MIME-Version: 1.0
In-Reply-To: <CALCETrUJWp-99dbU0Yq08jCkf7N+N+hD_5FUVdKc+aY6fZhLmA@mail.gmail.com>
References: <20170321045713.GE23490@yexl-desktop> <CAJcbSZGE9P3p6GV=7QHeDUesJZuR4mE=9tkMK+6SRbOoc-2kAg@mail.gmail.com>
 <CA+55aFyz-snHPk=GkE7mHzd3r8nzWy4DnFEnNGLrn_ck4rPKwA@mail.gmail.com>
 <CAJcbSZHFYz4wpBftvS40Gv9X5Epuk43wLStv9-7axWd3965fcw@mail.gmail.com>
 <CAJcbSZFuJiJE0=rS-ZGPLCMUgaz+dPzHUm-iMa=J1E4H0qAHFg@mail.gmail.com>
 <CA+55aFzS=X5J9QXbF_oXdO3KTrGjzA6BpGfdLo8BiuuU81cOPw@mail.gmail.com> <CALCETrUJWp-99dbU0Yq08jCkf7N+N+hD_5FUVdKc+aY6fZhLmA@mail.gmail.com>
From: Andy Lutomirski <luto@kernel.org>
Date: Tue, 21 Mar 2017 16:51:21 -0700
Message-ID: <CALCETrUsJPdW7cOs7-iKmeLZufZ8hGRpyC99-H0dy0Nw-pCCSw@mail.gmail.com>
Subject: Re: [lkp-robot] [x86] 69218e4799: BUG:kernel_hang_in_boot_stage
To: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Thomas Garnier <thgarnie@google.com>,
        kernel test robot <xiaolong.ye@intel.com>,
        Ingo Molnar <mingo@kernel.org>,
        Alexander Potapenko <glider@google.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        Andrey Ryabinin <aryabinin@virtuozzo.com>,
        Andy Lutomirski <luto@kernel.org>,
        Ard Biesheuvel <ard.biesheuvel@linaro.org>,
        Boris Ostrovsky <boris.ostrovsky@oracle.com>,
        Borislav Petkov <bp@suse.de>, Chris Wilson <chris@chris-wilson.co.uk>,
        Christian Borntraeger <borntraeger@de.ibm.com>,
        Dmitry Vyukov <dvyukov@google.com>,
        Frederic Weisbecker <fweisbec@gmail.com>,
        Jiri Kosina <jikos@kernel.org>, Joerg Roedel <joro@8bytes.org>,
        Jonathan Corbet <corbet@lwn.net>, Josh Poimboeuf <jpoimboe@redhat.com>,
        Juergen Gross <jgross@suse.com>, Kees Cook <keescook@chromium.org>,
        Len Brown <len.brown@intel.com>, Lorenzo Stoakes <lstoakes@gmail.com>,
        "Luis R . Rodriguez" <mcgrof@kernel.org>,
        Matt Fleming <matt@codeblueprint.co.uk>,
        Michal Hocko <mhocko@suse.com>, Paolo Bonzini <pbonzini@redhat.com>,
        Paul Gortmaker <paul.gortmaker@windriver.com>,
        Pavel Machek <pavel@ucw.cz>, Peter Zijlstra <peterz@infradead.org>,
        =?UTF-8?B?UmFkaW0gS3LEjW3DocWZ?= <rkrcmar@redhat.com>,
        "Rafael J . Wysocki" <rjw@rjwysocki.net>,
        Rusty Russell <rusty@rustcorp.com.au>,
        Stanislaw Gruszka <sgruszka@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Tim Chen <tim.c.chen@linux.intel.com>,
        Vitaly Kuznetsov <vkuznets@redhat.com>, zijun_hu <zijun_hu@htc.com>,
        LKML <linux-kernel@vger.kernel.org>,
        Stephen Rothwell <sfr@canb.auug.org.au>, LKP <lkp@01.org>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1661
Lines: 42

On Tue, Mar 21, 2017 at 3:32 PM, Andy Lutomirski <luto@amacapital.net> wrote:
> On Tue, Mar 21, 2017 at 2:11 PM, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
>> On Tue, Mar 21, 2017 at 1:25 PM, Thomas Garnier <thgarnie@google.com> wrote:
>>> The issue seems to be related to exceptions happening in close pages
>>> to the fixmap GDT remapping.
>>>
>>> The original page fault happen in do_test_wp_bit which set a fixmap
>>> entry to test WP flag. If I grow the number of processors supported
>>> increasing the distance between the remapped GDT page and the WP test
>>> page, the error does not reproduce.
>>>
>>> I am still looking at the exact distance between repro and no-repro as
>>> well as the exact root cause.
>>
>> Hmm. Have we set the GDT limit incorrectly, somehow? The GDT *can*
>> cover 8k entries, which at 8 bytes each would be 64kB.
>
> The QEMU barf says the GDT limit is 0xff, for better or for worse.
>
>>
>> So somebody trying to load an invalid segment (say, 0xffff) might end
>> up causing an access to the GDT base + 64k - 8.
>>
>> It is also possible that the CPU might do a page table writability
>> check *before* it does the limit check. That would sound odd, though.
>> Might be a CPU errata.
>>
>

> There's presumably something genuinely wrong with our GDT.

This is suspicious.  I added this code in test_wp_bit:

    if (memcmp(get_current_gdt_ro(), get_current_gdt_rw(), 4096) != 0) {
        pr_err("Oh crap\n");
        BUG_ON(1);
    }

It printed "Oh crap" and blew up.  Methinks something's wrong with the
fixmap.  Is it possible that we're crossing a PMD boundary and failing
to translate the addresses right?