October 2021 Archives by author

• Messages sorted by: [ thread ] [ subject ] [ date ]
• More info on this list...

Starting: Fri Oct 1 04:08:09 PDT 2021
Ending: Sun Oct 31 22:12:51 PDT 2021
Messages: 318

• [Mlir-commits] [mlir] a1d1c31 - Add a `check-mlir-build-only` build target that only builds the dependencies of the `check-mlir` test target (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 237d18a - Fix memory leaks in mlir/test/CAPI/ir.c   Mehdi Amini
• [Mlir-commits] [mlir] db79f4a - Free memory leak on duplicate interface registration   Mehdi Amini
• [Mlir-commits] [mlir] 107198f - Fix memory leaks in mlir/unittests/MLIRTableGenTests   Mehdi Amini
• [Mlir-commits] [mlir] 57d9ade - Fix memory leaks in MLIR unit-tests (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 9312cb6 - Fix Undefined Behavior in MLIR Diagnostic: don't call memcpy with a nullptr source   Mehdi Amini
• [Mlir-commits] [mlir] fe48ecb - Fix memory leak in mlir-cpu-runner/sgemm_naive_codegen.mlir (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 4b28638 - Fix multiple memory leaks in mlir-cpu-runner tests (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] bac4529 - Fix/disable more MLIR tests exposing leaks in ASAN builds (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 2da3fac - Fix memory leak in MLIR SPIRV ModuleCombiner   Mehdi Amini
• [Mlir-commits] [mlir] 51b9f0b - Fix memory leaks in MLIR integration tests for vector dialect (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 5de44d2 - Disable leak check for the MLIR Sparse CPU integration tests (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 903facd - Disable leak check for the MLIR Linalg CPU integration tests (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] cb2e0eb - Fix last leaky MLIR integration test (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 86f5028 - Exclude MLIR python binding tests from Sanitizer tests for now   Mehdi Amini
• [Mlir-commits] [mlir] bce0c64 - Fix ASAN execution for the MLIR Python tests   Mehdi Amini
• [Mlir-commits] [mlir] 00b7d95 - Stop stripping the `std.` prefix when printing operations in a region with a defined default dialect   Mehdi Amini
• [Mlir-commits] [mlir] 82cd8b8 - Fix test-rsqrt.mlir to accept AMD's approximation of rsqrt as well   Mehdi Amini
• [Mlir-commits] [mlir] 70a8d61 - evert "[mlir] Limit Python dependency to Development.Module when possible."   Mehdi Amini
• [Mlir-commits] [mlir] 8c9f506 - Disable mlir/test/mlir-cpu-runner/async-group.mlir with ASAN   Mehdi Amini
• [Mlir-commits] [mlir] dda810c - Fix a comment at call-site to match the declared parameter (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 6f5c685 - Avoid one branch in Operation::getOpOperand()/getOperand() (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] d0d991c - Improve fatal error message when an Attribute or Type wasn't initialized by a dialect (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] e2f16be - Fix clang-tidy warnings in MLIR Python bindings (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] cb11ddb - Revert "[MLIR][OpenMP] Add support for ordered construct"   Mehdi Amini
• [Mlir-commits] [mlir] a8c1d9d - Add a clear() method on the PassManager (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] f431d38 - Make Python MLIR Operation not iterable   Mehdi Amini
• [Mlir-commits] [mlir] 693c61b - [mlir] Enable loop peeling for "reduction" dimensions of tiled_loop.   Alexander Belyaev
• [Mlir-commits] [mlir] 96cee29 - [mlir] Allow polynomial approximations for N-d vectors.   Alexander Belyaev
• [Mlir-commits] [mlir] a9a0ea9 - [mlir] Update Erf approximation.   Alexander Belyaev
• [Mlir-commits] [mlir] 7d79a25 - [mlir] Use float literals to make Windows build happy.   Alexander Belyaev
• [Mlir-commits] [mlir] 16b8f4d - [mlir][sparse] add a "release" operation to sparse tensor dialect   Aart Bik
• [Mlir-commits] [mlir] 57c56cf - X86Vector: relax checks in rsqrt's integration test   Aart Bik
• [Mlir-commits] [mlir] 849f016 - [mlir][sparse] accept affine subscripts in outer dimensions of dense memrefs   Aart Bik
• [Mlir-commits] [mlir] 9fc35c3 - [mlir][sparse] add action enum, group enums together, few asserts   Aart Bik
• [Mlir-commits] [mlir] 35517a2 - [mlir][sparse] add init sparse tensor operation   Aart Bik
• [Mlir-commits] [mlir] a652e5b - [mlir][sparse] emergency fix after constant -> arith.constant change   Aart Bik
• [Mlir-commits] [mlir] b24788a - [mlir][sparse] implement sparse tensor init operation   Aart Bik
• [Mlir-commits] [mlir] e9b1c97 - [mlir][sparse] run less combinations of SpMM in test (to reduce runtime)   Aart Bik
• [Mlir-commits] [mlir] 9d1db3d - [mlir][sparse] generalize sparse_tensor.convert on static/dynamic dimension sizes   Aart Bik
• [Mlir-commits] [mlir] bd5494d - [mlir][sparse] make index type explicit in public API of support library   Aart Bik
• [Mlir-commits] [mlir] 1b15160 - [mlir][sparse] lower trivial tensor.cast on identical sparse tensors   Aart Bik
• [Mlir-commits] [mlir] c8d5dcb - [mlir][sparse] refactor loop sequence codegen   Aart Bik
• [Mlir-commits] [mlir] 1e6ef0c - [mlir][sparse] refine trait of sparse_tensor.convert   Aart Bik
• [Mlir-commits] [mlir] 947e14b - [mlir][sparse] move conversion test back to original CHECK testing   Aart Bik
• [Mlir-commits] [mlir] 185960d - [mlir][sparse] fix conversion bug when changing pointer/index sizes   Aart Bik
• [Mlir-commits] [mlir] 00040d7 - [mlir][sparse] fix broken asan test   Aart Bik
• [Mlir-commits] [mlir] 0121c96 - [mlir][sparse] refine the mixed width sparse conversion test   Aart Bik
• [Mlir-commits] [mlir] 08b63db - [MLIR][GPU] Add GPU launch op support for dynamic shared memory   Uday Bondhugula
• [Mlir-commits] [mlir] 0b83a35 - [MLIR][NFC] Drop unnecessary use of OpBuilder in build trip count map   Uday Bondhugula
• [Mlir-commits] [mlir] 1e39d32 - [MLIR] Add OrOp folding rule for constant one operand   Uday Bondhugula
• [Mlir-commits] [mlir] b2217b3 - [MLIR] Fix affine loop unroll corner case for full unroll   Uday Bondhugula
• [Mlir-commits] [mlir] 05fb260 - [MLIR] Fix assert crash when an unregistered dialect op is encountered   Uday Bondhugula
• [Mlir-commits] [mlir] 41a8b46 - [MLIR] Fix AffineExpr getLargestKnownDivisor for ceildiv and floordiv   Uday Bondhugula
• [Mlir-commits] [mlir] 57b9b29 - [MLIR][LLVM] Add llvm.mlir.global_ctors/dtors and translation support   Uday Bondhugula
• [Mlir-commits] [mlir] 10a80c4 - [mlir] Implement replacement of SymbolRefAttrs in Dialect attributes using SubElementAttr interface   Markus Böck
• [Mlir-commits] [mlir] 3077c13 - [mlir][NFC] Make InsertionGuard properly move constructible   Markus Böck
• [Mlir-commits] [mlir] a875e6e - [mlir][NFC] Fix typo in comment   Markus Böck
• [Mlir-commits] [mlir] 286a7a4 - [mlir][docs] Fix obvious spelling mistakes in documentation   Markus Böck
• [Mlir-commits] [mlir] eaf2588 - [mlir][Linalg] Add support for min/max reduction vectorization in linalg.generic   Diego Caballero
• [Mlir-commits] [mlir] 5c1d356 - [mlir][Linalg] Enable vectorization of explicit broadcasts   Diego Caballero
• [Mlir-commits] [mlir] eeb09fd - [mlir][Linalg] Enable vectorization of 'mul', 'and', 'or' and 'xor' reductions   Diego Caballero
• [Mlir-commits] [mlir] a4bccf7 - [mlir][doc] fix typos.   Jian Cai
• [Mlir-commits] [mlir] 7a79c6a - [mlir][OpenMP] OpenMP Synchronization Hints stored as IntegerAttr   Kiran Chandramohan
• [Mlir-commits] [mlir] 3b01cf9 - [mlir][openmp] Add an interface for Outlineable OpenMP ops   Valentin Clement
• [Mlir-commits] [mlir] 2fa2248 - [mlir] Placeholder used in predicate should be the base type   Chia-hung Duan
• [Mlir-commits] [mlir] 519663b - [MLIR] Add an option to disable `maxIterations` in greedy pattern rewrites   Frederik Gossen
• [Mlir-commits] [mlir] bf28849 - [mlir][linalg] Retire PoolingMaxOp/PoolingMinOp/PoolingSumOp.   Tobias Gysi
• [Mlir-commits] [mlir] 32a7d60 - [mli][linalg] Change tensor size in unit test (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] e826db6 - [mlir][linalg] Move generalization pattern to Transforms (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] a744c7e - [mlir][linalg] Update OpDSL to use the newly introduced min and max ops.   Tobias Gysi
• [Mlir-commits] [mlir] 3fe7fe4 - [mlir][linalg] Add unsigned min/max/cast function to OpDSL.   Tobias Gysi
• [Mlir-commits] [mlir] 1ebd197 - [mlir][linalg] Add generalization to CodegenStrategy.   Tobias Gysi
• [Mlir-commits] [mlir] 23800b0 - [mlir][linalg] Add loop interchange to CodegenStrategy.   Tobias Gysi
• [Mlir-commits] [mlir] 8ed2e8e - [mlir][linalg] Retire Linalg ConvOp.   Tobias Gysi
• [Mlir-commits] [mlir] 97d67f7 - [mlir][linalg] Add vector transfer lowering patters to codegen strategy.   Tobias Gysi
• [Mlir-commits] [mlir] c8faeb1 - [mlir][linalg] Add switch to disable/enable vector transfer lowering.   Tobias Gysi
• [Mlir-commits] [mlir] 63638ed - [mlir][linalg] Enable the vector transfer lowering by default.   Tobias Gysi
• [Mlir-commits] [mlir] eaa5275 - [mlir][linalg] Verify every LinalgOp has a body.   Tobias Gysi
• [Mlir-commits] [mlir] a8f69be - [mlir][linalg] Expose flag to control nofold attribute when padding.   Tobias Gysi
• [Mlir-commits] [mlir] 3f335ff - [mlir][linalg] Fix FusionOnTensors header and make local method static (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] 31a949f - [mlir][linalg] Remove unused method (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] 1eae247 - [mlir][linalg] Use OpBuilder in rewriteAsPaddedOp (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] e83d846 - [mlir][linalg] Adapt hoistPaddingOnTensors signature to support patterns (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] d0ec4a8 - [mlir][linalg] Add pad and hoist test pass.   Tobias Gysi
• [Mlir-commits] [mlir] 6638112 - [mlir][linalg] Add padding pass to strategy passes.   Tobias Gysi
• [Mlir-commits] [mlir] 93769e8 - [mlir] [test] Include mlir_tools_dir in PATH to fix mlir-reduce   Michał Górny
• [Mlir-commits] [mlir] 0f567f0 - [mlir] [test] Add missing tool substitutions   Michał Górny
• [Mlir-commits] [mlir] 80e3936 - [lldb, mlir] Migrate from getNumArgOperands and arg_operands (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 939a808 - Use llvm::is_contained (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] d8e4170 - Ensure newlines at the end of files (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 4ba9d9c - Use StringRef::contains (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] b33c211 - [TableGen] Use llvm::erase_value (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] db2b1e9 - [Utils] Fix a warning in DialectConversion.cpp   Kazu Hirata
• [Mlir-commits] [mlir] a88867a - [IR] Fix a warning   Kazu Hirata
• [Mlir-commits] [mlir] f67d57c - [mlir][Shape] Add a pattern to turn extract from shape_of into tensor.dim   Benjamin Kramer
• [Mlir-commits] [mlir] 898e809 - [mlir] Fix a crash when creating a 1d zero element LLVM constant   Benjamin Kramer
• [Mlir-commits] [mlir] d009f6e - [mlir] Convert ConstShapeOp to a static tensor type.   Adrian Kuegel
• [Mlir-commits] [mlir] 2bb208d - [mlir] Don't allow dynamic extent tensor types for ConstShapeOp.   Adrian Kuegel
• [Mlir-commits] [mlir] 376c138 - [mlir] Use empty() calls where possible.   Adrian Kuegel
• [Mlir-commits] [mlir] 9fb1086 - [mlir][python] Add a __contains__ method to the python bindings for DictionaryAttr.   Adrian Kuegel
• [Mlir-commits] [mlir] ecbee48 - [Builders.h] Silence a warning by adding a cast.   Chris Lattner
• [Mlir-commits] [mlir] 5627225 - Return failure on failure in convertBlockSignature.   Stella Laurenzo
• [Mlir-commits] [mlir] 5d6d30e - [mlir] Extend C and Python API to support bulk loading of DenseElementsAttr.   Stella Laurenzo
• [Mlir-commits] [mlir] c5f445d - [mlir][python] Temporarily disable test for converting unsupported DenseElementsAttr types to a buffer.   Stella Laurenzo
• [Mlir-commits] [mlir] a201829 - Fix parsing of hex-format index dense tensor attributes.   Stella Laurenzo
• [Mlir-commits] [mlir] 3d92722 - [mlir][python] Add a warning for cmake version < 3.15.   Stella Laurenzo
• [Mlir-commits] [mlir] fe6d993 - [mlir] Ability to build CAPI dylibs from out of tree projects against installed LLVM.   Stella Laurenzo
• [Mlir-commits] [mlir] 8e54f24 - Disable add_mlir_aggregate() debug file generation.   Stella Laurenzo
• [Mlir-commits] [mlir] a897590 - Add MLIR_INSTALL_AGGREGATE_OBJECTS and default it to ON.   Stella Laurenzo
• [Mlir-commits] [mlir] d86688f - [mlir][python] Segment MLIR Python test dialect to avoid testonly dependency.   Stella Laurenzo
• [Mlir-commits] [mlir] 288f1f8 - Fix MLIR LLVMIR test after 4723c9b3c6c46632a5d66e65d198899894b1e2c5   Roman Lebedev
• [Mlir-commits] [mlir] 4271269 - Revert "[IR] `IRBuilderBase::CreateAdd()`: short-circuit `x + 0` --> `x`"   Roman Lebedev
• [Mlir-commits] [mlir] 1e4cfe5 - [mlir][SPIRVToLLVM] Propagate location attribute from spv.GlobalVariable to llvm.mlir.global   Weiwei Li
• [Mlir-commits] [mlir] c0a6381 - [mlir][SPIRVToLLVM] Solve ExecutionModeOp redefinition and add OpTypeSampledImage into SPV_Type   Weiwei Li
• [Mlir-commits] [mlir] b983783 - [MLIR][linalg] Preserve location during elementwise fusion   Geoffrey Martin-Noble
• [Mlir-commits] [mlir] b096ac9 - [MLIR] Improve debug messages in BuiltinTypes   Geoffrey Martin-Noble
• [Mlir-commits] [mlir] efc6fe9 - [MLIR][TOSA] Drop "OnTensors" suffix   Geoffrey Martin-Noble
• [Mlir-commits] [mlir] 40b9c39 - [MLIR][LLVM] Add memset intrinsic   William S. Moses
• [Mlir-commits] [mlir] 0210a3d - [MLIR] Update DRR doc with returnType directive   Jeff Niu
• [Mlir-commits] [mlir] 3f00e10 - [mlir][OpenMP]Support for modifiers in workshare loops   Mats Petersson
• [Mlir-commits] [mlir] c3dcf39 - [mlir] Restrict to requiring traits when using InferTensorType trait.   Jacques Pienaar
• [Mlir-commits] [mlir] 04d76d3 - [mlir][python] Add nameloc getter   Jacques Pienaar
• [Mlir-commits] [mlir] 65c9907 - [mlir][ods] Enable emitting getter/setter prefix   Jacques Pienaar
• [Mlir-commits] [mlir] 965ec6d - [mlir] Add folder for shape.add   Jacques Pienaar
• [Mlir-commits] [mlir] 62bf850 - [mlir] Flipping Test dialect to prefixed form _Both   Jacques Pienaar
• [Mlir-commits] [mlir] 6a99423 - [mlir] Expand prefixing to OpFormatGen   Jacques Pienaar
• [Mlir-commits] [mlir] 42e9af9 - [mlir] Rename to avoid overlap in accessor prefixing   Jacques Pienaar
• [Mlir-commits] [mlir] cfb72fd - [mlir] Switch arith, llvm, std & shape dialects to accessors prefixed both form.   Jacques Pienaar
• [Mlir-commits] [mlir] ac14b83 - [mlir] Give GenericAtomicRMW region a name   Jacques Pienaar
• [Mlir-commits] [mlir] b288d08 - [mlir-c] Avoid compiler warning   Jacques Pienaar
• [Mlir-commits] [mlir] 332ce23 - [mlir][ods] Fix incorrect accessing of segment_sizes   Jacques Pienaar
• [Mlir-commits] [mlir] 0ef217d - [mlir] Fix missing prefix for region accessor on OpAdaptor   Jacques Pienaar
• [Mlir-commits] [mlir] dde9636 - [mlir] Flip accessors to prefixed form (NFC)   Jacques Pienaar
• [Mlir-commits] [mlir] 2fa76d4 - [mlir][ods] Fix incorrectly generated attribute name.   Jacques Pienaar
• [Mlir-commits] [mlir] e244a6f - [mlir] Replace report_fatal_error(std::string) uses with report_fatal_error(Twine)   Simon Pilgrim
• [Mlir-commits] [mlir] cfb53d8 - [NFC] Make some includes explicit   Nikita Popov
• [Mlir-commits] [mlir] fd96133 - [MLIR] Rename Shape dialect's `join` to `meet`.   Alexandre Rames
• [Mlir-commits] [mlir] 0a06068 - [MLIR] Remove the unused `ArithmeticCastOp`.   Alexandre Rames
• [Mlir-commits] [mlir] 782a97a - [mlir][capi] Add TypeID to MLIR C-API   Daniel Resnick
• [Mlir-commits] [mlir] 1760d8b - [mlir][ODS] Support result type inference in custom assembly format   Daniel Resnick
• [Mlir-commits] [mlir] d023298 - [MLIR] Fix unused tablegen template arg warnings   Cullen Rhodes
• [Mlir-commits] [mlir] b8ffcb1 - [mlir:Pass] Generate a reproducer as early as possible   River Riddle
• [Mlir-commits] [mlir] a77cd55 - [mlir] Add support for specifying printing flags when adding an op to a Diagnostic   River Riddle
• [Mlir-commits] [mlir] 20211b3 - [mlir] Fix tsan failure in PassCrashRecovery   River Riddle
• [Mlir-commits] [mlir] d781361 - [mlir] Add AnyAttrOf tablegen attribute constraint   River Riddle
• [Mlir-commits] [mlir] 6d6ad76 - [mlir][NFC] Provide accessor for TableGen record for constraints   River Riddle
• [Mlir-commits] [mlir] b7144ab - [NFC] Clean up a few methods within GreedyPatternRewriter   River Riddle
• [Mlir-commits] [mlir] 5652ecc - [mlir:GreedyPatternRewriter] Add debug logging for pattern rewriter actions   River Riddle
• [Mlir-commits] [mlir] 01b55f1 - [NFC] Tidy up DialectConversion.cpp   River Riddle
• [Mlir-commits] [mlir] 015192c - [mlir:DialectConversion] Restructure how argument/target materializations get invoked   River Riddle
• [Mlir-commits] [mlir] 2321ac1 - [mlir] Remove the deprecated OpConversionPattern::matchAndRewrite methods   River Riddle
• [Mlir-commits] [mlir] f1b9221 - [MLIR][Math] Add erf to math dialect   Sean Silva
• [Mlir-commits] [mlir] a36aca5 - [mlir] Avoid including <alloca.h> on OpenBSD   Brad Smith
• [Mlir-commits] [mlir] 6b1f653 - [mlir][linalg][bufferize] tensor.cast may require a copy   Matthias Springer
• [Mlir-commits] [mlir] 1d24b8c - [mlir][linalg][bufferize][NFC] Change bufferizableInPlaceAnalysis signature   Matthias Springer
• [Mlir-commits] [mlir] 56bf688 - [mlir][linalg][bufferize][NFC] Simplify getAliasingOpResult()   Matthias Springer
• [Mlir-commits] [mlir] a046154 - [mlir][linalg][bufferize] Add bufferRelation to op interface   Matthias Springer
• [Mlir-commits] [mlir] 89b2f29 - [mlir][linalg][bufferize] Fix/add missing case to getAliasingOpOperand   Matthias Springer
• [Mlir-commits] [mlir] 7dfd3bb - [mlir][linalg][bufferize][NFC] API change of aliasesNonWritableBuffer   Matthias Springer
• [Mlir-commits] [mlir] f8453ea - [mlir][linalg][bufferize] Rewrite "write into non-writable memory" detection   Matthias Springer
• [Mlir-commits] [mlir] a6ddb7c - [mlir][linalg][bufferize][NFC] Reduce code duplication around bufferizableInPlaceAnalysis   Matthias Springer
• [Mlir-commits] [mlir] fd31620 - [mlir][linalg][bufferize][NFC] Conflict detection funcs take OpOperand and OpResult   Matthias Springer
• [Mlir-commits] [mlir] 7c3a810 - [mlir][linalg][bufferize] Put buffer copying in separate function   Matthias Springer
• [Mlir-commits] [mlir] 0292236 - [mlir][linalg][bufferize][NFC] Simplify allocateBuffersForResults   Matthias Springer
• [Mlir-commits] [mlir] 0ee5323 - [mlir][linalg][bufferize][NFC] Simplify getAliasingOpOperand signature   Matthias Springer
• [Mlir-commits] [mlir] 441485c - [mlir][linalg][bufferize][NFC] Remove InSpaceSpec from bufferizesToMemoryWrite   Matthias Springer
• [Mlir-commits] [mlir] c161084 - [mlir] Make OpResult usable with DenseSet etc.   Matthias Springer
• [Mlir-commits] [mlir] d3cb6bf - [mlir][linalg][bufferize] Rewrite conflict detection   Matthias Springer
• [Mlir-commits] [mlir] 7dd7078 - [mlir][linalg][bufferize] Handle scf::ForOp correctly in bufferizesToMemoryRead   Matthias Springer
• [Mlir-commits] [mlir] e7bb8dd - [mlir][linalg][bufferize] Relax rules for extract_slice/insert_slice matching   Matthias Springer
• [Mlir-commits] [mlir] 252386a - [mlir] Add enclosingOpOk parameter to properlyDominates   Matthias Springer
• [Mlir-commits] [mlir] fd26ca4 - [mlir][scf] Add insideMutuallyExclusiveBranches helper   Matthias Springer
• [Mlir-commits] [mlir] 9c55e71 - [mlir][linalg][bufferize] Bufferize using PostOrder traversal   Matthias Springer
• [Mlir-commits] [mlir] 65ef43e - [mlir][linalg][bufferize][NFC] Check return value of getResultBuffer   Matthias Springer
• [Mlir-commits] [mlir] 36ec848 - [mlir][linalg][bufferize][NFC] Change findValueInReverseUseDefChain signature   Matthias Springer
• [Mlir-commits] [mlir] c5501a7 - [mlir][linalg][bufferize] Eliminate InitTensorOps of InsertSliceOp sources   Matthias Springer
• [Mlir-commits] [mlir] 7a7e93f - [mlir][linalg][bufferize] Avoid creating copies that are never read   Matthias Springer
• [Mlir-commits] [mlir] 94213bc - [mlir][linalg][bufferize] Fix bug in getInplaceableOpResult   Matthias Springer
• [Mlir-commits] [mlir] 5f8228d - [mlir][linalg][bufferize] Fix bufferizesToMemoryWrite for TiledLoopOp   Matthias Springer
• [Mlir-commits] [mlir] 3bbc869 - [mlir][linalg][bufferize] Support scf::IfOp   Matthias Springer
• [Mlir-commits] [mlir] 7bce6bb - [mlir][linalg][bufferize] Fix crash when bufferizing CallOpInterface   Matthias Springer
• [Mlir-commits] [mlir] 5b98e4e - [mlir][linalg][bufferize] Add analysis fuzzer option   Matthias Springer
• [Mlir-commits] [mlir] 4e14bac - [mlir][linalg][bufferize] tensor::CastOp is an alias-only op   Matthias Springer
• [Mlir-commits] [mlir] 81e8c39 - [mlir][linalg][bufferize][NFC] Add bufferizesToAliasOnly   Matthias Springer
• [Mlir-commits] [mlir] d5a4c86 - [mlir][tosa] tosa.cast support for unsigned integers   Rob Suderman
• [Mlir-commits] [mlir] 4c48f7e - [mlir][tosa] Create basic dynamic shape support for several ops.   Rob Suderman
• [Mlir-commits] [mlir] 95e4b71 - [mlir][tosa] Fix tosa average_pool2d to linalg type issue   Rob Suderman
• [Mlir-commits] [mlir] 59dd418 - [mlir][tosa] Fix tosa.cast UiToFp32 for tosa-to-linalg   Rob Suderman
• [Mlir-commits] [mlir] 4ada6c2 - [mlir][tosa] Adds a canonicalization to the transpose op if the perms are a no op   Rob Suderman
• [Mlir-commits] [mlir] 9c62bb5 - Implementation of `ReshapeNoopOptimization` canonicalizer.   Rob Suderman
• [Mlir-commits] [mlir] 58901a5 - [mlir][tosa] Correct tosa.avg_pool2d for specification error   Rob Suderman
• [Mlir-commits] [mlir] a35f54c - [tosa][mlir] Add bailout to TosaMakeBroadcastable for unranked case   Rob Suderman
• [Mlir-commits] [mlir] b0c4aaf - Allow only valid vector.shape_cast transitive folding   Ahmed Taei
• [Mlir-commits] [mlir] 21f9e4a - Avoid infinity arithmetics when computing exp approximations   Ahmed Taei
• [Mlir-commits] [mlir] 813fa79 - Don't drop in_bounds when vector-transfer-collapse-inner-most-dims   Ahmed Taei
• [Mlir-commits] [mlir] a3dd4e7 - Drop transfer_read inner most unit dimensions   Ahmed S. Taei
• [Mlir-commits] [mlir] 7aebdfc - [mlir] Limit Python dependency to Development.Module when possible.   Mike Urbach
• [Mlir-commits] [mlir] 55e76c7 - [mlir] Limit Python dependency to Development.Module when possible.   Mike Urbach
• [Mlir-commits] [mlir] d576f45 - [MLIR][OpenMP] Added parseClauses   Shraiysh Vaishay
• [Mlir-commits] [mlir] 10e0878 - [MLIR][OpenMP][NFC] Moved Synchronization Hint related functions   Shraiysh Vaishay
• [Mlir-commits] [mlir] c4c7e06 - [MLIR][OpenMP] Shifted hint from CriticalOp to CriticalDeclareOp   Shraiysh Vaishay
• [Mlir-commits] [mlir] a81672b - [NFC][MLIR][OpenMP] Splitting the WsLoop tests.   Shraiysh Vaishay
• [Mlir-commits] [mlir] 9fb52cb - [MLIR][OpenMP] Added omp.atomic.read and omp.atomic.write   Shraiysh Vaishay
• [Mlir-commits] [mlir] 30bd11f - [MLIR][OpenMP] Fixed the missing inclusive clause in omp.wsloop and fix order clause   Shraiysh Vaishay
• [Mlir-commits] [mlir] fab634b - [mlir] Tighten strided layout specification.   Nicolas Vasilache
• [Mlir-commits] [mlir] 8096759 - [mlir][Linalg] NFC - Add support to specify that a tensor value is known to bufferize to writeable memory   Nicolas Vasilache
• [Mlir-commits] [mlir] af9dce1 - [mlir][Linalg] Allow operand-less scf::ExecuteRegionOp to encapsulate scf::YieldOp   Nicolas Vasilache
• [Mlir-commits] [mlir] 26b3e92 - [mlir][Linalg] Don't return early from inPlaceAnalysis   Nicolas Vasilache
• [Mlir-commits] [mlir] 31270eb - [mlir][Vector] Let vector.multi_reduction reduce down to a scalar.   Nicolas Vasilache
• [Mlir-commits] [mlir] 8f1650c - [mlir][Linalg] NFC - Refactor vector.broadcast op verification logic and make it available as a precondition in Linalg vectorization.   Nicolas Vasilache
• [Mlir-commits] [mlir] 67b1053 - [mlir][Vector] Allow a 0-d for for vector transfer ops.   Nicolas Vasilache
• [Mlir-commits] [mlir] 47f7938 - [mlir][Vector] Add support for lowering 0-d transfers to load/store.   Nicolas Vasilache
• [Mlir-commits] [mlir] 0c74b12 - [mlir][Vector] NFC - Add test to exercise lowering of vector.transfer to scf   Nicolas Vasilache
• [Mlir-commits] [mlir] 753a67b - [mlir][Linalg] Refactor and improve vectorization to add support for reduction into 0-d tensors.   Nicolas Vasilache
• [Mlir-commits] [mlir] 0a7f81a - mlir][Vector] Fix spuriously disabled test.   Nicolas Vasilache
• [Mlir-commits] [mlir] b24c91f - [mlir][Vector][Bigfix] Fix vector transfer to store lowering to insert a proper ExtractOp   Nicolas Vasilache
• [Mlir-commits] [mlir] 012c0cc - [mlir] NFC - Avoid unused symbol in opt mode.   Nicolas Vasilache
• [Mlir-commits] [mlir] 0eeaad3 - [mlir][Linalg] Fix insertion point in comprehensive bufferization   Nicolas Vasilache
• [Mlir-commits] [mlir] 82dd977 - [mlir][Linalg] Tighten canonicalization of InsertSliceOp that triggers infinite loop   Nicolas Vasilache
• [Mlir-commits] [mlir] 6080271 - [mlir][vector] NFC - Refactor and extract a helper StructuredGenerator class   Nicolas Vasilache
• [Mlir-commits] [mlir] 6bb7d24 - [mlir][Linalg] Add a first vectorization pattern for conv1d in NWCxWCF format.   Nicolas Vasilache
• [Mlir-commits] [mlir] 203accf - [mlir][Linalg] Improve conv vectorization for the stride==1 case.   Nicolas Vasilache
• [Mlir-commits] [mlir] eda2ebd - [mlir][Vector] NFC - Extract rewrites related to insert/extract strided slice in a separate file.   Nicolas Vasilache
• [Mlir-commits] [mlir] 489fec2 - [mlir][Linalg] NFC - Drop Optional in favor of FailureOr   Nicolas Vasilache
• [Mlir-commits] [mlir] 89d55d3 - [mlir][Linalg] Retire CodegenStrategy::transform   Nicolas Vasilache
• [Mlir-commits] [mlir] 4703a07 - [mlir][Linalg] NFC - Reorganize options nesting.   Nicolas Vasilache
• [Mlir-commits] [mlir] e03b443 - Revert "[mlir][Linalg] NFC - Reorganize options nesting."   Nicolas Vasilache
• [Mlir-commits] [mlir] 1b702ee - [mlir][Linalg] NFC - Reorganize options nesting.   Nicolas Vasilache
• [Mlir-commits] [mlir] 176a0ea - [mlr][Linalg] NFC - Add option to hook vector.multi_reduction lowering to strategies.   Nicolas Vasilache
• [Mlir-commits] [mlir] d054b80 - [mlir][Vector] NFC - Add option to hook vector.transpose lowering to strategies.   Nicolas Vasilache
• [Mlir-commits] [mlir] 00ac874 - [mlir][Vector] Add InsertStridedSliceOp -> ShuffleOp for the rank-1 cases.   Nicolas Vasilache
• [Mlir-commits] [mlir] 8e4c806 - [mlir][Linalg] NFC - Add additional control to lower vector.shape_cast ops   Nicolas Vasilache
• [Mlir-commits] [mlir] cd392c0 - [mlir][Linalg] NFC - Make more option names consistent.   Nicolas Vasilache
• [Mlir-commits] [mlir] 505afd1 - [mlir] Clean up boolean flags usage in LIT tests   Vladislav Vinogradov
• [Mlir-commits] [mlir] c6390f1 - [mlir] Fix AsmPrinter for types with sub elements   Vladislav Vinogradov
• [Mlir-commits] [mlir] d6296c3 - [mlir] Allow to use vformat utility with MLIR classes   Vladislav Vinogradov
• [Mlir-commits] [mlir] e41ebbe - [mlir][RFC] Refactor layout representation in MemRefType   Vladislav Vinogradov
• [Mlir-commits] [mlir] 17664a2 - [mlir][NFC] Fix leak in unit test   Vladislav Vinogradov
• [Mlir-commits] [mlir] 8447067 - [MLIR] Add `KeywordOrString` handling to AsmParser   Andrew Young
• [Mlir-commits] [mlir] 44b22f6 - [MLIR] Expose optional attribute parsing functions   Andrew Young
• [Mlir-commits] [mlir] a3f4259 - [mlir][linalg] Include InitTensorOp in tiling canonicalization   Lei Zhang
• [Mlir-commits] [mlir] 070b0af - [mlir][spirv] Fix path in define_enum.sh script   Lei Zhang
• [Mlir-commits] [mlir] 83e074a - [mlir] Add an 'cppNamespace' field to availability   Lei Zhang
• [Mlir-commits] [mlir] 7a89444 - [mlir][spirv] Add ops and patterns for lowering standard max/min ops   Lei Zhang
• [Mlir-commits] [mlir] 3964c1d - [mlir][vector] Split populateVectorContractLoweringPatterns   Lei Zhang
• [Mlir-commits] [mlir] 4cd7ff6 - [mlir][linalg] Constant fold linalg.generic that are transposes   Lei Zhang
• [Mlir-commits] [mlir] bdd37c9 - [mlir][tensor] Add some folders for insert/extract slice ops   Lei Zhang
• [Mlir-commits] [mlir] 519b350 - [mlir][vector] Add folder for no-op InsertStridedSliceOp   Lei Zhang
• [Mlir-commits] [mlir] c788cad - [mlir][linalg] Fix FoldConstantTranspose execution inefficiency   Lei Zhang
• [Mlir-commits] [mlir] 5d001f5 - [mlir] Fix a bug in Affine LICM.   Amy Zhuang
• [Mlir-commits] [mlir] 5ce368c - [mlir] Vectorize induction variables   Amy Zhuang
• [Mlir-commits] [mlir] b9ae741 - [mlir] Fix getVectorReductionOp   Amy Zhuang
• [Mlir-commits] [mlir] 8276ac1 - [mlir] Add alignment attribute to memref.global   Eugene Zhulenev
• [Mlir-commits] [mlir] e2a37bb - [mlir] Add alignment option to constant tensor bufferization pass   Eugene Zhulenev
• [Mlir-commits] [mlir] bf32bb7 - [mlir] Update approximation range for Tanh operation   Eugene Zhulenev
• [Mlir-commits] [mlir] f97f946 - Canonicalize max/min operations on integers.   Eugene Zhulenev
• [Mlir-commits] [mlir] 0d9b478 - [mlir] Reduce the number of iterations in async microbenchmarks   Eugene Zhulenev
• [Mlir-commits] [mlir] 627fa0b - [mlir] MathApproximations: unroll virtual vectors into hardware vectors for ISA specific operation   Eugene Zhulenev
• [Mlir-commits] [mlir] ec32d54 - [mlir] MathApproximations: scalars shape must be 0-rank   Eugene Zhulenev
• [Mlir-commits] [mlir] ed9e52f - [mlir][python] Usability improvements for Python bindings   Alex Zinenko
• [Mlir-commits] [mlir] 3a3a09f - [mlir][python] Provide more convenient wrappers for std.ConstantOp   Alex Zinenko
• [Mlir-commits] [mlir] 255a690 - [mlir][python] Provide more convenient constructors for std.CallOp   Alex Zinenko
• [Mlir-commits] [mlir] 01d696e - [mlir] rename the "packing" flag of linalg.pad_tensor to "nofold"   Alex Zinenko
• [Mlir-commits] [mlir] b164f23 - [mlir][python] support taking ops instead of values in op constructors   Alex Zinenko
• [Mlir-commits] [mlir] bacb0ca - [mlir] add user-level documentation for Python bindings   Alex Zinenko
• [Mlir-commits] [mlir] e67cbbe - [mlir][python] Expose CallSiteLoc Python side   Alex Zinenko
• [Mlir-commits] [mlir] 78f2dae - [mlir][python] Provide some methods and properties for API completeness   Alex Zinenko
• [Mlir-commits] [mlir] 90a6c3c - [mlir] Fix typos in the Python bindings doc   Alex Zinenko
• [Mlir-commits] [mlir] 7fd6f40 - [mlir][python] Add custom constructor for memref load   Alex Zinenko
• [Mlir-commits] [mlir] 2b55e14 - [mlir] fix python bindings cmake   Alex Zinenko
• [Mlir-commits] [mlir] a04c0b7 - [mlir][python] Fix MemRefType IsAFunction in Python bindings   Alex Zinenko
• [Mlir-commits] [mlir] 18fbd5f - [mlir][python] Better support for variadic regions in Python bindings   Alex Zinenko
• [Mlir-commits] [mlir] d5429a1 - [mlir][python] Add 'loc' property to ops   Alex Zinenko
• [Mlir-commits] [mlir] 310736e - [mlir] fix region property generation in python bindings   Alex Zinenko
• [Mlir-commits] [mlir] 14c9207 - [mlir] support interfaces in Python bindings   Alex Zinenko
• [Mlir-commits] [mlir] 24685aa - [mlir][python] allow for detaching operations from a block   Alex Zinenko
• [Mlir-commits] [mlir] fb3d83a - [mlir] Fix redundant return in the void method.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 8f0c673 - [MLIR] fix arith dialect build failure   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 4281946 - [mlir][Tensor] Add ReifyRankedShapedTypeOpInterface to tensor.extract_slice.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] c6828e0 - [mlir] Make ConversionTarget dynamic legality callbacks composable   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 7c97e32 - [mlir][linalg] Fix generic reduction vectorization   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] aa71f48 - [mlir] update new linalg vectorization tests after vectorization fix   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] cc83c24 - [mlir][vector] Add canonicalization extract + splat   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 8c1553f - [mlir][spirv] Add memory semantics verify for atomic operations   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] afad0cd - [mlir][vector] Refactor linalg vectorization for reductions   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] cb3aa49 - [MLIR][arith] fix references to std.constant in comments   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 44610c0 - [MLIR][ODS] default-valued strings should be in quotes   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 52d6c5d - [MLIR] Generalize Affine dependence analysis using Affine Relations   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 4a5ff56 - [mlir][docs] Fix name of get arith->LLVM patterns in docs   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 2ea5e7b - [mlir] SPIR-V: add sin, cos, log, sqrt OCL ops   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 21bb463 - [mlir] fix bugs with NamedAttrList   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] dc2be87 - [MLIR][OpenMP] Add support for ordered construct   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] b37e518 - [MLIR][OpenMP] Add support for ordered construct   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 1d8cc45 - [mlir][vector] Add patterns to convert multidimreduce to vector.contract   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 93d0ade - [mlir][linalg] Remove special case for contraction vectorization   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 5c0369e - Fix escaping in RewriterGen.cpp.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 95935e8 - Make genAttributeVerifier escape the summary.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 516884f - [MLIR] Fix FloorDivSIOpConverter that was failing for index type after the arithmetic op refactor   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] f5f5926 - [MLIR] FlatAffineValueConstraints: Fix bug in mergeSymbolIds   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] c86f218 - [mlir][Linalg] Allow comprehensive bufferization to use callbacks for alloc/dealloc.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 5fb46a9 - Revert "[mlir][Linalg] Allow comprehensive bufferization to use callbacks for alloc/dealloc."   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 16e530d - When generating C++ code, use C++ string escaping.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 20bd6fb - [mlir] gen_spirv_dialect.py: Some support for OCL ops generation   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] eacd6e1 - [mlir][GPUtoNVVM] Relax restriction on wmma op lowering   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 3be3c94 - [MLIR] FlatAffineConstraints: Ensure dimensionalities match when calling mergeLocalIds   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 6e2c0e6 - [mlir][spirv] Add conversions from arith.bitcast, std.br, std.cond_br to spirv.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] e858275 - [mlir][arith] `traits` need to be passed to parent op   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] ca01034 - [mlir][sparse] Factoring out getZero() and avoiding unnecessary Type params   wren romano
• [Mlir-commits] [mlir] 14fffda - [mlir][sparse] Factoring out allocaIndices()   wren romano
• [Mlir-commits] [mlir] af7ac1d - [mlir][sparse] Sharing calls to adaptor.getOperands()[0]   wren romano
• [Mlir-commits] [mlir] 63d4fc9 - [mlir][sparse] Factoring out helper functions for generating constants   wren romano
• [Mlir-commits] [mlir] 5167c36 - [mlir][sparse] Misc code cleanup   wren romano
• [Mlir-commits] [mlir] bd0cae6 - [mlir][sparse] Renaming variables for consistency/clarity   wren romano
• [Mlir-commits] [mlir] 28882b6 - [mlir][sparse] Implementing sparse=>dense conversion.   wren romano
• [Mlir-commits] [mlir] 5389cdc - [mlir][sparse] Adding dynamic-size support for sparse=>dense conversion   wren romano
• [Mlir-commits] [mlir] 30a64c9 - [mlir][sparse] Renaming CPP macros for clarity   wren romano
• [Mlir-commits] [mlir] 6be36fd - [mlir][sparse] Improve handling of dynamic-sizes for sparse=>dense conversion   wren romano

Last message date: Sun Oct 31 22:12:51 PDT 2021
Archived on: Sun Oct 31 22:12:54 PDT 2021

• Messages sorted by: [ thread ] [ subject ] [ date ]
• More info on this list...