This is an automated email from the git hooks/post-receive script.
unknown user pushed a change to branch master in repository llvm.
from 615f3cc In visitSTORE, always use FindBetterChain, rather than only w [...] new 2c96583 Revert "In visitSTORE, always use FindBetterChain, rather tha [...]
The 1 revisions listed above as "new" are entirely new to this repository and will be described in separate emails. The revisions listed as "adds" were already present in the repository and have only been added to this reference.
Summary of changes: lib/CodeGen/SelectionDAG/DAGCombiner.cpp | 416 ++++++---- lib/CodeGen/TargetLoweringBase.cpp | 2 +- lib/Target/AMDGPU/AMDGPUISelLowering.cpp | 10 + test/CodeGen/AArch64/argument-blocks.ll | 2 +- test/CodeGen/AArch64/arm64-abi.ll | 5 +- test/CodeGen/AArch64/arm64-memset-inline.ll | 4 +- test/CodeGen/AArch64/ldst-opt.ll | 13 +- test/CodeGen/AArch64/merge-store.ll | 5 +- test/CodeGen/AArch64/vector_merge_dep_check.ll | 3 +- test/CodeGen/AMDGPU/debugger-insert-nops.ll | 24 +- test/CodeGen/AMDGPU/insert_vector_elt.ll | 5 +- test/CodeGen/AMDGPU/merge-stores.ll | 17 +- test/CodeGen/AMDGPU/private-element-size.ll | 12 +- test/CodeGen/AMDGPU/si-triv-disjoint-mem-access.ll | 17 +- test/CodeGen/ARM/2012-10-04-AAPCS-byval-align8.ll | 3 +- test/CodeGen/ARM/alloc-no-stack-realign.ll | 100 ++- test/CodeGen/ARM/ifcvt10.ll | 2 + test/CodeGen/ARM/memset-inline.ll | 10 +- test/CodeGen/ARM/static-addr-hoisting.ll | 6 +- test/CodeGen/BPF/undef.ll | 65 +- test/CodeGen/MSP430/Inst16mm.ll | 2 +- test/CodeGen/Mips/cconv/arguments-float.ll | 24 +- test/CodeGen/Mips/cconv/arguments-varargs.ll | 44 +- test/CodeGen/Mips/fastcc.ll | 76 +- test/CodeGen/Mips/load-store-left-right.ll | 126 ++- test/CodeGen/Mips/micromips-li.ll | 2 +- test/CodeGen/Mips/mips64-f128.ll | 2 +- test/CodeGen/Mips/mno-ldc1-sdc1.ll | 46 +- test/CodeGen/Mips/msa/f16-llvm-ir.ll | 14 +- test/CodeGen/Mips/msa/i5_ld_st.ll | 32 +- test/CodeGen/Mips/o32_cc_byval.ll | 54 +- test/CodeGen/Mips/o32_cc_vararg.ll | 4 +- test/CodeGen/PowerPC/anon_aggr.ll | 31 +- test/CodeGen/PowerPC/complex-return.ll | 12 +- test/CodeGen/PowerPC/jaggedstructs.ll | 16 +- test/CodeGen/PowerPC/ppc64-align-long-double.ll | 57 +- test/CodeGen/PowerPC/structsinmem.ll | 28 +- test/CodeGen/PowerPC/structsinregs.ll | 60 +- test/CodeGen/SystemZ/unaligned-01.ll | 5 +- test/CodeGen/Thumb/2010-07-15-debugOrdering.ll | 2 +- test/CodeGen/Thumb/stack-access.ll | 8 +- test/CodeGen/X86/2010-09-17-SideEffectsInChain.ll | 2 +- test/CodeGen/X86/2012-11-28-merge-store-alias.ll | 2 +- test/CodeGen/X86/MergeConsecutiveStores.ll | 14 +- test/CodeGen/X86/avx512-mask-op.ll | 4 + test/CodeGen/X86/chain_order.ll | 4 +- .../CodeGen/X86/clear_upper_vector_element_bits.ll | 56 +- test/CodeGen/X86/combiner-aa-0.ll | 20 + test/CodeGen/X86/combiner-aa-1.ll | 23 + test/CodeGen/X86/copy-eflags.ll | 17 +- test/CodeGen/X86/dag-merge-fast-accesses.ll | 12 +- .../X86/dont-trunc-store-double-to-float.ll | 6 +- .../extractelement-legalization-store-ordering.ll | 8 +- test/CodeGen/X86/i256-add.ll | 8 +- test/CodeGen/X86/i386-shrink-wrapping.ll | 5 +- test/CodeGen/X86/live-range-nosubreg.ll | 5 +- test/CodeGen/X86/merge-consecutive-loads-128.ll | 20 +- test/CodeGen/X86/merge-consecutive-loads-256.ll | 8 +- .../X86/merge-store-partially-alias-loads.ll | 8 +- test/CodeGen/X86/pr18023.ll | 31 + test/CodeGen/X86/split-store.ll | 20 +- test/CodeGen/X86/stores-merging.ll | 11 +- test/CodeGen/X86/vector-compare-results.ll | 730 ++++++++--------- test/CodeGen/X86/vector-shuffle-variable-128.ll | 906 ++++++++++++--------- test/CodeGen/X86/vector-shuffle-variable-256.ll | 232 +++--- test/CodeGen/X86/win32-eh.ll | 157 ++-- test/CodeGen/XCore/varargs.ll | 2 +- 67 files changed, 1995 insertions(+), 1682 deletions(-) create mode 100644 test/CodeGen/X86/combiner-aa-0.ll create mode 100644 test/CodeGen/X86/combiner-aa-1.ll create mode 100644 test/CodeGen/X86/pr18023.ll