Chris Johns commented on a discussion on cpukit/include/rtems/cpuuse_backtrace.h: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/843#note_136895 > + */ > +typedef struct { > + /** > + * @brief Frame address (program counter). > + */ > + uintptr_t address; > + > + /** > + * @brief Resolved symbol name (32 bytes, null-terminated). > + */ > + char symbol_name[32]; > + > + /** > + * @brief Object/module name containing the symbol (32 bytes, > null-terminated). > + */ > + char object_name[32]; If the module has been deleted this symbol would not be valid. And if a new module is loaded with a different symbol would there be a clash. I suppose the nature of this data is determined by the life time of the struct. If it is transient then maybe copies are OK? This however comes back to the original issue of length and very long symbols. -- View it on GitLab: https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/843#note_136895 You're receiving this email because of your account on gitlab.rtems.org.
_______________________________________________ bugs mailing list [email protected] http://lists.rtems.org/mailman/listinfo/bugs
