Re: nfs invalidates lose pte dirty bits

[Andrea Arcangeli <andrea@suse.de>]

On Fri, Dec 23, 2005 at 09:41:55AM +0100, Trond Myklebust wrote:
> On Fri, 2005-12-23 at 03:36 +0100, Andrea Arcangeli wrote:
> > On Thu, Dec 22, 2005 at 06:30:49PM -0500, Trond Myklebust wrote:
> > > See the latest git release where we introduce the nfs_sync_mapping()
> > > helper.
> > 
> > So you also still break completely with threaded programs, did you
> > consider that while fixing the most obvious problem? Isn't that a
> > problem too? What about my suggestion of invalidate_inode_clean_pages?
> 
> It is only a problem when doing mmap writes. In the case of ordinary

Yes, those changes are all about mmap writes.

> However if the user is doing mmap writes while the file is in the
> process of being modified on the server, then they are doing something
> wrong anyway. The small race between nfs_sync_mapping() and
> invalidate_inode_pages2() is the least of their problems.

I'm talking about spurious revalidates, I don't think the testcase I'm
dealing with is really needing an invalidate, it's a spurious one
(perhaps triggered by flock), but I'm lucky it's single threaded so
current fix will work for them.