Re: [rfc][patch] store-free path walking

[Andi Kleen]

> A few years ago when K7s were not ancient yet, after hearing
> argument for and against non-temporal stores,
> I decided to finally figure it for myself.
> 
> I tested kernel build workload on two kernels with the only
> one difference - clear_page with and without non-temporal stores.
> 
> "Non-temporal stores" kernel was faster, not slower. Just a little bit,
> but reproducibly.

I did the same experiments and in my case (K8) they were slower in 
real-world benchmarks for clear_page (but faster in micro benchmarks)

But I didn't propose non temporal anyways.

-Andi

-- 
ak@xxxxxxxxxxxxxxx -- Speaking for myself only.
--
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/