Re: [PATCH] x86 idle: repair large-server 50-watt idle-powerregression

[Ingo Molnar]

* H. Peter Anvin <hpa@xxxxxxxxx> wrote:

> CLFLUSH is only (guaranteed to be) ordered with respect to MFENCE 
> according to the SDM.

Yes, I quoted the relevant text a couple of hours ago, CLFLUSH is 
pretty special, so it's not ordered against atomics or serializing 
instructions for example - only ordered against explict MFENCE calls, 
which on x86 is mb()/smp_mb().

Thanks,

	Ingo
--
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/