From 95792b92eaa5093e7f214ee74973ff1066b1089d Mon Sep 17 00:00:00 2001
From: Eric Astor <epastor@google.com>
Date: Tue, 29 Sep 2020 16:57:25 -0400
Subject: [PATCH] [ms] [llvm-ml] Add MASM hex float support

Implement MASM's syntax for specifying floats in raw hexadecimal bytes.

Reviewed By: thakis

Differential Revision: https://reviews.llvm.org/D87401
---
 include/llvm/MC/MCParser/MCAsmLexer.h |  4 ++++
 lib/MC/MCParser/AsmLexer.cpp          |  5 +++++
 lib/MC/MCParser/MasmParser.cpp        | 17 +++++++++++++++++
 test/tools/llvm-ml/builtin_types.test |  8 ++++++++
 tools/llvm-mc/llvm-mc.cpp             |  5 +++++
 tools/llvm-ml/llvm-ml.cpp             |  2 ++
 6 files changed, 41 insertions(+)

diff --git a/include/llvm/MC/MCParser/MCAsmLexer.h b/include/llvm/MC/MCParser/MCAsmLexer.h
index 1e449a7f59d..e2f3301d2f2 100644
--- a/include/llvm/MC/MCParser/MCAsmLexer.h
+++ b/include/llvm/MC/MCParser/MCAsmLexer.h
@@ -49,6 +49,7 @@ protected: // Can only create subclasses.
   bool SkipSpace = true;
   bool AllowAtInIdentifier;
   bool IsAtStartOfStatement = true;
+  bool LexMasmHexFloats = false;
   bool LexMasmIntegers = false;
   bool UseMasmDefaultRadix = false;
   unsigned DefaultRadix = 10;
@@ -159,6 +160,9 @@ public:
 
   unsigned getMasmDefaultRadix() const { return DefaultRadix; }
   void setMasmDefaultRadix(unsigned Radix) { DefaultRadix = Radix; }
+
+  /// Set whether to lex masm-style hex float literals, such as 3f800000r.
+  void setLexMasmHexFloats(bool V) { LexMasmHexFloats = V; }
 };
 
 } // end namespace llvm
diff --git a/lib/MC/MCParser/AsmLexer.cpp b/lib/MC/MCParser/AsmLexer.cpp
index 12a71d69ed7..d8a20341bfb 100644
--- a/lib/MC/MCParser/AsmLexer.cpp
+++ b/lib/MC/MCParser/AsmLexer.cpp
@@ -350,6 +350,11 @@ AsmToken AsmLexer::LexDigit() {
       return LexFloatLiteral();
     }
 
+    if (LexMasmHexFloats && (*CurPtr == 'r' || *CurPtr == 'R')) {
+      ++CurPtr;
+      return AsmToken(AsmToken::Real, StringRef(TokStart, CurPtr - TokStart));
+    }
+
     unsigned Radix = 0;
     if (*CurPtr == 'h' || *CurPtr == 'H') {
       // hexadecimal number
diff --git a/lib/MC/MCParser/MasmParser.cpp b/lib/MC/MCParser/MasmParser.cpp
index 352d9473a23..d0a52657d66 100644
--- a/lib/MC/MCParser/MasmParser.cpp
+++ b/lib/MC/MCParser/MasmParser.cpp
@@ -3425,10 +3425,13 @@ bool MasmParser::parseRealValue(const fltSemantics &Semantics, APInt &Res) {
   // We don't truly support arithmetic on floating point expressions, so we
   // have to manually parse unary prefixes.
   bool IsNeg = false;
+  SMLoc SignLoc;
   if (getLexer().is(AsmToken::Minus)) {
+    SignLoc = getLexer().getLoc();
     Lexer.Lex();
     IsNeg = true;
   } else if (getLexer().is(AsmToken::Plus)) {
+    SignLoc = getLexer().getLoc();
     Lexer.Lex();
   }
 
@@ -3450,6 +3453,20 @@ bool MasmParser::parseRealValue(const fltSemantics &Semantics, APInt &Res) {
       Value = APFloat::getZero(Semantics);
     else
       return TokError("invalid floating point literal");
+  } else if (IDVal.consume_back("r") || IDVal.consume_back("R")) {
+    // MASM hexadecimal floating-point literal; no APFloat conversion needed.
+    // To match ML64.exe, ignore the initial sign.
+    unsigned Size = Value.getSizeInBits(Semantics);
+    if (Size != (IDVal.size() << 2))
+      return TokError("invalid floating point literal");
+
+    // Consume the numeric token.
+    Lex();
+
+    Res = APInt(Size, IDVal, 16);
+    if (SignLoc.isValid())
+      return Warning(SignLoc, "MASM-style hex floats ignore explicit sign");
+    return false;
   } else if (errorToBool(
                  Value.convertFromString(IDVal, APFloat::rmNearestTiesToEven)
                      .takeError())) {
diff --git a/test/tools/llvm-ml/builtin_types.test b/test/tools/llvm-ml/builtin_types.test
index b99c491cb8d..f04e318b1b0 100644
--- a/test/tools/llvm-ml/builtin_types.test
+++ b/test/tools/llvm-ml/builtin_types.test
@@ -72,6 +72,14 @@ t6_double REAL8 1.3
 ; CHECK-LABEL: t6_double:
 ; CHECK-NEXT: .quad 4608533498688228557
 
+t7_single_hex REAL4 3f800000r
+t7_double_hex REAL8 3FF0000000000000R
+
+; CHECK-LABEL: t7_single_hex:
+; CHECK-NEXT: .long 1065353216
+; CHECK-LABEL: t7_double_hex:
+; CHECK-NEXT: .quad 4607182418800017408
+
 .code
 
 END
diff --git a/tools/llvm-mc/llvm-mc.cpp b/tools/llvm-mc/llvm-mc.cpp
index 66b55abc489..f8352f36ad7 100644
--- a/tools/llvm-mc/llvm-mc.cpp
+++ b/tools/llvm-mc/llvm-mc.cpp
@@ -169,6 +169,10 @@ static cl::opt<bool> LexMasmIntegers(
     "masm-integers",
     cl::desc("Enable binary and hex masm integers (0b110 and 0ABCh)"));
 
+static cl::opt<bool> LexMasmHexFloats(
+    "masm-hexfloats",
+    cl::desc("Enable MASM-style hex float initializers (3F800000r)"));
+
 static cl::opt<bool> NoExecStack("no-exec-stack",
                                  cl::desc("File doesn't need an exec stack"));
 
@@ -300,6 +304,7 @@ static int AssembleInput(const char *ProgName, const Target *TheTarget,
   Parser->setShowParsedOperands(ShowInstOperands);
   Parser->setTargetParser(*TAP);
   Parser->getLexer().setLexMasmIntegers(LexMasmIntegers);
+  Parser->getLexer().setLexMasmHexFloats(LexMasmHexFloats);
 
   int Res = Parser->Run(NoInitialTextSection);
 
diff --git a/tools/llvm-ml/llvm-ml.cpp b/tools/llvm-ml/llvm-ml.cpp
index 460a566f321..3a3984286e4 100644
--- a/tools/llvm-ml/llvm-ml.cpp
+++ b/tools/llvm-ml/llvm-ml.cpp
@@ -177,6 +177,7 @@ static int AsLexInput(SourceMgr &SrcMgr, MCAsmInfo &MAI, raw_ostream &OS) {
   Lexer.setBuffer(SrcMgr.getMemoryBuffer(SrcMgr.getMainFileID())->getBuffer());
   Lexer.setLexMasmIntegers(true);
   Lexer.useMasmDefaultRadix(true);
+  Lexer.setLexMasmHexFloats(true);
 
   bool Error = false;
   while (Lexer.Lex().isNot(AsmToken::Eof)) {
@@ -208,6 +209,7 @@ static int AssembleInput(const char *ProgName, const Target *TheTarget,
   Parser->setTargetParser(*TAP);
   Parser->getLexer().setLexMasmIntegers(true);
   Parser->getLexer().useMasmDefaultRadix(true);
+  Parser->getLexer().setLexMasmHexFloats(true);
 
   int Res = Parser->Run(/*NoInitialTextSection=*/true);