[OCaml][test] Use correct data layout string. #68781
Merged
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
core.ml would previously set a data layout string of "e" and check that it remained "e". This is fragile when we have the data layout string auto-upgrade facility, and indeed broke when D86310 upgraded this one. As the auto-upgrade logic is not what is being tested here, it seems easier to just use the data layout string that the target expects.
nikic
approved these changes
Oct 11, 2023
fadlyas07
pushed a commit
to greenforce-project/llvm-project
that referenced
this pull request
Oct 14, 2023
* llvm-project/main: [clang][Interp][NFC] Pass PrimType to visitZeroInitializer() [StatepointLowering] Take return attributes of `gc.result` into account (llvm#68439) [gn build] Port f445be9 [gn build] Port 311bc66 [AArch64][compiler-rt] Only build SME ABI routines when compiler supports asm. (llvm#68991) [mlir][mlir-query] Fix shared build. NFC [RISCV] Add CompressPat for c.beqz/bnez with X0 in the first operand. (llvm#69042) Canonically identical types are allowed in compound expressions in C [AArch64][GlobalISel] Expand converage of FMA. [clang] Enable Wenum-constexpr-conversion also in system headers and … (llvm#67528) [AArch64] Additional GISel test for FMA. NFC [Clang] Implement the 'counted_by' attribute (llvm#68750) [mlir][Linalg]: Optimize linalg generic in transform::PromoteOp to avoid unnecessary copies (llvm#68555) [clang][Interp] Implement compound assign operators on bitfields (llvm#67306) [lldb] Add SBType::FindDirectNestedType() function (llvm#68705) Use llvm::endianness::{big,little,native} (NFC) [RISCV] Move hasOneUse() call after opcode check. [clang] Stop including llvm/ADT/StringMap.h (NFC) [Test] Pre-submit tests for llvm#68972 (llvm#69040) [tools] Stop including llvm/ADT/StringMap.h (NFC) Stop including llvm/ADT/STLFunctionalExtras.h (NFC) [RISCV][GISel] Legalize G_SMULO/G_UMULO (llvm#67635) [llvm] Stop including llvm/ADT/StringMap.h (NFC) [clang] Remove unused using decls (NFC) [PowerPC][JITLink] Support R_PPC64_GOT_TLSGD_PCREL34 (llvm#68660) [BOLT] Use llvm::erase_if (NFC) [AST] Stop including llvm/ADT/StringMap.h (NFC) [libc++][NFC] Fix slightly incorrect comment in PSTL documentation [libcxx] Refactoring SIMD function names in PSTL CPU backend (llvm#69029) NFC: Precommit GISel checks for arm64-indexed-memory.ll [GlobalISel][NFC] Add distinct CHECK/SDAG/GISEL run lines to test. [CodeLayout] CDSortImpl: remove two conditions that cannot trigger. NFC [RISCV][SDAG] Fix constant narrowing when narrowing loads (llvm#69015) [RISCV][GISel] Add isel patterns for ADDIW/SRLIW/SRAIW/SLLIW and remove custom selection. (llvm#68470) [LLDB] Fix buildbots [mlir][mlir-query] Introduce mlir-query tool with autocomplete support [LLDB][NFC] Create a namespace for the DWARF plugin (llvm#68150) [scudo] Fix type mismatch in trusty. (llvm#69024) [MSVC, ARM64] Add __prefetch intrinsic (llvm#67174) [gn] port 99d92d1 [scudo] Add specific die functions for linux specific failures. (llvm#68650) Reland "[mlir][arith] Canonicalization patterns for `arith.select` (llvm#67809)" (llvm#68941) [LV] Use LatchVPBB directly instead of going through region (NFC). [lldb][DataFormatter] VectorType: fix format for arrays with size not a power-of-2 (llvm#68907) [clang-format][NFC] Annotate control statement r_braces (llvm#68621) [libc++][NFC] Fix typo in filename [libc++] Use -nostdlib++ on GCC unconditionally (llvm#68832) [HWASAN] Mark built-ins as not built-ins to prevent optimizations (llvm#68936) [AMDGPU] Fix comments about afn and arcp in fast unsafe fdiv handling (llvm#68982) [LLDB] Fix type formatting empty c-strings (llvm#68924) [libc++] Introduce make_test_jthread for jthread tests (llvm#68837) [LLDB][NFC] Remove dead code (llvm#68927) [mlir] Fix distinct attr mismatch error reporting (llvm#68938) [lldb] Fix `po` alias by printing fix-its to the console. (llvm#68755) [mlir][sparse] cleanup sparse tensor materialization parameter setup (llvm#68956) [CodeLayout] Fixing initialization of empty ranges (llvm#68917) [llvm][TableGen][Jupyter] Link to tutorial notebook from README Lower std::string's alignment requirement from 16 to 8. (llvm#68807) [ASan] Recognize lea r10, [rip + XX] (llvm#68910) [lldb][DataFormatter] unordered_map: account for new libc++ __hash_node layout (llvm#68574) [mlir][Tosa] fix fp16/bf16 support for AvgPool2d (llvm#68718) [lldb][test] Temporarily disable TestQueueFromStdModule.py (llvm#68970) [libc++] Re-apply "Remove UB in list, forward_list and __hash_table" [InstCombine] Fold Ext(i1) Pred shr(A, BW - 1) => i1 Pred A s< 0 (llvm#68244) [gn build] Port 2cea1ba [AArch64][NFC] Refactor NEON, SVE and SME classes and multiclasses fo… (llvm#68800) [gn] port 46cb8d9 [clang][Interp] Support AddOffset with 128bit offsets (llvm#68679) Fix typo in CMake var [lldb] fix release build (llvm#68979) Re-apply '[AArch64] Enable "sink-and-fold" in MachineSink by default (llvm#67432)' [NFC][AArch64][LV] Reorganise LV tests using symbols from SLEEF (llvm#68207) [MLIR][Presburger] clang-format and clang-tidy [mlir][arith] Fix canon pattern for large ints in chained arith (llvm#68900) [BOLT][RISCV] Recognize mapping syms with encoded ISA (llvm#68964) Revert "[Sema] Add check for bitfield assignments to integral types" (llvm#68963) [mlir][ODS] Omit printing default-valued attributes in oilists (llvm#68880) [mlir][linalg] Add NHWC + FHWC Img2Col (llvm#68708) [NFC] Fixed typo in LangRef.rst (llvm#68789) [mlir][doc] Trim summary text during DocGen (llvm#68477) [ADT] Stop including identity.h (NFC) [CodeGenPrepare] Check types when unmerging GEPs across indirect branches (llvm#68587) [AMDGPU] Make ubsan happy (llvm#68959) Revert "Add Documentation for Execution Results Handling in Clang-Repl (llvm#65650)" [llvm-remarkutil] Silence a GCC warning about missing returns after a fully covered switch. NFC. [llvm-remarkutil] Fix building with Xcode 12 [Clang] Fix tautological assertion in `Sema::CheckX86BuiltinTileDuplicate` [flang][runtime] Fix IsContiguous for zero and one element arrays (llvm#68869) [llvm] Use llvm::erase_if (NFC) [ELF] Restore the --call-graph-profile-sort=hfsort default before llvm#68638 [mlir] Apply ClangTidy fix (NFC) [mlir][sparse] remove unused sparse tensor iterator (llvm#68951) [ArmSVE][NVVM][Bazel] Added Features to BUILD.bazel file (llvm#68949) [Support] Stop including cstddef (NFC) Use llvm::endianness::{big,little,native} (NFC) [mlir][sparse] add a forwarding insertion to SparseTensorStorage (llvm#68939) [AMDGPU] Use llvm::endianness::little (NFC) [IR] Move isConvergenceControlIntrinsic under "namespace llvm" (NFC) [llvm] Remove "using namespace llvm;" from header files (NFC) [PowerPC][Atomics] Remove redundant block to clear reservation (llvm#68430) [SVE][InstCombine] Delete redundante sel instructions with ptrue [SVE][InstCombine] Precommit tests for select + ptrue [mlir][affine] ValueBoundsConstraintSet: Fully compose affine.apply (llvm#68899) [PowerPC][JITLink] Support R_PPC64_GOT_PCREL34 (llvm#68658) [mlir][sparse] remove dead code from utils (llvm#68943) [mlir][nfc] Rename type constraint for scalable vectors (llvm#68808) Revert "[mlir][arith] Canonicalization patterns for `arith.select` (llvm#67809)" [libc++] Remove libc++'s own <setjmp.h> header (llvm#68806) quick fix for TestDumpDwo [mlir][nvvm] Fix circular dependency in (llvm#68934) [TSAN] add support for riscv64 (llvm#68735) [CodeLayout][NFC] Using MergedVector to avoid extra vector allocations (llvm#68724) [Driver] Have -rdynamic be a no-op on Haiku (llvm#67872) [AMDGPU] Change the representation of double literals in operands (llvm#68740) [libc++] Check formatting with clang-format 17 (llvm#68928) [clang][Interp] Fix crash during `InterpStack` printing (llvm#68246) [InstCombine] Improve logic for adding flags to shift instructions. [InstCombine] Tests for adding flags to shifts; NFC [ValueTracking] Do more thorough non-zero check in `isKnownToBePowerOfTwo` when `OrZero` is no set. [ValueTracking] Add support for non-splat vecs in cmpExcludesZero Just a small QOL change. [ValueTracking] Add tests for `cmpExcludesZero` for non-splat vecs; NFC Fix minimal-throw-catch.ll on x86 mac Revert "Reapply "InstCombine: Introduce SimplifyDemandedUseFPClass"" [mlir][sparse] implement sparse_tensor.reorder_coo (llvm#68916) [gn build] Port 1c9035d [SystemZ][z/OS] Add UtcClock extension to chrono.h/.cpp (llvm#67846) [ValueTracking] Add better support for ConstantRange(And) [ValueTracking] Add better support for ConstantRange(Shl) [ValueTracking] Add more tests for constant ranges; NFC [libc++][mdspan] Fix extents CTAD (llvm#68737) Add `target modules dump separate-debug-info` (llvm#66035) [Sema] Add check for bitfield assignments to larger integral types (llvm#68276) [clang][NFC] Specify Type and ExtQuals as having 16-byte alignment (llvm#68377) [libc++] Improve the output of the generated-output CI job (llvm#68903) [gn build] Port 31c2cf1 [gn] port 0ce6255 (HipStdPar) [mlir][ArmSVE] Avoid UBSAN issue with VectorType::Builder (NFC) [mlir][sparse] introduce sparse_tensor.reorder_coo operation (llvm#68827) [gn] port f445be9 [clang][Tests] Fix shared build. NFC [Remarks] Fix '-fpermissive'. NFC [ARM] Correctly handle .inst in IT and VPT blocks (llvm#68902) [Flang][OpenMP] NFC: Fix test [AArch64] Fix incorrect big-endian spill in foldMemoryOperandImpl (llvm#65601) [Remarks] Introduce count subcommand for llvm-remarkutil. (llvm#66214) Reland: [MLIR][Transforms] Fix Mem2Reg removal order to respect dominance (llvm#68877) LoopVectorize/test: clean up reduction.ll; generate using UTC (NFC) (llvm#68890) [mlir][ArmSVE] Add convert_to/from_svbool ops (llvm#68586) [InstCombine] Extend `sext`/`zext` boolean additions to vectors [llvm-readtapi] Fix bad symlink with readtapi tool (llvm#68835) Add Documentation for Execution Results Handling in Clang-Repl (llvm#65650) [IR][NFC] Assert that FunctionType::getParamType is in-bounds (llvm#68887) [mlir][nvvm] Move `BasicPtxBuilder` Interface to Its Own File (NFC) (llvm#68095) Reland "[compiler-rt][aarch64] Add SME ABI support routines." (llvm#68875) [AArch64] Fix schedmodel pre/post-index loads and stores for TSV110 [test] precommit sched model for tsv110, NFC [PhaseOrdering] Regenerate test checks (NFC) [gn] port 2b05fa8 (InterpTests) [libcxx] <experimental/simd> Add generate constructor of class simd (llvm#66737) [LoopVectorize] Regenerate test checks (NFC) [Clang] Regenerate test checks (NFC) [bazel] add llvm/HipStdPar (llvm#68883) [flang]Check for dominance in loop versioning (llvm#68797) [flang][openmp] Don't mark loop variables with explicit DSA as predetermined (llvm#68723) [PhaseOrdering] Add test for switch with different GEP types (NFC) Revert "[mlir][ODS] Omit printing default-valued attributes in oilists (llvm#68694)" [mlir][ODS] Omit printing default-valued attributes in oilists (llvm#68694) [clang][Tests] Add clangTooling dep to Interp unittest [AMDGPU] Update ASAN tests with update_test_checks. (llvm#68688) [clang][dataflow] Various formatting tweaks to HTMLLogger. (llvm#68392) Revert "[compiler-rt][aarch64] Add SME ABI support routines." (llvm#68875) [clang][Interp] Fix expected values for Pointer API [clang][Interp][NFC] Remove Pointer.h include from Function.h [llvm-lib] [Object] Use ECSYMBOLS section for ARM64EC importlib symbols. (llvm#68328) [SystemZ][z/OS] Add vararg support to z/OS (llvm#68834) [HIP][LLVM][Opt] Add LLVM support for `hipstdpar` [flang][driver] add command line arguments for alias tags pass [InstCombine] Canonicalize `(X +/- Y) & Y` into `~X & Y` when Y is a power of 2 (llvm#67915) [MachineSink] Reduce the number of unnecessary invalidations of StoreInstrCache (NFC) (llvm#68676) [AArch64] Fix postinc operands for Cortex-A57 scheduling [MachineSink] Use LLVM ADTs (NFC) (llvm#68677) [bazel] update llvm-config.h.cmake ArgList: Simplify 3-opt overload of AddAllArgs. NFC [bazel] fix linter complains for libc bazel [mlir][python] Improve sanitization of python names (llvm#68801) [compiler-rt][aarch64] Add SME ABI support routines. [AMDGPU] Add lit() asm operand modifier for SP3 compatibility. (llvm#68839) [PowerPC] Use zext instead of anyext in custom and combine (llvm#68784) [LoongArch] Support finer-grained DBAR hints for LA664+ (llvm#68787) [mlir][tosa] Add aggressiveReduceConstant argument for the constant reduce optimization (llvm#68765) [BasicBlockSections] Introduce the path cloning profile format to BasicBlockSectionsProfileReader. (llvm#67214) [mlir][Bazel] Don't duplicate files in data and test attribute. [BOLT] Return proper minimal alignment from BF (llvm#67707) [Support] Stop including MathExtras.h (NFC) [Affine][Bazel] Added AffineOps to BUILD.bazel file (llvm#68842) [mlir][memref] Fix index delinearization for CollapseShapeOp folding (llvm#68833) [Driver] Group together linker options using addAllArgs (llvm#68349) [PowerPC][Atomics] Simplify atomicrmw i128 patterns. NFC. (llvm#68779) [libc] Mark operator== const to avoid ambiguity in C++20. (llvm#68805) [XCOFF][OBJECT] get symbol size by calling XCOFF interfaces (llvm#67304) [AIX] recognize vsr in inline asm for AIX (llvm#68476) [MergeFuncs] Use sizeWithoutDebug to decide if we create a thunk (llvm#68627) [RA] Disable split around hint register if optimize for size (llvm#68619) [mlir][Transform] Create a transform interpreter and a preloader pass (llvm#68661) [lldb-vscode] Update installation instructions (llvm#68234) [mlir][python] python binding for the affine.store op (llvm#68816) [clang][dataflow] Add support for lambda captures (llvm#68558) [libunwind] Avoid reading OOB for non-existent .eh_frame_hdr (llvm#68815) [libunwind] Fix wrong end argument passed to decodeEHHdr() (llvm#68813) [mlir][sparse] remove tests (llvm#68826) [llvm-cov gcov] Make recursive propagateCounts iterative. NFC [X86] Use indirect addressing for high 2GB of x32 address space [mlir][sparse] fix unused variable warning on release build (llvm#68824) [Support] Add `\g<ref>` backreferences in Regex::sub() (llvm#67220) [AArch64] Fix postinc operands for Neoverse-N2 scheduling [flang] Add missing dependency FIRCodeGen -> FIRAnalysis [mlir][sparse] simplify ConvertOp rewriting rules (llvm#68350) [Driver] Silence stdlib warning when linking C on FreeBSD (llvm#68011) [asan] Ensure __asan_register_elf_globals is called in COMDAT asan.module_ctor (llvm#67745) [flang] use TBAAForest in TBAABuilder [mlir][sparse] refactor dim2lvl/lvl2dim passing into MapRef (llvm#68649) [lld] Don't allow -dynamicbase:no on ARM64EC. [flang][openacc] Relax type check for private recipe on acc.serial (llvm#68814) [mlir][bufferization] Transfer `restrict` during empty tensor elimination (llvm#68729) [SlotIndexes] Simplify getInstructionFromIndex. NFCI. [SlotIndexes] Use upper/lower bound terminology for MBB searches. NFC. (llvm#68802) [flang] add missing dependency FIRTransforms -> FIRAnalysis [OpenMPIRBuilder] Add ThreadLimit and NumTeams clauses to teams construct (llvm#68364) [MLIR][NVGPU] Introduce `warpgroup.init.accumulator` Op (llvm#67530) Revert "Reland: [MLIR][Transforms] Fix Mem2Reg removal order to respect dominance (llvm#68767)" [clang] __is_trivially_equality_comparable for types containing lambdas (llvm#68506) [Flang][OpenMP] NFC: Port a few parallel tests to HLFIR flow [AArch64][GlobalISel] Support more types for TRUNC (llvm#66927) [mlgo][nfc] Remove / fix vestigial references to Tensorflow [SlotIndexes] Simplify SlotIndex() and isSameInstr(). NFC. [AArch64][PAC] Check authenticated LR value during tail call [flang][FIR] add FIR TBAA pass [Local] Merge constant / non-constant code paths (NFC) [flang][FIR] add FirAliasAnalysisOpInterface (llvm#68317) [ConstraintElim] Add test for llvm#68751 (NFC) [libcxx] [docs] Update the MinGW build example (llvm#68790) [libcxx] [docs] Remove mention of a MSVC/Debug mode caveat (llvm#68791) [libc++] Run clang-format on newly-added test to fix CI Revert "Diagnose problematic uses of constructor/destructor attribute (llvm#67673)" Diagnose problematic uses of constructor/destructor attribute (llvm#67673) Fix LLVM sphinx bot Revert "Diagnose problematic uses of constructor/destructor attribute (llvm#67673)" [SVE][CodeGenPrepare] Sink address calculations that match SVE gather/scatter addressing modes. (llvm#66996) Diagnose problematic uses of constructor/destructor attribute (llvm#67673) [flang] Use object before converts in fir.dispatch (llvm#68589) Revert "[HIP][LLVM][Opt] Add LLVM support for `hipstdpar`" [HIP][LLVM][Opt] Add LLVM support for `hipstdpar` [TableGen] New bang operator `!repr`. (llvm#68716) [AArch64] Fix postinc operands for Neoverse-N1 scheduling [AMDGPU] Add encoding/decoding support for non-result-returning ATOMIC_CSUB instructions (llvm#68684) [LoongArch] Add some atomic tests (llvm#68766) [PowerPC] Add test for llvm#68783 (NFC) [OCaml][test] Use correct data layout string. (llvm#68781) [EarlyIfConversion] Simplify condition after llvm#65729 [VirtRegMap] Simplify condition after llvm#65729 [GVN] Always require LoopInfo [X86] Align i128 to 16 bytes in x86 datalayouts [AArch64] Fix postinc operands for Neoverse-V1 scheduling [mlir][docgen] Add option to enable Hugo Docs specific features (llvm#68725) [clang][Interp] Remove expected-no-directives lines Clean up strange uses of getAnalysisIfAvailable (llvm#65729) [AMDGPU] Update test remat-smrd.mir [clang][C2x] Remove confusing diagnostic auto storage class specifier (llvm#68710) [clang][Interp] Limit MaxBitInt to 128 bits [BOLT] Preserve label annotations for injected functions (llvm#68713) [BOLT][NFC] Add allocator id to MCPlusBuilder::setLabel (llvm#68707) Reland: [MLIR][Transforms] Fix Mem2Reg removal order to respect dominance (llvm#68767) CycleInfo: Fix splitCriticalEdge (llvm#68584) [clang][RISCV] Fix the condition of checking signature in getIndex (llvm#67403) [clang][Interp] Add basic support for _BitInt (llvm#68069) [Docs] Fix broken LLVM doc pages (llvm#68631) Use llvm::endianness::{big,little,native} (NFC) [llvm] Drop unaligned from calls to llvm::support::endian::{read,write} (NFC) [mlir][Bazel] Adjustments for 8c67c48 [mlir][Bazel] Fix wrong dependency path. [Interp] Use llvm::endianness::native (NFC) [Mesh][Bazel] Added MeshOps to BUILD.bazel file (llvm#68759) Use llvm::endianness (NFC) [libc++][NFC] Fix typo in synopsis comment [Driver] Rename AddAllArgs (initialization list overload) to addAllArgs [clang][driver] Support option '-r' for target AVR (llvm#68484) [RISCV] Add Stackmap/Statepoint/Patchpoint support without targets [X86] Fix value-extending/truncating loads and stores of __ptr32/__ptr64 pointers (llvm#67168) [llvm] Use llvm::endianness::{big,little,native} (NFC) [MLIR] Refactor FuncOpToLLVM to operate on FunctionOpInterface instead of FuncOp (NFC) (llvm#68665) Move global namespace cl::opt inside llvm:: or internalize them [RISCV] Make PostRAScheduler a target feature (llvm#68692) [Clang][PowerPC] Support tune CPU in target attribute (llvm#68681) [LoongArch] Improve codegen for atomic ops (llvm#67391) [FlowSensitive] Stop including llvm/ADT/Any.h (NFC) [CodeGen] Use a range-based for loop (NFC) [llvm] Drop unaligned from calls to llvm::support::endian::{read,write} (NFC) [flang] Remove unused using decls (NFC) [analyzer] Compute length of string literal initializers (llvm#66990) (llvm#68368) [dsymutil] Keep DSYMUTIL_REPRODUCER_PATH in addition to LLVM_DIAGNOSTIC_DIR Revert "[RISCV] Shrink vslideup's LMUL when lowering fixed insert_subvector (llvm#65997)" Revert "[lldb] Fix `po` alias by printing fix-its to the console. (llvm#68452)" [lldb] Fix `po` alias by printing fix-its to the console. (llvm#68452) [MachO] Fix unaligned load in extractSections (llvm#68741) [mlir][linalg] Enable CollapseLinalgDimensions to collapse memref based operations (llvm#68522) [compiler-rt] During profile flushing, setup SIGKILL mask earlier (llvm#68622) [clang-tidy][Doc][NFC] Regenerate list.rst & clang-analyzer doc [MLIR] Add a new Mesh dialect (llvm#68007) Revert "[MLIR][Transforms] Fix Mem2Reg removal order to respect dominance (llvm#68687)" (llvm#68732) [libcxx] Allow string to use SSO in constant evaluation. (llvm#66576) [RISCV][GlobalISel] Select G_PTR_ADD (llvm#67605) [RISCV][GlobalISel] Select G_ICMP, G_LOAD, G_STORE, G_ZEXTLOAD (llvm#67619) [RISCV][GlobalISel] Lower calls to variadic functions (llvm#68271) [gn build] Port f901254 [libc++][hardening] Undeprecate safe mode (llvm#68391) [OpenMP] Disable flaky libomptarget AMDGPU test [libc++] Remove libc++'s own <limits.h> (llvm#65472) [libc] Fix missing warp sync for the NVPTX assert [Sparc] Remove duplicate ALU and SETHI instructions (NFCI) (llvm#66851) [dsymutil] Rename DSYMUTIL_REPRODUCER_PATH to LLVM_DIAGNOSTIC_DIR [flang][openacc][NFC] Fix TODO messages [flang][openacc] Support assumed shape array in firstprivate recipe (llvm#68640) [ConstraintElim] Support decrementing inductions with step -1. (llvm#68644) [GVN] Drop Clobber dependency if store may overwrite only the same value (llvm#68322) [GVN][NFC] Add pre-commit test [ELF] Making cdsort default for function reordering (llvm#68638) [RISCV][NFC] Correct/clarify RISCVPushPopOptimizer.cpp header comment [RISCV][NFC] Correct RISCVInstrInfoZc.td header comment - Zc* extensions no longer experimental [RISCV][NFC] Remove mention of experimental extensions from RISCVInstrInfoZb.td header [lld] [MTE] Drop MTE globals for fully static executables, not ban (llvm#68217) [AArch64][GlobalISel] Legalize G_VECREDUCE_ADD bigger types (llvm#68202) [mlir][TosaToLinalg] Lower tosa reduce operations to linalg reduce op (llvm#68480) Fix attributes of llvm.amdgcn.make.buffer.rsrc (llvm#68609) [LLVM][DWARF] Create thread safe context for DWO/DWP DWARFContext (llvm#68262) [libc][bazel] Define libc namespace in a separate file and drop the `release` flag (llvm#68563) [mlir][python] Handle dashes in op name [AArch64] Fix x18 being used by nest ptrs with MSVC ABI (llvm#68008) [MLIR][Transforms] Fix Mem2Reg removal order to respect dominance (llvm#68687) [mlir][spirv] SCFToSPIRV: fix WhileOp block args types conversion (llvm#68588) [AArch64] Fix postinc operands for Cortex-A55 scheduling [clang][Interp] Basic support for bit fields [clang][ExprConst] Short-circuit ConstantExpr evaluation Revert "[HIP][LLVM][Opt] Add LLVM support for `hipstdpar`" in order to address build breakage. [HIP][LLVM][Opt] Add LLVM support for `hipstdpar` [clang][Interp] Handle CXXScalarValueInitExprs (llvm#67147) [ValueTracking] Use SimplifyQuery in haveNoCommonBitsSet() (NFC) [BOLT][runtime] Add start & fini symbols (llvm#68505) [flang][hlfir] Cast actual cst len character to stmt func dummy type (llvm#68598) [BOLT] Err when linking objects of different architectures (llvm#66770) [clang][Interp] Fix returning nullptr from functions (llvm#67229) [AArch64] Fix postinc operands for Cortex-A53 scheduling [ValueTracking] Use SimplifyQuery for the overflow APIs (NFC) [Analysis] Move SimplifyQuery into separate header (NFC) [SCEV] Make invalidation in SCEVCallbackVH more thorough (llvm#68316) [gn build] Port 71ae858 [clang][Interp] Fix intap tests if int128 is not supported [clang][analyzer] Add C++ array delete checker [clang][analyzer] Rename DeleteWithNonVirtualDtorChecker to CXXDeleteChecker Revert "[X86] Support EGPR (R16-R31) for APX (llvm#67702)" [llvm-rc] [test] Convert the recently added menuex tests to Unix newlines. NFC. [Sema] Handle large shift counts in GetExprRange (llvm#68590) [llvm-rc] [test] Fix a misplaced extra argument in a test comment. NFC. [bazel] fix build for 4790578 [mlir][ArmSVE] Add convert.from/to.svbool intrinsics (llvm#68418) [bazel] fix build for 4790578 [LVI][CVP] Treat undef like a full range (llvm#68190) [clang][Interp][NFC] Move int128 tests to their own file [AArch64] Fix postinc operands for Cortex-A510 scheduling [clang]Avoid diagnose invalid consteval call for invalid function decl (llvm#68646) [AMDGPU] Use absolute relocations when compiling for AMDPAL and Mesa3D (llvm#67791) [gitattributes] Don't mark all llvm-rc test Inputs as binary (llvm#68583) [Aarch64] Materialize immediates with 64-bit ORR + EOR if shorter (llvm#68287) [clang] [MinGW] Explicitly always pass the -fno-use-init-array (llvm#68571) [MLIR][TOSA] Remove failed test cases (llvm#68664) Replace hard coded numbers from 462d583 with regex so the test passes on downstream projects that may define additional opcodes. [Clang] Fix missing diagnostic for non-standard layout type in `offsetof` (llvm#65246) [clang] Fix several issues in the generated AttrHasAttributeImpl.inc -fsanitize=alignment: check memcpy/memmove arguments (llvm#67766) [mlir] Make overloads of SymbolTable::replaceAllSymbolUses consistent. (llvm#68320) [JITLink] Allow pre-existing eh-frame CIE edges on FDEs. [mlir] remove some GCC warning llvm#68409 (llvm#68528) [MLIR][TOSA] Add tosa.slice operation conversion failure scenario (llvm#68578) [RISCV] Simplify PatSetCC_m and PatFprFprDynFrm_m (llvm#68562) [X86] Support EGPR (R16-R31) for APX (llvm#67702) [C++20] [Modules] Don't emit function bodies which is noinline and av… (llvm#68501) [X86][NFC]Update test cases after D159250 (llvm#68517) Support big endian in llvm-symbolizer's data location dwarf info parser (llvm#67284) [mlir][sparse] Extract `StorageSpecifierToLLVMPass` from bufferization pipeline (llvm#68635) [mlir][bufferization][NFC] Simplify `bufferizeOp` function signature (llvm#68625) [mlir][bufferization] Allow cyclic function graphs without tensors (llvm#68632) [flang][openacc] Support array with dynamic extent in private recipe (llvm#68624) [mlir][sparse] Fix errors in doc and tests (llvm#68641) [lldb] add stop-at-user-entry option to process launch (llvm#67019) [flang][runtime] Workaround cuda-11.8 compilation issue. (llvm#68459) [gn] port 24b0c43 [Github] Add PR author name to subscription email (llvm#68440) [Sanitizer][Docs] Reformat CMake invocation in docs Reapply "[scudo] Make local cache be agnostic to the type of node in … (llvm#68633) [Sanitizer][Docs] Improve docs on building Asan (llvm#68636) [OpenACC][Bazel] Added OpenACCOpsInterfaces to BUILD.bazel file (llvm#68639) [llvm][objdump] Remove support for printing the embedded Bitcode section in MachO files. (llvm#68457) [ASan][Windows] Fix rip-relative instruction replacement (llvm#68432) [mlir][sparse] add expanded size to API (llvm#68614) [mlir][sparse] rename map utility (llvm#68611) Remove LLDB introspection entrypoints from the shim (llvm#68450) [flang][hlfir] address char_convert issues as mentioned in llvm#64315 (llvm#67570) [VectorCombine]Fix a crash during long vector analysis. [Debuginfod] Add \n to llvm-debuginfod-find error [mlir][python] generate value builders (llvm#68308) Revert "[scudo] Make local cache be agnostic to the type of node in f… (llvm#68626) [clang-cl] Document behavior difference of strict aliasing in clang-cl vs clang. (llvm#68460) Revert "[compiler-rt] Allow Fuchsia to use 64-bit allocator for RISCV (llvm#68343)" [LLD] [MinGW] Handle the --dll option (llvm#68575) [libcxx] [test] Quote the python executable in the executor (llvm#68208) [scudo] Make local cache be agnostic to the type of node in freelist (llvm#67379) [clang-tidy] Add support for optional parameters in config. [X86] Fix logic for optimizing movmsk(bitcast(shuffle(x))); PR67287 [X86] Add tests for incorrectly optimizing out shuffle used in `movmsk`; PR67287 [NVPTX] Improve lowering of v4i8 (llvm#67866) Revert "[Clang] Implement the 'counted_by' attribute" (llvm#68603) [RISCV] Generaize reduction tree matching to fp sum reductions (llvm#68599) [mlir][arith] Canonicalization patterns for `arith.select` (llvm#67809) Annotate enum r brace [clang-format][NFC] Annotate more r_braces [flang][openacc] Added acc::RecipeInterface for getting alloca insertion point. (llvm#68464) [lldb][NFCI] Remove use of ConstString from FilterRule in StructuredDataDarwinLog (llvm#68347) [FrontEnd] Fix a warning [mlir][nvvm] Introduce `elect.sync` Op (llvm#68323) [flang][openacc] Support allocatable and pointer array in private recipe Revert "[mlir][tools] Introduce tblgen-to-irdl tool (llvm#66865)" Revert "[flang][openacc] Support allocatable and pointer array in private recipe (llvm#68422)" [flang][openacc] Support allocatable and pointer array in private recipe (llvm#68422) [mlir][sparse] move variable into assert to avoid 'unused' error (llvm#68604) [mlir][tosa] Add verifier for `ArgMax` operator (llvm#68410) [mlir][sparse] replace specialized buffer setup with util code (llvm#68461) [InstCombine] Fold zext-of-icmp with no shift (llvm#68503) [InstCombine] Precommit test for PR68465 [AArch64][LoopVectorize] Use upper bound trip count instead of the constant TC when choosing max VF (llvm#67697) [clang-tidy] Improve `ExceptionSpecAnalyzer`s handling of conditional noexcept expressions (llvm#68359) Revert "[SCEV] Don't invalidate past dependency-breaking instructions" [Sink] Fix bugs of sinking unreachable BB from phi (llvm#68576) [GlobalISel] Add support for *_fpmode intrinsics [ConstantFold] Avoid some uses of ConstantExpr::getSExt() (NFC) [mlir][vector] Restore assert and fix typos (llvm#68581) [OpenMPIRBuilder] Remove wrapper function in `createTask`, `createTeams` (llvm#67723) [VP][NFC] Add 32-bit test for VP (llvm#68582) [OpenMP] Fix setting visibility on declare target variables [libc++] LWG 3821 uses_allocator_construction_args should have overload for pair-like (llvm#66939) [SCEV] Don't require positive BTC when non-zero is sufficient [IndVars] Add test for phi select exit value with large BTC (NFC) Revert "[MachineLICM] Relax overlay conservative PHI check (llvm#67186)" (llvm#68580) Reapply [Verifier] Sanity check alloca size against DILocalVariable fragment size [VP] IR expansion for bitreverse/bswap (llvm#68504) [clang-tidy] add namespace qualifier NFC (llvm#68579) [analyzer][NFC] Remove outdated FIXME comment (llvm#68211) [NFS][CodeMoverUtils] Add comment saying not ready for production usage. (llvm#68573) [MachineLICM] Relax overlay conservative PHI check (llvm#67186) Revert "[CodeGen] Really renumber slot indexes before register allocation (llvm#67038)" [MemCpyOpt] Fix the invalid code modification for GEP (llvm#68479) [CodeGen] Really renumber slot indexes before register allocation (llvm#67038) [mlir][bazel] Fix after 7bbfd2a [DAG] Remove unused variable 'VT' in DAGCombiner.cpp (NFC) [VectorCombine] Rename foldBitcastShuf -> foldBitcastShuffle. NFC. [LV] Cache call vectorization decisions (llvm#66521) [DAG] foldSelectOfBinops - correctly handle select of binops where ResNo != 0 [Docs] Fix GEP type in example (llvm#68533) Fix Wunused-variable warning. NFC. Fix Wparentheses warning. NFC. [mlir][ArmSVE] Restructure sources to match ArmSME dialect (NFC) (llvm#68399) [AArch64][SME] Zero reserved bytes when allocating a new TPIDR2 object (llvm#68411) [clang-format][NFC] Make InsertNewlineAtEOF a little more efficient [SPIRV] Fix SPV_KHR_expect_assume support (llvm#67793) Use llvm::endianness{,::little,::native} (NFC) Update MLIR conversion to LLVMFunc to account better for properties (llvm#67406) [flang] Set func.func arg attributes for procedure designators (llvm#68420) [flang][hlfir] use fir.type_info to skip runtime call if nofinal is set (llvm#68397) [lldb][Docs] Use RST link format in IntelPT doc [lldb][Docs] Fix typo in debugging lldb doc [llvm] Drop unaligned from calls to llvm::support::endian::{read,write} (NFC) [Sparc] Replace CMP instructions with InstAlias (NFCI) (llvm#66859) [Driver] Hook up Haiku ARM support (llvm#67222) Change-Id: I5c57794f25a7a054205738606c2318e8b9d4c4c3 Signed-off-by: greenforce-auto-merge <greenforce-auto-merge@users.noreply.github.com>
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.
core.ml would previously set a data layout string of "e" and check that it remained "e". This is fragile when we have the data layout string auto-upgrade facility, and indeed broke when D86310 upgraded this one. As the auto-upgrade logic is not what is being tested here, it seems easier to just use the data layout string that the target expects.