[RelLookupTableConverter] Drop unnamed_addr for GVs in operands to avoid generating GOTPCREL relocations

Author: dianqk

llvm/lib/Transforms/Utils/RelLookupTableConverter.cpp

@@ -21,7 +21,10 @@
 
 using namespace llvm;
 
-static bool shouldConvertToRelLookupTable(Module &M, GlobalVariable &GV) {
+static bool
+shouldConvertToRelLookupTable(Module &M, GlobalVariable &GV,
+                              SmallVectorImpl<GlobalVariable *> &GVOps,
+                              bool ShouldDropUnnamedAddr) {
   // If lookup table has more than one user,
   // do not generate a relative lookup table.
   // This is to simplify the analysis that needs to be done for this pass.
@@ -85,6 +88,9 @@ static bool shouldConvertToRelLookupTable(Module &M, GlobalVariable &GV) {
         !GlovalVarOp->isDSOLocal() ||
         !GlovalVarOp->isImplicitDSOLocal())
       return false;
+
+    if (ShouldDropUnnamedAddr)
+      GVOps.push_back(GlovalVarOp);
   }
 
   return true;
@@ -108,24 +114,8 @@ static GlobalVariable *createRelLookupTable(Function &Func,
   uint64_t Idx = 0;
   SmallVector<Constant *, 64> RelLookupTableContents(NumElts);
 
-  Triple TT = M.getTargetTriple();
-  // FIXME: This should be removed in the future.
-  bool ShouldDropUnnamedAddr =
-      // Drop unnamed_addr to avoid matching pattern in
-      // `handleIndirectSymViaGOTPCRel`, which generates GOTPCREL relocations
-      // not supported by the GNU linker and LLD versions below 18 on aarch64.
-      TT.isAArch64()
-      // Apple's ld64 (and ld-prime on Xcode 15.2) miscompile something on
-      // x86_64-apple-darwin. See
-      // https://github.com/rust-lang/rust/issues/140686 and
-      // https://github.com/rust-lang/rust/issues/141306.
-      || (TT.isX86() && TT.isOSDarwin());
-
   for (Use &Operand : LookupTableArr->operands()) {
     Constant *Element = cast<Constant>(Operand);
-    if (ShouldDropUnnamedAddr)
-      if (auto *GlobalElement = dyn_cast<GlobalValue>(Element))
-        GlobalElement->setUnnamedAddr(GlobalValue::UnnamedAddr::None);
     Type *IntPtrTy = M.getDataLayout().getIntPtrType(M.getContext());
     Constant *Base = llvm::ConstantExpr::getPtrToInt(RelLookupTable, IntPtrTy);
     Constant *Target = llvm::ConstantExpr::getPtrToInt(Element, IntPtrTy);
@@ -199,10 +189,28 @@ static bool convertToRelativeLookupTables(
 
   bool Changed = false;
 
+  Triple TT = M.getTargetTriple();
+  // FIXME: This should be removed in the future.
+  bool ShouldDropUnnamedAddr =
+      // Drop unnamed_addr to avoid matching pattern in
+      // `handleIndirectSymViaGOTPCRel`, which generates GOTPCREL relocations
+      // not supported by the GNU linker and LLD versions below 18 on aarch64.
+      TT.isAArch64()
+      // Apple's ld64 (and ld-prime on Xcode 15.2) miscompile something on
+      // x86_64-apple-darwin. See
+      // https://github.com/rust-lang/rust/issues/140686 and
+      // https://github.com/rust-lang/rust/issues/141306.
+      || (TT.isX86() && TT.isOSDarwin());
   for (GlobalVariable &GV : llvm::make_early_inc_range(M.globals())) {
-    if (!shouldConvertToRelLookupTable(M, GV))
+    SmallVector<GlobalVariable *, 4> GVOps;
+
+    if (!shouldConvertToRelLookupTable(M, GV, GVOps, ShouldDropUnnamedAddr))
       continue;
 
+    if (ShouldDropUnnamedAddr)
+      for (auto *GVOp : GVOps)
+        GVOp->setUnnamedAddr(GlobalValue::UnnamedAddr::None);
+
     convertToRelLookupTable(GV);
 
     // Remove the original lookup table.

llvm/test/Transforms/RelLookupTableConverter/unnamed_addr.ll

@@ -42,9 +42,9 @@
 ; x86_64-apple-darwin: @y2 = internal constant ptr @x1
 ; x86_64-apple-darwin: @y3 = internal constant ptr @x0
 ; x86_64-apple-darwin: @load_relative_2.table.rel = private unnamed_addr constant [4 x i32] [i32 trunc (i64 sub (i64 ptrtoint (ptr @y3 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y2 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y1 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y0 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32)], align 4
-; x86_64-apple-darwin: @b0 = private unnamed_addr constant [8 x i8] c"00000000"
-; x86_64-apple-darwin: @b1 = private unnamed_addr constant [8 x i8] c"11111111"
-; x86_64-apple-darwin: @b2 = private unnamed_addr constant [8 x i8] c"22222222"
+; x86_64-apple-darwin: @b0 = private constant [8 x i8] c"00000000"
+; x86_64-apple-darwin: @b1 = private constant [8 x i8] c"11111111"
+; x86_64-apple-darwin: @b2 = private constant [8 x i8] c"22222222"
 ; x86_64-apple-darwin: @load_relative_3.table.rel = private unnamed_addr constant [3 x i32] [i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b0, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b1, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b2, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32)], align 4
 ;.
 ; aarch64: @a0 = private constant i32 0
@@ -60,9 +60,9 @@
 ; aarch64: @y2 = internal constant ptr @x1
 ; aarch64: @y3 = internal constant ptr @x0
 ; aarch64: @load_relative_2.table.rel = private unnamed_addr constant [4 x i32] [i32 trunc (i64 sub (i64 ptrtoint (ptr @y3 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y2 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y1 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr @y0 to i64), i64 ptrtoint (ptr @load_relative_2.table.rel to i64)) to i32)], align 4
-; aarch64: @b0 = private unnamed_addr constant [8 x i8] c"00000000"
-; aarch64: @b1 = private unnamed_addr constant [8 x i8] c"11111111"
-; aarch64: @b2 = private unnamed_addr constant [8 x i8] c"22222222"
+; aarch64: @b0 = private constant [8 x i8] c"00000000"
+; aarch64: @b1 = private constant [8 x i8] c"11111111"
+; aarch64: @b2 = private constant [8 x i8] c"22222222"
 ; aarch64: @load_relative_3.table.rel = private unnamed_addr constant [3 x i32] [i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b0, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b1, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32), i32 trunc (i64 sub (i64 ptrtoint (ptr getelementptr inbounds (i8, ptr @b2, i64 8) to i64), i64 ptrtoint (ptr @load_relative_3.table.rel to i64)) to i32)], align 4
 ;.
 ; x86_64: @a0 = private unnamed_addr constant i32 0