linux-kernel - Re: [PATCH v3 8/8] x86, mm, gup: prevent get_page() race with munmap in paravirt guest

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: <1aacc7ac-87b0-e22e-a265-ea175506844d@suse.cz>
Date:   Mon, 16 Dec 2019 16:11:14 +0100
From:   Vlastimil Babka <vbabka@...e.cz>
To:     Peter Zijlstra <peterz@...radead.org>,
        Vitaly Kuznetsov <vkuznets@...hat.com>
Cc:     Ajay Kaher <akaher@...are.com>, gregkh@...uxfoundation.org,
        stable@...r.kernel.org, torvalds@...ux-foundation.org,
        punit.agrawal@....com, akpm@...ux-foundation.org,
        kirill.shutemov@...ux.intel.com, willy@...radead.org,
        will.deacon@....com, mszeredi@...hat.com, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, srivatsab@...are.com,
        srivatsa@...il.mit.edu, amakhalov@...are.com, srinidhir@...are.com,
        bvikas@...are.com, anishs@...are.com, vsirnapalli@...are.com,
        srostedt@...are.com, Oscar Salvador <osalvador@...e.de>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        Juergen Gross <jgross@...e.com>,
        Borislav Petkov <bp@...en8.de>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.org>
Subject: Re: [PATCH v3 8/8] x86, mm, gup: prevent get_page() race with munmap
 in paravirt guest

On 12/16/19 2:47 PM, Peter Zijlstra wrote:
> On Mon, Dec 16, 2019 at 02:30:44PM +0100, Vitaly Kuznetsov wrote:
>> Peter Zijlstra <peterz@...radead.org> writes:
>>
>>> On Tue, Dec 17, 2019 at 02:15:48AM +0530, Ajay Kaher wrote:
>>>> From: Vlastimil Babka <vbabka@...e.cz>
>>>>
>>>> The x86 version of get_user_pages_fast() relies on disabled interrupts to
>>>> synchronize gup_pte_range() between gup_get_pte(ptep); and get_page() against
>>>> a parallel munmap. The munmap side nulls the pte, then flushes TLBs, then
>>>> releases the page. As TLB flush is done synchronously via IPI disabling
>>>> interrupts blocks the page release, and get_page(), which assumes existing
>>>> reference on page, is thus safe.
>>>> However when TLB flush is done by a hypercall, e.g. in a Xen PV guest, there is
>>>> no blocking thanks to disabled interrupts, and get_page() can succeed on a page
>>>> that was already freed or even reused.
>>>>
>>>> We have recently seen this happen with our 4.4 and 4.12 based kernels, with
>>>> userspace (java) that exits a thread, where mm_release() performs a futex_wake()
>>>> on tsk->clear_child_tid, and another thread in parallel unmaps the page where
>>>> tsk->clear_child_tid points to. The spurious get_page() succeeds, but futex code
>>>> immediately releases the page again, while it's already on a freelist. Symptoms
>>>> include a bad page state warning, general protection faults acessing a poisoned
>>>> list prev/next pointer in the freelist, or free page pcplists of two cpus joined
>>>> together in a single list. Oscar has also reproduced this scenario, with a
>>>> patch inserting delays before the get_page() to make the race window larger.
>>>>
>>>> Fix this by removing the dependency on TLB flush interrupts the same way as the
>>>
>>> This is suppsed to be fixed by:
>>>
>>> arch/x86/Kconfig:       select HAVE_RCU_TABLE_FREE              if PARAVIRT
>>>
>>
>> Yes,

Well, that commit fixes the "page table can be freed under us" part. But
this patch is about the "get_page() will succeed on a page that's being
freed" part. Upstream fixed that unknowingly in 4.13 by a gup.c
refactoring that would be too risky to backport fully.

>> but HAVE_RCU_TABLE_FREE was enabled on x86 only in 4.14:
>>
>> commit 9e52fc2b50de3a1c08b44f94c610fbe998c0031a
>> Author: Vitaly Kuznetsov <vkuznets@...hat.com>
>> Date:   Mon Aug 28 10:22:51 2017 +0200
>>
>>     x86/mm: Enable RCU based page table freeing (CONFIG_HAVE_RCU_TABLE_FREE=y)
>>
>> and, if I understood correctly, Ajay is suggesting the patch for older
>> stable kernels (4.9 and 4.4 I would guess).
> 
> It wasn't at all clear this was targeted at old kernels (I only got this
> one patch).
> 
> And why can't those necro kernels do backports of the upstream solution?
>