Re: x86: A fast way to check capabilities of the current cpu

[Tejun Heo]

On Fri, Jan 21, 2011 at 11:21:02AM -0600, Christoph Lameter wrote:
> > I don't think percpu_read_stable() can be used here.  It's not
> > guaranteed to be stable across different cpus.
> 
> Why would that matter? The caller has to disabled preemption anyways since
> otherwise the processor may change which means that the result of the
> operation is useless.

Because

	preempt_disable();
	this_cpu_has();
	preempt_enable();
	preempt_disable();
	this_cpu_has();
	preempt_enable();

might malfunction.  percpu_read_stable() is pretty much applicable
only to stuff local to the thread.

> > Also, can we just implement what's necessary on top of this_cpu_has()?
> > this_cpu_has() already has constant handling, so there's no need to
> > add this_cpu_test_bit() at this point.
> 
> Not sure what you mean. this_cpu_test_bit is necessary because
> test_cpu_cap expects a regular pointer and performs a regular load.
> this_cpu_constant_test_bit handles the segment prefix necessary for a per
> cpu load.
> 
> The constant refers to the bit.

Oh, you're right.  Sorry about that.  Can you please then add a
comment noting that the operation is x86 only?  Maybe prefix it with
x86_?

Thanks.

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/