cuda-kat
cuda-kat copied to clipboard
Consider using sized integer types for the builtins
While nVIDIA's own C headers for builtin wrappers use the fundamental types int
, unsigned
, unsigned long long
etc. - the builtins are actually based on exact parameter sizes, not the wishy-washy C integer types. Should we then not make our builtins reflect that, by taking only intNN_t
's and uintNN_t
s as integer parameters?