On Thu, 4 Jan 96 10:59:24 PST 
 greywolf@defender.VAS.viewlogic.com (Incorrigible punster -- do not incorrige) wrote:

 > Interesting approach, though I don't see why it's considered inelegant.
 > You have a flag which says "do this", and if it works, and it's relatively
 > simple, what else matters?

Well, because it adds a special-case global.  "Ick."

 > The only other way I can think of to do this is to implement another lower-
 > half of a driver as a "*dumpstrat()" part of a [bc]devsw, and I don't
 > believe such a generic routine exists.
 > 
 > It might be nice to have a "sd_dumpstrat()" and/or "st_dumpstrat()",
 > since then you wouldn't have to check that "inelegant" flag above...

Having a "dumpstrat" doesn't really help the fact that the SCSI driver 
wanted to do DVMA ... That's below the level of the disk driver itself.

Actually, Gordon's suggestion of using the SR_IMMED (?? ... I have the 
mail somewhere, but I know what he's referring to) flag to force PIO will 
work, because that's really only used during autoconfiguration, at the 
moment, as far as I can tell.

--------------------------------------------------------------------------
Jason R. Thorpe                                       thorpej@nas.nasa.gov
NASA Ames Research Center                               Home: 408.866.1912
NAS: M/S 258-6                                          Work: 415.604.0935
Moffett Field, CA 94035                                Pager: 415.428.6939