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;
Date:   Fri, 20 May 2022 10:25:04 +1000
From:   Nicholas Piggin <npiggin@gmail.com>
Subject: Re: [PATCH v6 20/29] init/main: Delay initialization of the lockup
 detector after smp_init()
To:     Ricardo Neri <ricardo.neri-calderon@linux.intel.com>
Cc:     Andi Kleen <ak@linux.intel.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Lu Baolu <baolu.lu@linux.intel.com>,
        David Woodhouse <dwmw2@infradead.org>,
        Stephane Eranian <eranian@google.com>,
        iommu@lists.linux-foundation.org, Joerg Roedel <joro@8bytes.org>,
        linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org,
        "Ravi V. Shankar" <ravi.v.shankar@intel.com>,
        Ricardo Neri <ricardo.neri@intel.com>,
        Suravee Suthikulpanit <Suravee.Suthikulpanit@amd.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Tony Luck <tony.luck@intel.com>, x86@kernel.org
References: <20220506000008.30892-1-ricardo.neri-calderon@linux.intel.com>
        <20220506000008.30892-21-ricardo.neri-calderon@linux.intel.com>
        <1652178524.7j2o02lrl8.astroid@bobo.none>
        <20220513231640.GA9074@ranerica-svr.sc.intel.com>
In-Reply-To: <20220513231640.GA9074@ranerica-svr.sc.intel.com>
MIME-Version: 1.0
Message-Id: <1653005608.3e5k2jq3of.astroid@bobo.none>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Excerpts from Ricardo Neri's message of May 14, 2022 9:16 am:
> On Tue, May 10, 2022 at 08:38:22PM +1000, Nicholas Piggin wrote:
>> Excerpts from Ricardo Neri's message of May 6, 2022 9:59 am:
>> > Certain implementations of the hardlockup detector require support for
>> > Inter-Processor Interrupt shorthands. On x86, support for these can on=
ly
>> > be determined after all the possible CPUs have booted once (in
>> > smp_init()). Other architectures may not need such check.
>> >=20
>> > lockup_detector_init() only performs the initializations of data
>> > structures of the lockup detector. Hence, there are no dependencies on
>> > smp_init().
>>=20
>=20
> Thank you for your feedback Nicholas!
>=20
>> I think this is the only real thing which affects other watchdog types?
>=20
> Also patches 18 and 19 that decouple the NMI watchdog functionality from
> perf.
>=20
>>=20
>> Not sure if it's a big problem, the secondary CPUs coming up won't
>> have their watchdog active until quite late, and the primary could
>> implement its own timeout in __cpu_up for secondary coming up, and
>> IPI it to get traces if necessary which is probably more robust.
>=20
> Indeed that could work. Another alternative I have been pondering is to b=
oot
> the system with the perf-based NMI watchdog enabled. Once all CPUs are up
> and running, switch to the HPET-based NMI watchdog and free the PMU count=
ers.

Just to cover smp_init()? Unless you could move the watchdog=20
significantly earlier, I'd say it's probably not worth bothering
with.

Yes the boot CPU is doing *some* work that could lock up, but most=20
complexity is in the secondaries coming up and they won't have their own=20
watchdog coverage for a good chunk of that anyway.

If anything I would just add some timeout warning or IPI or something in
those wait loops in x86's __cpu_up code if you are worried about=20
catching issues here. Actually the watchdog probably wouldn't catch any
of those anyway because they either run with interrupts enabled or
touch_nmi_watchdog()! So yeah that'd be pretty pointless.

Thanks,
Nick