Changes

Summary

1. [X86] EltsFromConsecutiveLoads - replace SmallBitVector with APInt (NFC). Minor refactor to simplify some incoming patches to improve broadcast loads.
2. Fix test to pass on LLP64 targets
3. Fix clang tests broken by r353547 that depend on InstrProf
4. [GlobalISel] Regex the opcodes in unit test to fix non-deterministic ordering Differential Revision: https://reviews.llvm.org/D57988
5. [CodeGen][X86] Don't scalarize vector saturating add/sub Now that we have vector support for [US](ADD|SUB)O we no longer need to scalarize when expanding [US](ADD|SUB)SAT. This matches what the cost model already does. Differential Revision: https://reviews.llvm.org/D57348
6. fix -Wextra-semi warnings
7. Make LIBCXX_STANDARD_VER configurable
8. [AArch64] Regenerate bswap tests
9. [X86] Add basic bitreverse/bswap combine tests Shows missing SimplifyDemandedBits support
10. [DAG] Add optional AllowUndefs to isNullOrNullSplat No change in default behaviour (AllowUndefs = false)
11. [DAGCombine] Simplify funnel shifts with undef/zero args to bitshifts Now that we have SimplifyDemandedBits support for funnel shifts (rL353539), we need to simplify funnel shifts back to bitshifts in cases where either argument has been folded to undef/zero. Differential Revision: https://reviews.llvm.org/D58009

Revision 353655 by rksimon: [X86] EltsFromConsecutiveLoads - replace SmallBitVector with APInt (NFC). Minor refactor to simplify some incoming patches to improve broadcast loads.
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/lib/Target/X86/X86ISelLowering.cpp (diff)	llvm.src/lib/Target/X86/X86ISelLowering.cpp
Revision 353654 by rnk: Fix test to pass on LLP64 targets
Change Type	Path in Repository	Path in Workspace
	/cfe/trunk/test/Analysis/osobjectcstylecastchecker_test.cpp (diff)	clang.src/test/Analysis/osobjectcstylecastchecker_test.cpp
Revision 353653 by rnk: Fix clang tests broken by r353547 that depend on InstrProf
Change Type	Path in Repository	Path in Workspace
	/cfe/trunk/test/CoverageMapping/unused_names.c (diff)	clang.src/test/CoverageMapping/unused_names.c
	/cfe/trunk/test/Profile/cxx-lambda.cpp (diff)	clang.src/test/Profile/cxx-lambda.cpp
	/cfe/trunk/test/Profile/cxx-rangefor.cpp (diff)	clang.src/test/Profile/cxx-rangefor.cpp
	/cfe/trunk/test/Profile/cxx-stmt-initializers.cpp (diff)	clang.src/test/Profile/cxx-stmt-initializers.cpp
	/cfe/trunk/test/Profile/cxx-templates.cpp (diff)	clang.src/test/Profile/cxx-templates.cpp
	/cfe/trunk/test/Profile/cxx-throws.cpp (diff)	clang.src/test/Profile/cxx-throws.cpp
	/cfe/trunk/test/Profile/cxx-virtual-destructor-calls.cpp (diff)	clang.src/test/Profile/cxx-virtual-destructor-calls.cpp
Revision 353652 by mgrang: [GlobalISel] Regex the opcodes in unit test to fix non-deterministic ordering Differential Revision: https://reviews.llvm.org/D57988
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/test/CodeGen/AArch64/GlobalISel/legalize-ext-csedebug-output.mir (diff)	llvm.src/test/CodeGen/AArch64/GlobalISel/legalize-ext-csedebug-output.mir
Revision 353651 by nikic: [CodeGen][X86] Don't scalarize vector saturating add/sub Now that we have vector support for [US](ADD|SUB)O we no longer need to scalarize when expanding [US](ADD|SUB)SAT. This matches what the cost model already does. Differential Revision: https://reviews.llvm.org/D57348
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/lib/CodeGen/SelectionDAG/TargetLowering.cpp (diff)	llvm.src/lib/CodeGen/SelectionDAG/TargetLowering.cpp
	/llvm/trunk/test/CodeGen/X86/sadd_sat.ll (diff)	llvm.src/test/CodeGen/X86/sadd_sat.ll
	/llvm/trunk/test/CodeGen/X86/sadd_sat_vec.ll (diff)	llvm.src/test/CodeGen/X86/sadd_sat_vec.ll
	/llvm/trunk/test/CodeGen/X86/ssub_sat.ll (diff)	llvm.src/test/CodeGen/X86/ssub_sat.ll
	/llvm/trunk/test/CodeGen/X86/ssub_sat_vec.ll (diff)	llvm.src/test/CodeGen/X86/ssub_sat_vec.ll
Revision 353650 by ericwf: fix -Wextra-semi warnings
Change Type	Path in Repository	Path in Workspace
	/libcxx/trunk/include/chrono (diff)	libcxx.src/include/chrono
Revision 353649 by ericwf: Make LIBCXX_STANDARD_VER configurable
Change Type	Path in Repository	Path in Workspace
	/libcxx/trunk/CMakeLists.txt (diff)	libcxx.src/CMakeLists.txt
	/libcxx/trunk/lib/CMakeLists.txt (diff)	libcxx.src/lib/CMakeLists.txt
Revision 353648 by rksimon: [AArch64] Regenerate bswap tests
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/test/CodeGen/AArch64/arm64-rev.ll (diff)	llvm.src/test/CodeGen/AArch64/arm64-rev.ll
Revision 353647 by rksimon: [X86] Add basic bitreverse/bswap combine tests Shows missing SimplifyDemandedBits support
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/test/CodeGen/X86/combine-bitreverse.ll	llvm.src/test/CodeGen/X86/combine-bitreverse.ll
	/llvm/trunk/test/CodeGen/X86/combine-bswap.ll	llvm.src/test/CodeGen/X86/combine-bswap.ll
Revision 353646 by rksimon: [DAG] Add optional AllowUndefs to isNullOrNullSplat No change in default behaviour (AllowUndefs = false)
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/include/llvm/CodeGen/SelectionDAGNodes.h (diff)	llvm.src/include/llvm/CodeGen/SelectionDAGNodes.h
	/llvm/trunk/lib/CodeGen/SelectionDAG/DAGCombiner.cpp (diff)	llvm.src/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
	/llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAG.cpp (diff)	llvm.src/lib/CodeGen/SelectionDAG/SelectionDAG.cpp
Revision 353645 by rksimon: [DAGCombine] Simplify funnel shifts with undef/zero args to bitshifts Now that we have SimplifyDemandedBits support for funnel shifts (rL353539), we need to simplify funnel shifts back to bitshifts in cases where either argument has been folded to undef/zero. Differential Revision: https://reviews.llvm.org/D58009
Change Type	Path in Repository	Path in Workspace
	/llvm/trunk/lib/CodeGen/SelectionDAG/DAGCombiner.cpp (diff)	llvm.src/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
	/llvm/trunk/test/CodeGen/X86/funnel-shift.ll (diff)	llvm.src/test/CodeGen/X86/funnel-shift.ll