Bring r325915 back.

The tests that failed on a windows host have been fixed.

Original message:

Start setting dso_local for COFF.

With this there are still some GVs where we don't set dso_local
because setGVProperties is never called. I intend to fix that in
followup commits. This is just the bare minimum to teach
shouldAssumeDSOLocal what it should do for COFF.

llvm-svn: 325940

1 files changed, 13 insertions, 1 deletions

diff --git a/clang/lib/CodeGen/CodeGenModule.cpp b/clang/lib/CodeGen/CodeGenModule.cpp
index e8577b0153f..03bf2169157 100644
--- a/clang/lib/CodeGen/CodeGenModule.cpp
+++ b/clang/lib/CodeGen/CodeGenModule.cpp
@@ -713,8 +713,20 @@ void CodeGenModule::setGlobalVisibility(llvm::GlobalValue *GV,
 
 static bool shouldAssumeDSOLocal(const CodeGenModule &CGM,
                                  llvm::GlobalValue *GV) {
+  // DLLImport explicitly marks the GV as external.
+  if (GV->hasDLLImportStorageClass())
+    return false;
+
   const llvm::Triple &TT = CGM.getTriple();
-  // Only handle ELF for now.
+  // Every other GV is local on COFF.
+  // Make an exception for windows OS in the triple: Some firmware builds use
+  // *-win32-macho triples. This (accidentally?) produced windows relocations
+  // without GOT tables in older clang versions; Keep this behaviour.
+  // FIXME: even thread local variables?
+  if (TT.isOSBinFormatCOFF() || (TT.isOSWindows() && TT.isOSBinFormatMachO()))
+    return true;
+
+  // Only handle COFF and ELF for now.
   if (!TT.isOSBinFormatELF())
     return false;