From mboxrd@z Thu Jan  1 00:00:00 1970
From: Mike Frysinger <vapier@gentoo.org>
Date: Sat, 19 Nov 2011 15:17:27 -0500
Subject: [U-Boot] [PATCH] Add PR1 Appliance - ISDN PRI board
In-Reply-To: <003901cca6ce$76650ef0$2901a8c0@dpn>
References: <014501cca57b$1d65e800$2901a8c0@dpn>
	<201111190042.52468.vapier@gentoo.org>
	<003901cca6ce$76650ef0$2901a8c0@dpn>
Message-ID: <201111191517.28399.vapier@gentoo.org>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On Saturday 19 November 2011 10:18:16 Dimitar Penev wrote:
> >> +void board_reset(void)
> >> +{
> >> + /* workaround for weak pull ups on ssel */
> >> + if (CONFIG_BFIN_BOOT_MODE == BFIN_BOOT_SPI_MASTER)
> >> +  bfin_reset_boot_spi_cs(GPIO_PF10);
> >> +}
> >
> >do you actually have this problem in the hardware ?  or is this just left
> >over
> >from copying the old bf537 file ?
> 
> PR1 Appliance seems to boot from SPI FLASH properly without the
> board_reset().
> From the other side we use 10KOhm on the SSEL pin the same as in
> BF537-STAMP I think
> so I am not sure why we will have no issue if it is observed in
> BF537-STAMP. Removing  board_reset() for now

i don't think the issue was ever observed on a bf537-stamp board.  in looking
at the convoluted history, the original report came in due to a customer
board:	http://blackfin.uclinux.org/gf/tracker/1401

unfortunately, the kernel and u-boot blackfin ports at that time were so bad,
that the change went into common code.  which means later on, as things were
untangled, they were pushed out into the boards files.  part of that is my
fault for not fully figuring out what was going on.

so this should get punted for the bf537-stamp board and others.  i'll take
care of that.

> >> +#define CONFIG_SYS_NO_FLASH            /* we have only NAND */
> >
> >not true ... you have SPI too ;)
> 
> Or do you refer to my comment only? Fixed.

yes

> >> +#define CONFIG_NET_MULTI 1
> >
> >no longer needed -> delete
> 
> The included patch is with the above line removed as per your advice.
> 
> If I test with the 2011R1-RC3 from the ADI repository I still need it
> however.
> Note that I have tested the patch (with CONFIG_NET_MULTI ) against the ADI
> repository
> (not the mainstream) and this is what I can confirm as working on PR1
> Appliance.

right, NET_MULTI is needed for the last release, but to get into mainline
u-boot, you'll need to drop it :).  so i'll merge it into the u-boot mainline,
but it won't be part of ADI's 2011R1 release.  it should make it into the
release after that though (2012R1).

what do you think of the attached patch ?
-mike