sparc erc32 FPU confusion

Jiri Gaisler jiri at gaisler.com
Wed May 31 10:52:02 UTC 2006



Aleix Conchillo Flaqué wrote:
> On 31 May 2006, at 10:51, Jiri Gaisler wrote:
> 
>> Tasks that are not marked as FPU tasks could therefore trap
>> unless they are compiled with lower optimization effort,
>> OR with -msoft-float.
> 
> ....
> 
>> an FPU is present and enabled. The kernel can (and must)
>> therefore always be compiled with -msoft-float to avoid
>> spurious FPU instructions due to gcc optimizations. If
>> your application uses the FPU, these parts should be
>> compiled without -msoft-float. If you share code between
>> FPU tasks and non-FPU tasks, you must decrease the
>> optimization level to avoid unexpected FPU instructions
>> as explained above.
>>
> 
> So, this means that the kernel can still be compiled with -O4 (like  
> before) as it is forcing to not use the FPU with the -msoft-float?

Yes, this is exactly why the -msoft-float is used for the kernel.

> 
> Do you know from which version of gcc this is the behaviour?

I have seen it in gcc-3.4.4, so probably it was introduced
in the gcc-3.4.x series. I have not tested gcc-3.3.x .

Jiri.




More information about the users mailing list