diff options
| author | Yan, Zheng <zheng.z.yan@intel.com> | 2013-05-03 10:15:52 +0400 |
|---|---|---|
| committer | Theodore Ts'o <tytso@mit.edu> | 2013-05-03 10:15:52 +0400 |
| commit | e30b5dca15dea86aa697f9d58ff646294fe80d3d (patch) | |
| tree | 549d1ff3f8e5ecb9675a2937113aa99e64ef3ea1 /include/trace | |
| parent | 0d606e2c9fccdd4e67febf1e2da500e1bfe9e045 (diff) | |
| download | linux-e30b5dca15dea86aa697f9d58ff646294fe80d3d.tar.xz | |
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 <lkp@linux.intel.com>
Signed-off-by: Yan, Zheng <zheng.z.yan@intel.com>
Signed-off-by: Zheng Liu <wenqing.lz@taobao.com>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Diffstat (limited to 'include/trace')
| -rw-r--r-- | include/trace/events/ext4.h | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/include/trace/events/ext4.h b/include/trace/events/ext4.h index d0e686402df8..8ee15b97cd38 100644 --- a/include/trace/events/ext4.h +++ b/include/trace/events/ext4.h @@ -2139,7 +2139,7 @@ TRACE_EVENT(ext4_es_remove_extent, __entry->lblk, __entry->len) ); -TRACE_EVENT(ext4_es_find_delayed_extent_enter, +TRACE_EVENT(ext4_es_find_delayed_extent_range_enter, TP_PROTO(struct inode *inode, ext4_lblk_t lblk), TP_ARGS(inode, lblk), @@ -2161,7 +2161,7 @@ TRACE_EVENT(ext4_es_find_delayed_extent_enter, (unsigned long) __entry->ino, __entry->lblk) ); -TRACE_EVENT(ext4_es_find_delayed_extent_exit, +TRACE_EVENT(ext4_es_find_delayed_extent_range_exit, TP_PROTO(struct inode *inode, struct extent_status *es), TP_ARGS(inode, es), |