On Fri, 2007-10-12 at 04:14 +1000, Nick Piggin wrote:
quoted text > On Friday 12 October 2007 20:37, Peter Zijlstra wrote:
> > On Fri, 2007-10-12 at 02:57 +1000, Nick Piggin wrote:
> > > On Friday 12 October 2007 19:03, Peter Zijlstra wrote:
> > > > Subject: mm: avoid dirtying shared mappings on mlock
> > > >
> > > > Suleiman noticed that shared mappings get dirtied when mlocked.
> > > > Avoid this by teaching make_pages_present about this case.
> > > >
> > > > Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > > > Acked-by: Suleiman Souhlal <suleiman@google.com>
> > >
> > > Umm, I don't see the other piece of this thread, so I don't
> > > know what the actual problem was.
> > >
> > > But I would really rather not do this. If you do this, then you
> > > now can get random SIGBUSes when you write into the memory if it
> > > can't allocate blocks or ... (some other filesystem specific
> > > condition).
> >
> > I'm not getting this, make_pages_present() only has to ensure all the
> > pages are read from disk and in memory. How is this different from a
> > read-scan?
>=20
> I guess because we've mlocked a region that has PROT_WRITE access...
> but actually, I suppose mlock doesn't technically require that we
> can write to the memory, only that the page isn't swapped out.
>=20
> Still, it is nice to be able to have a reasonable guarantee of
> writability.
>=20
>=20
> > The pages will still be read-only due to dirty tracking, so the first
> > write will still do page_mkwrite().
>=20
> Which can SIGBUS, no?
Sure, but that is no different than any other mmap'ed write. I'm not
seeing how an mlocked region is special here.
I agree it would be nice if mmap'ed writes would have better error
reporting than SIGBUS, but such is life.
