llvm-svn: 44257

llvm-svn: 44255

llvm-svn: 44254

llvm-svn: 44253

any sense it is important that ParamAttr::None gets
treated the same as not supplying an attribute at
all.  Rather than stripping ParamAttr::None out of
the list of attributes, assert if ParamAttr::None
is seen.  Fix up the bitcode reader which liked to
insert ParamAttr::None all over the place.  Patch
based on one by Török Edwin.

llvm-svn: 44250

"setcc" -> "icmp op" in comments. No functionality change.

llvm-svn: 44249

llvm-svn: 44248

block-level expressions are evaluated the same as regular expressions. Test
case provided by Nuno Lopes.

llvm-svn: 44247

"block-expressions" when visiting arbitrary expressions (via calls to
"Visit()").  This results in a refactoring where a dataflow analysis no
longer needs to always special case when handling block-expressions versus
non-block expressions.

Updated LiveVariables and UninitializedValues to conform to the slightly
altered interface of these visitor classes.

Thanks to Nuno Lopes for providing a test case that illustrated some
fundamental problems in the current design of the CFGXXXStmtVisitor classes
and how they were used.

llvm-svn: 44246

llvm-svn: 44244

Fix a couple of problems:
1. Don't assume the VT-1 is a VT that is half the size.
2. Treat vectors of FP in the vector path, not the FP path.

This has a couple of remaining problems before it will work with
the code in PR1811: the code below this change assumes that it can
use extload/shift/or to construct the result, which isn't right for
vectors.

This also doesn't handle vectors of 1 or vectors that aren't pow-2.

llvm-svn: 44243

llvm-svn: 44242

llvm-svn: 44240

llvm-svn: 44239

and ImmutableSet classes in the main LLVM libraries.

llvm-svn: 44237

llvm-svn: 44236

llvm-svn: 44235

llvm-svn: 44234

llvm-svn: 44233

llvm-svn: 44228

llvm-svn: 44227

llvm-svn: 44226

for correct propagation/update of liveness information within subexpressions
of Block-Level expressions.  Test case provided by Nuno Lopes.

llvm-svn: 44225

state that occurred in subexpressions of Block-Level expressions. Bug and fix
provided by Nuno Lopes.

llvm-svn: 44224

llvm-svn: 44222

llvm-svn: 44221

the StmtIterator referring to the initializers of a chain of Decls was equal
to the "end" iterator. The particular bug manifested when an iterator was
created on a chain of decls with no initializers.

Thanks to Nuno Lopes for reporting this bug and providing a patch.

llvm-svn: 44220

a valid but empty buffer if stdin is empty.

llvm-svn: 44219

llvm-svn: 44218

	call x86_fastcallcc void @func( i32* %X, i64 0 )
not:
	callx86_fastcallcc  void @func( i32* %X, i64 0 )

This fixes Codegen/X86/fast-cc-merge-stack-adj.ll

llvm-svn: 44217

llvm-svn: 44216

llvm-svn: 44215

drops a dependency on flex and lets us make future progress more 
easily.  Yay for 2 fewer .cvs files to make silly conflicts with.

llvm-svn: 44213

llvm-svn: 44212

llvm-svn: 44211

one dependency on flex and gets rid of two ".cvs" files.

llvm-svn: 44210

Now that we are passing back "free standing decls", make sure -ast-dump works like -ast-print.
Also added a cast to be safe...

llvm-svn: 44209

Make sure Sema::ParsedFreeStandingDeclSpec() returns a decl representing the type.
Adding basic printing to StmtPrinter::PrintRawDecl().

llvm-svn: 44208

llvm-svn: 44204

http://llvm.org/viewvc/llvm-project?rev=44199&view=rev

This patch completely broke serialization due to an invariant I assumed but
did not hold. The assumed invariant was that all pointer IDs emitted by a call
to BatchEmitOwnedPtrs would be consecutive. This is only the case if there has
been no forward references to an owned pointer (and hence already registered
with the Serializer object).

llvm-svn: 44203