Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [PATCH 2/2] KVM: x86/mmu: Don't step down in the TDP iterator
 when zapping all SPTEs
To:     Sean Christopherson <seanjc@google.com>
Cc:     Ben Gardon <bgardon@google.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>,
        Wanpeng Li <wanpengli@tencent.com>,
        Jim Mattson <jmattson@google.com>,
        Joerg Roedel <joro@8bytes.org>, kvm <kvm@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>
References: <20210812050717.3176478-1-seanjc@google.com>
 <20210812050717.3176478-3-seanjc@google.com>
 <CANgfPd8HSYZbqmi21XQ=XeMCndXJ0+Ld0eZNKPWLa1fKtutiBA@mail.gmail.com>
 <YRVVWC31fuZiw9tT@google.com>
 <928be04d-e60e-924c-1f3a-cb5fef8b0042@redhat.com>
 <YRVbamoQhvPmrEgK@google.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <7a95b2f6-a7ad-5101-baa5-6a19194695a3@redhat.com>
Date:   Thu, 12 Aug 2021 19:38:43 +0200
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101
 Thunderbird/78.11.0
MIME-Version: 1.0
In-Reply-To: <YRVbamoQhvPmrEgK@google.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 12/08/21 19:33, Sean Christopherson wrote:
> On Thu, Aug 12, 2021, Paolo Bonzini wrote:
>> On 12/08/21 19:07, Sean Christopherson wrote:
>>> Yeah, I was/am on the fence too, I almost included a blurb in the cover letter
>>> saying as much.  I'll do that for v2 and let Paolo decide.
>>
>> I think it makes sense to have it.  You can even use the ternary operator
> 
> Hah, yeah, I almost used a ternary op.  Honestly don't know why I didn't, guess
> my brain flipped a coin.
> 
>>
>> 	/*
>> 	 * When zapping everything, all entries at the top level
>> 	 * ultimately go away, and the levels below go down with them.
>> 	 * So do not bother iterating all the way down to the leaves.
> 
> The subtle part is that try_step_down() won't actually iterate down because it
> explicitly rereads and rechecks the SPTE.
> 
> 	if (iter->level == iter->min_level)
> 		return false;
> 
> 	/*
> 	 * Reread the SPTE before stepping down to avoid traversing into page
> 	 * tables that are no longer linked from this entry.
> 	 */
> 	iter->old_spte = READ_ONCE(*rcu_dereference(iter->sptep));  \
>                                                                       ---> this is the code that is avoided
> 	child_pt = spte_to_child_pt(iter->old_spte, iter->level);   /
> 	if (!child_pt)
> 		return false;

Ah, right - so I agree with Ben that it's not too important.

Paolo