[llvm] r174926 - [ms-inline asm] Add support for lexing hexidecimal integers with a [hH] suffix.

[Chad Rosier]

Author: mcrosier
Date: Mon Feb 11 19:00:01 2013
New Revision: 174926

URL: http://llvm.org/viewvc/llvm-project?rev=174926&view=rev
Log:
[ms-inline asm] Add support for lexing hexidecimal integers with a [hH] suffix.
Part of rdar://12470373

Added:
    llvm/trunk/test/MC/X86/intel-syntax-hex.s
Modified:
    llvm/trunk/lib/MC/MCParser/AsmLexer.cpp

Modified: llvm/trunk/lib/MC/MCParser/AsmLexer.cpp
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/lib/MC/MCParser/AsmLexer.cpp?rev=174926&r1=174925&r2=174926&view=diff
==============================================================================
--- llvm/trunk/lib/MC/MCParser/AsmLexer.cpp (original)
+++ llvm/trunk/lib/MC/MCParser/AsmLexer.cpp Mon Feb 11 19:00:01 2013
@@ -162,21 +162,43 @@ static void SkipIgnoredIntegerSuffix(con
     CurPtr += 3;
 }
 
+// Look ahead to search for first non-hex digit, if it's [hH], then we treat the
+// integer as a hexadecimal, possibly with leading zeroes.
+static unsigned doLookAhead(const char *&CurPtr, unsigned DefaultRadix) {
+  const char *FirstHex = 0;
+  const char *LookAhead = CurPtr;
+  while (1) {
+    if (isdigit(*LookAhead)) {
+      ++LookAhead;
+    } else if (isxdigit(*LookAhead)) {
+      if (!FirstHex)
+        FirstHex = LookAhead;
+      ++LookAhead;
+    } else {
+      break;
+    }
+  }
+  bool isHex = *LookAhead == 'h' || *LookAhead == 'H';
+  CurPtr = isHex || !FirstHex ? LookAhead : FirstHex;
+  if (isHex)
+    return 16;
+  return DefaultRadix;
+}
+
 /// LexDigit: First character is [0-9].
 ///   Local Label: [0-9][:]
 ///   Forward/Backward Label: [0-9][fb]
 ///   Binary integer: 0b[01]+
 ///   Octal integer: 0[0-7]+
-///   Hex integer: 0x[0-9a-fA-F]+
+///   Hex integer: 0x[0-9a-fA-F]+ or [0x]?[0-9][0-9a-fA-F]*[hH]
 ///   Decimal integer: [1-9][0-9]*
 AsmToken AsmLexer::LexDigit() {
   // Decimal integer: [1-9][0-9]*
   if (CurPtr[-1] != '0' || CurPtr[0] == '.') {
-    while (isdigit(*CurPtr))
-      ++CurPtr;
-
+    unsigned Radix = doLookAhead(CurPtr, 10);
+    bool isHex = Radix == 16;
     // Check for floating point literals.
-    if (*CurPtr == '.' || *CurPtr == 'e') {
+    if (!isHex && (*CurPtr == '.' || *CurPtr == 'e')) {
       ++CurPtr;
       return LexFloatLiteral();
     }
@@ -184,15 +206,20 @@ AsmToken AsmLexer::LexDigit() {
     StringRef Result(TokStart, CurPtr - TokStart);
 
     long long Value;
-    if (Result.getAsInteger(10, Value)) {
+    if (Result.getAsInteger(Radix, Value)) {
       // Allow positive values that are too large to fit into a signed 64-bit
       // integer, but that do fit in an unsigned one, we just convert them over.
       unsigned long long UValue;
-      if (Result.getAsInteger(10, UValue))
-        return ReturnError(TokStart, "invalid decimal number");
+      if (Result.getAsInteger(Radix, UValue))
+        return ReturnError(TokStart, !isHex ? "invalid decimal number" :
+                           "invalid hexdecimal number");
       Value = (long long)UValue;
     }
 
+    // Consume the [bB][hH].
+    if (Radix == 2 || Radix == 16)
+      ++CurPtr;
+
     // The darwin/x86 (and x86-64) assembler accepts and ignores ULL and LL
     // suffixes on integer literals.
     SkipIgnoredIntegerSuffix(CurPtr);
@@ -243,6 +270,10 @@ AsmToken AsmLexer::LexDigit() {
     if (StringRef(TokStart, CurPtr - TokStart).getAsInteger(0, Result))
       return ReturnError(TokStart, "invalid hexadecimal number");
 
+    // Consume the optional [hH].
+    if (*CurPtr == 'h' || *CurPtr == 'H')
+      ++CurPtr;
+
     // The darwin/x86 (and x86-64) assembler accepts and ignores ULL and LL
     // suffixes on integer literals.
     SkipIgnoredIntegerSuffix(CurPtr);
@@ -251,15 +282,17 @@ AsmToken AsmLexer::LexDigit() {
                     (int64_t)Result);
   }
 
-  // Must be an octal number, it starts with 0.
-  while (*CurPtr >= '0' && *CurPtr <= '9')
-    ++CurPtr;
-
-  StringRef Result(TokStart, CurPtr - TokStart);
+  // Either octal or hexidecimal.
   long long Value;
-  if (Result.getAsInteger(8, Value))
+  unsigned Radix = doLookAhead(CurPtr, 8);
+  StringRef Result(TokStart, CurPtr - TokStart);
+  if (Result.getAsInteger(Radix, Value))
     return ReturnError(TokStart, "invalid octal number");
 
+  // Consume the [hH].
+  if (Radix == 16)
+    ++CurPtr;
+
   // The darwin/x86 (and x86-64) assembler accepts and ignores ULL and LL
   // suffixes on integer literals.
   SkipIgnoredIntegerSuffix(CurPtr);

Added: llvm/trunk/test/MC/X86/intel-syntax-hex.s
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/MC/X86/intel-syntax-hex.s?rev=174926&view=auto
==============================================================================
--- llvm/trunk/test/MC/X86/intel-syntax-hex.s (added)
+++ llvm/trunk/test/MC/X86/intel-syntax-hex.s Mon Feb 11 19:00:01 2013
@@ -0,0 +1,26 @@
+// RUN: llvm-mc -triple x86_64-unknown-unknown -x86-asm-syntax=intel %s | FileCheck %s
+// rdar://12470373
+
+// Checks to make sure we parse the hexidecimal suffix properly.
+// CHECK: movl $10, %eax
+  mov eax, 10
+// CHECK: movl $16, %eax
+  mov eax, 10h
+// CHECK: movl $16, %eax
+  mov eax, 10H
+// CHECK: movl $4294967295, %eax
+  mov eax, 0ffffffffh
+// CHECK: movl $4294967295, %eax
+  mov eax, 0xffffffff
+// CHECK: movl $4294967295, %eax
+  mov eax, 0xffffffffh
+// CHECK: movl $15, %eax
+  mov eax, 0fh
+// CHECK: movl $162, %eax
+  mov eax, 0a2h
+// CHECK: movl $162, %eax
+  mov eax, 0xa2
+// CHECK: movl $162, %eax
+  mov eax, 0xa2h
+// CHECK: movl $674, %eax
+  mov eax, 2a2h