Re: [PATCH v2 00/15] Avoid MAP_FIXED gap exposure

[Bert Karwatzki <spasswolf@web.de>]

* Andrew Morton <akpm@linux-foundation.org> [240626 16:59]:
> On Tue, 25 Jun 2024 15:11:30 -0400 "Liam R. Howlett" <Liam.Howlett@oracle.com> wrote:
>
> > It is now possible to walk the vma tree using the rcu read locks and is
> > beneficial to do so to reduce lock contention.  Doing so while a
> > MAP_FIXED mapping is executing means that a reader may see a gap in the
> > vma tree that should never logically exist - and does not when using the
> > mmap lock in read mode.  The temporal gap exists because mmap_region()
> > calls munmap() prior to installing the new mapping.
>
> What are the consequences when this race hits?  IOW, why do we need to
> change anything?
>

If I understand this correctly the plan is to replace mmap_read_lock(mm) by
rcu_read_lock(). So the consequences of a visible gap could be tested by
replacing mmap_read_lock(mm) by rcu_read_lock() within the old code. If this is
the case I'm willing to test it.

Bert Karwatzki