use byte swapping loads and stores.

shrinks the code a bit on sparc64.

ok jmatthew@

make qla_iocb_seg structs 4 byte aligned and use htolem32 to set it.

shaves ~256 bytes off of qla_put_data_seg on sparc64

ok kettenis@ jmatthew@

Always copy out all mailbox registers and use msleep in a loop to ensure that
post-attach mailbox operations work reliably.

set adapter buswidth according to the number of loop ids supported by the
firmware (2048 for 2k login firmware, 256 otherwise)

pointed out by deraadt@, ok dlg@

Rather than scanning local loop ids sequentially, just ask the chip which
local devices it has logged in to. This will also be useful for hotplug
later on.

missing define for qla2100 support

Add support for ISP2200 based adapters.

This involves a bit of a hack in the interrupt processing path, where we
ignore the interrupt status if we're expecting a mailbox command to complete,
but we don't have the mailbox semaphore and the response ring hasn't moved.
Typically next time through, the interrupt status is the same and we have the
semaphore so we can read the results.

Stop using 23xx-specific interrupt status codes everywhere. Instead, classify
into one of 4 types in qla_read_isr.

Introduce qla(4), a new driver for Qlogic fibre channel HBAs (only ISP23xx
so far). Works reasonably well now, so it's going in the tree so others can
try it out.

with some help and fixes from dlg@, general encouragement from basically
everyone