linux-kernel - Re: [PATCH] KVM: VMX: Check if bus lock vmexit was preempted

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <0b8714ed-c3ae-01ee-d2fd-fe7542d01596@redhat.com>
Date:   Thu, 23 Sep 2021 08:37:21 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     Hao Xiang <hao.xiang@...ux.alibaba.com>,
        Xiaoyao Li <xiaoyao.li@...el.com>, kvm@...r.kernel.org
Cc:     linux-kernel@...r.kernel.org, chenyi.qiang@...el.com,
        shannon.zhao@...ux.alibaba.com,
        Sean Christopherson <seanjc@...gle.com>
Subject: Re: [PATCH] KVM: VMX: Check if bus lock vmexit was preempted

On 23/09/21 02:59, Hao Xiang wrote:
> EXIT_REASON.bus_lock_detected may or may not be set when 
> exit_reason=EXIT_REASON_BUS_LOCK,  It clould depend on hardware 
> implementaion. No matter when intel states it clearly, I think it is 
> better that we avoid repeated assignment by adding additional check 
> condition in vmx_handle_exit.  Of course , it is also ok that 
> hand_bus_lock_vmexit do nothing , but the code is not clear, and the 
> code logic will be inconsistent with spec description.

For 5.16 we'll go with something like Sean's sketch, that sets the 
bus_lock_detected bit on EXIT_REASON_BUS_LOCK.

Paolo