[llvm] Skip debug machine instructions in AMDGPU SIWholeQuadMode pass. (PR #134518)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-amdgpu

Author: None (kpyzhov)

<details>
<summary>Changes</summary>



---
Full diff: https://github.com/llvm/llvm-project/pull/134518.diff


1 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp (+6) 


``````````diff
diff --git a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
index 57ae7d63218dd..8ae134a7b8091 100644
--- a/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
+++ b/llvm/lib/Target/AMDGPU/SIWholeQuadMode.cpp
@@ -1300,6 +1300,12 @@ void SIWholeQuadMode::processBlock(MachineBasicBlock &MBB, BlockInfo &BI,
 
   for (unsigned Idx = 0;; ++Idx) {
     MachineBasicBlock::iterator Next = II;
+
+    if (II != IE && II->isDebugInstr()) {
+      II = ++Next;
+      continue;
+    }
+
     char Needs = StateExact | StateWQM; // Strict mode is disabled by default.
     char OutNeeds = 0;
 

``````````

</details>


https://github.com/llvm/llvm-project/pull/134518