Skip site navigation (1)Skip section navigation (2)
Date:      13 May 1999 07:43:42 -0700
From:      Frank McConnell <fmc@reanimators.org>
To:        Warner Losh <imp@harmony.village.org>
Cc:        freebsd-stable@FreeBSD.ORG
Subject:   Re: Is aha broken?
Message-ID:  <199905131443.HAA00961@daemonweed.reanimators.org>
In-Reply-To: Warner Losh's message of "Wed, 12 May 1999 22:18:21 -0600"
References:  <199905121429.HAA01982@daemonweed.reanimators.org>  <Pine.BSF.4.10.9905111124070.34934-100000@germanium.xtalwind.net> <199905130418.WAA26127@harmony.village.org>

next in thread | previous in thread | raw e-mail | index | archive | help
Warner Losh <imp@harmony.village.org> wrote:
> I've never used aha on the SMP machine.  Usually this error is an
> indication that one of two things.  First one is a termination error.
> The second is a IRQ/DRQ mismatch.  Usually a termination. 

I thought about termination before writing, but the cabling is the
same cabling that it had when it was in the 2.2.1-STABLE box, I
thought it was terminated properly (but does the 1542CP do
auto-termination on its internal connector if there's nothing
connected there? -- can't find TFM), I've been over it just in case I
wiggled a connector loose during the box swap, and I've replaced a 6'
cable with a 3' just in case.  No joy doing that.

But this morning I changed the card's idea of its IRQ to 9 and it all
works with no "create_intr" complaint.  I'm guessing that IRQ 11 was
falling foul of (and being changed to -1 by) the "#ifdef APIC_IO" code
just before the call to register_intr() in config_isadev_c(), but as
I'm not sure what is going on in that code I think I'm just going to
rejoice and do a backup now that I can get to my tape drive.

> The driver doesn't use the hardware status bits that can tell if the
> cable configuration is illegal, nor does it tell you if the
> termination is bad.  Plug and play is poorly supported.  It seems that
> works sometimes.  However, I've had reports of problems which I've
> been unable to recreate here.

OK, here's a couple of additional data points: plug and play is disabled
on this card, and this is all now plugged into a Tyan S1832DL "Tiger 100"
motherboard with two CPUs.

-Frank McConnell


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199905131443.HAA00961>