On Tue, 5 Mar 2019, Geert Uytterhoeven wrote:
On Tue, Mar 5, 2019 at 3:58 AM Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx> wrote:
On Tue, 5 Mar 2019, Finn Thain wrote:
Looks bogus to me.
If you change memcpy to __builtin_memcpy, then we avoid the macro and the
warning changes to,
./include/linux/string.h:456:3: warning: '__builtin_memcpy' forming offset [7, 8] is out of the bounds [0, 6] [-Warray-bounds]
__builtin_memcpy(dest, src, dest_len);
The compiler has nothing to complain about here. dest is known to be
id->fr and dest_len is known to be sizeof(id->fr).
The error message indicates that gcc has applied the bounds [0, 6] to dest
when in fact those are the bounds for src.
My mistake. GCC is right, it seems memcpy will read past the end of
"5.0.0+".
But only if the else branch is taken, which is not the case.
You and I know that, because we can see what values get passed to
memcpy_and_pad(). But how is gcc to know that?
If we replace strlen with __builtin_strlen, this problem goes away. It's
interesting that the kernel's strlen implementation in
include/linux/string.h can't achieve this.
--
Gr{oetje,eeting}s,
Geert
