diff options
| author | Jaegeuk Kim <jaegeuk.kim@samsung.com> | 2014-04-02 08:34:36 +0200 |
|---|---|---|
| committer | Jaegeuk Kim <jaegeuk.kim@samsung.com> | 2014-04-07 02:50:58 +0200 |
| commit | 6b4afdd794783fe515b50838aa36591e3feea990 (patch) | |
| tree | 5f35f32e27db8deaf248cf1e7a4930aab0822a73 /fs/f2fs/file.c | |
| parent | f2fs: fix to cover io->bio with io_rwsem (diff) | |
| download | linux-6b4afdd794783fe515b50838aa36591e3feea990.tar.xz linux-6b4afdd794783fe515b50838aa36591e3feea990.zip | |
f2fs: introduce f2fs_issue_flush to avoid redundant flush issue
Some storage devices show relatively high latencies to complete cache_flush
commands, even though their normal IO speed is prettry much high. In such
the case, it needs to merge cache_flush commands as much as possible to avoid
issuing them redundantly.
So, this patch introduces a mount option, "-o flush_merge", to mitigate such
the overhead.
If this option is enabled by user, F2FS merges the cache_flush commands and then
issues just one cache_flush on behalf of them. Once the single command is
finished, F2FS sends a completion signal to all the pending threads.
Note that, this option can be used under a workload consisting of very intensive
concurrent fsync calls, while the storage handles cache_flush commands slowly.
Signed-off-by: Jaegeuk Kim <jaegeuk.kim@samsung.com>
Diffstat (limited to 'fs/f2fs/file.c')
| -rw-r--r-- | fs/f2fs/file.c | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 6ba26680c468..302d552afea5 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -186,7 +186,7 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) ret = wait_on_node_pages_writeback(sbi, inode->i_ino); if (ret) goto out; - ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); + ret = f2fs_issue_flush(F2FS_SB(inode->i_sb)); } out: trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret); |