From: David Hildenbrand <david@redhat.com>
To: Qi Zheng <zhengqi.arch@bytedance.com>,
akpm@linux-foundation.org, tglx@linutronix.de,
hannes@cmpxchg.org, mhocko@kernel.org, vdavydov.dev@gmail.com
Cc: linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
linux-mm@kvack.org, songmuchun@bytedance.com
Subject: Re: [PATCH 0/7] Free user PTE page table pages
Date: Mon, 19 Jul 2021 13:28:38 +0200 [thread overview]
Message-ID: <089e710c-fb06-e731-6d50-7858d6b9ecdf@redhat.com> (raw)
In-Reply-To: <5ce5fb25-df1d-b807-8807-595b8a7bfc63@redhat.com>
On 19.07.21 09:34, David Hildenbrand wrote:
> On 18.07.21 06:30, Qi Zheng wrote:
>> Hi,
>>
>> This patch series aims to free user PTE page table pages when all PTE entries
>> are empty.
>>
>> The beginning of this story is that some malloc libraries(e.g. jemalloc or
>> tcmalloc) usually allocate the amount of VAs by mmap() and do not unmap those VAs.
>> They will use madvise(MADV_DONTNEED) to free physical memory if they want.
>> But the page tables do not be freed by madvise(), so it can produce many
>> page tables when the process touches an enormous virtual address space.
>
> ... did you see that I am actually looking into this?
>
> https://lkml.kernel.org/r/bae8b967-c206-819d-774c-f57b94c4b362@redhat.com
>
> and have already spent a significant time on it as part of my research,
> which is *really* unfortunate and makes me quite frustrated at the
> beginning of the week alreadty ...
>
> Ripping out page tables is quite difficult, as we have to stop all page
> table walkers from touching it, including the fast_gup, rmap and page
> faults. This usually involves taking the mmap lock in write. My approach
> does page table reclaim asynchronously from another thread and do not
> rely on reference counts.
FWIW, I had a quick peek and I like the simplistic approach using
reference counting, although it seems to come with a price. By hooking
using pte_alloc_get_map_lock() instead of pte_alloc_map_lock, we can
handle quite some cases easily.
There are cases where we might immediately see a reuse after discarding
memory (especially, with virtio-balloon free page reporting), in which
case it's suboptimal to immediately discard instead of waiting a bit if
there is a reuse. However, the performance impact seems to be
comparatively small.
I do wonder if the 1% overhead you're seeing is actually because of
allcoating/freeing or because of the reference count handling on some
hot paths.
I'm primarily looking into asynchronous reclaim, because it somewhat
makes sense to only reclaim (+ pay a cost) when there is really need to
reclaim memory -- similar to our shrinker infrastructure.
--
Thanks,
David / dhildenb
next prev parent reply other threads:[~2021-07-19 11:28 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-07-18 4:30 [PATCH 0/7] Free user PTE page table pages Qi Zheng
2021-07-18 4:30 ` [PATCH 1/7] mm: fix the deadlock in finish_fault() Qi Zheng
2021-07-18 21:28 ` Kirill A. Shutemov
2021-07-19 9:53 ` Qi Zheng
2021-07-20 23:14 ` Andrew Morton
2021-07-21 2:21 ` Qi Zheng
2021-07-18 4:30 ` [PATCH 2/7] mm: introduce pte_install() helper Qi Zheng
2021-07-18 21:31 ` Kirill A. Shutemov
2021-07-19 10:20 ` Qi Zheng
2021-07-18 4:30 ` [PATCH 3/7] mm: remove redundant smp_wmb() Qi Zheng
2021-07-27 13:39 ` Muchun Song
2021-07-18 4:30 ` [PATCH 4/7] mm: rework the parameter of lock_page_or_retry() Qi Zheng
2021-07-18 4:30 ` [PATCH 5/7] mm: free user PTE page table pages Qi Zheng
2021-07-18 6:19 ` Mika Penttilä
2021-07-19 12:56 ` Qi Zheng
2021-07-19 13:55 ` Mika Penttilä
2021-07-19 14:12 ` Qi Zheng
2021-07-19 14:17 ` Mika Penttilä
2021-07-18 22:01 ` Kirill A. Shutemov
2021-07-19 13:55 ` Qi Zheng
2021-07-18 4:30 ` [PATCH 6/7] mm: defer freeing PTE page table for a grace period Qi Zheng
2021-07-18 4:30 ` [PATCH 7/7] mm: use mmu_gather to free PTE page table Qi Zheng
2021-07-19 7:34 ` [PATCH 0/7] Free user PTE page table pages David Hildenbrand
2021-07-19 11:28 ` David Hildenbrand [this message]
2021-07-19 12:42 ` Muchun Song
2021-07-19 13:30 ` Muchun Song
2021-07-20 4:00 ` 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=089e710c-fb06-e731-6d50-7858d6b9ecdf@redhat.com \
--to=david@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhocko@kernel.org \
--cc=songmuchun@bytedance.com \
--cc=tglx@linutronix.de \
--cc=vdavydov.dev@gmail.com \
--cc=zhengqi.arch@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).