From:   ebiederm@xmission.com (Eric W. Biederman)
To:     Linus Torvalds <torvalds@linux-foundation.org>
Cc:     Brian Gerst <brgerst@gmail.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Oleg Nesterov <oleg@redhat.com>,
        Al Viro <viro@zeniv.linux.org.uk>,
        Kees Cook <keescook@chromium.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        "the arch\/x86 maintainers" <x86@kernel.org>,
        H Peter Anvin <hpa@zytor.com>,
        Andy Lutomirski <luto@kernel.org>
References: <87y26nmwkb.fsf@disp2133>
        <20211020174406.17889-9-ebiederm@xmission.com>
        <874k8htmb2.fsf@email.froward.int.ebiederm.org>
        <CAMzpN2jkK5sAv-Kg_kVnCEyVySiqeTdUORcC=AdG1gV6r8nUew@mail.gmail.com>
        <87ilwxrvu9.fsf@email.froward.int.ebiederm.org>
        <CAHk-=widK1vko2EN9PtV3jTo02u-expXxozui-fsK-0uKrcGHg@mail.gmail.com>
        <87ee7lqe6k.fsf@email.froward.int.ebiederm.org>
        <CAHk-=whf-nR4xDSEDtXeowFn4QQKY3s_MXoM8Ha-dPiOzFzUxQ@mail.gmail.com>
        <CAHk-=wjU1Nf+zzFgwE5bAGSq2W9DifmKDsAOFuZcaaHq3vEQ8w@mail.gmail.com>
Date:   Fri, 12 Nov 2021 15:43:23 -0600
In-Reply-To: <CAHk-=wjU1Nf+zzFgwE5bAGSq2W9DifmKDsAOFuZcaaHq3vEQ8w@mail.gmail.com>
        (Linus Torvalds's message of "Fri, 12 Nov 2021 13:24:37 -0800")
Message-ID: <87wnldnj7o.fsf@email.froward.int.ebiederm.org>
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/26.1 (gnu/linux)
MIME-Version: 1.0
Content-Type: text/plain
Subject: Re: [PATCH 09/20] signal/vm86_32: Replace open coded BUG_ON with an actual BUG_ON
Precedence: bulk

Linus Torvalds <torvalds@linux-foundation.org> writes:

> On Fri, Nov 12, 2021 at 1:23 PM Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
>>
>> But I could see some quick test hack doing it - the IVT at boot is
>> actually not at zero, but at fffxxxxx. 8086 is magic.
>
> .. and it's been too long, and I'm too lazy to check - it may be that
> vm86 mode doesn't even do that magic boot-time address thing.
>
> It's not like we really care about vm86 mode any more, since pretty
> much nobody users it.

As I recall at boot CS == 0xffff0000 EIP == 0x0000fff0 and the cpu is in
16bit mode.  Which means the cpu runs the instructions in the last
16bytes of memory at boot up.  Which is just enough for a jump somewhere
else.  Such as 64K backwards where there is enough space to actual have
enough code to do something.

I don't think vm86 even attempts to emulate that behavior as it is only
concerned about 16bit only cpus and emulation.

In the nobody cares camp I have just sent you a pull request to remove
the ancient (except it wasn't a BUG_ON) and problematic test in the
BUG_ON.

I think that is enough to resolve this.

Eric