Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII;
 format=flowed
Content-Transfer-Encoding: 7bit
Date:   Wed, 30 Dec 2020 10:06:19 +0000
From:   Marc Zyngier <maz@kernel.org>
To:     Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc:     David Brazdil <dbrazdil@google.com>, kvmarm@lists.cs.columbia.edu,
        Mark Rutland <mark.rutland@arm.com>,
        Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
        Suzuki K Poulose <suzuki.poulose@arm.com>,
        Sudeep Holla <sudeep.holla@arm.com>,
        linux-kernel@vger.kernel.org, James Morse <james.morse@arm.com>,
        linux-arm-kernel@lists.infradead.org,
        Catalin Marinas <catalin.marinas@arm.com>,
        kernel-team@android.com, Will Deacon <will@kernel.org>,
        Julien Thierry <julien.thierry.kdev@gmail.com>
Subject: Re: [PATCH] KVM: arm64: Allow PSCI SYSTEM_OFF/RESET to return
In-Reply-To: <20201229170412.GN1551@shell.armlinux.org.uk>
References: <20201229160059.64135-1-dbrazdil@google.com>
 <20201229170412.GN1551@shell.armlinux.org.uk>
User-Agent: Roundcube Webmail/1.4.9
Message-ID: <0e1d555404e4ae8edcf6737735dc0eb7@kernel.org>
Precedence: bulk

On 2020-12-29 17:04, Russell King - ARM Linux admin wrote:
> On Tue, Dec 29, 2020 at 04:00:59PM +0000, David Brazdil wrote:
>> The KVM/arm64 PSCI relay assumes that SYSTEM_OFF and SYSTEM_RESET 
>> should
>> not return, as dictated by the PSCI spec. However, there is firmware 
>> out
>> there which breaks this assumption, leading to a hyp panic. Make KVM
>> more robust to broken firmware by allowing these to return.
> 
> Are you sure you should just return?
> 
> We've had issues in the past with Linux reboot(2) that returns
> to userspace, allowing on 32-bit ARM for example watchdogs to
> unexpectedly continue being serviced.

I don't think this changes anything compared to the case where
the PSCI relay isn't enabled. The EL1 part of the kernel would
see the SYSTEM_RESET call return, and handle it accordingly
(stay in a while(1) loop).

This is consistent with the PSCI relay design goal of being
invisible to the EL1 kernel.

Thanks,

         M.
-- 
Jazz is not dead. It just smells funny...