[LLD][COFF] Deduplicate common chunks when linking COFF files.

lld/COFF/Chunks.cpp

@@ -777,7 +777,7 @@ uint32_t SectionChunk::getSectionNumber() const {
   return s.getIndex() + 1;
 }
 
-CommonChunk::CommonChunk(const COFFSymbolRef s) : sym(s) {
+CommonChunk::CommonChunk(const COFFSymbolRef s) : active(false), sym(s) {
   // The value of a common symbol is its size. Align all common symbols smaller
   // than 32 bytes naturally, i.e. round the size up to the next power of two.
   // This is what MSVC link.exe does.

lld/COFF/Chunks.h

@@ -522,6 +522,8 @@ class CommonChunk : public NonSectionChunk {
   uint32_t getOutputCharacteristics() const override;
   StringRef getSectionName() const override { return ".bss"; }
 
+  bool active;
+
 private:
   const COFFSymbolRef sym;
 };

lld/COFF/Symbols.h

@@ -236,6 +236,8 @@ class DefinedCommon : public DefinedCOFF {
                 CommonChunk *c = nullptr)
       : DefinedCOFF(DefinedCommonKind, f, n, s), data(c), size(size) {
     this->isExternal = true;
+    if (c)
+      c->active = true;
   }
 
   static bool classof(const Symbol *s) {

lld/COFF/Writer.cpp

@@ -1114,6 +1114,10 @@ void Writer::createSections() {
         sc->printDiscardedMessage();
       continue;
     }
+    if (auto *cc = dyn_cast<CommonChunk>(c)) {
+      if (!cc->active)
+        continue;
+    }
     StringRef name = c->getSectionName();
     if (shouldStripSectionSuffix(sc, name, ctx.config.mingw))
       name = name.split('$').first;

lld/test/COFF/common-dedup.ll

@@ -0,0 +1,38 @@
+; REQUIRES: x86
+; RUN: split-file %s %t.dir
+; RUN: llc %t.dir/t1.ll -o %t.t1.obj --filetype=obj
+; RUN: llc %t.dir/t2.ll -o %t.t2.obj --filetype=obj
+; RUN: lld-link %t.t1.obj %t.t2.obj  -entry:main -out:%t.exe
+; RUN: llvm-readobj --section-headers %t.exe | FileCheck %s
+
+; Make sure that the data section contains just one copy of @a, not two.
+; CHECK: Name: .data
+; CHECK-NEXT: VirtualSize: 0x1000
+
+;--- t1.ll
+target triple = "x86_64-pc-windows-msvc"
+@a = common global [4096 x i8] zeroinitializer
+
+define i32 @usea() {
+  %ref_common = load i32, ptr @a
+  ret i32 %ref_common
+}
+
+;--- t2.ll
+target triple = "x86_64-pc-windows-msvc"
+@a = common global [4096 x i8] zeroinitializer
+
+define i32 @useb() {
+  %ref_common = load i32, ptr @a
+  ret i32 %ref_common
+}
+
+declare i32 @usea()
+
+define dso_local i32 @main() local_unnamed_addr {
+entry:
+  %a = tail call i32 @usea()
+  %b = tail call i32 @useb()
+  %add = add nsw i32 %a, %b
+  ret i32 %add
+}