From ac0c18f2c693f0e7a44dbbb36b14d5141e5d20e5 Mon Sep 17 00:00:00 2001 From: Kemeng Shi Date: Wed, 28 Feb 2024 17:19:53 +0800 Subject: fs/writeback: avoid to writeback non-expired inode in kupdate writeback In kupdate writeback, only expired inode (have been dirty for longer than dirty_expire_interval) is supposed to be written back. However, kupdate writeback will writeback non-expired inode left in b_io or b_more_io from last wb_writeback. As a result, writeback will keep being triggered unexpected when we keep dirtying pages even dirty memory is under threshold and inode is not expired. To be more specific: Assume dirty background threshold is > 1G and dirty_expire_centisecs is > 60s. When we running fio -size=1G -invalidate=0 -ioengine=libaio --time_based -runtime=60... (keep dirtying), the writeback will keep being triggered as following: wb_workfn wb_do_writeback wb_check_background_flush /* * Wb dirty background threshold starts at 0 if device was idle and * grows up when bandwidth of wb is updated. So a background * writeback is triggered. */ wb_over_bg_thresh /* * Dirtied inode will be written back and added to b_more_io list * after slice used up (because we keep dirtying the inode). */ wb_writeback Writeback is triggered per dirty_writeback_centisecs as following: wb_workfn wb_do_writeback wb_check_old_data_flush /* * Write back inode left in b_io and b_more_io from last wb_writeback * even the inode is non-expired and it will be added to b_more_io * again as slice will be used up (because we keep dirtying the * inode) */ wb_writeback Fix this by moving non-expired inode to dirty list instead of more io list for kupdate writeback in requeue_inode. Test as following: /* make it more easier to observe the issue */ echo 300000 > /proc/sys/vm/dirty_expire_centisecs echo 100 > /proc/sys/vm/dirty_writeback_centisecs /* create a idle device */ mkfs.ext4 -F /dev/vdb mount /dev/vdb /bdi1/ /* run buffer write with fio */ fio -name test -filename=/bdi1/file -size=800M -ioengine=libaio -bs=4K \ -iodepth=1 -rw=write -direct=0 --time_based -runtime=60 -invalidate=0 Fio result before fix (run three tests): 1360MB/s 1329MB/s 1455MB/s Fio result after fix (run three tests): 1737MB/s 1729MB/s 1789MB/s Writeback for non-expired inode is gone as expeted. Observe this with trace writeback_start and writeback_written as following: echo 1 > /sys/kernel/debug/tracing/events/writeback/writeback_start/enab echo 1 > /sys/kernel/debug/tracing/events/writeback/writeback_written/enable cat /sys/kernel/tracing/trace_pipe Signed-off-by: Kemeng Shi Link: https://lore.kernel.org/r/20240228091958.288260-2-shikemeng@huaweicloud.com Reviewed-by: Jan Kara Signed-off-by: Christian Brauner --- fs/fs-writeback.c | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) (limited to 'fs/fs-writeback.c') diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index e4f17c53ddfc..fe634f00f4d9 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1561,7 +1561,8 @@ static void inode_sleep_on_writeback(struct inode *inode) * thread's back can have unexpected consequences. */ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, - struct writeback_control *wbc) + struct writeback_control *wbc, + unsigned long dirtied_before) { if (inode->i_state & I_FREEING) return; @@ -1594,7 +1595,8 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, * We didn't write back all the pages. nfs_writepages() * sometimes bales out without doing anything. */ - if (wbc->nr_to_write <= 0) { + if (wbc->nr_to_write <= 0 && + !inode_dirtied_after(inode, dirtied_before)) { /* Slice used up. Queue for next turn. */ requeue_io(inode, wb); } else { @@ -1862,6 +1864,11 @@ static long writeback_sb_inodes(struct super_block *sb, unsigned long start_time = jiffies; long write_chunk; long total_wrote = 0; /* count both pages and inodes */ + unsigned long dirtied_before = jiffies; + + if (work->for_kupdate) + dirtied_before = jiffies - + msecs_to_jiffies(dirty_expire_interval * 10); while (!list_empty(&wb->b_io)) { struct inode *inode = wb_inode(wb->b_io.prev); @@ -1967,7 +1974,7 @@ static long writeback_sb_inodes(struct super_block *sb, spin_lock(&inode->i_lock); if (!(inode->i_state & I_DIRTY_ALL)) total_wrote++; - requeue_inode(inode, tmp_wb, &wbc); + requeue_inode(inode, tmp_wb, &wbc, dirtied_before); inode_sync_complete(inode); spin_unlock(&inode->i_lock); -- cgit v1.2.3