mirror of
https://github.com/ml-explore/mlx.git
synced 2025-12-16 01:49:05 +08:00
Faster general unary op (#2472)
* faster general unary op * faster general ops + reorg * fix + comment * binary two * copy general
This commit is contained in:
15
mlx/backend/cuda/binary/equal.cu
Normal file
15
mlx/backend/cuda/binary/equal.cu
Normal file
@@ -0,0 +1,15 @@
|
||||
// Copyright © 2025 Apple Inc.
|
||||
|
||||
#include "mlx/backend/cuda/binary/binary.cuh"
|
||||
|
||||
namespace mlx::core {
|
||||
void Equal::eval_gpu(const std::vector<array>& inputs, array& out) {
|
||||
nvtx3::scoped_range r("Equal::eval_gpu");
|
||||
auto& s = out.primitive().stream();
|
||||
if (equal_nan_) {
|
||||
binary_op_gpu<cu::NaNEqual>(inputs, out, name(), s);
|
||||
} else {
|
||||
binary_op_gpu<cu::Equal>(inputs, out, name(), s);
|
||||
}
|
||||
}
|
||||
} // namespace mlx::core
|
||||
Reference in New Issue
Block a user