Skip to content

Coding Style

Chao Liu edited this page Feb 23, 2022 · 19 revisions
  • Integer type

    • only index_t and long_index_t should be used for index calculation in GPU code, int or long is not allowed
    • if integer is used as tensor data, then use int32_t oruint32_t instead of int or unsigned
  • Naming style

    • Use camel case for all Classes, member function, Template argument. NameForClass
    • Use void function_name()
    • class_member_
    • local_variable
  • Must add bracket for all if and else

Clone this wiki locally