You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
CUDA provides only limited vector widths and only for variable
width integer types. We use our own vector type and some C++
templates to get more flexible vectors. We aim to get 128bit loads
by adjusting the width based on the type size.
Signed-off-by: Joseph Schuchart <[email protected]>
0 commit comments