[mips] Process numeric register name in the .set assignment directive

Now LLVM assembler cannot process the following code and generates an
error. GNU tools support .set assignment directive with numeric register
name.

```
.set r4, 4

test.s:1:11: error: invalid token in expression
  .set r4, $4
           ^
```

This patch teach assembler to handle such directives correctly.
Unfortunately a numeric register name cannot be represented as an
expression. That's why we have to maintain a separate `StringMap`
in the `MipsAsmParser` to keep mapping between aliases names and
register numbers.

Differential revision: https://reviews.llvm.org/D47464

llvm-svn: 333428

lib/Target/Mips/AsmParser/MipsAsmParser.cpp

@@ -146,6 +146,9 @@ class MipsAsmParser : public MCTargetAsmParser {
   /// If true, then CpSaveLocation is a register, otherwise it's an offset.
   bool     CpSaveLocationIsRegister;
 
+  // Map of register aliases created via the .set directive.
+  StringMap<AsmToken> RegisterSets;
+
   // Print a warning along with its fix-it message at the given range.
   void printWarningWithFixIt(const Twine &Msg, const Twine &FixMsg,
                              SMRange Range, bool ShowColors = true);
@@ -183,6 +186,9 @@ class MipsAsmParser : public MCTargetAsmParser {
   OperandMatchResultTy
   matchAnyRegisterNameWithoutDollar(OperandVector &Operands,
                                     StringRef Identifier, SMLoc S);
+  OperandMatchResultTy matchAnyRegisterWithoutDollar(OperandVector &Operands,
+                                                     const AsmToken &Token,
+                                                     SMLoc S);
   OperandMatchResultTy matchAnyRegisterWithoutDollar(OperandVector &Operands,
                                                      SMLoc S);
   OperandMatchResultTy parseAnyRegister(OperandVector &Operands);
@@ -5991,6 +5997,19 @@ bool MipsAsmParser::searchSymbolAlias(OperandVector &Operands) {
           llvm_unreachable("Should never ParseFail");
       }
     }
+  } else if (Sym->isUnset()) {
+    // If symbol is unset, it might be created in the `parseSetAssignment`
+    // routine as an alias for a numeric register name.
+    // Lookup in the aliases list.
+    auto Entry = RegisterSets.find(Sym->getName());
+    if (Entry != RegisterSets.end()) {
+      OperandMatchResultTy ResTy =
+          matchAnyRegisterWithoutDollar(Operands, Entry->getValue(), S);
+      if (ResTy == MatchOperand_Success) {
+        Parser.Lex();
+        return true;
+      }
+    }
   }
 
   return false;
@@ -6060,10 +6079,8 @@ MipsAsmParser::matchAnyRegisterNameWithoutDollar(OperandVector &Operands,
 }
 
 OperandMatchResultTy
-MipsAsmParser::matchAnyRegisterWithoutDollar(OperandVector &Operands, SMLoc S) {
-  MCAsmParser &Parser = getParser();
-  auto Token = Parser.getLexer().peekTok(false);
-
+MipsAsmParser::matchAnyRegisterWithoutDollar(OperandVector &Operands,
+                                             const AsmToken &Token, SMLoc S) {
   if (Token.is(AsmToken::Identifier)) {
     LLVM_DEBUG(dbgs() << ".. identifier\n");
     StringRef Identifier = Token.getIdentifier();
@@ -6090,6 +6107,12 @@ MipsAsmParser::matchAnyRegisterWithoutDollar(OperandVector &Operands, SMLoc S) {
   return MatchOperand_NoMatch;
 }
 
+OperandMatchResultTy
+MipsAsmParser::matchAnyRegisterWithoutDollar(OperandVector &Operands, SMLoc S) {
+  auto Token = getLexer().peekTok(false);
+  return matchAnyRegisterWithoutDollar(Operands, Token, S);
+}
+
 OperandMatchResultTy
 MipsAsmParser::parseAnyRegister(OperandVector &Operands) {
   MCAsmParser &Parser = getParser();
@@ -6849,11 +6872,24 @@ bool MipsAsmParser::parseSetAssignment() {
     return reportParseError("unexpected token, expected comma");
   Lex(); // Eat comma
 
-  if (Parser.parseExpression(Value))
+  if (getLexer().is(AsmToken::Dollar) &&
+      getLexer().peekTok().is(AsmToken::Integer)) {
+    // Parse assignment of a numeric register:
+    //   .set r1,$1
+    Parser.Lex(); // Eat $.
+    RegisterSets[Name] = Parser.getTok();
+    Parser.Lex(); // Eat identifier.
+    getContext().getOrCreateSymbol(Name);
+  } else if (!Parser.parseExpression(Value)) {
+    // Parse assignment of an expression including
+    // symbolic registers:
+    //   .set  $tmp, $BB0-$BB1
+    //   .set  r2, $f2
+    MCSymbol *Sym = getContext().getOrCreateSymbol(Name);
+    Sym->setVariableValue(Value);
+  } else {
     return reportParseError("expected valid expression after comma");
-
-  MCSymbol *Sym = getContext().getOrCreateSymbol(Name);
-  Sym->setVariableValue(Value);
+  }
 
   return false;
 }

test/MC/Mips/mips_directives.s

@@ -52,11 +52,12 @@ $BB0_4:
     .set FPU_MASK,$f7
     .set  $tmp7, $BB0_4-$BB0_2
     .set f6,$f6
+    .set r1,$1
 # CHECK:    abs.s   $f6, $f7           # encoding: [0x46,0x00,0x39,0x85]
 # CHECK:    lui     $1, %hi($tmp7)     # encoding: [0x3c,0x01,A,A]
 # CHECK:                               #   fixup A - offset: 0, value: %hi($tmp7), kind: fixup_Mips_HI16
     abs.s  f6,FPU_MASK
-    lui $1, %hi($tmp7)
+    lui r1, %hi($tmp7)
 
 # CHECK:    .set mips32r2
 # CHECK:    ldxc1   $f0, $zero($5)     # encoding: [0x4c,0xa0,0x00,0x01]
@@ -89,4 +90,4 @@ $BB0_4:
 # CHECK:   balign   $5, $6, 3           # encoding: [0x7c,0xc5,0x1c,0x31]
    .set dspr2
    append   $7, $10, 2
-   balign   $5, $6, 3
+   balign   $5, $6, 3