[llvm] 2147e29 - [AMDGPU] Tests for unnecessary S_WAIT_XCNT insertion (#145688)
via llvm-commits
llvm-commits at lists.llvm.org
Wed Jul 23 05:24:06 PDT 2025
Author: Jay Foad
Date: 2025-07-23T13:24:03+01:00
New Revision: 2147e29f641f22822e2b0e10c978b28b06db1aeb
URL: https://github.com/llvm/llvm-project/commit/2147e29f641f22822e2b0e10c978b28b06db1aeb
DIFF: https://github.com/llvm/llvm-project/commit/2147e29f641f22822e2b0e10c978b28b06db1aeb.diff
LOG: [AMDGPU] Tests for unnecessary S_WAIT_XCNT insertion (#145688)
Hardware does an implicit "S_WAIT_XCNT 0" between SMEM and VMEM
instructions, so there will never be outstanding address translations
for both SMEM and VMEM at the same time.
Added:
Modified:
llvm/test/CodeGen/AMDGPU/wait-xcnt.mir
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/AMDGPU/wait-xcnt.mir b/llvm/test/CodeGen/AMDGPU/wait-xcnt.mir
index f4b95236b2fa1..af8b9e7c8cd82 100644
--- a/llvm/test/CodeGen/AMDGPU/wait-xcnt.mir
+++ b/llvm/test/CodeGen/AMDGPU/wait-xcnt.mir
@@ -966,3 +966,45 @@ body: |
$vgpr2 = V_MOV_B32_e32 $vgpr2, implicit $exec
$sgpr0 = S_MOV_B32 0
...
+
+# TODO: Unnecessary wait before overwriting vgpr0.
+---
+name: overwrite_vgpr_after_smem
+tracksRegLiveness: true
+machineFunctionInfo:
+ isEntryFunction: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1, $sgpr0_sgpr1
+ ; GCN-LABEL: name: overwrite_vgpr_after_smem
+ ; GCN: liveins: $vgpr0_vgpr1, $sgpr0_sgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: $vgpr2 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+ ; GCN-NEXT: $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+ ; GCN-NEXT: S_WAIT_XCNT 0
+ ; GCN-NEXT: $vgpr0 = V_MOV_B32_e32 0, implicit $exec
+ $vgpr2 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+ $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+ $vgpr0 = V_MOV_B32_e32 0, implicit $exec
+...
+
+# TODO: Unnecessary wait before overwriting sgpr0.
+---
+name: overwrite_sgpr_after_vmem
+tracksRegLiveness: true
+machineFunctionInfo:
+ isEntryFunction: true
+body: |
+ bb.0:
+ liveins: $vgpr0_vgpr1, $sgpr0_sgpr1
+ ; GCN-LABEL: name: overwrite_sgpr_after_vmem
+ ; GCN: liveins: $vgpr0_vgpr1, $sgpr0_sgpr1
+ ; GCN-NEXT: {{ $}}
+ ; GCN-NEXT: $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+ ; GCN-NEXT: $vgpr2 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+ ; GCN-NEXT: S_WAIT_XCNT 0
+ ; GCN-NEXT: $sgpr0 = S_MOV_B32 0
+ $sgpr2 = S_LOAD_DWORD_IMM $sgpr0_sgpr1, 0, 0
+ $vgpr2 = GLOBAL_LOAD_DWORD $vgpr0_vgpr1, 0, 0, implicit $exec
+ $sgpr0 = S_MOV_B32 0
+...
More information about the llvm-commits
mailing list