Tail duplication can mix incompatible registers in phi nodes

Do not tail duplicate blocks where the successor has a phi node,
and the corresponding value in that phi node uses a subregister.

http://reviews.llvm.org/D13922

llvm-svn: 250877

diff --git a/llvm/lib/CodeGen/TailDuplication.cpp b/llvm/lib/CodeGen/TailDuplication.cpp
index c2e8b93..ff86dab 100644 (file)
--- a/llvm/lib/CodeGen/TailDuplication.cpp
+++ b/llvm/lib/CodeGen/TailDuplication.cpp
@@ -607,6 +607,27 @@ TailDuplicatePass::shouldTailDuplicate(const MachineFunction &MF,
       return false;
   }
 
+  // Check if any of the successors of TailBB has a PHI node in which the
+  // value corresponding to TailBB uses a subregister.
+  // If a phi node uses a register paired with a subregister, the actual
+  // "value type" of the phi may differ from the type of the register without
+  // any subregisters. Due to a bug, tail duplication may add a new operand
+  // without a necessary subregister, producing an invalid code. This is
+  // demonstrated by test/CodeGen/Hexagon/tail-dup-subreg-abort.ll.
+  // Disable tail duplication for this case for now, until the problem is
+  // fixed.
+  for (auto SB : TailBB.successors()) {
+    for (auto &I : *SB) {
+      if (!I.isPHI())
+        break;
+      unsigned Idx = getPHISrcRegOpIdx(&I, &TailBB);
+      assert(Idx != 0);
+      MachineOperand &PU = I.getOperand(Idx);
+      if (PU.getSubReg() != 0)
+        return false;
+    }
+  }
+
   if (HasIndirectbr && PreRegAlloc)
     return true;
 

diff --git a/llvm/test/CodeGen/Hexagon/tail-dup-subreg-abort.ll b/llvm/test/CodeGen/Hexagon/tail-dup-subreg-abort.ll
new file mode 100644 (file)
index 0000000..82dae2c
--- /dev/null
+++ b/llvm/test/CodeGen/Hexagon/tail-dup-subreg-abort.ll
@@ -0,0 +1,28 @@
+; RUN: llc -march=hexagon -O2 -disable-cgp < %s
+; REQUIRES: asserts
+;
+; Tail duplication can ignore subregister information on PHI nodes, and as
+; a result, generate COPY instructions between registers of different classes.
+; This could lead to HexagonInstrInfo::copyPhysReg aborting on an unhandled
+; src/dst combination.
+;
+define i32 @foo(i32 %x, i64 %y) nounwind {
+entry:
+  %a = icmp slt i32 %x, 0
+  %lo = trunc i64 %y to i32
+  br i1 %a, label %next, label %tail
+tail:
+  br label %join
+next:
+  %c = icmp eq i32 %x, 0
+  br i1 %c, label %b1, label %tail
+b1:
+  %t1 = lshr i64 %y, 32
+  %hi = trunc i64 %t1 to i32
+  br label %join
+join:
+  %val = phi i32 [ %hi, %b1 ], [ %lo, %tail ]
+  ret i32 %val
+}
+
+