From: David Hildenbrand <david@redhat.com>
To: Qi Zheng <zhengqi.arch@bytedance.com>, Jason Gunthorpe <jgg@nvidia.com>
Cc: akpm@linux-foundation.org, tglx@linutronix.de,
kirill.shutemov@linux.intel.com, mika.penttila@nextfour.com,
linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, songmuchun@bytedance.com,
zhouchengming@bytedance.com
Subject: Re: [PATCH v3 00/15] Free user PTE page table pages
Date: Thu, 11 Nov 2021 12:19:20 +0100 [thread overview]
Message-ID: <27d73395-70b4-fe4a-4c8d-415b43ff9c1f@redhat.com> (raw)
In-Reply-To: <9ee06b52-4844-7996-fa34-34fc7d4fdc10@bytedance.com>
On 11.11.21 12:08, Qi Zheng wrote:
>
>
> On 11/11/21 5:22 PM, David Hildenbrand wrote:
>> On 11.11.21 04:58, Qi Zheng wrote:
>>>
>>>
>>> On 11/11/21 1:37 AM, David Hildenbrand wrote:
>>>>>> It would still be a fairly coarse-grained locking, I am not sure if that
>>>>>> is a step into the right direction. If you want to modify *some* page
>>>>>> table in your process you have exclude each and every page table walker.
>>>>>> Or did I mis-interpret what you were saying?
>>>>>
>>>>> That is one possible design, it favours fast walking and penalizes
>>>>> mutation. We could also stick a lock in the PMD (instead of a
>>>>> refcount) and still logically be using a lock instead of a refcount
>>>>> scheme. Remember modify here is "want to change a table pointer into a
>>>>> leaf pointer" so it isn't an every day activity..
>>>>
>>>> It will be if we somewhat frequent when reclaim an empty PTE page table
>>>> as soon as it turns empty. This not only happens when zapping, but also
>>>> during writeback/swapping. So while writing back / swapping you might be
>>>> left with empty page tables to reclaim.
>>>>
>>>> Of course, this is the current approach. Another approach that doesn't
>>>> require additional refcounts is scanning page tables for empty ones and
>>>> reclaiming them. This scanning can either be triggered manually from
>>>> user space or automatically from the kernel.
>>>
>>> Whether it is introducing a special rwsem or scanning an empty page
>>> table, there are two problems as follows:
>>>
>>> #1. When to trigger the scanning or releasing?
>>
>> For example when reclaiming memory, when scanning page tables in
>> khugepaged, or triggered by user space (note that this is the approach I
>> originally looked into). But it certainly requires more locking thought
>> to avoid stopping essentially any page table walker.
>>
>>> #2. Every time to release a 4K page table page, 512 page table
>>> entries need to be scanned.
>>
>> It would happen only when actually trigger reclaim of page tables
>> (again, someone has to trigger it), so it's barely an issue.
>>
>> For example, khugepaged already scans the page tables either way.
>>
>>>
>>> For #1, if the scanning is triggered manually from user space, the
>>> kernel is relatively passive, and the user does not fully know the best
>>> timing to scan. If the scanning is triggered automatically from the
>>> kernel, that is great. But the timing is not easy to confirm, is it
>>> scanned and reclaimed every time zap or try_to_unmap?
>>>
>>> For #2, refcount has advantages.
>>>
>>>>
>>>>>
>>>>> There is some advantage with this thinking because it harmonizes well
>>>>> with the other stuff that wants to convert tables into leafs, but has
>>>>> to deal with complicated locking.
>>>>>
>>>>> On the other hand, refcounts are a degenerate kind of rwsem and only
>>>>> help with freeing pages. It also puts more atomics in normal fast
>>>>> paths since we are refcounting each PTE, not read locking the PMD.
>>>>>
>>>>> Perhaps the ideal thing would be to stick a rwsem in the PMD. read
>>>>> means a table cannot be come a leaf. I don't know if there is space
>>>>> for another atomic in the PMD level, and we'd have to use a hitching
>>>>> post/hashed waitq scheme too since there surely isn't room for a waitq
>>>>> too..
>>>>>
>>>>> I wouldn't be so quick to say one is better than the other, but at
>>>>> least let's have thought about a locking solution before merging
>>>>> refcounts :)
>>>>
>>>> Yes, absolutely. I can see the beauty in the current approach, because
>>>> it just reclaims "automatically" once possible -- page table empty and
>>>> nobody is walking it. The downside is that it doesn't always make sense
>>>> to reclaim an empty page table immediately once it turns empty.
>>>>
>>>> Also, it adds complexity for something that is only a problem in some
>>>> corner cases -- sparse memory mappings, especially relevant for some
>>>> memory allocators after freeing a lot of memory or running VMs with
>>>> memory ballooning after inflating the balloon. Some of these use cases
>>>> might be good with just triggering page table reclaim manually from user
>>>> space.
>>>>
>>>
>>> Yes, this is indeed a problem. Perhaps some flags can be introduced so
>>> that the release of page table pages can be delayed in some cases.
>>> Similar to the lazyfree mechanism in MADV_FREE?
>>
>> The issue AFAIU is that once your refcount hits 0 (no more references,
>> no more entries), the longer you wait with reclaim, the longer others
>> have to wait for populating a fresh page table because the "page table
>> to be reclaimed" is still stuck around. You'd have to keep the refcount
>> increased for a while, and only drop it after a while. But when? And
>> how? IMHO it's not trivial, but maybe there is an easy way to achieve it.
>>
>
> For running VMs with memory ballooning after inflating the balloon, is
> this a hot behavior? Even if it is, it is already facing the release and
> reallocation of physical pages. The overhead after introducing
> pte_refcount is that we need to release and re-allocate page table page.
> But 2MB physical pages only corresponds to 4KiB of PTE page table page.
> So maybe the overhead is not big.
The cases that come to my mind are
a) Swapping on shared memory with concurrent access
b) Reclaim on file-backed memory with concurrent access
c) Free page reporting as implemented by virtio-balloon
In all of these cases, you can have someone immediately re-access the
page table and re-populate it.
For something mostly static (balloon inflation, memory allocator), it's
not that big of a deal I guess.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2021-11-11 11:19 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-10 10:54 [PATCH v3 00/15] Free user PTE page table pages Qi Zheng
2021-11-10 10:54 ` [PATCH v3 01/15] mm: do code cleanups to filemap_map_pmd() Qi Zheng
2021-11-10 10:54 ` [PATCH v3 02/15] mm: introduce is_huge_pmd() helper Qi Zheng
2021-11-11 13:46 ` kernel test robot
2021-11-10 10:54 ` [PATCH v3 03/15] mm: move pte_offset_map_lock() to pgtable.h Qi Zheng
2021-11-10 10:54 ` [PATCH v3 04/15] mm: rework the parameter of lock_page_or_retry() Qi Zheng
2021-11-10 10:54 ` [PATCH v3 05/15] mm: add pmd_installed_type return for __pte_alloc() and other friends Qi Zheng
2021-11-10 10:54 ` [PATCH v3 06/15] mm: introduce refcount for user PTE page table page Qi Zheng
2021-11-11 0:37 ` kernel test robot
2021-11-10 10:54 ` [PATCH v3 07/15] mm/pte_ref: add support for user PTE page table page allocation Qi Zheng
2021-11-11 15:17 ` kernel test robot
2021-11-10 10:54 ` [PATCH v3 08/15] mm/pte_ref: initialize the refcount of the withdrawn PTE page table page Qi Zheng
2021-11-10 10:54 ` [PATCH v3 09/15] mm/pte_ref: add support for the map/unmap of user " Qi Zheng
2021-11-10 10:54 ` [PATCH v3 10/15] mm/pte_ref: add support for page fault path Qi Zheng
2021-11-10 10:54 ` [PATCH v3 11/15] mm/pte_ref: take a refcount before accessing the PTE page table page Qi Zheng
2021-11-10 10:54 ` [PATCH v3 12/15] mm/pte_ref: update the pmd entry in move_normal_pmd() Qi Zheng
2021-11-10 10:54 ` [PATCH v3 13/15] mm/pte_ref: free user PTE page table pages Qi Zheng
2021-11-14 14:43 ` [mm/pte_ref] afcc9fb874: kernel_BUG_at_include/linux/pte_ref.h kernel test robot
2021-11-10 10:54 ` [PATCH v3 14/15] Documentation: add document for pte_ref Qi Zheng
2021-11-10 14:39 ` Jonathan Corbet
2021-11-11 5:40 ` Qi Zheng
2021-11-10 10:54 ` [PATCH v3 15/15] mm/pte_ref: use mmu_gather to free PTE page table pages Qi Zheng
2021-11-10 12:56 ` [PATCH v3 00/15] Free user " Jason Gunthorpe
2021-11-10 13:25 ` David Hildenbrand
2021-11-10 13:59 ` Qi Zheng
2021-11-10 14:38 ` Jason Gunthorpe
2021-11-10 15:37 ` David Hildenbrand
2021-11-10 16:39 ` Jason Gunthorpe
2021-11-10 17:37 ` David Hildenbrand
2021-11-10 17:49 ` Jason Gunthorpe
2021-11-11 3:58 ` Qi Zheng
2021-11-11 9:22 ` David Hildenbrand
2021-11-11 11:08 ` Qi Zheng
2021-11-11 11:19 ` David Hildenbrand [this message]
2021-11-11 12:00 ` Qi Zheng
2021-11-11 12:20 ` David Hildenbrand
2021-11-11 12:32 ` Qi Zheng
2021-11-11 12:51 ` David Hildenbrand
2021-11-11 13:01 ` Qi Zheng
2021-11-10 16:49 ` Matthew Wilcox
2021-11-10 16:53 ` David Hildenbrand
2021-11-10 16:56 ` Jason Gunthorpe
2021-11-10 13:54 ` Qi Zheng
-- strict thread matches above, loose matches on Subject: below --
2021-11-10 8:40 Qi Zheng
2021-11-10 8:52 ` Qi Zheng
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=27d73395-70b4-fe4a-4c8d-415b43ff9c1f@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=jgg@nvidia.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mika.penttila@nextfour.com \
--cc=songmuchun@bytedance.com \
--cc=tglx@linutronix.de \
--cc=zhengqi.arch@bytedance.com \
--cc=zhouchengming@bytedance.com \
/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).