On 2014-05-14 16:04:47, tytso@xxxxxxx wrote:
> On Wed, May 14, 2014 at 11:02:47AM -0400, Chris Mason wrote:
> > >> linux-fsdevel might seem as a good candidate for it, but still I
> > >> think that it deserves a separate ML to point people to.
> I'm personally in favor of using linux-fsdevel since it might
> encourage more fs developers who aren't using xfstests yet to start
> using it.
> For example, we started investigating using xfstests to test unionfs,
> and pretty quickly found problems. (I suspect the same problem exists
> in AUFS, BTW, but I've been focusing on unionfs because it's simpler
> and less scary.) The patches to enable the use of xfstests to test
> unionfs are still pretty rough, but hopefully we'll get those sent to
> Dave once they are cleaned up a bit.
Oh, that sounds interesting. I haven't seen these patches, but I expect
they would be pretty easy for me to extend for testing eCryptfs. That
has been on my todo list for a long time but I haven't spent much time
working on eCryptfs lately.
BTW, you can use this email as a supporting data point for your first
Description: Digital signature