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;
MIME-Version: 1.0
References: <20230414172922.812640-1-rananta@google.com> <20230414172922.812640-8-rananta@google.com>
 <ZF5xLrr2tEYdLL1i@linux.dev> <CAJHc60wUu3xB4J8oJ+FCxerDad1TzZLCMgHYGFfv0K-hzC0qmw@mail.gmail.com>
 <ZGPPj1AXS0Uah2Ug@linux.dev>
In-Reply-To: <ZGPPj1AXS0Uah2Ug@linux.dev>
From:   Raghavendra Rao Ananta <rananta@google.com>
Date:   Tue, 16 May 2023 11:54:55 -0700
Message-ID: <CAJHc60wF19h9ixEQK5SS+oGA41xNqrFQaH9VE6sS0DXjQFCh7A@mail.gmail.com>
Subject: Re: [PATCH v3 7/7] KVM: arm64: Use TLBI range-based intructions for unmap
To:     Oliver Upton <oliver.upton@linux.dev>
Cc:     Marc Zyngier <maz@kernel.org>, James Morse <james.morse@arm.com>,
        Suzuki K Poulose <suzuki.poulose@arm.com>,
        Ricardo Koller <ricarkol@google.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Jing Zhang <jingzhangos@google.com>,
        Colton Lewis <coltonlewis@google.com>,
        linux-arm-kernel@lists.infradead.org, kvmarm@lists.linux.dev,
        linux-kernel@vger.kernel.org, kvm@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, May 16, 2023 at 11:46=E2=80=AFAM Oliver Upton <oliver.upton@linux.d=
ev> wrote:
>
> On Tue, May 16, 2023 at 10:21:33AM -0700, Raghavendra Rao Ananta wrote:
> > On Fri, May 12, 2023 at 10:02=E2=80=AFAM Oliver Upton <oliver.upton@lin=
ux.dev> wrote:
> > > >  int kvm_pgtable_stage2_unmap(struct kvm_pgtable *pgt, u64 addr, u6=
4 size)
> > > >  {
> > > > +     int ret;
> > > > +     struct stage2_unmap_data unmap_data =3D {
> > > > +             .pgt =3D pgt,
> > > > +             /*
> > > > +              * If FEAT_TLBIRANGE is implemented, defer the indivi=
dial PTE
> > > > +              * TLB invalidations until the entire walk is finishe=
d, and
> > > > +              * then use the range-based TLBI instructions to do t=
he
> > > > +              * invalidations. Condition this upon S2FWB in order =
to avoid
> > > > +              * a page-table walk again to perform the CMOs after =
TLBI.
> > > > +              */
> > > > +             .skip_pte_tlbis =3D system_supports_tlb_range() &&
> > > > +                                     stage2_has_fwb(pgt),
> > >
> > > Why can't the underlying walker just call these two helpers directly?
> > > There are static keys behind these...
> > >
> > I wasn't aware of that. Although, I tried to look into the
> > definitions, but couldn't understand how static keys are at play here.
> > By any chance are you referring to the alternative_has_feature_*()
> > implementations when checking for cpu capabilities?
>
> Ah, right, these were recently changed to rely on alternative patching
> in commit 21fb26bfb01f ("arm64: alternatives: add alternative_has_feature=
_*()").
> Even still, the significance remains as the alternative patching
> completely eliminates a conditional branch on the presence of a
> particular feature.
>
> Initializing a local with the presence/absence of a feature defeats such
> an optimization.
>
Thanks for the info! Introduction of stage2_unmap_defer_tlb_flush()
(in patch-7/7) would call these functions as needed and get rid of
'skip_pte_tlbis'.

- Raghavendra
> --
> Thanks,
> Oliver