linux-kernel - Re: [RFC PATCH] KVM: x86/mmu: replace BUG() with KVM

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] [day] [month] [year] [list]

Message-ID: <CAL715WKymGc60gLHT9_g+RDs3TgSBCZqdZJD0D3Z=ZrCgH05nA@mail.gmail.com>
Date:   Wed, 23 Nov 2022 15:48:50 -0800
From:   Mingwei Zhang <mizhang@...gle.com>
To:     Sean Christopherson <seanjc@...gle.com>
Cc:     Paolo Bonzini <pbonzini@...hat.com>,
        "H. Peter Anvin" <hpa@...or.com>, kvm@...r.kernel.org,
        linux-kernel@...r.kernel.org,
        Nagareddy Reddy <nspreddy@...gle.com>,
        Jim Mattson <jmattson@...gle.com>,
        David Matlack <dmatlack@...gle.com>
Subject: Re: [RFC PATCH] KVM: x86/mmu: replace BUG() with KVM_BUG() in shadow mmu

On Wed, Nov 23, 2022 at 3:18 PM Sean Christopherson <seanjc@...gle.com> wrote:
>
> On Wed, Nov 23, 2022, Mingwei Zhang wrote:
> > diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
> > index 4736d7849c60..075d31b0db9c 100644
> > --- a/arch/x86/kvm/mmu/mmu.c
> > +++ b/arch/x86/kvm/mmu/mmu.c
> > @@ -955,12 +955,12 @@ static void pte_list_remove(u64 *spte, struct kvm_rmap_head *rmap_head)
> >
> >       if (!rmap_head->val) {
> >               pr_err("%s: %p 0->BUG\n", __func__, spte);
> > -             BUG();
> > +             KVM_BUG();
>
> This won't compile.  KVM_BUG() isn't a direct replacement for BUG(), it's more
> akin to WARN().
>
> And that's why I suggested this be RFC: @kvm needs to be plumbed down here in order
> to use KVM_BUG().  I don't mind that too much, it's just a little unfortunate.

I wonder if using kvm_get_running_vcpu()->kvm is safe here? Assuming
we can, then @kvm plumbing shouldn't be a problem.