diff options
| author | Jay Foad <jay.foad@amd.com> | 2025-07-23 11:35:57 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-07-23 11:35:57 +0100 |
| commit | 756ac65987b84b7427c25d76f069a04a4a817a5c (patch) | |
| tree | 7c3d20b61e8fcdafeff8123054e215d7ad70b5ab /llvm/test/CodeGen/AMDGPU | |
| parent | 33455825428f9e1b7998a66e228da7f6d483acf8 (diff) | |
| download | llvm-756ac65987b84b7427c25d76f069a04a4a817a5c.zip llvm-756ac65987b84b7427c25d76f069a04a4a817a5c.tar.gz llvm-756ac65987b84b7427c25d76f069a04a4a817a5c.tar.bz2 | |
[CodeGen] Add a pass for testing finalizeBundle (#149813)
This allows for unit testing of finalizeBundle with standard MIR tests
using update_mir_test_checks.py.
Diffstat (limited to 'llvm/test/CodeGen/AMDGPU')
| -rw-r--r-- | llvm/test/CodeGen/AMDGPU/finalizebundle.mir | 18 |
1 files changed, 18 insertions, 0 deletions
diff --git a/llvm/test/CodeGen/AMDGPU/finalizebundle.mir b/llvm/test/CodeGen/AMDGPU/finalizebundle.mir new file mode 100644 index 0000000..ea1ae04 --- /dev/null +++ b/llvm/test/CodeGen/AMDGPU/finalizebundle.mir @@ -0,0 +1,18 @@ +# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5 +# RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -passes=finalizebundle-test %s -o - | FileCheck %s + +--- +name: test_overlap +body: | + bb.0: + liveins: $vgpr0_vgpr1 + ; CHECK-LABEL: name: test_overlap + ; CHECK: liveins: $vgpr0_vgpr1 + ; CHECK-NEXT: {{ $}} + ; CHECK-NEXT: BUNDLE implicit-def $vgpr2_vgpr3, implicit-def $vgpr2, implicit-def $vgpr2_lo16, implicit-def $vgpr2_hi16, implicit-def $vgpr3, implicit-def $vgpr3_lo16, implicit-def $vgpr3_hi16, implicit-def $vgpr3_vgpr4, implicit-def $vgpr4, implicit-def $vgpr4_lo16, implicit-def $vgpr4_hi16, implicit $vgpr0_vgpr1, implicit $exec, implicit $vgpr1_vgpr2 { + ; CHECK-NEXT: $vgpr2_vgpr3 = V_LSHLREV_B64_pseudo_e32 1, $vgpr0_vgpr1, implicit $exec + ; CHECK-NEXT: $vgpr3_vgpr4 = V_LSHLREV_B64_pseudo_e32 1, $vgpr1_vgpr2, implicit $exec + ; CHECK-NEXT: } + $vgpr2_vgpr3 = V_LSHLREV_B64_pseudo_e32 1, $vgpr0_vgpr1, implicit $exec + $vgpr3_vgpr4 = V_LSHLREV_B64_pseudo_e32 1, $vgpr1_vgpr2, implicit $exec +... |