On Tue, May 12, 2020 at 12:59:49PM +1000, Dave Chinner wrote: > Concurrent rm of same 50 million inodes: > > unpatched patched > machine A: 8m30s 3m09s > machine B: 4m02s 4m51s This actually is a significant slow down on machine B, which might be the more common one. Can you find out how that happens and if we can mitigate it?
