On Tue, Dec 04, 2018 at 11:21:14AM +0200, Nikolay Borisov wrote: > > > On 3.12.18 г. 18:06 ч., Josef Bacik wrote: > > The cleaner thread usually takes care of delayed iputs, with the > > exception of the btrfs_end_transaction_throttle path. The cleaner > > thread only gets woken up every 30 seconds, so instead wake it up to do > > it's work so that we can free up that space as quickly as possible. > > This description misses any rationale whatsoever about why the cleaner > needs to be woken up more frequently than 30 seconds (and IMO this is > the most important question that needs answering). > Yeah I'll add that. > Also have you done any measurements of the number of processed delayed > inodes with this change. Given the behavior you so desire why not just > make delayed iputs to be run via schedule_work on the global workqueue > and be done with it? I'm sure the latency will be better than the > current 30 seconds one :) We already have the cleaner thread to do this work, and it sets up for the snapshot drop stuff to be run as well. We could probably add another delayed work thing, but I would rather do that in a different patch. Thanks, Josef
