Skip to contentSkip to conversation

AT91RM9200 boots from NOR Flash or not

U-Boot Jens Scharsig 4 participants 9 messages Nov 6, 2008 – Nov 8, 2008

lore
Share & tools
lore
#0 lore raw

Hello,

I am a little bit confused. We have designed a new AT91RM9200 based board. It should boot from 16-bit NOR Flash. I have read many relevant article in the forum, but I don't know , which board should I use as a base. I have look at the actual source (2008.10 and git), and I think, u-boot can't boot without a preloader as long CONFIG_AT91RM9200 is defined. Is there an open patch to resolve this?
Can you get an example for a working implementation?

Best regarts

Jens Scharsig

#1 lore raw
Quoted 10 lines · Jens Scharsig

Hello,

I am a little bit confused. We have designed a new AT91RM9200 based board. It should boot from 16-bit NOR Flash. I have read many relevant article in the forum, but I don't know , which board should I use as a base. I have look at the actual source (2008.10 and git), and I think, u-boot can't boot without a preloader as long CONFIG_AT91RM9200 is defined. Is there an open patch to resolve this?
Can you get an example for a working implementation?

Quoted 10 lines

Best regarts

Jens Scharsig

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
http://lists.denx.de/mailman/listinfo/u-boot

It's true that it won't boot when CONFIG_AT91RM9200 is defined. I dont know why it is like this since older versions of U-boot didn't have this problem (it changed around 1.3.0 if I remember correctly). It's almost like even if you do not define CONFIG_SKIP_LOWLEVEL_INIT and/or CONFIG_SKIP_RELOCATE_UBOOT, it still works like they were. Maybe it's because Atmel usually supplies their own Boot.bin along with the Development Kit (the configuration I would say you should use as example for your own) You can resolve this by editing two lines in the start.S file of the ARM920T cpu specific directory. In version 2008.10 its line 181 you can delete this, or change to ifdef, and again in line 264 you can either delete the if statement, or just make it branch to the lowlevel_init no matter what.

I know its what someone would call a hack, but it works :)

Mvh. / Best Regards

Martin Hejnfelt

#2 lore raw
Quoted 11 lines

for your own) You can resolve this by editing two lines in the start.S file of the ARM920T cpu specific directory. In version 2008.10 its line 181 you can delete this, or change to ifdef, and again in line 264 you can either delete the if statement, or just make it branch to the lowlevel_init no matter what.

I know its what someone would call a hack, but it works :)

Mvh. / Best Regards

Martin Hejnfelt

Thanx Martin,

the board speaks to me.

Regards

Jens Scharsig

#3 lore raw

Dear Jens Scharsig,

Quoted 2 lines · In message <[email protected]> you wrote:

the board speaks to me.

A patch, a patch - submit a patch, please.

Best regards,

Wolfgang Denk

Signature · 5 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Unser Kopf ist rund, damit das Denken die Richtung wechseln kann.
                                                   -- Francis Picabia
#4 lore raw
Quoted 11 lines

Dear Jens Scharsig,

Quoted 1 line · In message <[email protected]> you wrote:

the board speaks to me.

A patch, a patch - submit a patch, please.

Best regards,

Wolfgang Denk

Dear Wolfgang Denk,

I can do this in few weeks. But I have only our costum board. So I can't test the interaktion with other boards.

Best regarts Jens

#5 lore raw

Dear Jens,

Quoted 3 lines · in message <[email protected]> you wrote:

I can do this in few weeks. But I have only our costum board. So I can't test the interaktion with other boards.

I can test on the DK, and eventually on the cmc_pu2.

Best regards,

Wolfgang Denk

Signature · 6 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
This cultural mystique surrounding the  biological  function  --  you
realize humans are overly preoccupied with the subject.
	-- Kelinda the Kelvan, "By Any Other Name", stardate 4658.9
#6 lore raw
In reply to Martin Hejnfelt

Martin Hejnfelt sez,

Quoted 9 lines

Message from Jens Scharsig,

Quoted 8 lines

I am a little bit confused. We have designed a new AT91RM9200 based board. It should boot from 16-bit NOR Flash. I have read many relevant article in the forum, but I don't know , which board should I use as a base. I have look at the actual source (2008.10 and> git), and I think, u-boot can't boot without a preloader as long> CONFIG_AT91RM9200 is defined. Is there an open patch to resolve this? Can you get an example for a working implementation?

I built u-boot 1.3.4 for the AT91RM9200DK and was able to hack in support for the flash chip I'm using. You might try the same.

The setup I'm using is to store boot.bin from atmel in flash at . And a copy of u-boot.gz at 10010000.

The way I did was to enable the AT91RM9200's boot serial port boot loader and download Atmels loader.bin via XMODEM. Then download uboot.bin via XMODEM. Once that it downloaded you should get a prompt.

For auto booting I have Atmels boot program boot.bin stored in flash at 10000000-10005fff. And gzipped u-boot stored at 10010000-1001ffff.

Quoted 4 lines

It's true that it won't boot when CONFIG_AT91RM9200 is defined. I don't know why it is like this since older versions of U-boot didn't have this problem (it changed around 1.3.0 if I remember correctly).

Hmmm...

Quoted 11 lines

It's almost like even if you do not define CONFIG_SKIP_LOWLEVEL_INIT and/or CONFIG_SKIP_RELOCATE_UBOOT, it still works like they were. Maybe it's because Atmel usually supplies their own Boot.bin along with the Development Kit (the configuration I would say you should use as example for your own) You can resolve this by editing two lines in the start.S file of the ARM920T cpu specific directory. In version 2008.10 its line 181 you can delete this, or change to ifdef, and again in line 264 you can either delete the if statement, or just make it branch to the lowlevel_init no matter what.

Quoted 1 line

I know its what someone would call a hack, but it works :)

This is the second place I've seen this. I've been using atmels boot.bin loader to uncompress u-boot and it seems to work okay.

Matt Harper
Tehama-Wireless.

#7 lore raw

Dear Pink Boy,

Quoted 6 lines · In message <[email protected]> you wrote:

I built u-boot 1.3.4 for the AT91RM9200DK and was able to hack in support for the flash chip I'm using. You might try the same.

The setup I'm using is to store boot.bin from atmel in flash at . And a copy of u-boot.gz at 10010000.

This is not needed. U-Boot can boot from NOR directly, without need for boot.bin.

Best regards,

Wolfgang Denk

Signature · 5 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
The C-shell doesn't parse. It adhoculates.
    - Casper.Dik at Holland.Sun.COM in <[email protected]>
#8 lore raw
In reply to Jens Scharsig

Dear Jens Scharsig,

Quoted 8 lines · In message <[email protected]> you wrote:

I am a little bit confused. We have designed a new AT91RM9200 based board. It should boot from 16-bit NOR Flash. I have read many relevant article in the forum, but I don't know , which board should I use as a base. I have look at the actual source (2008.10 and git), and I think, u-boot can't boot without a preloader as long CONFIG_AT91RM9200 is defined. Is there an open patch to resolve this?
Can you get an example for a working implementation?

At least the at91rm9200dk and cmc_pu2 are supposed to boot from NOR flash - they did in earlier versions of U=-Boot, but probably have not undergone much testing lately. I have only the AT91RM9200DK in immediate access at the moment - this one boots 1.2.0-g14da5f76 just fine.

If recent versions should not work any more it would be nice if someone with suitable hardware could run a git bisect to locate the culprit. Thanks in advance.

Best regards,

Wolfgang Denk

Signature · 7 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
If a packet hits a pocket on a socket on a port,
And the bus is interrupted as a very last resort,
And the address of the memory makes your floppy disk abort,
Then the socket packet pocket has an error to report! - Ken Burchill?