[libc++] Refactor node creation and destruction in list and forward_list #65614
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
github-actions
bot
added
the
libc++
EricWF
approved these changes
Sep 8, 2023
The UB fixes are the ones I am picking up in https://reviews.llvm.org/D101206. These fixes are about ensuring that we begin and end the lifetime of node classes, which are currently never created. We allocate the nodes and then we construct their various members individually, but we never actually call their constructor or destructor. |
ldionne
force-pushed
the
review/refactor-list-nodes
branch
2 times, most recently
from
86063a0
to
16c3d05
Compare
This removes a lot of code duplication, makes the code simpler and prepares the terrain for https://reviews.llvm.org/D101206, which will fix some UB in the node-based containers. This also allows removing the dependency of list and forward_list on unique_ptr by using __allocation_guard instead.
ldionne
force-pushed
the
review/refactor-list-nodes
branch
from
16c3d05
to
95b2e16
Compare
ZijunZhaoCCK
pushed a commit
to ZijunZhaoCCK/llvm-project
that referenced
this pull request
Sep 19, 2023
…ist (llvm#65614) This removes a lot of code duplication, makes the code simpler and prepares the terrain for https://reviews.llvm.org/D101206, which will fix some UB in the node-based containers. This also allows removing the dependency of list and forward_list on unique_ptr by using __allocation_guard instead.
fadlyas07
pushed a commit
to greenforce-project/llvm-project
that referenced
this pull request
Sep 23, 2023
* llvm-project/main:
[Workflow] Roll back some settings since they caused more issues
[DAGCombiner][AArch64] Fix incorrect cast VT in `takeInexpensiveLog2`
[clang][Interp][NFC] Enable more existing SemaCXX tests
Propagate the volatile qualifier of exp to store /load operations .
[libc] Fix cyclical dependency on errno matcher for NVPTX architectures
[Workflow] Use a higher max depth
[Workflow] Try to fix code-formatter failing to find changes in some cases.
[clang][Interp][NFC] Enable constexpr-single-element-array test
Fix: Distinguish CFI Metadata Checks in MergeFunctions Pass (#65963)
[clang][Interp][NFC] Refactor ByteCodeEmitter a bit
[ConstraintElimination] Extend unsigned-to-signed fact transfer
[ConstraintElimination] Introduce tests for PR66173
[gn build] Port e5f169f91a86
Revert "[ORC][LLJIT] Move enable-debugger-support utility out of LLJITBuilder."
Reapply "[AArch64] Merge LDRSWpre-LD[U]RSW pair into LDPSWpre"
[Support] Add [[nodiscard]] (NFC)
Revert "[NVPTX] Improve lowering of v2i16 logical ops. (#67073)"
[mlir][complex] Support Fastmath flag in the conversion of exp,expm1 (#67001)
Remove expected-fail for Linux on TestStepOverWatchpoint.py
Re-enable StepOverWatchpoint test, make it more focused
[Vectorize] Use range-based for loops (NFC)
[release notes] Add forgotten case to list of changed manglings.
Use llvm::drop_begin and llvm::drop_end (NFC)
Use llvm::is_contained (NFC)
[llvm] Use llvm::is_contained (NFC)
[RISCV][GISel] Emit G_CONSTANT 0 as a copy from X0. (#67202)
[RISCV][GISel] Add test showing missed opportunity to use X0 for the LHS of sub for negate.
[mlir][sparse] make high and 2:4 a level format, not a level property (#67206)
[llvm] Remove uses of Type::getPointerTo() (NFC)
[gn build] Port e1a5bb59b91d
[ORC][LLJIT] Move enable-debugger-support utility out of LLJITBuilder.
[libunwind] Relax a REQUIRES on a test that passes on FreeBSD
[ORC][LLJIT] Fix typo in assert message.
[ORC][LLJIT] Add a Pre-PlatformSetup-Setup function.
[mlir][sparse] add lvlToDim field to sparse tensor encoding (#67194)
[Driver] Implement ToolChain on Haiku (#66038)
[sanitizer] Add more internal symbolizer tests
Revert "Implement [[msvc::no_unique_address]] (#65675)" (#67198)
[mlir] Fix some cmake dependencies in LLVMIR Dialect (#66956)
coroutines doc: fix a RST critical
[mlir][TOSA] Fix shape inference when operand was inferred (#66906)
Revert "Fix attr docs from previous no-unique-adderss change (#67195)"
Fix attr docs from previous no-unique-adderss change (#67195)
[hwasan] Enabled internal_symbolizer with hwasan
[BasicBlockSections] Split cold parts of custom-section functions. (#66731)
[RISCV] Handle EltType > XLEN case in VMV_V_X_VL to VMV_S_X_VL fold
[Corosplit][DebugInfo] Don't add EntryValue ops in variadic DIExpressions (#67179)
[hwasan] Fixing false invalid-free with disabled tagging (#67169)
[Driver] Fix detection of libc++ with empty sysroot. (#66947)
Implement [[msvc::no_unique_address]] (#65675)
[VPlan] Ensure start value of phis is the first op at construction (NFC)
[RISCV][GISel] Implement instruction selection for G_PHI and G_BRCOND.
[NFC][ObjectSizeOffsetVisitor] Remove redundant equality check
[VPlan] Fix invalid IR in unit test input, run verifier.
[NVPTX] Improve lowering of v2i16 logical ops. (#67073)
Add command line option --no-trap-after-noreturn (#67051)
LoopRotationUtils: Special case zero-branch weight cases (#66681)
[libc] Refactor scanf reader to match printf (#66023)
[mlir][AMDGPU] Move to new buffer resource intrinsics
[scudo] Convert assert to CHECK macro. (#67184)
[llvm-objdump] Add support for the PT_OPENBSD_NOBTCFI segment type. (#67127)
[BasicBlockSections] Introduce the basic block sections profile version 1. (#65506)
[PowerPC] Improve kill flag computation and add verification after MI peephole
[gn build] Manually port 8e87dc10
[Fuchsia] Build with -fvisibility=default (#67067)
Revert "[dataflow] use true/false literals in formulas, rather than variables"
[lldb] Fix a warning
[dsymutil] Remove paper trail warnings (#67041)
[bazel] Port 8e87dc10b80f298a682bf01fab190a57e6d5af57 (RISCVPostLegalizerCombiner)
Revert "[ORC] Add N_SO and N_OSO stabs entries to MachO debug objects."
[ORC] Add N_SO and N_OSO stabs entries to MachO debug objects.
[hwasan] Disable test with internal_symbolizer
[flang][hlfir] Fixed some finalization/deallocation issues. (#67047)
[RISCV] Ruse an existing variable to shorten a line [nfc]
[lldb][NFC] Move some ctors and tors to cpp files (#67165)
[Atomic][doc] Fix outdated hook name and description (#66989)
[RISCV][NFC] Remove rdty arg of PseudoLoad and the default rdty value of PseudoFloatLoad (#67014)
[RISCV][GISel] Add a post legalizer combiner and enable a couple comb… (#67053)
[RISCV] Truncate constants to eltwidth before checking simm5 when con… (#67062)
[scudo] Avoid deprecated-volatile warning in HybridMutex::delayLoop (#67135)
[lldb][NFCI] Change parameter type in UserExpression::GetObjectPointer (#67055)
[Coverage] Fix a warning
[mlir][sparse] replace "sparse compiler" with "sparsifier" in doc (#67082)
[libc] Update integration test's linking options (#67158)
[NFC]Do not use recursion for CounterMappingContext::evaluate (#66961)
[mlir][sparse] Add more error messages and avoid crashing in new parser (#67034)
[libc][NFC] Extend ErrnoSetterMatcher to test expected inequalities. (#67153)
[RISCV] Check for COPY_TO_REGCLASS in usesAllOnesMask (#67037)
[AMDGPU] Create matchPERM helper from performOrCombine PERM matching code.
[workflow] Tweak code formatting workflow to be faster
[dataflow] use true/false literals in formulas, rather than variables
[MLIR][NVGPU] Introduce `nvgpu.mbarrier.group` for multiple mbarrier use (#65951)
[SLP]Do not try to reorder possible strided nodes.
[lldb] Require paused process and frame for "register info" command (#67124)
Revert "[AMDGPU] Introduce real and keep fake True16 instructions."
AMDGPU: Teach isArgPassedInSGPR() about cs_chain* calling convention (#67086)
[IRTranslator] Set NUW flag for inbounds gep and load/store offsets
[X86] SandyBridge ymm broadcast loads use port5 + port23
[X86] SandyBridge masked stores use 2uop from port01
[libc] Fix pthread_create_test for 32 bit systems (#66564)
[mlir][nvgpu] Fix shared build. NFC
[libc] Add ${CMAKE_CROSSCOMPILING_EMULATOR} to custom test cmdlines (#66565)
[ConstraintElim] Move just before loop simplification pipeline.
[mlir][bufferization] Remove `deallocationFn` from `BufferizationOptions` (#67128)
[mlir][bufferization][NFC] Remove yielded tensor analysis (#67126)
[X86] SandyBridge DPPS folded instructions use an extra port5 resource
[clang-cl] Fix value of __FUNCTION__ and __FUNC__ in MSVC mode for c++. (#66120)
[SPIRV] Add support for SPV_KHR_bit_instructions (#66215)
[include-cleaner] Respect the UsingShadowDecl when find headers for ambiguous std symbols. (#66485)
[AArch64] Expand log/exp tests. NFC
Coroutines: Handle non-zero stack address space (#67092)
[gn build] Port eb02ee44d325
[bazel] Port for 22e2b805ef3c4caa971fad3656976b6e0017aff3
[MLIR][UB] Add inliner interface for UB dialect (#67115)
[InstrRef][NFC] Improve diagram illustrating locations (#66944)
[Flang][OpenMP] Create versions of a few tests with HLFIR lowering
[AArch64] Move PAuth codegen down the machine pipeline
[CodeGen] Remove extraneous spaces at start of line in TargetOpcodes.def. NFC
[bazel] Port for 2388222695d25639cf3a2e08b4a8e97da4d55313
[AArch64] Expand Sin/Cos GlobalISel testing. NFC
[Flang][OpenMP] NFC: Move PFT tests to a separate subdirectory
[AMDGPU] Update RUN line in test (NFC)
[LSR] Regenerate test checks (NFC)
Reland "[SimplifyCFG] Hoist common instructions on switch" (#67077)
[StackColoring] Handle fixed object index
[mlir][ArmSME] Add `enable_arm_streaming_ignore` attribute (#66911)
[MLIR][linalg] Fix unpack rewriter for dynamic shapes (#67096)
[ConstraintElim] Support adding facts from switch terminators. (#67061)
[AMDGPU] Don't suppress printing the .l and .h register suffixes.
[ConstraintElim] Add additional switch case and use i8 instead of i32.
[DAG] getNode() - remove oneuse limit from (zext (trunc (assertzext x))) -> (assertzext x) fold (REAPPLIED)
docs: Fix misplaced apostrophe (#67103)
[AMDGPU] Introduce real and keep fake True16 instructions.
[StackColoring] Handle SEH catch object stack slots conservatively
[AMDGPU] Have a subtarget feature to control use of real True16 instructions.
[MLIR][NVGPU] Adding `nvgpu.warpgroup.mma` Op for Hopper GPUs (#65440)
[AArch64] Don't rely on (zext (trunc x)) pattern to detect zext_inreg MULL patterns - use value tracking directly
[AArch64] LowerMUL - use SDValue directly instead of SDNode. NFC.
[AArch64][SVE2] Do not emit RSHRNB for large shifts (#66672)
[dataflow] Parse formulas from text (#66424)
[clang][Sema][NFC] _or_null -> _if_present
[Serialization] Do less redundant work computing affecting module maps (#66933)
[flang][runtime] Finalize polymorphic components using dynamic type (#67040)
[AMDGPU] Add True16 register classes.
[mlir][ArmSME] Add tile slice to vector intrinsics (#66910)
[clangd] Allow --query-driver to match a dot-normalized form of the path (#66757)
build-docs: Add option to disable doxygen/sphinx docs (#66928)
[mlir] Fix incorrect check in ptx builder test (nfc) (#67097)
Reland: [Workflow] Add new code format helper.
[LSR] Simplify type check for opaque pointers (NFC)
Revert "[SLP]Use source vector type as the original vector type instead of"
[Analysis] Use drop_begin (NFC)
[clang] Enable descriptions for --print-supported-extensions (#66715)
[llvm] Use range-based for loops (NFC)
[SCEVExpander] Clarify absence of no-op casts (NFC)
[libunwind][WebAssembly] Support Wasm EH
[libc++abi][WebAssembly] Support Wasm EH
[SCEV] Fix -Wunused-variable in ScalarEvolutionExpander.cpp (NFC)
[Analysis] Use std::clamp (NFC)
Revert "[DAG] getNode() - remove oneuse limit from (zext (trunc (assertzext x))) -> (assertzext x) fold"
[SCEV] Require that addrec operands dominate the loop
[flang] Deallocate local allocatable at end of their scopes (#67036)
[ARM] Add some extra testing for MVE postinc loops. NFC
[Clang][LLVM][Coroutines] Prevent __coro_gro from outliving __promise (#66706)
[libc] Pull more definitions from linux/stat.h (#67071)
[analyzer] Fix taint sink rules for exec-like functions (#66358)
[clang] Use a range-based for loop (NFC)
[NFC][hwasan] Remove redundant constant
[AArch64][GlobalISel] Avoid running the shl(zext(a), C) -> zext(shl(a, C)) combine. (#67045)
[AArch64][GlobalISel] Split offsets of consecutive stores to aid STP … (#66980)
[mlir][Vector] Add support for Value indices to vector.extract/insert
[AMDGPU][MC][GFX11] Always output wait_vdst and wait_exp (#66610)
[NFC][hwasan] Return to one loop in CopyShadow
[mlir][sparse] update doc for new surface syntax encoding (#67072)
[hwasan] Optimize shadow shapshot size (#67068)
[X86] Use RIP-relative for non-globals in medium code model in classifyLocalReference() (#67070)
[hwasan] Fix consts after 6eaaf0916
[ELF][test] Improve GOTPCRELX tests
[NFC][CodeGen] Create method to clear registers (#66958)
Lazy deference underlying object for shared/weak/unique_ptr synthetic… (#67069)
[mlir] Move supplemental patterns before op replacement (#66959)
[hwasan] Store some report data early (#66682)
[-Wunsafe-buffer-usage] Use `Strategy` to determine whether to fix a parameter
[libc] Template the printf / scanf parser class (#66277)
[VE] Add Core to CMakeLists.txt for VE unittest
[NFC][hwasan] Add a few const
watch set expression's default type was wrong with new modify type
[MSVC, ARM64] Add _Copy* and _Count* intrinsics (#66554)
[NFC][hwasan] Remove unused members
[NFC][HWASAN] Extract BaseReport::PrintTags
[runtimes] Simplify testing configurations on GCC by using -nostdlib++ (#67021)
[TEST][hwasan] Check more details in overflow test (#67059)
[runtimes][NFC] Remove old Lit annotations for gcc-12 and clang-14
[scudo] Fine tune busy-waiting in HybridMutex
[BOLT][AArch64] Fix instrumentation deadloop
[Github] Add compiler-rt:scudo label to scudo prs (#66829)
[NVPTX] improve lowering for common byte-extraction operations. (#66945)
[libc++] When run in verbose mode, ssh.py should print to stderr, not stdout
[libc++] Simplify ssh.py by not passing args everywhere
[ConstraintElim] Add extra switch test coverage.
I can't make this test fail locally, but it is failing on the macOS bots as well. I'll have to figure out a different way to test this.
Ensure we disambiguate NTTP names when doing TemplateTree comparisons
[CodeLayout] Refactor std::vector uses, namespace, and EdgeCountT. NFC
[ConstraintElim] Add phase ordering tests for pipeline adjustment.
Add mangling for type trait expressions.
[-Wunsafe-buffer-usage] Group parameter fix-its
[libc] Fix and simplify the implementation of 'fread' on the GPU (#66948)
[BPF][DebugInfo] Show CO-RE relocations in llvm-objdump
[builtins][CMake] Replace custom target for lse_builtin symlinks (#66936)
[mlir][index] Implement folders for CastSOp and CastUOp (#66960)
[AArch64] Separate PNR into its own Register Class (#65306)
[libc] Fix Off By One Errors In Printf Long Double (#66957)
[SLP]Use source vector type as the original vector type instead of artificial for better cost estimation.
Reland: [sanitizer_symbolizer] Add StackTracePrinter virtual class (#66689)
[clang-format] Don't split "DPI"/"DPI-C" in Verilog imports (#66951)
[BOLT] Fix data race in MCPlusBuilder::getOrCreateAnnotationIndex (#67004)
[AArch64] Pre-commit some tests for D152828 (NFC)
[AArch64] Correctly determine if {ADD,SUB}{W,X}rs instructions are cheap
[libc++] Fix how we run codesign in the test suite when --codesign_identity is provided
[libc++] Add a --verbose option to ssh.py to help debug failures
[AArch64] Refactor AArch64InstrInfo::isAsCheapAsAMove (NFC)
Revert "[InlineCost] Check for conflicting target attributes early"
[SLp]Fix a crash because of wrong deps between vectorized nodes.
ISel/RISCV: remove dead code corresponding to VP_FSH[L|R] (#67035)
Set -rpath-link only if the path is nonempty.
[libc][NFC] Remove unused function from the RPC server
[flang] Centralize automatic deallocation code in lowering (#67003)
[mlir][tosa] Constant optimizations for reduce operations
[CodeGen] Improve compilation time with VLIWMachineScheduler (#66942)
[mlir][Interfaces][NFC] Better documentation for `RegionBranchOpInterface` (#66920)
[mlir][memref][transform] Add new alloca_to_global op. (#66511)
[libc] Remove the 'rpc_reset' routine from the RPC implementation (#66700)
CostModel/RISCV: fix typos in fround test, vector length (#67025)
[mlir][Interfaces] Clean up `DestinationStyleOpInterface` (#67015)
[runtimes] Workaround a subtle linker issue on macOS in the CI
[SelectionDAG] Flags are dropped when creating a new FMUL (#66701)
[NFC][AsmPrinter] Refactor constructVariableDIE (#66435)
Add TMA Store operation to the NVVM dialect
[clang-tidy] Update llvmlibc-implementation-in-namespace to new rules (#66504)
[SLP][NFC]Unify add() member function in CostEstimator, NFC.
[libc] Update siginfo_t to match kernel definition (#66560)
[MLIR][IR] Rename `Block::hasTerminator` to `mightHaveTerminator` (#66870)
[LSR] Regenerate test checks (NFC)
[AIX] cxx_std_23 is currently not a known feature to ibm-clang (#66952)
[SLP][NFC]Improve compile time by storing all nodes for the given scalar.
[SCEVExpander] Drop ExpandTy argument from expandIVInc() (NFC)
[SCEVExpander] Drop Ty argument from expandAddToGEP() (NFC)
[MLIR] Better document for basic PTX Builder Interface (NFC) (#67016)
[SCEVExpander] Remove some unnecessary effective type handling (NFC)
[lldb][AArch64] Initialise m_sve_state in NativeRegisterContextLinux
[flang] Generate valid IR on GOTO DO body (#66084)
Reland "[lldb][AArch64] Linux corefile support for SME"
[mlir][linalg][transform] Fix test breakage from 69bc1cbb. (NFC)
[mlir][linalg][transform] Rename {masked_vectorize => vectorize => vectorize_children_and...}. (#66575)
[InstCombine] Remove instcombine-infinite-loop-threshold option
[mlir][Vector] NFC - Fix doc generation (#67007)
[SCEVExpander] Remove unnecessary expandCodeForImpl() wrapper (NFC)
[RISCV] Shrink vslideup's LMUL when lowering fixed insert_subvector (#65997)
Revert "[lldb][AArch64] Linux corefile support for SME"
[flang] Avoid cycles during instantiation of derived types
Revert "[AArch64][GlobalISel] Adopt dup(load) -> LD1R patterns from SelectionDAG"
[clang][Sema] Fix crash introduced in b2cd9db589335d5885c04df83003a623cf2f05ff (#66954)
[lldb][AArch64] Linux corefile support for SME
[IndVars] Check expansion safety during LFTR
[libc++] Move a few XFAILs to UNSUPPORTED for consistency
[libc++] Temporarily disable std::cin tests on Windows -> Linux cross compiler
[SlotIndexes] Remove unused EXPENSIVE_CHECKS machinery
[NFC] clang-format lld/COFF/Driver.cpp and lld/Common/Filesystem.cpp
[mlir] more bazel fixes for vector attributes
[AMDGPU] Precommit test for D159533 (#66965)
[DAG] getNode() - remove oneuse limit from (zext (trunc (assertzext x))) -> (assertzext x) fold
[mlir][python] smaller scope for vector enumgen (#66992)
[UpdateTestChecks] Add support for SPIRV in update_llc_test_checks.py (#66213)
[IndVars] Add test for #66986 (NFC)
[gn build] Port 75e862077834
Reland "[lldb] Add 'modify' type watchpoints, make it default (#66308)"
[CMake] Fully delete the deprecated LLVM_USE_CRT* (#66850)
[AMDGPU] Select gfx1150 SALU Float instructions (#66885)
Revert "[InstrProf][compiler-rt] Enable MC/DC Support in LLVM Source-based Code Coverage (1/3)"
[AArch64] Add an aarch64-enable-ext-to-tbl option. NFC
[SPIRV][DX] Share one test between backends (#65975)
[LV] Directly add fast-math flags to select recipe (NFC).
[AMDGPU] Remove Code Object V2 (#65715)
[mlir][linalg][NFC] Make `LinalgOp` inherit from `DestinationStyleOpInterface` (#66995)
[RISCV] Add overlapping constraints label
[gn build] Port a7b78cac9a77
Revert "[lldb] Add 'modify' type watchpoints, make it default (#66308)"
[RISCV] Move sifive_files to riscv_files (#66872)
[libc++] Implement LWG-3655: The INVOKE operation and union types
[AMDGPU] Add gfx1150 SALU Float instructions (#66884)
[libc][clang-tidy] Add llvm-header-guard to get consistant naming and prevent file copy/paste issues. (#66477)
[LV] Relax over-strict assertion for reduction exit value selects.
[libc++] Add test coverage for unordered containers comparison (#66692)
[runtimes] Fix link order of system librarires on Apple platforms (#66940)
[mlir][Interfaces] `LoopLikeOpInterface`: Add helpers to get region iter_args and inits (#66925)
[clang][www] Change 'clang-test' to 'check-clang' on the hacking webpage
[InstCombine] Add pre-commit tests for PR66733. NFC.
Fix induction variable type in scf.for py binding.
[SimpleLoopUnswitch] Fix exponential unswitch
[clang] Update Clang version from 17 to 18 in scan-build.1.
[Docs] Add requirements.txt for sphinx and extension
[X86] Add test for #66984 (NFC)
[clang][AST] fix lack comparison of declRefExpr in ASTStructuralEquivalence (#66041)
[BOLT][RISCV] Carry-over annotations when fixing calls (#66763)
Re-apply "[Parse] Split incremental-extensions" (#66446)
[Demangle] demangle builtin type transformations (#65902)
[AMDGPU] [SIFrameLowering] Use LiveRegUnits instead of LivePhysRegs.
[MC] Actually make .pseudoprobe created sections deterministic
[NFC][hwasan] Store shadow bytes early (#66682)
Revert "[NFC][hwasan] Store shadow bytes early (#66682)"
[clang][AST][ASTImporter] improve AST comparasion on VarDecl & GotoStmt (#66976)
Temoprarily stop running this test on Intel macs
Reland [InlineCost] Enable the cost benefit analysis for Sample PGO (#66457)
[NFC][hwasan] Store shadow bytes early (#66682)
[NFC][hwasan] Stored tail early (#66682)
[NFC][hwasan] Use stored chunk in TailOverwrittenReport (#66682)
[NFC][hwasan] Set more fields by value and make them const (#66682)
[NFC][hwasan] More consts in BaseReport (#66682)
[Flang][OpenMP][MLIR] Fix memory leak caused by D149368 causing sanitizer error and fix iterator invalidation error
[Bazel] Follow up for ac182deee8287 (D76445), RISCVGenGlobalISel.inc
Revert "[InlineCost] Enable the cost benefit analysis for Sample PGO (#66457)"
Recommmit "[RISCV] Improve contant materialization to end with 'not' if the cons… (#66950)"
[clang] NFCI: Use `FileEntryRef` in `FileManagerTest`
[clang] NFCI: Use `FileEntryRef` in `SymbolCollector`
[InlineCost] Enable the cost benefit analysis for Sample PGO (#66457)
[MC] Make .pseudo_probe created sections deterministic after D91878
-fsanitize=function: Add a (bugged) test case for a sugared function type
-fsanitize=function: Add a MSVC test case
[SystemZ] Fix IWYU
Copy and paste error in a file header.
The test: test_run_then_attach_wait_interrupt was flakey on Linux & Windows.
[Sema] Fix fixit cast printing inside macros (#66853)
Revert "[RISCV] Improve contant materialization to end with 'not' if the cons… (#66950)"
[RISCV] Improve contant materialization to end with 'not' if the cons… (#66950)
Revert "[MC] Make .pseudo_probe sections deterministic after D91878"
[clang] NFCI: Use `FileEntryRef` in `PreamblePatch`
[gn build] Port 933ad5c897ee
[gn] expand dummy target from 2ed813ff2696
[RISCV] Add llvm/test/Transforms/VectorCombine/RISCV/lit.local.cfg
[mlir][sparse] Generates python bindings for SparseTensorTransformOps. (#66937)
[VectorCombine][RISCV] Convert VPIntrinsics with splat operands to splats (#65706)
[fuzzer] Relax the FileCheck string for test fuzzer-custommutator.test (#66343)
[Doc] Clarify the comments about how handleErrors works. (#65655)
[mlir][python] Expose AsmState python side. (#66819)
[ELF] -r --compress-debug-sections: update implicit addends for .rel.debug_* referencing STT_SECTION symbols (#66804)
[InlineCost]Account for switch instructons when the switch condition could be simplified as a result of inlines.
Revert "[Coverage] Fix -Wswitch after D138847"
[ELF][test] Improve non-abs-reloc.s to test non-STT_SECTION local symbol
Revert "[Coverage][llvm-cov] Enable MC/DC Support in LLVM Source-based Code Coverage (2/3)"
Revert "Remove some raciness from the TestProcessAttach.test_run_then_attach_wait_interrupt"
[lldb] Fix stdcpp type summary mistakenly marked as regex (NFC) (#66949)
[Coverage] Fix -Wswitch after D138847
[X86] fix combineSubSetcc to handle a large constant (#66941)
[HWASAN] Prevent SEGV in report near inaccessible memory (#66861)
[mlir] Fix warnings
[clang-tidy][NFC] Update documentation for bugprone-undefined-memory-manipulation
[scudo] Remove hard-coded use of AndroidConfig. (#66845)
[MC] Make .pseudo_probe sections deterministic after D91878
[lldb] Add 'modify' type watchpoints, make it default (#66308)
[Coverage][llvm-cov] Enable MC/DC Support in LLVM Source-based Code Coverage (2/3)
Fix sphinx bot build
Remove some raciness from the TestProcessAttach.test_run_then_attach_wait_interrupt command.
[SLP]Do not gather node, if the instruction, that does not require scheduling, is previously vectorized.
Remove #include that the cp-to-llvm script added.
Implement mangling rules for C++20 concepts and requires-expressions.
[clang-tidy] Fix support for typedefs in readability-identifier-naming (#66835)
[clang-tidy][NFC] Add release notes entry for D127036
[clang-tidy] Warn about arrays in bugprone-undefined-memory-manipulation
[bazel] Port D149368 mlir openmp
[libc++abi][NFC] Remove mentions of LIBCXXABI_SHARED_LIBRARIES
[MLGO] Move MBB Profile Dump test to Generic (#66856)
Revert D155711 "[SimplifyCFG] Hoist common instructions on Switch."
[MLIR][OpenMP] Initial Lowering of Declare Target for Data
[SLP][NFC]Add a test with the reused main op instruction, NFC.
[DAGCombiner] Extend `combineFMulOrFDivWithIntPow2` to work for non-splat float vecs
[DAGCombiner] Fold IEEE `fmul`/`fdiv` by Pow2 to `add`/`sub` of exp
[AMDGPU] Add tests for folding `fmul`/`fdiv` by Pow2 to `add`/`sub` of exp; NFC
[X86] Add tests for folding `fmul`/`fdiv` by Pow2 to `add`/`sub` of exp; NFC
[LiveRange] Verify Other LiveRange in Join API (#66809)
[bazel] Port 4fbb5f93506317f063753c2b4aecebc86d63264e
[Clang][C] Fixed a bug where we reject an _Atomic qualified integer in a switch statment
[libc][Obvious] Fix incorrect RPC opcode for `clearerr`
[SLP]Fix a crash when trying to find operand with re-vectorized main instruction.
[mlir] regenerate linalg named ops yaml (#65475)
[mlir] introduce transform.loop.forall_to_for (#65474)
[mlir] Bazel fixes for 1b8b55644313216e6b0fa233bbd8b01fee23f99f (#66929)
[mlir][sparse] remove most bufferization.alloc_tensor ops from sparse (#66847)
[Driver] Some improvements for path handling on NetBSD (#66863)
[SelectionDAG] [NFC] Add pre-commit test for PR66701. (#66796)
[AArch64] Add some tests for setcc known bits fold. NFC
Cleanup fallback NOT checks
[AArch64][GlobalISel] Adopt dup(load) -> LD1R patterns from SelectionDAG
[clang][CodeGen] The `eh_typeid_for` intrinsic needs special care too (#65699)
[ELF] Postpone "unable to move location counter backward" error (#66854)
[NFC][AsmPrinter] Refactor FrameIndexExprs as a std::set (#66433)
[ELF][test] Add a test to demonstrate #66836
[X86] Regenerate pr39098.ll
[AMDGPU] Regenerate always-uniform.ll
[flang][hlfir] Fixed cleanup code placement indeterminism in OrderedAssignments. (#66811)
[SPIRV] Fix OpConstant float and double printing
[gn] Add dummy build file for VETests
[lldb][AArch64] Invalidate cached VG value before reconfiguring SVE registers
[mlir][Vector] Add fastmath flags to vector.reduction (#66905)
[NFC] Fix spelling 'constanst' -> 'constants'
[AMDGPU] Convert tests rotr.ll and rotl.ll to be auto-generated (#66828)
Revert "[clang] Don't inherit dllimport/dllexport to exclude_from_explicit_instantiation members during explicit instantiation (#65961)"
[lldb-vscode] Use auto summaries whenever variables don't have a summary (#66551)
[VE] Add TargetParser to CMakeLists.txt for VE unittest
Warn on align directive with non-zero fill value in virtual sections (#66792)
[mlir][VectorOps] Extend vector.constant_mask to support 'all true' scalable dims (#66638)
[libc++] Sort available features before printing them
[X86] X86DAGToDAGISel::matchIndexRecursively - replace hard coded recursion limit with SelectionDAG::MaxRecursionDepth. NFCI.
[AArch64][SME] Enable TPIDR2 lazy-save for za_preserved
[libc++] Simplify how the global stream tests are written (#66842)
[SPIR-V] Fix 64-bit integer literal printing (#66686)
[MLIR][Presburger] Template Matrix to allow MPInt and Fraction; use IntMatrix for integer matrices (#66897)
[clang][dataflow] Identify post-visit state changes in the HTML logger. (#66746)
[X86] Fix an assembler bug of CMPCCXADD. (#66748)
[X86] Align other variants to use void * as 512 variants. (#66310)
[mlir][bufferization][NFC] Move memref specific implementation of AllocationOpInterface to memref dialect directory (#66637)
[Docs] Update ExceptionHandling example (NFC)
[clang][Interp][NFC] Small code refactoring
[RISCV] Add tests where bin ops of splats could be scalarized. NFC (#65747)
[mlir][bufferization][scf] Implement BufferDeallocationOpInterface for scf.reduce.return (#66886)
[SROA] Don't shrink volatile load past end
[NFC][AMDGPU] Remove redundant hasSideEffects=1
[NFC][AMDGPU] Perform a single lookup in map in SIInsertWaitcnts::isPreheaderToFlush
[VE] Add unittest for intrinsics (#66730)
[analyzer] Fix StackAddrEscapeChecker crash on temporary object fields (#66493)
[NFC][Clang] Address reviews about overrideFunctionFeaturesWithTargetFeatures (#65938)
[X86] Add test cases for gnux32 large constants Issue #55061
[X86] combineINSERT_SUBVECTOR - attempt to combine concatenated shuffles
[X86] vector-interleaved tests - add AVX512-SLOW/AVX512-FAST common prefixes to reduce duplication
[IR] Fix a memory leak if Function::dropAllReferences() is followed by setHungoffOperand
[analyzer] Fix crash analyzing _BitInt() in evalIntegralCast (#66782)
[mlir][gpu][bufferization] Implement BufferDeallocationOpInterface for gpu.terminator (#66880)
[RISCV] Add missing V extensions for zvk-invalid-features.c (#66875)
[Flang][OpenMP] Move FIR lowering tests to a separate directory (#66779)
[clang] [C23] Fix crash with _BitInt running clang-tidy (#65889)
[lldb][API] Remove debug print in TestRunLocker.py
[LLVM] convergence verifier should visit all instructions (#66200)
[GuardUtils] Revert llvm::isWidenableBranch change (#66411)
[reland][libc][cmake] Tidy compiler includes (#66783) (#66878)
[lldb][AArch64] Add SME's streaming vector control register
[mlir][vector] Make ReorderElementwiseOpsOnBroadcast support vector.splat (#66596)
[SimpleLoopUnswitch] Fix reversed branch during condition injection
[OpenMP][VE] Limit the number of threads to create (#66729)
Revert "Fix a bug with cancelling "attach -w" after you have run a process previously (#65822)"
[clang-format][NFC] Clean up signatures of some parser functions (#66569)
[lldb][Docs] Fix typo in style docs
Revert "[Workflow] Add new code format helper. (#66684)"
[flang] Prevent IR name clashes between BIND(C) and external procedures (#66777)
[NFC][InferAlignment] Swap extern declaration and definition of EnableInferAlignmentPass
[TableGen][GlobalISel] Use `GIM_SwitchOpcode` in Combiners (#66864)
[AMDGPU] Run twoaddr tests with -early-live-intervals (#66775)
[RISCV] Fix bugs about register list of Zcmp push/pop. (#66073)
[gn build] Port 0f152a55d3e4
[Workflow] Add new code format helper. (#66684)
[mlir][IR] Trigger `notifyOperationRemoved` callback for nested ops (#66771)
[ELF][test] Improve -r tests for local symbols
[InferAlignment] Enable InferAlignment pass by default
[InstCombine] Use a cl::opt to control calls to getOrEnforceKnownAlignment in LoadInst and StoreInst
[InferAlignment] Implement InferAlignmentPass
[InferAlignment] Create tests for InferAlignment pass
[RISCV] Install sifive_vector.h to riscv-resource-headers (#66330)
[MC,X86] Property report error for modifiers with incorrect size
[clang][dataflow] Reorder checks to protect against a null pointer dereference. (#66764)
[ELF] Remove a R_ARM_PCA special case from relocateNonAlloc
[ELF][test] Improve tests about non-SHF_ALLOC sections relocated by non-ABS relocations
[libc++] Warn if an unsupported compiler is used
[NFC][hwasan] Collect heap allocations early (#66682)
[NFC][hwasan] Extract announce_by_id (#66682)
[NFC][hwasan] Extract a few BaseReport::Copy methods (#66682)
[NFC][hwasan] Clang-format c557621176f5f38b5757a325cc72be0a11a91c78
[NFC][hwasan] Find overflow candidate early (#66682)
[NFC][hwasan] Make ShowHeapOrGlobalCandidate a method (#66682)
[DependencyScanningFilesystem] Make sure the local/shared cache filename lookups use only absolute paths (#66122)
[RISCV] Prefer Zcmp push/pop instead of save-restore calls. (#66046)
[builtins][NFC] Avoid using CRT_LDBL_128BIT in tests (#66832)
[lldb][NFCI] Remove unused struct ConstString::StringIsEqual
[Sparc] Remove LEA instructions (NFCI) (#65850)
[Docs] Fix table after previous document update
[mlir][sparse] unifies sparse_tensor.sort_coo/sort into one operation. (#66722)
A test was changing directory and then incorrectly restoring the directory to the "testdir" which is the build directory for that test, not the original source directory. That caused subsequent tests to fail.
[IR] Add "Large Data Threshold" module metadata (#66797)
[SimplifyCFG] Hoist common instructions on Switch.
[SimplifyCFG] Pre-commit test for extending HoistThenElseCodeToIf.
Fix test added in D150987 to account for different path separators which was causing the test to fail on Windows.
[bazel] Port c649f29c24c9fc1502d8d53e0c96c3d24b31de1a (llvm-nm --line-numbers)
[AMDGPU] Add ASM and MC updates for preloading kernargs
[lit] Apply aa71680f2948's fix to an additional test
[libc] Fix printf config not working (#66834)
[clang] Improve CI output when trailing whitespace is found (#66649)
[libc++] Fix __threading_support when used with C11 threading (#66780)
[EarlyCSE] Compare GEP instructions based on offset (#65875)
[AMDGPU] Use inreg for hint to preload kernel arguments
[InstrProf][compiler-rt] Enable MC/DC Support in LLVM Source-based Code Coverage (1/3)
[lit] Fix a test fail under windows
[mlir][spirv] Define KHR cooperative matrix properties (#66823)
[mlir][spirv] Suffix NV cooperative matrix props with `_nv` (#66820)
[Clang] Static member initializers are not immediate escalating context. (#66021)
[AsmPrint] Dump raw frequencies in `-mbb-profile-dump` (#66818)
Explicitly set triple on line-numbers.test
[llvm][docs] Update active CoC Commitee members (#66814)
[mlir][sparse][gpu] free all buffers allocated for spGEMM (#66813)
[-Wunsafe-bugger-usage] Clean tests: remove nondeterministic ordering
Revert "[libc][cmake] Tidy compiler includes (#66783)" (#66822)
[llvm-nm] Add --line-numbers flag
[RISCV] Match strided load via DAG combine (#66800)
[OpenMP][Docs][NFC] Update documentation
[libc][cmake] Tidy compiler includes (#66783)
[libc++][NFC] Clean up std::__call_once
[mlir][spirv] Add conversions for Arith's `maxnumf` and `minnumf` (#66696)
[BOLT][runtime] Test for outline-atomics support
[flang] Correct handling of assumed-rank allocatables in ALLOCATE (#66718)
[github] Make branch workflow more robust (#66781)
[SLP]Fix PR66795: Check correct deps for vectorized inst with multiple vectorized node uses.
[Coverage] Skip visiting ctor member initializers with invalid source locations.
[ORC] Add writePointers to ExecutorProcessControl's MemoryAccess
Revert "Revert "[flang][hlfir] Fixed assignment/finalization order for user-defined assignments. (#66736)""
Revert "[flang][hlfir] Fixed assignment/finalization order for user-defined assignments. (#66736)"
Let the c(xx)_status pages reflect that clang 17 is released
Fix a bug with cancelling "attach -w" after you have run a process previously (#65822)
[X86] Use RIP-relative addressing for data under large data threshold for medium code model
[flang] Follow memory source through more operations (#66713)
Addressed review comments to use ThreadSafe instead of !ThreadSafe
[RISCV] Require alignment when forming gather with larger element type
[flang][hlfir] Fixed assignment/finalization order for user-defined assignments. (#66736)
[lldb] Fix build after d5a62b78b8ae
[Profile] Delete coverage-debug-info-correlate.cpp test on mac as debug info correlation not working on mac for unkown reasons.
[lldb][NFCI] Remove unneccessary allocation in ScriptInterpreterPythonImpl::GetSyntheticTypeName (#66724)
[lldb][NFCI] Remove unneeded ConstString from intel-pt plugin (#66721)
[RISCV] Expand test coverage for widening gather and strided load idioms
[SSP] Accessing __stack_chk_guard when using LTO (#66535)
[RISCV] Add more instructions for the short forward branch optimization. (#66789)
[gn build] Port 93fde2ea1b2c
[mlir][sparse] More allocate -> empty tensor migration (#66720)
[mlir][spirv][gpu] Convert remaining wmma ops to KHR coop matrix (#66455)
[RISCV] Add a pass to rewrite rd to x0 for non-computational instrs whose return values are unused
[lit] Echo full RUN lines in case of external shells (#66408)
[lit] Improve test output from lit's internal shell
[lit] Drop "Script:", make -v and -a imply -vv
[libc][math] Extract non-MPFR math tests into libc-math-smoke-tests.
[RISCV] Fix bad isel predicate handling for Ztso. (#66739)
[TwoAddressInstruction] Use isPlainlyKilled in processTiedPairs (#65976)
[DAGCombiner] Combine vp.strided.store with unit stride to vp.store (#66774)
[DAGCombiner] Combine vp.strided.load with unit stride to vp.load (#66766)
[mlir][Interfaces] `LoopLikeOpInterface`: Support ops with multiple regions (#66754)
[mlir][SCF] `ForOp`: Remove `getIterArgNumberForOpOperand` (#66629)
In ExprRequirement building, treat OverloadExpr as dependent (#66683)
[SROA] Remove unnecessary IsStorePastEnd handling (NFCI)
[RISCV] Add a combine to form masked.store from unit strided store
[RISCV] Recognize veyron-v1 processor in clang driver. (#66703)
[libunwind][AIX] Fix up TOC register if unw_getcontext is called from a different module (#66549)
[mlgo] Fix tests post PR #66334
[SPIRV] Test basic float and int types (#66282)
[clang][TSA] Thread safety cleanup functions
[flang] Add comdats to functions with linkonce linkage (#66516)
[bazel] Port for 571df0132daa903ed2c5ad5776e4d264b823de40
[SPIRV] Fix bug in emitting GLSL ext inst names
[FuncSpec] Adjust the names of specializations and promoted stack values
[MLIR][OpenMP] Alterations to LLVM-IR lowering of OpenMP Dialect due to Bounds and MapEntry operations
[Flang][MLIR][OpenMP] Update OMPEarlyOutlining to support Bounds, MapEntry and declare target globals
[Flang][OpenMP] Add lowering from PFT to new MapEntry and Bounds operations and tie them to relevant Target operations
[OpenMP][MLIR] Refactor and extend current map support by adding MapInfoOp and DataBoundsOp operations to the OpenMP Dialect
[Dexter] Associate parser errors with correct file (#66765)
Fix math-errno issue (#66381)
[IR] `SingleBlock::push_back` operations to the back of the block (#66655)
[InstCombine] Make flag drop during select equiv fold more generic
[InstCombine] Add test for #62450 (NFC)
[lldb][AArch64] Add testing for SME's ZA and SVG registers
[lldb][AArch64] Implement resizing of SME's ZA register
[RISCV] Add tests for memory constraint A
[LLD] Improve the lit tests added by 272bd6f9cc86bf6b4dd6bd51e85c46db10e8b86a
[mlir] avoid crash in transform.sequence verifier (#66756)
[RISCV] Merge RV32/RV64 CHECK lines in strided vp load/store tests. NFC
[lldb][AArch64] Add SME streaming vector length pseduo register
[lldb][AArch64] Add SME's Array Storage (ZA) register
[CodeGen] Renumber slot indexes before register allocation (#66334)
[AMDGPU] Fix line endings in a test
Move CallInst::CreateFree to IRBuilderBase
[clang][CFG] Cleanup functions
[analyzer] TaintPropagation checker strlen() should not propagate (#66086)
[NFC] Preparatory work for D153131 (#66750)
[lldb][AArch64] Document how to control the SVE/SSVE tests
[SPIR-V] Preserve pointer address space for load/gep instructions
[mlir][linalg][transform][python] Allow no args in MaskedVectorize. (#66541)
[mlir][linalg][transform] Add some debug output to vectorization. (NFC) (#66520)
[lldb] Correct expected output for variable on 32 bit platforms
[AArch64] Remove copy instruction between uaddlv with v4i16/v8i16 and dup (#66508)
[SPIR-V] Remove -opaque-pointers=0 from LITs, fixes for opaque pointers support
[examples] Remove unused variable 'BB' in BrainF.cpp (NFC)
[lldb][Docs] Document our major differences from the LLVM style (#66345)
[mlir][IR] Change `MutableArrayRange` to enumerate `OpOperand &` (#66622)
[llvm] Move CallInst::CreateMalloc to IRBuilderBase::CreateMalloc
[clang][dataflow] Fix two null pointer dereferences in `getMemberForAccessor()`. (#66742)
[NFC][NewGVN] Update assume_dominating_icmp.ll (#66711)
AMDGPU: Fix handling of -0 in round lowering (#65761)
[mlir][bazel] Small adjustment to https://github.com/llvm/llvm-project/commit/7db4a6f278a8b983a9ce86fa32f61ca9f9be5156
[gn build] Port 44532a9dd4a0
[llvm-libgcc][CMake] Refactor llvm-libgcc (#65455)
Revert "[lldb] Add 'modify' type watchpoints, make it default (#66308)"
[CodeLayout] Fix X1_Y_X2 and Y_X2_X1 testing for jumps from Y (#66592)
[ELF][test] Improve relocations referencing STT_SECTION tests for -r
[mlir] Quality of life improvements to python API types. (#66723)
[libc] Specify path for making include/ subdirs (#66589)
[RISCV] Fix inline asm error for block address (#66640)
[BOLT][YAML] Only read first profile per function
[InstCombine] Add pre-commit tests for PR66606. NFC.
[mlir][c] Expose AsmState. (#66693)
[RISCV][NFC] Simplify the sp-offset reduction by spimm of CM.PUSH/POP. (#66667)
[RISCV][MC]Add support for Binary MCExpr
[gn build] Port 3692267ca8f9
[lldb] Add 'modify' type watchpoints, make it default (#66308)
[clang] Fix null dereference on return in lambda attribute statement expr (#66643)
[bazel] Port cbd475040f8952cfc55b9e13dd5ce6c4f6434cd3 (mlir/lib/Dialect/MLProgram)
[NFC][hwasan] Collect heap related data early (#66682)
[LoopPredication] Fix division by zero in case of zero branch weights (#66506)
[NFC][hwasan] Make PrintAddressDescription method of BaseReport (#66682)
[NFC][hwasan] Remove default argument
[NFC][hwasan] Add access_size into base report (#66682)
[LAA] Analyze pointers forked by a phi (#65834)
[NFC][hwasan] Store thread id in SavedStackAllocations (#66682)
[mlir][tensor] Bufferize tensor.reshape with non-identity layouts (#65654)
[NFC][hwasan] Move PrintAddressDescription (#66682)
[NFC][hwasan] Extract BaseReport (#66682)
[mlir][TilingInterface] Make the tiling set tile sizes function use `OpFoldResult`. (#66566)
[NFC][hwasan] Move Report classes together (#66682)
[NFC][hwasan] Use unnamed namespace and static (#66682)
[NFC][hwasan] Create *Report classes (#66682)
[mlir][mlprogram] Add `mlprogram-pipeline-globals` optimization pass
[NFC][hwasan] Remove default argument
[RISCV] Combine a gather to a larger element type (#66694)
[driver] Address missed feedback from https://reviews.llvm.org/D158476
[driver] Search for compatible Android runtime directories
[AArch64] Move LDR_PXI from isStoreToStackSlot to isLoadFromStackSlot (#65658)
[WPD][LLD] Add option to validate RTTI is enabled on all native types and prevent devirtualization on types with native RTTI
[mlir][sparse] fix logical error when generating sort_coo. (#66690)
[Clang] Fix a crash in the diagnostic emission of invalid immediate calls (#66699)
[LLDB] Attempt to fix DumpDataExtractorTest
Ignore out-of-bounds printf() format string in ASan test (#66695)
Revert "[analyzer] Fix crash analyzing _BitInt() in evalIntegralCast (#65887)"
[LLDB] Attempt to fix DumpDataExtractorTest
[lld][WebAssembly] Improve error message on adding LTO object post-LTO. NFC (#66688)
[scudo] Remove AndroidSvelteConfig. (#66444)
[VPlan] Fold (MUL A, 1) -> A as VPlan2VPlan transform.
[analyzer] Fix crash analyzing _BitInt() in evalIntegralCast (#65887)
Disable MSAN test: ELF_perf.s
[SimplifyLibCalls] Drop redundant CreatePointerCast in optimizeSPrintFString
[MemCpyOpt] Drop redundant CreatePointerCast
[DSE] Simplify based on opaque pointers
[flang] Fix CFI_CDESC_T(rank) for C (#66260)
[clang] Correct source locations for instantiations of function templates.
[flang] Prioritize DATA object error messages a little better (#66258)
[llvm-profgen] Ignore inline frames with an emtpy function name (#66678)
[flang] Extension: accept NULL([MOLD=]) for ALLOCATABLE INTENT(IN) du… (#66256)
[TargetMachine] Remove DefaultOptions (#66673)
[flang] Fix bogus error w/ COMMON & EQUIVALENCE (#66254)
[driver] Perform fallback target searches for stdlib
[driver] Conditionally include installed libc++ headers for Android
[runtimes] Add a mechanism to use cache files for a runtimes build
[gn build] Port 0218ea4aaa54
[flang] Allow empty SEQUENCE types (#66252)
[bazel] Port 8677aaa1a32a (RISCV pre-legalizer combiners)
[libc++] Implement ranges::ends_with
[flang] Accept NULL() actual for optional allocatable dummy argument (#66251)
[RISCV] Narrow indices of fixed vector gather/scatter nodes
[NFC][Clang] Remove tailing white spaces in `llvm/include/llvm/Frontend/OpenMP/OMP.td`
[flang] Fix bogus warning about missing FINAL on assumed-rank (#66250)
[InstCombine] Fix typo in the test `icmp-range.ll`. NFC.
[BOLT][NFC] Refactor RI::discoverFileObjects()
[BOLT] Speedup symbol table sort
[flang] Fix characterization of result of function returning procedur… (#66248)
[Profile] Pass InstrProfSymtab by reference.
[gn] port 8677aaa1a32a (RISCV pre-legalizer combiners)
[flang] Fixed -Werror build after #65919. (#66679)
Revert "[MLIR][Presburger] Template Matrix to allow MPInt and Fraction (#65272)"
[libc++] Fix `std::pair`'s pair-like constructor's incorrect assumption (#66585)
[RISCV][GISel] Add initial pre-legalizer combiners copying from AArch64.
[lldb] Fix the TestValueAPI test
[InstCombine] Add `sext` version of pre-commit tests for PR65852. NFC.
[flang] Remove unused function IsListDirectedFieldComplete
[MLIR][Presburger] Template Matrix to allow MPInt and Fraction (#65272)
[flang] Fix compile-time infinite loop (#66246)
[NewGVN] Set parent to the temporal instructions that are generated during phi-of-ops optimization (#66314)
[mlir][Vector] Add a rewrite pattern for better low-precision ext(bit… (#66648)
[InstSimplify] Add pre-commits for PR#66606. NFC.
[flang][hlfir] Alias analysis for host associated accesses. (#65919)
[flang][runtime] Enforce proper termination of list-directed input va… (#66244)
Skip tests with earlier versions of clang
[mlir][sparse] fix merge conflict in sddmm test (#66670)
[LLDB] Add a setting for printing ValueObject hex values without leading zeroes (#66548)
Revert "[ARM] Always lower direct calls as direct when the outliner is enabled (#66434)"
[Profile] Pass InstrProfSymtab instead of IndexedInstrProfReader when creating BinaryCoverageReader
[RISCV] Match constant indices of non-index type when forming strided ops (#65777)
[flang] Allow continuation of D lines (#66242)
[flang] Fix SIZEOF() expression rewriting (#66241)
[RISCV] Merge some test checks rvv/fixed-vectors-masked-gather.ll [nfc]
[AArch64] Remove the Z#_HI register definitions (#66353)
[SelectionDAG][RISCV] Mask constants to narrow size in TargetLowering::expandUnalignedStore.
[RISCV] Add test case to show bad codegen for unaligned i64 store of a large constant.
[libc++][lit] Atomically update the persistent cache (#66538)
[mlir][sparse] refine sparse fusion with empty tensors materialization (#66563)
[SelectionDAG][RISCV][PowerPC][X86] Use TargetConstant for immediates for ISD::PREFETCH. (#66601)
[flang] Fold NORM2() (#66240)
[mlir][llvm] Use zeroinitializer for TargetExtType (#66510)
[flang][runtime] INQUIRE(UNIT=-666, EXIST=x) should be .FALSE. (#66239)
[RISCV] Fix crash when legalizing mgather/scatter on rv32
Revert "Reapply [Verifier] Sanity check alloca size against DILocalVariable fragment size"
[flang] Accept pointer-valued function results as ASSOCIATED() arguments (#66238)
[VP] Add missing functional_intrinsic properties and add static_assert. NFC (#66199)
[libc][GPU] Fix dependencies for externally installed stub files (#66653)
[Flang][OpenMP] Add trivial conversion pattern for omp.ordered_region (#66085)
[libc++][CI][AIX] modify the equivalence classes of regex_match for locale "cs_CZ.ISO8859-2"
[mlir][bufferization] Remove allow-return-allocs and create-deallocs pass options, remove bufferization.escape attribute (#66619)
[DAG] getNode() - begin generalizing the (zext (trunc (assertzext x))) -> (assertzext x) fold.
[libc++] [string_view] Remove operators made redundant by C++20 (#66206)
[mlir] Fix Bazel build
[InstCombine] Add pre-commit tests for PR65073. NFC.
[Github] Add a new line before the line separator to avoid paragraphs being treated as titles
[ArgPromotion] Update allocsize indices after promotion
[TwoAddressInstruction] Update LiveIntervals after INSERT_SUBREG with undef read (#66211)
[mlir][Vector] Fix -Wunused-function in VectorEmulateNarrowType.cpp (NFC)
[libc++] Simplify how we select modules flavors in the test suite (#66385)
Fix unused variable (#66644)
[mlir][bufferization] Empty tensor elimination for materialize_in_destination (#65468)
[InstSimplify] Fold icmp of `X and/or C1` and `X and/or C2` into constant (#65905)
[Flang][OpenMP] Run Flang-specific OpenMP MLIR passes in bbc
[mlir][Vector] Add a rewrite pattern for better low-precision bitcast… (#66387)
[libc] Add GPU config file using the new format (#66635)
[flang] Lower PRIVATE component names safely (#66076)
[IR] Remove unnecessary bitcast from CreateMalloc()
[SVE] Fix crash when costing getelementptr with scalable target type.
[scudo] Use MemMap in Vector
[libc++] Implement LWG3545: std::pointer_traits should be SFINAE-friendly. (#65177)
[MemCpyOpt] fix miscompile for non-dominated use of src alloca for stack-move optimization (#66618)
[lldb][Docs] Link up the newly restored data formatters page
[Bolt] fix a relocation bug for R_AARCH64_CALL26
[SDag] Fold saddo[_carry] with bitwise-not argument to ssubo[_carry] (#66571)
update_mir_test_checks.py: match undef vreg subreg definitions (#66627)
[mlir][Interfaces][NFC] LoopLikeOpInterface: Consistent TD formatting (#66097)
[mlir][bufferization] Switch tests to new deallocation pass pipeline (#66517)
[llvm][documentation] Fix coroutines documentation (#66420)
[clang][Sema] Remove irrelevant diagnostics from constraint satisfaction failure
[ValueTracking] Remove unused Depth parameter (NFC)
[clang][Interp] Allow zero-init of primitives with an empty init list
[AMDGPU] Fix non-deterministic iteration order in SIFixSGPRCopies (#66617)
[clang][dataflow] Model the fields that are accessed via inline accessors (#66368)
[clang-transformer] Allow stencils to read from system headers. (#66480)
[mlir][IR] Change `MutableOperandRange::operator[]` to return an `OpOperand &` (#66515)
[Clang][test] Replace legacy -target with --target=. (#66572)
[RISCV] Only emit .option when extension is supported
[RISCV][NFC] precommit for D159399
[mlir][SCF] scf.for: Consistent API around `initArgs` (#66512)
[VectorCombine] Fix a spot in commit 068357d9b09cd635b1c2f126d119ce9afecb28f7
[bazel] Port for 88e3358f331d727e7bbcddb2821ef89d25d1ab7a
[flang][runtime] zero size allocation in source allocation (#66124)
[MLIR] Allow comparison of opaque properties (#66378)
[SCEV] Fix incorrect nsw inference for multiply of addrec (#66500)
[ORC] Fix -Wunused-but-set-variable in JITLoaderPerf.cpp (NFC)
[gn build] Port 88e3358f331d
[ORC][JITLink] Non-debuginfo JITLink perf jitdump support.
[RISCV] Teach RISCVMergeBaseOffset to handle BlockAddress
[RISCV] Add tests for merging base offset of BlockAddress
[VectorCombine] Enable transform 'scalarizeLoadExtract' for scalable vector types (#65443)
[WholeProgramDevirt] Use llvm:: qualifier to implement declared functions. NFC
[JITLink][PowerPC][RFC] Make JITLink default in LLJIT for ppc64 elfv2abi
[clang-tidy] Avoid checking magic numbers if _BitInt (#65888)
[RISCV] Use getConstantOperandVal. NFC
[RISCV] Reuse existing XLenVT variable. NFC
[InstCombine] Fix transforms of two select patterns (#65845)
[InstSimplify] Add signed version of pre-commit tests for PR65905. NFC.
[MLIR][Presburger] Implement convertVarKind for PresburgerRelation
[InstSimplify] Add pre-commit tests for PR65905. NFC.
[mlir][bytecode] Check that bytecode source buffer is sufficiently aligned. (#66380)
[InstCombine] Add pre-commit tests for PR65845
[clang] fix stray whitespace in release notes to fix CI failure
[VPlan] Remove stray indent when printing scalar steps recipe.
[RISCV] Combine trunc (sra sext (x), zext (y)) to sra (x, smin (y, scalarsizeinbits(y) - 1)) (#65728)
[AArch64][GlobalISel] Add lowering for constant BIT/BIF/BSP (#65897)
[Support] Remove redundant forwarding functions read/write (NFC) (#66051)
[clang][Docs] Document X86 interrupt attribute (#65662)
[libc++] Fix potentially flaky test joinable.pass.cpp
[OCaml][NPM] Add OCaml bindings to new pass manager
[gn build] Port 695138ca8405
[SDAG][RISCV] Avoid expanding is-power-of-2 pattern on riscv32/64 with zbb
[libc++] implement std::`jthread`
Fix PR#62594 : static lambda call operator is not convertible to function pointer on win32
[gn] port ac182deee828 (RISCVGISel.td)
[BOLT] Incorporate umask into the output file permission
[BOLT] Skip the validation of CFG after it is finalized
workflows/pr-receive: Ignore pull requests with 10 or more commits (#66320)
[clang-format] Properly indent lines inside Verilog case structure (#65861)
[InstSimplify] Add pre-commit tests for PR65910. NFC.
[Github] Fix yml syntax
[InstCombine] Handle constant arms in `select` of `srem` fold
[InstCombine] Use `ConstantInt::getBool` instead of `Constant::getIntegerValue`. NFC.
[SDAG][RISCV] Avoid neg instructions when lowering atomic_load_sub with a constant rhs
[Github] Do not label clang-format files as clang (#66388)
[ELF] Make -t work with --format=binary
[Driver] Change default PCH extension name from .gch to .pch (#66165)
[NFC][hwasan] Use InternalScopedString in ScopedReport
[NFC][hwasan] Remove unused mutex
[NFC][hwasan] Add SANITIZER_GUARDED_BY to a variable
[clang-tidy][NFC] Minor cleanup in ClangTidyMain.cpp
[ELF] Implement getImplicitAddend and enable checkDynamicRelocsDefault for Hexagon
[NFC][sanitizer] Remove InternalScopedString::Append nullptr workaround
[NFC][sanitizer] Remove temp buffer
[NFC][sanitizer] Replace a few AppendF with Append
[ELF] Implement getImplicitAddend and enable checkDynamicRelocsDefault for AMDGPU
[ELF] Implement getImplicitAddend and enable checkDynamicRelocsDefault for PPC32
[NFC][sanitizer] Add InternalScopedString::Append (#66559)
[ELF] Enable checkDynamicRelocsDefault for PPC64
[ELF] checkDynamicRelocsDefault: invert the condition
Use assert instead of branch-to-unreachable
[LAA] Improve the output remark for LoopVectorize (#65832)
[ELF] checkDynamicRelocsDefault: invert the condition
[ELF] Correct a comment about ^=. NFC
[ELF] Remove a special case from ExprValue::getSectionOffset. NFC
[NFC][sanitizer] Rename InternalScopedString::append to AppendF (#66558)
[RISCV] Prefer vrgatherei16 for shuffles (#66291)
[RISCV] Optimize gather/scatter to unit-stride memop + shuffle (#66279)
[RISCV][GlobalISel] Select ALU GPR instructions
[GitHub] Fix `mlir:core` label definition (#66556)
[gn build] Port f4f85e0ab405
[llvm-profdata] Remove MD5 collision check in D147740 (#66544)
Revert "[NFC][sanitizer_symbolizer]Add StackTracePrinter class (#66530)"
[RISCV] Move narrowIndex to be a DAG combine over target independent nodes
[RISCV] Normalize gather/scatter addressing to UNSIGNED_SCALAR
[TLI] Add extend as explicit parameter to shouldRemoveExtendFromGSIndex [nfc]
[libc] Add a developer doc about adding new config options. (#66432)
[mlgo] Fix regalloc tests
[clang] Remove spurious trailing whitespace
[HWASan] Delete the print-memory-usage-android.c test (#66550)
[clang-format] Fix a bug in ContinuationIndenter (#66354)
[Coverage] Add coverage for constructor member initializers. (#66441)
[TOSA] Change EightK MaxScale to 256 (#66536)
[NFC][sanitizer_symbolizer]Add StackTracePrinter class (#66530)
AMDGPU/GlobalISel: Don't pointlessly check for convergent intrinsics
[runtimes] Add CODEOWNERS entries for libc++, libc++abi and libunwind (#66416)
[mlir][sparse] Migrate tests to use new syntax (#66543)
Object: account for short output names (#66532)
[RA] Split a virtual register in cold blocks if it is not assigned preferred physical register
[mlir] Add `maxnumf` and `minnumf` to `AtomicRMWKind` (#66442)
[RISCV] Avoid toggling VL for hidden splat case in constant buildvector lowering
[mlir][transform] Fix crash in transform.get_parent_op. (#66492)
[test] Don't check COMPILER_RT_STANDALONE_BUILD for deps (#66259)
[NFC][Coverage] Rename coverage function attribute name from "Cov Function Name" to "Coverage Function Name"
Fix lld Sphinx build
[SLP]Improve canreuseExtracts for reordering analysis.
[clang][Interp] Handle AttributedStmts (#66495)
[mlir][sparse][tensor] replace bufferization with empty tensor (#66450)
Fix test case in 32db121b29f78e4c41116b2a8f1c730f9522b202.
[lldb] Revive internal data formatter documentation (#66527)
[DebugInfo] Process single-location debug values in variadic form when producing DWARF
[Coverage] Allow Clang coverage to be used with debug info correlation.
Fix-forward a -Wconversion scudo bug.
[ARM] Always lower direct calls as direct when the outliner is enabled (#66434)
[InlineAsm] break up Data Bitfield::Element into corresponding fields (#66297)
[sanitizer_common] Add internal_wcs[n]cpy functions (#66529)
Extend `retcon.once` coroutines lowering to optionally produce a normal result (#66333)
IRBuilder: avoid crash when seeking to start of a BasicBlock with only DebugInfo (#66266)
[lldb-vscode][NFC] Access a pointee type in a simpler way
[OMPIRBuilder] Fix shared clause for task construct
[Github] Use our own issue labeler fork (#65900)
[libc][cmake] Report invalid clang-tidy path (#66475)
[Github] Fix issue number escaping
[lldb][Docs] Update links in Data Formatter section (#66474)
[mlir][vector] Fix vector.broadcast lowering for scalable vectors (#66344)
[mlir][linalg][nfc] Fix typo in a file name
[TextAPI] Add missing flag input for <=tbdv3 formats
[AMDGPU] Tweak tuple weight calculation. NFC. (#66490)
[mlir][Linalg] Move test
Add a or(phi,phi) test without loops
[DX] Support pipeline state masks (#66425)
[Github] Improve formating of PR diffs in bot notifications (#66118)
[AArch64][GlobalISel] Select llvm.aarch64.neon.st* intrinsics (#65491)
[mlir][IR] Insert operations before `SingleBlock`'s terminator (#65959)
[libc++] Improve the verbosity of configuration errors when a compiler flag is not supported (#66379)
[libc++] Update Apple CI jobs to run the C++23 configuration (#66367)
[libc++][NFC] Introduce named states in std::call_once (#66289)
[BOLT] Fix deadloop bug in taildup
Fix clang-tidy sphinx docs
[libc++] Refactor node creation and destruction in list and forward_list (#65614)
[libc++] Implement P2538R1 "ADL-proof std::projected" (#65411)
[GitHub] Add teams for ADT and Support (#65519)
[gn build] Port 72d4d4e3b9f6
Reapply [analyzer] Simplify SVal for simple NonLoc->Loc casts
[AArch64] Guard against a invalid size request in performVecReduceAddCombine
[clang-tidy]add new check `bugprone-compare-pointer-to-member-virtual-function` (#66055)
Reapply [Verifier] Sanity check alloca size against DILocalVariable fragment size
[update_mir_test_checks] Fix new test in non-X86 builds
[clang-format] Generate the style options
[clang-format] Disable string breaking in JS for now (#66372)
[clang][dataflow] Add a test for context-sensitive analysis on a self-referential class. (#66359)
[MemoryBuiltins] Use SmallDenseMap for visited map (NFC)
[IRSymtab] Replace linear time lookup with DenseSet (#66376)
[update_mir_test_checks] Handle multiple defs of vreg (#66483)
[update_mir_test_checks] Precommit test for multiple defs of vreg (#66482)
[AArch64][GlobalISel] Select llvm.aarch64.neon.ld* intrinsics (#65630)
[SCEV] Add tests for PR66066 (NFC)
[ConstantRange] Optimize smul nowrap with constant (NFC)
[clang][Interp][NFC] Remove unused includes
Revert "[MachineLICM] Handle Subloops"
Disable writing an output file to the source directory in Driver test
Revert "[analyzer] Simplify SVal for simple NonLoc->Loc casts (#66463)"
[Github] add spirv backend team (#66075)
[clang][Interp] Diagnose unknown parameter values
[clang][Interp] Check pointer inc/dec ops for initialization
[analyzer] Simplify SVal for simple NonLoc->Loc casts (#66463)
[SLP]Fix PR66176: SLP incorrectly reorders select operands.
[CMake][Z3]Don't attempt to compile / run if cross-compiling (#66355)
[AArch64][SME] Emit Zero instruction for NewZA functions
[AMDGPU] Remove repeated -mtriple options from RUN lines (#66486)
[BOLT][RISCV] Add support for linker relaxation
[GVN] Also remove phi nodes from VN table (PR65447)
[mlir][Conversion] Allow lowering to fixed arrays of scalable vectors
[flang][hlfir] Add hlfir.minval intrinsic (#66306)
Revert "[mlir][bufferization] Switch tests to new deallocation pass pipeline (#66471)"
[libc] Add invoke / invoke_result type traits (#65750)
[mlir][bufferization] Switch tests to new deallocation pass pipeline (#66471)
[clang] Fix wrong warning about missing init for flexible array members (#66341)
Revert "[compiler-rt] Check for and use -lunwind when linking with -nodefaultlibs (#65912)"
[mlir][linalg][transform][python] Clean up _ext.py test. (#66469)
[compiler-rt] Check for and use -lunwind when linking with -nodefaultlibs (#65912)
[mlir][bazel] Fix build.
[MemoryBuiltins] Cache the result of ObjectOffsetSizeVisitor::visit. #64796 (#65326)
[LoongArch] Fix incorrect instruction 'and' in pattern
[GlobalISel] Check for unsupported Windows features on invoke (#65864)
[compiler-rt] Allow to build on CYGWIN (#66348)
[LLD] [MinGW] Ignore --sort-common option (#66336)
[clang][Interp] Visit Logical-not operand as bool
[Assignment Tracking] Trim assignments for untagged out of bounds stores (#66095)
[CMake] Fix BUILD_SHARED_LIBS build on Solaris
[clang][Interp][NFC] Fix a comment typo
[clang][Interp] Convert logical binop operands to bool
[mlir][bufferization] Define a pipeline for buffer deallocation (#66352)
[clang][Interp] Specify C standard in test case
[MLIR] SM_90 integratation test of TMA `128x64xf16` and `64x64xf16` with 128b Swizzling (#65954)
[MemCpyOpt] Avoid infinite loop in processMemSetMemCpyDependence (PR54983)
[mlir][transform][lingalg][python] Replace pdl.operation => transform.any_op. (#66392)
[clang][Interp] Lazily visit unknown global declarations
[CMake] Remove SetTargetTriple (#66464)
[AMDGPU] Correctly restore FP mode in FDIV32 lowering (#66346)
[IRDL] Fix CMake to generate IRDL dialect docs (#66371)
[Sparc] Don't emit __multi3 on 32-bit SPARC (#66362)
[lldb] Add config and decorator for wide character support in Editline (#66447)
[GVN] Invalidate MDA when deduplicating phi nodes
[RISCV] Specify FilterClassField to filter out unneeded pseudos (#65460)
[InstCombine] Remove unnecessary one-use-check (#66419)
[TextAPI] Consolidate TextAPI Reader/Writer APIs. (#66108)
[mlir][vector] Cleanup VectorUnroll and create a generic tile iteration utility
[mlir][spirv][gpu] Add conversion for load/store/mad coop matrix ops (#66311)
[mlir] Add Python bindings for DenseResourceElementsAttr. (#66319)
[Driver] Also respect SysRoot for the system library paths on DragonFly (#66056)
[scudo] Add missing header in combined_test (#66449)
[JITlink][AArch32] Implement ELF::R_ARM_MOVT_ABS and R_ARM_MOVW_ABS_NC (#66219)
[SLP][NFC]Add a test with incorrect reduction of poisoned logical bool.
[mlir][sparse] Migrate more tests to use new syntax (#66443)
[lsan][fuchsia] Add extra check for allocator cache to avoid overflow
[mlir][sparse][gpu] refine type of workspace size variables (#66438)
Fix the calculation of fragmented bytes in secondary (#66422)
[mlir][arith] Add LLVM lowering for `maxnumf`, `minnumf` ops (#66431)
[X86] Introduce a large data threshold for the medium code model
[mlir][gpu][NVPTX] Enable NVIDIA GPU JIT compilation path (#66220)
[mlir][memref] Use dedicated ops in `AtomicRMWOpConverter` (#66437)
[mlir][arith] Introduce `minnumf` and `maxnumf` operations (#66429)
Mark test added in 5a58e98 as requiring ppc, not x86 since it tries to use the powerpc64le target.
[NFC][CodeGen] Change CodeGenOpt::Level/CodeGenFileType into enum classes (#66295)
[GitHub][mlir][spirv] Add missing patterns for SPIR-V in mlir (#66423)
[libc][Obvious] Fix missing entrypoints after moving to generic
[mlir][spirv] Support coop matrix in `spirv.CompositeConstruct` (#66399)
Mark LWG 2426 as complete.
[libc] Implement more input functions on the GPU (#66288)
Add comments (NFC) (#66427)
[HIP] Fix comdat of template kernel handle (#66283)
[mlir][VectorToGPU] Update memref stride preconditions on `nvgpu.mma.sync` path
Add .swift_ast to list of Swift sections (#66426)
[SLP]Do not account non-instructions for external use.
[SLP][NFC]Add a test for non-instruction with external use.
[scudo] Add -Wconversion for tests and clean-up warnings. (#66147)
[libc++] Make sure LWG2070 is implemented as a DR (#65998)
[BOLT][NFC] Simplify RI::selectFunctionsToProcess
[clang][dataflow] Ignore assignment where base class's operator is used (#66364)
Reland "[clang] Add experimental option to omit the RTTI component from the vtable when -fno-rtti is used"
[gn build] Port 71e3642619dd
[AArch64] Relax binary format switch in AArch64MCInstLower::LowerSymbolOperand to all…
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This removes a lot of code duplication, makes the code simpler and prepares the terrain for https://reviews.llvm.org/D101206, which will fix some UB in the node-based containers.
This also allows removing the dependency of list and forward_list on unique_ptr by using __allocation_guard instead.