[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20111216021254.GG14839@one.firstfloor.org>
Date: Fri, 16 Dec 2011 03:12:54 +0100
From: Andi Kleen <andi@...stfloor.org>
To: "H. Peter Anvin" <hpa@...or.com>
Cc: Andi Kleen <andi@...stfloor.org>, Borislav Petkov <bp@...64.org>,
linux-kernel@...r.kernel.org, kay.sievers@...y.org, trenn@...e.de,
Andi Kleen <ak@...ux.intel.com>, davej@...hat.com
Subject: Re: [PATCH 07/10] cpufreq: Add support for x86 cpuinfo auto loading
On Thu, Dec 15, 2011 at 05:19:46PM -0800, H. Peter Anvin wrote:
> On 12/08/2011 06:37 AM, Andi Kleen wrote:
> >> Well, there is CPUID_8000_0007_EDX[7] which denotes the presence
> >> of the P-state registers. It probably could be added to
> >> init_scattered_cpuid_features() but there's still special handling
> >> needed for K8.
> >
> > Dave is probably right that it's not worth moving the list.
> >
> >> Can you do both a family and a feature bit test with the autoprobing
> >> code?
> >
>
> Another similar question: is it possible to key a load on more than one
> CPU feature? It's fairly common that you need more than one feature for
> a specific piece of code to function, and it would be good to be able to
> represent it.
Right now it's not possible. The driver just errors out if it can't
find some additional capability.
I considered it originally, but all the cases I looked at were quite
obscure (unlikely to happen) so I opted for this simpler setup.
-Andi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists