llvm-svn: 63181

llvm-svn: 63180

llvm-svn: 63179

llvm-svn: 63178

llvm-svn: 63177

llvm-svn: 63176

Patch by Alexei Svitkine: Refactor Sema::ParseAST API to allow clients to pass as an argument a TranslationUnit object whose contents live beyond the call to ParseAST.

llvm-svn: 63175

accessibility of the ivar (related to objc2's
non-fragile abi).

llvm-svn: 63166

Changed section names for meta-data (to match current gcc).

llvm-svn: 63163

llvm-svn: 63162

Performance impact (clang -fsyntax-only INPUTS/Cocoa_h.m):
  non-PTH: 0.4% improvement
  PTH: 0.8% improvement

llvm-svn: 63159

llvm-svn: 63158

llvm-svn: 63157

snaroff% time ../../Release-Asserts/bin/clang INPUTS/Cocoa_h.m
0.179u 0.051s 0:00.23 95.6%	0+0k 0+0io 0pf+0w
snaroff% time ../../Release-Asserts/bin/clang INPUTS/Cocoa_h.m -disable-free
0.169u 0.052s 0:00.22 95.4%	0+0k 0+0io 0pf+0w

llvm-svn: 63153

llvm-svn: 63150

llvm-svn: 63141

Also changed FunctionTypeProto to be allocated with 8-byte alignment (noticed by Doug). I couldn't think of any reason to allocate on 16-byte boundaries. If anyone remembers why we were doing this, let me know!

llvm-svn: 63137

Remove many references to ASTContext::getAllocator(), replacing them with calls to the recently added placement new (which uses ASTContext's allocator for memory). Also added ASTContext::Deallocate().

This will simplify runtime replacement of ASTContext's allocator. Keeping the allocator private (and removing getAllocator() entirely) is also goodness.

llvm-svn: 63135

If an input constraint refers to an output constraint, it should have the same constraint info as the output constraint. Fixes PR3417

llvm-svn: 63127

Still more work to do in this area.

llvm-svn: 63126

llvm-svn: 63116

.def file for each library.  This means that adding a diagnostic
to sema doesn't require all the other libraries to be rebuilt.

Patch by Anders Johnsen!

llvm-svn: 63111

Fix bug in BasicStore::getLValueElement where if the base of an array subscript expression was an ElementRegion we stacked another ElementRegion on top of that.

This fixes PR 3422.

llvm-svn: 63110

of a type.  The implementation is currently something of a hack, but is 
sufficient for now and allows clients to be built on it.

llvm-svn: 63108

llvm-svn: 63101

__extension__.  This sort of construct shows up in the gcc source code.

llvm-svn: 63100

as reported to the user and as manipulated by #line.  This is what __FILE__,
__INCLUDE_LEVEL__, diagnostics and other things should follow (but not 
dependency generation!).  

This patch also includes several cleanups along the way: 

- SourceLocation now has a dump method, and several other places 
  that did similar things now use it.
- I cleaned up some code in AnalysisConsumer, but it should probably be
  simplified further now that NamedDecl is better.
- TextDiagnosticPrinter is now simplified and cleaned up a bit.

This patch is a prerequisite for #line, but does not actually provide 
any #line functionality.

llvm-svn: 63098

llvm-svn: 63097

llvm-svn: 63094

address space we used up.  Some interesting data:

For c99-intconst-1.c:
6912762 SLocEntry's allocated, 25592386B of Sloc address space used.

For cocoa.h:
26469 SLocEntry's allocated, 10278752B of Sloc address space used.

For carbon.h:
27364 SLocEntry's allocated, 12398141B of Sloc address space used.

Clearly 2G of sloc address space should be enough for anyone?!

llvm-svn: 63093

llvm-svn: 63092

 - gcc appears to be classifying <1 x double> as INTEGER which is
   odd. Will investigate later.

llvm-svn: 63086

llvm-svn: 63085

- Add the distinction between the 'bug type' and the 'bug description'

HTMLDiagnostics:
- Output the bug type field as HTML comments

scan-build:
- Use the bug type field instead of the bug description for the HTML table.
- Radar filing now automatically picks up the bug description in the title (addresses <rdar://problem/6265970>)

llvm-svn: 63084

 - Code quality is poor, but simple.

llvm-svn: 63083

PTH: Use Token::setLiteralData() to directly store a pointer to cached spelling data in the PTH file.  This removes a ton of code for looking up spellings using sourcelocations in the PTH file.  This simplifies both PTH-generation and reading.

Performance impact for -fsyntax-only on Cocoa.h (with Cocoa.h in the PTH file):
- PTH generation time improves by 5%
- PTH reading improves by 0.3%.

llvm-svn: 63072

llvm-svn: 63071

case correctly.

llvm-svn: 63068

as well (for nonfragile-abi).

llvm-svn: 63062

  - When it's safe, ActionResult uses the low bit of the pointer for
  the "invalid" flag rather than a separate "bool" value. This keeps
  GCC from generating some truly awful code, for a > 3x speedup in the
  result-passing microbenchmark.
  - When DISABLE_SMART_POINTERS is defined, store an ActionResult
  within ASTOwningResult rather than an ASTOwningPtr. Brings the
  performance benefits of the above to smart pointers with
  DISABLE_SMART_POINTERS defined.

Sadly, these micro-benchmark performance improvements don't seem to
make much of a difference on Cocoa.h right now. However, they're
harmless and might help with future optimizations.

llvm-svn: 63061