> It seems to be always loaded by the common interrupt code (and %ebx is a
> call-saved register, so calling the interrupt handlers and returning
> doesn't clobber it).
True enough. I guess I should of done this from the beginning ...
> But testing it may be a good idea ;^p
Of course I will, but it is hard (and subject to Heisenberg principals)
to test that we preempt whenever necessary. Only with the patch I sent
you earlier am I seeing no missed preemptions ... now to work on the
egregiously long-held locks.
Robert Love
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/