CUDASecurity

General CUDA security rules

Nested Rules

CUDASecurity-ARR01	Do not form or use out-of-bounds array subscripts for CUDA kernels
CUDASecurity-CON01	Prevent data races when accessing bit-fields from multiple threads
CUDASecurity-CON02	When data must be accessed by multiple threads, provide a mutex and guarantee no adjacent data is also accessed
CUDASecurity-CON03	Avoid race conditions when using library functions
CUDASecurity-CON04	Do not allow data races in multithreaded code
CUDASecurity-CON05	Avoid deadlock by locking in a predefined order
CUDASecurity-CON06	Wrap functions that can fail spuriously in a loop
CUDASecurity-DCL01	Do not declare an identifier with conflicting Memory attribute specifiers
CUDASecurity-DCL02	Avoid information leakage when passing a structure across a trust boundary
CUDASecurity-ERR01	Detect and handle CUDA runtime API errors
CUDASecurity-EXP01	Do not call va_arg with an argument of the incorrect type
CUDASecurity-EXP02	Do not rely on side effects in operands to sizeof, _Alignof, or _Generic
CUDASecurity-MSC01	Do not call va_arg() on a va_list that has an indeterminate value

Options

• Setting an option for this rule means setting the default for all nested rules.

• This rule shares the following common options: exclude_in_macros, exclude_messages_in_system_headers, excludes, extend_exclude_to_macro_invocations, includes, justification_checker, languages, post_processing, provider, report_at, severity

• The following places define options that affect this rule: Stylechecks, Analysis-GlobalOptions

• This rule has no individual options.