linux-kernel - Re: [PATCH] x86/fpu: Disable BH while while loading FPU registers in __fpu__restore

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: <20181119151948.g27vuuavpgmyqynn@linutronix.de>
Date:   Mon, 19 Nov 2018 16:19:48 +0100
From:   Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To:     Dave Hansen <dave.hansen@...el.com>
Cc:     Borislav Petkov <bp@...en8.de>, x86@...nel.org,
        Ingo Molnar <mingo@...nel.org>, linux-kernel@...r.kernel.org,
        Andy Lutomirski <luto@...nel.org>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Radim Krčmář <rkrcmar@...hat.com>,
        kvm@...r.kernel.org, "Jason A. Donenfeld" <Jason@...c4.com>,
        Rik van Riel <riel@...riel.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>
Subject: Re: [PATCH] x86/fpu: Disable BH while while loading FPU registers in
 __fpu__restore_sig()

On 2018-11-19 07:08:44 [-0800], Dave Hansen wrote:
> On 11/19/18 7:06 AM, Sebastian Andrzej Siewior wrote:
> > On 2018-11-19 07:04:35 [-0800], Dave Hansen wrote:
> >> Does the local_bh_disable() itself survive?
> > Not in __fpu__restore_sig(). I do have:
> > | static inline void __fpregs_changes_begin(void)
> > | {
> > |        preempt_disable();
> > |        local_bh_disable();
> > | }
> > 
> > and __fpregs_changes_begin() is introduced as part of the series.
> 
> OK, so can we just comment *that*, please?  Basically, why do we need botj?

let me do this then.
local_bh_disable() should be enough. However I had a discussion with
PeterZ that this (local_bh_disable()) also acting as preempt_disable())
is an implementation detail and should be avoided. It is not true
Preempt-RT for instance.

Sebastian