Oleg Shyshkov cad08503b8 [mlir][memref] Lower copy of memrefs with outer size-1 dims to intrinsic memcpy. ...

With this change, more `memref.copy` will be lowered to the efficient `memcpy`. For example,

```
memref.copy %subview, %alloc : memref<1x576xf32, strided<[704, 1]>> to memref<1x576xf32>
```

Differential Revision: https://reviews.llvm.org/D150448

2023-05-12 17:18:13 +02:00

..

convert-alloca-scope.mlir

[mlir][MemRef] Add option to -finalize-memref-to-llvm to emit opaque pointers

2023-02-08 10:15:44 +01:00

convert-dynamic-memref-ops.mlir

Add a lowering for memref.dealloc with unranked memrefs.

2023-02-16 14:19:16 +01:00

convert-static-memref-ops.mlir

[mlir][MemRef|Tensor] Fix the handling of DimOp

2023-02-16 11:38:19 +01:00

expand-then-convert-to-llvm.mlir

[MemRefToLLVM] Add a method in MemRefDescriptor to get the buffer addr

2023-04-26 06:12:18 +02:00

generic-functions.mlir

[mlir][MemRef] Add option to -finalize-memref-to-llvm to emit opaque pointers

2023-02-08 10:15:44 +01:00

invalid.mlir

Add generic type attribute mapping infrastructure, use it in GpuToX

2023-02-09 18:00:46 +00:00

memref-to-llvm.mlir

[mlir][memref] Lower copy of memrefs with outer size-1 dims to intrinsic memcpy.

2023-05-12 17:18:13 +02:00

typed-pointers.mlir

[mlir] Handle different pointer sizes in unranked memref descriptors

2023-02-09 19:14:58 +00:00