Message-Id: <199512101821.KAA04338@vandys-lap.cisco.com>
To: Pai-Hsiang Hsiao <shawn@iis7.MIT.EDU>
Cc: VSTa mailing list <vsta@cisco.com>
Subject: Re: Interrupts arrive to fast? 
In-Reply-To: Your message of "Mon, 11 Dec 1995 01:42:59 +0800."
             <Pine.SUN.3.90.951211011817.1804A-100000@iis7> 
Date: Sun, 10 Dec 1995 10:21:09 -0800
From: Andy Valencia <vandys@cisco.com>

[Pai-Hsiang Hsiao <shawn@iis7> writes:]

>In cam server, after receiving M_ISR, driver speciic interrupts handling
>routines are called (in cam_ccb_wait()).

So you have seen at least one M_ISR message?

>But when I attach the codes for AHA2940 to cam, it hangs in msg_receive.
>(in cam_ccb_wait()).

>I add some watch points in kernel, in order to figure out what's going on
>in the dark side. It seems that interrupts arrive too fast.

If you have IDE as well, it can be really nice to boot the system up
entirely and then run cam under the debugger.

>The machine does not halt, the output from kernel are buffered, and being
>displayed after key strikes. Times the interrupts arrive are more than
>thousands, and are stored in m->arg1. (That's why I think I have no chance
>to receive the message)

arg1 only tallies interrupts up if the message is already queued under its
destination port, so it sounds like the M_ISR message is waiting to be
received on a queue somewhere.

>I don't know whether the problem is in my codes or not, if you have
>experience on this before, please give me some advices on this.

Mike Larson can comment on the SCSI aspect.  Overall, it sounds like the
sequence of interrupts and card interactions might be subtly different from
the 1542, and you're getting out of step with the controller.  Or, the
interrupts for the 2940 are getting directed to the wrong port.

							Andy