Linux-mm Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "David Hildenbrand (Arm)" <david@kernel.org>
To: Joanne Koong <joannelkoong@gmail.com>,
	akpm@linux-foundation.org, ljs@kernel.org
Cc: baohua@kernel.org, usama.arif@linux.dev, hannes@cmpxchg.org,
	kas@kernel.org, alex@ghiti.fr, ziy@nvidia.com,
	baolin.wang@linux.alibaba.com, liam@infradead.org,
	npache@redhat.com, ryan.roberts@arm.com, dev.jain@arm.com,
	lance.yang@linux.dev, vbabka@kernel.org, rppt@kernel.org,
	surenb@google.com, mhocko@suse.com, willy@infradead.org,
	linux-mm@kvack.org
Subject: Re: [PATCH v4 4/4] mm: add anonymous mTHP folios to the deferred split list
Date: Tue, 17 Nov 2026 23:21:08 +0100	[thread overview]
Message-ID: <0d75991d-9381-440c-8d7c-07bca53cc8ff@kernel.org> (raw)
In-Reply-To: <20261103212004.1356327-5-joannelkoong@gmail.com>

On 11/3/26 22:20, Joanne Koong wrote:
> Unlike for PMD-sized folios, an anonymous mTHP folio doesn't get added
> to the deferred split list at fault or collapse time. As a result, a
> fully mapped mTHP folio that is mostly zero-filled doesn't get split by
> the deferred split shrinker when the system is under memory pressure.
> 
> At Meta we would like to deploy 2M THP=always on arm64 with 64k base
> pages, as 2M gives the contpte benefits while the PMD size (512M) is too
> big to use. Without underused splitting this causes memory regressions,
> as the unused parts of those folios can never be broken down and
> reclaimed.
> 
> Add anonymous mTHP folios to the deferred split list from
> map_anon_folio_pte_nopf(), mirroring what map_anon_folio_pmd_nopf()
> already does for PMD-sized folios. This covers both the fault path and
> the khugepaged mTHP collapse path. If there is memory pressure, a
> zero-filled mTHP can then be split with its zero pages remapped to the
> shared zero page and reclaimed.
> 
> Only queue mTHP folios of at least 2M for underused splitting. With
> thp_underused() scaling khugepaged/max_ptes_none to the folio size, any
> folio of order 2 or above could in principle be underused, but for small
> folios it is unlikely to be worth it. An order-2 folio on 4k base pages
> can free at most four pages and on systems that only enable small mTHP
> sizes (eg Android's order-2-only configuration) zero-filled subpages
> tend to be few and short-lived.
> 
> The minimum is hard-coded rather than a knob, since admins have no good
> way of choosing it, and is a size rather than an order since what a
> split recovers depends on the folio's size in bytes, and the same order
> means very different sizes across different base page sizes. It can be
> lowered later if data shows smaller sizes are worth it. PMD-sized folios
> are exempt from the minimum in order to maintain backwards compatibility
> since on some architectures (s390, with a 1M PMD) the PMD size can be
> below 2M.
> 
> To minimize overhead on the common order-0 fault path, the
> deferred_split_folio() call is guarded by an inline folio_test_large()
> check.
> 
> Suggested-by: Usama Arif <usama.arif@linux.dev>
> Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
> ---
>  mm/huge_memory.c | 14 ++++++++++++++
>  mm/memory.c      |  2 ++
>  2 files changed, 16 insertions(+)
> 
> diff --git a/mm/huge_memory.c b/mm/huge_memory.c
> index 8f90fb609e57..cda3c886ec53 100644
> --- a/mm/huge_memory.c
> +++ b/mm/huge_memory.c
> @@ -4575,6 +4575,13 @@ static bool thp_underused_never(unsigned long max_ptes_none)
>  	return max_ptes_none == HPAGE_PMD_NR - 1;
>  }
>  
> +/*
> + * Smallest mTHP folio that is queued for underused splitting. Kept
> + * conservative, as small folios are unlikely to free enough memory to be
> + * worth queueing and scanning.
> + */
> +#define THP_UNDERUSED_MIN_SIZE	SZ_2M

Working on orders would be nice.

I think you can use something like

	order_base_2(SZ_2M >> PAGE_SHIFT)

for that.

And then you can just do something like

	max(order_base_2(SZ_2M >> PAGE_SHIFT), HPAGE_PMD_ORDER)

And place the comment next to this definition.

> +
>  /* partially_mapped=false won't clear PG_partially_mapped folio flag */
>  void deferred_split_folio(struct folio *folio, bool partially_mapped)
>  {
> @@ -4600,6 +4607,13 @@ void deferred_split_folio(struct folio *folio, bool partially_mapped)
>  		 */
>  		if (thp_underused_never(khugepaged_max_ptes_none))
>  			return;
> +		/*
> +		 * Keep queueing PMD-sized folios as before, even where the PMD
> +		 * size is below the minimum (eg 1M on s390).
> +		 */
> +		if (folio_size(folio) < THP_UNDERUSED_MIN_SIZE &&
> +		    !is_pmd_order(folio_order(folio)))
> +			return;
>  	}

Shame that we already did a function call just to figure that out. Shouldn't we
just replace the folio_test_large() thingy below by helper that embeds this
directly, with a focus on underused shrinking?

>  
>  	/*
> diff --git a/mm/memory.c b/mm/memory.c
> index 0c9ed04528d8..44376321180b 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -5477,6 +5477,8 @@ void map_anon_folio_pte_nopf(struct folio *folio, pte_t *pte,
>  	folio_add_lru_vma(folio, vma);
>  	set_ptes(vma->vm_mm, addr, pte, entry, nr_pages);
>  	update_mmu_cache_range(NULL, vma, addr, pte, nr_pages);
> +	if (folio_test_large(folio))
> +		deferred_split_folio(folio, false);

Do we also ahve to take care of mTHP placed by khugepaged?

-- 
Cheers,

David


      parent reply	other threads:[~2026-11-17 22:21 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-03 21:20 [PATCH v4 0/4] mm: split underused anonymous mTHP folios Joanne Koong
2026-11-03 21:20 ` [PATCH v4 1/4] mm/huge_memory: scale thp_underused() to the folio size Joanne Koong
2026-11-05 12:20   ` Kiryl Shutsemau
2026-11-10  3:54   ` Zi Yan
2026-11-12 23:50     ` Joanne Koong
2026-11-10  8:00   ` Baolin Wang
2026-11-17 22:12   ` David Hildenbrand (Arm)
2026-11-03 21:20 ` [PATCH v4 2/4] mm/huge_memory: don't queue folios when underused splitting is inactive Joanne Koong
2026-11-03 22:15   ` Usama Arif
2026-11-05 12:23   ` Kiryl Shutsemau
2026-11-10  8:13   ` Baolin Wang
2026-11-03 21:20 ` [PATCH v4 3/4] mm/huge_memory: add a per-order counter for underused splits Joanne Koong
2026-11-03 22:20   ` Usama Arif
2026-11-05 12:26     ` Kiryl Shutsemau
2026-11-06  0:06       ` Joanne Koong
2026-11-10  8:10   ` Baolin Wang
2026-11-17 22:22   ` David Hildenbrand (Arm)
2026-11-03 21:20 ` [PATCH v4 4/4] mm: add anonymous mTHP folios to the deferred split list Joanne Koong
2026-11-03 22:20   ` Barry Song
2026-11-04 21:21   ` sashiko-bot
2026-11-04 23:50     ` Joanne Koong
2026-11-05 12:37       ` Kiryl Shutsemau
2026-11-06  0:05         ` Joanne Koong
2026-11-09 22:29           ` Joanne Koong
2026-11-17 22:14             ` David Hildenbrand (Arm)
2026-11-17 22:21   ` David Hildenbrand (Arm) [this message]

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=0d75991d-9381-440c-8d7c-07bca53cc8ff@kernel.org \
    --to=david@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=alex@ghiti.fr \
    --cc=baohua@kernel.org \
    --cc=baolin.wang@linux.alibaba.com \
    --cc=dev.jain@arm.com \
    --cc=hannes@cmpxchg.org \
    --cc=joannelkoong@gmail.com \
    --cc=kas@kernel.org \
    --cc=lance.yang@linux.dev \
    --cc=liam@infradead.org \
    --cc=linux-mm@kvack.org \
    --cc=ljs@kernel.org \
    --cc=mhocko@suse.com \
    --cc=npache@redhat.com \
    --cc=rppt@kernel.org \
    --cc=ryan.roberts@arm.com \
    --cc=surenb@google.com \
    --cc=usama.arif@linux.dev \
    --cc=vbabka@kernel.org \
    --cc=willy@infradead.org \
    --cc=ziy@nvidia.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