Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=utf-8
Mime-Version: 1.0 (1.0)
Subject: Re: [PATCH v4] x86/power: Fix 'nosmt' vs. hibernation triple fault during resume
From:   Andy Lutomirski <luto@amacapital.net>
In-Reply-To: <nycvar.YFH.7.76.1905312251350.1962@cbobk.fhfr.pm>
Date:   Fri, 31 May 2019 14:22:27 -0700
Cc:     Andy Lutomirski <luto@kernel.org>,
        "Rafael J. Wysocki" <rjw@rjwysocki.net>,
        Josh Poimboeuf <jpoimboe@redhat.com>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        the arch/x86 maintainers <x86@kernel.org>,
        Pavel Machek <pavel@ucw.cz>, Ingo Molnar <mingo@redhat.com>,
        Borislav Petkov <bp@alien8.de>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Linux PM <linux-pm@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <98E57C7E-24E2-4EB8-A14E-FCA80316F812@amacapital.net>
References: <nycvar.YFH.7.76.1905282326360.1962@cbobk.fhfr.pm> <20190531051456.fzkvn62qlkf6wqra@treble> <nycvar.YFH.7.76.1905311045240.1962@cbobk.fhfr.pm> <5564116.e9OFvgDRbB@kreacher> <CALCETrUpseta+NrhVwzzVFTe-BkBHtDUJBO22ci3mAsVR+XOog@mail.gmail.com> <nycvar.YFH.7.76.1905311628330.1962@cbobk.fhfr.pm> <B7AC83ED-3F11-42B9-8506-C842A5937B50@amacapital.net> <nycvar.YFH.7.76.1905311651450.1962@cbobk.fhfr.pm> <CALCETrUQzZTRnvmOS09UvRM9UCGEDvSdbJtkeeEa2foMf+hF2w@mail.gmail.com> <nycvar.YFH.7.76.1905312251350.1962@cbobk.fhfr.pm>
To:     Jiri Kosina <jikos@kernel.org>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk


> On May 31, 2019, at 2:05 PM, Jiri Kosina <jikos@kernel.org> wrote:
>=20
>> On Fri, 31 May 2019, Andy Lutomirski wrote:
>>=20
>> The Intel SDM Vol 3 34.10 says:
>>=20
>> If the HLT instruction is restarted, the processor will generate a
>> memory access to fetch the HLT instruction (if it is
>> not in the internal cache), and execute a HLT bus transaction. This
>> behavior results in multiple HLT bus transactions
>> for the same HLT instruction.
>=20
> Which basically means that both hibernation and kexec have been broken in=20=

> this respect for gazillions of years, and seems like noone noticed. Makes=20=

> one wonder what the reason for that might be.
>=20
> Either SDM is not precise and the refetch actually never happens for real=20=

> (or is always in these cases satisfied from I$ perhaps?), or ... ?
>=20
> So my patch basically puts things back where they have been for ages=20
> (while mwait is obviously much worse, as that gets woken up by the write=20=

> to the monitored address, which inevitably does happen during resume), but=
=20
> seems like SDM is suggesting that we've been in a grey zone wrt RSM at=20
> least for all those ages.
>=20
> So perhaps we really should ditch resume_play_dead() altogether=20
> eventually, and replace it with sending INIT IPI around instead (and then=20=

> waking the CPUs properly via INIT INIT START). I'd still like to do that=20=

> for 5.3 though, as that'd be slightly bigger surgery, and conservatively=20=

> put things basically back to state they have been up to now for 5.2.
>=20


Seems reasonable to me.  I would guess that it mostly works because SMI isn=E2=
=80=99t all that common and the window where it matters is short.  Or maybe t=
he SDM is misleading.=