On 03/07/2014 02:06 PM, Dave Jones wrote:
> On Fri, Mar 07, 2014 at 10:38:56PM +0100, Borislav Petkov wrote:
>
> > * Another option would be if we change the f/m/s of "qemu64" so that
> > the test doesn't fire.
> >
> > What also makes me wonder is why is this thing even called qemu64?? AMD
> > family 6 was 32-bit only CPUs so 64 is kinda wrong IMO. I mean, the
> > kernel code is in "ifdef CONFIG_X86_32" block so "qemu64" is patently
> > wrong naming.
>
> Additionally, fam:6 model:6 stepping 3 never existed in the real world afaict.
> I used to keep x86info's stepping db pretty up to date, and that only has
> knowledge
> of stepping 1 & 2.[*] Modelling qemu on something from the real world might
> be
> a better idea than inventing new special cases.
>
> > Oh, and the thing has CPUID_EXT2_LM which is also a WTH moment for me.
> > Paolo, what's going on here?
>
> Yeah, this is a mess, there should be no family < 0xf with LM set.
>
> Dave
For AMD, at least. Intel obviously have a bunch of chips with family ==
6 and LM.
Perhaps Fengguang should switch from "qemu64,+smep,+smap" to something
like "Haswell,+smep,+smap" instead for the time being at least?
-hpa
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
