> Your change also did not do any bounds checking at all, so how is this > now safer? > > confused, > > greg k-h I assumed bounds checking wasn't necessary here because obj is allocated at the start of the function with enough space (sizeof(struct bpf_object) + strlen(path) + 1). My main motivation for the change was the deprecation of strcpy(). However, thinking about it now, I'm not entirely sure memcpy is even needed in this context. I'd really appreciate any feedback or clarification on the best approach here.
