From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paolo Bonzini Subject: Re: [PATCH RFC] kvm: x86: add halt_poll module parameter Date: Thu, 05 Feb 2015 21:14:36 +0100 Message-ID: <54D3CF2C.6030203@redhat.com> References: <1423152325-5094-1-git-send-email-pbonzini@redhat.com> <54D3BC94.9010007@siemens.com> <54D3C296.7030207@redhat.com> <54D3C314.3070903@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Cc: rkrcmar@redhat.com, mtosatti@redhat.com To: Rik van Riel , Jan Kiszka , linux-kernel@vger.kernel.org, kvm@vger.kernel.org Return-path: In-Reply-To: <54D3C314.3070903@redhat.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On 05/02/2015 20:23, Rik van Riel wrote: >> > 3) long term anyway we want it to auto tune, which is better than tuning >> > it per-VM. > We may want to auto tune it per VM. We may even want to auto tune it per VCPU. > However, if we make auto tuning work well, I do not > think we want to expose a user visible tunable per > VM, and commit to keeping that kind of interface > around forever. Exactly. We probably want module parameters to tune the minimum/maximum values (which includes the special cases of disabling polling altogether, and disabling the autotuning while leaving polling enabled), but committing to a per-VM interface is premature. Paolo