#3209 introduced an unintentional performance regression by removing the `inline` keyword on some functions. https://github.com/NVIDIA/cuda-python/pull/3209#discussion_r4392260454
Activity
leofang commentedon Oct 28, 2026
Closed by #3255