Hi,
Due to the recently discussed IO stalls in 2.4, I had a
bit of a look at the request allocator in 2.5 and found a
problem. The same problem does exist in 2.4 as well. It
could cause the same symptoms as these stalls. It would
be nice to know if its really the cause.

It would be nice if anyone interested could test.
Patch compiles, boots and survives tiobench, etc. here.
Jens thought the patch (2.5 version) looked OK.

Thanks,
Nick

Previously:
* request queue fills up
* process 1 calls get_request, sleeps
* a couple of requests are freed
* process 2 calls get_request, proceeds
* a couple of requests are freed
* process 2 calls get_request, proceeds
...

Now as unlikely as it seems, it could be a problem. Its a fairness
problem that process 2 can skip ahead of process 1 anyway.

With the patch:
* request queue fills up
* any process calling get_request will sleep
* once the queue gets below the batch watermark, processes
 start being worken, and may allocate.