linux-kernel - Re: [PATCH] generic arch_futex_atomic_op

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: <alpine.DEB.2.21.1907151513450.1722@nanos.tec.linutronix.de>
Date:   Mon, 15 Jul 2019 15:15:10 +0200 (CEST)
From:   Thomas Gleixner <tglx@...utronix.de>
To:     Arnd Bergmann <arnd@...db.de>
cc:     Vasily Averin <vvs@...tuozzo.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        linux-arch <linux-arch@...r.kernel.org>,
        Ingo Molnar <mingo@...hat.com>
Subject: Re: [PATCH] generic arch_futex_atomic_op_inuser() cleanup

On Mon, 15 Jul 2019, Arnd Bergmann wrote:

> On Mon, Jul 15, 2019 at 12:29 PM Vasily Averin <vvs@...tuozzo.com> wrote:
> >
> > Looks like this code is dead and therefore looks strange.
> > I've found it during manual code review and decided to send patch
> > to pay your attention to this problem.
> > Probably it's better to remove this code at all?
> >
> > On 7/15/19 1:27 PM, Vasily Averin wrote:
> > > Access to 'op' variable does not require pagefault_disable(),
> > > 'ret' variable should be initialized before using,
> > > 'oldval' variable can be replaced by constant.
> > >
> > > Signed-off-by: Vasily Averin <vvs@...tuozzo.com>
> 
> I'm not following the reasoning for any of the changes here. Why do you
> think we don't need the pagefault_disable() around get_user()/put_user(),
> and which part of the funtion is dead code?

All of it. If you change the function to

{
	return -ENOSYS;
}

then it is equivalent (except for the pointless pagefault_disable/enable()
pair which protects absolutely nothing).

Thanks,

	tglx