Subject: Re: TECHNICAL INFTECHNICAL INFO ON THE AST CARD WANTED
To: Roland McGrath <roland@frob.com>
From: Brian Buhrow <buhrow@cats.ucsc.edu>
List: tech-kern
Date: 08/11/1994 08:53:36
I don't know what the original driver you wrote looked like, but the
one which is in the NetBSD distribution relies on the com driver itself to
set all the bits on the individual uarts. Thus, flags = 1 is a necessary
addition to the configuration file that wants to take advantage of the ast
driver. I believe we could add something to the ast multiplexer that puts
the appropriate values on the softc structure before control is ever passed
to the com driver. This would obviate the need for the flags stuff in the
configuration file and would make the use of the card, virtually "dummy
proof". Quite frankly, until I could actually see what sort of interrupt
activity was going on on my system, I had no idea what the problem was.
There were no hints about the problem in the documentation for the card,
which assumes you'll be running in enhanced mode under SCO Xenix. Thanks
to Bill for pointing out the problem, I'm a happy ast user, but I do have a
few more gray hairs because of the confusion going on.
-Brian
On Aug 11, 2:35am, Roland McGrath wrote:
} Subject: Re: TECHNICAL INFTECHNICAL INFO ON THE AST CARD WANTED
} When I wrote the ast driver originally, it disabled MCR_IENABLE for
} the multiplexed ports. I can't see a reason why it was changed not
} to.
>-- End of excerpt from Roland McGrath
.
------------------------------------------------------------------------------