Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <0cfb1f41-381f-6621-0fe7-14a54059a90d@redhat.com>
Date:   Fri, 29 Apr 2022 10:48:26 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.8.0
Subject: Re: [PATCH v3] KVM: SEV: Mark nested locking of vcpu->lock
Content-Language: en-US
To:     Hillf Danton <hdanton@sina.com>
Cc:     Peter Gonda <pgonda@google.com>,
        Sean Christopherson <seanjc@google.com>,
        LKML <linux-kernel@vger.kernel.org>
References: <20220407195908.633003-1-pgonda@google.com>
 <CAFNjLiXC0AdOw5f8Ovu47D==ex7F0=WN_Ocirymz4xL=mWvC5A@mail.gmail.com>
 <CAMkAt6r-Mc_YN-gVHuCpTj4E1EmcvyYpP9jhtHo5HRHnoNJAdA@mail.gmail.com>
 <CAMkAt6r+OMPWCbV_svUyGWa0qMzjj2UEG29G6P7jb6uH6yko2w@mail.gmail.com>
 <62e9ece1-5d71-f803-3f65-2755160cf1d1@redhat.com>
 <CAMkAt6q6YLBfo2RceduSXTafckEehawhD4K4hUEuB4ZNqe2kKg@mail.gmail.com>
 <4c0edc90-36a1-4f4c-1923-4b20e7bdbb4c@redhat.com>
 <CAMkAt6oL5qi7z-eh4z7z8WBhpc=Ow6WtcJA5bDi6-aGMnz135A@mail.gmail.com>
 <CAMkAt6rmDrZfN5DbNOTsKFV57PwEnK2zxgBTCbEPeE206+5v5w@mail.gmail.com>
 <20220429010312.4013-1-hdanton@sina.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
In-Reply-To: <20220429010312.4013-1-hdanton@sina.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 4/29/22 03:03, Hillf Danton wrote:
> Wonder if local lock classes [1] help.
> 
> [1]https://lore.kernel.org/lkml/165055518776.3745911.9346998911322224736.stgit@dwillia2-desk3.amr.corp.intel.com/

No, they wouldn't.  Local lock classes are more of a per-subsystem lock, 
while here the issue is that we are taking an arbitrary amount of locks 
at the same time.

Technically it would be possible to put a struct lock_class_key in 
struct kvm_vcpu, but that wouldn't scale and would actually _reduce_ the 
likelihood of lockdep reporting bad things.

The effectiveness of lockdep comes exactly from using the same map for 
all locks in the class, so that AB/BA scenarios are caught throughout 
the whole life of the system.  If each lock has a separate they would be 
caught only if the "B" is exactly the same mutex in both AB and BA cases.

Paolo