Re: file metadata via fs API (was: [GIT PULL] Filesystem Information)

[Al Viro]

On Tue, Aug 11, 2020 at 08:20:24AM -0700, Linus Torvalds wrote:

> I don't think this works for the reasons Al says, but a slight
> modification might.
> 
> IOW, if you do something more along the lines of
> 
>        fd = open(""foo/bar", O_PATH);
>        metadatafd = openat(fd, "metadataname", O_ALT);
> 
> it might be workable.
> 
> So you couldn't do it with _one_ pathname, because that is always
> fundamentally going to hit pathname lookup rules.
> 
> But if you start a new path lookup with new rules, that's fine.

Except that you suddenly see non-directory dentries get children.
And a lot of dcache-related logics needs to be changed if that
becomes possible.

I agree that xattrs are garbage, but this approach won't be
a straightforward solution.  Can those suckers be passed to
...at() as starting points?  Can they be bound in namespace?
Can something be bound *on* them?  What do they have for inodes
and what maintains their inumbers (and st_dev, while we are at
it)?  Can _they_ have secondaries like that (sensu Swift)?
Is that a flat space, or can they be directories?

Only a part of the problems is implementation-related (and those are
not trivial at all); most the fun comes from semantics of those things.
And answers to the implementation questions are seriously dependent upon
that...