[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <fdc94c1f-fb39-edc2-b7eb-cea87a69023d@redhat.com>
Date: Thu, 12 Apr 2018 22:23:05 +0200
From: Paolo Bonzini <pbonzini@...hat.com>
To: Sean Christopherson <sean.j.christopherson@...el.com>
Cc: KarimAllah Ahmed <karahmed@...zon.de>, x86@...nel.org,
linux-kernel@...r.kernel.org, kvm@...r.kernel.org, hpa@...or.com,
jmattson@...gle.com, mingo@...hat.com, rkrcmar@...hat.com,
tglx@...utronix.de
Subject: Re: [PATCH 06/10] KVM/nVMX: Use kvm_vcpu_map when mapping the virtual
APIC page
On 12/04/2018 19:57, Sean Christopherson wrote:
> On Thu, Apr 12, 2018 at 04:38:39PM +0200, Paolo Bonzini wrote:
>> On 21/02/2018 18:47, KarimAllah Ahmed wrote:
>>> +
>>> + if (kvm_vcpu_map(vcpu, gpa_to_gfn(vmcs12->virtual_apic_page_addr), map))
>>> + vmcs_write64(VIRTUAL_APIC_PAGE_ADDR, gfn_to_gpa(map->pfn));
>>
>> This should in principle be pfn_to_hpa. However, pfn_to_hpa is unused;
>> let's just remove it and do "<< PAGE_SHIFT". Unlike gfn_to_gpa, where
>> in principle the shift could be different, pfn_to_hpa is *by definition*
>> a shift left by PAGE_SHIFT.
>
> Any reason not to use PFN_PHYS instead of the handcoded shift?
That works too of course. It's just not used much (or at all?) in KVM.
Paolo
Powered by blists - more mailing lists