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

Add contiguous_copy_gpu util for copying array (#2379)

Browse Source
This commit is contained in:
Cheng
2025-07-18 22:44:25 +09:00
committed by GitHub
parent 31fc530c76
commit 45adec102c
20 changed files with 40 additions and 67 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
mlx/backend/cuda/logsumexp.cu
View File

@@ -108,8 +108,7 @@ void LogSumExp::eval_gpu(const std::vector<array>& inputs, array& out) {
if (x.flags().contiguous && x.strides()[x.ndim() - 1] == 1) {
return x;
} else {
auto x_copy = array(x.shape(), x.dtype(), nullptr, {});
copy_gpu(x, x_copy, CopyType::General, s);
array x_copy = contiguous_copy_gpu(x, s);
encoder.add_temporary(x_copy);
return x_copy;
}