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>
 <7a95b2f6-a7ad-5101-baa5-6a19194695a3@redhat.com>
 <YRVebIjxEv87I55b@google.com>
From:   Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <b08a7751-20c3-26fc-522e-c4cf274d9a6c@redhat.com>
Date:   Fri, 13 Aug 2021 09:27:35 +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: <YRVebIjxEv87I55b@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 19:46, Sean Christopherson wrote:
>>> 	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.
> Ya.  There is a measurable performance improvement, but it's really only
> meaningful when there aren't many SPTEs to zap, otherwise the cost of zapping
> completely dominates the time.

I don't understand.  When try_step_down is called by tdp_iter_next, all 
it does is really just the READ_ONCE, because spte_to_child_pt will see 
a non-present PTE and return immediately.  Why do two, presumably cache 
hot, reads cause a measurable performance improvement?

Paolo