Add simple optimization: check for (and skip) spaces and tabs immediately

before lexing a token.  This speeds the common case where tokens are
separated by small amount of whitespace.  This makes a slight but
reproducible positive effect lexing a preprocessed carbon.h.

llvm-svn: 38691

1 files changed, 9 insertions, 0 deletions

diff --git a/clang/Lex/Lexer.cpp b/clang/Lex/Lexer.cpp
index 9cf0dae0a0c..967d4e76758 100644
--- a/clang/Lex/Lexer.cpp
+++ b/clang/Lex/Lexer.cpp
@@ -928,6 +928,15 @@ LexNextToken:
   // CurPtr - Cache BufferPtr in an automatic variable.
   const char *CurPtr = BufferPtr;
 
+  // Small amounts of horizontal whitespace is very common between tokens.
+  if ((*CurPtr == ' ') || (*CurPtr == '\t')) {
+    ++CurPtr;
+    while ((*CurPtr == ' ') || (*CurPtr == '\t'))
+      ++CurPtr;
+    BufferPtr = CurPtr;
+    Result.SetFlag(LexerToken::LeadingSpace);
+  }
+  
   unsigned SizeTmp, SizeTmp2;   // Temporaries for use in cases below.
   
   // Read a character, advancing over it.