Re: [PATCH]: v9fs: add readpage support

[Eric Van Hensbergen <ericvh@gmail.com>]

On 1/13/06, Andrew Morton <akpm@osdl.org> wrote:
> Eric Van Hensbergen <ericvh@gmail.com> wrote:
> >
> > I went looking for an example of how to do this better.  More or less,
> > v9fs reads and writes are similar to DirectIO since they don't go
> > through the page-cache.
>
> hm.  Why not?
>

At the moment we'd rather not cache anything with v9fs as it hides
operations from the servers, and in the case of synthetic servers it
can be a bad thing.  There is a somewhat well-understood strategy for
how to do cacheing of static files in a sane manner under 9P, but we
were holding off on trying to move that into v9fs for the time being. 
The only reason we added the read-only mmap support back in was to
support users who were trying to access executables over 9P
connections.

> >
> > Now, that being said, it still seems to me to be a bit heavy weight --
> > do folks have a better pointer to code that I can use as an example of
> > how to do this more efficiently?
>
> Not really.  If that's what you need to do then that's the way to do it.
> We've had nasty races and other problems wrt invalidate_inode_pages2 and
> pagefaults, so I suggest you test that mix carefully.
>
> Have you tried fsx-linux?  It's really good for finding data integrity
> bugs.  There's a copy in
> http://www.zip.com.au/~akpm/linux/patches/stuff/ext3-tools.tar.gz.
>

We've been regression testing with vanilla fsx, I'll upgrade to
fsx-linux and make sure we are clean.

>
> I'd suggest that you want the mapping->nrpages test - it'll be a useful
> speedup in the common case.
>

Yeah, as I was tracing through the invalidate_inode_pages2() this
morning I realized this is probably a good idea.  Should have a new
patch to you by the end of the weekend.

          -eric