Am I right in thinking that Russell worked out a cunning fix that circumvents
the K StrongARM bug on ARMLinux, from version 2.2.6 or thereabouts.
Is how it works documented anywhere?
Therefore, could the ARM BSD source be similarly patched, so that BSD users on
K didn't have to upgrade [to T or to Linux :-)]?
Nick
----- Forwarded message from Kjetil B Thomassen <[EMAIL PROTECTED]> -----
Return-Path: <[EMAIL PROTECTED]>
Received: from mail.netbsd.org (redmail.netbsd.org [155.53.200.193])
by flirble.org (8.9.1/8.8.5) with SMTP id MAA19649
for <[EMAIL PROTECTED]>; Tue, 30 Nov 1999 12:20:54 GMT
Received: (qmail 9675 invoked by uid 605); 30 Nov 1999 12:20:32 -0000
Received: (qmail 9669 invoked from network); 30 Nov 1999 12:20:27 -0000
Received: from login-2.eunet.no ([EMAIL PROTECTED])
by redmail.netbsd.org with SMTP; 30 Nov 1999 12:20:27 -0000
Received: from login-1.eunet.no ([EMAIL PROTECTED] [193.71.71.238])
by login-2.eunet.no (8.9.3/8.9.3/GN) with ESMTP id NAA64583;
Tue, 30 Nov 1999 13:20:25 +0100 (CET)
(envelope-from [EMAIL PROTECTED])
Received: from localhost (kjetilbt@localhost)
by login-1.eunet.no (8.9.3/8.8.8) with ESMTP id NAA30414;
Tue, 30 Nov 1999 13:20:24 +0100 (CET)
(envelope-from [EMAIL PROTECTED])
X-Authentication-Warning: login-1.eunet.no: kjetilbt owned process doing -bs
Date: Tue, 30 Nov 1999 13:20:24 +0100 (CET)
From: Kjetil B Thomassen <[EMAIL PROTECTED]>
X-Sender: [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
To: Angelo Melis <[EMAIL PROTECTED]>
cc: [EMAIL PROTECTED]
Subject: Re: Two questions ...
In-Reply-To: <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: [EMAIL PROTECTED]
Precedence: list
Delivered-To: [EMAIL PROTECTED]
On Tue, 30 Nov 1999, Angelo Melis wrote:
> Some time ago I installed NetBSD-arm32 1.4.1 on my RISC PC and after some
> trouble with the install kernel I got it installed. And it works OK. I even
> have access to the internet throu the Atomwide serial card.
>
> But this weekend I tried to compile some stuff I downloaded and got some
> problems. the compiler crashed with a signal 11 (????) and core-dumped.
> Sometimes cc1 and sometimes cpp crashed. After deleting the coredump and
> restarting the make process ... it all worked.
>
> Today I tried to update some packages on my system and noticed the same
> problems. But it is not only cc1 or cpp, but also pkg_delete or some other
> program crashes with a signal 11.
>
> My question is ... what is signal 11 ? And how can I make sure that this
> doesn't keep happening, because it is annoying.
This sounds like a revision J or K StrongARM SA-110. There is a bug in
these that can lead to problems like this.
In order to solve this, you have to upgrade your SA-110 to revison S
or T, something that Simtec can do for GBP 100. See an earlier article
regarding this.
Another way to solve it, would be to replace your SA-110 with an
ARM610 or ARM710, but this would of course degrade performance.
There are ways to code around this problem, but they are too complex
for anyone to have done anything about this.
To figure out which revision your SA-110 is, just look at the dmesg
output, or check the letter on the physical chip. My revision K claims
that it is revision 2, so you need one that is newer than that.
This whole problem was discussed in great detail a year or two ago,
so if you check the mailing list archive, you will be able to get
more details.
> Secondly ... after booting ... I get an error from kvm_read (?) with an
> illegal address (0x0). Is this something I must worry about ? Is it connected
> to the core dump problem ?
It may be, but I don't know.
Kjetil B.
mailto:[EMAIL PROTECTED]
http://home.eunet.no/~kjetilbt/
----- End forwarded message -----
unsubscribe: body of `unsubscribe linux-arm' to [EMAIL PROTECTED]
++ Please use [EMAIL PROTECTED] for ++
++ kernel-related discussions. ++
