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: <20221108115543.1425199-1-apatel@ventanamicro.com> <mvmiljpd4ht.fsf@suse.de>
In-Reply-To: <mvmiljpd4ht.fsf@suse.de>
From:   Anup Patel <anup@brainfault.org>
Date:   Sun, 27 Nov 2022 17:49:34 +0530
Message-ID: <CAAhSdy2FnvRYo6AySDpKAt+O+8RzyH8fNaZekjyqs1pnmg_RQA@mail.gmail.com>
Subject: Re: [PATCH] RISC-V: KVM: Exit run-loop immediately if xfer_to_guest fails
To:     Andreas Schwab <schwab@suse.de>
Cc:     Anup Patel <apatel@ventanamicro.com>,
        Paolo Bonzini <pbonzini@redhat.com>,
        Atish Patra <atishp@atishpatra.org>,
        Palmer Dabbelt <palmer@dabbelt.com>,
        Paul Walmsley <paul.walmsley@sifive.com>,
        Andrew Jones <ajones@ventanamicro.com>, kvm@vger.kernel.org,
        kvm-riscv@lists.infradead.org, linux-riscv@lists.infradead.org,
        linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Tue, Nov 8, 2022 at 5:48 PM Andreas Schwab <schwab@suse.de> wrote:
>
> On Nov 08 2022, Anup Patel wrote:
>
> > If xfer_to_guest_mode_handle_work() fails in the run-loop then exit
> > the run-loop immediately instead of doing it after some more work.
> >
> > Signed-off-by: Anup Patel <apatel@ventanamicro.com>
> > Reviewed-by: Andrew Jones <ajones@ventanamicro.com>
> > ---
> >  arch/riscv/kvm/vcpu.c | 5 +++--
> >  1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/arch/riscv/kvm/vcpu.c b/arch/riscv/kvm/vcpu.c
> > index 71ebbc4821f0..17d5b3f8c2ee 100644
> > --- a/arch/riscv/kvm/vcpu.c
> > +++ b/arch/riscv/kvm/vcpu.c
> > @@ -984,8 +984,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu)
> >       while (ret > 0) {
> >               /* Check conditions before entering the guest */
> >               ret = xfer_to_guest_mode_handle_work(vcpu);
> > -             if (!ret)
> > -                     ret = 1;
> > +             if (ret)
> > +                     continue;
>
> If that is supposed to exit the loop, it would be clearer to just use
> break.

This is a convention within the run-loop that we continue whenever
whenever "ret" is no longer suitable to continue. I don't see any
particular advantage in breaking this convention over here.

>
> > +             ret = 1;
>
> There is a condition on ret <= 0 later in the loop that no longer can be
> true.

Yes, for now the "ret <= 0" check is useless and the compiler will
optimize this comparison. We will be soon having more stuff added
to run-loop (such as AIA update) which will make "ret <= 0" check
useful again.

>
> --
> Andreas Schwab, SUSE Labs, schwab@suse.de
> GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE  1748 E4D4 88E3 0EEA B9D7
> "And now for something completely different."

Queued this patch for Linux-6.2

Thanks,
Anup