On Mon, Oct 08, 2018 at 01:24:55PM +1100, Dave Chinner wrote: > On Mon, Oct 08, 2018 at 09:02:58AM +1100, Dave Chinner wrote: > > On Sun, Oct 07, 2018 at 12:13:19PM +0200, Christoph Hellwig wrote: > > > On Sat, Oct 06, 2018 at 11:43:31AM +0200, Christoph Hellwig wrote: > > > > I suspect the asserts actually are what is incorrect. But given > > > > how late we are in the cycle I've just dropped the patch and kicked > > > > off xfstests runs (now including rmap, sigh..). > > > > > > Without this patch test runs including rmap succeed. Do you want > > > me to resend with the patch dropped (thing should just apply without > > > it as-is). > > > > I'll drop it and see what happens... > > Different problems after dropping that patch and re-instating the > rest of the series. The shutdown/io error stress tests now hang > randomly in unmount reclaiming inodes. e.g. generic/388: Ah, my fault. False alarm. I had a patch that I knew was broken at the end of the series that I hadn't commented out. All prior testing was without it, but when I reset branch without this patch I simply popped everything back in, including the broken patch at the end. Retesting now. -Dave. -- Dave Chinner david@xxxxxxxxxxxxx
