[U-Boot-Users] AMD dbau1550

[U-Boot-Users] AMD dbau1550

[Leif Lindholm]

Hi

I've been working on getting u-boot 1.1.1 up and running on <subject>,
in little endian mode (which someone asked for a few days ago I
believe).

I've been able to get as far as "Starting kernel ..." after downloading
the kernel u-boot image through tftp.

I know this tecnically means u-boot is finished, however the same kernel
(in srec format) boots fine from the original yamon loader on the board.

When I halt the processor through jtag (bdi2000), the PC is always at
0xbfc00424 (somewhere in ROM exception land?).

The loaded image has been configured using 
tools/mkimage -C gzip -a 0x80100000 -e 0x80246040 -A mips -O linux -T kernel -d vmlinux.gz vmlinux.u-boot

0x80246040 is the location of "kernel_entry".

Does anyone have any hints about how I could work my way past this problem?

regards
	/
		Leif

[U-Boot-Users] AMD dbau1550

[Dan Malek]

On May 24, 2004, at 11:22 AM, Leif Lindholm wrote:

> Does anyone have any hints about how I could work my way past this 
> problem?

The kernel expects lots of setup done by yamon for board configuration,
environment and kernel start parameters.  Without major changes to the
kernel start up functions to extract similar information from a u-boot
environment, the kernel isn't going to boot up.


	-- Dan

[U-Boot-Users] AMD dbau1550

[Thomas Lange]

The first thing I would check, is the parameters passed to kernel.
E.g. wrong memory size will make things really bad.

Default behaviour for dbau1x00 u-boot is to pass mem size in bytes.
Does your kernel expect that?

I suggest activating DEBUG macro in lib_mips/mips_linux.c

/Thomas

Leif Lindholm wrote:
> Hi
> 
> I've been working on getting u-boot 1.1.1 up and running on <subject>,
> in little endian mode (which someone asked for a few days ago I
> believe).
> 
> I've been able to get as far as "Starting kernel ..." after downloading
> the kernel u-boot image through tftp.
> 
> I know this tecnically means u-boot is finished, however the same kernel
> (in srec format) boots fine from the original yamon loader on the board.
> 
> When I halt the processor through jtag (bdi2000), the PC is always at
> 0xbfc00424 (somewhere in ROM exception land?).
> 
> The loaded image has been configured using 
> tools/mkimage -C gzip -a 0x80100000 -e 0x80246040 -A mips -O linux -T kernel -d vmlinux.gz vmlinux.u-boot
> 
> 0x80246040 is the location of "kernel_entry".
> 
> Does anyone have any hints about how I could work my way past this problem?
> 
> regards
> 	/
> 		Leif
> 
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: Oracle 10g
> Get certified on the hottest thing ever to hit the market... Oracle 10g. 
> Take an Oracle 10g class now, and we'll give you the exam FREE.
> http://ads.osdn.com/?ad_id=3149&alloc_id=8166&op=click
> _______________________________________________
> U-Boot-Users mailing list
> U-Boot-Users at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/u-boot-users
> 

[U-Boot-Users] AMD dbau1550 (resolved)

[Leif Lindholm]

Hi, and thanks for your suggestions.

It turned out, however, to be a simple RTFM problem.
I was trying to load the ELF version of the kernel, which made my load
address and entry point invalid.

Since I used the dbau1x00 code as base, the parameter/memory size
passing was ok (after I changed the hard-coded 64Mb to hard coded 128Mb
:)

Am looking into some way to integrate this stuff into mainline.

regards
	/
		Leif

[U-Boot-Users] AMD dbau1550 (resolved)

[Wolfgang Denk]

In message <1085509772.19990.39.camel@callisto.smurfnet.se> you wrote:
> 
> Am looking into some way to integrate this stuff into mainline.

See the README, section "Submitting Patches".

Best regards,

Wolfgang Denk

-- 
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd at denx.de
Faith may be defined briefly as an illogical belief in the  occurence
of the improbable.                                    - H. L. Mencken