linux-kernel - Re: Fix built-in early-load Intel microcode alignment

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20200116191632.GB117689@otc-nc-03>
Date:   Thu, 16 Jan 2020 11:16:32 -0800
From:   "Raj, Ashok" <ashok.raj@...el.com>
To:     Jari Ruusu <jari.ruusu@...il.com>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Andy Lutomirski <luto@...capital.net>,
        Luis Chamberlain <mcgrof@...nel.org>,
        Borislav Petkov <bp@...en8.de>,
        Fenghua Yu <fenghua.yu@...el.com>, johannes.berg@...el.com,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        stable <stable@...r.kernel.org>,
        Hans de Goede <hdegoede@...hat.com>,
        Andy Lutomirski <luto@...nel.org>,
        Ashok Raj <ashok.raj@...el.com>
Subject: Re: Fix built-in early-load Intel microcode alignment

Hi Jari

Sorry for the delay, just returned after a long vacation. 

On Thu, Jan 16, 2020 at 08:55:52AM +0200, Jari Ruusu wrote:
> On 1/15/20, Linus Torvalds <torvalds@...ux-foundation.org> wrote:
> > However, the most likely cause is that you have a borderline dodgy
> > system, and the microcode update then just triggers a pre-existing
> > problem.
> 
> For that particular processor model, there appears to be microcode
> updates for four steppings: 9 10 11 and 12. My model is stepping
> 9, so it appears to be early commercially sold version of that
> model. Probably more problems on it than on later steppings.

I don't suspect the alignment issue during microcode load to trigger
any hangs after couple days. Its only used to copy to some internal store.
So once the ucode is loaded and if you look at /proc/cpuinfo and it reflects
the latest microcode you are past the alignment issue. I suspect its
something else.

Can you please also document the OEM, BIOS versions, and also both the
old and new microcode versions after the update.

Would suggest logging the hang issue in public github for microcode issues.
This would let the product folks look at them.

https://github.com/intel/Intel-Linux-Processor-Microcode-Data-Files

> 
> > But it might be worth it if the intel people could check up with their
> > microcode people on this anyway - if there is _one_ report of "my
> > system locks up with newer ucode", that's one thing. But if Jari isn't
> > alone...
> 

Cheers,
Ashok