From e30b5dca15dea86aa697f9d58ff646294fe80d3d Mon Sep 17 00:00:00 2001 From: "Yan, Zheng" Date: Fri, 3 May 2013 02:15:52 -0400 Subject: ext4: fix fio regression We (Linux Kernel Performance project) found a regression introduced by commit: f7fec032aa ext4: track all extent status in extent status tree The commit causes about 20% performance decrease in fio random write test. Profiler shows that rb_next() uses a lot of CPU time. The call stack is: rb_next ext4_es_find_delayed_extent ext4_map_blocks _ext4_get_block ext4_get_block_write __blockdev_direct_IO ext4_direct_IO generic_file_direct_write __generic_file_aio_write ext4_file_write aio_rw_vect_retry aio_run_iocb do_io_submit sys_io_submit system_call_fastpath io_submit td_io_getevents io_u_queued_complete thread_main main __libc_start_main The cause is that ext4_es_find_delayed_extent() doesn't have an upper bound, it keeps searching until a delayed extent is found. When there are a lots of non-delayed entries in the extent state tree, ext4_es_find_delayed_extent() may uses a lot of CPU time. Reported-by: LKP project Signed-off-by: Yan, Zheng Signed-off-by: Zheng Liu Cc: "Theodore Ts'o" --- fs/ext4/file.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'fs/ext4/file.c') diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 64848b5..1bc4523 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -464,7 +464,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) * If there is a delay extent at this offset, * it will be as a data. */ - ext4_es_find_delayed_extent(inode, last, &es); + ext4_es_find_delayed_extent_range(inode, last, last, &es); if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { if (last != start) dataoff = last << blkbits; @@ -547,7 +547,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) * If there is a delay extent at this offset, * we will skip this extent. */ - ext4_es_find_delayed_extent(inode, last, &es); + ext4_es_find_delayed_extent_range(inode, last, last, &es); if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { last = es.es_lblk + es.es_len; holeoff = last << blkbits; -- cgit v1.1