llvm-project/mlir/lib/Conversion
Yi Zhang e1318078a4 Support non identity layout map for reshape ops in MemRefToLLVM lowering
This change borrows the ideas from `computeExpanded/CollapsedLayoutMap`
and computes the dynamic strides at runtime for the memref descriptors.

Differential Revision: https://reviews.llvm.org/D124001
2022-04-26 13:03:53 -04:00
..
AffineToStandard [mlir] Trim a huge number of unnecessary dependencies on the Func dialect 2022-03-01 12:10:04 -08:00
ArithmeticToLLVM [mlir] Move SelectOp from Standard to Arithmetic 2022-02-02 14:45:12 -08:00
ArithmeticToSPIRV [mlir][spirv] Check nullptr before usage to fix crash 2022-04-04 17:20:00 -04:00
ArmNeon2dToIntr [mlir] Update FuncOp conversion passes to Pass/InterfacePass<FunctionOpInterface> 2022-03-08 12:25:32 -08:00
AsyncToLLVM [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
BufferizationToMemRef [mlir][bufferize] Support layout maps in bufferization.clone lowering 2022-03-16 22:29:22 +09:00
ComplexToLLVM [mlir] Change ComplexToLLVM to be a generic pass 2022-03-08 12:25:32 -08:00
ComplexToStandard [mlir] Update FuncOp conversion passes to Pass/InterfacePass<FunctionOpInterface> 2022-03-08 12:25:32 -08:00
ControlFlowToLLVM [mlir] Rename the Standard dialect to the Func dialect 2022-03-01 12:10:04 -08:00
ControlFlowToSPIRV [mlir][SPIRV] Split up StandardToSPIRV now that the Standard dialect is gone 2022-03-02 13:14:36 -08:00
FuncToLLVM [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
FuncToSPIRV [mlir][spirv] Convert func.call to spv.FunctionCall 2022-03-26 19:21:23 +08:00
GPUCommon [mlir:FunctionOpInterface] Rename the "type" attribute to "function_type" 2022-03-16 17:07:04 -07:00
GPUToNVVM [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
GPUToROCDL [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
GPUToSPIRV [mlir:FunctionOpInterface] Rename the "type" attribute to "function_type" 2022-03-16 17:07:04 -07:00
GPUToVulkan [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
LLVMCommon [mlir][LLVM] Fix crash when using opaque pointers in function signatures 2022-04-23 00:10:31 +02:00
LinalgToLLVM [mlir] Split out a new ControlFlow dialect from Standard 2022-02-06 14:51:16 -08:00
LinalgToSPIRV [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
LinalgToStandard [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
MathToLLVM [mlir][Pass] Deprecate FunctionPass in favor of OperationPass<FuncOp> 2022-01-18 19:52:44 -08:00
MathToLibm [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
MathToSPIRV [mlir][spirv] Add pattern to lower math.copysign 2022-04-01 12:06:47 -04:00
MemRefToLLVM Support non identity layout map for reshape ops in MemRefToLLVM lowering 2022-04-26 13:03:53 -04:00
MemRefToSPIRV [mlir:OpConversionPattern] Add overloads for taking an Adaptor instead of ArrayRef 2021-09-24 17:51:41 +00:00
NVGPUToNVVM [mlir][nvgpu] Move mma.sync and ldmatrix in nvgpu dialect 2022-04-14 23:44:52 +00:00
OpenACCToLLVM [mlir] Finish replacing OwningRewritePatternList with RewritePatternSet 2022-01-26 23:11:02 -08:00
OpenACCToSCF [mlir] Trim a huge number of unnecessary dependencies on the Func dialect 2022-03-01 12:10:04 -08:00
OpenMPToLLVM Added lowering support for atomic read and write constructs 2022-04-21 12:19:13 +05:30
PDLToPDLInterp [mlir:PDL] Remove the ConstantParams support from native Constraints/Rewrites 2022-03-19 13:28:24 -07:00
ReconcileUnrealizedCasts [mlir][NFC] Update the Builtin dialect to use "Both" accessors 2022-03-08 12:25:32 -08:00
SCFToControlFlow [mlir] Split out a new ControlFlow dialect from Standard 2022-02-06 14:51:16 -08:00
SCFToGPU [mlir] Update FuncOp conversion passes to Pass/InterfacePass<FunctionOpInterface> 2022-03-08 12:25:32 -08:00
SCFToOpenMP [MLIR][OpenMP] Place alloca scope within wsloop in scf.parallel to omp lowering 2022-03-02 12:46:58 -05:00
SCFToSPIRV [mlir][SPIRV] Split up StandardToSPIRV now that the Standard dialect is gone 2022-03-02 13:14:36 -08:00
SPIRVCommon [mlir][spirv] Add math.fma lowering to spirv 2022-01-19 10:57:05 -08:00
SPIRVToLLVM [mlir:FunctionOpInterface] Rename the "type" attribute to "function_type" 2022-03-16 17:07:04 -07:00
ShapeToStandard [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
TensorToSPIRV [mlir][SPIRV] Split up StandardToSPIRV now that the Standard dialect is gone 2022-03-02 13:14:36 -08:00
TosaToArith [tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard 2022-04-04 12:22:12 -07:00
TosaToLinalg [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
TosaToSCF [mlir:NFC] Remove the forward declaration of FuncOp in the mlir namespace 2022-04-18 12:01:55 -07:00
TosaToTensor [tosa] Add option to disable tosa.apply_scale lowering in TosaToStandard 2022-04-04 12:22:12 -07:00
VectorToGPU [mlir] Flip Vector dialect accessors used to prefixed form. 2022-03-28 11:24:47 -07:00
VectorToLLVM [mlir][vector][nfc] Rename index optimizations option 2022-03-29 11:33:22 +01:00
VectorToROCDL [mlir] Flip Vector dialect accessors used to prefixed form. 2022-03-28 11:24:47 -07:00
VectorToSCF [mlir] Flip Vector dialect accessors used to prefixed form. 2022-03-28 11:24:47 -07:00
VectorToSPIRV [mlir] Flip Vector dialect accessors used to prefixed form. 2022-03-28 11:24:47 -07:00
CMakeLists.txt [mlir][nvgpu] Move mma.sync and ldmatrix in nvgpu dialect 2022-04-14 23:44:52 +00:00
PassDetail.h [mlir][NFC] Drop a few unnecessary includes from Pass.h 2022-04-07 23:42:47 -07:00