driver SPI for MPC5200

driver SPI for MPC5200

[iida]

i am tring to find a driver SPI for MPC5200.
In  linuxppc-2.4.25-denx
there is cpm_spi.c in arch/ppc/8260_io
It can be used in mpc5200 ?
Or what i have to modify to use in mpc5200
Thanks

--
Atenciosamente,
Renato Iida
Z Tecnologia em Comunicação
www.ztec.com.br


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

Re: driver SPI for MPC5200

[Wolfgang Denk]

In message <40BC902D.7080804@ztec.com.br> you wrote:
>
> i am tring to find a driver SPI for MPC5200.
> In  linuxppc-2.4.25-denx
> there is cpm_spi.c in arch/ppc/8260_io
> It can be used in mpc5200 ?

No, it cannot. It's a driver for the MPC8260's CPM; the MPC5200  does
not have such a thing.

> Or what i have to modify to use in mpc5200

You need to write a new driver.

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
"It ain't so much the things we don't know that get  us  in  trouble.
It's  the  things  we know that ain't so." - Artemus Ward aka Charles
Farrar Brown

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

no core dump from the kernel on SIGSEGV

[David Ho]

Hi all,

I'm looking at making the kernel core dump on a process that seg vaulted.
What I did was recompiled the kernel specifying a non-zero core file size
rlim[RLIMIT_CORE] = { RLIM_INFINITY, RLIM_INFINITY }.  I was still not
getting a core dump.

The following test program was used to signal a seg vault (SIGSEGV).

#include <stdio.h>

int main(int argc, char** argv) {

        unsigned int* pointer = (unsigned int*) 0x12345678;

        /* generate a SIGSEGV - invalid memory reference */
        *pointer = 0xdeadbeef;

        return 0;
}

Anyone can shred some light on this?

Thanks a lot,
David


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

Re: no core dump from the kernel on SIGSEGV

[Eugene Surovegin]

On Tue, Jun 01, 2004 at 04:22:34PM -0400, David Ho wrote:
>
> Hi all,
>
> I'm looking at making the kernel core dump on a process that seg vaulted.
> What I did was recompiled the kernel specifying a non-zero core file size
> rlim[RLIMIT_CORE] = { RLIM_INFINITY, RLIM_INFINITY }.  I was still not
> getting a core dump.
>

DO NOT TOUCH the kernel!

Please, read about ulimit shell command.

Eugene


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

Re: no core dump from the kernel on SIGSEGV

[David Ho]

> >
> > I'm looking at making the kernel core dump on a process that seg
vaulted.
> > What I did was recompiled the kernel specifying a non-zero core file
size
> > rlim[RLIMIT_CORE] = { RLIM_INFINITY, RLIM_INFINITY }.  I was still not
> > getting a core dump.
> >
>
> DO NOT TOUCH the kernel!
>
> Please, read about ulimit shell command.

Okay, you point taken.  I assume ulimit will have the same effect of
assigning it a non-zero value.
What is troubling me is why it doesn't go to the core_dump function in
fs/binfmt_elf.c

David


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

Re: driver SPI for MPC5200

[Kate Alhola]

Wolfgang Denk wrote:

>>Or what i have to modify to use in mpc5200
>>
>>
>
>You need to write a new driver.
>
>

I have been doin such thing a while. The base idea is to hae similar
subsystem tha i2c has.
Least in my application i like to controll multiple diferent devices via
spi  and i like
that the spi connected targets look a like "normal" devides to linux
programs.
At the moment i am working spi core and drivers for tsc2301 ( system
management
chip / adc/audio ), MMC ( multimedi card ) and frame buffer device for
small 128x64
graphical LCD ( UC1606 Chip )

The thing that i am waiting is to get something clear to this DMA
problem that
has been discussed here lately.

The MPC5200 SPI device does not support Bestcomm DMA at all but PSC can
be used
in SPI mode with DMA. With DMA i can update display or read MMC block
with just one interupt without polling etc needed with normal SPI.


Kate

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