Jan Kara 5052b069ac jbd2: fix dbench4 performance regression for 'nobarrier' mounts ... Commit b685d3d65a "block: treat REQ_FUA and REQ_PREFLUSH as synchronous" removed REQ_SYNC flag from WRITE_FUA implementation. Since JBD2 strips REQ_FUA and REQ_FLUSH flags from submitted IO when the filesystem is mounted with nobarrier mount option, journal superblock writes ended up being async writes after this patch and that caused heavy performance regression for dbench4 benchmark with high number of processes. In my test setup with HP RAID array with non-volatile write cache and 32 GB ram, dbench4 runs with 8 processes regressed by ~25%. Fix the problem by making sure journal superblock writes are always treated as synchronous since they generally block progress of the journalling machinery and thus the whole filesystem. Fixes: b685d3d65a CC: stable@vger.kernel.org Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Theodore Ts'o <tytso@mit.edu>		2017-04-29 21:07:30 -04:00
..
checkpoint.c	block,fs: use REQ_* flags directly	2016-11-01 09:43:26 -06:00
commit.c	fs/jbd2, locking/mutex, sched/wait: Use mutex_lock_io() for journal->j_checkpoint_mutex	2017-01-14 11:30:06 +01:00
journal.c	jbd2: fix dbench4 performance regression for 'nobarrier' mounts	2017-04-29 21:07:30 -04:00
Kconfig	jbd2: remove debug dependency on debug_fs and update Kconfig help text	2013-06-12 23:07:51 -04:00
Makefile
recovery.c	fs: have ll_rw_block users pass in op and flags separately	2016-06-07 13:41:38 -06:00
revoke.c	jbd2: don't leak memory if setting up journal fails	2017-03-15 15:08:48 -04:00
transaction.c	jbd2: don't leak modified metadata buffers on an aborted journal	2017-02-04 23:14:19 -05:00