From 83c517c44eb3087b284ee73c49decb7aaf38e273 Mon Sep 17 00:00:00 2001
From: Teresa Johnson <tejohnson@google.com>
Date: Wed, 30 Mar 2016 18:15:08 +0000
Subject: [PATCH] Restore "[ThinLTO] Serialize the Module SourceFileName
 to/from LLVM assembly"

This restores commit 264869, with a fix for windows bots to properly
escape '\' in the path when serializing out. Added test.

llvm-svn: 264884
---
 llvm/lib/AsmParser/LLLexer.cpp                   |  1 +
 llvm/lib/AsmParser/LLParser.cpp                  | 17 +++++++++++++++++
 llvm/lib/AsmParser/LLParser.h                    |  1 +
 llvm/lib/AsmParser/LLToken.h                     |  1 +
 llvm/lib/IR/AsmWriter.cpp                        | 22 ++++++++++++++++++++++
 llvm/test/Assembler/source-filename-backslash.ll |  8 ++++++++
 llvm/test/Assembler/source-filename.ll           |  8 ++++++++
 7 files changed, 58 insertions(+)
 create mode 100644 llvm/test/Assembler/source-filename-backslash.ll
 create mode 100644 llvm/test/Assembler/source-filename.ll

diff --git a/llvm/lib/AsmParser/LLLexer.cpp b/llvm/lib/AsmParser/LLLexer.cpp
index d238513..46ffc64 100644
--- a/llvm/lib/AsmParser/LLLexer.cpp
+++ b/llvm/lib/AsmParser/LLLexer.cpp
@@ -533,6 +533,7 @@ lltok::Kind LLLexer::LexIdentifier() {
   KEYWORD(notail);
   KEYWORD(target);
   KEYWORD(triple);
+  KEYWORD(source_filename);
   KEYWORD(unwind);
   KEYWORD(deplibs);             // FIXME: Remove in 4.0.
   KEYWORD(datalayout);
diff --git a/llvm/lib/AsmParser/LLParser.cpp b/llvm/lib/AsmParser/LLParser.cpp
index 76bd99e..39c6135 100644
--- a/llvm/lib/AsmParser/LLParser.cpp
+++ b/llvm/lib/AsmParser/LLParser.cpp
@@ -239,6 +239,10 @@ bool LLParser::ParseTopLevelEntities() {
     case lltok::kw_define:  if (ParseDefine()) return true; break;
     case lltok::kw_module:  if (ParseModuleAsm()) return true; break;
     case lltok::kw_target:  if (ParseTargetDefinition()) return true; break;
+    case lltok::kw_source_filename:
+      if (ParseSourceFileName())
+        return true;
+      break;
     case lltok::kw_deplibs: if (ParseDepLibs()) return true; break;
     case lltok::LocalVarID: if (ParseUnnamedType()) return true; break;
     case lltok::LocalVar:   if (ParseNamedType()) return true; break;
@@ -336,6 +340,19 @@ bool LLParser::ParseTargetDefinition() {
 }
 
 /// toplevelentity
+///   ::= 'source_filename' '=' STRINGCONSTANT
+bool LLParser::ParseSourceFileName() {
+  assert(Lex.getKind() == lltok::kw_source_filename);
+  std::string Str;
+  Lex.Lex();
+  if (ParseToken(lltok::equal, "expected '=' after source_filename") ||
+      ParseStringConstant(Str))
+    return true;
+  M->setSourceFileName(Str);
+  return false;
+}
+
+/// toplevelentity
 ///   ::= 'deplibs' '=' '[' ']'
 ///   ::= 'deplibs' '=' '[' STRINGCONSTANT (',' STRINGCONSTANT)* ']'
 /// FIXME: Remove in 4.0. Currently parse, but ignore.
diff --git a/llvm/lib/AsmParser/LLParser.h b/llvm/lib/AsmParser/LLParser.h
index 96f864a..fcece62 100644
--- a/llvm/lib/AsmParser/LLParser.h
+++ b/llvm/lib/AsmParser/LLParser.h
@@ -261,6 +261,7 @@ namespace llvm {
     bool ValidateEndOfModule();
     bool ParseTargetDefinition();
     bool ParseModuleAsm();
+    bool ParseSourceFileName();
     bool ParseDepLibs();        // FIXME: Remove in 4.0.
     bool ParseUnnamedType();
     bool ParseNamedType();
diff --git a/llvm/lib/AsmParser/LLToken.h b/llvm/lib/AsmParser/LLToken.h
index f06b5b8..c5a7490 100644
--- a/llvm/lib/AsmParser/LLToken.h
+++ b/llvm/lib/AsmParser/LLToken.h
@@ -59,6 +59,7 @@ namespace lltok {
     kw_notail,
     kw_target,
     kw_triple,
+    kw_source_filename,
     kw_unwind,
     kw_deplibs,                 // FIXME: Remove in 4.0
     kw_datalayout,
diff --git a/llvm/lib/IR/AsmWriter.cpp b/llvm/lib/IR/AsmWriter.cpp
index ab56f080..4750cf6 100644
--- a/llvm/lib/IR/AsmWriter.cpp
+++ b/llvm/lib/IR/AsmWriter.cpp
@@ -2203,6 +2203,22 @@ void AssemblyWriter::writeOperandBundles(ImmutableCallSite CS) {
   Out << " ]";
 }
 
+/// Escape any backslashes in the source file (e.g. Windows paths)
+/// before emitting, so that it is parsed properly by the lexer on input.
+static void EscapeBackslashes(std::string Str,
+                              SmallVectorImpl<char> &Res) {
+  for (auto C : Str) {
+    switch (C) {
+    default:
+      break;
+    case '\\':
+      Res.push_back('\\');
+      break;
+    }
+    Res.push_back(C);
+  }
+}
+
 void AssemblyWriter::printModule(const Module *M) {
   Machine.initialize();
 
@@ -2215,6 +2231,12 @@ void AssemblyWriter::printModule(const Module *M) {
       M->getModuleIdentifier().find('\n') == std::string::npos)
     Out << "; ModuleID = '" << M->getModuleIdentifier() << "'\n";
 
+  if (!M->getSourceFileName().empty()) {
+    SmallString<128> EscapedName;
+    EscapeBackslashes(M->getSourceFileName(), EscapedName);
+    Out << "source_filename = \"" << EscapedName << "\"\n";
+  }
+
   const std::string &DL = M->getDataLayoutStr();
   if (!DL.empty())
     Out << "target datalayout = \"" << DL << "\"\n";
diff --git a/llvm/test/Assembler/source-filename-backslash.ll b/llvm/test/Assembler/source-filename-backslash.ll
new file mode 100644
index 0000000..27b5198
--- /dev/null
+++ b/llvm/test/Assembler/source-filename-backslash.ll
@@ -0,0 +1,8 @@
+
+; Make sure that llvm-as/llvm-dis properly assemble/disassemble the
+; source_filename.
+
+; RUN: llvm-as < %s | llvm-dis | FileCheck %s
+
+; CHECK: source_filename = "C:\\path\\with\\backslashes\\test.cc"
+source_filename = "C:\\path\\with\\backslashes\\test.cc"
diff --git a/llvm/test/Assembler/source-filename.ll b/llvm/test/Assembler/source-filename.ll
new file mode 100644
index 0000000..11284fa
--- /dev/null
+++ b/llvm/test/Assembler/source-filename.ll
@@ -0,0 +1,8 @@
+
+; Make sure that llvm-as/llvm-dis properly assemble/disassemble the
+; source_filename.
+
+; RUN: llvm-as < %s | llvm-dis | FileCheck %s
+
+; CHECK: source_filename = "test.cc"
+source_filename = "test.cc"
-- 
2.7.4