linux-kernel - Re: [PATCH -next for tip:x86/pti] x86/tlb: drop unneeded local vars in enable_l1d_flush_for

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

Message-ID: <20201001081958.GO2628@hirez.programming.kicks-ass.net>
Date:   Thu, 1 Oct 2020 10:19:58 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Singh, Balbir" <sblbir@...zon.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Lukas Bulwahn <lukas.bulwahn@...il.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.org>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "x86@...nel.org" <x86@...nel.org>,
        "H. Peter Anvin" <hpa@...or.com>,
        Nathan Chancellor <natechancellor@...il.com>,
        Nick Desaulniers <ndesaulniers@...gle.com>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "clang-built-linux@...glegroups.com" 
        <clang-built-linux@...glegroups.com>,
        "kernel-janitors@...r.kernel.org" <kernel-janitors@...r.kernel.org>,
        "linux-safety@...ts.elisa.tech" <linux-safety@...ts.elisa.tech>
Subject: Re: [PATCH -next for tip:x86/pti] x86/tlb: drop unneeded local vars
 in enable_l1d_flush_for_task()


Your MUA is having trouble wrapping text at 78 chars.

On Thu, Oct 01, 2020 at 09:49:30AM +1000, Singh, Balbir wrote:

> So this is the change I am playing with, I don't like the idea of
> killing the task, but it's better than silently not flushing, I guess
> system administrators will learn with time not to correctly the
> affinity of tasks flushing L1D. For the affinity bits, not being able
> to change the affinity is better, but not being able to provide
> feedback on as to why is a bit weird as well, but I wonder if there
> are other cases where we might want to lock the affinity of a task for
> it's lifetime.

You can't really do that, hot-unplug can (and will) destroy any affinity
setting, and if the task/admin wants to recover it needs to be able to
re-set affinity after that.