linux-kernel - Re: [PATCH v2 1/6] Fix "x86/alternatives: Lockdep-enforce text_mutex in text

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: <E8934715-B8B1-4E37-A37F-18D040885C4F@vmware.com>
Date:   Thu, 6 Sep 2018 20:57:38 +0000
From:   Nadav Amit <namit@...are.com>
To:     Peter Zijlstra <peterz@...radead.org>
CC:     Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...hat.com>, X86 ML <x86@...nel.org>,
        Arnd Bergmann <arnd@...db.de>,
        linux-arch <linux-arch@...r.kernel.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Jiri Kosina <jkosina@...e.cz>,
        Andy Lutomirski <luto@...nel.org>,
        Kees Cook <keescook@...omium.org>,
        Dave Hansen <dave.hansen@...el.com>
Subject: Re: [PATCH v2 1/6] Fix "x86/alternatives: Lockdep-enforce text_mutex
 in text_poke*()"

at 1:25 PM, Peter Zijlstra <peterz@...radead.org> wrote:

> On Thu, Sep 06, 2018 at 07:58:40PM +0000, Nadav Amit wrote:
>>> With that CR3 trickery, we can rid ourselves of the text_mutex
>>> requirement, since concurrent text_poke is 'safe'. That would clean up
>>> the kgdb code quite a bit.
>> 
>> I don’t know. I’m somewhat worried with multiple mechanisms potentially
>> changing the same code at the same time - and maybe ending up with some
>> mess.
> 
> kgdb only pokes INT3, that should be pretty safe.

Maybe I misunderstand your point. If you want me to get rid of text_mutex
completely, I am afraid it will be able to cause mess by changing the same
piece of code through kprobes and the static-keys mechanism.

I doubt it would work today without failing, but getting rid of text_mutex
is likely to make it even worse.