Mapping of exception masks between source and target architectures with different
numbers of exception masks enables a binary translator to translate code from the
source to the target architecture and to determine an appropriate state for the
source architecture if an exception is raised when executing the translated code.