[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <87wnda40am.fsf@redhat.com>
Date: Tue, 21 Jun 2022 15:24:17 +0200
From: Vitaly Kuznetsov <vkuznets@...hat.com>
To: Paolo Bonzini <pbonzini@...hat.com>, kvm@...r.kernel.org
Cc: Sean Christopherson <seanjc@...gle.com>,
Maxim Levitsky <mlevitsk@...hat.com>,
Wanpeng Li <wanpengli@...cent.com>,
Jim Mattson <jmattson@...gle.com>,
Michael Kelley <mikelley@...rosoft.com>,
Siddharth Chandrasekaran <sidcha@...zon.de>,
Yuan Yao <yuan.yao@...ux.intel.com>,
linux-hyperv@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 17/39] KVM: x86: hyper-v: L2 TLB flush
Paolo Bonzini <pbonzini@...hat.com> writes:
> On 6/13/22 15:39, Vitaly Kuznetsov wrote:
>> - tlb_flush_fifo = kvm_hv_get_tlb_flush_fifo(vcpu);
>> + tlb_flush_fifo = kvm_hv_get_tlb_flush_fifo(vcpu, is_guest_mode(vcpu));
>>
>
> Any reason to add this parameter?
Yes) Other users (all from kvm_hv_flush_tlb()) set this parameter based
on whether the calling vCPU (*not* the destination vCPU, which is the
first parameter) is in guest mode or not.
> It is always set to is_guest_mode(vcpu) and, even if it wasn't, I
> would add the parameter directly in patch 11.
I can move to Patch11 if necessary.
--
Vitaly
Powered by blists - more mailing lists