AHC / SCSI UPDATE

[Satoshi Asami]

Justin:

 * Sorry for not responding sooner, but I don't read this list regularly
 * anymore...

Which list you are talking about?  (I hope that's not scsi! :)

 * Remeber that it's not really the type of device that matters, but the
 * possibility of starvation.  If you have lots of concurrent I/O going on
 * to multiple disks on a single chain, you can still experience this problem
 * (Hi Satoshi!).

Yeah, we've seen it too.  Lots of IBM DCHS drives on a couple of
chains, lots of random I/O, and sooner or later we'll get the "timed
out while idle" on one of the drives.  (And then that drive spins
down, which has been a headache for us for quite a while...however,
recently Justin & co. have given us patches to fix this and we're
testing it now.)

Satoshi