 subgroupMul(i8vec4);\n"
- "uint8_t subgroupMul(uint8_t);\n"
- "u8vec2 subgroupMul(u8vec2);\n"
- "u8vec3 subgroupMul(u8vec3);\n"
- "u8vec4 subgroupMul(u8vec4);\n"
-
- "int8_t subgroupMin(int8_t);\n"
- "i8vec2 subgroupMin(i8vec2);\n"
- "i8vec3 subgroupMin(i8vec3);\n"
- "i8vec4 subgroupMin(i8vec4);\n"
- "uint8_t subgroupMin(uint8_t);\n"
- "u8vec2 subgroupMin(u8vec2);\n"
- "u8vec3 subgroupMin(u8vec3);\n"
- "u8vec4 subgroupMin(u8vec4);\n"
-
- "int8_t subgroupMax(int8_t);\n"
- "i8vec2 subgroupMax(i8vec2);\n"
- "i8vec3 subgroupMax(i8vec3);\n"
- "i8vec4 subgroupMax(i8vec4);\n"
- "uint8_t subgroupMax(uint8_t);\n"
- "u8vec2 subgroupMax(u8vec2);\n"
- "u8vec3 subgroupMax(u8vec3);\n"
- "u8vec4 subgroupMax(u8vec4);\n"
-
- "int8_t subgroupAnd(int8_t);\n"
- "i8vec2 subgroupAnd(i8vec2);\n"
- "i8vec3 subgroupAnd(i8vec3);\n"
- "i8vec4 subgroupAnd(i8vec4);\n"
- "uint8_t subgroupAnd(uint8_t);\n"
- "u8vec2 subgroupAnd(u8vec2);\n"
- "u8vec3 subgroupAnd(u8vec3);\n"
- "u8vec4 subgroupAnd(u8vec4);\n"
-
- "int8_t subgroupOr(int8_t);\n"
- "i8vec2 subgroupOr(i8vec2);\n"
- "i8vec3 subgroupOr(i8vec3);\n"
- "i8vec4 subgroupOr(i8vec4);\n"
- "uint8_t subgroupOr(uint8_t);\n"
- "u8vec2 subgroupOr(u8vec2);\n"
- "u8vec3 subgroupOr(u8vec3);\n"
- "u8vec4 subgroupOr(u8vec4);\n"
-
- "int8_t subgroupXor(int8_t);\n"
- "i8vec2 subgroupXor(i8vec2);\n"
- "i8vec3 subgroupXor(i8vec3);\n"
- "i8vec4 subgroupXor(i8vec4);\n"
- "uint8_t subgroupXor(uint8_t);\n"
- "u8vec2 subgroupXor(u8vec2);\n"
- "u8vec3 subgroupXor(u8vec3);\n"
- "u8vec4 subgroupXor(u8vec4);\n"
-
- "int8_t subgroupInclusiveAdd(int8_t);\n"
- "i8vec2 subgroupInclusiveAdd(i8vec2);\n"
- "i8vec3 subgroupInclusiveAdd(i8vec3);\n"
- "i8vec4 subgroupInclusiveAdd(i8vec4);\n"
- "uint8_t subgroupInclusiveAdd(uint8_t);\n"
- "u8vec2 subgroupInclusiveAdd(u8vec2);\n"
- "u8vec3 subgroupInclusiveAdd(u8vec3);\n"
- "u8vec4 subgroupInclusiveAdd(u8vec4);\n"
-
- "int8_t subgroupInclusiveMul(int8_t);\n"
- "i8vec2 subgroupInclusiveMul(i8vec2);\n"
- "i8vec3 subgroupInclusiveMul(i8vec3);\n"
- "i8vec4 subgroupInclusiveMul(i8vec4);\n"
- "uint8_t subgroupInclusiveMul(uint8_t);\n"
- "u8vec2 subgroupInclusiveMul(u8vec2);\n"
- "u8vec3 subgroupInclusiveMul(u8vec3);\n"
- "u8vec4 subgroupInclusiveMul(u8vec4);\n"
-
- "int8_t subgroupInclusiveMin(int8_t);\n"
- "i8vec2 subgroupInclusiveMin(i8vec2);\n"
- "i8vec3 subgroupInclusiveMin(i8vec3);\n"
- "i8vec4 subgroupInclusiveMin(i8vec4);\n"
- "uint8_t subgroupInclusiveMin(uint8_t);\n"
- "u8vec2 subgroupInclusiveMin(u8vec2);\n"
- "u8vec3 subgroupInclusiveMin(u8vec3);\n"
- "u8vec4 subgroupInclusiveMin(u8vec4);\n"
-
- "int8_t subgroupInclusiveMax(int8_t);\n"
- "i8vec2 subgroupInclusiveMax(i8vec2);\n"
- "i8vec3 subgroupInclusiveMax(i8vec3);\n"
- "i8vec4 subgroupInclusiveMax(i8vec4);\n"
- "uint8_t subgroupInclusiveMax(uint8_t);\n"
- "u8vec2 subgroupInclusiveMax(u8vec2);\n"
- "u8vec3 subgroupInclusiveMax(u8vec3);\n"
- "u8vec4 subgroupInclusiveMax(u8vec4);\n"
-
- "int8_t subgroupInclusiveAnd(int8_t);\n"
- "i8vec2 subgroupInclusiveAnd(i8vec2);\n"
- "i8vec3 subgroupInclusiveAnd(i8vec3);\n"
- "i8vec4 subgroupInclusiveAnd(i8vec4);\n"
- "uint8_t subgroupInclusiveAnd(uint8_t);\n"
- "u8vec2 subgroupInclusiveAnd(u8vec2);\n"
- "u8vec3 subgroupInclusiveAnd(u8vec3);\n"
- "u8vec4 subgroupInclusiveAnd(u8vec4);\n"
-
- "int8_t subgroupInclusiveOr(int8_t);\n"
- "i8vec2 subgroupInclusiveOr(i8vec2);\n"
- "i8vec3 subgroupInclusiveOr(i8vec3);\n"
- "i8vec4 subgroupInclusiveOr(i8vec4);\n"
- "uint8_t subgroupInclusiveOr(uint8_t);\n"
- "u8vec2 subgroupInclusiveOr(u8vec2);\n"
- "u8vec3 subgroupInclusiveOr(u8vec3);\n"
- "u8vec4 subgroupInclusiveOr(u8vec4);\n"
-
- "int8_t subgroupInclusiveXor(int8_t);\n"
- "i8vec2 subgroupInclusiveXor(i8vec2);\n"
- "i8vec3 subgroupInclusiveXor(i8vec3);\n"
- "i8vec4 subgroupInclusiveXor(i8vec4);\n"
- "uint8_t subgroupInclusiveXor(uint8_t);\n"
- "u8vec2 subgroupInclusiveXor(u8vec2);\n"
- "u8vec3 subgroupInclusiveXor(u8vec3);\n"
- "u8vec4 subgroupInclusiveXor(u8vec4);\n"
-
- "int8_t subgroupExclusiveAdd(int8_t);\n"
- "i8vec2 subgroupExclusiveAdd(i8vec2);\n"
- "i8vec3 subgroupExclusiveAdd(i8vec3);\n"
- "i8vec4 subgroupExclusiveAdd(i8vec4);\n"
- "uint8_t subgroupExclusiveAdd(uint8_t);\n"
- "u8vec2 subgroupExclusiveAdd(u8vec2);\n"
- "u8vec3 subgroupExclusiveAdd(u8vec3);\n"
- "u8vec4 subgroupExclusiveAdd(u8vec4);\n"
-
- "int8_t subgroupExclusiveMul(int8_t);\n"
- "i8vec2 subgroupExclusiveMul(i8vec2);\n"
- "i8vec3 subgroupExclusiveMul(i8vec3);\n"
- "i8vec4 subgroupExclusiveMul(i8vec4);\n"
- "uint8_t subgroupExclusiveMul(uint8_t);\n"
- "u8vec2 subgroupExclusiveMul(u8vec2);\n"
- "u8vec3 subgroupExclusiveMul(u8vec3);\n"
- "u8vec4 subgroupExclusiveMul(u8vec4);\n"
-
- "int8_t subgroupExclusiveMin(int8_t);\n"
- "i8vec2 subgroupExclusiveMin(i8vec2);\n"
- "i8vec3 subgroupExclusiveMin(i8vec3);\n"
- "i8vec4 subgroupExclusiveMin(i8vec4);\n"
- "uint8_t subgroupExclusiveMin(uint8_t);\n"
- "u8vec2 subgroupExclusiveMin(u8vec2);\n"
- "u8vec3 subgroupExclusiveMin(u8vec3);\n"
- "u8vec4 subgroupExclusiveMin(u8vec4);\n"
-
- "int8_t subgroupExclusiveMax(int8_t);\n"
- "i8vec2 subgroupExclusiveMax(i8vec2);\n"
- "i8vec3 subgroupExclusiveMax(i8vec3);\n"
- "i8vec4 subgroupExclusiveMax(i8vec4);\n"
- "uint8_t subgroupExclusiveMax(uint8_t);\n"
- "u8vec2 subgroupExclusiveMax(u8vec2);\n"
- "u8vec3 subgroupExclusiveMax(u8vec3);\n"
- "u8vec4 subgroupExclusiveMax(u8vec4);\n"
-
- "int8_t subgroupExclusiveAnd(int8_t);\n"
- "i8vec2 subgroupExclusiveAnd(i8vec2);\n"
- "i8vec3 subgroupExclusiveAnd(i8vec3);\n"
- "i8vec4 subgroupExclusiveAnd(i8vec4);\n"
- "uint8_t subgroupExclusiveAnd(uint8_t);\n"
- "u8vec2 subgroupExclusiveAnd(u8vec2);\n"
- "u8vec3 subgroupExclusiveAnd(u8vec3);\n"
- "u8vec4 subgroupExclusiveAnd(u8vec4);\n"
-
- "int8_t subgroupExclusiveOr(int8_t);\n"
- "i8vec2 subgroupExclusiveOr(i8vec2);\n"
- "i8vec3 subgroupExclusiveOr(i8vec3);\n"
- "i8vec4 subgroupExclusiveOr(i8vec4);\n"
- "uint8_t subgroupExclusiveOr(uint8_t);\n"
- "u8vec2 subgroupExclusiveOr(u8vec2);\n"
- "u8vec3 subgroupExclusiveOr(u8vec3);\n"
- "u8vec4 subgroupExclusiveOr(u8vec4);\n"
-
- "int8_t subgroupExclusiveXor(int8_t);\n"
- "i8vec2 subgroupExclusiveXor(i8vec2);\n"
- "i8vec3 subgroupExclusiveXor(i8vec3);\n"
- "i8vec4 subgroupExclusiveXor(i8vec4);\n"
- "uint8_t subgroupExclusiveXor(uint8_t);\n"
- "u8vec2 subgroupExclusiveXor(u8vec2);\n"
- "u8vec3 subgroupExclusiveXor(u8vec3);\n"
- "u8vec4 subgroupExclusiveXor(u8vec4);\n"
-
- "int8_t subgroupClusteredAdd(int8_t, uint);\n"
- "i8vec2 subgroupClusteredAdd(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredAdd(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredAdd(i8vec4, uint);\n"
- "uint8_t subgroupClusteredAdd(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredAdd(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredAdd(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredAdd(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredMul(int8_t, uint);\n"
- "i8vec2 subgroupClusteredMul(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredMul(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredMul(i8vec4, uint);\n"
- "uint8_t subgroupClusteredMul(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredMul(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredMul(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredMul(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredMin(int8_t, uint);\n"
- "i8vec2 subgroupClusteredMin(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredMin(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredMin(i8vec4, uint);\n"
- "uint8_t subgroupClusteredMin(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredMin(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredMin(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredMin(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredMax(int8_t, uint);\n"
- "i8vec2 subgroupClusteredMax(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredMax(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredMax(i8vec4, uint);\n"
- "uint8_t subgroupClusteredMax(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredMax(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredMax(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredMax(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredAnd(int8_t, uint);\n"
- "i8vec2 subgroupClusteredAnd(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredAnd(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredAnd(i8vec4, uint);\n"
- "uint8_t subgroupClusteredAnd(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredAnd(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredAnd(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredAnd(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredOr(int8_t, uint);\n"
- "i8vec2 subgroupClusteredOr(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredOr(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredOr(i8vec4, uint);\n"
- "uint8_t subgroupClusteredOr(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredOr(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredOr(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredOr(u8vec4, uint);\n"
-
- "int8_t subgroupClusteredXor(int8_t, uint);\n"
- "i8vec2 subgroupClusteredXor(i8vec2, uint);\n"
- "i8vec3 subgroupClusteredXor(i8vec3, uint);\n"
- "i8vec4 subgroupClusteredXor(i8vec4, uint);\n"
- "uint8_t subgroupClusteredXor(uint8_t, uint);\n"
- "u8vec2 subgroupClusteredXor(u8vec2, uint);\n"
- "u8vec3 subgroupClusteredXor(u8vec3, uint);\n"
- "u8vec4 subgroupClusteredXor(u8vec4, uint);\n"
-
- "int8_t subgroupQuadBroadcast(int8_t, uint);\n"
- "i8vec2 subgroupQuadBroadcast(i8vec2, uint);\n"
- "i8vec3 subgroupQuadBroadcast(i8vec3, uint);\n"
- "i8vec4 subgroupQuadBroadcast(i8vec4, uint);\n"
- "uint8_t subgroupQuadBroadcast(uint8_t, uint);\n"
- "u8vec2 subgroupQuadBroadcast(u8vec2, uint);\n"
- "u8vec3 subgroupQuadBroadcast(u8vec3, uint);\n"
- "u8vec4 subgroupQuadBroadcast(u8vec4, uint);\n"
-
- "int8_t subgroupQuadSwapHorizontal(int8_t);\n"
- "i8vec2 subgroupQuadSwapHorizontal(i8vec2);\n"
- "i8vec3 subgroupQuadSwapHorizontal(i8vec3);\n"
- "i8vec4 subgroupQuadSwapHorizontal(i8vec4);\n"
- "uint8_t subgroupQuadSwapHorizontal(uint8_t);\n"
- "u8vec2 subgroupQuadSwapHorizontal(u8vec2);\n"
- "u8vec3 subgroupQuadSwapHorizontal(u8vec3);\n"
- "u8vec4 subgroupQuadSwapHorizontal(u8vec4);\n"
-
- "int8_t subgroupQuadSwapVertical(int8_t);\n"
- "i8vec2 subgroupQuadSwapVertical(i8vec2);\n"
- "i8vec3 subgroupQuadSwapVertical(i8vec3);\n"
- "i8vec4 subgroupQuadSwapVertical(i8vec4);\n"
- "uint8_t subgroupQuadSwapVertical(uint8_t);\n"
- "u8vec2 subgroupQuadSwapVertical(u8vec2);\n"
- "u8vec3 subgroupQuadSwapVertical(u8vec3);\n"
- "u8vec4 subgroupQuadSwapVertical(u8vec4);\n"
-
- "int8_t subgroupQuadSwapDiagonal(int8_t);\n"
- "i8vec2 subgroupQuadSwapDiagonal(i8vec2);\n"
- "i8vec3 subgroupQuadSwapDiagonal(i8vec3);\n"
- "i8vec4 subgroupQuadSwapDiagonal(i8vec4);\n"
- "uint8_t subgroupQuadSwapDiagonal(uint8_t);\n"
- "u8vec2 subgroupQuadSwapDiagonal(u8vec2);\n"
- "u8vec3 subgroupQuadSwapDiagonal(u8vec3);\n"
- "u8vec4 subgroupQuadSwapDiagonal(u8vec4);\n"
-
- "bool subgroupAllEqual(int16_t);\n"
- "bool subgroupAllEqual(i16vec2);\n"
- "bool subgroupAllEqual(i16vec3);\n"
- "bool subgroupAllEqual(i16vec4);\n"
- "bool subgroupAllEqual(uint16_t);\n"
- "bool subgroupAllEqual(u16vec2);\n"
- "bool subgroupAllEqual(u16vec3);\n"
- "bool subgroupAllEqual(u16vec4);\n"
-
- "int16_t subgroupBroadcast(int16_t, uint);\n"
- "i16vec2 subgroupBroadcast(i16vec2, uint);\n"
- "i16vec3 subgroupBroadcast(i16vec3, uint);\n"
- "i16vec4 subgroupBroadcast(i16vec4, uint);\n"
- "uint16_t subgroupBroadcast(uint16_t, uint);\n"
- "u16vec2 subgroupBroadcast(u16vec2, uint);\n"
- "u16vec3 subgroupBroadcast(u16vec3, uint);\n"
- "u16vec4 subgroupBroadcast(u16vec4, uint);\n"
-
- "int16_t subgroupBroadcastFirst(int16_t);\n"
- "i16vec2 subgroupBroadcastFirst(i16vec2);\n"
- "i16vec3 subgroupBroadcastFirst(i16vec3);\n"
- "i16vec4 subgroupBroadcastFirst(i16vec4);\n"
- "uint16_t subgroupBroadcastFirst(uint16_t);\n"
- "u16vec2 subgroupBroadcastFirst(u16vec2);\n"
- "u16vec3 subgroupBroadcastFirst(u16vec3);\n"
- "u16vec4 subgroupBroadcastFirst(u16vec4);\n"
-
- "int16_t subgroupShuffle(int16_t, uint);\n"
- "i16vec2 subgroupShuffle(i16vec2, uint);\n"
- "i16vec3 subgroupShuffle(i16vec3, uint);\n"
- "i16vec4 subgroupShuffle(i16vec4, uint);\n"
- "uint16_t subgroupShuffle(uint16_t, uint);\n"
- "u16vec2 subgroupShuffle(u16vec2, uint);\n"
- "u16vec3 subgroupShuffle(u16vec3, uint);\n"
- "u16vec4 subgroupShuffle(u16vec4, uint);\n"
-
- "int16_t subgroupShuffleXor(int16_t, uint);\n"
- "i16vec2 subgroupShuffleXor(i16vec2, uint);\n"
- "i16vec3 subgroupShuffleXor(i16vec3, uint);\n"
- "i16vec4 subgroupShuffleXor(i16vec4, uint);\n"
- "uint16_t subgroupShuffleXor(uint16_t, uint);\n"
- "u16vec2 subgroupShuffleXor(u16vec2, uint);\n"
- "u16vec3 subgroupShuffleXor(u16vec3, uint);\n"
- "u16vec4 subgroupShuffleXor(u16vec4, uint);\n"
-
- "int16_t subgroupShuffleUp(int16_t, uint delta);\n"
- "i16vec2 subgroupShuffleUp(i16vec2, uint delta);\n"
- "i16vec3 subgroupShuffleUp(i16vec3, uint delta);\n"
- "i16vec4 subgroupShuffleUp(i16vec4, uint delta);\n"
- "uint16_t subgroupShuffleUp(uint16_t, uint delta);\n"
- "u16vec2 subgroupShuffleUp(u16vec2, uint delta);\n"
- "u16vec3 subgroupShuffleUp(u16vec3, uint delta);\n"
- "u16vec4 subgroupShuffleUp(u16vec4, uint delta);\n"
-
- "int16_t subgroupShuffleDown(int16_t, uint delta);\n"
- "i16vec2 subgroupShuffleDown(i16vec2, uint delta);\n"
- "i16vec3 subgroupShuffleDown(i16vec3, uint delta);\n"
- "i16vec4 subgroupShuffleDown(i16vec4, uint delta);\n"
- "uint16_t subgroupShuffleDown(uint16_t, uint delta);\n"
- "u16vec2 subgroupShuffleDown(u16vec2, uint delta);\n"
- "u16vec3 subgroupShuffleDown(u16vec3, uint delta);\n"
- "u16vec4 subgroupShuffleDown(u16vec4, uint delta);\n"
-
- "int16_t subgroupAdd(int16_t);\n"
- "i16vec2 subgroupAdd(i16vec2);\n"
- "i16vec3 subgroupAdd(i16vec3);\n"
- "i16vec4 subgroupAdd(i16vec4);\n"
- "uint16_t subgroupAdd(uint16_t);\n"
- "u16vec2 subgroupAdd(u16vec2);\n"
- "u16vec3 subgroupAdd(u16vec3);\n"
- "u16vec4 subgroupAdd(u16vec4);\n"
-
- "int16_t subgroupMul(int16_t);\n"
- "i16vec2 subgroupMul(i16vec2);\n"
- "i16vec3 subgroupMul(i16vec3);\n"
- "i16vec4 subgroupMul(i16vec4);\n"
- "uint16_t subgroupMul(uint16_t);\n"
- "u16vec2 subgroupMul(u16vec2);\n"
- "u16vec3 subgroupMul(u16vec3);\n"
- "u16vec4 subgroupMul(u16vec4);\n"
-
- "int16_t subgroupMin(int16_t);\n"
- "i16vec2 subgroupMin(i16vec2);\n"
- "i16vec3 subgroupMin(i16vec3);\n"
- "i16vec4 subgroupMin(i16vec4);\n"
- "uint16_t subgroupMin(uint16_t);\n"
- "u16vec2 subgroupMin(u16vec2);\n"
- "u16vec3 subgroupMin(u16vec3);\n"
- "u16vec4 subgroupMin(u16vec4);\n"
-
- "int16_t subgroupMax(int16_t);\n"
- "i16vec2 subgroupMax(i16vec2);\n"
- "i16vec3 subgroupMax(i16vec3);\n"
- "i16vec4 subgroupMax(i16vec4);\n"
- "uint16_t subgroupMax(uint16_t);\n"
- "u16vec2 subgroupMax(u16vec2);\n"
- "u16vec3 subgroupMax(u16vec3);\n"
- "u16vec4 subgroupMax(u16vec4);\n"
-
- "int16_t subgroupAnd(int16_t);\n"
- "i16vec2 subgroupAnd(i16vec2);\n"
- "i16vec3 subgroupAnd(i16vec3);\n"
- "i16vec4 subgroupAnd(i16vec4);\n"
- "uint16_t subgroupAnd(uint16_t);\n"
- "u16vec2 subgroupAnd(u16vec2);\n"
- "u16vec3 subgroupAnd(u16vec3);\n"
- "u16vec4 subgroupAnd(u16vec4);\n"
-
- "int16_t subgroupOr(int16_t);\n"
- "i16vec2 subgroupOr(i16vec2);\n"
- "i16vec3 subgroupOr(i16vec3);\n"
- "i16vec4 subgroupOr(i16vec4);\n"
- "uint16_t subgroupOr(uint16_t);\n"
- "u16vec2 subgroupOr(u16vec2);\n"
- "u16vec3 subgroupOr(u16vec3);\n"
- "u16vec4 subgroupOr(u16vec4);\n"
-
- "int16_t subgroupXor(int16_t);\n"
- "i16vec2 subgroupXor(i16vec2);\n"
- "i16vec3 subgroupXor(i16vec3);\n"
- "i16vec4 subgroupXor(i16vec4);\n"
- "uint16_t subgroupXor(uint16_t);\n"
- "u16vec2 subgroupXor(u16vec2);\n"
- "u16vec3 subgroupXor(u16vec3);\n"
- "u16vec4 subgroupXor(u16vec4);\n"
-
- "int16_t subgroupInclusiveAdd(int16_t);\n"
- "i16vec2 subgroupInclusiveAdd(i16vec2);\n"
- "i16vec3 subgroupInclusiveAdd(i16vec3);\n"
- "i16vec4 subgroupInclusiveAdd(i16vec4);\n"
- "uint16_t subgroupInclusiveAdd(uint16_t);\n"
- "u16vec2 subgroupInclusiveAdd(u16vec2);\n"
- "u16vec3 subgroupInclusiveAdd(u16vec3);\n"
- "u16vec4 subgroupInclusiveAdd(u16vec4);\n"
-
- "int16_t subgroupInclusiveMul(int16_t);\n"
- "i16vec2 subgroupInclusiveMul(i16vec2);\n"
- "i16vec3 subgroupInclusiveMul(i16vec3);\n"
- "i16vec4 subgroupInclusiveMul(i16vec4);\n"
- "uint16_t subgroupInclusiveMul(uint16_t);\n"
- "u16vec2 subgroupInclusiveMul(u16vec2);\n"
- "u16vec3 subgroupInclusiveMul(u16vec3);\n"
- "u16vec4 subgroupInclusiveMul(u16vec4);\n"
-
- "int16_t subgroupInclusiveMin(int16_t);\n"
- "i16vec2 subgroupInclusiveMin(i16vec2);\n"
- "i16vec3 subgroupInclusiveMin(i16vec3);\n"
- "i16vec4 subgroupInclusiveMin(i16vec4);\n"
- "uint16_t subgroupInclusiveMin(uint16_t);\n"
- "u16vec2 subgroupInclusiveMin(u16vec2);\n"
- "u16vec3 subgroupInclusiveMin(u16vec3);\n"
- "u16vec4 subgroupInclusiveMin(u16vec4);\n"
-
- "int16_t subgroupInclusiveMax(int16_t);\n"
- "i16vec2 subgroupInclusiveMax(i16vec2);\n"
- "i16vec3 subgroupInclusiveMax(i16vec3);\n"
- "i16vec4 subgroupInclusiveMax(i16vec4);\n"
- "uint16_t subgroupInclusiveMax(uint16_t);\n"
- "u16vec2 subgroupInclusiveMax(u16vec2);\n"
- "u16vec3 subgroupInclusiveMax(u16vec3);\n"
- "u16vec4 subgroupInclusiveMax(u16vec4);\n"
-
- "int16_t subgroupInclusiveAnd(int16_t);\n"
- "i16vec2 subgroupInclusiveAnd(i16vec2);\n"
- "i16vec3 subgroupInclusiveAnd(i16vec3);\n"
- "i16vec4 subgroupInclusiveAnd(i16vec4);\n"
- "uint16_t subgroupInclusiveAnd(uint16_t);\n"
- "u16vec2 subgroupInclusiveAnd(u16vec2);\n"
- "u16vec3 subgroupInclusiveAnd(u16vec3);\n"
- "u16vec4 subgroupInclusiveAnd(u16vec4);\n"
-
- "int16_t subgroupInclusiveOr(int16_t);\n"
- "i16vec2 subgroupInclusiveOr(i16vec2);\n"
- "i16vec3 subgroupInclusiveOr(i16vec3);\n"
- "i16vec4 subgroupInclusiveOr(i16vec4);\n"
- "uint16_t subgroupInclusiveOr(uint16_t);\n"
- "u16vec2 subgroupInclusiveOr(u16vec2);\n"
- "u16vec3 subgroupInclusiveOr(u16vec3);\n"
- "u16vec4 subgroupInclusiveOr(u16vec4);\n"
-
- "int16_t subgroupInclusiveXor(int16_t);\n"
- "i16vec2 subgroupInclusiveXor(i16vec2);\n"
- "i16vec3 subgroupInclusiveXor(i16vec3);\n"
- "i16vec4 subgroupInclusiveXor(i16vec4);\n"
- "uint16_t subgroupInclusiveXor(uint16_t);\n"
- "u16vec2 subgroupInclusiveXor(u16vec2);\n"
- "u16vec3 subgroupInclusiveXor(u16vec3);\n"
- "u16vec4 subgroupInclusiveXor(u16vec4);\n"
-
- "int16_t subgroupExclusiveAdd(int16_t);\n"
- "i16vec2 subgroupExclusiveAdd(i16vec2);\n"
- "i16vec3 subgroupExclusiveAdd(i16vec3);\n"
- "i16vec4 subgroupExclusiveAdd(i16vec4);\n"
- "uint16_t subgroupExclusiveAdd(uint16_t);\n"
- "u16vec2 subgroupExclusiveAdd(u16vec2);\n"
- "u16vec3 subgroupExclusiveAdd(u16vec3);\n"
- "u16vec4 subgroupExclusiveAdd(u16vec4);\n"
-
- "int16_t subgroupExclusiveMul(int16_t);\n"
- "i16vec2 subgroupExclusiveMul(i16vec2);\n"
- "i16vec3 subgroupExclusiveMul(i16vec3);\n"
- "i16vec4 subgroupExclusiveMul(i16vec4);\n"
- "uint16_t subgroupExclusiveMul(uint16_t);\n"
- "u16vec2 subgroupExclusiveMul(u16vec2);\n"
- "u16vec3 subgroupExclusiveMul(u16vec3);\n"
- "u16vec4 subgroupExclusiveMul(u16vec4);\n"
-
- "int16_t subgroupExclusiveMin(int16_t);\n"
- "i16vec2 subgroupExclusiveMin(i16vec2);\n"
- "i16vec3 subgroupExclusiveMin(i16vec3);\n"
- "i16vec4 subgroupExclusiveMin(i16vec4);\n"
- "uint16_t subgroupExclusiveMin(uint16_t);\n"
- "u16vec2 subgroupExclusiveMin(u16vec2);\n"
- "u16vec3 subgroupExclusiveMin(u16vec3);\n"
- "u16vec4 subgroupExclusiveMin(u16vec4);\n"
-
- "int16_t subgroupExclusiveMax(int16_t);\n"
- "i16vec2 subgroupExclusiveMax(i16vec2);\n"
- "i16vec3 subgroupExclusiveMax(i16vec3);\n"
- "i16vec4 subgroupExclusiveMax(i16vec4);\n"
- "uint16_t subgroupExclusiveMax(uint16_t);\n"
- "u16vec2 subgroupExclusiveMax(u16vec2);\n"
- "u16vec3 subgroupExclusiveMax(u16vec3);\n"
- "u16vec4 subgroupExclusiveMax(u16vec4);\n"
-
- "int16_t subgroupExclusiveAnd(int16_t);\n"
- "i16vec2 subgroupExclusiveAnd(i16vec2);\n"
- "i16vec3 subgroupExclusiveAnd(i16vec3);\n"
- "i16vec4 subgroupExclusiveAnd(i16vec4);\n"
- "uint16_t subgroupExclusiveAnd(uint16_t);\n"
- "u16vec2 subgroupExclusiveAnd(u16vec2);\n"
- "u16vec3 subgroupExclusiveAnd(u16vec3);\n"
- "u16vec4 subgroupExclusiveAnd(u16vec4);\n"
-
- "int16_t subgroupExclusiveOr(int16_t);\n"
- "i16vec2 subgroupExclusiveOr(i16vec2);\n"
- "i16vec3 subgroupExclusiveOr(i16vec3);\n"
- "i16vec4 subgroupExclusiveOr(i16vec4);\n"
- "uint16_t subgroupExclusiveOr(uint16_t);\n"
- "u16vec2 subgroupExclusiveOr(u16vec2);\n"
- "u16vec3 subgroupExclusiveOr(u16vec3);\n"
- "u16vec4 subgroupExclusiveOr(u16vec4);\n"
-
- "int16_t subgroupExclusiveXor(int16_t);\n"
- "i16vec2 subgroupExclusiveXor(i16vec2);\n"
- "i16vec3 subgroupExclusiveXor(i16vec3);\n"
- "i16vec4 subgroupExclusiveXor(i16vec4);\n"
- "uint16_t subgroupExclusiveXor(uint16_t);\n"
- "u16vec2 subgroupExclusiveXor(u16vec2);\n"
- "u16vec3 subgroupExclusiveXor(u16vec3);\n"
- "u16vec4 subgroupExclusiveXor(u16vec4);\n"
-
- "int16_t subgroupClusteredAdd(int16_t, uint);\n"
- "i16vec2 subgroupClusteredAdd(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredAdd(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredAdd(i16vec4, uint);\n"
- "uint16_t subgroupClusteredAdd(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredAdd(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredAdd(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredAdd(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredMul(int16_t, uint);\n"
- "i16vec2 subgroupClusteredMul(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredMul(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredMul(i16vec4, uint);\n"
- "uint16_t subgroupClusteredMul(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredMul(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredMul(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredMul(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredMin(int16_t, uint);\n"
- "i16vec2 subgroupClusteredMin(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredMin(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredMin(i16vec4, uint);\n"
- "uint16_t subgroupClusteredMin(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredMin(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredMin(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredMin(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredMax(int16_t, uint);\n"
- "i16vec2 subgroupClusteredMax(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredMax(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredMax(i16vec4, uint);\n"
- "uint16_t subgroupClusteredMax(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredMax(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredMax(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredMax(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredAnd(int16_t, uint);\n"
- "i16vec2 subgroupClusteredAnd(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredAnd(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredAnd(i16vec4, uint);\n"
- "uint16_t subgroupClusteredAnd(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredAnd(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredAnd(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredAnd(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredOr(int16_t, uint);\n"
- "i16vec2 subgroupClusteredOr(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredOr(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredOr(i16vec4, uint);\n"
- "uint16_t subgroupClusteredOr(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredOr(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredOr(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredOr(u16vec4, uint);\n"
-
- "int16_t subgroupClusteredXor(int16_t, uint);\n"
- "i16vec2 subgroupClusteredXor(i16vec2, uint);\n"
- "i16vec3 subgroupClusteredXor(i16vec3, uint);\n"
- "i16vec4 subgroupClusteredXor(i16vec4, uint);\n"
- "uint16_t subgroupClusteredXor(uint16_t, uint);\n"
- "u16vec2 subgroupClusteredXor(u16vec2, uint);\n"
- "u16vec3 subgroupClusteredXor(u16vec3, uint);\n"
- "u16vec4 subgroupClusteredXor(u16vec4, uint);\n"
-
- "int16_t subgroupQuadBroadcast(int16_t, uint);\n"
- "i16vec2 subgroupQuadBroadcast(i16vec2, uint);\n"
- "i16vec3 subgroupQuadBroadcast(i16vec3, uint);\n"
- "i16vec4 subgroupQuadBroadcast(i16vec4, uint);\n"
- "uint16_t subgroupQuadBroadcast(uint16_t, uint);\n"
- "u16vec2 subgroupQuadBroadcast(u16vec2, uint);\n"
- "u16vec3 subgroupQuadBroadcast(u16vec3, uint);\n"
- "u16vec4 subgroupQuadBroadcast(u16vec4, uint);\n"
-
- "int16_t subgroupQuadSwapHorizontal(int16_t);\n"
- "i16vec2 subgroupQuadSwapHorizontal(i16vec2);\n"
- "i16vec3 subgroupQuadSwapHorizontal(i16vec3);\n"
- "i16vec4 subgroupQuadSwapHorizontal(i16vec4);\n"
- "uint16_t subgroupQuadSwapHorizontal(uint16_t);\n"
- "u16vec2 subgroupQuadSwapHorizontal(u16vec2);\n"
- "u16vec3 subgroupQuadSwapHorizontal(u16vec3);\n"
- "u16vec4 subgroupQuadSwapHorizontal(u16vec4);\n"
-
- "int16_t subgroupQuadSwapVertical(int16_t);\n"
- "i16vec2 subgroupQuadSwapVertical(i16vec2);\n"
- "i16vec3 subgroupQuadSwapVertical(i16vec3);\n"
- "i16vec4 subgroupQuadSwapVertical(i16vec4);\n"
- "uint16_t subgroupQuadSwapVertical(uint16_t);\n"
- "u16vec2 subgroupQuadSwapVertical(u16vec2);\n"
- "u16vec3 subgroupQuadSwapVertical(u16vec3);\n"
- "u16vec4 subgroupQuadSwapVertical(u16vec4);\n"
-
- "int16_t subgroupQuadSwapDiagonal(int16_t);\n"
- "i16vec2 subgroupQuadSwapDiagonal(i16vec2);\n"
- "i16vec3 subgroupQuadSwapDiagonal(i16vec3);\n"
- "i16vec4 subgroupQuadSwapDiagonal(i16vec4);\n"
- "uint16_t subgroupQuadSwapDiagonal(uint16_t);\n"
- "u16vec2 subgroupQuadSwapDiagonal(u16vec2);\n"
- "u16vec3 subgroupQuadSwapDiagonal(u16vec3);\n"
- "u16vec4 subgroupQuadSwapDiagonal(u16vec4);\n"
-
- "bool subgroupAllEqual(int64_t);\n"
- "bool subgroupAllEqual(i64vec2);\n"
- "bool subgroupAllEqual(i64vec3);\n"
- "bool subgroupAllEqual(i64vec4);\n"
- "bool subgroupAllEqual(uint64_t);\n"
- "bool subgroupAllEqual(u64vec2);\n"
- "bool subgroupAllEqual(u64vec3);\n"
- "bool subgroupAllEqual(u64vec4);\n"
-
- "int64_t subgroupBroadcast(int64_t, uint);\n"
- "i64vec2 subgroupBroadcast(i64vec2, uint);\n"
- "i64vec3 subgroupBroadcast(i64vec3, uint);\n"
- "i64vec4 subgroupBroadcast(i64vec4, uint);\n"
- "uint64_t subgroupBroadcast(uint64_t, uint);\n"
- "u64vec2 subgroupBroadcast(u64vec2, uint);\n"
- "u64vec3 subgroupBroadcast(u64vec3, uint);\n"
- "u64vec4 subgroupBroadcast(u64vec4, uint);\n"
-
- "int64_t subgroupBroadcastFirst(int64_t);\n"
- "i64vec2 subgroupBroadcastFirst(i64vec2);\n"
- "i64vec3 subgroupBroadcastFirst(i64vec3);\n"
- "i64vec4 subgroupBroadcastFirst(i64vec4);\n"
- "uint64_t subgroupBroadcastFirst(uint64_t);\n"
- "u64vec2 subgroupBroadcastFirst(u64vec2);\n"
- "u64vec3 subgroupBroadcastFirst(u64vec3);\n"
- "u64vec4 subgroupBroadcastFirst(u64vec4);\n"
-
- "int64_t subgroupShuffle(int64_t, uint);\n"
- "i64vec2 subgroupShuffle(i64vec2, uint);\n"
- "i64vec3 subgroupShuffle(i64vec3, uint);\n"
- "i64vec4 subgroupShuffle(i64vec4, uint);\n"
- "uint64_t subgroupShuffle(uint64_t, uint);\n"
- "u64vec2 subgroupShuffle(u64vec2, uint);\n"
- "u64vec3 subgroupShuffle(u64vec3, uint);\n"
- "u64vec4 subgroupShuffle(u64vec4, uint);\n"
-
- "int64_t subgroupShuffleXor(int64_t, uint);\n"
- "i64vec2 subgroupShuffleXor(i64vec2, uint);\n"
- "i64vec3 subgroupShuffleXor(i64vec3, uint);\n"
- "i64vec4 subgroupShuffleXor(i64vec4, uint);\n"
- "uint64_t subgroupShuffleXor(uint64_t, uint);\n"
- "u64vec2 subgroupShuffleXor(u64vec2, uint);\n"
- "u64vec3 subgroupShuffleXor(u64vec3, uint);\n"
- "u64vec4 subgroupShuffleXor(u64vec4, uint);\n"
-
- "int64_t subgroupShuffleUp(int64_t, uint delta);\n"
- "i64vec2 subgroupShuffleUp(i64vec2, uint delta);\n"
- "i64vec3 subgroupShuffleUp(i64vec3, uint delta);\n"
- "i64vec4 subgroupShuffleUp(i64vec4, uint delta);\n"
- "uint64_t subgroupShuffleUp(uint64_t, uint delta);\n"
- "u64vec2 subgroupShuffleUp(u64vec2, uint delta);\n"
- "u64vec3 subgroupShuffleUp(u64vec3, uint delta);\n"
- "u64vec4 subgroupShuffleUp(u64vec4, uint delta);\n"
-
- "int64_t subgroupShuffleDown(int64_t, uint delta);\n"
- "i64vec2 subgroupShuffleDown(i64vec2, uint delta);\n"
- "i64vec3 subgroupShuffleDown(i64vec3, uint delta);\n"
- "i64vec4 subgroupShuffleDown(i64vec4, uint delta);\n"
- "uint64_t subgroupShuffleDown(uint64_t, uint delta);\n"
- "u64vec2 subgroupShuffleDown(u64vec2, uint delta);\n"
- "u64vec3 subgroupShuffleDown(u64vec3, uint delta);\n"
- "u64vec4 subgroupShuffleDown(u64vec4, uint delta);\n"
-
- "int64_t subgroupAdd(int64_t);\n"
- "i64vec2 subgroupAdd(i64vec2);\n"
- "i64vec3 subgroupAdd(i64vec3);\n"
- "i64vec4 subgroupAdd(i64vec4);\n"
- "uint64_t subgroupAdd(uint64_t);\n"
- "u64vec2 subgroupAdd(u64vec2);\n"
- "u64vec3 subgroupAdd(u64vec3);\n"
- "u64vec4 subgroupAdd(u64vec4);\n"
-
- "int64_t subgroupMul(int64_t);\n"
- "i64vec2 subgroupMul(i64vec2);\n"
- "i64vec3 subgroupMul(i64vec3);\n"
- "i64vec4 subgroupMul(i64vec4);\n"
- "uint64_t subgroupMul(uint64_t);\n"
- "u64vec2 subgroupMul(u64vec2);\n"
- "u64vec3 subgroupMul(u64vec3);\n"
- "u64vec4 subgroupMul(u64vec4);\n"
-
- "int64_t subgroupMin(int64_t);\n"
- "i64vec2 subgroupMin(i64vec2);\n"
- "i64vec3 subgroupMin(i64vec3);\n"
- "i64vec4 subgroupMin(i64vec4);\n"
- "uint64_t subgroupMin(uint64_t);\n"
- "u64vec2 subgroupMin(u64vec2);\n"
- "u64vec3 subgroupMin(u64vec3);\n"
- "u64vec4 subgroupMin(u64vec4);\n"
-
- "int64_t subgroupMax(int64_t);\n"
- "i64vec2 subgroupMax(i64vec2);\n"
- "i64vec3 subgroupMax(i64vec3);\n"
- "i64vec4 subgroupMax(i64vec4);\n"
- "uint64_t subgroupMax(uint64_t);\n"
- "u64vec2 subgroupMax(u64vec2);\n"
- "u64vec3 subgroupMax(u64vec3);\n"
- "u64vec4 subgroupMax(u64vec4);\n"
-
- "int64_t subgroupAnd(int64_t);\n"
- "i64vec2 subgroupAnd(i64vec2);\n"
- "i64vec3 subgroupAnd(i64vec3);\n"
- "i64vec4 subgroupAnd(i64vec4);\n"
- "uint64_t subgroupAnd(uint64_t);\n"
- "u64vec2 subgroupAnd(u64vec2);\n"
- "u64vec3 subgroupAnd(u64vec3);\n"
- "u64vec4 subgroupAnd(u64vec4);\n"
-
- "int64_t subgroupOr(int64_t);\n"
- "i64vec2 subgroupOr(i64vec2);\n"
- "i64vec3 subgroupOr(i64vec3);\n"
- "i64vec4 subgroupOr(i64vec4);\n"
- "uint64_t subgroupOr(uint64_t);\n"
- "u64vec2 subgroupOr(u64vec2);\n"
- "u64vec3 subgroupOr(u64vec3);\n"
- "u64vec4 subgroupOr(u64vec4);\n"
-
- "int64_t subgroupXor(int64_t);\n"
- "i64vec2 subgroupXor(i64vec2);\n"
- "i64vec3 subgroupXor(i64vec3);\n"
- "i64vec4 subgroupXor(i64vec4);\n"
- "uint64_t subgroupXor(uint64_t);\n"
- "u64vec2 subgroupXor(u64vec2);\n"
- "u64vec3 subgroupXor(u64vec3);\n"
- "u64vec4 subgroupXor(u64vec4);\n"
-
- "int64_t subgroupInclusiveAdd(int64_t);\n"
- "i64vec2 subgroupInclusiveAdd(i64vec2);\n"
- "i64vec3 subgroupInclusiveAdd(i64vec3);\n"
- "i64vec4 subgroupInclusiveAdd(i64vec4);\n"
- "uint64_t subgroupInclusiveAdd(uint64_t);\n"
- "u64vec2 subgroupInclusiveAdd(u64vec2);\n"
- "u64vec3 subgroupInclusiveAdd(u64vec3);\n"
- "u64vec4 subgroupInclusiveAdd(u64vec4);\n"
-
- "int64_t subgroupInclusiveMul(int64_t);\n"
- "i64vec2 subgroupInclusiveMul(i64vec2);\n"
- "i64vec3 subgroupInclusiveMul(i64vec3);\n"
- "i64vec4 subgroupInclusiveMul(i64vec4);\n"
- "uint64_t subgroupInclusiveMul(uint64_t);\n"
- "u64vec2 subgroupInclusiveMul(u64vec2);\n"
- "u64vec3 subgroupInclusiveMul(u64vec3);\n"
- "u64vec4 subgroupInclusiveMul(u64vec4);\n"
-
- "int64_t subgroupInclusiveMin(int64_t);\n"
- "i64vec2 subgroupInclusiveMin(i64vec2);\n"
- "i64vec3 subgroupInclusiveMin(i64vec3);\n"
- "i64vec4 subgroupInclusiveMin(i64vec4);\n"
- "uint64_t subgroupInclusiveMin(uint64_t);\n"
- "u64vec2 subgroupInclusiveMin(u64vec2);\n"
- "u64vec3 subgroupInclusiveMin(u64vec3);\n"
- "u64vec4 subgroupInclusiveMin(u64vec4);\n"
-
- "int64_t subgroupInclusiveMax(int64_t);\n"
- "i64vec2 subgroupInclusiveMax(i64vec2);\n"
- "i64vec3 subgroupInclusiveMax(i64vec3);\n"
- "i64vec4 subgroupInclusiveMax(i64vec4);\n"
- "uint64_t subgroupInclusiveMax(uint64_t);\n"
- "u64vec2 subgroupInclusiveMax(u64vec2);\n"
- "u64vec3 subgroupInclusiveMax(u64vec3);\n"
- "u64vec4 subgroupInclusiveMax(u64vec4);\n"
-
- "int64_t subgroupInclusiveAnd(int64_t);\n"
- "i64vec2 subgroupInclusiveAnd(i64vec2);\n"
- "i64vec3 subgroupInclusiveAnd(i64vec3);\n"
- "i64vec4 subgroupInclusiveAnd(i64vec4);\n"
- "uint64_t subgroupInclusiveAnd(uint64_t);\n"
- "u64vec2 subgroupInclusiveAnd(u64vec2);\n"
- "u64vec3 subgroupInclusiveAnd(u64vec3);\n"
- "u64vec4 subgroupInclusiveAnd(u64vec4);\n"
-
- "int64_t subgroupInclusiveOr(int64_t);\n"
- "i64vec2 subgroupInclusiveOr(i64vec2);\n"
- "i64vec3 subgroupInclusiveOr(i64vec3);\n"
- "i64vec4 subgroupInclusiveOr(i64vec4);\n"
- "uint64_t subgroupInclusiveOr(uint64_t);\n"
- "u64vec2 subgroupInclusiveOr(u64vec2);\n"
- "u64vec3 subgroupInclusiveOr(u64vec3);\n"
- "u64vec4 subgroupInclusiveOr(u64vec4);\n"
-
- "int64_t subgroupInclusiveXor(int64_t);\n"
- "i64vec2 subgroupInclusiveXor(i64vec2);\n"
- "i64vec3 subgroupInclusiveXor(i64vec3);\n"
- "i64vec4 subgroupInclusiveXor(i64vec4);\n"
- "uint64_t subgroupInclusiveXor(uint64_t);\n"
- "u64vec2 subgroupInclusiveXor(u64vec2);\n"
- "u64vec3 subgroupInclusiveXor(u64vec3);\n"
- "u64vec4 subgroupInclusiveXor(u64vec4);\n"
-
- "int64_t subgroupExclusiveAdd(int64_t);\n"
- "i64vec2 subgroupExclusiveAdd(i64vec2);\n"
- "i64vec3 subgroupExclusiveAdd(i64vec3);\n"
- "i64vec4 subgroupExclusiveAdd(i64vec4);\n"
- "uint64_t subgroupExclusiveAdd(uint64_t);\n"
- "u64vec2 subgroupExclusiveAdd(u64vec2);\n"
- "u64vec3 subgroupExclusiveAdd(u64vec3);\n"
- "u64vec4 subgroupExclusiveAdd(u64vec4);\n"
-
- "int64_t subgroupExclusiveMul(int64_t);\n"
- "i64vec2 subgroupExclusiveMul(i64vec2);\n"
- "i64vec3 subgroupExclusiveMul(i64vec3);\n"
- "i64vec4 subgroupExclusiveMul(i64vec4);\n"
- "uint64_t subgroupExclusiveMul(uint64_t);\n"
- "u64vec2 subgroupExclusiveMul(u64vec2);\n"
- "u64vec3 subgroupExclusiveMul(u64vec3);\n"
- "u64vec4 subgroupExclusiveMul(u64vec4);\n"
-
- "int64_t subgroupExclusiveMin(int64_t);\n"
- "i64vec2 subgroupExclusiveMin(i64vec2);\n"
- "i64vec3 subgroupExclusiveMin(i64vec3);\n"
- "i64vec4 subgroupExclusiveMin(i64vec4);\n"
- "uint64_t subgroupExclusiveMin(uint64_t);\n"
- "u64vec2 subgroupExclusiveMin(u64vec2);\n"
- "u64vec3 subgroupExclusiveMin(u64vec3);\n"
- "u64vec4 subgroupExclusiveMin(u64vec4);\n"
-
- "int64_t subgroupExclusiveMax(int64_t);\n"
- "i64vec2 subgroupExclusiveMax(i64vec2);\n"
- "i64vec3 subgroupExclusiveMax(i64vec3);\n"
- "i64vec4 subgroupExclusiveMax(i64vec4);\n"
- "uint64_t subgroupExclusiveMax(uint64_t);\n"
- "u64vec2 subgroupExclusiveMax(u64vec2);\n"
- "u64vec3 subgroupExclusiveMax(u64vec3);\n"
- "u64vec4 subgroupExclusiveMax(u64vec4);\n"
-
- "int64_t subgroupExclusiveAnd(int64_t);\n"
- "i64vec2 subgroupExclusiveAnd(i64vec2);\n"
- "i64vec3 subgroupExclusiveAnd(i64vec3);\n"
- "i64vec4 subgroupExclusiveAnd(i64vec4);\n"
- "uint64_t subgroupExclusiveAnd(uint64_t);\n"
- "u64vec2 subgroupExclusiveAnd(u64vec2);\n"
- "u64vec3 subgroupExclusiveAnd(u64vec3);\n"
- "u64vec4 subgroupExclusiveAnd(u64vec4);\n"
-
- "int64_t subgroupExclusiveOr(int64_t);\n"
- "i64vec2 subgroupExclusiveOr(i64vec2);\n"
- "i64vec3 subgroupExclusiveOr(i64vec3);\n"
- "i64vec4 subgroupExclusiveOr(i64vec4);\n"
- "uint64_t subgroupExclusiveOr(uint64_t);\n"
- "u64vec2 subgroupExclusiveOr(u64vec2);\n"
- "u64vec3 subgroupExclusiveOr(u64vec3);\n"
- "u64vec4 subgroupExclusiveOr(u64vec4);\n"
-
- "int64_t subgroupExclusiveXor(int64_t);\n"
- "i64vec2 subgroupExclusiveXor(i64vec2);\n"
- "i64vec3 subgroupExclusiveXor(i64vec3);\n"
- "i64vec4 subgroupExclusiveXor(i64vec4);\n"
- "uint64_t subgroupExclusiveXor(uint64_t);\n"
- "u64vec2 subgroupExclusiveXor(u64vec2);\n"
- "u64vec3 subgroupExclusiveXor(u64vec3);\n"
- "u64vec4 subgroupExclusiveXor(u64vec4);\n"
-
- "int64_t subgroupClusteredAdd(int64_t, uint);\n"
- "i64vec2 subgroupClusteredAdd(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredAdd(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredAdd(i64vec4, uint);\n"
- "uint64_t subgroupClusteredAdd(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredAdd(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredAdd(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredAdd(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredMul(int64_t, uint);\n"
- "i64vec2 subgroupClusteredMul(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredMul(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredMul(i64vec4, uint);\n"
- "uint64_t subgroupClusteredMul(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredMul(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredMul(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredMul(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredMin(int64_t, uint);\n"
- "i64vec2 subgroupClusteredMin(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredMin(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredMin(i64vec4, uint);\n"
- "uint64_t subgroupClusteredMin(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredMin(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredMin(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredMin(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredMax(int64_t, uint);\n"
- "i64vec2 subgroupClusteredMax(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredMax(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredMax(i64vec4, uint);\n"
- "uint64_t subgroupClusteredMax(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredMax(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredMax(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredMax(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredAnd(int64_t, uint);\n"
- "i64vec2 subgroupClusteredAnd(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredAnd(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredAnd(i64vec4, uint);\n"
- "uint64_t subgroupClusteredAnd(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredAnd(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredAnd(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredAnd(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredOr(int64_t, uint);\n"
- "i64vec2 subgroupClusteredOr(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredOr(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredOr(i64vec4, uint);\n"
- "uint64_t subgroupClusteredOr(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredOr(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredOr(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredOr(u64vec4, uint);\n"
-
- "int64_t subgroupClusteredXor(int64_t, uint);\n"
- "i64vec2 subgroupClusteredXor(i64vec2, uint);\n"
- "i64vec3 subgroupClusteredXor(i64vec3, uint);\n"
- "i64vec4 subgroupClusteredXor(i64vec4, uint);\n"
- "uint64_t subgroupClusteredXor(uint64_t, uint);\n"
- "u64vec2 subgroupClusteredXor(u64vec2, uint);\n"
- "u64vec3 subgroupClusteredXor(u64vec3, uint);\n"
- "u64vec4 subgroupClusteredXor(u64vec4, uint);\n"
-
- "int64_t subgroupQuadBroadcast(int64_t, uint);\n"
- "i64vec2 subgroupQuadBroadcast(i64vec2, uint);\n"
- "i64vec3 subgroupQuadBroadcast(i64vec3, uint);\n"
- "i64vec4 subgroupQuadBroadcast(i64vec4, uint);\n"
- "uint64_t subgroupQuadBroadcast(uint64_t, uint);\n"
- "u64vec2 subgroupQuadBroadcast(u64vec2, uint);\n"
- "u64vec3 subgroupQuadBroadcast(u64vec3, uint);\n"
- "u64vec4 subgroupQuadBroadcast(u64vec4, uint);\n"
-
- "int64_t subgroupQuadSwapHorizontal(int64_t);\n"
- "i64vec2 subgroupQuadSwapHorizontal(i64vec2);\n"
- "i64vec3 subgroupQuadSwapHorizontal(i64vec3);\n"
- "i64vec4 subgroupQuadSwapHorizontal(i64vec4);\n"
- "uint64_t subgroupQuadSwapHorizontal(uint64_t);\n"
- "u64vec2 subgroupQuadSwapHorizontal(u64vec2);\n"
- "u64vec3 subgroupQuadSwapHorizontal(u64vec3);\n"
- "u64vec4 subgroupQuadSwapHorizontal(u64vec4);\n"
-
- "int64_t subgroupQuadSwapVertical(int64_t);\n"
- "i64vec2 subgroupQuadSwapVertical(i64vec2);\n"
- "i64vec3 subgroupQuadSwapVertical(i64vec3);\n"
- "i64vec4 subgroupQuadSwapVertical(i64vec4);\n"
- "uint64_t subgroupQuadSwapVertical(uint64_t);\n"
- "u64vec2 subgroupQuadSwapVertical(u64vec2);\n"
- "u64vec3 subgroupQuadSwapVertical(u64vec3);\n"
- "u64vec4 subgroupQuadSwapVertical(u64vec4);\n"
-
- "int64_t subgroupQuadSwapDiagonal(int64_t);\n"
- "i64vec2 subgroupQuadSwapDiagonal(i64vec2);\n"
- "i64vec3 subgroupQuadSwapDiagonal(i64vec3);\n"
- "i64vec4 subgroupQuadSwapDiagonal(i64vec4);\n"
- "uint64_t subgroupQuadSwapDiagonal(uint64_t);\n"
- "u64vec2 subgroupQuadSwapDiagonal(u64vec2);\n"
- "u64vec3 subgroupQuadSwapDiagonal(u64vec3);\n"
- "u64vec4 subgroupQuadSwapDiagonal(u64vec4);\n"
-
- "bool subgroupAllEqual(float16_t);\n"
- "bool subgroupAllEqual(f16vec2);\n"
- "bool subgroupAllEqual(f16vec3);\n"
- "bool subgroupAllEqual(f16vec4);\n"
-
- "float16_t subgroupBroadcast(float16_t, uint);\n"
- "f16vec2 subgroupBroadcast(f16vec2, uint);\n"
- "f16vec3 subgroupBroadcast(f16vec3, uint);\n"
- "f16vec4 subgroupBroadcast(f16vec4, uint);\n"
-
- "float16_t subgroupBroadcastFirst(float16_t);\n"
- "f16vec2 subgroupBroadcastFirst(f16vec2);\n"
- "f16vec3 subgroupBroadcastFirst(f16vec3);\n"
- "f16vec4 subgroupBroadcastFirst(f16vec4);\n"
-
- "float16_t subgroupShuffle(float16_t, uint);\n"
- "f16vec2 subgroupShuffle(f16vec2, uint);\n"
- "f16vec3 subgroupShuffle(f16vec3, uint);\n"
- "f16vec4 subgroupShuffle(f16vec4, uint);\n"
-
- "float16_t subgroupShuffleXor(float16_t, uint);\n"
- "f16vec2 subgroupShuffleXor(f16vec2, uint);\n"
- "f16vec3 subgroupShuffleXor(f16vec3, uint);\n"
- "f16vec4 subgroupShuffleXor(f16vec4, uint);\n"
-
- "float16_t subgroupShuffleUp(float16_t, uint delta);\n"
- "f16vec2 subgroupShuffleUp(f16vec2, uint delta);\n"
- "f16vec3 subgroupShuffleUp(f16vec3, uint delta);\n"
- "f16vec4 subgroupShuffleUp(f16vec4, uint delta);\n"
-
- "float16_t subgroupShuffleDown(float16_t, uint delta);\n"
- "f16vec2 subgroupShuffleDown(f16vec2, uint delta);\n"
- "f16vec3 subgroupShuffleDown(f16vec3, uint delta);\n"
- "f16vec4 subgroupShuffleDown(f16vec4, uint delta);\n"
-
- "float16_t subgroupAdd(float16_t);\n"
- "f16vec2 subgroupAdd(f16vec2);\n"
- "f16vec3 subgroupAdd(f16vec3);\n"
- "f16vec4 subgroupAdd(f16vec4);\n"
-
- "float16_t subgroupMul(float16_t);\n"
- "f16vec2 subgroupMul(f16vec2);\n"
- "f16vec3 subgroupMul(f16vec3);\n"
- "f16vec4 subgroupMul(f16vec4);\n"
-
- "float16_t subgroupMin(float16_t);\n"
- "f16vec2 subgroupMin(f16vec2);\n"
- "f16vec3 subgroupMin(f16vec3);\n"
- "f16vec4 subgroupMin(f16vec4);\n"
-
- "float16_t subgroupMax(float16_t);\n"
- "f16vec2 subgroupMax(f16vec2);\n"
- "f16vec3 subgroupMax(f16vec3);\n"
- "f16vec4 subgroupMax(f16vec4);\n"
-
- "float16_t subgroupInclusiveAdd(float16_t);\n"
- "f16vec2 subgroupInclusiveAdd(f16vec2);\n"
- "f16vec3 subgroupInclusiveAdd(f16vec3);\n"
- "f16vec4 subgroupInclusiveAdd(f16vec4);\n"
-
- "float16_t subgroupInclusiveMul(float16_t);\n"
- "f16vec2 subgroupInclusiveMul(f16vec2);\n"
- "f16vec3 subgroupInclusiveMul(f16vec3);\n"
- "f16vec4 subgroupInclusiveMul(f16vec4);\n"
-
- "float16_t subgroupInclusiveMin(float16_t);\n"
- "f16vec2 subgroupInclusiveMin(f16vec2);\n"
- "f16vec3 subgroupInclusiveMin(f16vec3);\n"
- "f16vec4 subgroupInclusiveMin(f16vec4);\n"
-
- "float16_t subgroupInclusiveMax(float16_t);\n"
- "f16vec2 subgroupInclusiveMax(f16vec2);\n"
- "f16vec3 subgroupInclusiveMax(f16vec3);\n"
- "f16vec4 subgroupInclusiveMax(f16vec4);\n"
-
- "float16_t subgroupExclusiveAdd(float16_t);\n"
- "f16vec2 subgroupExclusiveAdd(f16vec2);\n"
- "f16vec3 subgroupExclusiveAdd(f16vec3);\n"
- "f16vec4 subgroupExclusiveAdd(f16vec4);\n"
-
- "float16_t subgroupExclusiveMul(float16_t);\n"
- "f16vec2 subgroupExclusiveMul(f16vec2);\n"
- "f16vec3 subgroupExclusiveMul(f16vec3);\n"
- "f16vec4 subgroupExclusiveMul(f16vec4);\n"
-
- "float16_t subgroupExclusiveMin(float16_t);\n"
- "f16vec2 subgroupExclusiveMin(f16vec2);\n"
- "f16vec3 subgroupExclusiveMin(f16vec3);\n"
- "f16vec4 subgroupExclusiveMin(f16vec4);\n"
-
- "float16_t subgroupExclusiveMax(float16_t);\n"
- "f16vec2 subgroupExclusiveMax(f16vec2);\n"
- "f16vec3 subgroupExclusiveMax(f16vec3);\n"
- "f16vec4 subgroupExclusiveMax(f16vec4);\n"
-
- "float16_t subgroupClusteredAdd(float16_t, uint);\n"
- "f16vec2 subgroupClusteredAdd(f16vec2, uint);\n"
- "f16vec3 subgroupClusteredAdd(f16vec3, uint);\n"
- "f16vec4 subgroupClusteredAdd(f16vec4, uint);\n"
-
- "float16_t subgroupClusteredMul(float16_t, uint);\n"
- "f16vec2 subgroupClusteredMul(f16vec2, uint);\n"
- "f16vec3 subgroupClusteredMul(f16vec3, uint);\n"
- "f16vec4 subgroupClusteredMul(f16vec4, uint);\n"
-
- "float16_t subgroupClusteredMin(float16_t, uint);\n"
- "f16vec2 subgroupClusteredMin(f16vec2, uint);\n"
- "f16vec3 subgroupClusteredMin(f16vec3, uint);\n"
- "f16vec4 subgroupClusteredMin(f16vec4, uint);\n"
-
- "float16_t subgroupClusteredMax(float16_t, uint);\n"
- "f16vec2 subgroupClusteredMax(f16vec2, uint);\n"
- "f16vec3 subgroupClusteredMax(f16vec3, uint);\n"
- "f16vec4 subgroupClusteredMax(f16vec4, uint);\n"
-
- "float16_t subgroupQuadBroadcast(float16_t, uint);\n"
- "f16vec2 subgroupQuadBroadcast(f16vec2, uint);\n"
- "f16vec3 subgroupQuadBroadcast(f16vec3, uint);\n"
- "f16vec4 subgroupQuadBroadcast(f16vec4, uint);\n"
-
- "float16_t subgroupQuadSwapHorizontal(float16_t);\n"
- "f16vec2 subgroupQuadSwapHorizontal(f16vec2);\n"
- "f16vec3 subgroupQuadSwapHorizontal(f16vec3);\n"
- "f16vec4 subgroupQuadSwapHorizontal(f16vec4);\n"
-
- "float16_t subgroupQuadSwapVertical(float16_t);\n"
- "f16vec2 subgroupQuadSwapVertical(f16vec2);\n"
- "f16vec3 subgroupQuadSwapVertical(f16vec3);\n"
- "f16vec4 subgroupQuadSwapVertical(f16vec4);\n"
-
- "float16_t subgroupQuadSwapDiagonal(float16_t);\n"
- "f16vec2 subgroupQuadSwapDiagonal(f16vec2);\n"
- "f16vec3 subgroupQuadSwapDiagonal(f16vec3);\n"
- "f16vec4 subgroupQuadSwapDiagonal(f16vec4);\n"
-
+ // Generate all flavors of subgroup ops.
+ static const char *subgroupOps[] =
+ {
+ "bool subgroupAllEqual(%s);\n",
+ "%s subgroupBroadcast(%s, uint);\n",
+ "%s subgroupBroadcastFirst(%s);\n",
+ "%s subgroupShuffle(%s, uint);\n",
+ "%s subgroupShuffleXor(%s, uint);\n",
+ "%s subgroupShuffleUp(%s, uint delta);\n",
+ "%s subgroupShuffleDown(%s, uint delta);\n",
+ "%s subgroupAdd(%s);\n",
+ "%s subgroupMul(%s);\n",
+ "%s subgroupMin(%s);\n",
+ "%s subgroupMax(%s);\n",
+ "%s subgroupAnd(%s);\n",
+ "%s subgroupOr(%s);\n",
+ "%s subgroupXor(%s);\n",
+ "%s subgroupInclusiveAdd(%s);\n",
+ "%s subgroupInclusiveMul(%s);\n",
+ "%s subgroupInclusiveMin(%s);\n",
+ "%s subgroupInclusiveMax(%s);\n",
+ "%s subgroupInclusiveAnd(%s);\n",
+ "%s subgroupInclusiveOr(%s);\n",
+ "%s subgroupInclusiveXor(%s);\n",
+ "%s subgroupExclusiveAdd(%s);\n",
+ "%s subgroupExclusiveMul(%s);\n",
+ "%s subgroupExclusiveMin(%s);\n",
+ "%s subgroupExclusiveMax(%s);\n",
+ "%s subgroupExclusiveAnd(%s);\n",
+ "%s subgroupExclusiveOr(%s);\n",
+ "%s subgroupExclusiveXor(%s);\n",
+ "%s subgroupClusteredAdd(%s, uint);\n",
+ "%s subgroupClusteredMul(%s, uint);\n",
+ "%s subgroupClusteredMin(%s, uint);\n",
+ "%s subgroupClusteredMax(%s, uint);\n",
+ "%s subgroupClusteredAnd(%s, uint);\n",
+ "%s subgroupClusteredOr(%s, uint);\n",
+ "%s subgroupClusteredXor(%s, uint);\n",
+ "%s subgroupQuadBroadcast(%s, uint);\n",
+ "%s subgroupQuadSwapHorizontal(%s);\n",
+ "%s subgroupQuadSwapVertical(%s);\n",
+ "%s subgroupQuadSwapDiagonal(%s);\n",
#ifdef NV_EXTENSIONS
- "uvec4 subgroupPartitionNV(float);\n"
- "uvec4 subgroupPartitionNV(vec2);\n"
- "uvec4 subgroupPartitionNV(vec3);\n"
- "uvec4 subgroupPartitionNV(vec4);\n"
- "uvec4 subgroupPartitionNV(int);\n"
- "uvec4 subgroupPartitionNV(ivec2);\n"
- "uvec4 subgroupPartitionNV(ivec3);\n"
- "uvec4 subgroupPartitionNV(ivec4);\n"
- "uvec4 subgroupPartitionNV(uint);\n"
- "uvec4 subgroupPartitionNV(uvec2);\n"
- "uvec4 subgroupPartitionNV(uvec3);\n"
- "uvec4 subgroupPartitionNV(uvec4);\n"
- "uvec4 subgroupPartitionNV(bool);\n"
- "uvec4 subgroupPartitionNV(bvec2);\n"
- "uvec4 subgroupPartitionNV(bvec3);\n"
- "uvec4 subgroupPartitionNV(bvec4);\n"
-
- "float subgroupPartitionedAddNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedAddNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedAddNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedAddNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedAddNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedAddNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedAddNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedAddNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedAddNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedAddNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedAddNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedAddNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedMulNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedMulNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedMulNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedMulNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedMulNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedMulNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedMulNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedMulNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedMulNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedMulNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedMulNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedMulNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedMinNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedMinNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedMinNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedMinNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedMinNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedMinNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedMinNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedMinNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedMinNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedMinNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedMinNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedMinNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedMaxNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedMaxNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedMaxNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedMaxNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedMaxNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedMaxNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedMaxNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedMaxNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedMaxNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedMaxNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedMaxNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedMaxNV(uvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedAndNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedAndNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedAndNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedAndNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedAndNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedAndNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedAndNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedAndNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedAndNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedAndNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedAndNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedAndNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedOrNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedOrNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedOrNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedOrNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedOrNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedOrNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedOrNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedOrNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedOrNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedOrNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedOrNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedOrNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedXorNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedXorNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedXorNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedXorNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedXorNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedXorNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedXorNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedXorNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedXorNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedXorNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedXorNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedXorNV(bvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedInclusiveAddNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedInclusiveAddNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedInclusiveAddNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedInclusiveAddNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedInclusiveAddNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveAddNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveAddNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveAddNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveAddNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveAddNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveAddNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveAddNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedInclusiveMulNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedInclusiveMulNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedInclusiveMulNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedInclusiveMulNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedInclusiveMulNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveMulNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveMulNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveMulNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveMulNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveMulNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveMulNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveMulNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedInclusiveMinNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedInclusiveMinNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedInclusiveMinNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedInclusiveMinNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedInclusiveMinNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveMinNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveMinNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveMinNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveMinNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveMinNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveMinNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveMinNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedInclusiveMaxNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedInclusiveMaxNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedInclusiveMaxNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedInclusiveMaxNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedInclusiveMaxNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveMaxNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveMaxNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveMaxNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveMaxNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveMaxNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveMaxNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveMaxNV(uvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedInclusiveAndNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveAndNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveAndNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveAndNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveAndNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveAndNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveAndNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveAndNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedInclusiveAndNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedInclusiveAndNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedInclusiveAndNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedInclusiveAndNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedInclusiveOrNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveOrNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveOrNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveOrNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveOrNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveOrNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveOrNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveOrNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedInclusiveOrNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedInclusiveOrNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedInclusiveOrNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedInclusiveOrNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedInclusiveXorNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedInclusiveXorNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedInclusiveXorNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedInclusiveXorNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedInclusiveXorNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedInclusiveXorNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedInclusiveXorNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedInclusiveXorNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedInclusiveXorNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedInclusiveXorNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedInclusiveXorNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedInclusiveXorNV(bvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedExclusiveAddNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedExclusiveAddNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedExclusiveAddNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedExclusiveAddNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedExclusiveAddNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveAddNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveAddNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveAddNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveAddNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveAddNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveAddNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveAddNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedExclusiveMulNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedExclusiveMulNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedExclusiveMulNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedExclusiveMulNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedExclusiveMulNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveMulNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveMulNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveMulNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveMulNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveMulNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveMulNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveMulNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedExclusiveMinNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedExclusiveMinNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedExclusiveMinNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedExclusiveMinNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedExclusiveMinNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveMinNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveMinNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveMinNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveMinNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveMinNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveMinNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveMinNV(uvec4, uvec4 ballot);\n"
-
- "float subgroupPartitionedExclusiveMaxNV(float, uvec4 ballot);\n"
- "vec2 subgroupPartitionedExclusiveMaxNV(vec2, uvec4 ballot);\n"
- "vec3 subgroupPartitionedExclusiveMaxNV(vec3, uvec4 ballot);\n"
- "vec4 subgroupPartitionedExclusiveMaxNV(vec4, uvec4 ballot);\n"
- "int subgroupPartitionedExclusiveMaxNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveMaxNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveMaxNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveMaxNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveMaxNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveMaxNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveMaxNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveMaxNV(uvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedExclusiveAndNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveAndNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveAndNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveAndNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveAndNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveAndNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveAndNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveAndNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedExclusiveAndNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedExclusiveAndNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedExclusiveAndNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedExclusiveAndNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedExclusiveOrNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveOrNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveOrNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveOrNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveOrNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveOrNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveOrNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveOrNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedExclusiveOrNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedExclusiveOrNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedExclusiveOrNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedExclusiveOrNV(bvec4, uvec4 ballot);\n"
-
- "int subgroupPartitionedExclusiveXorNV(int, uvec4 ballot);\n"
- "ivec2 subgroupPartitionedExclusiveXorNV(ivec2, uvec4 ballot);\n"
- "ivec3 subgroupPartitionedExclusiveXorNV(ivec3, uvec4 ballot);\n"
- "ivec4 subgroupPartitionedExclusiveXorNV(ivec4, uvec4 ballot);\n"
- "uint subgroupPartitionedExclusiveXorNV(uint, uvec4 ballot);\n"
- "uvec2 subgroupPartitionedExclusiveXorNV(uvec2, uvec4 ballot);\n"
- "uvec3 subgroupPartitionedExclusiveXorNV(uvec3, uvec4 ballot);\n"
- "uvec4 subgroupPartitionedExclusiveXorNV(uvec4, uvec4 ballot);\n"
- "bool subgroupPartitionedExclusiveXorNV(bool, uvec4 ballot);\n"
- "bvec2 subgroupPartitionedExclusiveXorNV(bvec2, uvec4 ballot);\n"
- "bvec3 subgroupPartitionedExclusiveXorNV(bvec3, uvec4 ballot);\n"
- "bvec4 subgroupPartitionedExclusiveXorNV(bvec4, uvec4 ballot);\n"
+ "uvec4 subgroupPartitionNV(%s);\n",
+ "%s subgroupPartitionedAddNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedMulNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedMinNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedMaxNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedAndNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedOrNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedXorNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveAddNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveMulNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveMinNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveMaxNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveAndNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveOrNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedInclusiveXorNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveAddNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveMulNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveMinNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveMaxNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveAndNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveOrNV(%s, uvec4 ballot);\n",
+ "%s subgroupPartitionedExclusiveXorNV(%s, uvec4 ballot);\n",
#endif
+ };
- "\n");
+ static const char *floatTypes[] = {
+ "float", "vec2", "vec3", "vec4",
+ "float16_t", "f16vec2", "f16vec3", "f16vec4",
+ };
+ static const char *doubleTypes[] = {
+ "double", "dvec2", "dvec3", "dvec4",
+ };
+ static const char *intTypes[] = {
+ "int8_t", "i8vec2", "i8vec3", "i8vec4",
+ "int16_t", "i16vec2", "i16vec3", "i16vec4",
+ "int", "ivec2", "ivec3", "ivec4",
+ "int64_t", "i64vec2", "i64vec3", "i64vec4",
+ "uint8_t", "u8vec2", "u8vec3", "u8vec4",
+ "uint16_t", "u16vec2", "u16vec3", "u16vec4",
+ "uint", "uvec2", "uvec3", "uvec4",
+ "uint64_t", "u64vec2", "u64vec3", "u64vec4",
+ };
+ static const char *boolTypes[] = {
+ "bool", "bvec2", "bvec3", "bvec4",
+ };
- if (profile != EEsProfile && version >= 400) {
- commonBuiltins.append(
- "bool subgroupAllEqual(double);\n"
- "bool subgroupAllEqual(dvec2);\n"
- "bool subgroupAllEqual(dvec3);\n"
- "bool subgroupAllEqual(dvec4);\n"
+ for (unsigned int i = 0; i < sizeof(subgroupOps)/sizeof(subgroupOps[0]); ++i) {
+ const char *op = subgroupOps[i];
- "double subgroupBroadcast(double, uint);\n"
- "dvec2 subgroupBroadcast(dvec2, uint);\n"
- "dvec3 subgroupBroadcast(dvec3, uint);\n"
- "dvec4 subgroupBroadcast(dvec4, uint);\n"
+ // Logical operations don't support float
+ bool logicalOp = strstr(op, "Or") || strstr(op, "And") ||
+ (strstr(op, "Xor") && !strstr(op, "ShuffleXor"));
+ // Math operations don't support bool
+ bool mathOp = strstr(op, "Add") || strstr(op, "Mul") || strstr(op, "Min") || strstr(op, "Max");
- "double subgroupBroadcastFirst(double);\n"
- "dvec2 subgroupBroadcastFirst(dvec2);\n"
- "dvec3 subgroupBroadcastFirst(dvec3);\n"
- "dvec4 subgroupBroadcastFirst(dvec4);\n"
+ const int bufSize = 256;
+ char buf[bufSize];
- "double subgroupShuffle(double, uint);\n"
- "dvec2 subgroupShuffle(dvec2, uint);\n"
- "dvec3 subgroupShuffle(dvec3, uint);\n"
- "dvec4 subgroupShuffle(dvec4, uint);\n"
-
- "double subgroupShuffleXor(double, uint);\n"
- "dvec2 subgroupShuffleXor(dvec2, uint);\n"
- "dvec3 subgroupShuffleXor(dvec3, uint);\n"
- "dvec4 subgroupShuffleXor(dvec4, uint);\n"
-
- "double subgroupShuffleUp(double, uint delta);\n"
- "dvec2 subgroupShuffleUp(dvec2, uint delta);\n"
- "dvec3 subgroupShuffleUp(dvec3, uint delta);\n"
- "dvec4 subgroupShuffleUp(dvec4, uint delta);\n"
-
- "double subgroupShuffleDown(double, uint delta);\n"
- "dvec2 subgroupShuffleDown(dvec2, uint delta);\n"
- "dvec3 subgroupShuffleDown(dvec3, uint delta);\n"
- "dvec4 subgroupShuffleDown(dvec4, uint delta);\n"
-
- "double subgroupAdd(double);\n"
- "dvec2 subgroupAdd(dvec2);\n"
- "dvec3 subgroupAdd(dvec3);\n"
- "dvec4 subgroupAdd(dvec4);\n"
-
- "double subgroupMul(double);\n"
- "dvec2 subgroupMul(dvec2);\n"
- "dvec3 subgroupMul(dvec3);\n"
- "dvec4 subgroupMul(dvec4);\n"
-
- "double subgroupMin(double);\n"
- "dvec2 subgroupMin(dvec2);\n"
- "dvec3 subgroupMin(dvec3);\n"
- "dvec4 subgroupMin(dvec4);\n"
-
- "double subgroupMax(double);\n"
- "dvec2 subgroupMax(dvec2);\n"
- "dvec3 subgroupMax(dvec3);\n"
- "dvec4 subgroupMax(dvec4);\n"
-
- "double subgroupInclusiveAdd(double);\n"
- "dvec2 subgroupInclusiveAdd(dvec2);\n"
- "dvec3 subgroupInclusiveAdd(dvec3);\n"
- "dvec4 subgroupInclusiveAdd(dvec4);\n"
-
- "double subgroupInclusiveMul(double);\n"
- "dvec2 subgroupInclusiveMul(dvec2);\n"
- "dvec3 subgroupInclusiveMul(dvec3);\n"
- "dvec4 subgroupInclusiveMul(dvec4);\n"
-
- "double subgroupInclusiveMin(double);\n"
- "dvec2 subgroupInclusiveMin(dvec2);\n"
- "dvec3 subgroupInclusiveMin(dvec3);\n"
- "dvec4 subgroupInclusiveMin(dvec4);\n"
-
- "double subgroupInclusiveMax(double);\n"
- "dvec2 subgroupInclusiveMax(dvec2);\n"
- "dvec3 subgroupInclusiveMax(dvec3);\n"
- "dvec4 subgroupInclusiveMax(dvec4);\n"
-
- "double subgroupExclusiveAdd(double);\n"
- "dvec2 subgroupExclusiveAdd(dvec2);\n"
- "dvec3 subgroupExclusiveAdd(dvec3);\n"
- "dvec4 subgroupExclusiveAdd(dvec4);\n"
-
- "double subgroupExclusiveMul(double);\n"
- "dvec2 subgroupExclusiveMul(dvec2);\n"
- "dvec3 subgroupExclusiveMul(dvec3);\n"
- "dvec4 subgroupExclusiveMul(dvec4);\n"
-
- "double subgroupExclusiveMin(double);\n"
- "dvec2 subgroupExclusiveMin(dvec2);\n"
- "dvec3 subgroupExclusiveMin(dvec3);\n"
- "dvec4 subgroupExclusiveMin(dvec4);\n"
-
- "double subgroupExclusiveMax(double);\n"
- "dvec2 subgroupExclusiveMax(dvec2);\n"
- "dvec3 subgroupExclusiveMax(dvec3);\n"
- "dvec4 subgroupExclusiveMax(dvec4);\n"
-
- "double subgroupClusteredAdd(double, uint);\n"
- "dvec2 subgroupClusteredAdd(dvec2, uint);\n"
- "dvec3 subgroupClusteredAdd(dvec3, uint);\n"
- "dvec4 subgroupClusteredAdd(dvec4, uint);\n"
-
- "double subgroupClusteredMul(double, uint);\n"
- "dvec2 subgroupClusteredMul(dvec2, uint);\n"
- "dvec3 subgroupClusteredMul(dvec3, uint);\n"
- "dvec4 subgroupClusteredMul(dvec4, uint);\n"
-
- "double subgroupClusteredMin(double, uint);\n"
- "dvec2 subgroupClusteredMin(dvec2, uint);\n"
- "dvec3 subgroupClusteredMin(dvec3, uint);\n"
- "dvec4 subgroupClusteredMin(dvec4, uint);\n"
-
- "double subgroupClusteredMax(double, uint);\n"
- "dvec2 subgroupClusteredMax(dvec2, uint);\n"
- "dvec3 subgroupClusteredMax(dvec3, uint);\n"
- "dvec4 subgroupClusteredMax(dvec4, uint);\n"
-
- "double subgroupQuadBroadcast(double, uint);\n"
- "dvec2 subgroupQuadBroadcast(dvec2, uint);\n"
- "dvec3 subgroupQuadBroadcast(dvec3, uint);\n"
- "dvec4 subgroupQuadBroadcast(dvec4, uint);\n"
-
- "double subgroupQuadSwapHorizontal(double);\n"
- "dvec2 subgroupQuadSwapHorizontal(dvec2);\n"
- "dvec3 subgroupQuadSwapHorizontal(dvec3);\n"
- "dvec4 subgroupQuadSwapHorizontal(dvec4);\n"
-
- "double subgroupQuadSwapVertical(double);\n"
- "dvec2 subgroupQuadSwapVertical(dvec2);\n"
- "dvec3 subgroupQuadSwapVertical(dvec3);\n"
- "dvec4 subgroupQuadSwapVertical(dvec4);\n"
-
- "double subgroupQuadSwapDiagonal(double);\n"
- "dvec2 subgroupQuadSwapDiagonal(dvec2);\n"
- "dvec3 subgroupQuadSwapDiagonal(dvec3);\n"
- "dvec4 subgroupQuadSwapDiagonal(dvec4);\n"
-
-#ifdef NV_EXTENSIONS
- "uvec4 subgroupPartitionNV(double);\n"
- "uvec4 subgroupPartitionNV(dvec2);\n"
- "uvec4 subgroupPartitionNV(dvec3);\n"
- "uvec4 subgroupPartitionNV(dvec4);\n"
-
- "double subgroupPartitionedAddNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedAddNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedAddNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedAddNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedMulNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedMulNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedMulNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedMulNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedMinNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedMinNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedMinNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedMinNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedMaxNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedMaxNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedMaxNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedMaxNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedInclusiveAddNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedInclusiveAddNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedInclusiveAddNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedInclusiveAddNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedInclusiveMulNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedInclusiveMulNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedInclusiveMulNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedInclusiveMulNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedInclusiveMinNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedInclusiveMinNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedInclusiveMinNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedInclusiveMinNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedInclusiveMaxNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedInclusiveMaxNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedInclusiveMaxNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedInclusiveMaxNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedExclusiveAddNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedExclusiveAddNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedExclusiveAddNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedExclusiveAddNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedExclusiveMulNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedExclusiveMulNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedExclusiveMulNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedExclusiveMulNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedExclusiveMinNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedExclusiveMinNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedExclusiveMinNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedExclusiveMinNV(dvec4, uvec4 ballot);\n"
-
- "double subgroupPartitionedExclusiveMaxNV(double, uvec4 ballot);\n"
- "dvec2 subgroupPartitionedExclusiveMaxNV(dvec2, uvec4 ballot);\n"
- "dvec3 subgroupPartitionedExclusiveMaxNV(dvec3, uvec4 ballot);\n"
- "dvec4 subgroupPartitionedExclusiveMaxNV(dvec4, uvec4 ballot);\n"
-#endif
-
- "\n");
+ if (!logicalOp) {
+ for (unsigned int j = 0; j < sizeof(floatTypes)/sizeof(floatTypes[0]); ++j) {
+ snprintf(buf, bufSize, op, floatTypes[j], floatTypes[j]);
+ commonBuiltins.append(buf);
+ }
+ if (profile != EEsProfile && version >= 400) {
+ for (unsigned int j = 0; j < sizeof(doubleTypes)/sizeof(doubleTypes[0]); ++j) {
+ snprintf(buf, bufSize, op, doubleTypes[j], doubleTypes[j]);
+ commonBuiltins.append(buf);
+ }
+ }
}
+ if (!mathOp) {
+ for (unsigned int j = 0; j < sizeof(boolTypes)/sizeof(boolTypes[0]); ++j) {
+ snprintf(buf, bufSize, op, boolTypes[j], boolTypes[j]);
+ commonBuiltins.append(buf);
+ }
+ }
+ for (unsigned int j = 0; j < sizeof(intTypes)/sizeof(intTypes[0]); ++j) {
+ snprintf(buf, bufSize, op, intTypes[j], intTypes[j]);
+ commonBuiltins.append(buf);
+ }
+ }
stageBuiltins[EShLangCompute].append(
"void subgroupMemoryBarrierShared();"
diff --git a/gtests/Spv.FromFile.cpp b/gtests/Spv.FromFile.cpp
index e6ee32b..0566c8f 100644
--- a/gtests/Spv.FromFile.cpp
+++ b/gtests/Spv.FromFile.cpp
@@ -451,6 +451,8 @@
"spv.subgroupExtendedTypesBallotNeg.comp",
"spv.subgroupExtendedTypesClustered.comp",
"spv.subgroupExtendedTypesClusteredNeg.comp",
+ "spv.subgroupExtendedTypesPartitioned.comp",
+ "spv.subgroupExtendedTypesPartitionedNeg.comp",
"spv.subgroupExtendedTypesShuffle.comp",
"spv.subgroupExtendedTypesShuffleNeg.comp",
"spv.subgroupExtendedTypesShuffleRelative.comp",
