[llvm] AMDGPU: Only allow 31bit positive offset for scratch (PR #71895)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-amdgpu

Author: Ruiling, Song (ruiling)

<details>
<summary>Changes</summary>

This just helps code safer as how the value interpreted by hardware might change.

---
Full diff: https://github.com/llvm/llvm-project/pull/71895.diff


1 Files Affected:

- (modified) llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp (+1-1) 


``````````diff
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp b/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
index cd810f0b43e50db..014026dd209da25 100644
--- a/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
@@ -1859,7 +1859,7 @@ bool AMDGPUDAGToDAGISel::SelectScratchSVAddr(SDNode *N, SDValue Addr,
       std::tie(SplitImmOffset, RemainderOffset)
         = TII->splitFlatOffset(COffsetVal, AMDGPUAS::PRIVATE_ADDRESS, true);
 
-      if (isUInt<32>(RemainderOffset)) {
+      if (RemainderOffset > 0 && isInt<32>(RemainderOffset)) {
         SDNode *VMov = CurDAG->getMachineNode(
           AMDGPU::V_MOV_B32_e32, SL, MVT::i32,
           CurDAG->getTargetConstant(RemainderOffset, SDLoc(), MVT::i32));

``````````

</details>


https://github.com/llvm/llvm-project/pull/71895