Re: [Linux-kernel-mentees] [PATCH net] rds: Prevent kernel-infoleak in rds_notify_queue_get()

[Leon Romanovsky]

On Sun, Aug 02, 2020 at 03:45:40PM -0700, Joe Perches wrote:
> On Sun, 2020-08-02 at 19:28 -0300, Jason Gunthorpe wrote:
> > On Sun, Aug 02, 2020 at 03:23:58PM -0700, Joe Perches wrote:
> > > On Sun, 2020-08-02 at 19:10 -0300, Jason Gunthorpe wrote:
> > > > On Sat, Aug 01, 2020 at 08:38:33AM +0300, Leon Romanovsky wrote:
> > > >
> > > > > I'm using {} instead of {0} because of this GCC bug.
> > > > > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53119
> > > >
> > > > This is why the {} extension exists..
> > >
> > > There is no guarantee that the gcc struct initialization {}
> > > extension also zeros padding.
> >
> > We just went over this. Yes there is, C11 requires it.
>
> c11 is not c90.  The kernel uses c90.

It is not accurate, kernel uses gnu89 dialect, which is C90 with some
C99 features [1]. In our case, we rely on GCC extension {} that doesn't
contradict standart [2] and fills holes with zeros too.

[1] Makefile:500
   496 KBUILD_CFLAGS   := -Wall -Wundef -Werror=strict-prototypes -Wno-trigraphs \
   497                    -fno-strict-aliasing -fno-common -fshort-wchar -fno-PIE \
   498                    -Werror=implicit-function-declaration -Werror=implicit-int \
   499                    -Wno-format-security \
   500                    -std=gnu89

[2] From GCC:
https://gcc.gnu.org/onlinedocs/gcc/C-Dialect-Options.html
"When a base standard is specified, the compiler accepts all programs
following that standard plus those using GNU extensions that do not
contradict it."

Thanks

>
>
>