| 1 | // Autogenerated by AMDGPUTargetDefEmitter.cpp |
| 2 | |
| 3 | #ifdef GET_AMDGPU_GPU_ENUM |
| 4 | #undef GET_AMDGPU_GPU_ENUM |
| 5 | GK_GENERIC, |
| 6 | GK_GENERIC_HSA, |
| 7 | GK_GFX600, |
| 8 | GK_GFX601, |
| 9 | GK_GFX602, |
| 10 | GK_GFX700, |
| 11 | GK_GFX701, |
| 12 | GK_GFX702, |
| 13 | GK_GFX703, |
| 14 | GK_GFX704, |
| 15 | GK_GFX705, |
| 16 | GK_GFX801, |
| 17 | GK_GFX802, |
| 18 | GK_GFX803, |
| 19 | GK_GFX805, |
| 20 | GK_GFX810, |
| 21 | GK_GFX900, |
| 22 | GK_GFX902, |
| 23 | GK_GFX904, |
| 24 | GK_GFX906, |
| 25 | GK_GFX908, |
| 26 | GK_GFX909, |
| 27 | GK_GFX90A, |
| 28 | GK_GFX90C, |
| 29 | GK_GFX942, |
| 30 | GK_GFX950, |
| 31 | GK_GFX1010, |
| 32 | GK_GFX1011, |
| 33 | GK_GFX1012, |
| 34 | GK_GFX1013, |
| 35 | GK_GFX1030, |
| 36 | GK_GFX1031, |
| 37 | GK_GFX1032, |
| 38 | GK_GFX1033, |
| 39 | GK_GFX1034, |
| 40 | GK_GFX1035, |
| 41 | GK_GFX1036, |
| 42 | GK_GFX1100, |
| 43 | GK_GFX1101, |
| 44 | GK_GFX1102, |
| 45 | GK_GFX1103, |
| 46 | GK_GFX1150, |
| 47 | GK_GFX1151, |
| 48 | GK_GFX1152, |
| 49 | GK_GFX1153, |
| 50 | GK_GFX1154, |
| 51 | GK_GFX1170, |
| 52 | GK_GFX1171, |
| 53 | GK_GFX1172, |
| 54 | GK_GFX1200, |
| 55 | GK_GFX1201, |
| 56 | GK_GFX1250_STRICT, |
| 57 | GK_GFX1250, |
| 58 | GK_GFX1251, |
| 59 | GK_GFX1310, |
| 60 | GK_GFX9_GENERIC, |
| 61 | GK_GFX9_4_GENERIC, |
| 62 | GK_GFX10_1_GENERIC, |
| 63 | GK_GFX10_3_GENERIC, |
| 64 | GK_GFX11_GENERIC, |
| 65 | GK_GFX11_7_GENERIC, |
| 66 | GK_GFX12_GENERIC, |
| 67 | GK_GFX12_5_GENERIC, |
| 68 | GK_GFX13_GENERIC, |
| 69 | #endif // GET_AMDGPU_GPU_ENUM |
| 70 | |
| 71 | #ifdef GET_AMDGPU_MAJOR_SUBARCH |
| 72 | #undef GET_AMDGPU_MAJOR_SUBARCH |
| 73 | struct AMDGPUMajorSubArchEntry { |
| 74 | Triple::SubArchType SubArch; |
| 75 | Triple::SubArchType Major; |
| 76 | }; |
| 77 | static constexpr std::array<AMDGPUMajorSubArchEntry, 49> AMDGPUMajorSubArch = {{ |
| 78 | {Triple::AMDGPUSubArch942, Triple::AMDGPUSubArch9_4}, |
| 79 | {Triple::AMDGPUSubArch950, Triple::AMDGPUSubArch9_4}, |
| 80 | {Triple::AMDGPUSubArch900, Triple::AMDGPUSubArch9}, |
| 81 | {Triple::AMDGPUSubArch902, Triple::AMDGPUSubArch9}, |
| 82 | {Triple::AMDGPUSubArch904, Triple::AMDGPUSubArch9}, |
| 83 | {Triple::AMDGPUSubArch906, Triple::AMDGPUSubArch9}, |
| 84 | {Triple::AMDGPUSubArch909, Triple::AMDGPUSubArch9}, |
| 85 | {Triple::AMDGPUSubArch90C, Triple::AMDGPUSubArch9}, |
| 86 | {Triple::AMDGPUSubArch1010, Triple::AMDGPUSubArch10_1}, |
| 87 | {Triple::AMDGPUSubArch1011, Triple::AMDGPUSubArch10_1}, |
| 88 | {Triple::AMDGPUSubArch1012, Triple::AMDGPUSubArch10_1}, |
| 89 | {Triple::AMDGPUSubArch1013, Triple::AMDGPUSubArch10_1}, |
| 90 | {Triple::AMDGPUSubArch1030, Triple::AMDGPUSubArch10_3}, |
| 91 | {Triple::AMDGPUSubArch1031, Triple::AMDGPUSubArch10_3}, |
| 92 | {Triple::AMDGPUSubArch1032, Triple::AMDGPUSubArch10_3}, |
| 93 | {Triple::AMDGPUSubArch1033, Triple::AMDGPUSubArch10_3}, |
| 94 | {Triple::AMDGPUSubArch1034, Triple::AMDGPUSubArch10_3}, |
| 95 | {Triple::AMDGPUSubArch1035, Triple::AMDGPUSubArch10_3}, |
| 96 | {Triple::AMDGPUSubArch1036, Triple::AMDGPUSubArch10_3}, |
| 97 | {Triple::AMDGPUSubArch1170, Triple::AMDGPUSubArch11_7}, |
| 98 | {Triple::AMDGPUSubArch1171, Triple::AMDGPUSubArch11_7}, |
| 99 | {Triple::AMDGPUSubArch1172, Triple::AMDGPUSubArch11_7}, |
| 100 | {Triple::AMDGPUSubArch1100, Triple::AMDGPUSubArch11}, |
| 101 | {Triple::AMDGPUSubArch1101, Triple::AMDGPUSubArch11}, |
| 102 | {Triple::AMDGPUSubArch1102, Triple::AMDGPUSubArch11}, |
| 103 | {Triple::AMDGPUSubArch1103, Triple::AMDGPUSubArch11}, |
| 104 | {Triple::AMDGPUSubArch1150, Triple::AMDGPUSubArch11}, |
| 105 | {Triple::AMDGPUSubArch1151, Triple::AMDGPUSubArch11}, |
| 106 | {Triple::AMDGPUSubArch1152, Triple::AMDGPUSubArch11}, |
| 107 | {Triple::AMDGPUSubArch1153, Triple::AMDGPUSubArch11}, |
| 108 | {Triple::AMDGPUSubArch1154, Triple::AMDGPUSubArch11}, |
| 109 | {Triple::AMDGPUSubArch1250, Triple::AMDGPUSubArch12_5}, |
| 110 | {Triple::AMDGPUSubArch1251, Triple::AMDGPUSubArch12_5}, |
| 111 | {Triple::AMDGPUSubArch1200, Triple::AMDGPUSubArch12}, |
| 112 | {Triple::AMDGPUSubArch1201, Triple::AMDGPUSubArch12}, |
| 113 | {Triple::AMDGPUSubArch1310, Triple::AMDGPUSubArch13}, |
| 114 | {Triple::AMDGPUSubArch600, Triple::AMDGPUSubArch6}, |
| 115 | {Triple::AMDGPUSubArch601, Triple::AMDGPUSubArch6}, |
| 116 | {Triple::AMDGPUSubArch602, Triple::AMDGPUSubArch6}, |
| 117 | {Triple::AMDGPUSubArch700, Triple::AMDGPUSubArch7}, |
| 118 | {Triple::AMDGPUSubArch701, Triple::AMDGPUSubArch7}, |
| 119 | {Triple::AMDGPUSubArch702, Triple::AMDGPUSubArch7}, |
| 120 | {Triple::AMDGPUSubArch703, Triple::AMDGPUSubArch7}, |
| 121 | {Triple::AMDGPUSubArch704, Triple::AMDGPUSubArch7}, |
| 122 | {Triple::AMDGPUSubArch705, Triple::AMDGPUSubArch7}, |
| 123 | {Triple::AMDGPUSubArch801, Triple::AMDGPUSubArch8}, |
| 124 | {Triple::AMDGPUSubArch802, Triple::AMDGPUSubArch8}, |
| 125 | {Triple::AMDGPUSubArch803, Triple::AMDGPUSubArch8}, |
| 126 | {Triple::AMDGPUSubArch805, Triple::AMDGPUSubArch8}, |
| 127 | }}; |
| 128 | #endif // GET_AMDGPU_MAJOR_SUBARCH |
| 129 | |
| 130 | #ifdef GET_AMDGPU_NAME_TABLE |
| 131 | #undef GET_AMDGPU_NAME_TABLE |
| 132 | |
| 133 | #ifdef __GNUC__ |
| 134 | #pragma GCC diagnostic push |
| 135 | #pragma GCC diagnostic ignored "-Woverlength-strings" |
| 136 | #endif |
| 137 | static constexpr char AMDGPUNameTableStorage[] = |
| 138 | "\0" |
| 139 | "16-bit-insts\0" |
| 140 | "add-min-max-insts\0" |
| 141 | "ashr-pk-insts\0" |
| 142 | "async-load-to-lds-insts\0" |
| 143 | "async-store-from-lds-insts\0" |
| 144 | "asynccnt\0" |
| 145 | "atomic-buffer-global-pk-add-f16-insts\0" |
| 146 | "atomic-buffer-pk-add-bf16-inst\0" |
| 147 | "atomic-ds-pk-add-16-insts\0" |
| 148 | "atomic-fmin-fmax-global-f32\0" |
| 149 | "atomic-fmin-fmax-global-f64\0" |
| 150 | "atomic-fadd-no-rtn-insts\0" |
| 151 | "atomic-fadd-rtn-insts\0" |
| 152 | "atomic-flat-pk-add-16-insts\0" |
| 153 | "atomic-global-pk-add-bf16-inst\0" |
| 154 | "bf16-cvt-insts\0" |
| 155 | "bf16-pk-insts\0" |
| 156 | "bf16-trans-insts\0" |
| 157 | "bf8-cvt-scale-insts\0" |
| 158 | "bvh-ray-tracing-insts\0" |
| 159 | "bitop3-insts\0" |
| 160 | "buffer-inv-inst\0" |
| 161 | "ci-insts\0" |
| 162 | "clusters\0" |
| 163 | "cube-insts\0" |
| 164 | "cvt-pknorm-vop2-insts\0" |
| 165 | "cvt-sr-pk-bf16-f32-inst\0" |
| 166 | "dl-insts\0" |
| 167 | "dpp\0" |
| 168 | "dot10-insts\0" |
| 169 | "dot11-insts\0" |
| 170 | "dot12-insts\0" |
| 171 | "dot13-insts\0" |
| 172 | "dot1-insts\0" |
| 173 | "dot2-insts\0" |
| 174 | "dot3-insts\0" |
| 175 | "dot4-insts\0" |
| 176 | "dot5-insts\0" |
| 177 | "dot6-insts\0" |
| 178 | "dot7-insts\0" |
| 179 | "dot8-insts\0" |
| 180 | "dot9-insts\0" |
| 181 | "exclusive-scan-insts\0" |
| 182 | "extended-image-insts\0" |
| 183 | "f16bf16-to-fp6bf6-cvt-scale-insts\0" |
| 184 | "f32-to-f16bf16-cvt-sr-insts\0" |
| 185 | "f32-to-fp6bf6-cvt-scale-insts\0" |
| 186 | "fp4-cvt-scale-insts\0" |
| 187 | "fp6bf6-cvt-scale-insts\0" |
| 188 | "fp8-conversion-insts\0" |
| 189 | "fp8-cvt-scale-insts\0" |
| 190 | "block16-cvt-scale-insts\0" |
| 191 | "fp8e5m3-insts\0" |
| 192 | "fp8-insts\0" |
| 193 | "fp6bf6-to-f16bf16f32-cvt-scale-insts\0" |
| 194 | "flat-buffer-global-fadd-f64-inst\0" |
| 195 | "flat-global-insts\0" |
| 196 | "gfx10-insts\0" |
| 197 | "gfx10-3-insts\0" |
| 198 | "gfx11-insts\0" |
| 199 | "gfx1250-insts\0" |
| 200 | "gfx1251-gemm-insts\0" |
| 201 | "gfx12-insts\0" |
| 202 | "gfx13-insts\0" |
| 203 | "gfx8-insts\0" |
| 204 | "gfx90a-insts\0" |
| 205 | "gfx940-insts\0" |
| 206 | "gfx950-insts\0" |
| 207 | "gfx9-insts\0" |
| 208 | "gws\0" |
| 209 | "image-insts\0" |
| 210 | "lds-barrier-arrive-atomic\0" |
| 211 | "lerp-inst\0" |
| 212 | "mai-insts\0" |
| 213 | "mcast-load-insts\0" |
| 214 | "mqsad-insts\0" |
| 215 | "mqsad-pk-insts\0" |
| 216 | "msad-insts\0" |
| 217 | "ocp-fp8-conversion-insts\0" |
| 218 | "permlane16-insts\0" |
| 219 | "permlane16-swap\0" |
| 220 | "permlane32-swap\0" |
| 221 | "pk-add-min-max-insts\0" |
| 222 | "prng-inst\0" |
| 223 | "qsad-insts\0" |
| 224 | "s-memrealtime\0" |
| 225 | "s-memtime-inst\0" |
| 226 | "swmmac-gfx1200-insts\0" |
| 227 | "swmmac-gfx1250-insts\0" |
| 228 | "s-wakeup-barrier-inst\0" |
| 229 | "sad-insts\0" |
| 230 | "setprio-inc-wg-inst\0" |
| 231 | "smem-prefetch-insts\0" |
| 232 | "tdm-insts\0" |
| 233 | "tanh-insts\0" |
| 234 | "tensor-cvt-lut-insts\0" |
| 235 | "transpose-load-f4f6-insts\0" |
| 236 | "vmem-to-lds-load-insts\0" |
| 237 | "vmem-pref-insts\0" |
| 238 | "wave-match-insts\0" |
| 239 | "wmma-128b-insts\0" |
| 240 | "wmma-256b-insts\0" |
| 241 | "wmma-n16-insts\0" |
| 242 | "wmma-f4-insts\0" |
| 243 | "wmma-16x16x128fp8-insts\0" |
| 244 | "xf32-insts\0" |
| 245 | "wavefrontsize32\0" |
| 246 | "wavefrontsize64\0" |
| 247 | "supports-wgp\0" |
| 248 | "supports-wave32\0" |
| 249 | "fast-fmaf\0" |
| 250 | "fast-denormal-f32\0" |
| 251 | "xnack-support\0" |
| 252 | "sramecc-support\0" |
| 253 | "xnack-on-off-modes\0" |
| 254 | "sramecc-on-off-modes\0" |
| 255 | "sgpr-init-bug\0" |
| 256 | "aperture-regs\0" |
| 257 | "get-doorbell-id\0" |
| 258 | "agpr-alloc\0" |
| 259 | "1536-physical-vgprs\0" |
| 260 | "1024-addressable-vgprs\0" |
| 261 | "half-addressable-physical-local-memory\0" |
| 262 | "back-off-barrier\0" |
| 263 | "lds-alloc-granularity-256\0" |
| 264 | "lds-alloc-granularity-512\0" |
| 265 | "lds-alloc-granularity-1024\0" |
| 266 | "lds-alloc-granularity-1280\0" |
| 267 | "lds-alloc-granularity-2048\0" |
| 268 | "lds-encoding-granularity-256\0" |
| 269 | "lds-encoding-granularity-512\0" |
| 270 | "lds-encoding-granularity-1024\0" |
| 271 | "lds-encoding-granularity-1280\0" |
| 272 | "lds-encoding-granularity-2048\0" |
| 273 | "generic\0" |
| 274 | "generic-hsa\0" |
| 275 | "gfx600\0" |
| 276 | "gfx6\0" |
| 277 | "gfx601\0" |
| 278 | "gfx602\0" |
| 279 | "gfx700\0" |
| 280 | "gfx7\0" |
| 281 | "gfx701\0" |
| 282 | "gfx702\0" |
| 283 | "gfx703\0" |
| 284 | "gfx704\0" |
| 285 | "gfx705\0" |
| 286 | "gfx801\0" |
| 287 | "gfx8\0" |
| 288 | "gfx802\0" |
| 289 | "gfx803\0" |
| 290 | "gfx805\0" |
| 291 | "gfx810\0" |
| 292 | "gfx900\0" |
| 293 | "gfx9\0" |
| 294 | "gfx902\0" |
| 295 | "gfx904\0" |
| 296 | "gfx906\0" |
| 297 | "gfx908\0" |
| 298 | "gfx909\0" |
| 299 | "gfx90a\0" |
| 300 | "gfx90c\0" |
| 301 | "gfx942\0" |
| 302 | "gfx950\0" |
| 303 | "gfx1010\0" |
| 304 | "gfx10\0" |
| 305 | "gfx1011\0" |
| 306 | "gfx1012\0" |
| 307 | "gfx1013\0" |
| 308 | "gfx1030\0" |
| 309 | "gfx1031\0" |
| 310 | "gfx1032\0" |
| 311 | "gfx1033\0" |
| 312 | "gfx1034\0" |
| 313 | "gfx1035\0" |
| 314 | "gfx1036\0" |
| 315 | "gfx1100\0" |
| 316 | "gfx11\0" |
| 317 | "gfx1101\0" |
| 318 | "gfx1102\0" |
| 319 | "gfx1103\0" |
| 320 | "gfx1150\0" |
| 321 | "gfx1151\0" |
| 322 | "gfx1152\0" |
| 323 | "gfx1153\0" |
| 324 | "gfx1154\0" |
| 325 | "gfx1170\0" |
| 326 | "gfx1171\0" |
| 327 | "gfx1172\0" |
| 328 | "gfx1200\0" |
| 329 | "gfx12\0" |
| 330 | "gfx1201\0" |
| 331 | "gfx1250-strict\0" |
| 332 | "gfx1250\0" |
| 333 | "gfx1251\0" |
| 334 | "gfx1310\0" |
| 335 | "gfx13\0" |
| 336 | "gfx9-generic\0" |
| 337 | "gfx9-4-generic\0" |
| 338 | "gfx10-1-generic\0" |
| 339 | "gfx10-3-generic\0" |
| 340 | "gfx11-generic\0" |
| 341 | "gfx11-7-generic\0" |
| 342 | "gfx12-generic\0" |
| 343 | "gfx12-5-generic\0" |
| 344 | "gfx13-generic\0" |
| 345 | "tahiti\0" |
| 346 | "pitcairn\0" |
| 347 | "verde\0" |
| 348 | "hainan\0" |
| 349 | "oland\0" |
| 350 | "kaveri\0" |
| 351 | "hawaii\0" |
| 352 | "kabini\0" |
| 353 | "mullins\0" |
| 354 | "bonaire\0" |
| 355 | "carrizo\0" |
| 356 | "iceland\0" |
| 357 | "tonga\0" |
| 358 | "fiji\0" |
| 359 | "polaris10\0" |
| 360 | "polaris11\0" |
| 361 | "tongapro\0" |
| 362 | "stoney\0" |
| 363 | "amdgpu6.00\0" |
| 364 | "amdgpu6.01\0" |
| 365 | "amdgpu6.02\0" |
| 366 | "amdgpu7.00\0" |
| 367 | "amdgpu7.01\0" |
| 368 | "amdgpu7.02\0" |
| 369 | "amdgpu7.03\0" |
| 370 | "amdgpu7.04\0" |
| 371 | "amdgpu7.05\0" |
| 372 | "amdgpu8.01\0" |
| 373 | "amdgpu8.02\0" |
| 374 | "amdgpu8.03\0" |
| 375 | "amdgpu8.05\0" |
| 376 | "amdgpu8.10\0" |
| 377 | "amdgpu9.00\0" |
| 378 | "amdgpu9.02\0" |
| 379 | "amdgpu9.04\0" |
| 380 | "amdgpu9.06\0" |
| 381 | "amdgpu9.08\0" |
| 382 | "amdgpu9.09\0" |
| 383 | "amdgpu9.0a\0" |
| 384 | "amdgpu9.0c\0" |
| 385 | "amdgpu9.42\0" |
| 386 | "amdgpu9.50\0" |
| 387 | "amdgpu9\0" |
| 388 | "amdgpu9.4\0" |
| 389 | "amdgpu10.10\0" |
| 390 | "amdgpu10.11\0" |
| 391 | "amdgpu10.12\0" |
| 392 | "amdgpu10.13\0" |
| 393 | "amdgpu10.30\0" |
| 394 | "amdgpu10.31\0" |
| 395 | "amdgpu10.32\0" |
| 396 | "amdgpu10.33\0" |
| 397 | "amdgpu10.34\0" |
| 398 | "amdgpu10.35\0" |
| 399 | "amdgpu10.36\0" |
| 400 | "amdgpu10.1\0" |
| 401 | "amdgpu10.3\0" |
| 402 | "amdgpu11.00\0" |
| 403 | "amdgpu11.01\0" |
| 404 | "amdgpu11.02\0" |
| 405 | "amdgpu11.03\0" |
| 406 | "amdgpu11.50\0" |
| 407 | "amdgpu11.51\0" |
| 408 | "amdgpu11.52\0" |
| 409 | "amdgpu11.53\0" |
| 410 | "amdgpu11.54\0" |
| 411 | "amdgpu11.70\0" |
| 412 | "amdgpu11.71\0" |
| 413 | "amdgpu11.72\0" |
| 414 | "amdgpu11\0" |
| 415 | "amdgpu11.7\0" |
| 416 | "amdgpu12.00\0" |
| 417 | "amdgpu12.01\0" |
| 418 | "amdgpu12\0" |
| 419 | "amdgpu12.50s\0" |
| 420 | "amdgpu12.50\0" |
| 421 | "amdgpu12.51\0" |
| 422 | "amdgpu12.5\0" |
| 423 | "amdgpu13.10\0" |
| 424 | "amdgpu13\0" |
| 425 | "amdgpu6\0" |
| 426 | "amdgpu7\0" |
| 427 | "amdgpu8\0" |
| 428 | "amdgpu\0" |
| 429 | ; |
| 430 | #ifdef __GNUC__ |
| 431 | #pragma GCC diagnostic pop |
| 432 | #endif |
| 433 | |
| 434 | static constexpr llvm::StringTable |
| 435 | AMDGPUNameTable = AMDGPUNameTableStorage; |
| 436 | #endif // GET_AMDGPU_NAME_TABLE |
| 437 | |
| 438 | #ifdef GET_AMDGPU_FEATURE_ENUM |
| 439 | #undef GET_AMDGPU_FEATURE_ENUM |
| 440 | FEAT_16_BIT_INSTS, |
| 441 | FEAT_ADD_MIN_MAX_INSTS, |
| 442 | FEAT_ASHR_PK_INSTS, |
| 443 | FEAT_ASYNC_LOAD_TO_LDS_INSTS, |
| 444 | FEAT_ASYNC_STORE_FROM_LDS_INSTS, |
| 445 | FEAT_ASYNCCNT, |
| 446 | FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, |
| 447 | FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, |
| 448 | FEAT_ATOMIC_DS_PK_ADD_16_INSTS, |
| 449 | FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, |
| 450 | FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, |
| 451 | FEAT_ATOMIC_FADD_NO_RTN_INSTS, |
| 452 | FEAT_ATOMIC_FADD_RTN_INSTS, |
| 453 | FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, |
| 454 | FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, |
| 455 | FEAT_BF16_CVT_INSTS, |
| 456 | FEAT_BF16_PK_INSTS, |
| 457 | FEAT_BF16_TRANS_INSTS, |
| 458 | FEAT_BF8_CVT_SCALE_INSTS, |
| 459 | FEAT_BVH_RAY_TRACING_INSTS, |
| 460 | FEAT_BITOP3_INSTS, |
| 461 | FEAT_BUFFER_INV_INST, |
| 462 | FEAT_CI_INSTS, |
| 463 | FEAT_CLUSTERS, |
| 464 | FEAT_CUBE_INSTS, |
| 465 | FEAT_CVT_PKNORM_VOP2_INSTS, |
| 466 | FEAT_CVT_SR_PK_BF16_F32_INST, |
| 467 | FEAT_DL_INSTS, |
| 468 | FEAT_DPP, |
| 469 | FEAT_DOT10_INSTS, |
| 470 | FEAT_DOT11_INSTS, |
| 471 | FEAT_DOT12_INSTS, |
| 472 | FEAT_DOT13_INSTS, |
| 473 | FEAT_DOT1_INSTS, |
| 474 | FEAT_DOT2_INSTS, |
| 475 | FEAT_DOT3_INSTS, |
| 476 | FEAT_DOT4_INSTS, |
| 477 | FEAT_DOT5_INSTS, |
| 478 | FEAT_DOT6_INSTS, |
| 479 | FEAT_DOT7_INSTS, |
| 480 | FEAT_DOT8_INSTS, |
| 481 | FEAT_DOT9_INSTS, |
| 482 | FEAT_EXCLUSIVE_SCAN_INSTS, |
| 483 | FEAT_EXTENDED_IMAGE_INSTS, |
| 484 | FEAT_F16BF16_TO_FP6BF6_CVT_SCALE_INSTS, |
| 485 | FEAT_F32_TO_F16BF16_CVT_SR_INSTS, |
| 486 | FEAT_F32_TO_FP6BF6_CVT_SCALE_INSTS, |
| 487 | FEAT_FP4_CVT_SCALE_INSTS, |
| 488 | FEAT_FP6BF6_CVT_SCALE_INSTS, |
| 489 | FEAT_FP8_CONVERSION_INSTS, |
| 490 | FEAT_FP8_CVT_SCALE_INSTS, |
| 491 | FEAT_BLOCK16_CVT_SCALE_INSTS, |
| 492 | FEAT_FP8E5M3_INSTS, |
| 493 | FEAT_FP8_INSTS, |
| 494 | FEAT_FP6BF6_TO_F16BF16F32_CVT_SCALE_INSTS, |
| 495 | FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, |
| 496 | FEAT_FLAT_GLOBAL_INSTS, |
| 497 | FEAT_GFX10_INSTS, |
| 498 | FEAT_GFX10_3_INSTS, |
| 499 | FEAT_GFX11_INSTS, |
| 500 | FEAT_GFX1250_INSTS, |
| 501 | FEAT_GFX1251_GEMM_INSTS, |
| 502 | FEAT_GFX12_INSTS, |
| 503 | FEAT_GFX13_INSTS, |
| 504 | FEAT_GFX8_INSTS, |
| 505 | FEAT_GFX90A_INSTS, |
| 506 | FEAT_GFX940_INSTS, |
| 507 | FEAT_GFX950_INSTS, |
| 508 | FEAT_GFX9_INSTS, |
| 509 | FEAT_GWS, |
| 510 | FEAT_IMAGE_INSTS, |
| 511 | FEAT_LDS_BARRIER_ARRIVE_ATOMIC, |
| 512 | FEAT_LERP_INST, |
| 513 | FEAT_MAI_INSTS, |
| 514 | FEAT_MCAST_LOAD_INSTS, |
| 515 | FEAT_MQSAD_INSTS, |
| 516 | FEAT_MQSAD_PK_INSTS, |
| 517 | FEAT_MSAD_INSTS, |
| 518 | FEAT_OCP_FP8_CONVERSION_INSTS, |
| 519 | FEAT_PERMLANE16_INSTS, |
| 520 | FEAT_PERMLANE16_SWAP, |
| 521 | FEAT_PERMLANE32_SWAP, |
| 522 | FEAT_PK_ADD_MIN_MAX_INSTS, |
| 523 | FEAT_PRNG_INST, |
| 524 | FEAT_QSAD_INSTS, |
| 525 | FEAT_S_MEMREALTIME, |
| 526 | FEAT_S_MEMTIME_INST, |
| 527 | FEAT_SWMMAC_GFX1200_INSTS, |
| 528 | FEAT_SWMMAC_GFX1250_INSTS, |
| 529 | FEAT_S_WAKEUP_BARRIER_INST, |
| 530 | FEAT_SAD_INSTS, |
| 531 | FEAT_SETPRIO_INC_WG_INST, |
| 532 | FEAT_SMEM_PREFETCH_INSTS, |
| 533 | FEAT_TDM_INSTS, |
| 534 | FEAT_TANH_INSTS, |
| 535 | FEAT_TENSOR_CVT_LUT_INSTS, |
| 536 | FEAT_TRANSPOSE_LOAD_F4F6_INSTS, |
| 537 | FEAT_VMEM_TO_LDS_LOAD_INSTS, |
| 538 | FEAT_VMEM_PREF_INSTS, |
| 539 | FEAT_WAVE_MATCH_INSTS, |
| 540 | FEAT_WMMA_128B_INSTS, |
| 541 | FEAT_WMMA_256B_INSTS, |
| 542 | FEAT_WMMA_N16_INSTS, |
| 543 | FEAT_WMMA_F4_INSTS, |
| 544 | FEAT_WMMA_16X16X128FP8_INSTS, |
| 545 | FEAT_XF32_INSTS, |
| 546 | FEAT_WAVEFRONTSIZE32, |
| 547 | FEAT_WAVEFRONTSIZE64, |
| 548 | FEAT_SUPPORTS_WGP, |
| 549 | FEAT_SUPPORTS_WAVE32, |
| 550 | FEAT_FAST_FMAF, |
| 551 | FEAT_FAST_DENORMAL_F32, |
| 552 | FEAT_XNACK_SUPPORT, |
| 553 | FEAT_SRAMECC_SUPPORT, |
| 554 | FEAT_XNACK_ON_OFF_MODES, |
| 555 | FEAT_SRAMECC_ON_OFF_MODES, |
| 556 | FEAT_SGPR_INIT_BUG, |
| 557 | FEAT_APERTURE_REGS, |
| 558 | FEAT_GET_DOORBELL_ID, |
| 559 | FEAT_AGPR_ALLOC, |
| 560 | FEAT_1536_PHYSICAL_VGPRS, |
| 561 | FEAT_1024_ADDRESSABLE_VGPRS, |
| 562 | FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, |
| 563 | FEAT_BACK_OFF_BARRIER, |
| 564 | FEAT_LDS_ALLOC_GRANULARITY_256, |
| 565 | FEAT_LDS_ALLOC_GRANULARITY_512, |
| 566 | FEAT_LDS_ALLOC_GRANULARITY_1024, |
| 567 | FEAT_LDS_ALLOC_GRANULARITY_1280, |
| 568 | FEAT_LDS_ALLOC_GRANULARITY_2048, |
| 569 | FEAT_LDS_ENCODING_GRANULARITY_256, |
| 570 | FEAT_LDS_ENCODING_GRANULARITY_512, |
| 571 | FEAT_LDS_ENCODING_GRANULARITY_1024, |
| 572 | FEAT_LDS_ENCODING_GRANULARITY_1280, |
| 573 | FEAT_LDS_ENCODING_GRANULARITY_2048, |
| 574 | NUM_FEATURES |
| 575 | #endif // GET_AMDGPU_FEATURE_ENUM |
| 576 | |
| 577 | #ifdef GET_AMDGPU_GPU_TABLE |
| 578 | #undef GET_AMDGPU_GPU_TABLE |
| 579 | static constexpr GPUKind AMDGPUFirstGPUKind = GK_GENERIC; |
| 580 | static constexpr GPUInfo AMDGPUGPUTable[] = { |
| 581 | {2415, Triple::NoSubArch, AMDGPUFeatureBitset({}), {6, 0, 0}, 0, 10, 32768, 32, 0}, |
| 582 | {2423, Triple::NoSubArch, AMDGPUFeatureBitset({}), {7, 0, 0}, 0, 10, 32768, 32, 0}, |
| 583 | {2435, Triple::AMDGPUSubArch600, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_LDS_ALLOC_GRANULARITY_256, FEAT_LDS_ENCODING_GRANULARITY_256}), {6, 0, 0}, 2442, 10, 32768, 32, 32}, |
| 584 | {2447, Triple::AMDGPUSubArch601, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_LDS_ALLOC_GRANULARITY_256, FEAT_LDS_ENCODING_GRANULARITY_256}), {6, 0, 1}, 2442, 10, 32768, 32, 32}, |
| 585 | {2454, Triple::AMDGPUSubArch602, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_LDS_ALLOC_GRANULARITY_256, FEAT_LDS_ENCODING_GRANULARITY_256}), {6, 0, 2}, 2442, 10, 32768, 32, 32}, |
| 586 | {2461, Triple::AMDGPUSubArch700, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 0}, 2468, 10, 65536, 32, 32}, |
| 587 | {2473, Triple::AMDGPUSubArch701, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 1}, 2468, 10, 65536, 32, 32}, |
| 588 | {2480, Triple::AMDGPUSubArch702, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 2}, 2468, 10, 65536, 16, 32}, |
| 589 | {2487, Triple::AMDGPUSubArch703, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 3}, 2468, 10, 65536, 16, 32}, |
| 590 | {2494, Triple::AMDGPUSubArch704, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 4}, 2468, 10, 65536, 32, 32}, |
| 591 | {2501, Triple::AMDGPUSubArch705, AMDGPUFeatureBitset({FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {7, 0, 5}, 2468, 10, 65536, 16, 32}, |
| 592 | {2508, Triple::AMDGPUSubArch801, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GFX8_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {8, 0, 1}, 2515, 10, 65536, 32, 32}, |
| 593 | {2520, Triple::AMDGPUSubArch802, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GFX8_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_DENORMAL_F32, FEAT_SGPR_INIT_BUG, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {8, 0, 2}, 2515, 10, 65536, 32, 32}, |
| 594 | {2527, Triple::AMDGPUSubArch803, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GFX8_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_DENORMAL_F32, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {8, 0, 3}, 2515, 10, 65536, 32, 32}, |
| 595 | {2534, Triple::AMDGPUSubArch805, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GFX8_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_DENORMAL_F32, FEAT_SGPR_INIT_BUG, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {8, 0, 5}, 2515, 10, 65536, 32, 32}, |
| 596 | {2541, Triple::AMDGPUSubArch810, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_GFX8_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {8, 1, 0}, 2515, 10, 65536, 16, 32}, |
| 597 | {2548, Triple::AMDGPUSubArch900, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 0}, 2555, 10, 65536, 32, 32}, |
| 598 | {2560, Triple::AMDGPUSubArch902, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 2}, 2555, 10, 65536, 32, 32}, |
| 599 | {2567, Triple::AMDGPUSubArch904, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 4}, 2555, 10, 65536, 32, 32}, |
| 600 | {2574, Triple::AMDGPUSubArch906, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 6}, 2555, 10, 65536, 32, 32}, |
| 601 | {2581, Triple::AMDGPUSubArch908, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT3_INSTS, FEAT_DOT4_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MAI_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 8}, 2555, 10, 65536, 32, 32}, |
| 602 | {2588, Triple::AMDGPUSubArch909, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 9}, 2555, 10, 65536, 32, 32}, |
| 603 | {2595, Triple::AMDGPUSubArch90A, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT3_INSTS, FEAT_DOT4_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX90A_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MAI_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_AGPR_ALLOC, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 10}, 2555, 8, 65536, 32, 32}, |
| 604 | {2602, Triple::AMDGPUSubArch90C, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 12}, 2555, 10, 65536, 32, 32}, |
| 605 | {2609, Triple::AMDGPUSubArch942, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BUFFER_INV_INST, FEAT_CI_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT3_INSTS, FEAT_DOT4_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FP8_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX90A_INSTS, FEAT_GFX940_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_LERP_INST, FEAT_MAI_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_XF32_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_AGPR_ALLOC, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 4, 2}, 2555, 8, 65536, 32, 32}, |
| 606 | {2616, Triple::AMDGPUSubArch950, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF8_CVT_SCALE_INSTS, FEAT_BITOP3_INSTS, FEAT_BUFFER_INV_INST, FEAT_CI_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT13_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT3_INSTS, FEAT_DOT4_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_F16BF16_TO_FP6BF6_CVT_SCALE_INSTS, FEAT_F32_TO_F16BF16_CVT_SR_INSTS, FEAT_FP4_CVT_SCALE_INSTS, FEAT_FP6BF6_CVT_SCALE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FP8_CVT_SCALE_INSTS, FEAT_FP8_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX90A_INSTS, FEAT_GFX940_INSTS, FEAT_GFX950_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_LERP_INST, FEAT_MAI_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PERMLANE32_SWAP, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_AGPR_ALLOC, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1280, FEAT_LDS_ENCODING_GRANULARITY_1280}), {9, 5, 0}, 2555, 8, 163840, 64, 32}, |
| 607 | {2623, Triple::AMDGPUSubArch1010, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 1, 0}, 2631, 20, 65536, 32, 32}, |
| 608 | {2637, Triple::AMDGPUSubArch1011, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 1, 1}, 2631, 20, 65536, 32, 32}, |
| 609 | {2645, Triple::AMDGPUSubArch1012, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 1, 2}, 2631, 20, 65536, 32, 32}, |
| 610 | {2653, Triple::AMDGPUSubArch1013, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 1, 3}, 2631, 20, 65536, 32, 32}, |
| 611 | {2661, Triple::AMDGPUSubArch1030, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 0}, 2631, 16, 65536, 32, 32}, |
| 612 | {2669, Triple::AMDGPUSubArch1031, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 1}, 2631, 16, 65536, 32, 32}, |
| 613 | {2677, Triple::AMDGPUSubArch1032, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 2}, 2631, 16, 65536, 32, 32}, |
| 614 | {2685, Triple::AMDGPUSubArch1033, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 3}, 2631, 16, 65536, 32, 32}, |
| 615 | {2693, Triple::AMDGPUSubArch1034, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 4}, 2631, 16, 65536, 32, 32}, |
| 616 | {2701, Triple::AMDGPUSubArch1035, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 5}, 2631, 16, 65536, 32, 32}, |
| 617 | {2709, Triple::AMDGPUSubArch1036, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 6}, 2631, 16, 65536, 32, 32}, |
| 618 | {2717, Triple::AMDGPUSubArch1100, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 0, 0}, 2725, 16, 65536, 32, 32}, |
| 619 | {2731, Triple::AMDGPUSubArch1101, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 0, 1}, 2725, 16, 65536, 32, 32}, |
| 620 | {2739, Triple::AMDGPUSubArch1102, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 0, 2}, 2725, 16, 65536, 32, 32}, |
| 621 | {2747, Triple::AMDGPUSubArch1103, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 0, 3}, 2725, 16, 65536, 32, 32}, |
| 622 | {2755, Triple::AMDGPUSubArch1150, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 5, 0}, 2725, 16, 65536, 32, 32}, |
| 623 | {2763, Triple::AMDGPUSubArch1151, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 5, 1}, 2725, 16, 65536, 32, 32}, |
| 624 | {2771, Triple::AMDGPUSubArch1152, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 5, 2}, 2725, 16, 65536, 32, 32}, |
| 625 | {2779, Triple::AMDGPUSubArch1153, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 5, 3}, 2725, 16, 65536, 32, 32}, |
| 626 | {2787, Triple::AMDGPUSubArch1154, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 5, 4}, 2725, 16, 65536, 32, 32}, |
| 627 | {2795, Triple::AMDGPUSubArch1170, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 7, 0}, 2725, 16, 65536, 32, 32}, |
| 628 | {2803, Triple::AMDGPUSubArch1171, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 7, 1}, 2725, 16, 65536, 32, 32}, |
| 629 | {2811, Triple::AMDGPUSubArch1172, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 7, 2}, 2725, 16, 65536, 32, 32}, |
| 630 | {2819, Triple::AMDGPUSubArch1200, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_SMEM_PREFETCH_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {12, 0, 0}, 2827, 16, 65536, 32, 32}, |
| 631 | {2833, Triple::AMDGPUSubArch1201, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_SMEM_PREFETCH_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {12, 0, 1}, 2827, 16, 65536, 32, 32}, |
| 632 | {2841, Triple::AMDGPUSubArch1250S, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ADD_MIN_MAX_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ASYNC_STORE_FROM_LDS_INSTS, FEAT_ASYNCCNT, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_LDS_BARRIER_ARRIVE_ATOMIC, FEAT_LERP_INST, FEAT_MCAST_LOAD_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PK_ADD_MIN_MAX_INSTS, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SWMMAC_GFX1250_INSTS, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SAD_INSTS, FEAT_SETPRIO_INC_WG_INST, FEAT_SMEM_PREFETCH_INSTS, FEAT_TDM_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_TRANSPOSE_LOAD_F4F6_INSTS, FEAT_VMEM_PREF_INSTS, FEAT_WMMA_N16_INSTS, FEAT_WAVEFRONTSIZE32, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1024_ADDRESSABLE_VGPRS, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_2048, FEAT_LDS_ENCODING_GRANULARITY_2048}), {12, 5, 0}, 2827, 16, 327680, 64, 45}, |
| 633 | {2856, Triple::AMDGPUSubArch1250, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ADD_MIN_MAX_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ASYNC_STORE_FROM_LDS_INSTS, FEAT_ASYNCCNT, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_BLOCK16_CVT_SCALE_INSTS, FEAT_FP8E5M3_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_LDS_BARRIER_ARRIVE_ATOMIC, FEAT_LERP_INST, FEAT_MCAST_LOAD_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PK_ADD_MIN_MAX_INSTS, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SWMMAC_GFX1250_INSTS, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SAD_INSTS, FEAT_SETPRIO_INC_WG_INST, FEAT_SMEM_PREFETCH_INSTS, FEAT_TDM_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_TRANSPOSE_LOAD_F4F6_INSTS, FEAT_VMEM_PREF_INSTS, FEAT_WMMA_N16_INSTS, FEAT_WMMA_F4_INSTS, FEAT_WMMA_16X16X128FP8_INSTS, FEAT_WAVEFRONTSIZE32, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1024_ADDRESSABLE_VGPRS, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_2048, FEAT_LDS_ENCODING_GRANULARITY_2048}), {12, 5, 0}, 2827, 16, 327680, 64, 45}, |
| 634 | {2864, Triple::AMDGPUSubArch1251, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ADD_MIN_MAX_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ASYNC_STORE_FROM_LDS_INSTS, FEAT_ASYNCCNT, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_BLOCK16_CVT_SCALE_INSTS, FEAT_FP8E5M3_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX1251_GEMM_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_LDS_BARRIER_ARRIVE_ATOMIC, FEAT_LERP_INST, FEAT_MCAST_LOAD_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PK_ADD_MIN_MAX_INSTS, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SWMMAC_GFX1250_INSTS, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SAD_INSTS, FEAT_SETPRIO_INC_WG_INST, FEAT_SMEM_PREFETCH_INSTS, FEAT_TDM_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_TRANSPOSE_LOAD_F4F6_INSTS, FEAT_VMEM_PREF_INSTS, FEAT_WMMA_N16_INSTS, FEAT_WMMA_F4_INSTS, FEAT_WMMA_16X16X128FP8_INSTS, FEAT_WAVEFRONTSIZE32, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1024_ADDRESSABLE_VGPRS, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_2048, FEAT_LDS_ENCODING_GRANULARITY_2048}), {12, 5, 1}, 2827, 16, 327680, 64, 45}, |
| 635 | {2872, Triple::AMDGPUSubArch1310, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_EXCLUSIVE_SCAN_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_F16BF16_TO_FP6BF6_CVT_SCALE_INSTS, FEAT_F32_TO_FP6BF6_CVT_SCALE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_BLOCK16_CVT_SCALE_INSTS, FEAT_FP6BF6_TO_F16BF16F32_CVT_SCALE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX12_INSTS, FEAT_GFX13_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SAD_INSTS, FEAT_SMEM_PREFETCH_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_WAVE_MATCH_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_1024}), {13, 1, 0}, 2880, 16, 196608, 32, 45}, |
| 636 | {2886, Triple::AMDGPUSubArch9, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {9, 0, 0}, 2555, 10, 65536, 32, 32}, |
| 637 | {2899, Triple::AMDGPUSubArch9_4, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BUFFER_INV_INST, FEAT_CI_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT3_INSTS, FEAT_DOT4_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX8_INSTS, FEAT_GFX90A_INSTS, FEAT_GFX940_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_LERP_INST, FEAT_MAI_INSTS, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_WAVEFRONTSIZE64, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_AGPR_ALLOC, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1280, FEAT_LDS_ENCODING_GRANULARITY_1280}), {9, 4, 0}, 2555, 8, 65536, 32, 32}, |
| 638 | {2914, Triple::AMDGPUSubArch10_1, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_XNACK_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_512, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 1, 0}, 2631, 20, 65536, 32, 32}, |
| 639 | {2930, Triple::AMDGPUSubArch10_3, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT1_INSTS, FEAT_DOT2_INSTS, FEAT_DOT5_INSTS, FEAT_DOT6_INSTS, FEAT_DOT7_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_S_MEMREALTIME, FEAT_S_MEMTIME_INST, FEAT_SAD_INSTS, FEAT_VMEM_TO_LDS_LOAD_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {10, 3, 0}, 2631, 16, 65536, 32, 32}, |
| 640 | {2946, Triple::AMDGPUSubArch11, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT12_INSTS, FEAT_DOT5_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_256B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 0, 3}, 2725, 16, 65536, 32, 32}, |
| 641 | {2960, Triple::AMDGPUSubArch11_7, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_GWS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {11, 7, 0}, 2725, 16, 65536, 32, 32}, |
| 642 | {2976, Triple::AMDGPUSubArch12, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BVH_RAY_TRACING_INSTS, FEAT_CI_INSTS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT10_INSTS, FEAT_DOT11_INSTS, FEAT_DOT12_INSTS, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_DOT9_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_QSAD_INSTS, FEAT_SWMMAC_GFX1200_INSTS, FEAT_SAD_INSTS, FEAT_SMEM_PREFETCH_INSTS, FEAT_WMMA_128B_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1536_PHYSICAL_VGPRS, FEAT_HALF_ADDRESSABLE_PHYSICAL_LOCAL_MEMORY, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_512}), {12, 0, 0}, 2827, 16, 65536, 32, 32}, |
| 643 | {2990, Triple::AMDGPUSubArch12_5, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ADD_MIN_MAX_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ASYNC_STORE_FROM_LDS_INSTS, FEAT_ASYNCCNT, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_BLOCK16_CVT_SCALE_INSTS, FEAT_FP8E5M3_INSTS, FEAT_FLAT_BUFFER_GLOBAL_FADD_F64_INST, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX12_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_LDS_BARRIER_ARRIVE_ATOMIC, FEAT_MCAST_LOAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PK_ADD_MIN_MAX_INSTS, FEAT_PRNG_INST, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SETPRIO_INC_WG_INST, FEAT_SMEM_PREFETCH_INSTS, FEAT_TDM_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_TRANSPOSE_LOAD_F4F6_INSTS, FEAT_VMEM_PREF_INSTS, FEAT_WAVEFRONTSIZE32, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_XNACK_SUPPORT, FEAT_SRAMECC_SUPPORT, FEAT_SRAMECC_ON_OFF_MODES, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_1024_ADDRESSABLE_VGPRS, FEAT_BACK_OFF_BARRIER, FEAT_LDS_ALLOC_GRANULARITY_2048, FEAT_LDS_ENCODING_GRANULARITY_2048}), {12, 5, 0}, 2827, 16, 327680, 64, 45}, |
| 644 | {3006, Triple::AMDGPUSubArch13, AMDGPUFeatureBitset({FEAT_16_BIT_INSTS, FEAT_ASHR_PK_INSTS, FEAT_ASYNC_LOAD_TO_LDS_INSTS, FEAT_ATOMIC_BUFFER_GLOBAL_PK_ADD_F16_INSTS, FEAT_ATOMIC_BUFFER_PK_ADD_BF16_INST, FEAT_ATOMIC_DS_PK_ADD_16_INSTS, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F32, FEAT_ATOMIC_FMIN_FMAX_GLOBAL_F64, FEAT_ATOMIC_FADD_NO_RTN_INSTS, FEAT_ATOMIC_FADD_RTN_INSTS, FEAT_ATOMIC_FLAT_PK_ADD_16_INSTS, FEAT_ATOMIC_GLOBAL_PK_ADD_BF16_INST, FEAT_BF16_CVT_INSTS, FEAT_BF16_PK_INSTS, FEAT_BF16_TRANS_INSTS, FEAT_BITOP3_INSTS, FEAT_CI_INSTS, FEAT_CLUSTERS, FEAT_CUBE_INSTS, FEAT_CVT_PKNORM_VOP2_INSTS, FEAT_CVT_SR_PK_BF16_F32_INST, FEAT_DL_INSTS, FEAT_DPP, FEAT_DOT7_INSTS, FEAT_DOT8_INSTS, FEAT_EXCLUSIVE_SCAN_INSTS, FEAT_EXTENDED_IMAGE_INSTS, FEAT_F16BF16_TO_FP6BF6_CVT_SCALE_INSTS, FEAT_F32_TO_FP6BF6_CVT_SCALE_INSTS, FEAT_FP8_CONVERSION_INSTS, FEAT_BLOCK16_CVT_SCALE_INSTS, FEAT_FP6BF6_TO_F16BF16F32_CVT_SCALE_INSTS, FEAT_FLAT_GLOBAL_INSTS, FEAT_GFX10_INSTS, FEAT_GFX10_3_INSTS, FEAT_GFX11_INSTS, FEAT_GFX1250_INSTS, FEAT_GFX12_INSTS, FEAT_GFX13_INSTS, FEAT_GFX8_INSTS, FEAT_GFX9_INSTS, FEAT_IMAGE_INSTS, FEAT_LERP_INST, FEAT_MQSAD_INSTS, FEAT_MQSAD_PK_INSTS, FEAT_MSAD_INSTS, FEAT_OCP_FP8_CONVERSION_INSTS, FEAT_PERMLANE16_INSTS, FEAT_PERMLANE16_SWAP, FEAT_PRNG_INST, FEAT_QSAD_INSTS, FEAT_S_WAKEUP_BARRIER_INST, FEAT_SAD_INSTS, FEAT_SMEM_PREFETCH_INSTS, FEAT_TANH_INSTS, FEAT_TENSOR_CVT_LUT_INSTS, FEAT_WAVE_MATCH_INSTS, FEAT_SUPPORTS_WGP, FEAT_SUPPORTS_WAVE32, FEAT_FAST_FMAF, FEAT_FAST_DENORMAL_F32, FEAT_APERTURE_REGS, FEAT_GET_DOORBELL_ID, FEAT_LDS_ALLOC_GRANULARITY_1024, FEAT_LDS_ENCODING_GRANULARITY_1024}), {13, 1, 0}, 2880, 16, 196608, 32, 45}, |
| 645 | }; |
| 646 | #endif // GET_AMDGPU_GPU_TABLE |
| 647 | |
| 648 | #ifdef GET_AMDGPU_FEATURE_NAME_TABLE |
| 649 | #undef GET_AMDGPU_FEATURE_NAME_TABLE |
| 650 | static constexpr StringTable::Offset AMDGPUFeatureNames[] = { |
| 651 | 1, |
| 652 | 14, |
| 653 | 32, |
| 654 | 46, |
| 655 | 70, |
| 656 | 97, |
| 657 | 106, |
| 658 | 144, |
| 659 | 175, |
| 660 | 201, |
| 661 | 229, |
| 662 | 257, |
| 663 | 282, |
| 664 | 304, |
| 665 | 332, |
| 666 | 363, |
| 667 | 378, |
| 668 | 392, |
| 669 | 409, |
| 670 | 429, |
| 671 | 451, |
| 672 | 464, |
| 673 | 480, |
| 674 | 489, |
| 675 | 498, |
| 676 | 509, |
| 677 | 531, |
| 678 | 555, |
| 679 | 564, |
| 680 | 568, |
| 681 | 580, |
| 682 | 592, |
| 683 | 604, |
| 684 | 616, |
| 685 | 627, |
| 686 | 638, |
| 687 | 649, |
| 688 | 660, |
| 689 | 671, |
| 690 | 682, |
| 691 | 693, |
| 692 | 704, |
| 693 | 715, |
| 694 | 736, |
| 695 | 757, |
| 696 | 791, |
| 697 | 819, |
| 698 | 849, |
| 699 | 869, |
| 700 | 892, |
| 701 | 913, |
| 702 | 933, |
| 703 | 957, |
| 704 | 971, |
| 705 | 981, |
| 706 | 1018, |
| 707 | 1051, |
| 708 | 1069, |
| 709 | 1081, |
| 710 | 1095, |
| 711 | 1107, |
| 712 | 1121, |
| 713 | 1140, |
| 714 | 1152, |
| 715 | 1164, |
| 716 | 1175, |
| 717 | 1188, |
| 718 | 1201, |
| 719 | 1214, |
| 720 | 1225, |
| 721 | 1229, |
| 722 | 1241, |
| 723 | 1267, |
| 724 | 1277, |
| 725 | 1287, |
| 726 | 1304, |
| 727 | 1316, |
| 728 | 1331, |
| 729 | 1342, |
| 730 | 1367, |
| 731 | 1384, |
| 732 | 1400, |
| 733 | 1416, |
| 734 | 1437, |
| 735 | 1447, |
| 736 | 1458, |
| 737 | 1472, |
| 738 | 1487, |
| 739 | 1508, |
| 740 | 1529, |
| 741 | 1551, |
| 742 | 1561, |
| 743 | 1581, |
| 744 | 1601, |
| 745 | 1611, |
| 746 | 1622, |
| 747 | 1643, |
| 748 | 1669, |
| 749 | 1692, |
| 750 | 1708, |
| 751 | 1725, |
| 752 | 1741, |
| 753 | 1757, |
| 754 | 1772, |
| 755 | 1786, |
| 756 | 1810, |
| 757 | 1821, |
| 758 | 1837, |
| 759 | 1853, |
| 760 | 1866, |
| 761 | 1882, |
| 762 | 1892, |
| 763 | 1910, |
| 764 | 1924, |
| 765 | 1940, |
| 766 | 1959, |
| 767 | 1980, |
| 768 | 1994, |
| 769 | 2008, |
| 770 | 2024, |
| 771 | 2035, |
| 772 | 2055, |
| 773 | 2078, |
| 774 | 2117, |
| 775 | 2134, |
| 776 | 2160, |
| 777 | 2186, |
| 778 | 2213, |
| 779 | 2240, |
| 780 | 2267, |
| 781 | 2296, |
| 782 | 2325, |
| 783 | 2355, |
| 784 | 2385, |
| 785 | }; |
| 786 | #endif // GET_AMDGPU_FEATURE_NAME_TABLE |
| 787 | |
| 788 | #ifdef GET_AMDGPU_GPU_ALIAS_TABLE |
| 789 | #undef GET_AMDGPU_GPU_ALIAS_TABLE |
| 790 | static constexpr GPUNameAlias AMDGPUGPUAliases[] = { |
| 791 | {3020, GK_GFX600}, |
| 792 | {3027, GK_GFX601}, |
| 793 | {3036, GK_GFX601}, |
| 794 | {3042, GK_GFX602}, |
| 795 | {3049, GK_GFX602}, |
| 796 | {3055, GK_GFX700}, |
| 797 | {3062, GK_GFX701}, |
| 798 | {3069, GK_GFX703}, |
| 799 | {3076, GK_GFX703}, |
| 800 | {3084, GK_GFX704}, |
| 801 | {3092, GK_GFX801}, |
| 802 | {3100, GK_GFX802}, |
| 803 | {3108, GK_GFX802}, |
| 804 | {3114, GK_GFX803}, |
| 805 | {3119, GK_GFX803}, |
| 806 | {3129, GK_GFX803}, |
| 807 | {3139, GK_GFX805}, |
| 808 | {3148, GK_GFX810}, |
| 809 | }; |
| 810 | #endif // GET_AMDGPU_GPU_ALIAS_TABLE |
| 811 | |
| 812 | #ifdef GET_AMDGPU_SUBARCH_NAME |
| 813 | #undef GET_AMDGPU_SUBARCH_NAME |
| 814 | static constexpr StringTable::Offset AMDGPUNoSubArchNameOffset = 3881; |
| 815 | struct AMDGPUSubArchNameEntry { |
| 816 | Triple::SubArchType SubArch; |
| 817 | StringTable::Offset NameOffset; |
| 818 | StringTable::Offset TripleNameOffset; |
| 819 | }; |
| 820 | static constexpr AMDGPUSubArchNameEntry AMDGPUSubArchNames[] = { |
| 821 | {Triple::AMDGPUSubArch600, 2435, 3155}, |
| 822 | {Triple::AMDGPUSubArch601, 2447, 3166}, |
| 823 | {Triple::AMDGPUSubArch602, 2454, 3177}, |
| 824 | {Triple::AMDGPUSubArch700, 2461, 3188}, |
| 825 | {Triple::AMDGPUSubArch701, 2473, 3199}, |
| 826 | {Triple::AMDGPUSubArch702, 2480, 3210}, |
| 827 | {Triple::AMDGPUSubArch703, 2487, 3221}, |
| 828 | {Triple::AMDGPUSubArch704, 2494, 3232}, |
| 829 | {Triple::AMDGPUSubArch705, 2501, 3243}, |
| 830 | {Triple::AMDGPUSubArch801, 2508, 3254}, |
| 831 | {Triple::AMDGPUSubArch802, 2520, 3265}, |
| 832 | {Triple::AMDGPUSubArch803, 2527, 3276}, |
| 833 | {Triple::AMDGPUSubArch805, 2534, 3287}, |
| 834 | {Triple::AMDGPUSubArch810, 2541, 3298}, |
| 835 | {Triple::AMDGPUSubArch900, 2548, 3309}, |
| 836 | {Triple::AMDGPUSubArch902, 2560, 3320}, |
| 837 | {Triple::AMDGPUSubArch904, 2567, 3331}, |
| 838 | {Triple::AMDGPUSubArch906, 2574, 3342}, |
| 839 | {Triple::AMDGPUSubArch908, 2581, 3353}, |
| 840 | {Triple::AMDGPUSubArch909, 2588, 3364}, |
| 841 | {Triple::AMDGPUSubArch90A, 2595, 3375}, |
| 842 | {Triple::AMDGPUSubArch90C, 2602, 3386}, |
| 843 | {Triple::AMDGPUSubArch942, 2609, 3397}, |
| 844 | {Triple::AMDGPUSubArch950, 2616, 3408}, |
| 845 | {Triple::AMDGPUSubArch9, 2886, 3419}, |
| 846 | {Triple::AMDGPUSubArch9_4, 2899, 3427}, |
| 847 | {Triple::AMDGPUSubArch1010, 2623, 3437}, |
| 848 | {Triple::AMDGPUSubArch1011, 2637, 3449}, |
| 849 | {Triple::AMDGPUSubArch1012, 2645, 3461}, |
| 850 | {Triple::AMDGPUSubArch1013, 2653, 3473}, |
| 851 | {Triple::AMDGPUSubArch1030, 2661, 3485}, |
| 852 | {Triple::AMDGPUSubArch1031, 2669, 3497}, |
| 853 | {Triple::AMDGPUSubArch1032, 2677, 3509}, |
| 854 | {Triple::AMDGPUSubArch1033, 2685, 3521}, |
| 855 | {Triple::AMDGPUSubArch1034, 2693, 3533}, |
| 856 | {Triple::AMDGPUSubArch1035, 2701, 3545}, |
| 857 | {Triple::AMDGPUSubArch1036, 2709, 3557}, |
| 858 | {Triple::AMDGPUSubArch10_1, 2914, 3569}, |
| 859 | {Triple::AMDGPUSubArch10_3, 2930, 3580}, |
| 860 | {Triple::AMDGPUSubArch1100, 2717, 3591}, |
| 861 | {Triple::AMDGPUSubArch1101, 2731, 3603}, |
| 862 | {Triple::AMDGPUSubArch1102, 2739, 3615}, |
| 863 | {Triple::AMDGPUSubArch1103, 2747, 3627}, |
| 864 | {Triple::AMDGPUSubArch1150, 2755, 3639}, |
| 865 | {Triple::AMDGPUSubArch1151, 2763, 3651}, |
| 866 | {Triple::AMDGPUSubArch1152, 2771, 3663}, |
| 867 | {Triple::AMDGPUSubArch1153, 2779, 3675}, |
| 868 | {Triple::AMDGPUSubArch1154, 2787, 3687}, |
| 869 | {Triple::AMDGPUSubArch1170, 2795, 3699}, |
| 870 | {Triple::AMDGPUSubArch1171, 2803, 3711}, |
| 871 | {Triple::AMDGPUSubArch1172, 2811, 3723}, |
| 872 | {Triple::AMDGPUSubArch11, 2946, 3735}, |
| 873 | {Triple::AMDGPUSubArch11_7, 2960, 3744}, |
| 874 | {Triple::AMDGPUSubArch1200, 2819, 3755}, |
| 875 | {Triple::AMDGPUSubArch1201, 2833, 3767}, |
| 876 | {Triple::AMDGPUSubArch12, 2976, 3779}, |
| 877 | {Triple::AMDGPUSubArch1250S, 2841, 3788}, |
| 878 | {Triple::AMDGPUSubArch1250, 2856, 3801}, |
| 879 | {Triple::AMDGPUSubArch1251, 2864, 3813}, |
| 880 | {Triple::AMDGPUSubArch12_5, 2990, 3825}, |
| 881 | {Triple::AMDGPUSubArch1310, 2872, 3836}, |
| 882 | {Triple::AMDGPUSubArch13, 3006, 3848}, |
| 883 | {Triple::AMDGPUSubArch6, 2435, 3857}, |
| 884 | {Triple::AMDGPUSubArch7, 2461, 3865}, |
| 885 | {Triple::AMDGPUSubArch8, 2508, 3873}, |
| 886 | }; |
| 887 | #endif // GET_AMDGPU_SUBARCH_NAME |
| 888 | |
| 889 | |