Re: [PATCH V5 00/12] Enable per-file/per-directory DAX operations V5

[Ira Weiny]

> > 
> > In summary:
> > 
> >  - Applications must call statx to discover the current S_DAX state.
> 
> Ok.
> 
> >  - There exists an advisory file inode flag FS_XFLAG_DAX that is set based on
> >    the parent directory FS_XFLAG_DAX inode flag.  (There is no way to change
> >    this flag after file creation.)
> > 
> >    If FS_XFLAG_DAX is set and the fs is on pmem then it will enable S_DAX at
> >    inode load time; if FS_XFLAG_DAX is not set, it will not enable S_DAX.
> >    Unless overridden...
> 
> Ok, fine with me. :)

:-D

> 
> >  - There exists a dax= mount option.
> > 
> >    "-o dax=off" means "never set S_DAX, ignore FS_XFLAG_DAX"
> >    	"-o nodax" means "dax=off"
> 
> I surveyed the three fses that support dax and found that none of the
> filesystems actually have a 'nodax' flag.  Now would be the time not to
> add such a thing, and make people specify dax=off instead.  It would
> be handy if we could have a single fsparam_enum for figuring out the dax
> mount options.

yes good point.

I'm working on updating the documentation patch and I think this might also
be better as:

	-o dax=never

Which is the opposite of 'always'.

> 
> >    "-o dax=always" means "always set S_DAX (at least on pmem), ignore FS_XFLAG_DAX"
> >    	"-o dax" by itself means "dax=always"
> >    "-o dax=iflag" means "follow FS_XFLAG_DAX" and is the default
> > 
> >  - There exists an advisory directory inode flag FS_XFLAG_DAX that can be
> >    changed at any time.  The flag state is copied into any files or
> >    subdirectories when they are created within that directory.  If programs
> >    require file access runs in S_DAX mode, they'll have to create those files
> 
> "...they must create..."

yes

> 
> >    inside a directory with FS_XFLAG_DAX set, or mount the fs with an
> >    appropriate dax mount option.
> 
> Otherwise seems ok to me.

Thanks!
Ira