linux-kernel - RE: [PATCH V2 4/6] x86/efi: Add efi page fault handler to fixup/recover from page faults caused by firmware

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] [day] [month] [year] [list]

Date:   Mon, 3 Sep 2018 17:16:01 +0000
From:   "Prakhya, Sai Praneeth" <sai.praneeth.prakhya@...el.com>
To:     Thomas Gleixner <tglx@...utronix.de>
CC:     "linux-efi@...r.kernel.org" <linux-efi@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "Neri, Ricardo" <ricardo.neri@...el.com>,
        "matt@...eblueprint.co.uk" <matt@...eblueprint.co.uk>,
        Lee Chun-Yi <jlee@...e.com>, Al Stone <astone@...hat.com>,
        Borislav Petkov <bp@...en8.de>, Ingo Molnar <mingo@...nel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Bhupesh Sharma <bhsharma@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Ard Biesheuvel <ard.biesheuvel@...aro.org>
Subject: RE: [PATCH V2 4/6] x86/efi: Add efi page fault handler to
 fixup/recover from page faults caused by firmware

> > The efi specific page fault handler offers us two advantages:
> > 1. Avoid panics/hangs caused by buggy firmware.
> > 2. Shout loud that the firmware is buggy and hence is not a kernel bug.
> >
> > Finally, this new mapping will not impact a reboot from kexec, as
> > kexec is only concerned about runtime memory regions.
> 
> No. This is just a horrible hack to make completely bogus firmware work and
> never fixed.
> 

Yes, that's true.

> The proper thing to do is to have a minimal page fault handler which does:
> 
>  1) Yell loudly if that ever happens
> 
>  2) Handles the reboot request gracefully
> 
>  3) Freeze and disable the EFI mess for all other cases
> 
> That does not require any hackery to make these mappings work from atomic
> context and keeps the mess confined to the EFI code where it belongs.
> 
> Ideally we just blacklist the offending system and be done with it.

This makes sense to me. I will implement the above suggested and as said should 
avoid the need for making mappings work from atomic context.

Regards,
Sai