on x86-64 linux.  This causes no regressions on
32 bit linux and 32 bit ppc.  More tests pass
on 64 bit ppc with no regressions.  I didn't
turn on eh on 64 bit linux because the intrinsics
needed to compile the eh runtime aren't done
yet.  But if you turn it on and link with the
mainline runtime then eh seems to work fine
on x86-64 linux with this patch.  Thanks to
Dale for testing.  The main point of the patch
is that if you output that some object is
encoded using 4 bytes you had better not output
8 bytes for it: the patch makes everything
consistent.

llvm-svn: 50825

llvm-svn: 50824

llvm-svn: 50823

Correctly check if the result of CStrInCStrNoCase is NULL to generate summaries; before we were inverting the condition.

llvm-svn: 50822

llvm-svn: 50821

Currently is sufficient to describe mod/ref behavior but will hopefully
eventually be extended for other purposes.

This isn't used by anything yet.

llvm-svn: 50820

properties. Couple of property tests will fail with this patch.
Will fix them next.

llvm-svn: 50818

llvm-svn: 50817

llvm-svn: 50816

llvm-svn: 50815

llvm-svn: 50805

Do not treat **instance** methods "copyWithZone:" and "mutableCopyWithZone:" from NSObject as allocators.

llvm-svn: 50802

llvm-svn: 50799

llvm-svn: 50798

llvm-svn: 50795

%ecx = op
store %cl<kill>, (addr)
(addr) = op %al

It's not safe to unfold the last operand and eliminate store even though %cl is marked kill. It's a sub-register use which means one of its super-register(s) may be used below.

llvm-svn: 50794

llvm-svn: 50793

llvm-svn: 50790

llvm-svn: 50789

llvm-svn: 50787

llvm-svn: 50786

llvm-svn: 50784

llvm-svn: 50783

fixed, revert this patch.

llvm-svn: 50782

Fix <rdar://problem/5879237> clang objc rewriter: ivars not accessible in structure generated for class

llvm-svn: 50781

llvm-svn: 50780

llvm-svn: 50779

llvm-svn: 50778

llvm-svn: 50777

we will need to strip all casts for intializer lookup.

llvm-svn: 50776

llvm-svn: 50775

Fix <rdar://problem/5888515> clang ObjC rewriter: Use objc.h for declarations instead of putting them in the preamble?

llvm-svn: 50774

PathDiagnosticClient of where to display a string (beyond just the SourceLocation).

llvm-svn: 50773

"#if 0" out some assumptions when auto-generating method summaries.

llvm-svn: 50772

llvm-svn: 50771

We need to update PHIs containing the exiting block, not the exit block.  We really should come up with better names for these.

llvm-svn: 50770

llvm-svn: 50769

llvm-svn: 50768

Fix <rdar://problem/5881225> clang ObjC Rewriter: Protocol structure defined twice when two interfaces use it in same file

llvm-svn: 50767

llvm-svn: 50766