Re: [Xen-devel] [PATCH] x86 fixes for 3.3 impacting distros (v1).

[H. Peter Anvin]

On 06/29/2012 02:52 PM, Cyclonus J wrote:
> 
> Peter,
> 
> hmm, It looks like option 1 doesn't have any perf regression, but it is still
> not acceptable? That is fine. If you prefer to have a software PAT table lookup, could you provide
> some details so I can try to get something align that direction?
> 

It has no perf regression, but it really buries deep in the code a
strange abstraction which only happens to work on Xen and is confusing
as hell.

The idea with a software PAT table is that the PAT numbers used by the
kernel should come from a table in the kernel instead of being
hard-coded.  That might take some work, and it remains to be seen if it
is practical.

It *may* be that we need to hard-code 0 as WB, still, but that should be
true on any sane platform.

	-hpa
--
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/