Gabriel Dos Reis <[EMAIL PROTECTED]> wrote: >>>>> Do you suppose the idiom is common enough that VRP could special-case >>>>> "arrays of size 1 at the end of a struct" ? And still obtain the >>>>> benefits of the optimisation in 99.99% of all >>>>> non-variable-length-tail-array cases? >>>> >>>> It makes sense to me. We could special case "arrays of size 1 at the end >>>> of the struct", and treat it as C99 flexible array members. Any other >>>> case could simply be considered broken. >>> >>> broken with respect to what? >> >> >> broken as in undefined behaviour? > > Could you explain in detail where you see the undefined behaviour?
Accessing the array beyond its size. From the tone of your concise answers, I deduce that this is not undefined behaviour as per the ISO C or ISO C++ standards (otherwise it would have been clear what it is undefined behaviour); in which case, I would appreciate if you could elaborate on when it is invalid to access an array outside its usual range. -- Giovanni Bajo
