Re: [PATCH v4 3/3] ext4: unconditionally enable the i_version counter

[Lukas Czerner <lczerner@redhat.com>]

On Fri, Aug 26, 2022 at 12:11:23PM -0400, Jeff Layton wrote:
> On Wed, 2022-08-24 at 18:03 +0200, Lukas Czerner wrote:
> > From: Jeff Layton <jlayton@kernel.org>
> > 
> > The original i_version implementation was pretty expensive, requiring a
> > log flush on every change. Because of this, it was gated behind a mount
> > option (implemented via the MS_I_VERSION mountoption flag).
> > 
> > Commit ae5e165d855d (fs: new API for handling inode->i_version) made the
> > i_version flag much less expensive, so there is no longer a performance
> > penalty from enabling it. xfs and btrfs already enable it
> > unconditionally when the on-disk format can support it.
> > 
> > Have ext4 ignore the SB_I_VERSION flag, and just enable it
> > unconditionally. While we're in here, remove the handling of
> > Opt_i_version as well, since we're almost to 5.20 anyway.
> > 
> > Ideally, we'd couple this change with a way to disable the i_version
> > counter (just in case), but the way the iversion mount option was
> > implemented makes that difficult to do. We'd need to add a new mount
> > option altogether or do something with tune2fs. That's probably best
> > left to later patches if it turns out to be needed.
> > 
> > [ Removed leftover bits of i_version from ext4_apply_options() since it
> > now can't ever be set in ctx->mask_s_flags -- lczerner ]
> > 
> > Cc: Dave Chinner <david@fromorbit.com>
> > Cc: Benjamin Coddington <bcodding@redhat.com>
> > Cc: Christoph Hellwig <hch@infradead.org>
> > Cc: Darrick J. Wong <djwong@kernel.org>
> > Signed-off-by: Jeff Layton <jlayton@kernel.org>
> > Signed-off-by: Lukas Czerner <lczerner@redhat.com>
> > Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
> > Reviewed-by: Jan Kara <jack@suse.cz>
> > ---
> > v3: Removed leftover bits of i_version from ext4_apply_options
> > v4: no change
> > 
> >  fs/ext4/inode.c |  5 ++---
> >  fs/ext4/super.c | 21 ++++-----------------
> >  2 files changed, 6 insertions(+), 20 deletions(-)
> > 
> > diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
> > index 2a220be34caa..c77d40f05763 100644
> > --- a/fs/ext4/inode.c
> > +++ b/fs/ext4/inode.c
> > @@ -5425,7 +5425,7 @@ int ext4_setattr(struct user_namespace *mnt_userns, struct dentry *dentry,
> >  			return -EINVAL;
> >  		}
> >  
> > -		if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size)
> > +		if (attr->ia_size != inode->i_size)
> >  			inode_inc_iversion(inode);
> >  
> >  		if (shrink) {
> > @@ -5735,8 +5735,7 @@ int ext4_mark_iloc_dirty(handle_t *handle,
> >  	 * ea_inodes are using i_version for storing reference count, don't
> >  	 * mess with it
> >  	 */
> > -	if (IS_I_VERSION(inode) &&
> > -	    !(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
> > +	if (!(EXT4_I(inode)->i_flags & EXT4_EA_INODE_FL))
> >  		inode_inc_iversion(inode);
> >  
> >  	/* the do_update_inode consumes one bh->b_count */
> > diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> > index 9a66abcca1a8..1c953f6d400e 100644
> > --- a/fs/ext4/super.c
> > +++ b/fs/ext4/super.c
> > @@ -1585,7 +1585,7 @@ enum {
> >  	Opt_inlinecrypt,
> >  	Opt_usrjquota, Opt_grpjquota, Opt_quota,
> >  	Opt_noquota, Opt_barrier, Opt_nobarrier, Opt_err,
> > -	Opt_usrquota, Opt_grpquota, Opt_prjquota, Opt_i_version,
> > +	Opt_usrquota, Opt_grpquota, Opt_prjquota,
> >  	Opt_dax, Opt_dax_always, Opt_dax_inode, Opt_dax_never,
> >  	Opt_stripe, Opt_delalloc, Opt_nodelalloc, Opt_warn_on_error,
> >  	Opt_nowarn_on_error, Opt_mblk_io_submit, Opt_debug_want_extra_isize,
> > @@ -1694,7 +1694,6 @@ static const struct fs_parameter_spec ext4_param_specs[] = {
> >  	fsparam_flag	("barrier",		Opt_barrier),
> >  	fsparam_u32	("barrier",		Opt_barrier),
> >  	fsparam_flag	("nobarrier",		Opt_nobarrier),
> > -	fsparam_flag	("i_version",		Opt_i_version),
> >  	fsparam_flag	("dax",			Opt_dax),
> >  	fsparam_enum	("dax",			Opt_dax_type, ext4_param_dax),
> >  	fsparam_u32	("stripe",		Opt_stripe),
> > @@ -2140,11 +2139,6 @@ static int ext4_parse_param(struct fs_context *fc, struct fs_parameter *param)
> >  	case Opt_abort:
> >  		ctx_set_mount_flag(ctx, EXT4_MF_FS_ABORTED);
> >  		return 0;
> > -	case Opt_i_version:
> > -		ext4_msg(NULL, KERN_WARNING, deprecated_msg, param->key, "5.20");
> > -		ext4_msg(NULL, KERN_WARNING, "Use iversion instead\n");
> > -		ctx_set_flags(ctx, SB_I_VERSION);
> > -		return 0;
> >  	case Opt_inlinecrypt:
> >  #ifdef CONFIG_FS_ENCRYPTION_INLINE_CRYPT
> >  		ctx_set_flags(ctx, SB_INLINECRYPT);
> > @@ -2814,14 +2808,6 @@ static void ext4_apply_options(struct fs_context *fc, struct super_block *sb)
> >  	sb->s_flags &= ~ctx->mask_s_flags;
> >  	sb->s_flags |= ctx->vals_s_flags;
> >  
> > -	/*
> > -	 * i_version differs from common mount option iversion so we have
> > -	 * to let vfs know that it was set, otherwise it would get cleared
> > -	 * on remount
> > -	 */
> > -	if (ctx->mask_s_flags & SB_I_VERSION)
> > -		fc->sb_flags |= SB_I_VERSION;
> > -
> >  #define APPLY(X) ({ if (ctx->spec & EXT4_SPEC_##X) sbi->X = ctx->X; })
> >  	APPLY(s_commit_interval);
> >  	APPLY(s_stripe);
> > @@ -2970,8 +2956,6 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb,
> >  		SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time);
> >  	if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME)
> >  		SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time);
> > -	if (sb->s_flags & SB_I_VERSION)
> > -		SEQ_OPTS_PUTS("i_version");
> >  	if (nodefs || sbi->s_stripe)
> >  		SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe);
> >  	if (nodefs || EXT4_MOUNT_DATA_FLAGS &
> > @@ -4640,6 +4624,9 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb)
> >  	sb->s_flags = (sb->s_flags & ~SB_POSIXACL) |
> >  		(test_opt(sb, POSIX_ACL) ? SB_POSIXACL : 0);
> >  
> > +	/* i_version is always enabled now */
> > +	sb->s_flags |= SB_I_VERSION;
> > +
> >  	if (le32_to_cpu(es->s_rev_level) == EXT4_GOOD_OLD_REV &&
> >  	    (ext4_has_compat_features(sb) ||
> >  	     ext4_has_ro_compat_features(sb) ||
> 
> Hi Lukas,
> 
> I know I had originally asked you to shepherd this patch into mainline,
> but I think it may be better to wait on it for now. Since I asked that,
> we've since found out that ext4 is bumping the i_version counter on
> atime updates. It'd be best to get that fixed before we turn this on
> unconditionally, since it could cause a performance regression in some
> cases. I'll plan to pick this back up for my latest i_version series if
> that sounds ok to you.
> 
> Sorry for the back and forth, and thanks again!

Hi Jeff,

sure, no problem. I can drop the patch. The rest of the series is still
valid though.

Thanks!
-Lukas

> 
> Cheers,
> -- 
> Jeff Layton <jlayton@kernel.org>
>