Matthias Springer 647aa6616f [mlir][SPIRVToLLVM] Set valid insertion point after op erasure (#146551) ...

Erasing/replacing an op, which is also the current insertion point,
invalidates the insertion point. Explicitly set the insertion point, so
that `copy` does not crash after the One-Shot Dialect Conversion
refactoring. (`ConversionPatternRewriter` will start behaving more like
a "normal" rewriter.)

2025-07-02 09:25:24 +02:00

..

AffineToStandard

…

AMDGPUToROCDL

[AMDGPU] Adding AMDGPU dialect wrapper for ROCDL transpose loads. (#145395)

2025-06-25 22:58:14 -04:00

ArithCommon

…

ArithToAMDGPU

[AMD][ROCDL] Add packed conversions fp8/bf8->bf16 and fp8/bf8->fp32 in ROCDL dialect (#131850)

2025-03-21 14:49:50 +00:00

ArithToArmSME

…

ArithToEmitC

[mlir][EmitC] Add pass that combines all available emitc conversions (#117549)

2025-05-01 07:24:01 -07:00

ArithToLLVM

[mlir][arith] Add overflow flags to arith.trunci (#144863)

2025-06-19 13:59:22 +02:00

ArithToSPIRV

[mlir][SPRIV][NFC] Avoid rollback in TypeCastingOpPattern (#136284)

2025-04-19 10:01:50 +02:00

ArmNeon2dToIntr

…

ArmSMEToLLVM

Emit inbounds and nuw attributes in memref. (#138984)

2025-05-20 14:16:22 -07:00

ArmSMEToSCF

[mlir][ArmSME] Audit ArmSME load/store ops (#139573)

2025-05-13 13:01:01 +01:00

AsyncToLLVM

[mlir][LLVM] Add OpBuilder & to lookupOrCreateFn functions (#136421)

2025-04-20 10:06:22 +02:00

BufferizationToMemRef

…

ComplexCommon

…

ComplexToLibm

…

ComplexToLLVM

…

ComplexToSPIRV

…

ComplexToStandard

[MLIR][NFC] Use base alias for constructor inheritance (#127756)

2025-02-19 16:05:25 +01:00

ControlFlowToLLVM

[mlir] Migrate away from ValueRange(std::nullopt) (NFC) (#145210)

2025-06-22 06:30:17 -07:00

ControlFlowToSCF

…

ControlFlowToSPIRV

[MLIR][NFC] Use base alias for constructor inheritance (#127756)

2025-02-19 16:05:25 +01:00

ConvertToEmitC

[mlir][EmitC] Add pass that combines all available emitc conversions (#117549)

2025-05-01 07:24:01 -07:00

ConvertToLLVM

…

FuncToEmitC

[mlir][EmitC] Add pass that combines all available emitc conversions (#117549)

2025-05-01 07:24:01 -07:00

FuncToLLVM

[mlir][Transforms] Add 1:N support to replaceUsesOfBlockArgument (#145171)

2025-06-23 12:07:00 +02:00

FuncToSPIRV

[MLIR][NFC] Use base alias for constructor inheritance (#127756)

2025-02-19 16:05:25 +01:00

GPUCommon

[mlir][gpu] Fix bug with GPU hardware intrinsic global location (#144923)

2025-06-22 22:09:44 -06:00

GPUToLLVMSPV

…

GPUToNVVM

[MLIR] Mark LLVM::FMAOp as legal (#144671)

2025-06-18 15:49:00 +02:00

GPUToROCDL

[AMDGPU] Set AS8 address width to 48 bits

2025-05-19 17:26:05 -07:00

GPUToSPIRV

[mlir][gpu] Add gpu.rotate operation (#142796)

2025-07-01 11:32:25 +01:00

IndexToLLVM

…

IndexToSPIRV

…

LinalgToStandard

…

LLVMCommon

[MLIR] Replace getVoidPtrType with getPtrType in ConvertToLLVMPattern (#145657)

2025-06-27 12:31:53 +02:00

MathToEmitC

…

MathToFuncs

[NFC] Use more isa and isa_and_nonnull instead dyn_cast for predicates (#137393)

2025-05-13 22:34:42 +08:00

MathToLibm

…

MathToLLVM

[mlir][math] Fix intrinsic conversions to LLVM for 0D-vector types (#141020)

2025-06-02 12:27:44 +01:00

MathToROCDL

[MLIR][ROCDL] Add conversion of math.erfc to AMD GPU library calls (#128899)

2025-02-27 14:32:25 -05:00

MathToSPIRV

[mlir][spirv] Add lowering of multiple math trig/hypb functions (#143604)

2025-06-11 09:20:40 -04:00

MemRefToEmitC

[mlir] Use llvm::is_contained (NFC) (#140467)

2025-05-19 06:18:46 -07:00

MemRefToLLVM

[MLIR] Replace getVoidPtrType with getPtrType in ConvertToLLVMPattern (#145657)

2025-06-27 12:31:53 +02:00

MemRefToSPIRV

[mlir][spirv]: Add Image to Vulkan Storage Class Map (#144899)

2025-06-23 17:44:33 +01:00

MeshToMPI

Silence "non-void function does not return a value in all control paths" in Release builds. NFC

2025-06-23 17:58:41 +02:00

MPIToLLVM

Lower allreduce (#144716)

2025-06-23 14:01:17 +02:00

NVGPUToNVVM

[MLIR][LLVM] Tail call support for inline asm op (#140826)

2025-05-22 15:30:31 -07:00

NVVMToLLVM

…

OpenACCToSCF

…

OpenMPToLLVM

[mlir] Use llvm::all_of (NFC) (#140464)

2025-05-18 18:13:49 -07:00

PDLToPDLInterp

[mlir] Migrate away from std::nullopt (NFC) (#145523)

2025-06-25 11:49:22 -07:00

ReconcileUnrealizedCasts

…

SCFToControlFlow

…

SCFToEmitC

[mlir][EmitC] Add pass that combines all available emitc conversions (#117549)

2025-05-01 07:24:01 -07:00

SCFToGPU

[mlir] Migrate away from TypeRange(std::nullopt) (NFC) (#145246)

2025-06-22 19:09:31 -07:00

SCFToOpenMP

[mlir][OpenMP] add attribute for privatization barrier (#140089)

2025-05-22 15:24:02 +01:00

SCFToSPIRV

[mlir] Migrate away from ValueRange(std::nullopt) (NFC) (#145210)

2025-06-22 06:30:17 -07:00

ShapeToStandard

…

SPIRVCommon

…

SPIRVToLLVM

[mlir][SPIRVToLLVM] Set valid insertion point after op erasure (#146551)

2025-07-02 09:25:24 +02:00

TensorToLinalg

…

TensorToSPIRV

[MLIR][NFC] Use base alias for constructor inheritance (#127756)

2025-02-19 16:05:25 +01:00

TosaToArith

[mlir][tosa] Add support for EXT-DOUBLEROUND and EXT-INEXACTROUND (#130337)

2025-03-10 19:32:30 +00:00

TosaToLinalg

[mlir] Migrate away from std::nullopt (NFC) (#145523)

2025-06-25 11:49:22 -07:00

TosaToMLProgram

[tosa] Change VariableOp to align with spec (#142240)

2025-06-03 17:41:33 +01:00

TosaToSCF

[mlir][tosa] Update ControlFlow variable names to match with TOSA v1.0 spec (#129790)

2025-03-05 01:17:42 +00:00

TosaToTensor

[mlir][tosa] Fix indexing in TosaToTensor (#140906)

2025-05-22 17:08:05 +03:00

UBToLLVM

…

UBToSPIRV

…

VectorToArmSME

[mlir][vector] Standardize base Naming Across Vector Ops (NFC) (#137859)

2025-05-12 09:44:50 +01:00

VectorToGPU

[mlir] Fix unused-variable warnings

2025-05-22 12:37:28 -07:00

VectorToLLVM

[mlir][Vector] Lower vector.to_elements to LLVM (#145766)

2025-06-26 10:36:08 -07:00

VectorToSCF

[mlir][vector] Restrict vector.insert/vector.extract to disallow 0-d vectors (#121458)

2025-06-26 09:47:06 +01:00

VectorToSPIRV

[mlir][vector] Standardise valueToStore Naming Across Vector Ops (NFC) (#134206)

2025-04-07 13:56:54 +01:00

VectorToXeGPU

[mlir][vector] Standardize base Naming Across Vector Ops (NFC) (#137859)

2025-05-12 09:44:50 +01:00

CMakeLists.txt

[mlir][EmitC] Add pass that combines all available emitc conversions (#117549)

2025-05-01 07:24:01 -07:00