AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20090331132408.4fa9b429@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:mail.onstor.net
NSV:
SSH:
R:<Bill.Fisher@onstor.com>
MAID:1
X-Sylpheed-Privacy-System:
X-Sylpheed-Sign:0
SCF:#mh/Mailbox/sent
RMID:#imap/andys@onstor.net@exch1.onstor.net/INBOX	0	49D26DE8.9060901@onstor.com
X-Sylpheed-End-Special-Headers: 1
Date: Tue, 31 Mar 2009 13:24:29 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: Bill Fisher <Bill.Fisher@onstor.com>
Subject: Re: tuxrx kernel crashes
Message-ID: <20090331132429.62c0a1cf@ripper.onstor.net>
In-Reply-To: <49D26DE8.9060901@onstor.com>
References: <20090330113442.63d15309@ripper.onstor.net>
	<20090330173221.44058853@ripper.onstor.net>
	<49D26DE8.9060901@onstor.com>
Organization: Onstor
X-Mailer: Sylpheed-Claws 2.6.0 (GTK+ 2.8.20; x86_64-pc-linux-gnu)
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

OK, the problem might be a merge issue, where something from my side
overwrote something from yours and broke the mgmt-bus code.  Like a
value in a header file or something.

Can you point me to your working tree?  I can diff a bunch of various
header files and hopefully find the problem.

On Tue, 31 Mar 2009 12:24:24 -0700 Bill Fisher <Bill.Fisher@onstor.com>
wrote:

> Andy Sharp wrote:
> > On Mon, 30 Mar 2009 11:34:42 -0700 Andrew Sharp <andy.sharp@onstor.com>
> > wrote:
> > 
> > 
> >>I tracked down the source of the crashes, it's because the mgmtbus code
> >>in git is not your latest code, at least I assume it isn't, because it's
> >>quite broken.  I looked through the mgmt_bus initialization code, and
> >>it's setting everything to the wrong addresses.  In fact, I think it's
> >>setting everything to the same address and also the wrong address.
> >>Hence the crash when it tries to access anything.
> >>
> >>Compare what you have in a working tree to what's in git in the mgmtbus
> >>driver directory and fix accordingly.
> >>
> Diffing the mgmt-bus.c and fixup-tuxrx.c code and there associated
> header files, between my working perforce tree and teh git tree
> shows no differences.
> 
> >>I actually spent about 10 minutes trying to fix it, but I think I only
> >>made it worse.  And then I decided that if this isn't your latest, then
> >>there's no point in me fixing just the initialization code.
> > 
> 
> 	You are obviously looking at the wrong code.
>  >
> > Hmm, possibly I was looking at the wrong code.  Or possibly not.  I
> > just discovered there's ssc-mgmt-bus directory and mgmt-bus directory,
> > and the latter seems like the correct one.  Yup, that's the one I was
> > looking at.  It must be missing some critical diffs from your working
> > code, I'm almost 100% positive of that.
> > 
> 
> The ssc-mgmtbus is the OLD code not the new stuff. That is now
> obsolete.
> 
> I will diff your kernel changes and see what has changed in that
> area. I have a perfecctly working tree that started prior to
> the git tree. Hence something is obviously different between
> them. I have proved that the mgmtbus and fixup code are
> identical, hence it stronly suggests that kernel mapping of
> that special region.
> 
> I will find the problem.
> 
> 
> 
> 
