[llvm] [AMDGPU] Rework dot4 signedness checks (PR #68757)

[Matt Arsenault via llvm-commits]

================
@@ -13056,32 +13023,74 @@ static bool isMul(const SDValue Op) {
           Opcode == AMDGPUISD::MUL_I24);
 }
 
-static std::optional<bool> checkSignedness(const SDValue &N,
-                                           ByteProvider<SDValue> &Src0,
-                                           ByteProvider<SDValue> &Src1) {
+static std::optional<bool>
+checkSignedness(const SDValue &N, ByteProvider<SDValue> &Src0,
+                ByteProvider<SDValue> &Src1, const SDValue &S0Op,
+                const SDValue &S1Op, const SelectionDAG &DAG) {
   auto MulOpcode = N.getOpcode();
-  std::optional<bool> IterIsSigned;
-  // Both sides of the tree must have the same signedness semantics.
-  if ((Src0.IsSigned != Src1.IsSigned) ||
-      (Src0.IsSigned.value_or(false) != Src1.IsSigned.value_or(false)))
-    return IterIsSigned;
-  // If we have a MUL_U24 op with signed semantics, then fail.
-  if (Src0.IsSigned.value_or(false) && MulOpcode == AMDGPUISD::MUL_U24)
-    return IterIsSigned;
-  // If we have a MUL_I24 op with unsigned semantics, then fail.
-  if (!Src0.IsSigned.value_or(true) && MulOpcode == AMDGPUISD::MUL_I24)
-    return IterIsSigned;
-
-  bool TopLevelSignedness =
-      MulOpcode == AMDGPUISD::MUL_I24 ||
-      (MulOpcode == ISD::MUL && N.getNode()->getFlags().hasNoSignedWrap() &&
-       !N.getNode()->getFlags().hasNoUnsignedWrap());
-
-  // In cases where we are accumulating into an i8 (for v_dot4), the
-  // ByteProvider will not have signedness info since the MSBs are dont-cares.
-  // In this case, we simply use the TopLevelSignedness of the instruction.
-  IterIsSigned = Src0.IsSigned.value_or(TopLevelSignedness);
-  return IterIsSigned;
+
+  // We have previously matched the signedness semantics of dot4 for this
+  // iteration via ValueTracking
+  if (MulOpcode == AMDGPUISD::MUL_U24 || MulOpcode == AMDGPUISD::MUL_I24)
+    return MulOpcode == AMDGPUISD::MUL_I24;
----------------
arsenm wrote:

This should give the same result if you just let it fall through the computeKnownBits handling?

https://github.com/llvm/llvm-project/pull/68757