linux-kernel - Re: [RFC PATCH 0/3] Add hugetlb MADV

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: <2ddd0a26-33fd-9cde-3501-f0584bbffefc@redhat.com>
Date:   Fri, 28 Jan 2022 10:55:00 +0100
From:   David Hildenbrand <david@...hat.com>
To:     Mike Kravetz <mike.kravetz@...cle.com>,
        linux-kernel@...r.kernel.org, linux-mm@...ck.org
Cc:     Michal Hocko <mhocko@...e.com>,
        Naoya Horiguchi <naoya.horiguchi@...ux.dev>,
        Axel Rasmussen <axelrasmussen@...gle.com>,
        Peter Xu <peterx@...hat.com>,
        Andrea Arcangeli <aarcange@...hat.com>,
        Mina Almasry <almasrymina@...gle.com>,
        Shuah Khan <shuah@...nel.org>,
        Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [RFC PATCH 0/3] Add hugetlb MADV_DONTNEED support

On 27.01.22 18:55, Mike Kravetz wrote:
> On 1/27/22 03:57, David Hildenbrand wrote:
>> On 13.01.22 19:03, Mike Kravetz wrote:
>>> Userfaultfd selftests for hugetlb does not perform UFFD_EVENT_REMAP
>>> testing.  However, mremap support was recently added in commit
>>> 550a7d60bd5e ("mm, hugepages: add mremap() support for hugepage backed
>>> vma").  While attempting to enable mremap support in the test, it was
>>> discovered that the mremap test indirectly depends on MADV_DONTNEED.
>>>
>>> hugetlb does not support MADV_DONTNEED.  However, the only thing
>>> preventing support is a check in can_madv_lru_vma().  Simply removing
>>> the check will enable support.
>>>
>>> This is sent as a RFC because there is no existing use case calling
>>> for hugetlb MADV_DONTNEED support except possibly the userfaultfd test.
>>> However, adding support makes sense as it is fairly trivial and brings
>>> hugetlb functionality more in line with 'normal' memory.
>>>
>>
>> Just a note:
>>
>> QEMU doesn't use huge anonymous memory directly (MAP_ANON | MAP_HUGE...)
>> but instead always goes either via hugetlbfs or via memfd. 
>>
>> For MAP_PRIVATE hugetlb mappings, fallocate(FALLOC_FL_PUNCH_HOLE) seems
>> to get the job done (IOW: also discards private anon pages). See the
>> comments in the QEMU code below. I remember that that is somewhat
>> inconsistent. For ordinary MAP_PRIVATE mapped files I remember that we
>> always need fallocate(FALLOC_FL_PUNCH_HOLE) + madvise(QEMU_MADV_DONTNEED)
>> to make sure
>>
>> a) All file pages are removed
>> b) All private anon pages are removed
>>
>> IIRC hugetlbfs really is different in that regard, but maybe other fs
>> behave similarly.
> 
> Yes it is really different.  And, some might even consider that a bug?
> Imagine if those private anon (COW) pages contain important data.  They
> could be unmapped/freed by some other process that has write access to
> the hugetlb file on which the private mapping is based.

Right, that's also what I once worried about in QEMU code. But then I
realized that any kind of fallocate(FALLOC_FL_PUNCH_HOLE) on a file
shared by multiple parties mapped MAP_PRIVATE might be bogus already.

Assume you have a VM running with MAP_SHARED on a file. The file
contains the VM memory state. Assume you pause the VM and want to
convert it into 2 instances that will continue running independently
based on the captured file state.

You'd have to MAP_PRIVATE the file such that both VMs start with the
original state and only see their modifications.

... but if one process decides to fallocate(FALLOC_FL_PUNCH_HOLE), for
example, due to memory ballooning, even a page that's still shared by
both processes (!COW), you'd corrupt the other VM.

So my assumption is that MAP_PRIVATE in combination with
fallocate(FALLOC_FL_PUNCH_HOLE) on a file mapped by more than one
process is just bogus already.

> 
> I believe this same issue exists for hugetlbfs ftruncate.  When fallocate
> hole punch support was added, it was based on the ftruncate functionality.
> 
> I am hesitant to change the behavior of hugetlb hole punch or truncate
> as people may be relying on that behavior today.  Your QEMU example is
> one such case.

Yes, I assume we're stuck with that.

-- 
Thanks,

David / dhildenb