Add support for macro parameters/arguments delimited by spaces,
to improve compatibility with GNU as.
Based on a patch by PaX Team.
Fixed assertion failures on non-Darwin and added additional test cases.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@164248 91177308-0d34-0410-b5e6-96231b3b80d8
diff --git a/lib/MC/MCParser/AsmParser.cpp b/lib/MC/MCParser/AsmParser.cpp
index 266d87e..2266b63 100644
--- a/lib/MC/MCParser/AsmParser.cpp
+++ b/lib/MC/MCParser/AsmParser.cpp
@@ -130,6 +130,9 @@
/// AssemblerDialect. ~OU means unset value and use value provided by MAI.
unsigned AssemblerDialect;
+ /// IsDarwin - is Darwin compatibility enabled?
+ bool IsDarwin;
+
public:
AsmParser(SourceMgr &SM, MCContext &Ctx, MCStreamer &Out,
const MCAsmInfo &MAI);
@@ -209,7 +212,8 @@
virtual void EatToEndOfStatement();
- bool ParseMacroArgument(MacroArgument &MA);
+ bool ParseMacroArgument(MacroArgument &MA,
+ AsmToken::TokenKind &ArgumentDelimiter);
bool ParseMacroArguments(const Macro *M, MacroArguments &A);
/// \brief Parse up to the end of statement and a return the contents from the
@@ -407,8 +411,8 @@
MCStreamer &_Out, const MCAsmInfo &_MAI)
: Lexer(_MAI), Ctx(_Ctx), Out(_Out), MAI(_MAI), SrcMgr(_SM),
GenericParser(new GenericAsmParser), PlatformParser(0),
- CurBuffer(0), MacrosEnabled(true), CppHashLineNumber(0),
- AssemblerDialect(~0U) {
+ CurBuffer(0), MacrosEnabled(true), CppHashLineNumber(0),
+ AssemblerDialect(~0U), IsDarwin(false) {
// Save the old handler.
SavedDiagHandler = SrcMgr.getDiagHandler();
SavedDiagContext = SrcMgr.getDiagContext();
@@ -429,6 +433,7 @@
} else if (_MAI.hasSubsectionsViaSymbols()) {
PlatformParser = createDarwinAsmParser();
PlatformParser->Initialize(*this);
+ IsDarwin = true;
} else {
PlatformParser = createELFAsmParser();
PlatformParser->Initialize(*this);
@@ -1471,6 +1476,8 @@
if (NParameters != 0 && NParameters != A.size())
return Error(L, "Wrong number of arguments");
+ // A macro without parameters is handled differently on Darwin:
+ // gas accepts no arguments and does no substitutions
while (!Body.empty()) {
// Scan for the next substitution.
std::size_t End = Body.size(), Pos = 0;
@@ -1537,15 +1544,23 @@
if (Parameters[Index].first == Argument)
break;
- // FIXME: We should error at the macro definition.
- if (Index == NParameters)
- return Error(L, "Parameter not found");
+ if (Index == NParameters) {
+ if (Body[Pos+1] == '(' && Body[Pos+2] == ')')
+ Pos += 3;
+ else {
+ OS << '\\' << Argument;
+ Pos = I;
+ }
+ } else {
+ for (MacroArgument::const_iterator it = A[Index].begin(),
+ ie = A[Index].end(); it != ie; ++it)
+ if (it->getKind() == AsmToken::String)
+ OS << it->getStringContents();
+ else
+ OS << it->getString();
- for (MacroArgument::const_iterator it = A[Index].begin(),
- ie = A[Index].end(); it != ie; ++it)
- OS << it->getString();
-
- Pos += 1 + Argument.size();
+ Pos += 1 + Argument.size();
+ }
}
// Update the scan point.
Body = Body.substr(Pos);
@@ -1560,22 +1575,97 @@
{
}
+static bool IsOperator(AsmToken::TokenKind kind)
+{
+ switch (kind)
+ {
+ default:
+ return false;
+ case AsmToken::Plus:
+ case AsmToken::Minus:
+ case AsmToken::Tilde:
+ case AsmToken::Slash:
+ case AsmToken::Star:
+ case AsmToken::Dot:
+ case AsmToken::Equal:
+ case AsmToken::EqualEqual:
+ case AsmToken::Pipe:
+ case AsmToken::PipePipe:
+ case AsmToken::Caret:
+ case AsmToken::Amp:
+ case AsmToken::AmpAmp:
+ case AsmToken::Exclaim:
+ case AsmToken::ExclaimEqual:
+ case AsmToken::Percent:
+ case AsmToken::Less:
+ case AsmToken::LessEqual:
+ case AsmToken::LessLess:
+ case AsmToken::LessGreater:
+ case AsmToken::Greater:
+ case AsmToken::GreaterEqual:
+ case AsmToken::GreaterGreater:
+ return true;
+ }
+}
+
/// ParseMacroArgument - Extract AsmTokens for a macro argument.
/// This is used for both default macro parameter values and the
/// arguments in macro invocations
-bool AsmParser::ParseMacroArgument(MacroArgument &MA) {
+bool AsmParser::ParseMacroArgument(MacroArgument &MA,
+ AsmToken::TokenKind &ArgumentDelimiter) {
unsigned ParenLevel = 0;
+ unsigned AddTokens = 0;
+
+ // gas accepts arguments separated by whitespace, except on Darwin
+ if (!IsDarwin)
+ Lexer.setSkipSpace(false);
for (;;) {
- if (Lexer.is(AsmToken::Eof) || Lexer.is(AsmToken::Equal))
+ if (Lexer.is(AsmToken::Eof) || Lexer.is(AsmToken::Equal)) {
+ Lexer.setSkipSpace(true);
return TokError("unexpected token in macro instantiation");
+ }
+
+ if (ParenLevel == 0 && Lexer.is(AsmToken::Comma)) {
+ // Spaces and commas cannot be mixed to delimit parameters
+ if (ArgumentDelimiter == AsmToken::Eof)
+ ArgumentDelimiter = AsmToken::Comma;
+ else if (ArgumentDelimiter != AsmToken::Comma) {
+ Lexer.setSkipSpace(true);
+ return TokError("expected ' ' for macro argument separator");
+ }
+ break;
+ }
+
+ if (Lexer.is(AsmToken::Space)) {
+ Lex(); // Eat spaces
+
+ // Spaces can delimit parameters, but could also be part an expression.
+ // If the token after a space is an operator, add the token and the next
+ // one into this argument
+ if (ArgumentDelimiter == AsmToken::Space ||
+ ArgumentDelimiter == AsmToken::Eof) {
+ if (IsOperator(Lexer.getKind())) {
+ // Check to see whether the token is used as an operator,
+ // or part of an identifier
+ const char *NextChar = getTok().getEndLoc().getPointer() + 1;
+ if (*NextChar == ' ')
+ AddTokens = 2;
+ }
+
+ if (!AddTokens && ParenLevel == 0) {
+ if (ArgumentDelimiter == AsmToken::Eof &&
+ !IsOperator(Lexer.getKind()))
+ ArgumentDelimiter = AsmToken::Space;
+ break;
+ }
+ }
+ }
// HandleMacroEntry relies on not advancing the lexer here
// to be able to fill in the remaining default parameter values
if (Lexer.is(AsmToken::EndOfStatement))
break;
- if (ParenLevel == 0 && Lexer.is(AsmToken::Comma))
- break;
// Adjust the current parentheses level.
if (Lexer.is(AsmToken::LParen))
@@ -1585,8 +1675,12 @@
// Append the token to the current argument list.
MA.push_back(getTok());
+ if (AddTokens)
+ AddTokens--;
Lex();
}
+
+ Lexer.setSkipSpace(true);
if (ParenLevel != 0)
return TokError("unbalanced parentheses in macro argument");
return false;
@@ -1595,6 +1689,9 @@
// Parse the macro instantiation arguments.
bool AsmParser::ParseMacroArguments(const Macro *M, MacroArguments &A) {
const unsigned NParameters = M ? M->Parameters.size() : 0;
+ // Argument delimiter is initially unknown. It will be set by
+ // ParseMacroArgument()
+ AsmToken::TokenKind ArgumentDelimiter = AsmToken::Eof;
// Parse two kinds of macro invocations:
// - macros defined without any parameters accept an arbitrary number of them
@@ -1603,7 +1700,7 @@
++Parameter) {
MacroArgument MA;
- if (ParseMacroArgument(MA))
+ if (ParseMacroArgument(MA, ArgumentDelimiter))
return true;
if (!MA.empty() || !NParameters)
@@ -3105,6 +3202,9 @@
return TokError("expected identifier in '.macro' directive");
MacroParameters Parameters;
+ // Argument delimiter is initially unknown. It will be set by
+ // ParseMacroArgument()
+ AsmToken::TokenKind ArgumentDelimiter = AsmToken::Eof;
if (getLexer().isNot(AsmToken::EndOfStatement)) {
for (;;) {
MacroParameter Parameter;
@@ -3113,21 +3213,19 @@
if (getLexer().is(AsmToken::Equal)) {
Lex();
- if (getParser().ParseMacroArgument(Parameter.second))
+ if (getParser().ParseMacroArgument(Parameter.second, ArgumentDelimiter))
return true;
}
Parameters.push_back(Parameter);
- if (getLexer().isNot(AsmToken::Comma))
+ if (getLexer().is(AsmToken::Comma))
+ Lex();
+ else if (getLexer().is(AsmToken::EndOfStatement))
break;
- Lex();
}
}
- if (getLexer().isNot(AsmToken::EndOfStatement))
- return TokError("unexpected token in '.macro' directive");
-
// Eat the end of statement.
Lex();