Re: [PATCH v5 0/20] fs: Fix missed inode write during fsync
Jan Kara <[email protected]>
| Newsgroups | org.kernel.vger.linux-ext4,org.kernel.vger.linux-fsdevel |
|---|---|
| Message-ID | <jckkvo72sng2r33ryty4yuduky6vl2plvsqim3fyqflp4vepwy@shqhgmis5pjr> |
On Sun 09-08-26 18:36:19, Askar Safin wrote: > Jan Kara <[email protected]>: > > The series strives to make sure sync_inode_metadata() -> > > writeback_single_inode() will end up to properly persisting not only the inode > > but also all metadata associated with the inode. This is beneficial for > > several reasons: > > It is well known that if one just created a file and wants to properly > persist it, they should not only call fsync on it, but also on its > parent directory. Yes. > Is it true that after your patchset this will not be needed anymore? No. This patch set only makes sure the inode itself and metadata such as indirect blocks etc. properly make it to the disk on fsync(2) and similar events which wasn't the case before for some filesystems. Honza -- Jan Kara <[email protected]> SUSE Labs, CR