The patch titled
Subject: mm: migrate LRU_REFS_MASK bits in folio_migrate_flags
has been added to the -mm mm-unstable branch. Its filename is
mm-migrate-lru_refs_mask-bits-in-folio_migrate_flags.patch
This patch will shortly appear at
https://git.kernel.org/pub/scm/linux/kernel/git/akpm/25-new.git/tree/patches/mm-migrate-lru_refs_mask-bits-in-folio_migrate_flags.patch
This patch will later appear in the mm-unstable branch at
git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next via the mm-everything
branch at git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm
and is updated there every 2-3 working days
------------------------------------------------------
From: Zhaoyang Huang <zhaoyang.huang@xxxxxxxxxx>
Subject: mm: migrate LRU_REFS_MASK bits in folio_migrate_flags
Date: Thu, 26 Sep 2024 13:06:47 +0800
Bits of LRU_REFS_MASK are not inherited during migration which lead to new
folio start from tier0 when MGLRU enabled. Try to bring as much bits of
folio->flags as possible since compaction and alloc_contig_range which
introduce migration do happen at times.
Link: https://lkml.kernel.org/r/20240926050647.5653-1-zhaoyang.huang@xxxxxxxxxx
Signed-off-by: Zhaoyang Huang <zhaoyang.huang@xxxxxxxxxx>
Suggested-by: Yu Zhao <yuzhao@xxxxxxxxxx>
Acked-by: David Hildenbrand <david@xxxxxxxxxx>
Cc: Matthew Wilcox <willy@xxxxxxxxxxxxx>
Cc: Yu Zhao <yuzhao@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/mm_inline.h | 10 ++++++++++
mm/migrate.c | 1 +
2 files changed, 11 insertions(+)
--- a/include/linux/mm_inline.h~mm-migrate-lru_refs_mask-bits-in-folio_migrate_flags
+++ a/include/linux/mm_inline.h
@@ -291,6 +291,12 @@ static inline bool lru_gen_del_folio(str
return true;
}
+static inline void folio_migrate_refs(struct folio *new, struct folio *old)
+{
+ unsigned long refs = READ_ONCE(old->flags) & LRU_REFS_MASK;
+
+ set_mask_bits(&new->flags, LRU_REFS_MASK, refs);
+}
#else /* !CONFIG_LRU_GEN */
static inline bool lru_gen_enabled(void)
@@ -313,6 +319,10 @@ static inline bool lru_gen_del_folio(str
return false;
}
+static inline void folio_migrate_refs(struct folio *new, struct folio *old)
+{
+
+}
#endif /* CONFIG_LRU_GEN */
static __always_inline
--- a/mm/migrate.c~mm-migrate-lru_refs_mask-bits-in-folio_migrate_flags
+++ a/mm/migrate.c
@@ -694,6 +694,7 @@ void folio_migrate_flags(struct folio *n
if (folio_test_idle(folio))
folio_set_idle(newfolio);
+ folio_migrate_refs(newfolio, folio);
/*
* Copy NUMA information to the new page, to prevent over-eager
* future migrations of this same page.
_
Patches currently in -mm which might be from zhaoyang.huang@xxxxxxxxxx are
mm-migrate-lru_refs_mask-bits-in-folio_migrate_flags.patch
mm-optimization-on-page-allocation-when-cma-enabled.patch
