Reland "[LoongArch] Add isSafeToMove hook to prevent unsafe instruction motion" (#167465)

heiher · dyung · commit a832a5222e48 · 2025-11-13T09:22:08.000Z
This patch introduces a new virtual method `TargetInstrInfo::isSafeToMove()` to allow backends to control whether a machine instruction can be safely moved by optimization passes. The `BranchFolder` pass now respects this hook when hoisting common code. By default, all instructions are considered safe to to move. For LoongArch, `isSafeToMove()` is overridden to prevent relocation-related instruction sequences (e.g. PC-relative addressing and calls) from being broken by instruction motion. Correspondingly, `isSchedulingBoundary()` is updated to reuse this logic for consistency. Relands #163725 (cherry picked from commit ea10026)
diff --git a/llvm/include/llvm/CodeGen/TargetInstrInfo.h b/llvm/include/llvm/CodeGen/TargetInstrInfo.h
@@ -1730,6 +1730,17 @@ class LLVM_ABI TargetInstrInfo : public MCInstrInfo {
     return true;
   }
 
+  /// Return true if it's safe to move a machine instruction.
+  /// This allows the backend to prevent certain special instruction
+  /// sequences from being broken by instruction motion in optimization
+  /// passes.
+  /// By default, this returns true for every instruction.
+  virtual bool isSafeToMove(const MachineInstr &MI,
+                            const MachineBasicBlock *MBB,
+                            const MachineFunction &MF) const {
+    return true;
+  }
+
   /// Test if the given instruction should be considered a scheduling boundary.
   /// This primarily includes labels and terminators.
   virtual bool isSchedulingBoundary(const MachineInstr &MI,
diff --git a/llvm/lib/CodeGen/BranchFolding.cpp b/llvm/lib/CodeGen/BranchFolding.cpp
@@ -1971,6 +1971,7 @@ bool BranchFolder::HoistCommonCodeInSuccs(MachineBasicBlock *MBB) {
   MachineBasicBlock::iterator FIB = FBB->begin();
   MachineBasicBlock::iterator TIE = TBB->end();
   MachineBasicBlock::iterator FIE = FBB->end();
+  MachineFunction &MF = *TBB->getParent();
   while (TIB != TIE && FIB != FIE) {
     // Skip dbg_value instructions. These do not count.
     TIB = skipDebugInstructionsForward(TIB, TIE, false);
@@ -1985,6 +1986,10 @@ bool BranchFolder::HoistCommonCodeInSuccs(MachineBasicBlock *MBB) {
       // Hard to reason about register liveness with predicated instruction.
       break;
 
+    if (!TII->isSafeToMove(*TIB, TBB, MF))
+      // Don't hoist the instruction if it isn't safe to move.
+      break;
+
     bool IsSafe = true;
     for (MachineOperand &MO : TIB->operands()) {
       // Don't attempt to hoist instructions with register masks.
diff --git a/llvm/lib/Target/LoongArch/LoongArchInstrInfo.cpp b/llvm/lib/Target/LoongArch/LoongArchInstrInfo.cpp
@@ -378,12 +378,9 @@ bool LoongArchInstrInfo::isBranchOffsetInRange(unsigned BranchOp,
   }
 }
 
-bool LoongArchInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
-                                              const MachineBasicBlock *MBB,
-                                              const MachineFunction &MF) const {
-  if (TargetInstrInfo::isSchedulingBoundary(MI, MBB, MF))
-    return true;
-
+bool LoongArchInstrInfo::isSafeToMove(const MachineInstr &MI,
+                                      const MachineBasicBlock *MBB,
+                                      const MachineFunction &MF) const {
   auto MII = MI.getIterator();
   auto MIE = MBB->end();
 
@@ -429,25 +426,25 @@ bool LoongArchInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
     auto MO2 = Lu32I->getOperand(2).getTargetFlags();
     if (MO0 == LoongArchII::MO_PCREL_HI && MO1 == LoongArchII::MO_PCREL_LO &&
         MO2 == LoongArchII::MO_PCREL64_LO)
-      return true;
+      return false;
     if ((MO0 == LoongArchII::MO_GOT_PC_HI || MO0 == LoongArchII::MO_LD_PC_HI ||
          MO0 == LoongArchII::MO_GD_PC_HI) &&
         MO1 == LoongArchII::MO_GOT_PC_LO && MO2 == LoongArchII::MO_GOT_PC64_LO)
-      return true;
+      return false;
     if (MO0 == LoongArchII::MO_IE_PC_HI && MO1 == LoongArchII::MO_IE_PC_LO &&
         MO2 == LoongArchII::MO_IE_PC64_LO)
-      return true;
+      return false;
     if (MO0 == LoongArchII::MO_DESC_PC_HI &&
         MO1 == LoongArchII::MO_DESC_PC_LO &&
         MO2 == LoongArchII::MO_DESC64_PC_LO)
-      return true;
+      return false;
     break;
   }
   case LoongArch::LU52I_D: {
     auto MO = MI.getOperand(2).getTargetFlags();
     if (MO == LoongArchII::MO_PCREL64_HI || MO == LoongArchII::MO_GOT_PC64_HI ||
         MO == LoongArchII::MO_IE_PC64_HI || MO == LoongArchII::MO_DESC64_PC_HI)
-      return true;
+      return false;
     break;
   }
   default:
@@ -487,7 +484,7 @@ bool LoongArchInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
         auto MO1 = LoongArchII::getDirectFlags(SecondOp->getOperand(2));
         auto MO2 = LoongArchII::getDirectFlags(Ld->getOperand(2));
         if (MO1 == LoongArchII::MO_DESC_PC_LO && MO2 == LoongArchII::MO_DESC_LD)
-          return true;
+          return false;
         break;
       }
       if (SecondOp == MIE ||
@@ -496,41 +493,53 @@ bool LoongArchInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
       auto MO1 = LoongArchII::getDirectFlags(SecondOp->getOperand(2));
       if (MO0 == LoongArchII::MO_PCREL_HI && SecondOp->getOpcode() == AddiOp &&
           MO1 == LoongArchII::MO_PCREL_LO)
-        return true;
+        return false;
       if (MO0 == LoongArchII::MO_GOT_PC_HI && SecondOp->getOpcode() == LdOp &&
           MO1 == LoongArchII::MO_GOT_PC_LO)
-        return true;
+        return false;
       if ((MO0 == LoongArchII::MO_LD_PC_HI ||
            MO0 == LoongArchII::MO_GD_PC_HI) &&
           SecondOp->getOpcode() == AddiOp && MO1 == LoongArchII::MO_GOT_PC_LO)
-        return true;
+        return false;
       break;
     }
     case LoongArch::ADDI_W:
     case LoongArch::ADDI_D: {
       auto MO = LoongArchII::getDirectFlags(MI.getOperand(2));
       if (MO == LoongArchII::MO_PCREL_LO || MO == LoongArchII::MO_GOT_PC_LO)
-        return true;
+        return false;
       break;
     }
     case LoongArch::LD_W:
     case LoongArch::LD_D: {
       auto MO = LoongArchII::getDirectFlags(MI.getOperand(2));
       if (MO == LoongArchII::MO_GOT_PC_LO)
-        return true;
+        return false;
       break;
     }
     case LoongArch::PseudoDESC_CALL: {
       auto MO = LoongArchII::getDirectFlags(MI.getOperand(2));
       if (MO == LoongArchII::MO_DESC_CALL)
-        return true;
+        return false;
       break;
     }
     default:
       break;
     }
   }
 
+  return true;
+}
+
+bool LoongArchInstrInfo::isSchedulingBoundary(const MachineInstr &MI,
+                                              const MachineBasicBlock *MBB,
+                                              const MachineFunction &MF) const {
+  if (TargetInstrInfo::isSchedulingBoundary(MI, MBB, MF))
+    return true;
+
+  if (!isSafeToMove(MI, MBB, MF))
+    return true;
+
   return false;
 }
 
diff --git a/llvm/lib/Target/LoongArch/LoongArchInstrInfo.h b/llvm/lib/Target/LoongArch/LoongArchInstrInfo.h
@@ -64,6 +64,9 @@ class LoongArchInstrInfo : public LoongArchGenInstrInfo {
   bool isBranchOffsetInRange(unsigned BranchOpc,
                              int64_t BrOffset) const override;
 
+  bool isSafeToMove(const MachineInstr &MI, const MachineBasicBlock *MBB,
+                    const MachineFunction &MF) const override;
+
   bool isSchedulingBoundary(const MachineInstr &MI,
                             const MachineBasicBlock *MBB,
                             const MachineFunction &MF) const override;
diff --git a/llvm/test/CodeGen/LoongArch/issue163681.ll b/llvm/test/CodeGen/LoongArch/issue163681.ll
@@ -0,0 +1,56 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 6
+; RUN: llc --mtriple=loongarch64 -code-model=large --verify-machineinstrs < %s \
+; RUN:   | FileCheck %s
+
+@.str = external constant [1 x i8]
+
+define void @caller(ptr %0) {
+; CHECK-LABEL: caller:
+; CHECK:       # %bb.0:
+; CHECK-NEXT:    addi.d $sp, $sp, -16
+; CHECK-NEXT:    .cfi_def_cfa_offset 16
+; CHECK-NEXT:    st.d $ra, $sp, 8 # 8-byte Folded Spill
+; CHECK-NEXT:    .cfi_offset 1, -8
+; CHECK-NEXT:    ld.w $a2, $zero, 0
+; CHECK-NEXT:    ld.d $a1, $a0, 0
+; CHECK-NEXT:    beqz $a2, .LBB0_2
+; CHECK-NEXT:  # %bb.1:
+; CHECK-NEXT:    pcalau12i $a0, %got_pc_hi20(.str)
+; CHECK-NEXT:    addi.d $a2, $zero, %got_pc_lo12(.str)
+; CHECK-NEXT:    lu32i.d $a2, %got64_pc_lo20(.str)
+; CHECK-NEXT:    lu52i.d $a2, $a2, %got64_pc_hi12(.str)
+; CHECK-NEXT:    ldx.d $a2, $a2, $a0
+; CHECK-NEXT:    move $a0, $zero
+; CHECK-NEXT:    jirl $ra, $zero, 0
+; CHECK-NEXT:    b .LBB0_3
+; CHECK-NEXT:  .LBB0_2:
+; CHECK-NEXT:    pcalau12i $a0, %got_pc_hi20(.str)
+; CHECK-NEXT:    addi.d $a2, $zero, %got_pc_lo12(.str)
+; CHECK-NEXT:    lu32i.d $a2, %got64_pc_lo20(.str)
+; CHECK-NEXT:    lu52i.d $a2, $a2, %got64_pc_hi12(.str)
+; CHECK-NEXT:    ldx.d $a2, $a2, $a0
+; CHECK-NEXT:    move $a0, $zero
+; CHECK-NEXT:    move $a3, $zero
+; CHECK-NEXT:    jirl $ra, $zero, 0
+; CHECK-NEXT:  .LBB0_3:
+; CHECK-NEXT:    st.d $zero, $zero, 0
+; CHECK-NEXT:    ld.d $ra, $sp, 8 # 8-byte Folded Reload
+; CHECK-NEXT:    addi.d $sp, $sp, 16
+; CHECK-NEXT:    ret
+  %2 = load i32, ptr null, align 4
+  %3 = icmp eq i32 %2, 0
+  %4 = load i64, ptr %0, align 8
+  br i1 %3, label %6, label %5
+
+5:                                                ; preds = %1
+  call void null(ptr null, i64 %4, ptr @.str)
+  br label %7
+
+6:                                                ; preds = %1
+  tail call void null(ptr null, i64 %4, ptr @.str, i32 0)
+  br label %7
+
+7:                                                ; preds = %6, %5
+  store ptr null, ptr null, align 8
+  ret void
+}
