[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <173cc977-58fd-9431-6199-c2eb8e898f7c@redhat.com>
Date: Wed, 17 Oct 2018 19:17:32 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: Jim Mattson <jmattson@...gle.com>
Cc: Vitaly Kuznetsov <vkuznets@...hat.com>,
Radim Krčmář <rkrcmar@...hat.com>,
Roman Kagan <rkagan@...tuozzo.com>,
"K. Y. Srinivasan" <kys@...rosoft.com>,
Haiyang Zhang <haiyangz@...rosoft.com>,
Stephen Hemminger <sthemmin@...rosoft.com>,
"Michael Kelley (EOSG)" <Michael.H.Kelley@...rosoft.com>,
LKML <linux-kernel@...r.kernel.org>,
Liran Alon <liran.alon@...cle.com>,
kvm list <kvm@...r.kernel.org>
Subject: Re: [PATCH v6 06/13] KVM: nVMX: optimize prepare_vmcs02{,_full} for
Enlightened VMCS case
On 17/10/2018 19:08, Jim Mattson wrote:
> I believe that ESXi reads GUEST_CS_AR_BYTES on every VM-exit to
> determine code size.
Which makes me wonder, maybe we should add GUEST_SS_AR_BYTES which is
where the CPL lives. But then your tests from last year didn't find it.
Paolo
> On Wed, Oct 17, 2018 at 10:02 AM, Paolo Bonzini <pbonzini@...hat.com> wrote:
>> On 17/10/2018 16:47, Vitaly Kuznetsov wrote:
>>>>> + if (!hv_evmcs || !(hv_evmcs->hv_clean_fields &
>>>>> + HV_VMX_ENLIGHTENED_CLEAN_FIELD_GUEST_GRP2)) {
>>>>> + vmcs_write16(GUEST_CS_SELECTOR, vmcs12->guest_cs_selector);
>>>>> + vmcs_write32(GUEST_CS_LIMIT, vmcs12->guest_cs_limit);
>>>>> + vmcs_write32(GUEST_CS_AR_BYTES, vmcs12->guest_cs_ar_bytes);
>>>>> + vmcs_writel(GUEST_ES_BASE, vmcs12->guest_es_base);
>>>>> + vmcs_writel(GUEST_CS_BASE, vmcs12->guest_cs_base);
>>>>> + }
>>>> For what it's worth, I suspect that these can be moved to
>>>> prepare_vmcs02_full. The initial implementation of shadow VMCS did not
>>>> expose "unrestricted guest" to the L1 hypervisor, and emulation does a
>>>> lot of accesses to CS (of course). Not sure how ES base ended up in
>>>> there and not DS base, though...
>>> I tried unshadowing all these fields and at least Hyper-V on KVM
>>> (without using eVMCS of course) experiences a 1200-1300 cpu cycles
>>> regression during tight cpuid loop test. I checked and this happens
>>> because it likes vmreading GUEST_CS_AR_BYTES a lot.
>>
>> Go figure. :) Liran, do you happen to know if ESX does something
>> similar with CS descriptor cache fields?
>>
>> Paolo
Powered by blists - more mailing lists