Re: barriers off by default?

[Jan Kara <jack@suse.cz>]

On Thu 15-05-08 15:44:04, Eric Sandeen wrote:
> Jan Kara wrote:
> 
> >> As I look at my shiny new 500G disks with 32MB of cache, I find myself
> >> wondering why the default for ext3 and ext4 is to have barriers disabled.
> >>
> >> This is a pretty dangerous default w.r.t. filesystem integrity on power
> >> loss, no?
> >>     
> >   JFYI: SUSE kernel carries for ages a patch which changes this default.
> > I'd be more than happy to drop it ;).
> >
> > 								Honza
> >   
> What do folks think of this?
> 
> the show_options change is a little funky since jbd may do a test 
> write and fail... (actually I was thinking maybe at fill_super we 
> should do a test barrier write and get it out of the way early...)
  Yes, the patch looks fine with me.

  Acked-by: Jan Kara <jack@suse.cz>

> diff --git a/Documentation/filesystems/ext3.txt b/Documentation/filesystems/ext3.txt
> index b45f3c1..daab1f5 100644
> --- a/Documentation/filesystems/ext3.txt
> +++ b/Documentation/filesystems/ext3.txt
> @@ -52,8 +52,16 @@ commit=nrsec	(*)	Ext3 can be told to sync all its data and metadata
>  			Setting it to very large values will improve
>  			performance.
>  
> -barrier=1		This enables/disables barriers.  barrier=0 disables
> -			it, barrier=1 enables it.
> +barrier=<0|1(*)>	This enables/disables the use of write barriers in
> +			the jbd code.  barrier=0 disables, barrier=1 enables.
> +			This also requires an IO stack which can support
> +			barriers, and if jbd gets an error on a barrier
> +			write, it will disable again with a warning.
> +			Write barriers enforce proper on-disk ordering
> +			of journal commits, making volatile disk write caches
> +			safe to use, at some performance penalty.  If
> +			your disks are battery-backed in one way or another,
> +			disabling barriers may safely improve performance.
>  
>  orlov		(*)	This enables the new Orlov block allocator. It is
>  			enabled by default.
> diff --git a/fs/ext3/super.c b/fs/ext3/super.c
> index fe3119a..d06e0f3 100644
> --- a/fs/ext3/super.c
> +++ b/fs/ext3/super.c
> @@ -555,6 +555,7 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
>  	struct super_block *sb = vfs->mnt_sb;
>  	struct ext3_sb_info *sbi = EXT3_SB(sb);
>  	struct ext3_super_block *es = sbi->s_es;
> +	journal_t *journal = EXT3_SB(sb)->s_journal;
>  	unsigned long def_mount_opts;
>  
>  	def_mount_opts = le32_to_cpu(es->s_default_mount_opts);
> @@ -613,8 +614,16 @@ static int ext3_show_options(struct seq_file *seq, struct vfsmount *vfs)
>  		seq_printf(seq, ",commit=%u",
>  			   (unsigned) (sbi->s_commit_interval / HZ));
>  	}
> -	if (test_opt(sb, BARRIER))
> -		seq_puts(seq, ",barrier=1");
> +	if (!test_opt(sb, BARRIER)) {
> +		seq_puts(seq, ",barrier=0");
> +	} else {
> +		/*
> +	 	* jbd inherits the barrier flag from ext3, and jbd may actually
> +	 	* turn off barriers if a write fails, so it's the real test.
> +	 	*/
> +		if (journal && !(journal->j_flags & JFS_BARRIER))
> +			seq_puts(seq, ",barrier=1(failed)");
> +	}
>  	if (test_opt(sb, NOBH))
>  		seq_puts(seq, ",nobh");
>  
> @@ -1589,6 +1598,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent)
>  	sbi->s_resgid = le16_to_cpu(es->s_def_resgid);
>  
>  	set_opt(sbi->s_mount_opt, RESERVATION);
> +	set_opt(sbi->s_mount_opt, BARRIER);
>  
>  	if (!parse_options ((char *) data, sb, &journal_inum, &journal_devnum,
>  			    NULL, 0))
> 
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR