[En-Nut-Discussion] newlib itoa() and RAM funny problem

[Ulrich Prinz]

Hi Uwe,

just built the toolchain again at home. Tried to run it against a kernel 
for a Cortex-A8 that sits here in a tablet.

I get some errors like
   CC      kernel/sys.o
/tmp/ccX7LrFA.s: Assembler messages:
/tmp/ccX7LrFA.s:283: Error: selected processor does not support 
requested special purpose register -- `mrs r3,cpsr'
/tmp/ccX7LrFA.s:290: Error: selected processor does not support 
requested special purpose register -- `msr cpsr_c,r3'
...
The original arm-eCross-eabi from Ole worked fine and produced pretty 
running kernels.

That looks like the compiler / assembler are not aware of Cortex series 
chips, or at least not of CortexA8 series.

I checked around and found that gcc optimizted for cortex are build with 
some extra flags and I miss these in your yagarto setup script.
So I'll try these and see if it adds some benefit.

Is there a naming convention for these?

best regards
Ulrich


Am 16.11.2012 11:06, schrieb Uwe Bonnes:
>>>>>> "Ulrich" == Ulrich Prinz <ulrich.prinz at googlemail.com> writes:
> ...
>
>      Ulrich> Now the joke of the story: The software crashes into Hardfault
>      Ulrich> of the CortexM3. I didn't figure out where and why it
>      Ulrich> crashes. It starts normally and even initializes the board. I
>      Ulrich> guess it crashes with the first printf().
>
>      Ulrich> I used several ARM toolchains on Linux all about gcc 4.5.x but
>      Ulrich> without any results.
>
> Yagarto uses   gcc-4.7.2.tar.bz2 now. Perhaps worth a try. If you have some
> test program, I could also try.
>
> Bye
>