[llvm] [WIP][ASAN] Add "asan_instrumented" llvm ir attribute to identify AddressSanitizer instrumented globals (PR #68865)

[via llvm-commits]

https://github.com/skc7 updated https://github.com/llvm/llvm-project/pull/68865

>From 8b24dcc15e821366b1150658af8efa51ff406a4e Mon Sep 17 00:00:00 2001
From: skc7 <Krishna.Sankisa at amd.com>
Date: Thu, 12 Oct 2023 14:05:48 +0530
Subject: [PATCH] [ASAN] Add AsanInstrumented llvm ir attribute to identify
 asan instrumented globals

---
 llvm/include/llvm/Bitcode/LLVMBitCodes.h  | 1 +
 llvm/include/llvm/IR/Attributes.h         | 3 +++
 llvm/include/llvm/IR/Attributes.td        | 3 +++
 llvm/lib/AsmParser/LLLexer.cpp            | 1 +
 llvm/lib/AsmParser/LLParser.cpp           | 8 ++++++++
 llvm/lib/Bitcode/Reader/BitcodeReader.cpp | 2 ++
 llvm/lib/Bitcode/Writer/BitcodeWriter.cpp | 2 ++
 llvm/lib/IR/Attributes.cpp                | 7 +++++++
 8 files changed, 27 insertions(+)

diff --git a/llvm/include/llvm/Bitcode/LLVMBitCodes.h b/llvm/include/llvm/Bitcode/LLVMBitCodes.h
index 52e76356a892e45..bb15e01d08d0670 100644
--- a/llvm/include/llvm/Bitcode/LLVMBitCodes.h
+++ b/llvm/include/llvm/Bitcode/LLVMBitCodes.h
@@ -713,6 +713,7 @@ enum AttributeKindCodes {
   ATTR_KIND_SKIP_PROFILE = 85,
   ATTR_KIND_MEMORY = 86,
   ATTR_KIND_NOFPCLASS = 87,
+  ATTR_KIND_ASAN_INSTRUMENTED = 88,
 };
 
 enum ComdatSelectionKindCodes {
diff --git a/llvm/include/llvm/IR/Attributes.h b/llvm/include/llvm/IR/Attributes.h
index db33b5400471685..eb02158f15a1254 100644
--- a/llvm/include/llvm/IR/Attributes.h
+++ b/llvm/include/llvm/IR/Attributes.h
@@ -251,6 +251,9 @@ class Attribute {
   /// Return the FPClassTest for nofpclass
   FPClassTest getNoFPClass() const;
 
+  /// Return if global variable is instrumented by AddrSanitizer.
+  bool isAsanInstrumented() const;
+
   /// The Attribute is converted to a string of equivalent mnemonic. This
   /// is, presumably, for writing out the mnemonics for the assembly writer.
   std::string getAsString(bool InAttrGrp = false) const;
diff --git a/llvm/include/llvm/IR/Attributes.td b/llvm/include/llvm/IR/Attributes.td
index aba1d718f7f72f9..1f81435d677d687 100644
--- a/llvm/include/llvm/IR/Attributes.td
+++ b/llvm/include/llvm/IR/Attributes.td
@@ -273,6 +273,9 @@ def SanitizeHWAddress : EnumAttr<"sanitize_hwaddress", [FnAttr]>;
 /// MemTagSanitizer is on.
 def SanitizeMemTag : EnumAttr<"sanitize_memtag", [FnAttr]>;
 
+/// Attribute to identify global variables instrumented by AddrSanitizer.
+def AsanInstrumented : EnumAttr<"asan_instrumented", []>;
+
 /// Speculative Load Hardening is enabled.
 ///
 /// Note that this uses the default compatibility (always compatible during
diff --git a/llvm/lib/AsmParser/LLLexer.cpp b/llvm/lib/AsmParser/LLLexer.cpp
index 466bdebc001f589..d3cf6ce8626f921 100644
--- a/llvm/lib/AsmParser/LLLexer.cpp
+++ b/llvm/lib/AsmParser/LLLexer.cpp
@@ -582,6 +582,7 @@ lltok::Kind LLLexer::LexIdentifier() {
   KEYWORD(no_sanitize_address);
   KEYWORD(no_sanitize_hwaddress);
   KEYWORD(sanitize_address_dyninit);
+  KEYWORD(asan_instrumented);
 
   KEYWORD(ccc);
   KEYWORD(fastcc);
diff --git a/llvm/lib/AsmParser/LLParser.cpp b/llvm/lib/AsmParser/LLParser.cpp
index 04eabc94cfc6abe..0d57065dc6b05ce 100644
--- a/llvm/lib/AsmParser/LLParser.cpp
+++ b/llvm/lib/AsmParser/LLParser.cpp
@@ -1525,6 +1525,14 @@ bool LLParser::parseFnAttributeValuePairs(AttrBuilder &B,
       continue;
     }
 
+    if (Token == lltok::kw_asan_instrumented) {
+      if (parseToken(lltok::kw_asan_instrumented,
+                     "expected 'asan_instrumented'"))
+        return true;
+      B.addAttribute(Attribute::AsanInstrumented);
+      continue;
+    }
+
     SMLoc Loc = Lex.getLoc();
     if (Token == lltok::kw_builtin)
       BuiltinLoc = Loc;
diff --git a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
index 1d1ec988a93d847..ff6e42970b9a8d4 100644
--- a/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
+++ b/llvm/lib/Bitcode/Reader/BitcodeReader.cpp
@@ -1980,6 +1980,8 @@ static Attribute::AttrKind getAttrFromCode(uint64_t Code) {
     return Attribute::NoSanitizeCoverage;
   case bitc::ATTR_KIND_NULL_POINTER_IS_VALID:
     return Attribute::NullPointerIsValid;
+  case bitc::ATTR_KIND_ASAN_INSTRUMENTED:
+    return Attribute::AsanInstrumented;
   case bitc::ATTR_KIND_OPT_FOR_FUZZING:
     return Attribute::OptForFuzzing;
   case bitc::ATTR_KIND_OPTIMIZE_FOR_SIZE:
diff --git a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
index e991d055f33474b..0209882cc33f20e 100644
--- a/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
+++ b/llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
@@ -789,6 +789,8 @@ static uint64_t getAttrKindEncoding(Attribute::AttrKind Kind) {
     return bitc::ATTR_KIND_SANITIZE_THREAD;
   case Attribute::SanitizeMemory:
     return bitc::ATTR_KIND_SANITIZE_MEMORY;
+  case Attribute::AsanInstrumented:
+    return bitc::ATTR_KIND_ASAN_INSTRUMENTED;
   case Attribute::SpeculativeLoadHardening:
     return bitc::ATTR_KIND_SPECULATIVE_LOAD_HARDENING;
   case Attribute::SwiftError:
diff --git a/llvm/lib/IR/Attributes.cpp b/llvm/lib/IR/Attributes.cpp
index 3d89d18e5822bee..7cef414f53ec357 100644
--- a/llvm/lib/IR/Attributes.cpp
+++ b/llvm/lib/IR/Attributes.cpp
@@ -408,6 +408,10 @@ FPClassTest Attribute::getNoFPClass() const {
   return static_cast<FPClassTest>(pImpl->getValueAsInt());
 }
 
+bool Attribute::isAsanInstrumented() const {
+  return hasAttribute(Attribute::AsanInstrumented);
+}
+
 static const char *getModRefStr(ModRefInfo MR) {
   switch (MR) {
   case ModRefInfo::NoModRef:
@@ -562,6 +566,9 @@ std::string Attribute::getAsString(bool InAttrGrp) const {
     return Result;
   }
 
+  if (hasAttribute(Attribute::AsanInstrumented))
+    return "asan_instrumented";
+
   // Convert target-dependent attributes to strings of the form:
   //
   //   "kind"