| From: |
| Jon Masters <jcm-AT-jonmasters.org> |
| To: |
| Henrique de Moraes Holschuh <hmh-AT-hmh.eng.br>, Andi Kleen <andi-AT-firstfloor.org> |
| Subject: |
| Re: Improve retpoline for Skylake |
| Date: |
| Mon, 15 Jan 2018 03:26:34 -0500 |
| Message-ID: |
| <985f979e-0740-5d8a-d6b8-b023105aa021@jonmasters.org> |
| Cc: |
| David Woodhouse <dwmw2-AT-infradead.org>, tglx-AT-linutronix.de, x86-AT-kernel.org, linux-kernel-AT-vger.kernel.org, pjt-AT-google.com, torvalds-AT-linux-foundation.org, gregkh-AT-linux-foundation.org, peterz-AT-infradead.org, luto-AT-amacapital.net, thomas.lendacky-AT-amd.com, arjan.van.de.ven-AT-intel.com |
On 01/12/2018 05:03 PM, Henrique de Moraes Holschuh wrote:
> On Fri, 12 Jan 2018, Andi Kleen wrote:
>>> Skylake still loses if it takes an SMI, right?
>>
>> SMMs are usually rare, especially on servers, and are usually
>> not very predictible, and even if you have
>
> FWIW, a data point: SMIs can be generated on demand by userspace on
> thinkpad laptops, but they will be triggered from within a kernel
> context. I very much doubt this is a rare pattern...
Sure. Just touch some "legacy" hardware that the vendor emulates in a
nasty SMI handler. It's definitely not acceptable to assume that SMIs
can't be generated under the control of some malicious user code.
Our numbers on Skylake weren't bad, and there seem to be all kinds of
corner cases, so again, it seems as if IBRS is the safest choice.
Jon.
