llvm
/
torch-mlir
mirror of https://github.com/llvm/torch-mlir
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
torch-mlir/test/Dialect/Numpy/ops.mlir

31 lines
1.6 KiB
MLIR
Raw Normal View History

Add builtin_ufunc and generic_ufunc ops.
2020-04-29 11:32:49 +08:00
// RUN: npcomp-opt -split-input-file %s | npcomp-opt | FileCheck --dump-input=fail %s
Add !numpy.any_dtype dialect type.
2020-04-30 09:20:42 +08:00
Add builtin_ufunc and generic_ufunc ops.
2020-04-29 11:32:49 +08:00
// -----
// CHECK-LABEL: @builtin_ufunc
Refactor old tracing tests and remove deprecated ops. * Old doctests to run under lit. * Old custom filecheck tests -> pytest directory (under lit). * Rename some old ufunc ops in the tracer.
2020-06-30 07:19:03 +08:00
func @builtin_ufunc(%arg0 : tensor<3xf64>, %arg1 : tensor<3xf64>) -> tensor<3xf64> {
%0 = numpy.builtin_ufunc_call<"numpy.add"> (%arg0, %arg1) : (tensor<3xf64>, tensor<3xf64>) -> tensor<3xf64>
return %0 : tensor<3xf64>
Add numpy.ufunc_call op.
2020-04-30 08:49:56 +08:00
}
Add support for "trailing_" and "out" variants of various ops. We already had the `promoteTrailingOutTensor` flag, but weren't using it. A inplaceVariantKernelName flag needed to be added. This change is a little dissatisfying, as the conversions done by the RecognizeKernelsPass are currently non-orthogonal. In particular, `kDropResultAndAliasArg0` probably won't work as intended if mixed with these (we probably need to promote kDropResultAndAliasArg0 to not be an arg-level thing anyway, as we have done with promoteTrailingOutTensor). This involved adding a new op `numpy.overwrite_array`. ``` numpy.overwrite_array %arg2 overwrites %arg0 : tensor<2x3xf32>, !numpy.ndarray<[2,3]:f32> ``` This models the destructive update behavior. Note that in the above op, we cannot simply RAUW %arg0 with a suitably conveted %arg2 (for example, %arg0 might have uses that are not dominated by %arg2, or might have an alias relation with some other array in the program). In general, we need a pass analogous to "SSA-formation" which knows how to see through these to uncover an underlying tensor program. Also, add tanh_out_e2e.py/div_inplace_e2e.py and fix some bitrot in refjit.py which is my running example I'm trying to get working.
2021-03-19 04:13:40 +08:00
// CHECK-LABEL: @ndarray_tensor_bridging
func @ndarray_tensor_bridging(%arg0: !numpy.ndarray<[2,3]:f32>, %arg1: !numpy.ndarray<[2,3]:f32>, %arg2: tensor<2x3xf32>) {
// CHECK-NEXT: numpy.copy_to_tensor
%t = numpy.copy_to_tensor %arg1 : (!numpy.ndarray<[2,3]:f32>) -> tensor<2x3xf32>
// CHECK-NEXT: numpy.create_array_from_tensor
%a = numpy.create_array_from_tensor %arg2 : (tensor<2x3xf32>) -> !numpy.ndarray<[2,3]:f32>
// CHECK-NEXT: numpy.overwrite_array
numpy.overwrite_array %arg2 overwrites %arg0 : tensor<2x3xf32>, !numpy.ndarray<[2,3]:f32>
return
}
Add torch-adjust-calling-conventions pass. This pass incorporates torch.type_bound info and also removes NoneType returns (eventually it will rewrite tuple types too, but can't yet because !basicpy.TupleType doesn't track element types). Recommend looking at adjust-calling-conventions.mlir first to see what it is doing, and holding your nose for the implementation of the pass. I decided to implement this with the conversion framework, because it gives us *some* goodies for type conversion -- mainly avoiding large amounts of tricky RAUW dances. Unfortunately, the conversion framework isn't a perfect fit for a couple reasons: - the incorporation of torch.type_bound is a context-sensitive rewrite (requires looking at the arg attr, not just the type). - NoneType conversion is 1->0, which requires some special handling - (not implemented yet) 1->N tuple type conversions require special handling. It's a little bit scary, but on balance doing it the other way would have its own downsides.
2021-04-02 08:36:18 +08:00
// CHECK-LABEL: @static_info_cast
func @static_info_cast(%arg0: !numpy.ndarray<[2,3]:f32>, %arg1: !numpy.ndarray<[?,3]:f32>, %arg2: !numpy.ndarray<*:f32>) {
// CHECK-NEXT: numpy.static_info_cast %arg0 : !numpy.ndarray<[2,3]:f32> to !numpy.ndarray<*:!numpy.any_dtype>
%0 = numpy.static_info_cast %arg0 : !numpy.ndarray<[2,3]:f32> to !numpy.ndarray<*:!numpy.any_dtype>
// CHECK-NEXT: numpy.static_info_cast %arg1 : !numpy.ndarray<[?,3]:f32> to !numpy.ndarray<[7,3]:f32>
%1 = numpy.static_info_cast %arg1 : !numpy.ndarray<[?,3]:f32> to !numpy.ndarray<[7,3]:f32>
// CHECK-NEXT: numpy.static_info_cast %arg2 : !numpy.ndarray<*:f32> to !numpy.ndarray<[?,?]:f32>
%2 = numpy.static_info_cast %arg2 : !numpy.ndarray<*:f32> to !numpy.ndarray<[?,?]:f32>
return
}