[U-Boot] SPI on PPC MPC85xx

[U-Boot] SPI on PPC MPC85xx

[Matthias]

Hi all,

I've been trying to setup SPI on our MPC85xx based board.

Unfortunately, compilation of mpc8xxx_spi.c fails due to a missing immap_t datatype.

I see, that this datatype is defined in the headers immap_83xx.h and immap_86xx.h but it is not in immap_85xx.h.

I am using u-boot-2009.11.1 but I also looked into the latest git sources, and there's no immap_t in that specific file neither.

Is there a reason why it is not implemented for MPC85xx? Or do I just need some more defines?

So far I only have CONFIG_MPC8XXX_SPI and undef CONFIG_SOFT_SPI.

Cheers
Matthias Dunda

[U-Boot] SPI on PPC MPC85xx

[Matthias Dunda]

Hi again,

unfortunately I have to push this message, as I received no comment on it, yet. 

Is there a reason why the immap_t is not implemented for mpc85xx?

I tried to copy it from mpc8360 but there are so many dependent mpc83xx datatypes in that struct that I postponed the work on that, because I think it'll lead to no success...

Even trying to use the soft SPI fails, because first thing the code does is using that immap_t datatype... :-(

Does anyone have a comment on this?

Cheers
Matthias

Am 25.10.2010 um 13:05 schrieb Matthias:

> Hi all,
> 
> I've been trying to setup SPI on our MPC85xx based board.
> 
> Unfortunately, compilation of mpc8xxx_spi.c fails due to a missing immap_t datatype.
> 
> I see, that this datatype is defined in the headers immap_83xx.h and immap_86xx.h but it is not in immap_85xx.h.
> 
> I am using u-boot-2009.11.1 but I also looked into the latest git sources, and there's no immap_t in that specific file neither.
> 
> Is there a reason why it is not implemented for MPC85xx? Or do I just need some more defines?
> 
> So far I only have CONFIG_MPC8XXX_SPI and undef CONFIG_SOFT_SPI.
> 
> Cheers
> Matthias Dunda
> 
> _______________________________________________
> U-Boot mailing list
> U-Boot at lists.denx.de
> http://lists.denx.de/mailman/listinfo/u-boot

[U-Boot] SPI on PPC MPC85xx

[Wolfgang Denk]

Dear Matthias Dunda,

In message <6A334922-E6EE-4DC3-AE44-F431570066C9@t1817.greatnet.de> you wrote:
> 
> unfortunately I have to push this message, as I received no comment on it, yet. 

See http://catb.org/esr/faqs/smart-questions.html#id586288

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
Every living thing wants to survive.
	-- Spock, "The Ultimate Computer", stardate 4731.3

[U-Boot] SPI on PPC MPC85xx

[Stefan Roese]

Hi Matthias,

On Wednesday 27 October 2010 08:55:42 Matthias Dunda wrote:
> unfortunately I have to push this message, as I received no comment on it,
> yet.
> 
> Is there a reason why the immap_t is not implemented for mpc85xx?
> 
> I tried to copy it from mpc8360 but there are so many dependent mpc83xx
> datatypes in that struct that I postponed the work on that, because I
> think it'll lead to no success...
> 
> Even trying to use the soft SPI fails, because first thing the code does is
> using that immap_t datatype... :-(
> 
> Does anyone have a comment on this?

I took a quick glance at this. There are different versions of SPI controllers 
in the MPC8xxx SoC's. As it seems mpc8xxx_spi.c does "only" support the "CPU-
mode" (meaning non-CPM/QE-mode) Freescale SPI controllers. This controller is 
implemented for example in the MPC8349. For your SoC (MPC8568) you need to 
implement the QE-mode which is currently not supported.
 
Cheers,
Stefan

--
DENX Software Engineering GmbH,      MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich,  Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-0 Fax: (+49)-8142-66989-80 Email: office at denx.de

[U-Boot] SPI on PPC MPC85xx

[Scott Wood]

On Wed, 27 Oct 2010 08:55:42 +0200
Matthias Dunda <user254@t1817.greatnet.de> wrote:

> Hi again,
> 
> unfortunately I have to push this message, as I received no comment on it, yet. 
> 
> Is there a reason why the immap_t is not implemented for mpc85xx?

immap_t gets to be a mess of ifdefs when you're trying to support a
bunch of different chips with slight differences, not to mention having
to manage all the reserved regions with obscure size values to keep
everything in the right spot (versus a simple offset of the start of a
register block that can be easily compared to the manual).

-Scott

[U-Boot] SPI on PPC MPC85xx

[Jens Gehrlein]

Hi Matthias,

Stefan Roese schrieb:
> Hi Matthias,
> 
> On Wednesday 27 October 2010 08:55:42 Matthias Dunda wrote:
>> unfortunately I have to push this message, as I received no comment on it,
>> yet.
>>
>> Is there a reason why the immap_t is not implemented for mpc85xx?
>>
>> I tried to copy it from mpc8360 but there are so many dependent mpc83xx
>> datatypes in that struct that I postponed the work on that, because I
>> think it'll lead to no success...
>>
>> Even trying to use the soft SPI fails, because first thing the code does is
>> using that immap_t datatype... :-(
>>
>> Does anyone have a comment on this?
> 
> I took a quick glance at this. There are different versions of SPI controllers 
> in the MPC8xxx SoC's. As it seems mpc8xxx_spi.c does "only" support the "CPU-
> mode" (meaning non-CPM/QE-mode) Freescale SPI controllers. This controller is 
> implemented for example in the MPC8349. For your SoC (MPC8568) you need to 
> implement the QE-mode which is currently not supported.

I didn't compare the MPC85xx QE SPI controller with the MPC8xx CPM SPI
controller, but maybe a look at arch/powerpc/cpu/mpc8xx/spi.c could
help you as a base for porting/creating a driver?

-- 
Kind regards,
Jens