> On Aug 4, 2023, at 3:38 AM, Kees Cook <keesc...@chromium.org> wrote:
>
> On Thu, Aug 03, 2023 at 09:31:24PM +0000, Qing Zhao wrote:
>> So, the basic question is:
>>
>> Given the following:
>>
>> struct fix {
>> int others;
>> int array[10];
>> }
>>
>> extern struct fix * alloc_buf ();
>>
>> int main ()
>> {
>> struct fix *p = alloc_buf ();
>> __builtin_object_size(p->array,0) == ?
>> }
>>
>> Given p->array, can the compiler determine that p points to an object that
>> has TYPE struct fix?
>>
>> If the answer is YES, then the current__builtin_object_size algorithm can be
>> improved to determine __builtin_object_size(p->array, 0) with the TYPE of
>> the struct fix.
>
> I think it is fine to leave __bos(..., 0) as-is. From the Linux kernel's
> use of __bos, we are almost exclusively only interesting the mode 1, not
> node 0. :)
Okay, that’s good to know.
Qing
>
> --
> Kees Cook
