February 2023 Archives by author

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

Starting: Wed Feb 1 00:11:34 PST 2023
Ending: Tue Feb 28 17:22:45 PST 2023
Messages: 333

• [Mlir-commits] [mlir] e8231ea - Simplify looping over Operation users in SliceAnalysis (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 0096d17 - Add `set(CMAKE_CXX_STANDARD 17)` to MLIR CMakeLists.txt   Mehdi Amini
• [Mlir-commits] [mlir] 2c9783c - Remove empty MLIRMemRefUtils library (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 72429a4 - Add a pass that builds a debug info scope for LLVMFuncOp (adding a DISubprogramAttr)   Mehdi Amini
• [Mlir-commits] [mlir] ab2b73d - Fix internal link in the MLIR doc   Mehdi Amini
• [Mlir-commits] [mlir] a85ea9a - [MLIR][doc] Clarify `allowUnregisteredDialects()` as "testing option" (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 5cf549e - Add info about the "testing only" aspect of `--allow-unregistered-dialect` in the cl::opt desc (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 8d2e0c1 - Refactor a MlirOptMainConfig class to hold the configuration of MlirOptMain (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] 01cfdf1 - [mlir-opt] Rename internally the variables backing --no-implicit-module to ExplicitModule (NFC)   Mehdi Amini
• [Mlir-commits] [mlir] f35ac8a - Fix SimplifyAllocConst pattern when we have alloc of negative sizes   Mehdi Amini
• [Mlir-commits] [mlir] 9cf9685 - [mlir][bufferize] Add noAnalysisFuncFilter to OneShotBufferizationOptions struct   Maya Amrami
• [Mlir-commits] [mlir] ace6072 - [mlir] PromoteBuffersToStackPass - Copy attributes of original AllocOp   Maya Amrami
• [Mlir-commits] [mlir] 4511128 - [mlir][bufferization] Consider defaultMemorySpace when bufferizing FuncOp   Maya Amrami
• [Mlir-commits] [mlir] 782045e - Revert "HHVM calling conventions."   Amir Ayupov
• [Mlir-commits] [mlir] af694df - [MLIR][OpenMP] Add conversion support from FIR to LLVM Dialect for OMP Target Data with region   Akash Banerjee
• [Mlir-commits] [mlir] 310deca - [mlir] Add loop bounds to scf.foreach_thread.   Alexander Belyaev
• [Mlir-commits] [mlir] eb2f946 - [mlir][scf] Rename ForeachThreadOp->ForallOp, PerformConcurrentlyOp->InParallelOp.   Alexander Belyaev
• [Mlir-commits] [mlir] cdf8f06 - [mlir] Add a folder for lbs, ubs, steps of scf.forall.   Alexander Belyaev
• [Mlir-commits] [mlir] 8a9d489 - [mlir] Clean-up math -> libm/llvm conversion.   Alexander Belyaev
• [Mlir-commits] [mlir] 25dbf8f - [mlir] Use foldDynamicIndexList instead of canonicalizeSubViewPart.   Alexander Belyaev
• [Mlir-commits] [mlir] 3bf1f0e - [mlir] Add missing patterns for lowering to Libm.   Alexander Belyaev
• [Mlir-commits] [mlir] a5cdcf4 - [mlir] Fix folding for scf.for(tensor.cast).   Alexander Belyaev
• [Mlir-commits] [mlir] 9fa61cb - [mlir] Insert tensor.cast only when needed when folding tensor.cast into extract_slice.   Alexander Belyaev
• [Mlir-commits] [mlir] 28b9277 - [mlir][sparse] add some documentation to storage layout (NFC)   Aart Bik
• [Mlir-commits] [mlir] e2e6e7a - [mlir][sparse] start using size_hint provided in allocation op   Aart Bik
• [Mlir-commits] [mlir] 3bd82f3 - [mlir][sparse] compute allocation size_hint   Aart Bik
• [Mlir-commits] [mlir] d453d73 - [mlir][vector] add proper verification to vector.print operation   Aart Bik
• [Mlir-commits] [mlir] e379b4e - [mlir][memref] annotate operand and result of realloc with proper memory attributes   Aart Bik
• [Mlir-commits] [mlir] c55e6af - [mlir][sparse] fix integration test with "VLA" run   Aart Bik
• [Mlir-commits] [mlir] 9db037d - [mlir][sparse] fixed pack op documentation and purity   Aart Bik
• [Mlir-commits] [mlir] 5561e17 - Revert "[mlir] Make the vast majority of integration and runner tests work on Windows"   Aart Bik
• [Mlir-commits] [mlir] 978c744 - [mlir][NFC] Clean and Fix broken Affine documentation   Uday Bondhugula
• [Mlir-commits] [mlir] e6edc1b - [MLIR] Fix non-deterministic generation from buffer-deallocation pass   Uday Bondhugula
• [Mlir-commits] [mlir] 687f782 - [MLIR] Fold away divs and mods in affine ops with operand info   Uday Bondhugula
• [Mlir-commits] [mlir] 8a583dd - [MLIR] Add replaceUsesWithIf on Operation   Uday Bondhugula
• [Mlir-commits] [mlir] 332b8cf - [MLIR][Affine] Fix affine.parallel op domain add   Uday Bondhugula
• [Mlir-commits] [mlir] a2802dd - [MLIR] Fix bug in addAffineParallelOpDomain upper bound constraint   Uday Bondhugula
• [Mlir-commits] [mlir] 83e9ef7 - [mlir][standalone] Extend sample with custom type   Marius Brehler
• [Mlir-commits] [mlir] eafca23 - [mlir][MemRef] Add required address space cast when lowering alloc to LLVM   Markus Böck
• [Mlir-commits] [mlir] ad48a0e - [mlir] Switch default Fold API to using FoldAdaptors   Markus Böck
• [Mlir-commits] [mlir] 0f827ee - [mlir][ods] add mechanism for deprecating an `OpBuilder` in ODS   Markus Böck
• [Mlir-commits] [mlir] 9b15fae - [mlir][doc] Remove outdated section about a deprecated `OpBuilder` use   Markus Böck
• [Mlir-commits] [mlir] 50ea17b - [mlir][MemRef] Add option to `-finalize-memref-to-llvm` to emit opaque pointers   Markus Böck
• [Mlir-commits] [mlir] cbd7aaa - [mlir][cf] Add support for opaque pointers to ControlFlowToLLVM lowering   Markus Böck
• [Mlir-commits] [mlir] a85bd1b - [mlir][FuncToLLVM] Add option for emitting opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 2ca4642 - [mlir][Async] Add option to LLVM lowering to use opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 81767f5 - [mlir][OpenMP] Add support for using Opaque Pointers in the OpenMP Dialect   Markus Böck
• [Mlir-commits] [mlir] cd4ca2d - [mlir] Port Conversion Passes to LLVM to use TableGen generated constructors and options   Markus Böck
• [Mlir-commits] [mlir] 4a2d458 - [mlir][Vector] add pass option to `convert-vector-to-llvm` to emit opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 7c4e45e - [mlir][SCFToOpenMP] Add pass option to emit LLVM opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 0aaf2e3 - [mlir][GPU] add required address space cast when lowering to LLVM   Markus Böck
• [Mlir-commits] [mlir] 009fc4c - [mlir][LLVM] Verify correct pointer casts with `llvm.bitcast`   Markus Böck
• [Mlir-commits] [mlir] 19b1e27 - [mlir][SPIRVToLLVM] Add pass option to emit opaque-pointers   Markus Böck
• [Mlir-commits] [mlir] 161b9d7 - [mlir] Make the vast majority of integration and runner tests work on Windows   Markus Böck
• [Mlir-commits] [mlir] 147f476 - [mlir][LLVM] Properly wrap code examples in markdown code blocks   Markus Böck
• [Mlir-commits] [mlir] 9048ea2 - Reland "[mlir] Make the vast majority of intgration and runner tests work on Windows"   Markus Böck
• [Mlir-commits] [mlir] b068c87 - [mlir][OpenACCToLLVM] Add pass option to emit opaque pointers   Markus Böck
• [Mlir-commits] [mlir] afcceec - [mlir][LLVM] Don't make `LLVM_IntPtrBase` a `BuildableType` to allow the use of opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 0e5aeae - [mlir][GPUToLLVM] Add support for emitting opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 475bbea - [mlir] Complety remove old `fold` API   Markus Böck
• [Mlir-commits] [mlir] 225f11c - [mlir] Partially revert removal of old `fold` method   Markus Böck
• [Mlir-commits] [mlir] 5934394 - [mlir][doc] Remove section about now removed `useFoldAPI` option   Markus Böck
• [Mlir-commits] [mlir] b38cf03 - [mlir][LinalgToLLVM] Add pass option for emitting opaque-pointers   Markus Böck
• [Mlir-commits] [mlir] 83b5979 - [mlir][LLVM] Enable opaque-pointers in e2e lowering test pass   Markus Böck
• [Mlir-commits] [mlir] f138840 - [mlir][Vector][Trivial] Move a conversion to LLVM test to the conversion passes test directory and port it to opaque pointers   Markus Böck
• [Mlir-commits] [mlir] 91be358 - [mlir][GPUToVulkan] Port conversion passes and `mlir-vulkan-runner` to opaque pointers   Markus Böck
• [Mlir-commits] [mlir] dff04fc - [mlir][SPIRV] Enable use of opaque-pointers in spirv-cpu-runner   Markus Böck
• [Mlir-commits] [mlir] 53689fd - [mlir][NVGPUToNVVM] Add option for emitting opaque pointers   Markus Böck
• [Mlir-commits] [mlir] efae6c9 - Revert "[mlir][linalg] Fix crash in vectorizer when expanding affine apply"   Diego Caballero
• [Mlir-commits] [mlir] f453589 - Revert "[mlir][linalg] Make Linalg vectorizer lower affine.apply"   Diego Caballero
• [Mlir-commits] [mlir] b1d8205 - [mlir][Vector] Add lowering support for 1-D masked multi-reductions   Diego Caballero
• [Mlir-commits] [mlir] d0613c1 - [mlir][LLVMIR] Add lowering support for vp.muladd intrinsic   Diego Caballero
• [Mlir-commits] [mlir] e9b82a5 - [mlir][Vector] Add LLVM lowering for masked reductions   Diego Caballero
• [Mlir-commits] [mlir] 9452356 - [mlir][Vector] Add support for masked vector.contract   Diego Caballero
• [Mlir-commits] [mlir] 1ac874c - [mlir][Vector] Add support for masked vector gather ops   Diego Caballero
• [Mlir-commits] [mlir] 1427277 - [mlir][Vector] Enable masking for static shapes   Diego Caballero
• [Mlir-commits] [mlir] c339f9e - [mlir][Vector] Support masking for more contraction flavors   Diego Caballero
• [Mlir-commits] [mlir] 51f235c - [mlir][Vector] Add folding for masked reductions and vector.mask   Diego Caballero
• [Mlir-commits] [mlir] 3a3ab21 - [mlir][Vector] Add support for high-order masked contractions   Diego Caballero
• [Mlir-commits] [mlir] bacf1aa - Revert "[mlir] Add function for checking if a block is inside a loop"   Kiran Chandramohan
• [Mlir-commits] [mlir] 22cdeb5 - [MLIR][OpenMP] Add Conversion for Atomic Update Op   Kiran Chandramohan
• [Mlir-commits] [mlir] 6d2d4a2 - [MLIR, OpenMP, Flang] Add Conversion to LLVM for Section Op   Kiran Chandramohan
• [Mlir-commits] [mlir] 6bb0ab0 - [MLIR] Propagate unpack through element-wise ops   Lorenzo Chelini
• [Mlir-commits] [mlir] 30d542f - [MLIR][Tensor] Introduce a pattern to propagate through `tensor.pad`   Lorenzo Chelini
• [Mlir-commits] [mlir] 1c22802 - [MLIR][Linalg] Change insertion point for `bubbleUpPackOpThroughElemGenericOp`   Lorenzo Chelini
• [Mlir-commits] [mlir] ee3efcf - Fix comment in `Vectorization.cpp` (NFC)   Lorenzo Chelini
• [Mlir-commits] [mlir] dcfd35d - [mlir]: Added IntArrayNthElemMaxValue & IntArrayNthElemInRange predicates (NFC)   Aviad Cohen
• [Mlir-commits] [mlir] 3810f76 - [mlir][tensor|memref] Harden the checks on dim op   Quentin Colombet
• [Mlir-commits] [mlir] 3818237 - [mlir][Tiling] Properly reject "buffer semantic" operations   Quentin Colombet
• [Mlir-commits] [mlir] 5a028cc - [mlir][LinAlgTransformOps][NFC] Use emitSilenceableError   Quentin Colombet
• [Mlir-commits] [mlir] 8b28500 - [mlir][MemRef][TransformOps] Fix error reporting for multibuffer   Quentin Colombet
• [Mlir-commits] [mlir] 4bc2357 - [mlir][MemRef|Tensor] Fix the handling of DimOp   Quentin Colombet
• [Mlir-commits] [mlir] 985f7ff - [mlir][gpu] Add support for integer types in gpu.subgroup_mma ops   Quinn Dawkins
• [Mlir-commits] [mlir] 5205c71 - [mlir][gpu] Add support for unsigned integer extend in vector to gpu.subgroup_mma lowering   Quinn Dawkins
• [Mlir-commits] [mlir] 21e6e70 - [mlir][linalg] Match element type of result when doing propagation of unpack through elementwise   Quinn Dawkins
• [Mlir-commits] [mlir] 5f2618f - [mlir][linalg] Allow constant exprs in pack/unpack propagation through elementwise   Quinn Dawkins
• [Mlir-commits] [mlir] bbf1d80 - [mlir][tensor] Fix transpose permutation in tensor.pack generalization pattern   Quinn Dawkins
• [Mlir-commits] [mlir] 87d02e0 - Recommit "[Support] change StringMap hash function from djbHash to xxHash"   Erik Desjardins
• [Mlir-commits] [mlir] 499abb2 - Add generic type attribute mapping infrastructure, use it in GpuToX   Krzysztof Drewniak
• [Mlir-commits] [mlir] d0f19ce - [mlir] Handle different pointer sizes in unranked memref descriptors   Krzysztof Drewniak
• [Mlir-commits] [mlir] 7fb9bbe - [mlir][Memref] Add memref.memory_space_cast and its lowerings   Krzysztof Drewniak
• [Mlir-commits] [mlir] 22f0c7a - [mlir][AMDGPU] 8-bit float usage in the AMDGPU dialect   Krzysztof Drewniak
• [Mlir-commits] [mlir] 584f643 - [MLIR][AMDGPU][ROCDL] Adding raw.buffer.atomic.fmax/smax/umin support   Krzysztof Drewniak
• [Mlir-commits] [mlir] dcee187 - [mlir] Add function for checking if a block is inside a loop   Tom Eccles
• [Mlir-commits] [mlir] 81a79ee - [mlir] Add function for checking if a block is inside a loop   Tom Eccles
• [Mlir-commits] [mlir] d768bf9 - [NFC][TargetParser] Replace uses of llvm/Support/Host.h   Archibald Elliott
• [Mlir-commits] [mlir] 4b815d8 - [mlir][NFC] Remove unused variable 'indexType' in GPUTransformOps.cpp   Jie Fu
• [Mlir-commits] [mlir] c0f504d - [mlir] Fix two build warnings in VectorToGPU (NFC)   Jie Fu
• [Mlir-commits] [mlir] 71251e8 - [mlir] Fix -Wsign-compare in SparseTensorRewriting.cpp and Sparsification.cpp (NFC)   Jie Fu
• [Mlir-commits] [mlir] 3de0bc4 - [MLIR] Reintroduce shared attribute names in the LLVM dialect   Frederik Gossen
• [Mlir-commits] [mlir] 31fc47e - [MLIR] Expose region equivalence check through OperationEquivalence   Frederik Gossen
• [Mlir-commits] [mlir] b12bcf3 - [MLIR] Add pass to deduplicate functions   Frederik Gossen
• [Mlir-commits] [mlir] e7b52c4 - [MLIR] Add primitive builders for scf.if   Frederik Gossen
• [Mlir-commits] [mlir] d8153ae - [mlir][llvm] Opaque pointer support for atomic and call ops.   Tobias Gysi
• [Mlir-commits] [mlir] f03b895 - [mlir][llvm] Use tablegen for enum conversion.   Tobias Gysi
• [Mlir-commits] [mlir] b6a2e33 - [mlir][llvm] Fix bug in constant import from LLVM IR.   Tobias Gysi
• [Mlir-commits] [mlir] 7dd974a - [mlir][llvm] Drop opaque ptr test in LLVM IR import.   Tobias Gysi
• [Mlir-commits] [mlir] 7f97895 - [mlir][llvm] Add extra attributes to the atomic ops.   Tobias Gysi
• [Mlir-commits] [mlir] fef08da - [mlir][llvm] Store memory op metadata using op attributes.   Tobias Gysi
• [Mlir-commits] [mlir] 771d9c0 - [mlir][func] Use the generated pass options in func to llvm.   Tobias Gysi
• [Mlir-commits] [mlir] 39da468 - Revert "[mlir][func] Use the generated pass options in func to llvm."   Tobias Gysi
• [Mlir-commits] [mlir] 5dc34eb - Reland "[mlir][func] Use the generated pass options in func to llvm."   Tobias Gysi
• [Mlir-commits] [mlir] 240c6f2 - [mlir][llvm] Improve LoadOp and StoreOp import.   Tobias Gysi
• [Mlir-commits] [mlir] bfea837 - [mlir][llvm] Reintroduce string based attribute setting.   Tobias Gysi
• [Mlir-commits] [mlir] 4469ec1 - [mlir][llvm] Import alias scope metadata from LLVM IR.   Tobias Gysi
• [Mlir-commits] [mlir] a048d73 - [MLIR][LLVM] Only disallow inlining for selected function attributes.   Tobias Gysi
• [Mlir-commits] [mlir] 6f4af64 - [mlir][llvm] Add atomic support to the LoadOp.   Tobias Gysi
• [Mlir-commits] [mlir] 9185896 - [mlir][llvm] Add atomic support to the StoreOp.   Tobias Gysi
• [Mlir-commits] [mlir] a52332d - [mlir][llvm] Add FastmathFlagsInterface only once (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] 3948f0a - [mlir][llvm] Rename LLVMOpsInterfaces.td to LLVMInterfaces.td (NFC).   Tobias Gysi
• [Mlir-commits] [mlir] 0fad18c - [CMake] Remove custom ccache CMake logic   Tobias Hieta
• [Mlir-commits] [mlir] fff762d - Revert "[CMake] Remove custom ccache CMake logic"   Tobias Hieta
• [Mlir-commits] [mlir] f4e69d9 - [mlir] Fix an unused variable warning   Kazu Hirata
• [Mlir-commits] [mlir] 0542390 - [mlir] Use std::optional instead of llvm::Optional (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 0dc18b6 - [mlir] Use std::optional instead of llvm::Optional (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 5382d28 - [mlir] Use std::optional instead of llvm::Optional (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] cbde212 - Use APInt::popcount instead of APInt::countPopulation (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] f8f3db2 - Use APInt::count{l, r}_{zero, one} (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] a7baaab - Use APInt::isZero instead of APInt::isNulLValue (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 4a05edd - [llvm] Use APInt::getZero instead of APInt::getNullValue (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] b7ffd96 - Use APInt::getAllOnes instead of APInt::getAllOnesValue (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 9e5d249 - Use APInt::isOne instead of APInt::isOneValue (NFC)   Kazu Hirata
• [Mlir-commits] [mlir] 96267b6 - [mlir] Add Float8E5M2FNUZ and Float8E4M3FNUZ types to MLIR   Chris Jackson
• [Mlir-commits] [mlir] df9500c - [mlir] Simplify a few cast implementations   Rahul Kayaith
• [Mlir-commits] [mlir] a0b2104 - [mlir] Use TypedValue in single result traits   Rahul Kayaith
• [Mlir-commits] [mlir] 0ce25b1 - [mlir] Require explicit casts when using TypedValue   Rahul Kayaith
• [Mlir-commits] [mlir] 82ae83a - [mlir] Silence a few -Wunused-but-set-parameter warnings   Rahul Kayaith
• [Mlir-commits] [mlir] 2aa1258 - [mlir][python] Don't emit diagnostics when printing invalid ops   Rahul Kayaith
• [Mlir-commits] [mlir] 81d0d2b - [mlir][sparse] Sparse reduction in lex order no longer produces dense output   Jim Kitchen
• [Mlir-commits] [mlir] f42e1a7 - Fix tsan problem where the per-thread shared_ptr() can be locked right before the cache is destroyed causing a race where it tries to remove an entry from a destroyed cache.   Benjamin Kramer
• [Mlir-commits] [mlir] e28bbfe - Revert "[mlir][linalg] Vectorize tensor.extract using contiguous loads"   Benjamin Kramer
• [Mlir-commits] [mlir] 58d246a - [mlir][spirv] Drop outdated TODO comment   Jakub Kuderski
• [Mlir-commits] [mlir] dc38cbc - [mlir][arith] Fold selection over constant vector conditions   Jakub Kuderski
• [Mlir-commits] [mlir] 892bf09 - [mlir][arith] Allow to specify `constFoldBinaryOp` result type   Jakub Kuderski
• [Mlir-commits] [mlir] 31c3528 - [mlir][spirv] Fix crash when decorating physical storage buffer pointers   Jakub Kuderski
• [Mlir-commits] [mlir] 1ce1e4a - [mlir][spirv] Validate float type bitwidths   Jakub Kuderski
• [Mlir-commits] [mlir] 85075b5 - [mlir] Use mlir::TypedValue to avoid compiler bug in MSVC.   Adrian Kuegel
• [Mlir-commits] [mlir] 01b9d35 - [mlir] Use the same name as the generated parameter name (NFC).   Adrian Kuegel
• [Mlir-commits] [mlir] 09ba324 - [NFC] [mlir] fix file header for SCFToControlFlow.h   Xiang Li
• [Mlir-commits] [mlir] cc4fb58 - [mlir] support complex type in DenseElementsAttr::get.   Xiang Li
• [Mlir-commits] [mlir] 1f07853 - [mlir][sparse] introduce sparse_tensor.pack operation   Peiming Liu
• [Mlir-commits] [mlir] a41672e - [mlir][sparse] implement lowering rules for sparse_tensor.pack operation   Peiming Liu
• [Mlir-commits] [mlir] 7fef8d6 - [mlir][sparse] implement bufferizableOpInterface for sparse_tensor.pack operation   Peiming Liu
• [Mlir-commits] [mlir] 7a8edea - [mlir][sparse] fix bug when packing tensor with 32 bit pointer width.   Peiming Liu
• [Mlir-commits] [mlir] ebf22c4 - [mlir][memref] fix typo in documentation (NFC)   Peiming Liu
• [Mlir-commits] [mlir] 0352690 - [mlir][sparse] make foreach operation support sparse tensor slices.   Peiming Liu
• [Mlir-commits] [mlir] 6dbca86 - [mlir][sparse] introduce sparse_tensor::unpack operation   Peiming Liu
• [Mlir-commits] [mlir] dc6427d - [mlir][sparse] implement lowering rules for sparse_tensor::unpack   Peiming Liu
• [Mlir-commits] [mlir] ce9ce66 - [mlir][sparse] fix a memory leakage when converting from a tensor slice   Peiming Liu
• [Mlir-commits] [mlir] 81cb70e - [mlir][sparse] fix a bug in UnpackOp converter.   Peiming Liu
• [Mlir-commits] [mlir] 9ed587e - [mlir][sparse] split sparse concatenate test cases smaller ones.   Peiming Liu
• [Mlir-commits] [mlir] 7d62bc9 - [mlir][sparse] disable sparse convolution test cases.   Peiming Liu
• [Mlir-commits] [mlir] e2e83f4 - [mlir][sparse] support coiteration over sparse tensor slices   Peiming Liu
• [Mlir-commits] [mlir] c738b43 - [mlir][sparse] introduce operations to query sparse tensor slice offset/strides at the given dimenion   Peiming Liu
• [Mlir-commits] [mlir] f30ca85 - [mlir][sparse] reduce the input size in stress_test.py to make it finish in time.   Peiming Liu
• [Mlir-commits] [mlir] 9e8d931 - [mlir][sparse] allow foreach operation to generate out-of-order loop on non-annotated tensor.   Peiming Liu
• [Mlir-commits] [mlir] 0af5f84 - [mlir][sparse] split reshape.mlir into expand/collapse_shape.mlir.   Peiming Liu
• [Mlir-commits] [mlir] e6758c8 - [mlir][sparse] comment out test cases in sparse_conversion with similiar behavior.   Peiming Liu
• [Mlir-commits] [mlir] b7d86f3 - [mlir][sparse] revert optimization for dense->csc conversion.   Peiming Liu
• [Mlir-commits] [mlir] 44ff23d - [mlir][sparse] unconditionally use IndexType for sparse_tensor.specifier   Peiming Liu
• [Mlir-commits] [mlir] 85dbb3f - [mlir][sparse] support sparse tensor element type conversion in codegen path   Peiming Liu
• [Mlir-commits] [mlir] 849529b - [mlir][sparse] fix performance bug in matmul with a sparse rhs due to suboptimal iteration graphs.   Peiming Liu
• [Mlir-commits] [mlir] 47fbb24 - [mlir][scf] Fix typo in description of option of TestSCFForUtilsPass (NFC).   Ingo Müller
• [Mlir-commits] [mlir] b716bf8 - [mlir][scf] Fix builder of WhileOp with region builder arguments.   Ingo Müller
• [Mlir-commits] [mlir] 6592b01 - [mlir][func] Add support for nested tuples to TestDecomposeCallGraphTypes.   Ingo Müller
• [Mlir-commits] [mlir] dc700f1 - [mlir][bufferization] Restrict function boundary buffer. to func.call.   Ingo Müller
• [Mlir-commits] [mlir] 4bba8bd - [mlir] Add RewriterBase::replaceAllUsesWith for Blocks.   Ingo Müller
• [Mlir-commits] [mlir] fd85a64 - [mlir][llvm] Add isVarArg flag to lookupOrCreateFn.   Ingo Müller
• [Mlir-commits] [mlir] 2fce909 - [mlir][affine] fix affine LICM pass for has effected memory's user   Jeff Niu
• [Mlir-commits] [mlir] f25cfd3 - [mlir] Reintroduce API for creating operations with a DictionaryAttr   Jeff Niu
• [Mlir-commits] [mlir] 2cfc66a - [mlir] Add a SharedSymbolTableCollection class   Jeff Niu
• [Mlir-commits] [mlir] 55bc18a - [mlir] Fix typo causing build error   Jeff Niu
• [Mlir-commits] [mlir] 1cb91b4 - [mlir] Add nontemporal field to memref.load/store and convey to llvm.load/store   Guray Ozen
• [Mlir-commits] [mlir] 057cae6 - [MLIR][Presburger] Matrix: hide expensive consistency check behind EXPENSIVE_CHECKS macro   Arjun P
• [Mlir-commits] [mlir] 6fa26f6 - [mlir][arith] Fold on extension of FP constants using arith.extf   Victor Perez
• [Mlir-commits] [mlir] be58d48 - [mlir] Canonicalize dynamic tensor.pad ops with constant inputs   George Petterson
• [Mlir-commits] [mlir] 44003d7 - Revert "Fix tsan problem where the per-thread shared_ptr() can be locked right before the cache is destroyed causing a race where it tries to remove an entry from a destroyed cache."   Mitch Phillips
• [Mlir-commits] [mlir] ee308c9 - [mlir][py] Fix infer return type invocation for variadics   Jacques Pienaar
• [Mlir-commits] [mlir] 393aa9c - [mlir][NFC] Add omitted operations to Transform dialect's doc   Jacques Pienaar
• [Mlir-commits] [mlir] 1d1a2eb - [mlir][py] Fix unused var   Jacques Pienaar
• [Mlir-commits] [mlir] a2dcd99 - [mlir][tosa] Add TOSA f64 type support for cast op   Jacques Pienaar
• [Mlir-commits] [mlir] 62570b7 - [mlir][linalg] Fix crash in vectorizer when expanding affine apply   Thomas Raoux
• [Mlir-commits] [mlir] a7686db - [mlir][gpu] Allow distributing to different level of IDs without failing   Thomas Raoux
• [Mlir-commits] [mlir] d18523c - [mlir][linalg] Check for tensor of 0 dims during vectorization   Thomas Raoux
• [Mlir-commits] [mlir] 2fc3c5c - [mlir][vector] Prevent duplicating operations during vector distribute   Thomas Raoux
• [Mlir-commits] [mlir] 288ae0b - [mlir][gpu] NFC change to pass threadID ops to rewriteOneForeachThreadToGpuThreads   Thomas Raoux
• [Mlir-commits] [mlir] 3cf7f22 - [mlir][vectorToGPU] Fix type used when folding transpose into read op   Thomas Raoux
• [Mlir-commits] [mlir] e3a88a4 - Revert "[mlir][vector] Prevent duplicating operations during vector distribute"   Thomas Raoux
• [Mlir-commits] [mlir] 0eabb88 - [mlir][gpu] NFC let user pick the threadID values when distributing foreach_thread   Thomas Raoux
• [Mlir-commits] [mlir] ddeb55a - [mlir] add option to multi-buffering   Thomas Raoux
• [Mlir-commits] [mlir] 17b2e73 - [mlir][LinAlg][Transform] Add a transform op for conv2d to im2col   Thomas Raoux
• [Mlir-commits] [mlir] b58daf9 - Add a lowering for memref.dealloc with unranked memrefs.   Johannes Reifferscheid
• [Mlir-commits] [mlir] 1fbccfd - [mlir][IRNumbering] Fix the dialect comparator to be strict   River Riddle
• [Mlir-commits] [mlir] 964997a - [mlir][core] Fix ValueRange printing in AsmPrinter   River Riddle
• [Mlir-commits] [mlir] 3941355 - [mlir][vector] Support 0-D vector when eliding single element reduction   Kai Sasaki
• [Mlir-commits] [mlir] 48f04ff - [mlir][affine] Normalize constant valued bound loop   Kai Sasaki
• [Mlir-commits] [mlir] de58a4f - [mlir][Analysis] Guard data flow analysis from no block function   Kai Sasaki
• [Mlir-commits] [mlir] 146f5d9 - [MLIR] NFC: fully scope use FastMathFlag.   Christian Sigg
• [Mlir-commits] [mlir] a7bf683 - [MLIR] NFC: fully scope use FastMathFlag.   Christian Sigg
• [Mlir-commits] [mlir] 618caf6 - [mlir] print-op-graph: StringMap=>map to stabilize iteration order   Fangrui Song
• [Mlir-commits] [mlir] 179db7e - [MLIR][OpenMP] Add support for depend clause   Prabhdeep Singh Soni
• [Mlir-commits] [mlir] 1ac248e - [mlir][bufferization][NFC] Rename getAliasingOpOperand/getAliasingOpResult   Matthias Springer
• [Mlir-commits] [mlir] b6ae3f8 - [mlir][tensor][bufferize] Implement getBufferType for CastOp   Matthias Springer
• [Mlir-commits] [mlir] 724a0e2 - [mlir] GreedyPatternRewriteDriver: Ignore scope when rewriting top-level ops   Matthias Springer
• [Mlir-commits] [mlir] d0a7bb9 - [mlir][SCF] Disallow multiple blocks in scf.if "else" region   Matthias Springer
• [Mlir-commits] [mlir] 330372f - [mlir][tensor][bufferize] tensor.empty does not define the result tensor contents   Matthias Springer
• [Mlir-commits] [mlir] 1fdf06d - [mlir][bufferization] Reads from tensors with undefined data are not a conflict   Matthias Springer
• [Mlir-commits] [mlir] c89c31a - [mlir][bufferization] Fix bufferization of repetitive regions   Matthias Springer
• [Mlir-commits] [mlir] cf2d374 - [mlir][bufferize][NFC] Merge AnalysisState and BufferizationAliasInfo   Matthias Springer
• [Mlir-commits] [mlir] 6d14b11 - [mlir][bufferize][NFC] OneShotAnalysis: Expose analysis hooks from AnalysisState   Matthias Springer
• [Mlir-commits] [mlir] dc7ad19 - [mlir][bufferize][NFC] Optimize read-only tensor detection   Matthias Springer
• [Mlir-commits] [mlir] 2b5a020 - [mlir][bufferization][NFC] Cache definitions of read tensors   Matthias Springer
• [Mlir-commits] [mlir] 9fa6b35 - [mlir][bufferization] Improve aliasing OpOperand/OpResult property   Matthias Springer
• [Mlir-commits] [mlir] 2441c07 - [mlir][bufferization] Support multiple leaves in EmptyTensorElimination   Matthias Springer
• [Mlir-commits] [mlir] 00a5ae1 - [mlir][bufferization] Better propagation of bufferizesToMemoryWrite through regions   Matthias Springer
• [Mlir-commits] [mlir] 8f7e740 - [mlir][bufferization] Add restrict and writable attrs to to_tensor   Matthias Springer
• [Mlir-commits] [mlir] 3ef062a - [mlir][transform] Add transform.get_result op   Matthias Springer
• [Mlir-commits] [mlir] 01581e2 - [mlir][linalg] Add bufferize_to_allocation transform op   Matthias Springer
• [Mlir-commits] [mlir] c645eb0 - [mlir][memref] Bufferize memref.tensor_store op   Matthias Springer
• [Mlir-commits] [mlir] 6dd9d18 - [mlir][linalg] Fix insertion point bug in D144022   Matthias Springer
• [Mlir-commits] [mlir] e44f405 - [mlir][bufferization] Fix bug in findValueInReverseUseDefChain   Matthias Springer
• [Mlir-commits] [mlir] fdb9e6a - [mlir][bufferization] Fix crash in EmptyTensorElimination   Matthias Springer
• [Mlir-commits] [mlir] 4cf936d - [mlir][transform] Add transform.get_defining_op op   Matthias Springer
• [Mlir-commits] [mlir] c653283 - This change makes `RewriterBase` symmetric to `OpBuilder`.   Matthias Springer
• [Mlir-commits] [mlir] 279c1d2 - [mlir] GreedyPatternRewriteDriver: Support optional Listener   Matthias Springer
• [Mlir-commits] [mlir] 93d640f - [mlir][SCF][Utils][NFC] Make some utils public for better reuse   Matthias Springer
• [Mlir-commits] [mlir] bafc4df - [mlir] RewriterBase::Listener: Add notifyOperationModified callback   Matthias Springer
• [Mlir-commits] [mlir] 2c40a0a - [mlir][IR][NFC] Rename RewriterBase::replaceUseIf to replaceUsesWithIf   Matthias Springer
• [Mlir-commits] [mlir] 9bdfa8d - [mlir][IR] Use Listener for IR callbacks in OperationFolder   Matthias Springer
• [Mlir-commits] [mlir] c080c1f - [mlir][GPU] Fix incorrect API usage in RewritePatterns   Matthias Springer
• [Mlir-commits] [mlir] bc8de51 - [mlir][IR] Add Iterator template option to IR walkers   Matthias Springer
• [Mlir-commits] [mlir] 61f3775 - [mlir][SCF] Fix incorrect API usage in RewritePatterns   Matthias Springer
• [Mlir-commits] [mlir] 3f9b044 - [mlir][Affine][NFC] Improve FlatAffineValueConstraint dump   Matthias Springer
• [Mlir-commits] [mlir] ca174f1 - [mlir] Pin for the PyPi requirements for mlir   Stella Stamenova
• [Mlir-commits] [mlir] 97fbbd6 - [mlir] Relax version requirement for PyYAML in mlir   Stella Stamenova
• [Mlir-commits] [mlir] 936819b - [mlir][tosa] make Select operator broadcastable in the pass   Rob Suderman
• [Mlir-commits] [mlir] 9d0b596 - [mlir][tosa] Fix segmentation fault in case of folding unranked tensor   Rob Suderman
• [Mlir-commits] [mlir] 0b92f8a - [mlir][NFC] Address filecheck_lint findings in tosa-to-linalg-named.mlir.   Robert Suderman
• [Mlir-commits] [mlir] 740e2e9 - [mlir][math] Math expansion for math.tan   Robert Suderman
• [Mlir-commits] [mlir] 889a117 - [mlir][llvm] Add structured loop metadata   Christian Ulmann
• [Mlir-commits] [mlir] a978757 - [mlir][llvm] Add missing license header (NFC)   Christian Ulmann
• [Mlir-commits] [mlir] b83caa3 - [mlir][llvm] Add support for loop metadata import   Christian Ulmann
• [Mlir-commits] [mlir] c6ac7e9 - [mlir][llvm] Add MD_prof import error handling   Christian Ulmann
• [Mlir-commits] [mlir] 851a897 - [mlir][llvm] Purge struct_attr   Christian Ulmann
• [Mlir-commits] [mlir] e630a50 - [mlir][llvm] Fuse MD_access_group & MD_loop import   Christian Ulmann
• [Mlir-commits] [mlir] fc2c791 - [mlir][llvm] Fix TBAA verfication crash   Christian Ulmann
• [Mlir-commits] [mlir] 4d7c879 - [mlir][llvm] Adapt loop metadata to match llvm   Christian Ulmann
• [Mlir-commits] [mlir] 7f249e4 - [milr][llvm] Add remaining loop metadata support   Christian Ulmann
• [Mlir-commits] [mlir] 544831a - [mlir] Drop unused arith conversion class (NFC)   Christian Ulmann
• [Mlir-commits] [mlir] d94399c - [mlir][llvm] Make LoopAnnotations non-discardable   Christian Ulmann
• [Mlir-commits] [mlir] 873ea45 - [mlir][llvm] Rename void debug type to null   Christian Ulmann
• [Mlir-commits] [mlir] 28542e9 - [mlir][llvm] Use before def debug intrinsic import   Christian Ulmann
• [Mlir-commits] [mlir] bf91cd6 - [mlir][llvm] Verify LLVM module before import   Christian Ulmann
• [Mlir-commits] [mlir] 276358f - [mlir][llvm] Make DI param optional to match LLVM   Christian Ulmann
• [Mlir-commits] [mlir] e133cb9 - [mlir][llvm] Add DINamespace attribute   Christian Ulmann
• [Mlir-commits] [mlir] 87a0479 - [mlir][llvm] Fuse access_group & loop export (NFC)   Christian Ulmann
• [Mlir-commits] [mlir] ddd1d1c - [mlir][llvm] Builders dont access null attr (NFC)   Christian Ulmann
• [Mlir-commits] [mlir] d9391a3 - [mlir][llvm] Stop exporting empty debug MD strings   Christian Ulmann
• [Mlir-commits] [mlir] 7cedd95 - [mlir][Linalg] Add a transform.structured.lower_unpack op   Nicolas Vasilache
• [Mlir-commits] [mlir] 89675aa - [mlir][Linalg] NFC - Extract inferGemmDims   Nicolas Vasilache
• [Mlir-commits] [mlir] 1528b70 - [mlir][Linalg] NFC - Extract inferGemmDims - Amend forgotten changes   Nicolas Vasilache
• [Mlir-commits] [mlir] 3f3f4d5 - [mlir][Linalg] Fix expected buffer semantics crash   Nicolas Vasilache
• [Mlir-commits] [mlir] 8ce6f7d - [mlir][Linalg] NFC - Fail gracefully instead of crashing in SplitReduction   Nicolas Vasilache
• [Mlir-commits] [mlir] 5ef7cea - [mlir][Vector] Significantly improve VectorToGPU.cpp   Nicolas Vasilache
• [Mlir-commits] [mlir] 3e0866b - [mlir][Vector] NFC - Fail gracefully on size mismatch instead of assert   Nicolas Vasilache
• [Mlir-commits] [mlir] 96179df - [mlir][Linalg] Add a transform dialect op to rewrite ops to destination passing style.   Nicolas Vasilache
• [Mlir-commits] [mlir] ec64038 - [mlir][MemRef] NFC - Add debug information to MultiBuffer.cpp   Nicolas Vasilache
• [Mlir-commits] [mlir] c624027 - [mlir][linalg][TransformOps] Connect hoistRedundantVectorTransfers   Nicolas Vasilache
• [Mlir-commits] [mlir] 8cc7189 - [mlir][Linalg][Transform] NFC - Replace assert by more graceful error   Nicolas Vasilache
• [Mlir-commits] [mlir] 4521b11 - [mlir][Linalg] Reimplement hoisting on tensors as a subset-based transformation   Nicolas Vasilache
• [Mlir-commits] [mlir] bb2ae98 - [mlir][Linalg] NFC - Apply cleanups to transforms   Nicolas Vasilache
• [Mlir-commits] [mlir] 6d2501b - [mlir][Linalg] Refactor transform.structured.pad to separate out hoisting   Nicolas Vasilache
• [Mlir-commits] [mlir] 6da0184 - [mlir][NFC] Use fully qualified C++ namespaces in .td files.   Vladislav Vinogradov
• [Mlir-commits] [mlir] 0cabe38 - [mlir][NFC] Use fully qualified C++ namespaces in .td files.   Vladislav Vinogradov
• [Mlir-commits] [mlir] 061201e - [mlir][linalg] Enhance padding LinalgOps to handle tensor.empty cases.   Hanhan Wang
• [Mlir-commits] [mlir] f71de25 - [mlir][tensor] Add shape inference methods to tensor::PackOp.   Hanhan Wang
• [Mlir-commits] [mlir] be75cf9 - [mlir][tensor] Fix a bug in tiling unpack op.   Hanhan Wang
• [Mlir-commits] [mlir] c6c41c3 - [mlir][linalg][tensor] Delete duplicate tests and fix typo in filename   Hanhan Wang
• [Mlir-commits] [mlir] 5fa9933 - [mlir][tensor] Do not use affine ops on non-perfect-tiling unpack cases.   Hanhan Wang
• [Mlir-commits] [mlir] 4d03651 - [mlir][tensor] Improve size inference in tiling tensor.pack ops.   Hanhan Wang
• [Mlir-commits] [mlir] 08f31b8 - [docs] Make consistent between MLIR tutorial doc and example code   Hsiangkai Wang
• [Mlir-commits] [mlir] bfe4ce3 - [mlir][sparse] Port the remaining integration tests to use SVE   Andrzej Warzynski
• [Mlir-commits] [mlir] 7301a7c - [mlir][linalg] Make Linalg vectorizer lower affine.apply   Andrzej Warzynski
• [Mlir-commits] [mlir] 7171244 - [mlir] Add vectorize_nd_extract attribute to the masked_vectorize Op   Andrzej Warzynski
• [Mlir-commits] [mlir] 89b144e - [mlir][linalg] Vectorize tensor.extract using contiguous loads   Andrzej Warzynski
• [Mlir-commits] [mlir] 779d54f - [mlir] Fix a -Wunused-variable warning, NFC   Haojian Wu
• [Mlir-commits] [mlir] 2c46051 - [mlir][spirv] Fix vector type mismatch in UnifyAliasedResourcePass   Lei Zhang
• [Mlir-commits] [mlir] 78e172f - [mlir][spirv] Support i32 memref.atomic_rmw conversion   Lei Zhang
• [Mlir-commits] [mlir] a1aad28 - [mlir][vector] NFC: Improve vector type accessor methods   Lei Zhang
• [Mlir-commits] [mlir] 80bb947 - [mlir] NFC: Prepare GPUToSPIRV tests for supporting 64bit index   Lei Zhang
• [Mlir-commits] [mlir] 9a4c768 - [mlir][spirv] Respect client API requirements for 64-bit index   Lei Zhang
• [Mlir-commits] [mlir] 85365b1 - [mlir][spirv] Fix Physical32/Physical64 support for OpenCL   Lei Zhang
• [Mlir-commits] [mlir] 8e628f7 - [mlir] Avoid building some shared libraries when PIC is off   Han Zhu
• [Mlir-commits] [mlir] 6e44f11 - [mlir] provide a base class for transform interpreter passes   Alex Zinenko
• [Mlir-commits] [mlir] 0242b96 - [mlir] more side effect verification in transform dialect   Alex Zinenko
• [Mlir-commits] [mlir] 749f7f9 - [mlir][transform] Fix apply WithPDLPatternsOp with non-pattern op   Alex Zinenko
• [Mlir-commits] [mlir] a702628 - [mlir] add support for transform dialect value handles   Alex Zinenko
• [Mlir-commits] [mlir] 7fdc2ed - [mlir] reallow null results in TransformEachOpTrait   Alex Zinenko
• [Mlir-commits] [mlir] 0b33f0d - [mlir][linalg] expose convolution dimension classifier   Alex Zinenko
• [Mlir-commits] [mlir] 8c7cfa3 - [mlir] fix LLVM IR translation of vector<... x index>   Alex Zinenko
• [Mlir-commits] [mlir] 115711c - [mlir][LinAlg][Transform][GPU] Add GPU memory hierarchy to the transform.promote op   Alex Zinenko
• [Mlir-commits] [mlir] b46e78c - [mlir][NFC] Cleanup Passes documentation   Alex Zinenko
• [Mlir-commits] [mlir] 9a29d87 - [mlir][sparse] Add checking parent op of SortOp   Alex Zinenko
• [Mlir-commits] [mlir] 3b1c86c - [mlir][sparse] Implement heap sort for sparse_tensor.sort.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] a150766 - [mlir][sparse] Implement hybrid quick sort for sparse_tensor.sort.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] e800c66 - [mlir][sparse] Extend readCOOIndices to support overhead types beyond index_type.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 86c8ea9 - [mlir] Add check for value that might be null.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] c2e248c - [mlir][sparse] Remove the expansion of symmetric MTX in the sparse tensor storage.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 09fd9ef - [mlir] Execute all requested translations in MlirTranslateMain   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] 27ea470 - [mlir][sparse] Add runtime support for reading a COO tensor and writing the data to the given indices and values buffers.   llvmlistbot at llvm.org
• [Mlir-commits] [mlir] bb4fc6b - [mlir][sparse] Adding `SparseTensorType::{operator==, hasSameDimToLvlMap}`   wren romano
• [Mlir-commits] [mlir] d950bdc - [mlir][sparse] misc code cleanup   wren romano
• [Mlir-commits] [mlir] ae7942e - [mlir][sparse] adding `SparseTensorType::get{Pointer, Index}Type` methods   wren romano

Last message date: Tue Feb 28 17:22:45 PST 2023
Archived on: Tue Feb 28 17:22:47 PST 2023

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