FYI, I have gotten some most strange results with the equivalent of:

     request_queue_t *q = kmalloc(...);
     blk_init_queue(q, ...)

The problem turns out to be that q->backing_dev_info.memory_backed is not
initialized, leaving it to be whatever was there before.  The end result is
that my "driver" never saw write requests, but reads came through fine.
Took a while to figure out that one...

Here's the fix, for what it's worth.  (I had, BTW, tried just zeroing the
whole structure in blk_init_queue(), but that makes IDE oops at boot - it
sets q->queuedata early).