Add support for the hammerhead (AVR32) board

 MAKEALL                                  |    1 +
 Makefile                                 |    3 +
 board/miromico/hammerhead/Makefile       |   40 +++++++
 board/miromico/hammerhead/config.mk      |    3 +
 board/miromico/hammerhead/eth.c          |   37 ++++++
 board/miromico/hammerhead/hammerhead.c   |  105 +++++++++++++++++

 board/miromico/hammerhead/u-boot.lds     |   73 ++++++++++++
 cpu/at32ap/at32ap700x/gpio.c             |   11 ++
 cpu/at32ap/at32ap700x/sm.h               |    2 +-
 cpu/at32ap/cpu.c                         |    5 +
 include/asm-avr32/arch-at32ap700x/clk.h  |    1 +
 include/asm-avr32/arch-at32ap700x/gpio.h |    3 +
 include/configs/hammerhead.h             |  179 ++++++++++++++++++++++++++++++
 net/eth.c                                |    4 +

Hi & welcome back from holiday :)

According to your request...

Date: Tue, 22 Jul 2008 14:20:15 +0200
From: linux at bohmer.net
To: u-boot-users at lists.sourceforge.net; wd at denx.de
CC: joakim.tjernlund at transmode.se
Subject: [U-Boot-Users] [patch 1/1] Set GD_FLG_RELOC for boards skipping relocation to RAM

If CONFIG_SKIP_RELOCATE_UBOOT is set the flag GD_FLG_RELOC is usually never set, because relocation to RAM is actually never done by U-boot itself. However, several pieces of code check if this flag is set at some time.

So, to make sure this flag is set on boards skipping relocation, this is added to the initialisation of U-boot at a moment where it is safe to do so.

Signed-off-by Remy Bohmer <[email protected]>

---
lib_arm/board.c | 16 ++++++++++++++++

1 file changed, 16 insertions(+)

Index: u-boot-git-almost-1.3.4/lib_arm/board.c
===================================================================

lib_arm/board.c 2008-07-21 12:25:32.000000000 +0200 renamed from lib_arm/board.c 2008-07-21 11:53:23.000000000 +0200 +11 -0

@@ -233,6 +233,17 @@ static int init_func_i2c (void)
 }
 #endif
 
+#ifdef CONFIG_SKIP_RELOCATE_UBOOT
+/*
+ * This routine sets the relocation done flag, because even if
+ * relocation is skipped, the flag is used by other generic code.
+ */
+static int reloc_init(void)
+{
+ gd->flags |= GD_FLG_RELOC;
+}
+#endif
+
 /*
 * Breathe some life into the board...
 *

@@ -262,6 +273,11 @@ int print_cpuinfo (void); /* test-only *
 
 init_fnc_t *init_sequence[] = {

cpu_init, /* basic cpu dependent setup */

lib_arm/board.c 2008-07-21 12:25:32.000000000 +0200 renamed from lib_arm/board.c 2008-07-21 11:53:23.000000000 +0200 +5 -0

+#if defined(CONFIG_SKIP_RELOCATE_UBOOT)
+ reloc_init, /* Set the relocation done flag, must
+ do this AFTER cpu_init(), but as soon
+ as possible */
+#endif

board_init, /* basic board dependent setup */
interrupt_init, /* set up exceptions */
env_init, /* initialize environment */

--

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
U-Boot-Users mailing list
U-Boot-Users at lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/u-boot-users

when would you not want to relocate ?

Quoted 1 line · Julien May

Hi & welcome back from holiday :)

Thanks :)

Quoted 1 line

According to your request...

This was rather more than I asked for, and it didn't apply to the hammerhead branch I already had.

After resolving the conflicts, I ended up with just a change to MAINTAINERS, so I've committed only that part. The rest is already there (except the net/eth.c stuff which shouldn't be necessary anymore after Ben's patch hit mainline)

Could you have a look at the result at

git://git.denx.de/u-boot-avr32.git hammerhead

If it looks ok to you (and still works), I'll fold it all into a single patch tomorrow. The commit messages are somewhat bogus...I'll fix it up at the same time.

Haavard

Currently I even cannot compile it, due to some not only to our board related errors. I'll check them and give you feedback asap or not later than tomorrow.

Quoted 3 lines · Julien May

Currently I even cannot compile it, due to some not only to our board related errors. I'll check them and give you feedback asap or not later than tomorrow.

Oh...right. I did fix those, but I forgot to pull them in.

Quoted 6 lines · Haavard Skinnemoen

Quoted 3 lines · Julien May

Currently I even cannot compile it, due to some not only to our board related errors. I'll check them and give you feedback asap or not later than tomorrow.

Oh...right. I did fix those, but I forgot to pull them in.

Done. I've also applied a couple of other fixes due to recent API changes, so it should at least build now.

Haavard

Quoted 2 lines · Haavard Skinnemoen

Done. I've also applied a couple of other fixes due to recent API changes, so it should at least build now.

Ah great!

Are those already committed to your hammerhead branch? Right now I couldn't pull anything new.

Maybe you can check those in so that I can pull and test the u-boot.

I am currently having problems in initializing the eth.

in net/eth.c the following is defined

static int __def_eth_init(bd_t *bis)

{
	return -1;
}

int cpu_eth_init(bd_t *bis) __attribute((weak, alias("__def_eth_init"))); int board_eth_init(bd_t *bis) __attribute((weak, alias("__def_eth_init")));

this is not calling my implementation of board_eth_init. I see and like the idea behind this but do not understand why my implementation is not getting called...

btw. I did update the implementation function header from void to int.

Quoted 11 lines · Julien May

I am currently having problems in initializing the eth.

in net/eth.c the following is defined

static int __def_eth_init(bd_t *bis)

{
	return -1;
}

int cpu_eth_init(bd_t *bis) __attribute((weak, alias("__def_eth_init"))); int board_eth_init(bd_t *bis) __attribute((weak, alias("__def_eth_init")));

Quoted 3 lines

this is not calling my implementation of board_eth_init. I see and like the idea behind this but do not understand why my implementation is not getting called...

Hmm, that's probably the "weak functions cannot be overriden by functions defined in their own file" crap that was discussed earlier. Can you try moving board_eth_init() into hammerhead.c?

Which reminds me...Ben posted a patch which did that for all the existing avr32 boards. I should probably apply it.

Quoted 2 lines

btw. I did update the implementation function header from void to int.

Yeah, that's probably a good idea too :-)

Haavard

Quoted 3 lines

Hmm, that's probably the "weak functions cannot be overriden by functions defined in their own file" crap that was discussed earlier. Can you try moving board_eth_init() into hammerhead.c?

Did so and it works now. I could make for this an incremental patch and send this to you.

Quoted 2 lines

Which reminds me...Ben posted a patch which did that for all the existing avr32 boards. I should probably apply it.

Otherwise I wait until you applied bens patch and test again hammerheads functionality in u-boot.

Quoted 4 lines

Did so and it works now. I could make for this an incremental patch and send this to you.

That would be great.

Please find below the incrementel patch. This patch just fixes the eth stuff of our board. As hopefully understood right the patch from ben sould fix the rest.

Hi John,

look at
http://www.itee.uq.edu.au/~listarch/microblaze-uclinux/archive/2008/05/msg00163.html

Please use

Quoted 46 lines

Hi, again.

I've used a new vendor/platform, I have the uclinux running in the same target a ML501 with an older version based on ise and edk 8.2, but I need approach the advantages of microblaze v7.00 and the PLB. Please thanks for your help, and excuse my English.

I don't know about inittab, and that init is selected in the application menu.

2008/7/24 Mostafa Ali <[email protected]>:

Quoted 33 lines

Look there are so many reasons for this problem, Please mention more detalis(like have you used one of the predefined platforms in the vendor/platform menu or you created a new one).

One of the reasons is that the kernel doesn't find a start up application. Make sure that inittab is exist in /romfs and init is selected in the application menu in petalinux configuration wizard. I need more de tails fro you, I'll be waiting.

--- On *Thu, 7/24/08, John Osorio <[email protected]>* wrote:

From: John Osorio <[email protected]>
Subject: [microblaze-uclinux] Help with kernel boot, PLEASE
To: microblaze-uclinux at itee.uq.edu.au
Date: Thursday, July 24, 2008, 8:36 AM

Hi, folks

I have a problem with the kernel boot, I'm using petalinux-v0.30-rc1 and kernel 2.6, my hardware is development under EDK 9.2_02 and ISE 9.2_04. When I'll try boot the image, the boot show in kermit the following:

## Booting image at 8c080000 ...

   Image Name:   PetaLinux Kernel 2.6
   Image Type:   Microblaze Linux Kernel Image (uncompressed)
   Data Size:    2678812 Bytes =  2.6 MB

Load Address: 90000000
Entry Point: 90000000
Verifying Checksum ... OK
OK

Quoted 14 lines

Quoted 8 lines

That's all I don't know, what's the problem, Please I need your Help, Bye. The boot is freeze, Kermit don't show anything

--
John Osorio

Signature · 1 line

John Osorio

No virus found in this incoming message.
Checked by AVG - http://www.avg.com
Version: 8.0.138 / Virus Database: 270.5.5/1571 - Release Date: 24.7.2008 05:42