All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sascha Hauer <s.hauer@pengutronix.de>
To: Roman Fietze <roman.fietze@telemotive.de>
Cc: barebox@lists.infradead.org
Subject: Re: Freescale i.MX35 using CSD1 and 256 MiB DDR2
Date: Tue, 26 Apr 2011 23:56:30 +0200	[thread overview]
Message-ID: <20110426215630.GL14770@pengutronix.de> (raw)
In-Reply-To: <201104261430.01267.roman.fietze@telemotive.de>

Hi Roman,

On Tue, Apr 26, 2011 at 02:30:01PM +0200, Roman Fietze wrote:
> Hello Barebox Mailing List Members,
> 
> I found this advice to patch RedBoot to support more than 128 MiB DDR2
> on a Freescale i.MX35 3-stack eval board by enabling CSD1 and properly
> configuring the ESDRAMC.
> 
>   http://www.spinics.net/lists/u-boot-v2/msg03422.html
> 
> They simply seem to modify the boot code by omitting some checks and
> adding MMU mapping entries.
> 
> My changes are not successful yet. It seems I'm simply not yet
> familiar enough with the i.MX35, barebox, or even both. The changes
> that I added to barebox are at the end of this mail.
> 
> It seems that barebox uses the DCD table to boot the chip in my case,
> so I added new entries there, and I can see them when dumping e.g. the
> ESDRAMC registers using barebox itself (yes, it still starts. I can
> still dump the RAM starting at 0x80000000, but when I try to dump
> 0x90000000 barebox simply hangs.
> 
> I checked the IOMUX GPRA, the PAD settings for CS2 and CS3, the
> ESDRAMC registers and all locations of Freescale's document containing
> CS2, CS3, CSD0, CSD1, etc. with no success.
> 
> I'm stuck. What am I missing? A pointer to the appropriate section of
> the documentation or any helpful link would already very helpful.

There are at least two bugs in your flash header, see below.

> 
> 
> diff --git a/arch/arm/boards/freescale-mx35-3-stack/flash_header.c b/arch/arm/boards/freescale-mx35-3-stack/flash_header.c
> index 4bee797..34edceb 100644
> --- a/arch/arm/boards/freescale-mx35-3-stack/flash_header.c
> +++ b/arch/arm/boards/freescale-mx35-3-stack/flash_header.c
> @@ -17,19 +17,28 @@ struct imx_dcd_entry __dcd_entry_section dcd_entry[] = {
>  	{ .ptr_type = 4, .addr = 0xB8001010, .val = 0x0000030C, },
>  	{ .ptr_type = 4, .addr = 0xB8001004, .val = 0x007ffc3f, },
>  	{ .ptr_type = 4, .addr = 0xB8001000, .val = 0x92220000, },
> +	{ .ptr_type = 4, .addr = 0xB800100C, .val = 0x007ffc3f, },
> +	{ .ptr_type = 4, .addr = 0xB8001008, .val = 0x92220000, },
>  	{ .ptr_type = 4, .addr = 0x80000400, .val = 0x12345678, },
> +	{ .ptr_type = 4, .addr = 0x90000400, .val = 0x12345678, },
>  	{ .ptr_type = 4, .addr = 0xB8001000, .val = 0xA2220000, },

There is the equivalent for this at 0xB8001008 missing

>  	{ .ptr_type = 4, .addr = 0x80000000, .val = 0x87654321, },
> +	{ .ptr_type = 4, .addr = 0x90000000, .val = 0x87654321, },
>  	{ .ptr_type = 4, .addr = 0x80000000, .val = 0x87654321, },
> +	{ .ptr_type = 4, .addr = 0x90000000, .val = 0x87654321, },
>  	{ .ptr_type = 4, .addr = 0xB8001000, .val = 0xB2220000, },
> +	{ .ptr_type = 4, .addr = 0xB8001004, .val = 0xB2220000, },

This should be 0xB8001008

>  	{ .ptr_type = 1, .addr = 0x80000233, .val = 0xda, },
> +	{ .ptr_type = 1, .addr = 0x90000233, .val = 0xda, },
>  	{ .ptr_type = 1, .addr = 0x82000780, .val = 0xda, },
> +	{ .ptr_type = 1, .addr = 0x92000780, .val = 0xda, },
>  	{ .ptr_type = 1, .addr = 0x82000400, .val = 0xda, },
> +	{ .ptr_type = 1, .addr = 0x92000400, .val = 0xda, },
>  	{ .ptr_type = 4, .addr = 0xB8001000, .val = 0x82226080, },
>  	{ .ptr_type = 4, .addr = 0xB8001004, .val = 0x007ffc3f, },
> +	{ .ptr_type = 4, .addr = 0xB8001008, .val = 0x82226080, },
>  	{ .ptr_type = 4, .addr = 0xB800100C, .val = 0x007ffc3f, },
>  	{ .ptr_type = 4, .addr = 0xB8001010, .val = 0x00000304, },
> -	{ .ptr_type = 4, .addr = 0xB8001008, .val = 0x00002000, },
>  };
>  
>  
>  
>  	/* Skip SDRAM initialization if we run from RAM */
> -	cmp	pc, #0x80000000
> +	cmp	pc, #CSD0_BASE_ADDR
>  	bls	1f
> -	cmp	pc, #0x90000000
> +	cmp	pc, #CSD1_BASE_ADDR

This should be 0xa0000000 since with 256MB Ram this is still in SDRAM.

I Hope this helps.

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

_______________________________________________
barebox mailing list
barebox@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/barebox

  reply	other threads:[~2011-04-26 21:56 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-04-26 12:30 Freescale i.MX35 using CSD1 and 256 MiB DDR2 Roman Fietze
2011-04-26 21:56 ` Sascha Hauer [this message]
2011-04-27  8:38   ` [PATCH] freescale-mx35-3-stack: support 256 MiB RAM Roman Fietze
2011-04-27  9:14     ` Marc Kleine-Budde
2011-04-27 10:32       ` Roman Fietze
2011-04-27 10:34         ` Marc Kleine-Budde
2012-01-20 15:04           ` Thomas Mayer
2012-01-23  9:26             ` Sascha Hauer
2012-01-23 14:04               ` Thomas Mayer

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110426215630.GL14770@pengutronix.de \
    --to=s.hauer@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=roman.fietze@telemotive.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.