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;
MIME-Version: 1.0
References: <20210719104735.3681732-1-qperret@google.com> <20210719104735.3681732-7-qperret@google.com>
In-Reply-To: <20210719104735.3681732-7-qperret@google.com>
From:   Fuad Tabba <tabba@google.com>
Date:   Tue, 20 Jul 2021 11:17:03 +0100
Message-ID: <CA+EHjTzjT2iv=9jsTNquSAD-_AWqpyCKVPiuCBWGzpxcAkWf9w@mail.gmail.com>
Subject: Re: [PATCH 06/14] KVM: arm64: Tolerate re-creating hyp mappings to
 set ignored bits
To:     Quentin Perret <qperret@google.com>
Cc:     maz@kernel.org, james.morse@arm.com, alexandru.elisei@arm.com,
        suzuki.poulose@arm.com, catalin.marinas@arm.com, will@kernel.org,
        linux-arm-kernel@lists.infradead.org, kvmarm@lists.cs.columbia.edu,
        linux-kernel@vger.kernel.org, ardb@kernel.org, qwandor@google.com,
        dbrazdil@google.com, kernel-team@android.com
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi Quentin,


On Mon, Jul 19, 2021 at 11:47 AM Quentin Perret <qperret@google.com> wrote:
>
> The current hypervisor stage-1 mapping code doesn't allow changing an
> existing valid mapping. Relax this condition by allowing changes that
> only target ignored bits, as that will soon be needed to annotate shared
> pages.
>
> Signed-off-by: Quentin Perret <qperret@google.com>
> ---
>  arch/arm64/kvm/hyp/pgtable.c | 18 ++++++++++++++++--
>  1 file changed, 16 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> index a0ac8c2bc174..34cf67997a82 100644
> --- a/arch/arm64/kvm/hyp/pgtable.c
> +++ b/arch/arm64/kvm/hyp/pgtable.c
> @@ -362,6 +362,17 @@ static int hyp_set_prot_attr(enum kvm_pgtable_prot prot, kvm_pte_t *ptep)
>         return 0;
>  }
>
> +static bool hyp_pte_needs_update(kvm_pte_t old, kvm_pte_t new)
> +{
> +       if (old == new)
> +               return false;
> +
> +       if (!kvm_pte_valid(old))
> +               return true;
> +
> +       return !WARN_ON((old ^ new) & ~KVM_PTE_LEAF_ATTR_IGNORED);

Wouldn't this return false if both ignored and non-ignored bits were
different, or is that not possible (judging by the WARN_ON)? If it is,
then it would need an update, wouldn't it?

Thanks,
/fuad


> +}
> +
>  static bool hyp_map_walker_try_leaf(u64 addr, u64 end, u32 level,
>                                     kvm_pte_t *ptep, struct hyp_map_data *data)
>  {
> @@ -371,9 +382,12 @@ static bool hyp_map_walker_try_leaf(u64 addr, u64 end, u32 level,
>         if (!kvm_block_mapping_supported(addr, end, phys, level))
>                 return false;
>
> -       /* Tolerate KVM recreating the exact same mapping */
> +       /*
> +        * Tolerate KVM recreating the exact same mapping, or changing ignored
> +        * bits.
> +        */
>         new = kvm_init_valid_leaf_pte(phys, data->attr, level);
> -       if (old != new && !WARN_ON(kvm_pte_valid(old)))
> +       if (hyp_pte_needs_update(old, new))
>                 smp_store_release(ptep, new);
>
>         data->phys += granule;
> --
> 2.32.0.402.g57bb445576-goog
>