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>
This commit is contained in:

committed by
Theodore Ts'o

parent
0d606e2c9f
commit
e30b5dca15
@@ -3642,7 +3642,7 @@ int ext4_find_delalloc_range(struct inode *inode,
|
||||
{
|
||||
struct extent_status es;
|
||||
|
||||
ext4_es_find_delayed_extent(inode, lblk_start, &es);
|
||||
ext4_es_find_delayed_extent_range(inode, lblk_start, lblk_end, &es);
|
||||
if (es.es_len == 0)
|
||||
return 0; /* there is no delay extent in this tree */
|
||||
else if (es.es_lblk <= lblk_start &&
|
||||
@@ -4608,9 +4608,10 @@ static int ext4_find_delayed_extent(struct inode *inode,
|
||||
struct extent_status es;
|
||||
ext4_lblk_t block, next_del;
|
||||
|
||||
ext4_es_find_delayed_extent(inode, newes->es_lblk, &es);
|
||||
|
||||
if (newes->es_pblk == 0) {
|
||||
ext4_es_find_delayed_extent_range(inode, newes->es_lblk,
|
||||
newes->es_lblk + newes->es_len - 1, &es);
|
||||
|
||||
/*
|
||||
* No extent in extent-tree contains block @newes->es_pblk,
|
||||
* then the block may stay in 1)a hole or 2)delayed-extent.
|
||||
@@ -4630,7 +4631,7 @@ static int ext4_find_delayed_extent(struct inode *inode,
|
||||
}
|
||||
|
||||
block = newes->es_lblk + newes->es_len;
|
||||
ext4_es_find_delayed_extent(inode, block, &es);
|
||||
ext4_es_find_delayed_extent_range(inode, block, EXT_MAX_BLOCKS, &es);
|
||||
if (es.es_len == 0)
|
||||
next_del = EXT_MAX_BLOCKS;
|
||||
else
|
||||
|
Reference in New Issue
Block a user