Re: [RFC PATCH v8 2/2] mm: introduce MAP_SHARED_VALIDATE, a mechanism to safely define new mmap flags

[Jan Kara <jack-AlSwsSmVLrQ@public.gmane.org>]

On Mon 11-09-17 13:10:30, Christoph Hellwig wrote:
> On Mon, Sep 11, 2017 at 11:47:14AM +0200, Jan Kara wrote:
> > On Fri 08-09-17 12:35:13, Dan Williams wrote:
> > > The mmap(2) syscall suffers from the ABI anti-pattern of not validating
> > > unknown flags. However, proposals like MAP_SYNC and MAP_DIRECT need a
> > > mechanism to define new behavior that is known to fail on older kernels
> > > without the support. Define a new MAP_SHARED_VALIDATE flag pattern that
> > > is guaranteed to fail on all legacy mmap implementations.
> > > 
> > > With this in place new flags can be defined as:
> > > 
> > >     #define MAP_new (MAP_SHARED_VALIDATE | val)
> > 
> > Is this changelog stale? Given MAP_SHARED_VALIDATE will be new mapping
> > type, I'd expect we define new flags just as any other mapping flags...
> > I see no reason why MAP_SHARED_VALIDATE should be or'ed to that.
> 
> Btw, I still think it should be a new hidden flag and not a new mapping
> type.  I brought this up last time, so maybe I missed the answer
> to my concern.

Hum, I don't remember your concern and the only comment from you to the
last posting which I've found is:

"no mmap3 syscall here :)

Do you also need to update the nommu mmap implementation?"

So I guess it's something else. So can you remind me or send a link?
Thanks!

							Honza
-- 
Jan Kara <jack-IBi9RG/b67k@public.gmane.org>
SUSE Labs, CR