From: David Hildenbrand <david@redhat.com>
To: "akash.tyagi" <akash.tyagi@mediatek.com>,
akpm@linux-foundation.org, vbabka@suse.cz,
matthias.bgg@gmail.com, angelogioacchino.delregno@collabora.com
Cc: surenb@google.com, mhocko@suse.com, jackmanb@google.com,
hannes@cmpxchg.org, ziy@nvidia.com, linux-mm@kvack.org,
linux-kernel@vger.kernel.org,
linux-arm-kernel@lists.infradead.org,
linux-mediatek@lists.infradead.org, wsd_upstream@mediatek.com
Subject: Re: [RFC PATCH] mm/page_alloc: Add PCP list for THP CMA
Date: Thu, 24 Jul 2025 11:52:31 +0200 [thread overview]
Message-ID: <67a54f31-e568-427a-8fc8-9791fd34e11b@redhat.com> (raw)
In-Reply-To: <20250724075301.1151817-1-akash.tyagi@mediatek.com>
On 24.07.25 09:53, akash.tyagi wrote:
> Currently, THP CMA pages share PCP lists with UNMOVABLE and RECLAIMABLE
> pages. This may result in CMA THP pages being allocated from the PCP
> list for other migratetypes. When this occurs, these pages may fail to
> be isolated, leading to CMA allocation failures when drivers request
> them.
Curious, did you run into that in practice?
Having MIGRATE_CMA pages allocated for unmovable allocations would
indeed be broken.
But, MIGRATE_PCPTYPES does not include MIGRATE_CMA. So there is also no
dedicated PCP list for VMA?
In free_unref_folios(), we have
"Non-isolated types over MIGRATE_PCPTYPES get added to the
MIGRATE_MOVABLE pcp list."
if (unlikely(migratetype >= MIGRATE_PCPTYPES))
migratetype = MIGRATE_MOVABLE;
So ... shouldn't that safe us here as well for THPs?
>
> This patch introduces a dedicated PCP list for the THP CMA migratetype,
> ensuring that CMA THP pages are not mixed with other migratetypes and
> remain available for CMA allocations as intended.
>
> Signed-off-by: akash.tyagi <akash.tyagi@mediatek.com>
> ---
> include/linux/mmzone.h | 10 ++++++++--
> mm/page_alloc.c | 5 +++++
> 2 files changed, 13 insertions(+), 2 deletions(-)
>
> diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
> index 283913d42d7b..dd93088ce851 100644
> --- a/include/linux/mmzone.h
> +++ b/include/linux/mmzone.h
> @@ -696,11 +696,17 @@ enum zone_watermarks {
>
> /*
> * One per migratetype for each PAGE_ALLOC_COSTLY_ORDER. Two additional lists
> - * are added for THP. One PCP list is used by GPF_MOVABLE, and the other PCP list
> - * is used by GFP_UNMOVABLE and GFP_RECLAIMABLE.
> + * are added for THP: one for GFP_MOVABLE, and one for GFP_UNMOVABLE and
> + * GFP_RECLAIMABLE. With CMA enabled, an extra THP PCP list is added for
> + * MIGRATE_CMA, allowing further distinction between MIGRATE_MOVABLE and
> + * MIGRATE_CMA for THP allocations.
> */
> #ifdef CONFIG_TRANSPARENT_HUGEPAGE
> +#ifdef CONFIG_CMA
> +#define NR_PCP_THP 3
> +#else
> #define NR_PCP_THP 2
> +#endif
> #else
> #define NR_PCP_THP 0
> #endif
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index 2ef3c07266b3..35f8041afbcc 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -519,6 +519,11 @@ static inline unsigned int order_to_pindex(int migratetype, int order)
> if (order > PAGE_ALLOC_COSTLY_ORDER) {
> VM_BUG_ON(order != HPAGE_PMD_ORDER);
>
> +#ifdef CONFIG_CMA
> + if (migratetype == MIGRATE_CMA)
> + return NR_LOWORDER_PCP_LISTS + 2;
> +#endif
> +
> movable = migratetype == MIGRATE_MOVABLE;
>
> return NR_LOWORDER_PCP_LISTS + movable;
--
Cheers,
David / dhildenb
next prev parent reply other threads:[~2025-07-24 9:52 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-07-24 7:53 akash.tyagi
2025-07-24 9:52 ` David Hildenbrand [this message]
2025-07-25 5:08 ` akash.tyagi
2025-07-25 7:04 ` David Hildenbrand
2025-07-25 14:27 ` Zi Yan
2025-07-29 12:30 ` akash.tyagi
2025-07-29 12:42 ` David Hildenbrand
2025-07-29 12:50 ` Matthew Wilcox
2025-08-04 18:31 ` Juan Yescas
2025-08-04 18:20 Juan Yescas
2025-08-04 18:49 ` David Hildenbrand
2025-08-04 19:00 ` Zi Yan
2025-08-04 19:10 ` David Hildenbrand
2025-08-05 1:24 ` Juan Yescas
2025-08-05 1:22 ` Juan Yescas
2025-08-05 9:54 ` Vlastimil Babka
2025-08-05 16:46 ` Juan Yescas
2025-08-05 17:12 ` Juan Yescas
2025-08-05 21:09 ` Vlastimil Babka
2025-08-06 21:54 ` Juan Yescas
2025-08-05 9:58 ` David Hildenbrand
2025-08-05 16:57 ` Juan Yescas
2025-08-05 21:08 ` David Hildenbrand
2025-08-06 21:44 ` Juan Yescas
2025-08-06 21:51 ` Juan Yescas
2025-09-09 20:07 ` Juan Yescas
2025-09-09 20:11 ` Juan Yescas
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=67a54f31-e568-427a-8fc8-9791fd34e11b@redhat.com \
--to=david@redhat.com \
--cc=akash.tyagi@mediatek.com \
--cc=akpm@linux-foundation.org \
--cc=angelogioacchino.delregno@collabora.com \
--cc=hannes@cmpxchg.org \
--cc=jackmanb@google.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mediatek@lists.infradead.org \
--cc=linux-mm@kvack.org \
--cc=matthias.bgg@gmail.com \
--cc=mhocko@suse.com \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
--cc=wsd_upstream@mediatek.com \
--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