linux-kernel - Re: [PATCH v5 01/13] KVM: x86: Cache total page count to avoid traversing the memslot array

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: <YW9Fi128rYxiF1v3@google.com>
Date:   Tue, 19 Oct 2021 22:24:11 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     "Maciej S. Szmigiero" <mail@...iej.szmigiero.name>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Wanpeng Li <wanpengli@...cent.com>,
        Jim Mattson <jmattson@...gle.com>,
        Igor Mammedov <imammedo@...hat.com>,
        Marc Zyngier <maz@...nel.org>,
        James Morse <james.morse@....com>,
        Julien Thierry <julien.thierry.kdev@...il.com>,
        Suzuki K Poulose <suzuki.poulose@....com>,
        Huacai Chen <chenhuacai@...nel.org>,
        Aleksandar Markovic <aleksandar.qemu.devel@...il.com>,
        Paul Mackerras <paulus@...abs.org>,
        Christian Borntraeger <borntraeger@...ibm.com>,
        Janosch Frank <frankja@...ux.ibm.com>,
        David Hildenbrand <david@...hat.com>,
        Cornelia Huck <cohuck@...hat.com>,
        Claudio Imbrenda <imbrenda@...ux.ibm.com>,
        Joerg Roedel <joro@...tes.org>, kvm@...r.kernel.org,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5 01/13] KVM: x86: Cache total page count to avoid
 traversing the memslot array

On Mon, Sep 20, 2021, Maciej S. Szmigiero wrote:
> From: "Maciej S. Szmigiero" <maciej.szmigiero@...cle.com>
> 
> There is no point in recalculating from scratch the total number of pages
> in all memslots each time a memslot is created or deleted.
> 
> Just cache the value and update it accordingly on each such operation so
> the code doesn't need to traverse the whole memslot array each time.
> 
> Signed-off-by: Maciej S. Szmigiero <maciej.szmigiero@...cle.com>
> ---
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index 28ef14155726..65fdf27b9423 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -11609,9 +11609,23 @@ void kvm_arch_commit_memory_region(struct kvm *kvm,
>  				const struct kvm_memory_slot *new,
>  				enum kvm_mr_change change)
>  {
> -	if (!kvm->arch.n_requested_mmu_pages)
> -		kvm_mmu_change_mmu_pages(kvm,
> -				kvm_mmu_calculate_default_mmu_pages(kvm));
> +	if (change == KVM_MR_CREATE)
> +		kvm->arch.n_memslots_pages += new->npages;
> +	else if (change == KVM_MR_DELETE) {
> +		WARN_ON(kvm->arch.n_memslots_pages < old->npages);
> +		kvm->arch.n_memslots_pages -= old->npages;
> +	}
> +
> +	if (!kvm->arch.n_requested_mmu_pages) {

Hmm, once n_requested_mmu_pages is set it can't be unset.  That means this can be
further optimized to skip avoid taking mmu_lock on flags-only changes (and
memslot movement).  E.g.

	if (!kvm->arch.n_requested_mmu_pages &&
	    (change == KVM_MR_CREATE || change == KVM_MR_DELETE)) {

	}

It's a little risky, but kvm_vm_ioctl_set_nr_mmu_pages() would need to be modified
to allow clearing n_requested_mmu_pages and it already takes slots_lock, so IMO
it's ok to force kvm_vm_ioctl_set_nr_mmu_pages() to recalculate pages if it wants
to allow reverting back to the default.

> +		u64 memslots_pages;
> +		unsigned long nr_mmu_pages;
> +
> +		memslots_pages = kvm->arch.n_memslots_pages * KVM_PERMILLE_MMU_PAGES;
> +		do_div(memslots_pages, 1000);
> +		nr_mmu_pages = max_t(typeof(nr_mmu_pages),
> +				     memslots_pages, KVM_MIN_ALLOC_MMU_PAGES);

"memslots_pages" is a bit of a misnomer.  Any objection to avoiding naming problems
by explicitly casting to an "unsigned long" and simply operating on nr_mmu_pages?

		nr_mmu_pages = (unsigned long)kvm->arch.n_memslots_pages;
		nr_mmu_pages *= (KVM_PERMILLE_MMU_PAGES / 1000);
		nr_mmu_pages = max(nr_mmu_pages, KVM_MIN_ALLOC_MMU_PAGES);
		kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);

E.g. the whole thing can be

	if (!kvm->arch.n_requested_mmu_pages &&
	    (change == KVM_MR_CREATE || change == KVM_MR_DELETE)) {
		unsigned long nr_mmu_pages;

		if (change == KVM_MR_CREATE) {
			kvm->arch.n_memslots_pages += new->npages;
		} else {
			WARN_ON(kvm->arch.n_memslots_pages < old->npages);
			kvm->arch.n_memslots_pages -= old->npages;
		}

		nr_mmu_pages = (unsigned long)kvm->arch.n_memslots_pages;
		nr_mmu_pages *= (KVM_PERMILLE_MMU_PAGES / 1000);
		nr_mmu_pages = max(nr_mmu_pages, KVM_MIN_ALLOC_MMU_PAGES);
		kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
	}

> +		kvm_mmu_change_mmu_pages(kvm, nr_mmu_pages);
> +	}
>  
>  	kvm_mmu_slot_apply_flags(kvm, old, new, change);
>