linux-kernel - Re: [PATCH v3 27/32] KVM: arm64: Refactor stage2_map_set_prot

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: <YEH3gKYY/Rz7W+50@google.com>
Date:   Fri, 5 Mar 2021 09:18:56 +0000
From:   Quentin Perret <qperret@...gle.com>
To:     Will Deacon <will@...nel.org>
Cc:     catalin.marinas@....com, maz@...nel.org, james.morse@....com,
        julien.thierry.kdev@...il.com, suzuki.poulose@....com,
        android-kvm@...gle.com, linux-kernel@...r.kernel.org,
        kernel-team@...roid.com, kvmarm@...ts.cs.columbia.edu,
        linux-arm-kernel@...ts.infradead.org, tabba@...gle.com,
        mark.rutland@....com, dbrazdil@...gle.com, mate.toth-pal@....com,
        seanjc@...gle.com, robh+dt@...nel.org
Subject: Re: [PATCH v3 27/32] KVM: arm64: Refactor stage2_map_set_prot_attr()

On Thursday 04 Mar 2021 at 20:03:36 (+0000), Will Deacon wrote:
> On Tue, Mar 02, 2021 at 02:59:57PM +0000, Quentin Perret wrote:
> > In order to ease its re-use in other code paths, refactor
> > stage2_map_set_prot_attr() to not depend on a stage2_map_data struct.
> > No functional change intended.
> > 
> > Signed-off-by: Quentin Perret <qperret@...gle.com>
> > ---
> >  arch/arm64/kvm/hyp/pgtable.c | 19 ++++++++-----------
> >  1 file changed, 8 insertions(+), 11 deletions(-)
> > 
> > diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
> > index 8e7059fcfd40..8aa01a9e2603 100644
> > --- a/arch/arm64/kvm/hyp/pgtable.c
> > +++ b/arch/arm64/kvm/hyp/pgtable.c
> > @@ -494,8 +494,7 @@ u64 kvm_get_vtcr(u64 mmfr0, u64 mmfr1, u32 phys_shift)
> >  	return vtcr;
> >  }
> >  
> > -static int stage2_map_set_prot_attr(enum kvm_pgtable_prot prot,
> > -				    struct stage2_map_data *data)
> > +static kvm_pte_t stage2_get_prot_attr(enum kvm_pgtable_prot prot)
> >  {
> >  	bool device = prot & KVM_PGTABLE_PROT_DEVICE;
> >  	kvm_pte_t attr = device ? PAGE_S2_MEMATTR(DEVICE_nGnRE) :
> > @@ -504,15 +503,15 @@ static int stage2_map_set_prot_attr(enum kvm_pgtable_prot prot,
> >  
> >  	if (prot & KVM_PGTABLE_PROT_NONE) {
> >  		if (prot != KVM_PGTABLE_PROT_NONE)
> > -			return -EINVAL;
> > +			return 0;
> 
> Hmm, does the architecture actually say that having all these attributes
> as 0 is illegal?

Hmm, that's a good point, that might not be the case. I assumed we would
have no use for this, but there we can easily avoid the restriction
so...

> If not, I think it would be better to keep the int return
> code and replace the 'data' parameter with a pointer to a kvm_pte_t.
> 
> Does that work?

I think so yes, I'll fix it up.

Cheers,
Quentin