Re: [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM

Subject: Re: [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM
From: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Date: Wed, 7 Jul 2021 15:58:41 +0200
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>, Sean Christopherson <seanjc@xxxxxxxxxx>, Wanpeng Li <wanpengli@xxxxxxxxxxx>, Vitaly Kuznetsov <vkuznets@xxxxxxxxxx>, Joerg Roedel <joro@xxxxxxxxxx>, Borislav Petkov <bp@xxxxxxxxx>, "H. Peter Anvin" <hpa@xxxxxxxxx>, Ingo Molnar <mingo@xxxxxxxxxx>, "open list:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <linux-kernel@xxxxxxxxxxxxxxx>, "maintainer:X86 ARCHITECTURE (32-BIT AND 64-BIT)" <x86@xxxxxxxxxx>, Jim Mattson <jmattson@xxxxxxxxxx>
In-reply-to: <9413056ebbd5997a35b446f2841589973484ba02.camel@redhat.com>
References: <20210623113002.111448-1-mlevitsk@redhat.com> <20210623113002.111448-3-mlevitsk@redhat.com> <6c4a69ce-595e-d5a1-7b4e-e6ce1afe1252@redhat.com> <43ef1a1ea488977db11d40ec9672b524ec816112.camel@redhat.com> <9413056ebbd5997a35b446f2841589973484ba02.camel@redhat.com>
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.11.0

On 07/07/21 14:57, Maxim Levitsky wrote:


Hi!
Any update? should I use a lock for this?


Yes please, even irq_lock can do.

Paolo

References:
- [PATCH 00/10] My AVIC patch queue
  - From: Maxim Levitsky
- [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM
  - From: Maxim Levitsky
- Re: [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM
  - From: Paolo Bonzini
- Re: [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM
  - From: Maxim Levitsky
- Re: [PATCH 02/10] KVM: x86: APICv: fix race in kvm_request_apicv_update on SVM
  - From: Maxim Levitsky