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/Conversion/TorchToTosa/cast_fp32_to_fp16.mlir

11 lines
609 B
MLIR
Raw Normal View History

allow tosa.cast to convert from f32 to f16 (#2934) According to the [official TOSA spec](https://www.mlplatform.org/tosa/tosa_spec.html#_cast), `tosa.cast` allows a cast from `fp32` to `fp16`. We were not previously accounting for this in the `TorchToTosa` lowering. Also did a tiny bit of cleanup in the code to make it easier to spot which conversions are currently allowed. --------- Co-authored-by: Srinath Avadhanula <srinath.avadhanula@getcruise.com>
2024-02-21 06:22:38 +08:00
// RUN: torch-mlir-opt <%s -convert-torch-to-tosa -split-input-file
[NFC reformat] Run pre-commit on all files and format misc. This is part 1 of ~3, formatting all miscellaneous text files and CPP files matched by a first run of pre-commit. These tend to be low change-traffic and are likely not disruptive. Subsequent patches will format Python files and remaining CPP files.
2024-04-28 05:08:09 +08:00
// CHECK: %{{.*}} = tosa.cast %{{.*}} : (tensor<1x32x220x220xf32>) -> tensor<1x32x220x220xf16>
allow tosa.cast to convert from f32 to f16 (#2934) According to the [official TOSA spec](https://www.mlplatform.org/tosa/tosa_spec.html#_cast), `tosa.cast` allows a cast from `fp32` to `fp16`. We were not previously accounting for this in the `TorchToTosa` lowering. Also did a tiny bit of cleanup in the code to make it easier to spot which conversions are currently allowed. --------- Co-authored-by: Srinath Avadhanula <srinath.avadhanula@getcruise.com>
2024-02-21 06:22:38 +08:00
func.func @forward(%arg0: !torch.vtensor<[1,32,220,220],f32>) -> !torch.vtensor<[1,32,220,220],f16> {
%int5 = torch.constant.int 5
%false = torch.constant.bool false
%none = torch.constant.none
%out = torch.aten.to.dtype %arg0, %int5, %false, %false, %none : !torch.vtensor<[1,32,220,220],f32>, !torch.int, !torch.bool, !torch.bool, !torch.none -> !torch.vtensor<[1,32,220,220],f16>
return %out : !torch.vtensor<[1,32,220,220],f16>
}