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

Andy Shevchenko andy.shevchenko at gmail.com
Fri Jul 31 07:05:04 UTC 2020


Sakari, JFYI. I remember during some reviews we have a discussion
about {0} vs {} and surprisingly they are not an equivalent.

On Fri, Jul 31, 2020 at 10:00 AM Leon Romanovsky <leon at kernel.org> wrote:
> On Fri, Jul 31, 2020 at 09:29:27AM +0300, Andy Shevchenko wrote:
> > On Friday, July 31, 2020, Greg Kroah-Hartman <gregkh at linuxfoundation.org>
> > wrote:
> > > On Fri, Jul 31, 2020 at 07:33:06AM +0200, Greg Kroah-Hartman wrote:
> > > > On Fri, Jul 31, 2020 at 07:53:01AM +0300, Leon Romanovsky wrote:
> > > > > On Thu, Jul 30, 2020 at 03:20:26PM -0400, Peilin Ye wrote:

...

> > > > > Of course, this is the difference between "{ 0 }" and "{}"
> > > initializations.
> > > >
> > > > Really?  Neither will handle structures with holes in it, try it and
> > > > see.
> >
> >
> > {} is a GCC extension, but I never thought it works differently.
>
> Yes, this is GCC extension and kernel relies on them very heavily.

I guess simple people who contribute to the kernel just haven't
realized (yet) that it's an extension and that's why we have plenty of
{} and {0} in the kernel.

> > > And if true, where in the C spec does it say that?


-- 
With Best Regards,
Andy Shevchenko


More information about the Linux-kernel-mentees mailing list