From: Kirill Tkhai <ktkhai@virtuozzo.com>
To: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: akpm@linux-foundation.org, dan.j.williams@intel.com,
mhocko@suse.com, keith.busch@intel.com,
kirill.shutemov@linux.intel.com,
alexander.h.duyck@linux.intel.com, ira.weiny@intel.com,
andreyknvl@google.com, arunks@codeaurora.org, vbabka@suse.cz,
cl@linux.com, riel@surriel.com, keescook@chromium.org,
hannes@cmpxchg.org, npiggin@gmail.com,
mathieu.desnoyers@efficios.com, shakeelb@google.com, guro@fb.com,
aarcange@redhat.com, hughd@google.com, jglisse@redhat.com,
mgorman@techsingularity.net, daniel.m.jordan@oracle.com,
jannh@google.com, kilobyte@angband.pl, linux-api@vger.kernel.org,
linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: Re: [PATCH v2 0/7] mm: process_vm_mmap() -- syscall for duplication a process mapping
Date: Tue, 28 May 2019 12:15:16 +0300 [thread overview]
Message-ID: <de6e4e89-66ac-da2f-48a6-4d98a728687a@virtuozzo.com> (raw)
In-Reply-To: <20190527233030.hpnnbi4aqnu34ova@box>
On 28.05.2019 02:30, Kirill A. Shutemov wrote:
> On Fri, May 24, 2019 at 05:00:32PM +0300, Kirill Tkhai wrote:
>> On 24.05.2019 14:52, Kirill A. Shutemov wrote:
>>> On Fri, May 24, 2019 at 01:45:50PM +0300, Kirill Tkhai wrote:
>>>> On 22.05.2019 18:22, Kirill A. Shutemov wrote:
>>>>> On Mon, May 20, 2019 at 05:00:01PM +0300, Kirill Tkhai wrote:
>>>>>> This patchset adds a new syscall, which makes possible
>>>>>> to clone a VMA from a process to current process.
>>>>>> The syscall supplements the functionality provided
>>>>>> by process_vm_writev() and process_vm_readv() syscalls,
>>>>>> and it may be useful in many situation.
>>>>>
>>>>> Kirill, could you explain how the change affects rmap and how it is safe.
>>>>>
>>>>> My concern is that the patchset allows to map the same page multiple times
>>>>> within one process or even map page allocated by child to the parrent.
>>>>>
>>>>> It was not allowed before.
>>>>>
>>>>> In the best case it makes reasoning about rmap substantially more difficult.
>>>>>
>>>>> But I'm worry it will introduce hard-to-debug bugs, like described in
>>>>> https://lwn.net/Articles/383162/.
>>>>
>>>> Andy suggested to unmap PTEs from source page table, and this make the single
>>>> page never be mapped in the same process twice. This is OK for my use case,
>>>> and here we will just do a small step "allow to inherit VMA by a child process",
>>>> which we didn't have before this. If someone still needs to continue the work
>>>> to allow the same page be mapped twice in a single process in the future, this
>>>> person will have a supported basis we do in this small step. I believe, someone
>>>> like debugger may want to have this to make a fast snapshot of a process private
>>>> memory (when the task is stopped for a small time to get its memory). But for
>>>> me remapping is enough at the moment.
>>>>
>>>> What do you think about this?
>>>
>>> I don't think that unmapping alone will do. Consider the following
>>> scenario:
>>>
>>> 1. Task A creates and populates the mapping.
>>> 2. Task A forks. We have now Task B mapping the same pages, but
>>> write-protected.
>>> 3. Task B calls process_vm_mmap() and passes the mapping to the parent.
>>>
>>> After this Task A will have the same anon pages mapped twice.
>>
>> Ah, sure.
>>
>>> One possible way out would be to force CoW on all pages in the mapping,
>>> before passing the mapping to the new process.
>>
>> This will pop all swapped pages up, which is the thing the patchset aims
>> to prevent.
>>
>> Hm, what about allow remapping only VMA, which anon_vma::rb_root contain
>> only chain and which vma->anon_vma_chain contains single entry? This is
>> a vma, which were faulted, but its mm never were duplicated (or which
>> forks already died).
>
> The requirement for the VMA to be faulted (have any pages mapped) looks
> excessive to me, but the general idea may work.
>
> One issue I see is that userspace may not have full control to create such
> VMA. vma_merge() can merge the VMA to the next one without any consent
> from userspace and you'll get anon_vma inherited from the VMA you've
> justed merged with.
>
> I don't have any valid idea on how to get around this.
Technically it is possible by creating boundary 1-page VMAs with another protection:
one above and one below the desired region, then map the desired mapping. But this
is not comfortable.
I don't think it's difficult to find a natural limitation, which prevents mapping
a single page twice if we want to avoid this at least on start. Another suggestion:
prohibit to map a remote process's VMA only in case of its vm_area_struct::anon_vma::root
is the same as root of one of local process's VMA.
What about this?
Thanks,
Kirill
next prev parent reply other threads:[~2019-05-28 9:15 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-05-20 14:00 [PATCH v2 0/7] mm: process_vm_mmap() -- syscall for duplication a process mapping Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 1/7] mm: Add process_vm_mmap() syscall declaration Kirill Tkhai
2019-05-21 0:28 ` Ira Weiny
2019-05-21 8:29 ` Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 2/7] mm: Extend copy_vma() Kirill Tkhai
2019-05-21 8:18 ` Kirill A. Shutemov
2019-05-21 8:48 ` Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 3/7] mm: Extend copy_page_range() Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 4/7] mm: Export round_hint_to_min() Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 5/7] mm: Introduce may_mmap_overlapped_region() helper Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 6/7] mm: Introduce find_vma_filter_flags() helper Kirill Tkhai
2019-05-20 14:00 ` [PATCH v2 7/7] mm: Add process_vm_mmap() Kirill Tkhai
2019-05-21 14:43 ` [PATCH v2 0/7] mm: process_vm_mmap() -- syscall for duplication a process mapping Andy Lutomirski
2019-05-21 15:52 ` Kirill Tkhai
2019-05-21 15:59 ` Kirill Tkhai
2019-05-21 16:20 ` Jann Horn
2019-05-21 17:03 ` Kirill Tkhai
2019-05-21 17:28 ` Jann Horn
2019-05-22 10:03 ` Kirill Tkhai
2019-05-21 16:43 ` Andy Lutomirski
2019-05-21 17:44 ` Kirill Tkhai
2019-05-23 16:19 ` Andy Lutomirski
2019-05-24 10:36 ` Kirill Tkhai
2019-05-22 15:22 ` Kirill A. Shutemov
2019-05-23 16:11 ` Kirill Tkhai
2019-05-24 10:45 ` Kirill Tkhai
2019-05-24 11:52 ` Kirill A. Shutemov
2019-05-24 14:00 ` Kirill Tkhai
2019-05-27 23:30 ` Kirill A. Shutemov
2019-05-28 9:15 ` Kirill Tkhai [this message]
2019-05-28 16:15 ` Kirill A. Shutemov
2019-05-29 14:33 ` Kirill Tkhai
2019-06-03 14:38 ` Kirill Tkhai
2019-06-03 14:56 ` Kirill Tkhai
2019-06-03 17:47 ` Kirill A. Shutemov
2019-06-04 9:32 ` Kirill Tkhai
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=de6e4e89-66ac-da2f-48a6-4d98a728687a@virtuozzo.com \
--to=ktkhai@virtuozzo.com \
--cc=aarcange@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=alexander.h.duyck@linux.intel.com \
--cc=andreyknvl@google.com \
--cc=arunks@codeaurora.org \
--cc=cl@linux.com \
--cc=dan.j.williams@intel.com \
--cc=daniel.m.jordan@oracle.com \
--cc=guro@fb.com \
--cc=hannes@cmpxchg.org \
--cc=hughd@google.com \
--cc=ira.weiny@intel.com \
--cc=jannh@google.com \
--cc=jglisse@redhat.com \
--cc=keescook@chromium.org \
--cc=keith.busch@intel.com \
--cc=kilobyte@angband.pl \
--cc=kirill.shutemov@linux.intel.com \
--cc=kirill@shutemov.name \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=mgorman@techsingularity.net \
--cc=mhocko@suse.com \
--cc=npiggin@gmail.com \
--cc=riel@surriel.com \
--cc=shakeelb@google.com \
--cc=vbabka@suse.cz \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).