On 7 Nov 2022 13:41:49 +0100 Jan Kara <jack@xxxxxxx> > So in fact I made sure (by debug counters) that sb_mark_inode_writeback() > and sb_clear_inode_writeback() get called the same number of times before > evict() gets called. So your debug patch would change nothing AFAICT... The same number you observed != the difference ihold + iput can make since the added ihold ensures the crash has nothing to do with the two functions mentioned if it reproduces. Right?
