Skip to content

to make TIE quantized linear operator to fall back to nnlib signed kernel for shapes not supported by the TIE kernel. #10637

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
May 2, 2025
Merged

to make TIE quantized linear operator to fall back to nnlib signed kernel for shapes not supported by the TIE kernel. #10637

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
26 changes: 26 additions & 0 deletions backends/cadence/hifi/operators/operators.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -36,6 +36,32 @@ ::executorch::aten::Tensor& div_out_mode(
::executorch::aten::optional<::executorch::aten::string_view> mode,
::executorch::aten::Tensor& out);

void quantized_linear_out(
__ET_UNUSED ::executorch::runtime::KernelRuntimeContext& ctx,
const ::executorch::aten::Tensor& in,
const ::executorch::aten::Tensor& weight,
const ::executorch::aten::Tensor& bias,
int64_t in_zero_point,
const ::executorch::aten::Tensor& weight_zero_point,
const ::executorch::aten::Tensor& out_multiplier,
const ::executorch::aten::Tensor& out_shift,
int64_t out_zero_point,
__ET_UNUSED const ::executorch::aten::optional<::executorch::aten::Tensor>& offset,
::executorch::aten::Tensor& out);

void quantized_linear_per_tensor_out(
__ET_UNUSED ::executorch::runtime::KernelRuntimeContext& ctx,
const ::executorch::aten::Tensor& in,
const ::executorch::aten::Tensor& weight,
const ::executorch::aten::Tensor& bias,
int64_t in_zero_point,
int64_t weight_zero_point,
int64_t out_multiplier,
int64_t out_shift,
int64_t out_zero_point,
__ET_UNUSED const ::executorch::aten::optional<::executorch::aten::Tensor>& offset,
::executorch::aten::Tensor& out);

} // namespace native
} // namespace HiFi
} // namespace impl
Expand Down
Loading