linux-kernel - Re: [PATCH] io_thread/x86: don't reset 'cs', 'ss', 'ds' and 'es' registers for io

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: <CAHk-=whZZ-g6qJ1Epox6c-_U-jSs+syA_7a8QKQkxZaudXXR_Q@mail.gmail.com>
Date:   Tue, 25 May 2021 10:23:39 -1000
From:   Linus Torvalds <torvalds@...ux-foundation.org>
To:     Olivier Langlois <olivier@...llion01.com>
Cc:     Jens Axboe <axboe@...nel.dk>, Stefan Metzmacher <metze@...ba.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Andy Lutomirski <luto@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        io-uring <io-uring@...r.kernel.org>,
        "the arch/x86 maintainers" <x86@...nel.org>
Subject: Re: [PATCH] io_thread/x86: don't reset 'cs', 'ss', 'ds' and 'es'
 registers for io_threads

On Tue, May 25, 2021 at 9:40 AM Olivier Langlois <olivier@...llion01.com> wrote:
>
> If I look in the create_io_thread() function, I can see that CLONE_VM
> isn't set...

That would indeed be horribly buggy, but I'm not seeing it:

                .flags          = ((lower_32_bits(flags) | CLONE_VM |
                                    CLONE_UNTRACED) & ~CSIGNAL),

Yeah, it has that odd "first create 'flags' without the CLONE_VM, but
that is only used for that

        lower_32_bits(flags)

thing, and then we explicitly add CLONE_VM in there in the actual
kernel_clone_args.

It's because of how 'kernel_thread()' is written (that has historical
reasons), and the oddity is copied from there.

           Linus