Re: [PATCH 0/6] mm: split underutilized THPs

[Usama Arif <usamaarif642@gmail.com>]

On 01/08/2024 17:27, David Hildenbrand wrote:
> On 01.08.24 18:22, Usama Arif wrote:
>>
>>
>> On 01/08/2024 07:09, Yu Zhao wrote:
>>> On Tue, Jul 30, 2024 at 6:54 AM Usama Arif <usamaarif642@gmail.com> wrote:
>>>>
>>>> The current upstream default policy for THP is always. However, Meta
>>>> uses madvise in production as the current THP=always policy vastly
>>>> overprovisions THPs in sparsely accessed memory areas, resulting in
>>>> excessive memory pressure and premature OOM killing.
>>>> Using madvise + relying on khugepaged has certain drawbacks over
>>>> THP=always. Using madvise hints mean THPs aren't "transparent" and
>>>> require userspace changes. Waiting for khugepaged to scan memory and
>>>> collapse pages into THP can be slow and unpredictable in terms of performance
>>>> (i.e. you dont know when the collapse will happen), while production
>>>> environments require predictable performance. If there is enough memory
>>>> available, its better for both performance and predictability to have
>>>> a THP from fault time, i.e. THP=always rather than wait for khugepaged
>>>> to collapse it, and deal with sparsely populated THPs when the system is
>>>> running out of memory.
>>>>
>>>> This patch-series is an attempt to mitigate the issue of running out of
>>>> memory when THP is always enabled. During runtime whenever a THP is being
>>>> faulted in or collapsed by khugepaged, the THP is added to a list.
>>>> Whenever memory reclaim happens, the kernel runs the deferred_split
>>>> shrinker which goes through the list and checks if the THP was underutilized,
>>>> i.e. how many of the base 4K pages of the entire THP were zero-filled.
>>>> If this number goes above a certain threshold, the shrinker will attempt
>>>> to split that THP. Then at remap time, the pages that were zero-filled are
>>>> not remapped, hence saving memory. This method avoids the downside of
>>>> wasting memory in areas where THP is sparsely filled when THP is always
>>>> enabled, while still providing the upside THPs like reduced TLB misses without
>>>> having to use madvise.
>>>>
>>>> Meta production workloads that were CPU bound (>99% CPU utilzation) were
>>>> tested with THP shrinker. The results after 2 hours are as follows:
>>>>
>>>>                              | THP=madvise |  THP=always   | THP=always
>>>>                              |             |               | + shrinker series
>>>>                              |             |               | + max_ptes_none=409
>>>> -----------------------------------------------------------------------------
>>>> Performance improvement     |      -      |    +1.8%      |     +1.7%
>>>> (over THP=madvise)          |             |               |
>>>> -----------------------------------------------------------------------------
>>>> Memory usage                |    54.6G    | 58.8G (+7.7%) |   55.9G (+2.4%)
>>>> -----------------------------------------------------------------------------
>>>> max_ptes_none=409 means that any THP that has more than 409 out of 512
>>>> (80%) zero filled filled pages will be split.
>>>>
>>>> To test out the patches, the below commands without the shrinker will
>>>> invoke OOM killer immediately and kill stress, but will not fail with
>>>> the shrinker:
>>>>
>>>> echo 450 > /sys/kernel/mm/transparent_hugepage/khugepaged/max_ptes_none
>>>> mkdir /sys/fs/cgroup/test
>>>> echo $$ > /sys/fs/cgroup/test/cgroup.procs
>>>> echo 20M > /sys/fs/cgroup/test/memory.max
>>>> echo 0 > /sys/fs/cgroup/test/memory.swap.max
>>>> # allocate twice memory.max for each stress worker and touch 40/512 of
>>>> # each THP, i.e. vm-stride 50K.
>>>> # With the shrinker, max_ptes_none of 470 and below won't invoke OOM
>>>> # killer.
>>>> # Without the shrinker, OOM killer is invoked immediately irrespective
>>>> # of max_ptes_none value and kill stress.
>>>> stress --vm 1 --vm-bytes 40M --vm-stride 50K
>>>>
>>>> Patches 1-2 add back helper functions that were previously removed
>>>> to operate on page lists (needed by patch 3).
>>>> Patch 3 is an optimization to free zapped tail pages rather than
>>>> waiting for page reclaim or migration.
>>>> Patch 4 is a prerequisite for THP shrinker to not remap zero-filled
>>>> subpages when splitting THP.
>>>> Patches 6 adds support for THP shrinker.
>>>>
>>>> (This patch-series restarts the work on having a THP shrinker in kernel
>>>> originally done in
>>>> https://lore.kernel.org/all/cover.1667454613.git.alexlzhu@fb.com/.
>>>> The THP shrinker in this series is significantly different than the
>>>> original one, hence its labelled v1 (although the prerequisite to not
>>>> remap clean subpages is the same).)
>>>>
>>>> Alexander Zhu (1):
>>>>    mm: add selftests to split_huge_page() to verify unmap/zap of zero
>>>>      pages
>>>>
>>>> Usama Arif (3):
>>>>    Revert "memcg: remove mem_cgroup_uncharge_list()"
>>>>    Revert "mm: remove free_unref_page_list()"
>>>>    mm: split underutilized THPs
>>>>
>>>> Yu Zhao (2):
>>>>    mm: free zapped tail pages when splitting isolated thp
>>>>    mm: don't remap unused subpages when splitting isolated thp
>>>
>>>   I would recommend shatter [1] instead of splitting so that
>>> 1) whoever underutilized their THPs get punished for the overhead;
>>> 2) underutilized THPs are kept intact and can be reused by others.
>>>
>>> [1] https://lore.kernel.org/20240229183436.4110845-3-yuzhao@google.com/
>>
>> The objective of this series is to reduce memory usage, while trying to keep the performance benefits you get of using THP=always. Punishing any applications performance is the opposite of what I am trying to do here.
>> For e.g. if there is only one main application running in production, and its using majority of the THPs, then reducing its performance doesn't make sense.
>>
> 
> I'm not sure if there would really be a performance degradation regarding the THP, after all we zap PTEs either way.
>

By performance I meant time/CPU used up for migration.
 
> Shattering will take longer because real migration is involved IIUC.
>

Yes, so thats what I want to avoid. If the system is CPU bound like the production workload I am testing, then spending any cycles on migration is going to make time/CPU performance worse.
 
Also, shattering isn't merged upstream, and it wouldn't make sense to make this series dependent on shattering.

>> Also, just going through the commit, and found the line "The advantage of shattering is that it keeps the original THP intact" a bit confusing. I am guessing the THP is freed? i.e. if a 2M THP has 10 non-zero filled base pages and the rest are zero-filled, then after shattering we will have 10*4K memory and not 2M+10*4K? Is it the case the THP is reused at next fault?
> 
> The idea is (as I understand it) to free the full THP abck to the buddy, replacing the individual pieces that are kept to freshly allocated order-0 pages from the buddy.
>