On Tue, Nov 20, 2018 at 3:02 AM Nick Desaulniers
<ndesaulni...@google.com> wrote:
>
> On Mon, Nov 19, 2018 at 4:37 AM Luc Van Oostenryck
> <luc.vanoostenr...@gmail.com> wrote:
> >
> > On Mon, Nov 19, 2018 at 07:31:43PM +0900, Masahiro Yamada wrote:
> > > The introduction of these dummy BUILD_BUG_ON stubs dates back to
> > > commit 903c0c7cdc21 ("sparse: define dummy BUILD_BUG_ON definition
> > > for sparse").
> > >
> > > At that time, BUILD_BUG_ON() was implemented with the negative array
> > > trick *and* the link-time trick, like this:
> > >
> > > extern int __build_bug_on_failed;
> > > #define BUILD_BUG_ON(condition) \
> > > do { \
> > > ((void)sizeof(char[1 - 2*!!(condition)])); \
> > > if (condition) __build_bug_on_failed = 1; \
> > > } while(0)
> > >
> > > Sparse is more strict about the negative array trick than GCC because
> > > Sparse requires the array length to be really constant.
> > >
> > > Here is the simple test code for the macro above:
> > >
> > > static const int x = 0;
> > > BUILD_BUG_ON(x);
> > >
> > > GCC is absolutely fine with it (-Wvla was not enabled at that time),
> > > but Sparse warns like this:
> > >
> > > error: bad constant expression
> > > error: cannot size expression
> > >
> > > (If you are using a newer version of Sparse, you will see a different
> > > warning message, "warning: Variable length array is used".)
> > >
> > > Anyway, Sparse was producing many false positive warnings, hence
> > > silenced.
> > >
> > > With the previous commit, the leftover negative array trick is gone.
> > > Sparse is fine with the current BUILD_BUG_ON(), which is implemented
> > > by using the 'error' attribute. (assuming your Sparse version supports
> > > -Wno-unknown-attribute option)
> > >
> > > I am keeping the stub for BUILD_BUG_ON_ZERO(). Otherwise, Sparse
> > > would complain about the following code, which GCC is fine with:
> > >
> > > static const int x = 0;
> > > int y = BUILD_BUG_ON_ZERO(x);
> > >
> > > Signed-off-by: Masahiro Yamada <yamada.masah...@socionext.com>
> > > Acked-by: Kees Cook <keesc...@chromium.org>
> >
> > Reviewed-by: Luc Van Oostenryck <luc.vanoostenr...@gmail.com>
>
> Clang builds not affected. Tested a quick arm64 defconfig build with
> Clang + this patch.
> Reviewed-by: Nick Desaulniers <ndesaulni...@google.com>
> Tested-by: Nick Desaulniers <ndesaulni...@google.com>
>
This patch can go in only when 1/3 is acceptable.
But, I see 1/3 is controversial.
--
Best Regards
Masahiro Yamada
