Karsten Kruse wrote:
> But even in this case a fs with softdep is safer than one without- It
>...
> Performance is a bonus, i think the safety alone is worth it.

Hmm, is this right?  My understanding of softdep (from mount(8)) was always
that it is a relaxation of the default synchronous metadata updating to
improve performance, with the updates ordered to ensure there is no safety
*cost* (or will be no safety cost once the bugs are out).  But does it carry
a safety *benefit*?  I didn't think so.

-Chap