diff options
| author | Jaegeuk Kim <jaegeuk@kernel.org> | 2022-03-03 13:21:27 -0800 |
|---|---|---|
| committer | Jaegeuk Kim <jaegeuk@kernel.org> | 2022-03-04 09:15:53 -0800 |
| commit | 7f8e249dccc4c530cf26c2c091f5bb64e701c262 (patch) | |
| tree | 10ac7c57fbc469361ed04b35bd4f0179a5665851 /fs/f2fs/Kconfig | |
| parent | 50c63009f6ab37eb78d8b4f9bc606a12b2b46505 (diff) | |
| download | linux-7f8e249dccc4c530cf26c2c091f5bb64e701c262.tar.gz linux-7f8e249dccc4c530cf26c2c091f5bb64e701c262.tar.bz2 linux-7f8e249dccc4c530cf26c2c091f5bb64e701c262.zip | |
f2fs: introduce F2FS_UNFAIR_RWSEM to support unfair rwsem
Unfair rwsem should be used when blk-cg is on. Otherwise, there is regression.
FYI, we noticed a -26.7% regression of aim7.jobs-per-min due to commit:
commit: e4544b63a7ee49e7fbebf35ece0a6acd3b9617ae ("f2fs: move f2fs to use reader-unfair rwsems")
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git master
in testcase: aim7
on test machine: 88 threads 2 sockets Intel(R) Xeon(R) Gold 6238M CPU @ 2.10GHz with 128G memory
with following parameters:
disk: 4BRD_12G
md: RAID0
fs: f2fs
test: sync_disk_rw
load: 100
cpufreq_governor: performance
ucode: 0x500320a
test-description: AIM7 is a traditional UNIX system level benchmark suite which is used to test and measure the performance of multiuser system.
test-url: https://sourceforge.net/projects/aimbench/files/aim-suite7/
Reported-by: kernel test robot <oliver.sang@intel.com>
Reviewed-by: Chao Yu <chao@kernel.org>
Signed-off-by: Jaegeuk Kim <jaegeuk@kernel.org>
Diffstat (limited to 'fs/f2fs/Kconfig')
| -rw-r--r-- | fs/f2fs/Kconfig | 7 |
1 files changed, 7 insertions, 0 deletions
diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig index f46a7339d6cf..03ef087537c7 100644 --- a/fs/f2fs/Kconfig +++ b/fs/f2fs/Kconfig @@ -143,3 +143,10 @@ config F2FS_IOSTAT Support getting IO statistics through sysfs and printing out periodic IO statistics tracepoint events. You have to turn on "iostat_enable" sysfs node to enable this feature. + +config F2FS_UNFAIR_RWSEM + bool "F2FS unfair rw_semaphore" + depends on F2FS_FS && BLK_CGROUP + help + Use unfair rw_semaphore, if system configured IO priority by block + cgroup. |