On Sat, Jun 13, 2020 at 6:00 PM Al Viro <viro@xxxxxxxxxxxxxxxxxx> wrote:
> On Sat, Jun 13, 2020 at 04:41:18PM +0100, Al Viro wrote:
> > On Sat, Jun 13, 2020 at 04:31:02PM +0100, Al Viro wrote:
> > > On Sat, Jun 13, 2020 at 07:12:36PM +0530, afzal mohammed wrote:
> > > > Hi,
> > > >
> > > > On Sat, Jun 13, 2020 at 01:56:15PM +0100, Al Viro wrote:
> > > >
> > > > > Incidentally, what about get_user()/put_user()? _That_ is where it's
> > > > > going to really hurt...
> > > >
> > > > All other uaccess routines are also planned to be added, posting only
> > > > copy_{from,to}_user() was to get early feedback (mentioned in the
> > > > cover letter)
> > >
> > > Sure, but what I mean is that I'd expect the performance loss to be
> > > dominated by that, not by copy_from_user/copy_to_user on large amounts
> > > of data. Especially on the loads like kernel builds - a lot of stat()
> > > and getdents() calls there.
> >
> > To clarify: stat() means small copy_to_user(), getdents() - a mix of
> > put_user() and small copy_to_user(). I would be very surprised if it
> > does not hurt a lot.
>
> PS: there's another fun issue here:
>
> fill a file with zeroes
> mmap that file in two areas, MAP_SHARED
> thread 1:
> munmap() the first area
> fill the second one with 'X'
> thread 2:
> write() from the first area into pipe
>
> One could expect that nothing by zeroes gets written into
> pipe - it might be a short write() (or -EFAULT), but finding
> any 'X' there would be a bug.
>
> Your patches allow for a possibility of write() doing
> get_user_pages_fast(), getting the first page just as
> munmap() is about to remove it from page tables and bugger
> off. Then thread 1 proceeds with the store (via the
> second area). And then thread 2 does memcpy() from that
> thing via a kmap_atomic()-created alias, observing the
> effect of the store.
>
> That might or might not be a POSIX violation, but it does
> look like a QoI issue...
I assume this problem exists in arch/um/kernel/skas/uaccess.c
and in Ingo's old x86 VMSPLIT_4G_4G patch as well, right?
I guess holding mmap_read_lock() would prevent it but make
it even more expensive.
Arnd
