The legacy interface is confusing. It's better to make virtio_disk
conform to the virtio spec. This is supported in QEMU since 4.2 by
disabling force-legacy for virtio-mmio.
Use refcnt to pin blocks into the cache
Replace flags/B_VALID with a boolean field valid
Use info[id].status to signal completion of disk interrupt
Pass a read/write flag to virtio_disk_rw
procinit() and map them high up (below TRAMPOLNE) with an empty
mapping below each stack. Never free a kernel stack.
Another way would be to allocate and map them dynamically, but then we
need to reload page table when switching processes in scheduler()
and/or have a kernel pagetable per proc (if we want k->stack to be the
same virtual address in each process).
One gotcha: kernel addresses are not equal to physical addresses for
stack addresses. A stack address must be translated if we need its
physical address (e.g., virtio passes a stack address to the disk).
