[linux-cifs-client] Re: [PATCH] cifs: make cifs_writepages use longer timeout

[Jeff Layton]

On Thu, 2 Apr 2009 13:15:39 -0500
Steve French <smfrench at gmail.com> wrote:

> When file is extended in write we could save the (previous) offset in
> cifs_file struct ...?  And then increase the saved offset to match the
> last write (past previous end of file).   Whenever we attempt a write
> more than a few megabytes past the saved range, we can use long op?
> Thoughts?
> 

Something like that might work. Basically I think we need to track the
current size as reported by the server separate from the i_size.

It's probably best suited to be in the CIFS_I struct rather than the
open file info. After a successful write past that value we'd update it.
We'd also update it when we get new FILE_*_INFO from the server.

Then we can use that value to decide how to time the op. At some
threshold we'll switch from a LONG to VLONG op.

I'll see what I can come up with...

-- 
Jeff Layton <jlayton at redhat.com>