> It does serve to highlight that the filesystem guts of any filesystem
> are likely to need to be able to do I/O at some times without
> worrying about accepting new requests from userland; the modern crop
> of filesystems just seem to extend that "some time" beyond the
> duration of a single, synchronous system call...

This feels very reminiscent, to me, of the "deadlock against itself"
issues I ran into when doing my pseudo-disk driver, and the similar
deadlock issue with swapping to raidframe: in all cases you have
something asynchronously filling requests, but which resource-contends
with the very things that are generating requests for it.

I wonder if there's some shared problem that could benefit from being
pulled out, abstracted, and solved in the abstract, so that all of
these and any similar future deadlock dangers could be addressed by it.

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B