llvm-svn: 203658

Summary:
Correct the match patterns and the lowerings that made the CodeGen tests pass despite the mistakes.

The original testcase that discovered the problem was SingleSource/UnitTests/SignlessType/factor.c in test-suite.
During review, we also found that some of the existing CodeGen tests were incorrect and fixed them:
* bitwise.ll: In bsel_v16i8 the IfSet/IfClear were reversed because bsel and bmnz have different operand orders and the test didn't correctly account for this. bmnz goes 'IfClear, IfSet, CondMask', while bsel goes 'CondMask, IfClear, IfSet'.
* vec.ll: In the cases where a bsel is emitted as a bmnz (they are the same operation with a different input tied to the result) the operands were in the wrong order.
* compare.ll and compare_float.ll: The bsel operand order was correct for a greater-than comparison, but a greater-than comparison instruction doesn't exist. Lowering this operation inverts the condition so the IfSet/IfClear need to be swapped to match.

The differences between BSEL, BMNZ, and BMZ and how they map to/from vselect are rather confusing. I've therefore added a note to MSA.txt to explain this in a single place in addition to the comments that explain each case.

Reviewers: matheusalmeida, jacksprat

Reviewed By: matheusalmeida

Differential Revision: http://llvm-reviews.chandlerc.com/D3028

llvm-svn: 203657

This is a compilation warning fix. No functional change.

llvm-svn: 203656

When the list of VFP registers to be saved was non-contiguous (so multiple
vpush/vpop instructions were needed) these were being ordered oddly, as in:
    vpush {d8, d9}
    vpush {d11}

This led to the layout in memory being [d11, d8, d9] which is ugly and doesn't
match the CFI_INSTRUCTIONs we're generating either (so Dwarf info would be
broken).

This switches the order of vpush/vpop (in both prologue and epilogue,
obviously) so that the Dwarf locations are correct again.

rdar://problem/16264856

llvm-svn: 203655

llvm-svn: 203654

llvm-svn: 203653

It can be used in multiple files now.

llvm-svn: 203652

This allows to use some code for mingw which was previously only
used for MSVC.

llvm-svn: 203651

They help fix mingw build.

llvm-svn: 203650

This is to help fix mingw build.

llvm-svn: 203649

if the thread is cancelled in pthread_cond_wait, it locks the mutex before
processing pthread_cleanup stack
but tsan was missing that, thus reporting false double-lock/wrong-unlock errors
see the test for details

llvm-svn: 203648

llvm-svn: 203647

llvm-svn: 203646

Caused by unknown tokens (e.g. "\n") not properly updating the state.

Example:
  const char* c = STRINGIFY(
  \na : b);

llvm-svn: 203645

In some cases the include is pushed "downstream" (or removed if
unused).

llvm-svn: 203644

llvm-svn: 203643

llvm-svn: 203642

llvm-svn: 203641

llvm-svn: 203640

This is unnecessary now that the flag handling has been fixed.  The flags will
be added properly in the main CMakeLists.txt after the config-ix inclusion which
performs the required check.

llvm-svn: 203639

Always use list(APPEND) as it will perform the desired action even if the list
is empty or previously unset.  The first set is harmless, however, the
subsequent set was overwriting the previous flag setup resulting in an improper
compilation command being generated.  This manifested as a build failure on
Linux when using cmake + ninja.

llvm-svn: 203638

add_definitions is meant for adding C preprocessor definitions.  Modern cmake
suggests use of the CMAKE_CXX_FLAGS for the purposes of pushing flags to the
compilation commands.  Simply switch to the modern form given that we are
already requiring a new enough cmake.

llvm-svn: 203637

llvm-svn: 203636

It seems gas can't handle CFI directives with VFP register names ("d12", etc.).
This broke us trying to build Chromium for Android after 201423.

A gas bug has been filed: https://sourceware.org/bugzilla/show_bug.cgi?id=16694

compnerd suggested making this conditional on whether we're using the integrated
assembler or not. I'll look into that in a follow-up patch.

Differential Revision: http://llvm-reviews.chandlerc.com/D3049

llvm-svn: 203635

llvm-svn: 203634

Unfortunately, XFAIL applies to the entire file.  Split the test into two
different units, enabling the proper application of XFAIL on the test which is
expected to fail on Windows.

llvm-svn: 203633

Reduce the stack usage as hopefully include paths are usually not too long.  Add
a test case for the path normalisation behaviour.

llvm-svn: 203632

THIRD TIME. Richard pointed out (again) that I'd switched the order of the instance variables; and thus failed to repair the ABI break. After this, I'm going to sit down and watch TV for the evening.

llvm-svn: 203631

When enabled, always validate the system headers when loading a module.
The end result of this is that when these headers change, we will notice
and rebuild the module.

llvm-svn: 203630

llvm-svn: 203628

This reverts commit 203504. We decided that it is better to disable this test on a per-buildbot basis in zorg rather than everywhere on Darwin.

llvm-svn: 203627

programs. Initialize lsdaEncoding to DW_EH_PE_omit and check for that
value to decide whether a value should be decoded.

llvm-svn: 203626

llvm-svn: 203625

This narrows the impact of r188833 after Dimitry pointed out that it's good to be
able to tell the difference between 'cc' and 'CC'.

llvm-svn: 203624

I'll fix this to not test assembly in a moment.

llvm-svn: 203623

llvm-svn: 203622

llvm-svn: 203621

llvm-svn: 203620

I could fold the callers into their one call site, but the indirection
(given how verbose choosing the section is) seemed helpful.

The use of a member function pointer's a bit "tricky", but seems limited
enough, the call sites are simple/clean/clear, and there's only one use.

llvm-svn: 203619

llvm-svn: 203618