linux-kernel - Re: [PATCH v2] x86/mce: Add Skylake quirk for patrol scrub reported errors

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20200927221917.GB4746@zn.tnic>
Date:   Mon, 28 Sep 2020 00:19:17 +0200
From:   Borislav Petkov <bp@...en8.de>
To:     "Luck, Tony" <tony.luck@...el.com>
Cc:     "Song, Youquan" <youquan.song@...el.com>,
        "x86@...nel.org" <x86@...nel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] x86/mce: Add Skylake quirk for patrol scrub reported
 errors

On Fri, Sep 25, 2020 at 04:06:20PM -0700, Luck, Tony wrote:
> In some ways that's pretty neat. But it would still be ugly if we need
> to extend it further for other issues. Especially if they don't have such
> a simple rule to adjust the severity.

I'm still secretly hoping that one fine day one of your minions will
come with a conversion series around the corner; series, which gets rid
of this unreadable mess and we get something *actually* readable like
mce_severity_amd().

And on that day I'll even take one patch per severities[] member so that
it is absolutely clear how the conversion has happened and review can be
good and catch any lurking errors.

One day... :-)

Because after that day, we can do arbitrary rules to adjust the severity
and there won't be any problems with extending it anymore because then
it'll be nice and flexible C code only.

> This would be better as a bit mask. I don't think we need this same
> hack on the next generation of CPUs ... but if we did, the bank numbers
> that would be affected don't form a continuous sequence.

Ok, feel free to adjust it how you think it is better.

> I'd need to stare at the placement of this in the sequence of rules at some
> non-Friday-afternoon time. It might be right, but as we've grumbled together
> many times before that code is full of surprise side effects.

Nah, I just put it there so that I can see the macros in the same window
and don't have to scroll - I'm purely relying on you here to place it in
the right spot.

Thx.

-- 
Regards/Gruss,
    Boris.

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