linux-kernel - Re: [PATCH V3 07/12] KVM: X86/MMU: Remove the check of the return value of to_shadow

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: <YtczcE8bxNJfWwZ5@google.com>
Date:   Tue, 19 Jul 2022 22:42:56 +0000
From:   Sean Christopherson <seanjc@...gle.com>
To:     Lai Jiangshan <jiangshanlai@...il.com>
Cc:     linux-kernel@...r.kernel.org, kvm@...r.kernel.org,
        Paolo Bonzini <pbonzini@...hat.com>,
        Vitaly Kuznetsov <vkuznets@...hat.com>,
        Maxim Levitsky <mlevitsk@...hat.com>,
        David Matlack <dmatlack@...gle.com>,
        Lai Jiangshan <jiangshan.ljs@...group.com>
Subject: Re: [PATCH V3 07/12] KVM: X86/MMU: Remove the check of the return
 value of to_shadow_page()

On Sat, May 21, 2022, Lai Jiangshan wrote:
> From: Lai Jiangshan <jiangshan.ljs@...group.com>
> 
> Remove the check of the return value of to_shadow_page() in
> mmu_free_root_page(), kvm_mmu_free_guest_mode_roots(), is_unsync_root()
> and is_tdp_mmu() because it can not return NULL.

s/can not/can no longer, to make it very clear the to_shadow_page() _could_ return
NULL in the past, but something changed.

And please explain _why_ it can no longer return NULL, even if that's just a
reference to kvm_mmu_alloc_local_shadow_page() or whatever it ends up being named.
That can also tie into the obsolete sp handling.

> Remove the check of the return value of to_shadow_page() in
> is_page_fault_stale() and is_obsolete_root() because it can not return
> NULL and the obsoleting for local shadow page is already handled by
> a different way.
> 
> When the obsoleting process is done, all the obsoleted non-local shadow
> pages are already unlinked from the local shadow pages by the help of
> the parent rmap from the children and the local shadow pages become
> theoretically valid again.  The local shadow page can be freed if
> is_obsolete_sp() return true, or be reused if is_obsolete_sp() becomes
> false.
> 
> Reviewed-by: David Matlack <dmatlack@...gle.com>
> Signed-off-by: Lai Jiangshan <jiangshan.ljs@...group.com>
> ---