JFFS2 deadlock with alloc_sem

[Roberts Nathan-mcg31137]

> > Has anyone seen this deadlock before? It seems to be a classic 
> > deadlock situation so I'm not sure if maybe I'm misinterpreting
things 
> > or the use case (several postmark tests running in parallel on a 
> > preemptible kernel) is especially vulnerable.
>
> I think Josh has spotted the real problem here. Does this help? If so,
as better fix will be forthcoming....

Thanks for the information. We tried the suggested modification and it
actually makes it much easier to reproduce the problem (so on one hand
it's a good thing.)

The steps we are using are:
1) Start postmark tests (usually 1 copy is sufficient but it is a little
easier to reproduce with multiple tests in parallel)
2) Wait for filesystem to get around 90% full 
3) Kill postmark (simple CTRL-C)
4) If filesystem is 100% full, delete postmark working directories
4) Restart postmark
5) goto 2

Normally within just a few iterations, we hang.