Re: [RFC] atomic create+open

[Miklos Szeredi]

> I think Miklos' point is that it's not an "optimisation" because it's
> not optional.  Optimisations are things where if you don't do them,
> the behaviour is still correct but slower.
> 
> As far as I can tell from this discussion, the atomic lookup+create is
> a non-optional requirement.

Exactly.

Trond, you wrote this in an earlier discussion:

> > so the filesystem can delay returning the error from the open
> > operation until the other errors have been sorted out by the lookup
> > code.
> 
> Intents are meant as optimisations, not replacements for existing
> operations. I'm therefore not really comfortable about having them
> return errors at all.

The case I described is not an optimization, so in that case you seem
to agree, that lookup intents are not the solution.

Miklos
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/