On Wed, Jan 03, 2024 at 02:37:08PM +0800, Daniel J Blueman wrote: > Hi Kent et al, > > On bcachefs-testing SHA 67503f8d4fc, I was able to trigger a BUG macro > at fs/bcachefs/extents.c:570 [1]. > > You'll need a "modprobe brd rd_size=536870912 rd_nr=8" before the > commands listed, however I wasn't able to reproduce this due to two > factors - existing filesystem state from previous runs, and a > background process corrupting some disk blocks on the first device. > The latter will also affect state in the former of course. > > I'll later figure out how to provide determinism in these areas for > reproducibility. Let me know if this testing is useful and if I should > be on master or downstream.
Yes, this testing is definitely useful. Let me know once you've got a reproducer and I'll turn it into a ktest test - thanks!
