[8802] in linux-scsi channel archive
Re: DEC RZ55 and Advansys ABP3925
daemon@ATHENA.MIT.EDU (Kenn Humborg)
Sat May 13 15:47:56 2000
Date: Sat, 13 May 2000 20:29:54 +0100
From: Kenn Humborg <kenn@linux.ie>
To: Eric Youngdale <eric@andante.org>
Cc: linux-scsi@vger.rutgers.edu, linux@advansys.com
Message-ID: <20000513202954.A3139@avalon.research.wombat.ie>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
In-Reply-To: <20000513194101.A1305@avalon.research.wombat.ie>; from kenn@linux.ie on Sat, May 13, 2000 at 07:41:01PM +0100
On Sat, May 13, 2000 at 07:41:01PM +0100, Kenn Humborg wrote:
> Scanning of ID 0 is now silent, and that makes things a bit
> different when the card interrupts after the scan of ID 1. From
> my reading of this, it looks like either:
>
> o The card is interrupting, but doesn't actually have any results
> for us. AscISR() doesn't find any work to do, so asc_isr_callback
> never gets called.
>
> o Something else is interrupting on IRQ 11. I don't think this is
> likely because I can see all my cards generating interrupts on
> other IRQs (via /proc/interrupts).
>
> I'm going to try printk()s in AscISR and friends and see if I can
> figure out what condition is causing asc_isr_callback() not to be
> called.
I've found out that AscISR() is calling AscIsrChipHalted().
AscIsrChipHalted() gets a halt code of 0 from the card, which
isn't something it's able to deal with. I've tried adding
a catch-all else clause to the end of AscIsrChipHalted() to
clear the halt condition and return both 0 and ERR. Neither
makes any difference.
I guess it is now definitely an Advansys issue, rather than a
generic Linux SCSI one. So over to linux@advansys.com!
Later,
Kenn
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@vger.rutgers.edu
