KVM Archive on lore.kernel.org
 help / color / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Wanpeng Li <kernellwp@gmail.com>, Marc Zyngier <maz@kernel.org>,
	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>,
	Ben Gardon <bgardon@google.com>
Subject: Re: [PATCH v2 09/10] KVM: Don't take mmu_lock for range invalidation unless necessary
Date: Mon, 19 Apr 2021 15:09:01 +0000
Message-ID: <YH2dDRBXJcbUcbLi@google.com> (raw)
In-Reply-To: <2a7670e4-94c0-9f35-74de-a7d5b1504ced@redhat.com>

On Mon, Apr 19, 2021, Paolo Bonzini wrote:
> On 19/04/21 10:49, Wanpeng Li wrote:
> > I saw this splatting:
> > 
> >   ======================================================
> >   WARNING: possible circular locking dependency detected
> >   5.12.0-rc3+ #6 Tainted: G           OE
> >   ------------------------------------------------------
> >   qemu-system-x86/3069 is trying to acquire lock:
> >   ffffffff9c775ca0 (mmu_notifier_invalidate_range_start){+.+.}-{0:0},
> > at: __mmu_notifier_invalidate_range_end+0x5/0x190
> > 
> >   but task is already holding lock:
> >   ffffaff7410a9160 (&kvm->mmu_notifier_slots_lock){.+.+}-{3:3}, at:
> > kvm_mmu_notifier_invalidate_range_start+0x36d/0x4f0 [kvm]
> 
> I guess it is possible to open-code the wait using a readers count and a
> spinlock (see patch after signature).  This allows including the
> rcu_assign_pointer in the same critical section that checks the number
> of readers.  Also on the plus side, the init_rwsem() is replaced by
> slightly nicer code.

Ugh, the count approach is nearly identical to Ben's original code.  Using a
rwsem seemed so clever :-/

> IIUC this could be extended to non-sleeping invalidations too, but I
> am not really sure about that.

Yes, that should be fine.

> There are some issues with the patch though:
> 
> - I am not sure if this should be a raw spin lock to avoid the same issue
> on PREEMPT_RT kernel.  That said the critical section is so tiny that using
> a raw spin lock may make sense anyway

If using spinlock_t is problematic, wouldn't mmu_lock already be an issue?  Or
am I misunderstanding your concern?

> - this loses the rwsem fairness.  On the other hand, mm/mmu_notifier.c's
> own interval-tree-based filter is also using a similar mechanism that is
> likewise not fair, so it should be okay.

The one concern I had with an unfair mechanism of this nature is that, in theory,
the memslot update could be blocked indefinitely.

> Any opinions?  For now I placed the change below in kvm/queue, but I'm
> leaning towards delaying this optimization to the next merge window.

I think delaying it makes sense.

> @@ -1333,9 +1351,22 @@ static struct kvm_memslots *install_new_memslots(struct kvm *kvm,
>  	WARN_ON(gen & KVM_MEMSLOT_GEN_UPDATE_IN_PROGRESS);
>  	slots->generation = gen | KVM_MEMSLOT_GEN_UPDATE_IN_PROGRESS;
> -	down_write(&kvm->mmu_notifier_slots_lock);
> +	/*
> +	 * This cannot be an rwsem because the MMU notifier must not run
> +	 * inside the critical section.  A sleeping rwsem cannot exclude
> +	 * that.

How on earth did you decipher that from the splat?  I stared at it for a good
five minutes and was completely befuddled.

> +	 */
> +	spin_lock(&kvm->mn_invalidate_lock);
> +	prepare_to_rcuwait(&kvm->mn_memslots_update_rcuwait);
> +	while (kvm->mn_active_invalidate_count) {
> +		set_current_state(TASK_UNINTERRUPTIBLE);
> +		spin_unlock(&kvm->mn_invalidate_lock);
> +		schedule();
> +		spin_lock(&kvm->mn_invalidate_lock);
> +	}
> +	finish_rcuwait(&kvm->mn_memslots_update_rcuwait);
>  	rcu_assign_pointer(kvm->memslots[as_id], slots);
> -	up_write(&kvm->mmu_notifier_slots_lock);
> +	spin_unlock(&kvm->mn_invalidate_lock);
>  	synchronize_srcu_expedited(&kvm->srcu);
> 

  reply index

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-02  0:56 [PATCH v2 00/10] KVM: Consolidate and optimize MMU notifiers Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 01/10] KVM: Assert that notifier count is elevated in .change_pte() Sean Christopherson
2021-04-02 11:08   ` Paolo Bonzini
2021-04-02  0:56 ` [PATCH v2 02/10] KVM: Move x86's MMU notifier memslot walkers to generic code Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 03/10] KVM: arm64: Convert to the gfn-based MMU notifier callbacks Sean Christopherson
2021-04-12 10:12   ` Marc Zyngier
2021-04-02  0:56 ` [PATCH v2 04/10] KVM: MIPS/MMU: " Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 05/10] KVM: PPC: " Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 06/10] KVM: Kill off the old hva-based " Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 07/10] KVM: Move MMU notifier's mmu_lock acquisition into common helper Sean Christopherson
2021-04-02  9:35   ` Paolo Bonzini
2021-04-02 14:59     ` Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 08/10] KVM: Take mmu_lock when handling MMU notifier iff the hva hits a memslot Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 09/10] KVM: Don't take mmu_lock for range invalidation unless necessary Sean Christopherson
2021-04-02  9:34   ` Paolo Bonzini
2021-04-02 14:59     ` Sean Christopherson
2021-04-19  8:49   ` Wanpeng Li
2021-04-19 13:50     ` Paolo Bonzini
2021-04-19 15:09       ` Sean Christopherson [this message]
2021-04-19 22:09         ` Paolo Bonzini
2021-04-20  1:17           ` Sean Christopherson
2021-04-02  0:56 ` [PATCH v2 10/10] KVM: x86/mmu: Allow yielding during MMU notifier unmap/zap, if possible Sean Christopherson
2021-04-02 12:17 ` [PATCH v2 00/10] KVM: Consolidate and optimize MMU notifiers Paolo Bonzini
2021-04-12 10:27   ` Marc Zyngier

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YH2dDRBXJcbUcbLi@google.com \
    --to=seanjc@google.com \
    --cc=bgardon@google.com \
    --cc=jmattson@google.com \
    --cc=joro@8bytes.org \
    --cc=kernellwp@gmail.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maz@kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=vkuznets@redhat.com \
    --cc=wanpengli@tencent.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

KVM Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/kvm/0 kvm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 kvm kvm/ https://lore.kernel.org/kvm \
		kvm@vger.kernel.org
	public-inbox-index kvm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.kvm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git