Cleanup. Document. Make sure that this build_vector optimization only runs before the op legalizer and that the used type is legal.

llvm-svn: 143358

The _fixed variants have a writeback operand, but not a stride operand.
Split the conditional flag to distinguish the cases.

llvm-svn: 143356

Example:

[11:33:09] johnny:/Volumes/data/lldb/svn/trunk/test $ ./dosep.ty -o "-v -n"
dotest.py options: -v -n
Running /Volumes/data/lldb/svn/trunk/test/dotest.py -v -n -p TestPublicAPIHeaders.py /Volumes/data/lldb/svn/trunk/test/api/check_public_api_headers
1: test_sb_api_directory (TestPublicAPIHeaders.SBDirCheckerCase)
   Test the SB API directory and make sure there's no unwanted stuff. ... ok

----------------------------------------------------------------------
Ran 1 test in 4.404s

OK
Running /Volumes/data/lldb/svn/trunk/test/dotest.py -v -n -p TestEmulations.py /Volumes/data/lldb/svn/trunk/test/arm_emulation
1: test_arm_emulations (TestEmulations.ARMEmulationTestCase) ... ok
2: test_thumb_emulations (TestEmulations.ARMEmulationTestCase) ... ok

----------------------------------------------------------------------
Ran 2 tests in 1.399s

OK

...

llvm-svn: 143355

llvm-svn: 143353

llvm-svn: 143352

llvm-svn: 143351

In x86_64, when calling an Objective-C method that returns a _Complex long double, make sure to use the objc_msgSend_fp2ret function which ensures that the return value will be {0, 0} if the receiver is nil.

llvm-svn: 143350

llvm-svn: 143349

llvm-svn: 143348

Useful to set up a virtual include hierarchy, e.g. to order diagnostics for ambiguous overloads from MemoryBuffers.

llvm-svn: 143347

I don't have any Debian system with one of these currently, and it seems
unlikely for one to show up suddenly. We can add more patterns here if
they become necessary.

llvm-svn: 143346

search logic. The Debian multiarch seems to have completely changed from
when these were originally added, and I'd like to remove a bunch of
them, but I'll be lazy and delay that until this logic is hoisted into
the driver where it belongs.

This should resolve PR11223.

llvm-svn: 143345

library search logic to "properly" handle multiarch installations. I've
tested this on both Debian unstable and the latest Ubuntu which both use
this setup, and this appears to work largely the same way as GCC does.
It isn't exactly the same, but it is close enough and more principled in
its behavior where it differs. This should resolve any failures to find
'crt1.o' etc on Debian-based Linux distributions. If folks find more
cases where we fail, please file bugs and CC me.

Test cases for all of the debian silliness are waiting both to simplify
the process of merging these down into the 3.0 release, and because
they're so crazy I haven't yet been able to really produce a fake tree
that represents what we need to test for. I'll eventually add them
though.

llvm-svn: 143344

llvm-svn: 143343

so that it will happen when we are doing code-completion, not reparsing.

llvm-svn: 143342

make sure that we keep track of locations of replaced decls as well.

llvm-svn: 143341

local DeclIDs came from.

llvm-svn: 143340

llvm-svn: 143339

that contains them.

llvm-svn: 143338

llvm-svn: 143337

llvm-svn: 143336

materialized temporaries.

llvm-svn: 143335

arbitrary pointers, if those pointers don't point to weak objects or literals.

llvm-svn: 143334

llvm-svn: 143332

llvm-svn: 143331

Another batch of ternary operators and some whitespace fixes
(Getting in sync with the clang_complete version of this file)

llvm-svn: 143330

result is not unspecified.

llvm-svn: 143329

llvm-svn: 143328

llvm-svn: 143327

-enable-dwarf-directory.

llvm-svn: 143326

This change is necessary to make this file python 2.4 compatible.

llvm-svn: 143324

instead of replacing it.

llvm-svn: 143323

We add a constructor to create a SourceLocation from a position in
a file and we use this SourceLocation to retrieve a cursor.

llvm-svn: 143322

llvm-svn: 143321

llvm-svn: 143320

llvm-svn: 143319

with the given predicate, it matches any condition and returns the
predicate - d'oh!  Original commit message:
The expression icmp eq (select (icmp eq x, 0), 1, x), 0 folds to false.
Spotted by my super-optimizer in 186.crafty and 450.soplex.  We really
need a proper infrastructure for handling generalizations of this kind
of thing (which occur a lot), however this case is so simple that I decided
to go ahead and implement it directly.

llvm-svn: 143318

llvm-svn: 143317

llvm-svn: 143316

X86: Emit logical shift by constant splat of <16 x i8> as a <8 x i16> shift and zero out the bits where zeros should've been shifted in.

llvm-svn: 143315