Re: Why my console output nothing

Re: Why my console output nothing

[rick.liu]

Dear Wolfgang

Thanks for ur reply,in your message you wrote:
======= 2003-05-13 15:46:00 =======

>Dear Rick,
>
>
>Chances for useful  replies  are  much  better  in  the  u-boot-users
>mailing list instead.

Later I will try :-)

>
>Chances for useful replies would be even better  if  you  provided  a
>little more information. "modify several line of ..." is not really a
>precise description of what you did.
>

you mean it's not the problem of my console configuration but the u-boot
program?
But I thought it might at least output some message of the board before it can
output the prompt "=>",is that right? Listing all my chang in the TQM860L.h
may help:

#define	CONFIG_CLOCKS_IN_MHZ	50 /*50Mhz osc connected to 860_EXTCLK*/
#define CONFIG_BOARD_TYPES	0     /*It's a simple board ,supported?*/
#define CFG_OR0_PRELIM	0xfff80920  /*boot flash is AM29LV040B,512k*/
#define CFG_BR0_PRELIM	0x40000401  /* magic numbers of my modify take
                                       /*reference from a working vxworks
BSP's
                                       /* romInit.s of the same board*/

/* #define SDRAM_BASE2_PRELIM	0x00000000*/
#define SDRAM_BASE4_PRELIM	0x00000000  /*cs4 is sdram*/

#define CFG_OR4_PRELIM	0xfe000f60
#define CFG_BR4_PRELIM	0x00000081

#define CFG_MAMR_11COL	0x10924111 /*11 col,here is the value I write to mamr
in tqm8xx.c*/

#define CONFIG_SCC1_ENET	1 /*should add a number?*/
/*#define CONFIG_FEC_ENET*/ /*no FEC*/

and in tqm8xx.c,I did like this:

1.changed upm_table according to my sdram;
2.modify the register's value in the function initdram()


>
>Porting U-Boot to new hardware requires more  than  just  tweaking  a
>line  here  and there. It will require both an in-depth understanding
>of the hardware and U-Boot.
>

:-),I am trying to learn,would u  recommend some helpful material about
u-boot?

>
>Did you read the README that comes with U-Boot?
>
yes

>Finally: did you order your BDI2000 yet? You will  probably  need  it
>soon.

I know that,and I will try to persuade my "boss".

Best regards,
Rick Liu





--http://www.eyou.com
--Îȶ¨¿É¿¿µÄÃâ·Ñµç×ÓÐÅÏä  ÓïÒôÓʼþ  Òƶ¯ÊéÇ©  ÈÕÀú·þÎñ  ÍøÂç´æ´¢...ÒÚÓÊδ¾¡


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Why my console output nothing

[Wolfgang Denk]

Hello,

in message <20030514152700.68497.qmail@eyou.com> you wrote:
>
> you mean it's not the problem of my console configuration but the u-boot
> program?

That might be the case. But obviously I can only speculate.

> But I thought it might at least output some message of the board before it can
> output the prompt "=>",is that right? Listing all my chang in the TQM860L.h

This is correct. But you need to get a couple of things right  before
you see a single character onthe console.

> and in tqm8xx.c,I did like this:

You should not change existing files in an incompatible way.  Instead
you  should  add  your  own set of files for your hardware. Otherwise
there will not be any chance to include your modifications  intot  he
public source tree.

> 1.changed upm_table according to my sdram;
> 2.modify the register's value in the function initdram()

I have no way to tell if your modifications might work. But  this  is
not  (yet)  relevant.  You will see output on the console long before
RAM is accessed.

> :-),I am trying to learn,would u  recommend some helpful material about
> u-boot?

Read the sources :-)

> >Finally: did you order your BDI2000 yet? You will  probably  need  it
> >soon.
>
> I know that,and I will try to persuade my "boss".

Tell him that you are stuck and cannot continue without the right tool.

Best regards,

Wolfgang Denk

--
Software Engineering:  Embedded and Realtime Systems,  Embedded Linux
Phone: (+49)-8142-4596-87  Fax: (+49)-8142-4596-88  Email: wd@denx.de
All a hacker needs is a tight PUSHJ, a loose pair of UUOs, and a warm
place to shift.

** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: Why my console output nothing

[Bret Indrelee]

On Wed, 14 May 2003, Wolfgang Denk wrote:
> in message <20030514152700.68497.qmail@eyou.com> you wrote:
> > >Finally: did you order your BDI2000 yet? You will  probably  need  it
> > >soon.
> >
> > I know that,and I will try to persuade my "boss".
>
> Tell him that you are stuck and cannot continue without the right tool.

It sometimes helps to do the math.

Figure out how many days of engineering time you would have to save in order
to pay for the BDI. In my experience, you've usually got at least one hardware
and one software engineer working together to bring a board up, so it really
doesn't take long to pay for it.

If time to market means anything to your firm, or if others are depending on
your completion before they can start working (you are on critical path for
schedule), it becomes even easier to justify.

We bought a couple, just so that there wouldn't be a problem finding one
when you needed it for debug. The background debug is well worth the price.

-Bret

--
Bret Indrelee                 QLogic Corporation
Bret.Indrelee@qlogic.com      6321 Bury Drive, St 13, Eden Prairie, MN 55346


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

boot PPClinux on Sandpoint with Atlimus board

[dong in kang]

 Hello,

 I'm trying to boot PPClinux on Sandpoint with Altimux board (using MPC755
and MPC107 bridge chip).
I don't use DINK32. I use Codewarrior v.6.0 with BDI 2000.
I'm trying to load the zImage.initrd on the SDRAM and directly run the
kernel without using bootloader.
I use MontaVista's hardhat linux. I configured the linux to work for
Sandpoint.
I use BDI2000 to initialize the bridge chip, which are less than 10
instructions long.
After initialization, I can read/write sdram.
And then I upload zImage.initrd at 0x800000.
It successfully decompresses the kernel and jump to 0x00000000.
>From there it works fine until it enables MMU and jump to 0xC0000000.
When it jumps to 0xC0000000, it only finds garbage there. All FFs at
0xC0000000.
I checked the register MSR and BATs. They look OK. But address translation
doesn't seem to occur.
I think I need to do more initialization before I upload zImage.initrd and
run it.
Could anybody tell me what should i do?

 Thanks,

 Dong-In Kang


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/

Re: boot PPClinux on Sandpoint with Atlimus board

[Mark A. Greer]

dong in kang wrote:

> Hello,
>
> I'm trying to boot PPClinux on Sandpoint with Altimux board (using MPC755
>and MPC107 bridge chip).
>I don't use DINK32. I use Codewarrior v.6.0 with BDI 2000.
>I'm trying to load the zImage.initrd on the SDRAM and directly run the
>kernel without using bootloader.
>I use MontaVista's hardhat linux. I configured the linux to work for
>Sandpoint.
>I use BDI2000 to initialize the bridge chip, which are less than 10
>instructions long.
>After initialization, I can read/write sdram.
>And then I upload zImage.initrd at 0x800000.
>It successfully decompresses the kernel and jump to 0x00000000.
>>From there it works fine until it enables MMU and jump to 0xC0000000.
>When it jumps to 0xC0000000, it only finds garbage there. All FFs at
>0xC0000000.
>
Well, you're glossing over a lot of steps any of which could be causing
you problems.  However, to address why you don't see anything at 0xc0000000:

You're looking there with your BDI, right?  If so, that's because
0xc0000000 is a virtual address and the 'md' telnet cmd needs a physical
address.  Try 'md 0'.  If things look okay.  Go into your System.map
file, find 'log_buf' and dump the data at the associated address
(converting the 0xCyyyyyyy to 0x0yyyyyyy).  You should see the printk
output that has accumulated (but not been dumped b/c you haven't reached
console_init() yet).  If you're lucky, you'll see the panic msg/reg dump
that will give you a clue as to what went wrong.

>I checked the register MSR and BATs. They look OK. But address translation
>doesn't seem to occur.
>I think I need to do more initialization before I upload zImage.initrd and
>run it.
>Could anybody tell me what should i do?
>
> Thanks,
>
> Dong-In Kang
>
>
Mark


** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/