building RTEMS 4.11 leon3 with floating point support?

[Cudmore, Alan P. (GSFC-5820)]

I have not analyzed it to determine what floating point code is being
generated. 

As soon as we try to FTP to the board, it generates an FP trap on GRMON.
When I enabled the floating point context support on the FTP task, it does
not crash. 

But I agree, it is worth trying to figure out where the FP code is being
generated in the FTP task.

Alan


On 9/2/15, 10:14 AM, "Joel Sherrill" <joel.sherrill at oarcorp.com> wrote:

>
>
>On 8/31/2015 5:47 AM, Cudmore, Alan P. (GSFC-5820) wrote:
>> Having a floating point configuration of the BSP makes sense. I was able
>> to rebuild the BSP with the hardware floating point compiler option and
>>it
>> works.
>>
>> I did get a floating point exception in the FTP task and had to change
>>the
>> FTP task create to enable floating point. Is it worth submitting a patch
>> to change that?
>
>What did it use the FPU for? Is this a case where GCC emits FPU
>instructions
>just to move data or calculate array indices?
>
>On some CPUs, when hard FP is turned on, GCC assumes the FPU can be used
>for some non-intuitive things and it is safer to implicitly make all tasks
>FP rather than letting people trip over implicit uses of the FPU.
>
>--joel
>
>> Thanks,
>> Alan
>>
>>
>> On 8/31/15, 2:40 AM, "Sebastian Huber"
>> <sebastian.huber at embedded-brains.de> wrote:
>>
>>> Hello Alan,
>>>
>>> I am not sure how Gaisler manages this in their RCC, but I think for
>>> standard RTEMS we need additional BSPs (e.g. via leon3_fp.cfg and
>>> ngmp_fp.cfg configuration files).
>>>
>>> On 17/08/15 17:29, Cudmore, Alan P. (GSFC-5820) wrote:
>>>> Hi,
>>>> We are currently trying to use the leon3 BSP for RTEMS 4.11.
>>>> We have used the 4.10 RCC/Gaisler tools with the Driver Manager in the
>>>> past, but we would like to be on 4.11 for the SMP support.
>>>>
>>>> When I build the 4.11 leon3 BSP, I seem to only have soft-float
>>>> support. When I try to build with the ‹enable-multilib configure
>>>> switch, the build fails. Before I start trying to determine the cause
>>>> of the compile failure, I wanted to make sure I am enabling floating
>>>> point support for the LEON3 correctly.
>>>>
>>>> Thanks,
>>>> Alan
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> devel mailing list
>>>> devel at rtems.org
>>>> http://lists.rtems.org/mailman/listinfo/devel
>>>
>>> --
>>> Sebastian Huber, embedded brains GmbH
>>>
>>> Address : Dornierstr. 4, D-82178 Puchheim, Germany
>>> Phone   : +49 89 189 47 41-16
>>> Fax     : +49 89 189 47 41-09
>>> E-Mail  : sebastian.huber at embedded-brains.de
>>> PGP     : Public key available on request.
>>>
>>> Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
>>>
>>
>> _______________________________________________
>> devel mailing list
>> devel at rtems.org
>> http://lists.rtems.org/mailman/listinfo/devel
>>
>
>-- 
>Joel Sherrill, Ph.D.             Director of Research & Development
>joel.sherrill at OARcorp.com        On-Line Applications Research
>Ask me about RTEMS: a free RTOS  Huntsville AL 35805
>Support Available                (256) 722-9985