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:   Wed, 17 Aug 2022 14:10:51 +0200
From:   Borislav Petkov <bp@alien8.de>
To:     Ashok Raj <ashok.raj@intel.com>
Cc:     Ingo Molnar <mingo@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Tony Luck <tony.luck@intel.com>,
        Dave Hansen <dave.hansen@intel.com>,
        LKML Mailing List <linux-kernel@vger.kernel.org>,
        X86-kernel <x86@kernel.org>,
        Andy Lutomirski <luto@amacapital.net>,
        Tom Lendacky <thomas.lendacky@amd.com>,
        Jacon Jun Pan <jacob.jun.pan@intel.com>
Subject: Re: [PATCH v3 3/5] x86/microcode: Avoid any chance of MCE's during
 microcode update
Message-ID: <Yvzay5jOu5XBJpWa@zn.tnic>
References: <20220817051127.3323755-1-ashok.raj@intel.com>
 <20220817051127.3323755-4-ashok.raj@intel.com>
 <Yvybq+hYT4tG/yAg@gmail.com>
 <Yvyfi9XC8bu0cOG+@gmail.com>
 <YvyiHGMbp2MtV0Vr@zn.tnic>
 <YvzXsf0mGEcOlZC5@araj-dh-work>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <YvzXsf0mGEcOlZC5@araj-dh-work>
Precedence: bulk

On Wed, Aug 17, 2022 at 11:57:37AM +0000, Ashok Raj wrote:
> You have this reversed. if you get an MCE and MCIP=1 -> shutdown

Yeah yeah.

> When MCE's happen during the update they are always fatal errors.

How did you decide that?

Because all CPUs are executing the loop and thus no user process?

> What we do here by setting MCIP=1, we promote to a more severe shutdown.

It probably should say somewhere that a shutdown is possible. Because if
the shutdown really happens, you get a black screen and no info as to
why...

> Ideally I would rather let the fallout happen since its observable vs a
> blind shutdown is what we are promoting to.

What fallout do you mean exactly?

> Shutdown, shutdown.. There is only 1 MCE no matter how many CPUs you have.

Because all CPUs are executing the loop? Or how do you decide this?

> Exception is the Local MCE which is recoverable, but only to user space.
> 
> If you get an error in the atomic we are polling, its a fatal error since
> SW can't recover and we shutdown.

Aha, I think you mean that: the MCE is fatal because if it happens on
any CPU, it will be in kernel mode.

> Overthinking :-).. If there is concensus, if Boris feels comfortable
> enough, i would drop this patch.

This is what we're doing right now - thinking about the consensus. And
Boris will feel comfortable once we've arrived at a sensible decision.

:-)

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette