Re: [RFC PATCH 1/8] mm: mmap: unmap large mapping by section

[Yang Shi]

On 3/21/18 3:15 PM, Matthew Wilcox wrote:
> On Wed, Mar 21, 2018 at 02:45:44PM -0700, Yang Shi wrote:
> > On 3/21/18 10:29 AM, Matthew Wilcox wrote:
> > > On Wed, Mar 21, 2018 at 09:31:22AM -0700, Yang Shi wrote:
> > > > On 3/21/18 6:08 AM, Michal Hocko wrote:
> > > > > Yes, this definitely sucks. One way to work that around is to split the
> > > > > unmap to two phases. One to drop all the pages. That would only need
> > > > > mmap_sem for read and then tear down the mapping with the mmap_sem for
> > > > > write. This wouldn't help for parallel mmap_sem writers but those really
> > > > > need a different approach (e.g. the range locking).
> > > > page fault might sneak in to map a page which has been unmapped before?
> > > >
> > > > range locking should help a lot on manipulating small sections of a large
> > > > mapping in parallel or multiple small mappings. It may not achieve too much
> > > > for single large mapping.
> > > I don't think we need range locking.  What if we do munmap this way:
> > >
> > > Take the mmap_sem for write
> > > Find the VMA
> > >     If the VMA is large(*)
> > >       Mark the VMA as deleted
> > >       Drop the mmap_sem
> > >       zap all of the entries
> > >       Take the mmap_sem
> > >     Else
> > >       zap all of the entries
> > > Continue finding VMAs
> > > Drop the mmap_sem
> > >
> > > Now we need to change everywhere which looks up a VMA to see if it needs
> > > to care the the VMA is deleted (page faults, eg will need to SIGBUS; mmap
> > Marking vma as deleted sounds good. The problem for my current approach is
> > the concurrent page fault may succeed if it access the not yet unmapped
> > section. Marking deleted vma could tell page fault the vma is not valid
> > anymore, then return SIGSEGV.
> >
> > > does not care; munmap will need to wait for the existing munmap operation
> > Why mmap doesn't care? How about MAP_FIXED? It may fail unexpectedly, right?
> Oh, I forgot about MAP_FIXED.  Yes, MAP_FIXED should wait for the munmap
> to finish.  But a regular mmap can just pretend that it happened before
> the munmap call and avoid the deleted VMAs.

But, my test shows race condition for reduced size mmap which calls 
do_munmap(). It may need wait for the munmap finish too.

So, in my patches, I just make the do_munmap() called from mmap() hold 
mmap_sem all the time.

Thanks,
Yang