Re: Lease semantic proposal

[Dan Williams]

On Tue, Oct 1, 2019 at 2:02 PM Ira Weiny <ira.weiny@xxxxxxxxx> wrote:
>
> On Mon, Sep 30, 2019 at 06:42:33PM +1000, Dave Chinner wrote:
> > On Wed, Sep 25, 2019 at 04:46:03PM -0700, Ira Weiny wrote:
> > > On Tue, Sep 24, 2019 at 08:26:20AM +1000, Dave Chinner wrote:
> > > > Hence, AFIACT, the above definition of a F_RDLCK|F_LAYOUT lease
> > > > doesn't appear to be compatible with the semantics required by
> > > > existing users of layout leases.
> > >
> > > I disagree.  Other than the addition of F_UNBREAK, I think this is consistent
> > > with what is currently implemented.  Also, by exporting all this to user space
> > > we can now write tests for it independent of the RDMA pinning.
> >
> > The current usage of F_RDLCK | F_LAYOUT by the pNFS code allows
> > layout changes to occur to the file while the layout lease is held.
>
> This was not my understanding.

I think you guys are talking past each other. F_RDLCK | F_LAYOUT can
be broken to allow writes to the file / layout. The new unbreakable
case would require explicit SIGKILL as "revocation method of last
resort", but that's the new incremental extension being proposed. No
changes to the current behavior of F_RDLCK | F_LAYOUT.

Dave, the question at hand is whether this new layout lease mode being
proposed is going to respond to BREAK_WRITE, or just BREAK_UNMAP. It
seems longterm page pinning conflicts really only care about
BREAK_UNMAP where pages that were part of the file are being removed
from the file. The unbreakable case can tolerate layout changes that
keep pinned pages mapped / allocated to the file.