[AMDGPU] Small correction in SITargetLowering::performOrCombine().

Differential Revision: https://reviews.llvm.org/D113203

diff --git a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
index 877b4500364e8f3df17ec2ca2d313ce8b526f863..519c5b936536127d8a89be400b874bae900df35c 100644 (file)
--- a/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/SIISelLowering.cpp
@@ -9484,7 +9484,8 @@ SDValue SITargetLowering::performOrCombine(SDNode *N,
   const ConstantSDNode *CRHS = dyn_cast<ConstantSDNode>(N->getOperand(1));
   if (CRHS) {
     if (SDValue Split
-          = splitBinaryBitConstantOp(DCI, SDLoc(N), ISD::OR, LHS, CRHS))
+          = splitBinaryBitConstantOp(DCI, SDLoc(N), ISD::OR,
+                                     N->getOperand(0), CRHS))
       return Split;
   }
 

diff --git a/llvm/test/CodeGen/AMDGPU/combine-reg-or-const.ll b/llvm/test/CodeGen/AMDGPU/combine-reg-or-const.ll
new file mode 100644 (file)
index 0000000..65ffba8
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/combine-reg-or-const.ll
@@ -0,0 +1,31 @@
+; RUN: llc  -mtriple=amdgcn-amd-amdhsa -o - %s | FileCheck %s
+
+; The OR instruction should not be eliminated by the "OR Combine" DAG optimization.
+
+; CHECK-LABEL: _Z11test_kernelPii:
+; CHECK: s_mul_i32
+; CHECK: s_sub_i32
+; CHECK: s_and_b32 [[S1:s[0-9]+]], {{s[0-9]+}}, {{s[0-9]+}}
+; CHECK: s_add_i32 [[S2:s[0-9]+]], {{s[0-9]+}}, [[S1]]
+; CHECK: s_or_b32 {{s[0-9]+}}, [[S2]], 0xc0
+
+define protected amdgpu_kernel void @_Z11test_kernelPii(i32 addrspace(1)* nocapture %Ad.coerce, i32 %s) local_unnamed_addr #5 {
+entry:
+  %rem.lhs.trunc = trunc i32 %s to i16
+  %rem4 = urem i16 %rem.lhs.trunc, 12
+  %rem.zext = zext i16 %rem4 to i32
+  %cmp = icmp eq i32 %s, 3
+  br i1 %cmp, label %if.then, label %if.end
+
+if.then:                                          ; preds = %entry
+  %idxprom = zext i32 %s to i64
+  %arrayidx3 = getelementptr inbounds i32, i32 addrspace(1)* %Ad.coerce, i64 %idxprom
+  %div = lshr i32 %rem.zext, 3
+  %or = or i32 %rem.zext, 192
+  %add = add nuw nsw i32 %or, %div
+  store i32 %add, i32 addrspace(1)* %arrayidx3, align 4
+  br label %if.end
+
+if.end:                                           ; preds = %if.then, %entry
+  ret void
+}