[Cbe-oss-dev] what's the expected behaviour for nosched contexts?

[Arnd Bergmann]

On Friday 01 June 2007, Jeremy Kerr wrote:
> > What about voluntary context  switches, e.g. for spu_acquire_saved? 
> > Are they allowed?
> 
> I would say no. The files under contexts that cause a spu_acquire_saved 
> are not present in the nosched_contents (although I see that a 
> signal{1,2} read will cause this, patch coming...).

I agree. One thing that users expect from using the _NOSCHED flag is
that you get somewhat predictable latencies for any operation, so I'd
like the definition of _NOSCHED to remain 'never do a context save
on this'.
This also means that we don't need to treat isolated context different
from non-isolated nosched contexts, except for the bit where isolated
more is entered or left.

	Arnd <><