Re: [PATCH v3 03/12] fs/xfs: Separate functionality of xfs_inode_supports_dax()

From: Ira Weiny <ira.weiny@intel.com>
To: Dave Chinner <david@fromorbit.com>
Cc: linux-kernel@vger.kernel.org,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	"Darrick J. Wong" <darrick.wong@oracle.com>,
	Dan Williams <dan.j.williams@intel.com>,
	Christoph Hellwig <hch@lst.de>,
	"Theodore Y. Ts'o" <tytso@mit.edu>, Jan Kara <jack@suse.cz>,
	linux-ext4@vger.kernel.org, linux-xfs@vger.kernel.org,
	linux-fsdevel@vger.kernel.org
Subject: Re: [PATCH v3 03/12] fs/xfs: Separate functionality of xfs_inode_supports_dax()
Date: Tue, 11 Feb 2020 08:13:49 -0800
Message-ID: <20200211161348.GA12866@iweiny-DESK2.sc.intel.com> (raw)
In-Reply-To: <20200211054748.GF10776@dread.disaster.area>

On Tue, Feb 11, 2020 at 04:47:48PM +1100, Dave Chinner wrote:
> On Sat, Feb 08, 2020 at 11:34:36AM -0800, ira.weiny@intel.com wrote:
> > From: Ira Weiny <ira.weiny@intel.com>
> > 

[snip]

> >  
> > +static bool
> > +xfs_inode_is_dax(
> > +	struct xfs_inode *ip)
> > +{
> > +	return (ip->i_d.di_flags2 & XFS_DIFLAG2_DAX) == XFS_DIFLAG2_DAX;
> > +}
> 
> I don't think these wrappers add any value at all - the naming of
> them is entirely confusing, too. e.g. "inode is dax" doesn't tell me
> that it is checking the on disk flags - it doesn't tell me how it is
> different to IS_DAX, or why I'd use one versus the other. And then
> xfs_inode_mount_is_dax() is just... worse.
> 
> Naming is hard. :)

Sure...  I'm particularly bad as well...

FWIW I don't see how xfs_inode_mount_is_dax() is worse, I rather think that is
pretty clear but I'm not going to quibble over names because I know I'm rubbish
at it and I'm certainly not enough of a FS person to make them clear...  ;-)

> 
> > +
> > +static bool
> > +xfs_inode_use_dax(
> > +	struct xfs_inode *ip)
> > +{
> > +	return xfs_inode_supports_dax(ip) &&
> > +		(xfs_inode_mount_is_dax(ip) ||
> > +		 xfs_inode_is_dax(ip));
> > +}
> 
> Urk. Naming - we're not "using dax" here, we are checkign to see if
> we should enable DAX on this inode. IOWs:

Well just to defend myself a little bit.  My thought was:

"When setting i_flags, should I use dax?"

> 
> static bool
> xfs_inode_enable_dax(
> 	struct xfs_inode *ip)
> {
> 	if (!xfs_inode_supports_dax(ip))
> 		return false;
> 
> 	if (ip->i_d.di_flags2 & XFS_DIFLAG2_DAX)
> 		return true;
> 	if (ip->i_mount->m_flags & XFS_MOUNT_DAX)
> 		return true;
> 	return false;
> }

Anyway, I'm good with this.

Changed for V4.

Thanks!
Ira

> 
> Cheers,
> 
> Dave.
> -- 
> Dave Chinner
> david@fromorbit.com