This is an automated email from the git hooks/post-receive script.
tcwg-buildslave pushed a change to branch linaro-local/ci/tcwg_kernel/llvm-master-arm-mainline-allyesconfig in repository toolchain/ci/llvm-project.
from 6280bc1cc34 [Flang][openmp][5.0] Add task_reduction clause. adds 05e6ac4eb81 [IROutliner] Removing a duplicate addition, causing overest [...] adds 5c951623bc8 [IROutliner] Refactoring errors in the cost model from past [...] adds f6515b05205 [PowerPC] Do not fold `cmp(d|w)` and `subf` instruction to [...] adds 09b3f3f22cb [benchmark] Fixed a build error when using CMake 3.15.1 + NDK-R20 adds 94257d12cb2 [RISCV] Remove unused method isUImm5NonZero() from RISCVAsm [...] adds a65092040ad [SVE] Fix inline assembly parsing crash adds 74e7cb26b9a [VE] Remove VA.needsCustom checks adds e43b3d1f5e0 Revert "[Sema] Fix deleted function problem in implicitly m [...] adds e0905553b42 [ArgPromotion] Delay dead GEP removal until doPromotion. adds 685c8b537af [AARCH64] Improve accumulator forwarding for Cortex-A57 model adds c287f90ccd3 [VE] Change default CPU name to "generic" adds 6c89f6fae49 [AArch64] Attempt to fix Mac tests with a more specific tri [...] adds 975b64b2937 [docs] Release notes for IsDecl in DIModule. adds 42652c1d6e2 [Sparc] Fixes for the internal assembler adds 901cc9b6f30 [ARM] Extend lowering for i64 reductions adds 23b41986527 [Support] Add KnownBits::icmp helpers. adds d38a0258a5f [AArch64] Add patterns for FMCLA*_indexed. adds 060cfd97954 [AArch64][SVE]Add cost model for masked gather and scatter [...] adds 4d7cb6da9fc [Sparc] SparcMCExpr::printVariantKind - fix Wcovered-switch [...] adds 82a29a62aba [OpenMP] Add definition/interface for target memory routines adds 9f8c0d15c7f DeclCXX - Fix getAs<> null-dereference static analyzer warn [...] adds e9f401d8a26 [IR] CallBase::getBundleOpInfoForOperand - ensure Current i [...] adds ed936aad781 [InterleavedAccess] Return correct 'modified' status. adds e2d3d501ef8 [RISCV][NFC] Add additional cmov tests adds c367258b5cc [SimplifyCFG] Enabled hoisting late in LTO pipeline. adds c55b609b777 [Hexagon] Fix bad SDNodeXForm adds 76bfbb74d38 [libomptarget][amdgpu] Call into deviceRTL instead of ockl adds f7463ca3cc5 [ProfileData] GCOVFile::readGCNO - silence undefined pointe [...] adds fe5d51a4897 [OpenMP] Add using bit flags to select Libomptarget Information adds dd6bb367d19 [LoopDeletion] Break backedge of loops when known not taken adds 7c63aac7bd4 Revert "[LoopDeletion] Break backedge of loops when known n [...] adds d8938c8bb54 CodeGen: Use Register adds 6976812129b [InstCombine] add tests for ashr+icmp; NFC adds dc9ac0e8207 [RISCV] Replace i32 with XLenVT in (add AddrFI, simm12) ise [...] adds b8f22f9d300 [NewPM][AMDGPU] Run InternalizePass when -amdgpu-internaliz [...] adds 848e8f938fd [llvm] Construct SmallVector with iterator ranges (NFC) adds 0edbc90ec56 [DebugInfo] Use llvm::append_range (NFC) adds eb198f4c3ce [llvm] Use llvm::any_of (NFC) adds a5f863e0765 [NewPM][AMDGPU] Port amdgpu-propagate-attributes-early/late adds e1833e7493a [NewPM][AMDGPU] Port amdgpu-unify-metadata adds 9a17bff4f71 [LoopNest] Allow empty basic blocks without loops adds 4034f9273ed Switching Clang UniqueInternalLinkageNamesPass scheduling t [...] adds 4d0aad96e43 [flang][openmp] Make Reduction clause part of OmpClause adds fe597efc30b [RISCV] Remove unused method RISCVInstPrinter::printSImm5Pl [...] adds fd323a897c6 [NewPM][AMDGPU] Port amdgpu-printf-runtime-binding adds 4e838ba9ea2 [NewPM][AMDGPU] Port amdgpu-always-inline adds 191552344bb [NewPM][AMDGPU] Make amdgpu-aa work with NewPM adds de6d43f16cb Revert "[LoopNest] Allow empty basic blocks without loops" adds 92be640bd7d [FPEnv][AMDGPU] Disable FSUB(-0,X)->FNEG(X) DAGCombine when [...] adds 58b6c5d932a [LoopUtils] reorder logic for creating reduction; NFC adds 97669575241 [LoopUtils] reduce code for creatng reduction; NFC adds aa169033892 [test] Pin backedge-id-bug-xfail.ll to legacy PM adds 8e293fe6ad0 [NewPM][AMDGPU] Pass TargetMachine to AMDGPUSimplifyLibCallsPass adds abbef2fd46d [ValueTracking] isGuaranteedNotToBePoison should return tru [...] adds efc82c4ad2b [NFC, Refactor] Modernize StorageClass from Specifiers.h to [...] adds 36263a7cccc [LoopUtils] remove redundant opcode parameter; NFC adds 2fd11e0b1ef Revert "[NFC, Refactor] Modernize StorageClass from Specifi [...] adds f67d3dbdb93 [clang] - Also look for devtoolset-10 adds b4f519bddda [NFCI] DwarfEHPrepare: update DomTree in non-permissive mod [...] adds 3fb57222c4c [NFCI] SimplifyCFG: switch to non-permissive DomTree update [...] adds ed9de61cc3e [SimplifyCFGPass] mergeEmptyReturnBlocks(): switch to non-p [...] adds a8604e3d5b7 [SimplifyCFG] simplifyIndirectBr(): switch to non-permissiv [...] adds 110b3d7855e [SimplifyCFG] SimplifyEqualityComparisonWithOnlyPredecessor [...] adds 32c47ebef18 [SimplifyCFG] SimplifyCondBranchToTwoReturns(): switch to n [...] adds e30fbbe9a53 [JumpThreading][NewPM] Skip when target has divergent CF adds c4f12a07a44 [WebAssembly] Remove old SDT_WebAssemblyCalls (NFC) adds f28b026d32c [InstSimplify] add a test for gep with poison operand (NFC) adds f665a8c5b8b [InstSimplify] gep with poison operand is poison adds ae614851631 [UpdateTestChecks] Fix PowerPC RE to support AIX assembly adds d51d72bbb91 [RISCV] Rename RVV intrinsics class (NFC) adds 2962f1149c8 [NFC] Add the getSizeInBytes() interface for MachineConstan [...] adds 48340fbe6a1 [NFC] [PowerPC] Update vec_constants test to reflect more patterns new b6c8feb29fc [NFC] [PowerPC] Remove dead code in BUILD_VECTOR peephole
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: clang/lib/AST/DeclCXX.cpp | 6 +- clang/lib/CodeGen/BackendUtil.cpp | 6 +- clang/lib/CodeGen/CGOpenMPRuntimeAMDGCN.cpp | 7 +- clang/lib/Driver/ToolChains/Gnu.cpp | 1 + clang/lib/Sema/SemaInit.cpp | 67 +-- clang/lib/Sema/SemaStmt.cpp | 36 +- .../CXX/class/class.init/class.copy.elision/p3.cpp | 50 -- clang/test/OpenMP/amdgcn_target_codegen.cpp | 10 +- flang/lib/Parser/openmp-parsers.cpp | 4 +- flang/lib/Parser/unparse.cpp | 2 - flang/lib/Semantics/check-omp-structure.cpp | 2 +- flang/lib/Semantics/check-omp-structure.h | 2 +- llvm/docs/ReleaseNotes.rst | 6 + llvm/include/llvm/Analysis/IVDescriptors.h | 9 +- llvm/include/llvm/Analysis/TargetTransformInfo.h | 11 +- .../llvm/Analysis/TargetTransformInfoImpl.h | 5 +- llvm/include/llvm/CodeGen/BasicTTIImpl.h | 6 +- llvm/include/llvm/CodeGen/LiveInterval.h | 7 +- llvm/include/llvm/CodeGen/MachineConstantPool.h | 6 +- llvm/include/llvm/CodeGen/MachineFrameInfo.h | 7 +- llvm/include/llvm/DebugInfo/CodeView/TypeRecord.h | 2 +- llvm/include/llvm/Frontend/OpenMP/OMP.td | 2 +- llvm/include/llvm/IR/DebugInfoMetadata.h | 5 +- llvm/include/llvm/IR/IntrinsicsRISCV.td | 16 +- llvm/include/llvm/IR/Metadata.h | 3 +- llvm/include/llvm/IR/PredIteratorCache.h | 2 +- llvm/include/llvm/Support/KnownBits.h | 31 ++ llvm/include/llvm/Transforms/Utils/LoopUtils.h | 3 +- llvm/lib/Analysis/IVDescriptors.cpp | 5 +- llvm/lib/Analysis/InstructionSimplify.cpp | 9 +- llvm/lib/Analysis/ScalarEvolution.cpp | 16 +- llvm/lib/Analysis/TargetTransformInfo.cpp | 4 + llvm/lib/Analysis/ValueTracking.cpp | 2 +- llvm/lib/CodeGen/AsmPrinter/AsmPrinter.cpp | 3 +- llvm/lib/CodeGen/DwarfEHPrepare.cpp | 2 +- llvm/lib/CodeGen/InterleavedAccessPass.cpp | 25 +- llvm/lib/CodeGen/MachineFunction.cpp | 12 +- llvm/lib/CodeGen/MachineSink.cpp | 6 +- llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp | 21 +- llvm/lib/DebugInfo/DWARF/DWARFDie.cpp | 3 +- llvm/lib/DebugInfo/MSF/MSFBuilder.cpp | 6 +- llvm/lib/DebugInfo/PDB/Native/NamedStreamMap.cpp | 2 +- llvm/lib/DebugInfo/PDB/Native/SymbolCache.cpp | 3 +- llvm/lib/DebugInfo/PDB/Native/TpiStreamBuilder.cpp | 2 +- .../ExecutionEngine/Orc/TargetProcessControl.cpp | 8 +- llvm/lib/IR/Constants.cpp | 2 +- llvm/lib/IR/Instructions.cpp | 2 +- llvm/lib/IR/Verifier.cpp | 4 +- llvm/lib/Passes/PassBuilder.cpp | 10 +- llvm/lib/ProfileData/GCOV.cpp | 2 +- llvm/lib/Support/KnownBits.cpp | 69 +++ llvm/lib/Support/SourceMgr.cpp | 2 +- llvm/lib/TableGen/Record.cpp | 4 +- llvm/lib/Target/AArch64/AArch64AsmPrinter.cpp | 3 +- llvm/lib/Target/AArch64/AArch64InstrInfo.cpp | 2 +- llvm/lib/Target/AArch64/AArch64InstrInfo.td | 21 + llvm/lib/Target/AArch64/AArch64SchedA57.td | 61 ++- llvm/lib/Target/AArch64/AArch64SchedA57WriteRes.td | 19 + .../Target/AArch64/AArch64TargetTransformInfo.cpp | 20 + .../Target/AArch64/AArch64TargetTransformInfo.h | 11 + llvm/lib/Target/AMDGPU/AMDGPU.h | 40 ++ llvm/lib/Target/AMDGPU/AMDGPUAliasAnalysis.cpp | 2 + llvm/lib/Target/AMDGPU/AMDGPUAliasAnalysis.h | 7 +- llvm/lib/Target/AMDGPU/AMDGPUAlwaysInlinePass.cpp | 20 +- llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp | 2 +- .../Target/AMDGPU/AMDGPUPrintfRuntimeBinding.cpp | 77 ++- .../Target/AMDGPU/AMDGPUPropagateAttributes.cpp | 19 + llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.cpp | 70 ++- llvm/lib/Target/AMDGPU/AMDGPUTargetMachine.h | 1 + llvm/lib/Target/AMDGPU/AMDGPUUnifyMetadata.cpp | 59 +-- llvm/lib/Target/ARM/ARMBaseInstrInfo.cpp | 2 +- llvm/lib/Target/ARM/ARMConstantIslandPass.cpp | 2 +- llvm/lib/Target/ARM/ARMISelLowering.cpp | 53 +- llvm/lib/Target/Hexagon/HexagonPatterns.td | 2 +- llvm/lib/Target/Mips/MipsConstantIslandPass.cpp | 2 +- llvm/lib/Target/PowerPC/PPCISelLowering.cpp | 11 - llvm/lib/Target/PowerPC/PPCInstrInfo.cpp | 8 + llvm/lib/Target/RISCV/AsmParser/RISCVAsmParser.cpp | 12 +- .../Target/RISCV/MCTargetDesc/RISCVInstPrinter.cpp | 9 - .../Target/RISCV/MCTargetDesc/RISCVInstPrinter.h | 2 - llvm/lib/Target/RISCV/RISCVFrameLowering.cpp | 2 +- llvm/lib/Target/RISCV/RISCVInstrInfo.td | 8 +- llvm/lib/Target/RISCV/RISCVInstrInfoV.td | 1 - llvm/lib/Target/Sparc/AsmParser/SparcAsmParser.cpp | 158 +++++- .../Target/Sparc/MCTargetDesc/SparcInstPrinter.cpp | 30 +- .../Sparc/MCTargetDesc/SparcMCCodeEmitter.cpp | 52 +- llvm/lib/Target/Sparc/MCTargetDesc/SparcMCExpr.cpp | 73 +-- llvm/lib/Target/Sparc/MCTargetDesc/SparcMCExpr.h | 1 + llvm/lib/Target/Sparc/SparcAsmPrinter.cpp | 2 +- llvm/lib/Target/Sparc/SparcISelLowering.cpp | 6 +- llvm/lib/Target/Sparc/SparcInstr64Bit.td | 6 +- llvm/lib/Target/Sparc/SparcInstrFormats.td | 4 +- llvm/lib/Target/Sparc/SparcInstrInfo.td | 25 +- llvm/lib/Target/VE/MCTargetDesc/VEMCTargetDesc.cpp | 2 +- llvm/lib/Target/VE/VE.td | 2 +- llvm/lib/Target/VE/VEISelLowering.cpp | 15 +- llvm/lib/Target/VE/VESubtarget.cpp | 2 +- .../lib/Target/WebAssembly/WebAssemblyInstrInfo.td | 2 - llvm/lib/Target/X86/X86MCInstLower.cpp | 5 +- llvm/lib/Transforms/IPO/ArgumentPromotion.cpp | 18 +- llvm/lib/Transforms/IPO/IROutliner.cpp | 4 +- llvm/lib/Transforms/IPO/PassManagerBuilder.cpp | 6 +- llvm/lib/Transforms/Scalar/JumpThreading.cpp | 4 + llvm/lib/Transforms/Scalar/SimplifyCFGPass.cpp | 7 +- llvm/lib/Transforms/Utils/LoopUtils.cpp | 112 ++--- llvm/lib/Transforms/Utils/SimplifyCFG.cpp | 107 ++-- llvm/lib/Transforms/Vectorize/LoopVectorize.cpp | 6 +- llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp | 6 +- llvm/lib/Transforms/Vectorize/VPlan.cpp | 2 +- .../AArch64/sve-getIntrinsicInstrCost-gather.ll | 37 ++ .../AArch64/sve-getIntrinsicInstrCost-scatter.ll | 40 ++ .../AArch64/inline-asm-constraints-bad-sve.ll | 9 + llvm/test/CodeGen/AArch64/neon-vcmla.ll | 95 ++++ llvm/test/CodeGen/AMDGPU/amdgpu-alias-analysis.ll | 2 + llvm/test/CodeGen/AMDGPU/clamp-modifier.ll | 2 +- llvm/test/CodeGen/AMDGPU/clamp.ll | 4 +- llvm/test/CodeGen/AMDGPU/fdiv32-to-rcp-folding.ll | 8 +- llvm/test/CodeGen/AMDGPU/fma-combine.ll | 4 +- llvm/test/CodeGen/AMDGPU/fneg-combines.ll | 42 +- .../force-alwaysinline-lds-global-address.ll | 2 + llvm/test/CodeGen/AMDGPU/fpext-free.ll | 2 +- llvm/test/CodeGen/AMDGPU/fptrunc.f16.ll | 4 +- llvm/test/CodeGen/AMDGPU/internalize.ll | 4 +- llvm/test/CodeGen/AMDGPU/known-never-snan.ll | 2 +- llvm/test/CodeGen/AMDGPU/llvm.amdgcn.fmad.ftz.ll | 4 +- .../CodeGen/AMDGPU/llvm.amdgcn.wavefrontsize.ll | 1 + llvm/test/CodeGen/AMDGPU/mad-combine.ll | 6 +- llvm/test/CodeGen/AMDGPU/mad-mix.ll | 8 +- llvm/test/CodeGen/AMDGPU/opencl-printf.ll | 1 + .../CodeGen/AMDGPU/propagate-attributes-clone.ll | 2 + .../propagate-attributes-flat-work-group-size.ll | 1 + .../AMDGPU/propagate-attributes-single-set.ll | 1 + llvm/test/CodeGen/AMDGPU/rcp-pattern.ll | 4 +- llvm/test/CodeGen/AMDGPU/rsq.ll | 2 +- llvm/test/CodeGen/AMDGPU/unify-metadata.ll | 1 + llvm/test/CodeGen/AMDGPU/v_mac.ll | 6 +- llvm/test/CodeGen/AMDGPU/v_mac_f16.ll | 12 +- .../CodeGen/Hexagon/isel-splat-vector-neg-i8.ll | 16 + llvm/test/CodeGen/PowerPC/aix-lr.ll | 40 +- llvm/test/CodeGen/PowerPC/pr47830.ll | 5 +- llvm/test/CodeGen/PowerPC/vec_constants.ll | 89 ++-- llvm/test/CodeGen/RISCV/rv32Zbt.ll | 560 +++++++++++++++++---- llvm/test/CodeGen/RISCV/rv64Zbt.ll | 260 +++++++++- llvm/test/CodeGen/RISCV/vararg.ll | 48 +- llvm/test/CodeGen/Thumb2/mve-vecreduce-add.ll | 122 +---- llvm/test/CodeGen/Thumb2/mve-vecreduce-addpred.ll | 200 +------- llvm/test/CodeGen/Thumb2/mve-vecreduce-mla.ll | 298 +---------- llvm/test/CodeGen/Thumb2/mve-vecreduce-mlapred.ll | 534 ++------------------ llvm/test/CodeGen/VE/Scalar/cpu.ll | 5 + llvm/test/MC/Sparc/sparc-asm-errors.s | 6 + llvm/test/MC/Sparc/sparc-ctrl-instructions.s | 7 + llvm/test/MC/Sparc/sparc-relocations.s | 4 + .../ArgumentPromotion/dead-gep-no-promotion.ll | 30 ++ llvm/test/Transforms/IROutliner/opt-remarks.ll | 108 +++- llvm/test/Transforms/IROutliner/outlining-calls.ll | 2 +- .../outlining-compatible-and-attribute-transfer.ll | 2 +- .../outlining-compatible-or-attribute-transfer.ll | 2 +- .../Transforms/IROutliner/outlining-cost-model.ll | 8 + .../IROutliner/outlining-different-constants.ll | 2 +- .../IROutliner/outlining-different-structure.ll | 2 +- .../IROutliner/outlining-isomorphic-predicates.ll | 2 +- llvm/test/Transforms/InstCombine/icmp-shr.ll | 419 +++++++++++++-- llvm/test/Transforms/InstSimplify/gep.ll | 12 +- .../X86/interleave-load-extract-shuffle-changes.ll | 58 +++ .../JumpThreading/divergent-target-test.ll | 2 + .../StructurizeCFG/AMDGPU/backedge-id-bug-xfail.ll | 2 +- .../Inputs/ppc_generated_funcs.ll | 1 + .../ppc_generated_funcs.ll.generated.expected | 60 +++ .../ppc_generated_funcs.ll.nogenerated.expected | 60 +++ .../tools/llvm-mca/AArch64/Cortex/forwarding-A57.s | 501 ++++++++++++++++++ llvm/tools/llvm-readobj/ELFDumper.cpp | 5 +- llvm/tools/opt/opt.cpp | 12 +- llvm/unittests/Analysis/ValueTrackingTest.cpp | 4 + llvm/unittests/Support/KnownBitsTest.cpp | 87 ++++ llvm/utils/TableGen/AsmMatcherEmitter.cpp | 5 +- llvm/utils/TableGen/GICombinerEmitter.cpp | 8 +- llvm/utils/UpdateTestChecks/asm.py | 11 +- llvm/utils/benchmark/include/benchmark/benchmark.h | 2 +- .../deviceRTLs/amdgcn/src/amdgcn_interface.h | 2 + .../deviceRTLs/amdgcn/src/target_impl.hip | 6 +- openmp/libomptarget/include/Debug.h | 44 +- openmp/libomptarget/include/SourceInfo.h | 9 +- openmp/libomptarget/plugins/cuda/src/rtl.cpp | 28 +- openmp/libomptarget/src/device.cpp | 23 +- openmp/libomptarget/src/interface.cpp | 42 +- openmp/libomptarget/src/private.h | 63 ++- openmp/libomptarget/test/offloading/info.c | 35 +- openmp/runtime/src/include/omp.h.var | 18 + openmp/runtime/src/include/omp_lib.f90.var | 97 ++++ openmp/runtime/src/include/omp_lib.h.var | 107 ++++ 190 files changed, 4017 insertions(+), 2107 deletions(-) delete mode 100644 clang/test/CXX/class/class.init/class.copy.elision/p3.cpp create mode 100644 llvm/test/Analysis/CostModel/AArch64/sve-getIntrinsicInstrCost- [...] create mode 100644 llvm/test/Analysis/CostModel/AArch64/sve-getIntrinsicInstrCost- [...] create mode 100644 llvm/test/CodeGen/Hexagon/isel-splat-vector-neg-i8.ll create mode 100644 llvm/test/CodeGen/VE/Scalar/cpu.ll create mode 100644 llvm/test/Transforms/ArgumentPromotion/dead-gep-no-promotion.ll create mode 100644 llvm/test/Transforms/InterleavedAccess/X86/interleave-load-extr [...] create mode 100644 llvm/test/tools/llvm-mca/AArch64/Cortex/forwarding-A57.s