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;
Date:   Fri, 26 May 2023 11:13:01 -0700
In-Reply-To: <20230425113932.3148-1-ubizjak@gmail.com>
Mime-Version: 1.0
References: <20230425113932.3148-1-ubizjak@gmail.com>
Message-ID: <ZHD2rYBCSe5OSYIU@google.com>
Subject: Re: [PATCH] KVM: x86/mmu: Add comment on try_cmpxchg64 usage in tdp_mmu_set_spte_atomic
From:   Sean Christopherson <seanjc@google.com>
To:     Uros Bizjak <ubizjak@gmail.com>
Cc:     kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
        Paolo Bonzini <pbonzini@redhat.com>,
        David Matlack <dmatlack@google.com>
Content-Type: text/plain; charset="us-ascii"
Precedence: bulk

On Tue, Apr 25, 2023, Uros Bizjak wrote:
> Commit aee98a6838d5 ("KVM: x86/mmu: Use try_cmpxchg64 in
> tdp_mmu_set_spte_atomic") removed the comment that iter->old_spte is
> updated when different logical CPU modifies the page table entry.
> Although this is what try_cmpxchg does implicitly, it won't hurt
> if this fact is explicitly mentioned in a restored comment.
> 
> Cc: Paolo Bonzini <pbonzini@redhat.com>
> Cc: Sean Christopherson <seanjc@google.com>
> Cc: David Matlack <dmatlack@google.com>
> Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
> ---
>  arch/x86/kvm/mmu/tdp_mmu.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kvm/mmu/tdp_mmu.c b/arch/x86/kvm/mmu/tdp_mmu.c
> index 7c25dbf32ecc..5d126b015086 100644
> --- a/arch/x86/kvm/mmu/tdp_mmu.c
> +++ b/arch/x86/kvm/mmu/tdp_mmu.c
> @@ -655,8 +655,16 @@ static inline int tdp_mmu_set_spte_atomic(struct kvm *kvm,
>  	 * Note, fast_pf_fix_direct_spte() can also modify TDP MMU SPTEs and
>  	 * does not hold the mmu_lock.
>  	 */
> -	if (!try_cmpxchg64(sptep, &iter->old_spte, new_spte))
> +	if (!try_cmpxchg64(sptep, &iter->old_spte, new_spte)) {
> +		/*
> +		 * The page table entry was modified by a different logical
> +		 * CPU. In this case the above try_cmpxchg updates
> +		 * iter->old_spte with the current value, so the caller
> +		 * operates on fresh data, e.g. if it retries
> +		 * tdp_mmu_set_spte_atomic().
> +		 */

If there's no objection, when applying I'll massage this to extend the comment
above the try_cmpxchg64(), e.g.

	/*
	 * Note, fast_pf_fix_direct_spte() can also modify TDP MMU SPTEs and
	 * does not hold the mmu_lock.  On failure, i.e. if a different logical
	 * CPU modified the SPTE, try_cmpxchg64() updates iter->old_spte with
	 * the current value, so the caller operates on fresh data, e.g. if it
	 * retries tdp_mmu_set_spte_atomic()
	 */
	if (!try_cmpxchg64(sptep, &iter->old_spte, new_spte))
		return -EBUSY;