On Mon, Apr 13, 2020 at 09:01:38AM -0700, Darrick J. Wong wrote: > On Sun, Apr 12, 2020 at 10:40:43PM -0700, ira.weiny@xxxxxxxxx wrote: [snip] > > > > -STATIC void > > +void > > xfs_diflags_to_iflags( > > - struct inode *inode, > > - struct xfs_inode *ip) > > + struct xfs_inode *ip, > > + bool init) > > { > > - uint16_t flags = ip->i_d.di_flags; > > - > > - inode->i_flags &= ~(S_IMMUTABLE | S_APPEND | S_SYNC | > > - S_NOATIME | S_DAX); > > - > > - if (flags & XFS_DIFLAG_IMMUTABLE) > > - inode->i_flags |= S_IMMUTABLE; > > - if (flags & XFS_DIFLAG_APPEND) > > - inode->i_flags |= S_APPEND; > > - if (flags & XFS_DIFLAG_SYNC) > > - inode->i_flags |= S_SYNC; > > - if (flags & XFS_DIFLAG_NOATIME) > > - inode->i_flags |= S_NOATIME; > > - if (xfs_inode_enable_dax(ip)) > > - inode->i_flags |= S_DAX; > > + struct inode *inode = VFS_I(ip); > > + unsigned int xflags = xfs_ip2xflags(ip); > > + unsigned int flags = 0; > > + > > + ASSERT(!(IS_DAX(inode) && init)); > > + > > + if (xflags & FS_XFLAG_IMMUTABLE) > > + flags |= S_IMMUTABLE; > > + if (xflags & FS_XFLAG_APPEND) > > + flags |= S_APPEND; > > + if (xflags & FS_XFLAG_SYNC) > > + flags |= S_SYNC; > > + if (xflags & FS_XFLAG_NOATIME) > > + flags |= S_NOATIME; > > + if (init && xfs_inode_enable_dax(ip)) > > + flags |= S_DAX; > > + > > + inode->i_flags &= ~(S_IMMUTABLE | S_APPEND | S_SYNC | S_NOATIME); > > I noticed that S_DAX drops out of the mask out operation here, which of > course resulted in an eyebrow-raise because the other four flags are > always set to whatever we just computed. :) > > Then I realized that yes, this is intentional since we can't change > S_DAX on the fly, and that S_DAX is never set i_flags on an inode that's > being initialized so we don't need to mask off S_DAX ever. > > Could we add a comment here to remind the reader that S_DAX is a bit > special? > > /* > * S_DAX can only be set during inode initialization and is never set by > * the VFS, so we cannot mask off S_DAX in i_flags. > */ > > With that added, > Reviewed-by: Darrick J. Wong <darrick.wong@xxxxxxxxxx> Added that comment. Thanks for the review! Ira > > --D