From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hollis Blanchard Subject: Re: [PATCH 0/2] kvm/e500v2: MMU optimization Date: Wed, 08 Sep 2010 09:07:07 -0700 Message-ID: <4C87B4AB.7010009@mentor.com> References: <1283938806-2981-1-git-send-email-yu.liu@freescale.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, agraf-l3A5Bk7waGM@public.gmane.org To: Liu Yu Return-path: In-Reply-To: <1283938806-2981-1-git-send-email-yu.liu-KZfg59tc24xl57MIdRCFDg@public.gmane.org> Sender: kvm-ppc-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: kvm.vger.kernel.org On 09/08/2010 02:40 AM, Liu Yu wrote: > The patchset aims at mapping guest TLB1 to host TLB0. > And it includes: > [PATCH 1/2] kvm/e500v2: Remove shadow tlb > [PATCH 2/2] kvm/e500v2: mapping guest TLB1 to host TLB0 > > The reason we need patch 1 is because patch 1 make things simple and flexible. > Only applying patch 1 aslo make kvm work. I've always thought the best long-term "optimization" on these cores is to share in the host PID allocation (i.e. __init_new_context()). This way, the TID in guest mappings would not overlap the TID in host mappings, and guest mappings could be demand-faulted rather than swapped wholesale. To do that, you would need to track the host PID in KVM data structures, I guess in the tlbe_ref structure. -- Hollis Blanchard Mentor Graphics, Embedded Systems Division From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hollis Blanchard Date: Wed, 08 Sep 2010 16:07:07 +0000 Subject: Re: [PATCH 0/2] kvm/e500v2: MMU optimization Message-Id: <4C87B4AB.7010009@mentor.com> List-Id: References: <1283938806-2981-1-git-send-email-yu.liu@freescale.com> In-Reply-To: <1283938806-2981-1-git-send-email-yu.liu-KZfg59tc24xl57MIdRCFDg@public.gmane.org> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Liu Yu Cc: kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, kvm-ppc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, agraf-l3A5Bk7waGM@public.gmane.org On 09/08/2010 02:40 AM, Liu Yu wrote: > The patchset aims at mapping guest TLB1 to host TLB0. > And it includes: > [PATCH 1/2] kvm/e500v2: Remove shadow tlb > [PATCH 2/2] kvm/e500v2: mapping guest TLB1 to host TLB0 > > The reason we need patch 1 is because patch 1 make things simple and flexible. > Only applying patch 1 aslo make kvm work. I've always thought the best long-term "optimization" on these cores is to share in the host PID allocation (i.e. __init_new_context()). This way, the TID in guest mappings would not overlap the TID in host mappings, and guest mappings could be demand-faulted rather than swapped wholesale. To do that, you would need to track the host PID in KVM data structures, I guess in the tlbe_ref structure. -- Hollis Blanchard Mentor Graphics, Embedded Systems Division