On Tue, Mar 01, 2016 at 09:24:11PM -0800, Darrick J. Wong wrote:
> I've rebased my trees and pushed them all to github.
>
> The for-dave-for-4.6 kernel and progs branches are the giant piles of patches
> against Dave's for-next integration trees which (I think) are being reviewed
> for 4.6.
>
> The for-dave branches are against upstream as they've always been.
BTW, what's the point of for-dave vs for-dave-for-4.6 for xfsprogs?
> New patches have been added on the end of the patchset.
>
> I noticed that generic/139 crashes for-dave with a 1k block size due something
> or other sending us bio->bi_bdev == NULL. This seems to be sorted out somehow
> in for-next. Other than that I haven't seen any problems... but I've only
> run against x64 on bare XFS. Will run other arches/NFS/etc tonight/tomorrow.
>
> The transaction block reservation complaints should be fixed now, and I
> think the transaction reservations have been fixed too... or at least they
> don't show up on the tinydisk test setup. But all that means is that someone
> else will find it, probably within the first 3 minutes of testing. :P
Passes on NFS without hitting the space reservation issue, and passes
on XFS without new regression. The odd transaction (not space)
reservation assert in xfs/140 that I started to myesteriously 100%
reproduce last week still is around on XFS. I'll see if I can fix that
or at least triage it further..
|