On Jun 02, 2009 08:07 -0400, Theodore Ts'o wrote: > When closing a file that had been previously truncated, force any > delay allocated blocks that to be allocated so that if the filesystem > is mounted with data=ordered, the data blocks will be pushed out to > disk along with the journal commit. Many application programs expect > this, so we do this to avoid zero length files if the system crashes > unexpectedly. > > @@ -3880,6 +3880,9 @@ void ext4_truncate(struct inode *inode) > if (!ext4_can_truncate(inode)) > return; > > + if (inode->i_size == 0) > + ei->i_state |= EXT4_STATE_DA_ALLOC_CLOSE; Since some applications open files with open(..., O_WRONLY|O_CREAT|O_TRUNC) to avoid re-using existing files (and avoiding the need to check if the file already exists to modify the flags), it would make sense to set EXT4_STATE_DA_ALLOC_CLOSE only if the file previously had some data in it. By the time we get to ext4_truncate() i_size is overwritten already, but it might make sense to also check i_disksize != 0 before setting this flag. Otherwise delayed allocation may be inadvertently disabled for these apps when it should not be. Cheers, Andreas -- Andreas Dilger Sr. Staff Engineer, Lustre Group Sun Microsystems of Canada, Inc. -- To unsubscribe from this list: send the line "unsubscribe linux-ext4" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
