mirror of https://github.com/llvm/torch-mlir
23 lines
1.1 KiB
MLIR
23 lines
1.1 KiB
MLIR
// RUN: npcomp-opt -lower-to-hybrid-tensor-memref-pipeline <%s | FileCheck %s --dump-input=fail
|
|
|
|
#map0 = affine_map<(d0) -> (d0)>
|
|
func @f(%arg0: tensor<?xf32>, %arg1: tensor<?xf32>) -> tensor<?xf32> {
|
|
%0 = "shape.shape_of"(%arg0) : (tensor<?xf32>) -> !shape.shape
|
|
%1 = "shape.shape_of"(%arg1) : (tensor<?xf32>) -> !shape.shape
|
|
%2 = "shape.broadcast"(%0, %1) : (!shape.shape, !shape.shape) -> !shape.shape
|
|
%3 = "shape.abort_if_error"(%2) : (!shape.shape) -> none
|
|
%4 = "tcp.island"(%3) ( {
|
|
%5 = "tcp.broadcast_to"(%arg0, %2) : (tensor<?xf32>, !shape.shape) -> tensor<?xf32>
|
|
%6 = "tcp.broadcast_to"(%arg1, %2) : (tensor<?xf32>, !shape.shape) -> tensor<?xf32>
|
|
// CHECK: alloc_memref
|
|
%7 = linalg.generic {args_in = 2 : i64, args_out = 1 : i64, indexing_maps = [#map0, #map0, #map0], iterator_types = ["parallel"]} %5, %6 {
|
|
^bb0(%arg2: f32, %arg3: f32): // no predecessors
|
|
%8 = addf %arg2, %arg3 : f32
|
|
linalg.yield %8 : f32
|
|
}: tensor<?xf32>, tensor<?xf32> -> tensor<?xf32>
|
|
"tcp.yield"(%7) : (tensor<?xf32>) -> ()
|
|
}) : (none) -> tensor<?xf32>
|
|
return %4 : tensor<?xf32>
|
|
}
|
|
|