/external/llvm-project/llvm/test/CodeGen/AMDGPU/ |
D | atomicrmw-nand.ll | 21 ; GCN-NEXT: s_or_b64 s[4:5], vcc, s[4:5] 25 ; GCN-NEXT: s_or_b64 exec, exec, s[4:5] 49 ; GCN-NEXT: s_or_b64 s[4:5], vcc, s[4:5] 53 ; GCN-NEXT: s_or_b64 exec, exec, s[4:5] 78 ; GCN-NEXT: s_or_b64 s[4:5], vcc, s[4:5] 82 ; GCN-NEXT: s_or_b64 exec, exec, s[4:5]
|
D | collapse-endcf.ll | 11 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC]] 46 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC_INNER]] 49 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC_OUTER]] 95 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC_OUTER]] 138 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC_INNER_IF_OUTER_ELSE]] 150 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC_ELSE]] 151 ; GCN: s_or_b64 exec, exec, [[SAVEEXEC_OUTER3]] 195 ; GCN-NEXT: s_or_b64 exec, exec, [[SAVEEXEC]] 217 ; GCN: s_or_b64 exec, exec, [[SAVEEXEC_OUTER:s\[[0-9:]+\]]] 220 ; GCN: s_or_b64 exec, exec, s{{\[[0-9]+:[0-9]+\]}} [all …]
|
D | endcf-loop-header.ll | 4 ; loop block. This intrinsic will be lowered to s_or_b64 by the code 10 ; CHECK: s_or_b64 exec, exec 13 ; CHECK-NOT: s_or_b64 exec, exec
|
D | i1-copy-from-loop.ll | 14 ; SI-NEXT: s_or_b64 exec, exec, s[14:15] 18 ; SI-NEXT: s_or_b64 s[4:5], s[14:15], s[4:5] 21 ; SI-NEXT: s_or_b64 s[8:9], s[8:9], s[12:13] 26 ; SI-NEXT: s_or_b64 s[10:11], s[10:11], exec 46 ; SI-NEXT: s_or_b64 exec, exec, s[4:5]
|
D | loop_break.ll | 47 ; GCN-NEXT: s_or_b64 s[4:5], s[4:5], exec 57 ; GCN-NEXT: s_or_b64 s[4:5], s[4:5], s[8:9] 61 ; GCN-NEXT: s_or_b64 s[0:1], s[8:9], s[0:1] 126 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], s[8:9] 136 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], s[8:9] 141 ; GCN-NEXT: s_or_b64 s[0:1], s[8:9], s[0:1] 145 ; GCN-NEXT: s_or_b64 exec, exec, s[0:1] 221 ; GCN-NEXT: s_or_b64 s[4:5], s[4:5], s[8:9] 231 ; GCN-NEXT: s_or_b64 s[4:5], s[4:5], s[8:9] 236 ; GCN-NEXT: s_or_b64 s[0:1], s[8:9], s[0:1] [all …]
|
D | divergent-branch-uniform-condition.ll | 32 ; ISA-NEXT: s_or_b64 exec, exec, s[8:9] 38 ; ISA-NEXT: s_or_b64 s[2:3], s[10:11], s[2:3] 41 ; ISA-NEXT: s_or_b64 s[4:5], s[4:5], s[8:9] 46 ; ISA-NEXT: s_or_b64 s[6:7], s[6:7], exec 60 ; ISA-NEXT: s_or_b64 exec, exec, s[2:3] 66 ; ISA-NEXT: s_or_b64 exec, exec, s[0:1]
|
D | multilevel-break.ll | 54 ; GCN-NEXT: s_or_b64 exec, exec, s[4:5] 56 ; GCN-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1] 68 ; GCN-NEXT: s_or_b64 exec, exec, s[8:9] 70 ; GCN-NEXT: s_or_b64 s[4:5], s[8:9], s[4:5] 79 ; GCN-NEXT: s_or_b64 s[2:3], s[2:3], exec 80 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], exec 89 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], s[10:11] 191 ; GCN-NEXT: s_or_b64 s[0:1], s[6:7], s[0:1] 194 ; GCN-NEXT: s_or_b64 s[4:5], s[4:5], s[6:7] 242 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], s[10:11] [all …]
|
D | atomic_optimizations_pixelshader.ll | 34 ; GFX7-NEXT: s_or_b64 exec, exec, s[10:11] 39 ; GFX7-NEXT: s_or_b64 exec, exec, s[8:9] 67 ; GFX8-NEXT: s_or_b64 exec, exec, s[10:11] 72 ; GFX8-NEXT: s_or_b64 exec, exec, s[8:9] 100 ; GFX9-NEXT: s_or_b64 exec, exec, s[10:11] 105 ; GFX9-NEXT: s_or_b64 exec, exec, s[8:9] 134 ; GFX1064-NEXT: s_or_b64 exec, exec, s[28:29] 139 ; GFX1064-NEXT: s_or_b64 exec, exec, s[8:9] 247 ; GFX8-NEXT: s_or_b64 exec, exec, s[10:11] 253 ; GFX8-NEXT: s_or_b64 exec, exec, s[8:9] [all …]
|
D | loop_exit_with_xor.ll | 11 ; GCN: s_or_b64 [[REG3:[^ ,]*]], [[REG2]], 41 ; GCN: s_or_b64 [[REG2:[^ ,]*]], [[REG1]], 66 ; GCN: s_or_b64 [[REG3]], [[REG2]],
|
D | si-annotate-cf.ll | 19 ; SI-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1] 23 ; SI-NEXT: s_or_b64 exec, exec, s[0:1] 43 ; FLAT-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1] 47 ; FLAT-NEXT: s_or_b64 exec, exec, s[0:1] 82 ; SI-NEXT: s_or_b64 exec, exec, s[6:7] 86 ; SI-NEXT: s_or_b64 s[2:3], s[0:1], s[2:3] 106 ; FLAT-NEXT: s_or_b64 exec, exec, s[6:7] 110 ; FLAT-NEXT: s_or_b64 s[2:3], s[0:1], s[2:3]
|
D | shift-i128.ll | 195 ; GCN-NEXT: s_or_b64 s[10:11], s[0:1], s[10:11] 232 ; GCN-NEXT: s_or_b64 s[10:11], s[0:1], s[10:11] 274 ; GCN-NEXT: s_or_b64 s[6:7], s[6:7], s[0:1] 462 ; GCN-NEXT: s_or_b64 s[0:1], s[16:17], s[18:19] 464 ; GCN-NEXT: s_or_b64 s[6:7], s[24:25], s[6:7] 483 ; GCN-NEXT: s_or_b64 s[6:7], s[10:11], s[6:7] 485 ; GCN-NEXT: s_or_b64 s[2:3], s[20:21], s[22:23] 531 ; GCN-NEXT: s_or_b64 s[6:7], s[24:25], s[6:7] 533 ; GCN-NEXT: s_or_b64 s[0:1], s[16:17], s[18:19] 553 ; GCN-NEXT: s_or_b64 s[6:7], s[8:9], s[6:7] [all …]
|
D | select-opt.ll | 74 ; GCN: s_or_b64 vcc, vcc, [[CMP1]] 91 ; GCN: s_or_b64 vcc, vcc, [[CMP1]] 107 ; GCN: s_or_b64 vcc, vcc, [[CMP1]] 123 ; GCN: s_or_b64 vcc, vcc, [[CMP1]]
|
D | sgpr-control-flow.ll | 129 ; SI-NEXT: s_or_b64 exec, exec, s[2:3] 175 ; SI-NEXT: s_or_b64 s[0:1], s[0:1], s[8:9] 192 ; SI-NEXT: s_or_b64 s[0:1], s[0:1], s[6:7] 194 ; SI-NEXT: s_or_b64 exec, exec, s[2:3]
|
D | atomic_optimizations_local_pointer.ll | 38 ; GFX7LESS-NEXT: s_or_b64 exec, exec, s[4:5] 67 ; GFX8-NEXT: s_or_b64 exec, exec, s[4:5] 96 ; GFX9-NEXT: s_or_b64 exec, exec, s[4:5] 128 ; GFX1064-NEXT: s_or_b64 exec, exec, s[4:5] 200 ; GFX7LESS-NEXT: s_or_b64 exec, exec, s[0:1] 233 ; GFX8-NEXT: s_or_b64 exec, exec, s[6:7] 265 ; GFX9-NEXT: s_or_b64 exec, exec, s[6:7] 301 ; GFX1064-NEXT: s_or_b64 exec, exec, s[6:7] 409 ; GFX8-NEXT: s_or_b64 exec, exec, s[2:3] 459 ; GFX9-NEXT: s_or_b64 exec, exec, s[2:3] [all …]
|
D | bfi_int.ll | 128 ; GCN: s_or_b64 169 ; GCN: s_or_b64 171 ; GCN: s_or_b64
|
/external/llvm/test/CodeGen/AMDGPU/ |
D | endcf-loop-header.ll | 4 ; loop block. This intrinsic will be lowered to s_or_b64 by the code 10 ; CHECK: s_or_b64 exec, exec 13 ; CHECK-NOT: s_or_b64 exec, exec
|
D | si-lower-control-flow-unreachable-block.ll | 9 ; GCN: s_or_b64 exec, exec 34 ; GCN: s_or_b64 exec, exec
|
D | valu-i1.ll | 55 ; SI: s_or_b64 exec, exec, [[BR_SREG]] 139 ; SI: s_or_b64 [[TMP:s\[[0-9]+:[0-9]+\]]], [[CMP]], [[COND_STATE]] 142 ; SI: s_or_b64 exec, exec, [[ORNEG2]] 143 ; SI: s_or_b64 [[COND_STATE]], [[ORNEG2]], [[TMP]] 148 ; SI: s_or_b64 exec, exec, [[COND_STATE]]
|
D | cgp-addressing-modes.ll | 45 ; GCN: s_or_b64 exec 72 ; GCN: s_or_b64 exec 99 ; GCN: s_or_b64 exec 235 ; GCN: s_or_b64 exec, exec 264 ; GCN: s_or_b64 exec, exec 297 ; GCN: s_or_b64 exec, exec 329 ; GCN: s_or_b64 exec, exec 360 ; GCN: s_or_b64 exec, exec 390 ; GCN: s_or_b64 exec, exec 428 ; GCN: s_or_b64 exec, exec
|
D | si-annotate-cf.ll | 8 ; SI: s_or_b64 30 ; FIXME: This could be folded into the s_or_b64 instruction 36 ; SI: s_or_b64 [[BREAK:s\[[0-9]+:[0-9]+\]]], vcc, [[ZERO]]
|
D | si-annotate-cfg-loop-assert.ll | 6 ; CHECK s_or_b64 exec, exec
|
D | or.ll | 87 ; SI: s_or_b64 158 ; SI: s_or_b64 s[{{[0-9]+:[0-9]+}}], vcc, s[{{[0-9]+:[0-9]+}}] 171 ; SI: s_or_b64 s[{{[0-9]+:[0-9]+}}], vcc, s[{{[0-9]+:[0-9]+}}]
|
/external/mesa3d/src/amd/compiler/tests/ |
D | test_optimizer.cpp | 175 writeout(0, bld.sop2(aco_opcode::s_or_b64, bld.def(bld.lm), bld.def(s1, scc), 181 writeout(1, bld.sop2(aco_opcode::s_or_b64, bld.def(bld.lm), bld.def(s1, scc), 187 writeout(2, bld.sop2(aco_opcode::s_or_b64, bld.def(bld.lm), bld.def(s1, scc), 218 writeout(6, bld.sop2(aco_opcode::s_or_b64, bld.def(bld.lm), bld.def(s1, scc), 226 writeout(7, bld.sop2(aco_opcode::s_or_b64, bld.def(bld.lm), bld.def(s1, scc),
|
/external/llvm-project/llvm/test/CodeGen/AMDGPU/GlobalISel/ |
D | divergent-control-flow.ll | 17 ; CHECK-NEXT: s_or_b64 exec, exec, s[4:5] 44 ; CHECK-NEXT: s_or_b64 exec, exec, s[4:5] 73 ; CHECK-NEXT: s_or_b64 exec, exec, s[4:5] 104 ; CHECK-NEXT: s_or_b64 exec, exec, s[4:5] 211 ; CHECK-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1]
|
D | orn2.ll | 474 ; GFX6-NEXT: s_or_b64 s[0:1], s[0:1], s[2:3] 482 ; GFX9-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1] 509 ; GFX6-NEXT: s_or_b64 s[0:1], s[2:3], s[0:1] 517 ; GFX9-NEXT: s_or_b64 s[0:1], s[0:1], s[2:3] 544 ; GFX6-NEXT: s_or_b64 s[0:1], s[0:1], s[2:3] 552 ; GFX9-NEXT: s_or_b64 s[0:1], s[2:3], s[4:5] 591 ; GFX6-NEXT: s_or_b64 s[0:1], s[0:1], s[4:5] 592 ; GFX6-NEXT: s_or_b64 s[2:3], s[2:3], s[4:5] 600 ; GFX9-NEXT: s_or_b64 s[0:1], s[2:3], s[6:7] 601 ; GFX9-NEXT: s_or_b64 s[2:3], s[4:5], s[6:7]
|