[U-Boot] OMAP5912 -- Status? -- Identified as ARM926EJS start.S problem

[U-Boot] OMAP5912 -- Status? -- Identified as ARM926EJS start.S problem

[James Harris]

Hi Igor,

Thanks for your prompt response.
I've made similar modifications to the omap5912osk over the past number of days with no success...
The problem appears deeper than just providing these definitions, which as you say, requires actual hardware to test with :)

I've gone back to 2010.09 release and it now works very nicely!

I chose that version as it is just prior to the relocation changes applied in the arch/arm/cpu/arm926ejs/start.S file:
2010-09-19 Heiko Schocher ARM: implement relocation for ARM926

That change would certainly cause out-of-box compilation failures, however I am not sure whether it was that change or possibly the following change:
2010-10-13 Albert Aribaud arm: implement ELF relocations
that causes the runtime failure...

I may investigate further, however, at this stage I can "move on" using the 2010.09 release.

As an aside: Using U-Boot 2010.09 also resolves a problem I was having with u-boot V1.1.1:
The V2.6.32 linux kernel could not be booted when mem=32M was passed using u-boot V1.1.1
However when I tried earlier today, through pure desperation, passing mem=16M it did boot.
Now under u-boot 2010.09 the Linux kernel boots just fine with mem=32M

Cheers,

James


.......................................................................
?
James Harris

T? +61 8 8104 5253
E?? jharris at ellex.com
.......................................................................

-----Original Message-----
From: Igor Grinberg [mailto:grinberg at compulab.co.il] 
Sent: Thursday, 19 May 2011 5:09 PM
To: James Harris
Cc: u-boot at lists.denx.de; rishi at ti.com
Subject: Re: [U-Boot] OMAP5912 -- Status?

Hi James,

On 05/19/11 10:07, James Harris wrote:

> Greetings all,
>
>  
>
> Over the past few days I've been trying to get u-boot 2011.03 for the 
> omap5912osk working.
>
>  
>
> Support for the target appears to have stagnated...
>
>  
>
> "Out of the box" the omap5912osk (and omap1610inn) target fails to 
> compile due to the following missing symbols:
>
> CONFIG_SYS_SDRAM_BASE
>
> CONFIG_SYS_INIT_SP_ADDR

I've send a patch [1] which hopefully fixes omap1610inn and omap1610h2.
I can't test is as I don't have this hardware.
It is currently waiting for Sandeep's approval.

[1]
http://www.mail-archive.com/u-boot at lists.denx.de/msg52157.html



--
Regards,
Igor.

[U-Boot] OMAP5912 -- Status? -- Identified as ARM926EJS start.S problem

[Igor Grinberg]

On 05/19/11 11:12, James Harris wrote:

> Hi Igor,
>
> Thanks for your prompt response.
> I've made similar modifications to the omap5912osk over the past number of days with no success...

:(

> The problem appears deeper than just providing these definitions, which as you say, requires actual hardware to test with :)

Exactly, I had to make it compile so I can compile test some other patch.
So I took some values from another board with the same SoC that looked sane to me.
And put there a "FIXME" comment...

> I've gone back to 2010.09 release and it now works very nicely!
>
> I chose that version as it is just prior to the relocation changes applied in the arch/arm/cpu/arm926ejs/start.S file:
> 2010-09-19 Heiko Schocher ARM: implement relocation for ARM926
>
> That change would certainly cause out-of-box compilation failures, however I am not sure whether it was that change or possibly the following change:
> 2010-10-13 Albert Aribaud arm: implement ELF relocations
> that causes the runtime failure...

Added Heiko and Albert to cc.

> I may investigate further, however, at this stage I can "move on" using the 2010.09 release.
>
> As an aside: Using U-Boot 2010.09 also resolves a problem I was having with u-boot V1.1.1:
> The V2.6.32 linux kernel could not be booted when mem=32M was passed using u-boot V1.1.1
> However when I tried earlier today, through pure desperation, passing mem=16M it did boot.
> Now under u-boot 2010.09 the Linux kernel boots just fine with mem=32M
>
> Cheers,
>
> James
>
>
> .......................................................................
>  
> James Harris
>
> T  +61 8 8104 5253
> E   jharris at ellex.com
> .......................................................................
>
> -----Original Message-----
> From: Igor Grinberg [mailto:grinberg at compulab.co.il] 
> Sent: Thursday, 19 May 2011 5:09 PM
> To: James Harris
> Cc: u-boot at lists.denx.de; rishi at ti.com
> Subject: Re: [U-Boot] OMAP5912 -- Status?
>
> Hi James,
>
> On 05/19/11 10:07, James Harris wrote:
>
>> Greetings all,
>>
>>  
>>
>> Over the past few days I've been trying to get u-boot 2011.03 for the 
>> omap5912osk working.
>>
>>  
>>
>> Support for the target appears to have stagnated...
>>
>>  
>>
>> "Out of the box" the omap5912osk (and omap1610inn) target fails to 
>> compile due to the following missing symbols:
>>
>> CONFIG_SYS_SDRAM_BASE
>>
>> CONFIG_SYS_INIT_SP_ADDR
> I've send a patch [1] which hopefully fixes omap1610inn and omap1610h2.
> I can't test is as I don't have this hardware.
> It is currently waiting for Sandeep's approval.
>
> [1]
> http://www.mail-archive.com/u-boot at lists.denx.de/msg52157.html
>
>
>
> --
> Regards,
> Igor.
>
>

-- 
Regards,
Igor.

[U-Boot] OMAP5912 -- Status? -- Identified as ARM926EJS start.S problem

[Wolfgang Denk]

Dear "James Harris",

In message <66F0ABBCB3D267479CB73B0F4C78908F02739008@ellex-adc1> you wrote:
> 
> I've made similar modifications to the omap5912osk over the past number of
> days with no success...
> The problem appears deeper than just providing these definitions, which as
> you say, requires actual hardware to test with :)

We still have an omap5912osk in our lab. If you can wait a bit for
the results, I can run a test or another.


Best regards,

Wolfgang Denk

-- 
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
Landru! Guide us!
	-- A Beta 3-oid, "The Return of the Archons", stardate 3157.4