AF:
NF:0
PS:10
SRH:1
SFN:
DSR:
MID:<20090511150905.381de8ff@ripper.onstor.net>
CFG:
PT:0
S:andy.sharp@onstor.com
RQ:
SSV:mail.onstor.net
NSV:
SSH:
R:<rendell.fong@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	2779531E7C760D4491C96305019FEEB52AC8BC1D6F@exch1.onstor.net
X-Sylpheed-End-Special-Headers: 1
Date: Mon, 11 May 2009 15:09:28 -0700
From: Andrew Sharp <andy.sharp@onstor.com>
To: Rendell Fong <rendell.fong@onstor.com>
Subject: Re: wka kernel available
Message-ID: <20090511150928.6ca8a7ee@ripper.onstor.net>
In-Reply-To: <2779531E7C760D4491C96305019FEEB52AC8BC1D6F@exch1.onstor.net>
References: <20090511143522.2c5975c7@ripper.onstor.net>
	<2779531E7C760D4491C96305019FEEB52AC8BC1D6F@exch1.onstor.net>
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

*%*%^%$#& that code is so broken.

We will emulate this, possibly fixing it somehwhat.  The

*shmem_ptr = 0;

isn't necessary, for instance, and its removal is possibly all that
is needed to fix it.

Do you need it now?  Does this memory in the WKA area?

On Mon, 11 May 2009 14:41:20 -0700 Rendell Fong
<rendell.fong@onstor.com> wrote:

> Take a look at sm-eee/eee_init.c: eee_initSync().
> 
> Both txrx and fp normally run this routine.  The while loop in there
> will poll forever waiting for the other side to reach the equivalent
> point during initialization before continuing on.
> 
> 
> > -----Original Message-----
> > From: Andy Sharp
> > Sent: Monday, May 11, 2009 2:35 PM
> > To: Rendell Fong
> > Subject: Re: wka kernel available
> > 
> > On Mon, 11 May 2009 14:15:05 -0700 Rendell Fong
> > <rendell.fong@onstor.com> wrote:
> > 
> > >
> > >
> > > > -----Original Message-----
> > > > From: Andy Sharp
> > > > Sent: Monday, May 11, 2009 1:39 PM
> > > > To: Rendell Fong
> > > > Subject: Re: wka kernel available
> > > >
> > > > On Mon, 11 May 2009 13:17:26 -0700 Rendell Fong
> > > > <rendell.fong@onstor.com> wrote:
> > > >
> > > > > The init of the eee code is staged and normally syncs up
> > > > > between the txrx and fp during the process.  This means that
> > > > > I won't be able to get too far unless the eee init code in
> > > > > both tuxrx and fp are running simultaneously.
> > > > >
> > > > > In tuxrx, do we have the equivalent of a startup routine (like
> > > > > sm-test/test.c) yet?  If so, where can I find it?
> > > >
> > > >
> > > > You need to give more exact details of what you need.
> > >
> > > Where in tuxrx during the init process is eee supposed to be
> > > initialized?
> > 
> > So, we need some synchronization method?  I suppose then that we
> > would want to emulate what EEE does now, but I don't know what that
> > is.
> > 
> > > > > Rendell
> > > > >
> > > > >
> > > > > > -----Original Message-----
> > > > > > From: Andy Sharp
> > > > > > Sent: Friday, May 08, 2009 6:57 PM
> > > > > > To: Rendell Fong
> > > > > > Cc: Bill Fisher
> > > > > > Subject: wka kernel available
> > > > > >
> > > > > > Hi Rendell,
> > > > > >
> > > > > > I don't know if this is all you need to get going, but I
> > > > > > created a kernel which sets aside 1MB for a WKA area.  Yes,
> > > > > > I know you only need 8KB, but hey, this is a small start,
> > > > > > so to speak.  I'll figure out the best way to do 8K later.
> > > > > > The address is 0xff00000.  Later on I'll figure out some
> > > > > > way to allocate this better and place it in the unused
> > > > > > mgmtbus area so the FP can just read it at init time.
> > > > > >
> > > > > > ~andys/rendell-wka-kernel.tar is a tarball of the kernel.
> > > > > > You should be able to take the vmlinux.bin right out of
> > > > > > there and load it on your tuxrx.  I'm giving you the whole
> > > > > > tar ball just in case you need to compile something or see
> > > > > > something.  The code that sets up the area is the last
> > > > > > function in arch/mips/onstor/tuxrx/setup.c.
> > > > > >
> > > > > > Cheers,
> > > > > >
> > > > > > a
> > > > > >
> > > > > > PS Bill, I finally fixed the code to put the mapped mgmtbus
> > > > > > PCI area into /proc/iomem so you can see it ~:^)
