linux-kernel - Re: x86/pti: smp_processor_id() called while preemptible in resume-from-sleep

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]

Date:   Sat, 30 Dec 2017 11:15:02 -0800
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Dave Hansen <dave.hansen@...ux.intel.com>
Cc:     Thomas Gleixner <tglx@...utronix.de>,
        Dominik Brodowski <linux@...inikbrodowski.net>,
        Andy Lutomirski <luto@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        "the arch/x86 maintainers" <x86@...nel.org>
Subject: Re: x86/pti: smp_processor_id() called while preemptible in resume-from-sleep

On Sat, Dec 30, 2017 at 11:03 AM, Dave Hansen
<dave.hansen@...ux.intel.com> wrote:
> On 12/30/2017 10:40 AM, Linus Torvalds wrote:
>> The __native_flush_tlb() function looks _very_ broken.
> ...
>> So I'd suggest moving the preempt_disable() up to the top of that
>> function, regardless of whether we could then remove that seemingly
>> stale TLB flush in that crazy
>> smpboot_setup/restore_warm_reset_vector() dance...
>
> If someone is calling __native_flush_tlb(), shouldn't they already be in
> a state where they can't be preempted?  It's fundamentally a one-cpu
> thing, both the actual CPU TLB flush _and_ the per-cpu variables.

Hmm. I think you're right.

> It seems like we might want to _remove_ the explicit
> preempt_dis/enable() from here:
>
>         preempt_disable();
>         native_write_cr3(__native_read_cr3());
>         preempt_enable();
>
> and add some warnings to ensure it's disabled when we enter
> __native_flush_tlb().

Agreed, that would certainly also be consistent.

The current code that disables preemption only selectively seems
insane to me. Either all or nothing, not this crazy half-way thing.

            Linus