linux/fs/ext3/fsync.c

117 lines
3.4 KiB
C
Raw Normal View History

/*
* linux/fs/ext3/fsync.c
*
* Copyright (C) 1993 Stephen Tweedie (sct@redhat.com)
* from
* Copyright (C) 1992 Remy Card (card@masi.ibp.fr)
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie (Paris VI)
* from
* linux/fs/minix/truncate.c Copyright (C) 1991, 1992 Linus Torvalds
*
* ext3fs fsync primitive
*
* Big-endian to little-endian byte-swapping/bitmaps by
* David S. Miller (davem@caip.rutgers.edu), 1995
*
* Removed unnecessary code duplication for little endian machines
* and excessive __inline__s.
* Andi Kleen, 1997
*
* Major simplications and cleanup - we only need to do the metadata, because
* we can depend on generic_block_fdatasync() to sync the data blocks.
*/
#include <linux/time.h>
#include <linux/blkdev.h>
#include <linux/fs.h>
#include <linux/sched.h>
#include <linux/writeback.h>
#include <linux/jbd.h>
#include <linux/ext3_fs.h>
#include <linux/ext3_jbd.h>
#include <trace/events/ext3.h>
/*
* akpm: A new design for ext3_sync_file().
*
* This is only called from sys_fsync(), sys_fdatasync() and sys_msync().
* There cannot be a transaction open by this task.
* Another task could have dirtied this inode. Its data can be in any
* state in the journalling system.
*
* What we do is just kick off a commit and wait on it. This will snapshot the
* inode to disk.
*/
int ext3_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
{
struct inode *inode = file->f_mapping->host;
struct ext3_inode_info *ei = EXT3_I(inode);
journal_t *journal = EXT3_SB(inode->i_sb)->s_journal;
int ret, needs_barrier = 0;
tid_t commit_tid;
trace_ext3_sync_file_enter(file, datasync);
if (inode->i_sb->s_flags & MS_RDONLY)
return 0;
ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
if (ret)
goto out;
/*
* Taking the mutex here just to keep consistent with how fsync was
* called previously, however it looks like we don't need to take
* i_mutex at all.
*/
mutex_lock(&inode->i_mutex);
J_ASSERT(ext3_journal_current_handle() == NULL);
/*
* data=writeback,ordered:
* The caller's filemap_fdatawrite()/wait will sync the data.
* Metadata is in the journal, we wait for a proper transaction
* to commit here.
*
* data=journal:
* filemap_fdatawrite won't do anything (the buffers are clean).
* ext3_force_commit will write the file data into the journal and
* will wait on that.
* filemap_fdatawait() will encounter a ton of newly-dirtied pages
* (they were dirtied by commit). But that's OK - the blocks are
* safe in-journal, which is all fsync() needs to ensure.
*/
if (ext3_should_journal_data(inode)) {
mutex_unlock(&inode->i_mutex);
ret = ext3_force_commit(inode->i_sb);
goto out;
}
if (datasync)
commit_tid = atomic_read(&ei->i_datasync_tid);
else
commit_tid = atomic_read(&ei->i_sync_tid);
ext3: fdatasync should skip metadata writeout when overwriting Currently fdatasync is identical to fsync in ext3. I think fdatasync should skip journal flush in data=ordered and data=writeback mode when it overwrites to already-instantiated blocks on HDD. When I_DIRTY_DATASYNC flag is not set, fdatasync should skip journal writeout because this indicates only atime or/and mtime updates. Following patch is the same approach of ext2's fsync code(ext2_sync_file). I did a performance test using the sysbench. #sysbench --num-threads=128 --max-requests=50000 --test=fileio --file-total-size=128G --file-test-mode=rndwr --file-fsync-mode=fdatasync run The result on ext3 was: -2.6.24 Operations performed: 0 Read, 50080 Write, 59600 Other = 109680 Total Read 0b Written 782.5Mb Total transferred 782.5Mb (12.116Mb/sec) 775.45 Requests/sec executed Test execution summary: total time: 64.5814s total number of events: 50080 total time taken by event execution: 3713.9836 per-request statistics: min: 0.0000s avg: 0.0742s max: 0.9375s approx. 95 percentile: 0.2901s Threads fairness: events (avg/stddev): 391.2500/23.26 execution time (avg/stddev): 29.0155/1.99 -2.6.24-patched Operations performed: 0 Read, 50009 Write, 61596 Other = 111605 Total Read 0b Written 781.39Mb Total transferred 781.39Mb (16.419Mb/sec) 1050.83 Requests/sec executed Test execution summary: total time: 47.5900s total number of events: 50009 total time taken by event execution: 2934.5768 per-request statistics: min: 0.0000s avg: 0.0587s max: 0.8938s approx. 95 percentile: 0.1993s Threads fairness: events (avg/stddev): 390.6953/22.64 execution time (avg/stddev): 22.9264/1.17 Filesystem I/O throughput was improved. Signed-off-by :Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Acked-by: Jan Kara <jack@suse.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-04-28 11:16:05 +02:00
if (test_opt(inode->i_sb, BARRIER) &&
!journal_trans_will_send_data_barrier(journal, commit_tid))
needs_barrier = 1;
log_start_commit(journal, commit_tid);
ret = log_wait_commit(journal, commit_tid);
/*
* In case we didn't commit a transaction, we have to flush
* disk caches manually so that data really is on persistent
* storage
*/
if (needs_barrier)
blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL);
mutex_unlock(&inode->i_mutex);
out:
trace_ext3_sync_file_exit(inode, ret);
return ret;
}