zhangyiss/mlx
1
0
Fork 0
mirror of https://github.com/ml-explore/mlx.git synced 2025-12-16 01:49:05 +08:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Faster general unary op (#2472)

Browse Source 
* faster general unary op

* faster general ops + reorg

* fix + comment

* binary two

* copy general
This commit is contained in:
Awni Hannun
2025-08-15 15:04:12 -07:00
committed by GitHub
parent dfb5022eab
commit 6441c21a94
62 changed files with 1215 additions and 203 deletions
Download Patch File Download Diff File Expand all files Collapse all files

27
mlx/backend/cuda/binary/bitwise_binary.cu Normal file
View File

@@ -0,0 +1,27 @@
// Copyright © 2025 Apple Inc.
#include "mlx/backend/cuda/binary/binary.cuh"
namespace mlx::core {
void BitwiseBinary::eval_gpu(const std::vector<array>& inputs, array& out) {
nvtx3::scoped_range r("BitwiseBinary::eval_gpu");
auto& s = out.primitive().stream();
switch (op_) {
case BitwiseBinary::And:
binary_op_gpu<cu::BitwiseAnd>(inputs, out, name(), s);
break;
case BitwiseBinary::Or:
binary_op_gpu<cu::BitwiseOr>(inputs, out, name(), s);
break;
case BitwiseBinary::Xor:
binary_op_gpu<cu::BitwiseXor>(inputs, out, name(), s);
break;
case BitwiseBinary::LeftShift:
binary_op_gpu<cu::LeftShift>(inputs, out, name(), s);
break;
case BitwiseBinary::RightShift:
binary_op_gpu<cu::RightShift>(inputs, out, name(), s);
break;
}
}
} // namespace mlx::core