Alexander Viro wrote:
>
> On Tue, 18 Apr 2000, Chris Mason wrote:
>
> > > ObReiserfs: what about readdir() vs. rebalancing and friends?
> >
> > I think you are asking about our code that detects when the tree has been
> > balanced underneath us, and starts the search over, right?
>
> Yes.
>
> > 3.6.4 is a bit cleaner than the code you read, but we haven't added the
> > tree locking yet.
Chris, you should make clear in this that what we currently have works, it just
uses overly large grained locks.
A casual reader could think otherwise.:-)
>
> Aha. I'll look at 3.6.4 when I'll get some sleep, OK?
>
> > It is in the list, which we are still prioritizing ;-)
>
> ... same for fhandle_to_dentry() here ;-)
In our case, we are choosing between having Alexei work on knfsd interaction
with reiserfs, and implementing seals. Seals reduce the granularity of locks in
reiserfs to per buffer granularity and avoid holding any locks for the length of
a disk I/O. I can describe seals if you or someone else is interested, but
first I must get some sleep.:-)
Hans
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 23 2000 - 21:00:14 EST