From 143937ca51cc6ae2fccc61a1cb916abb24cd34f5 Mon Sep 17 00:00:00 2001 From: Huang Ying Date: Wed, 15 Oct 2025 10:37:12 +0800 Subject: [PATCH 1/2] arm64, mm: avoid always making PTE dirty in pte_mkwrite() Current pte_mkwrite_novma() makes PTE dirty unconditionally. This may mark some pages that are never written dirty wrongly. For example, do_swap_page() may map the exclusive pages with writable and clean PTEs if the VMA is writable and the page fault is for read access. However, current pte_mkwrite_novma() implementation always dirties the PTE. This may cause unnecessary disk writing if the pages are never written before being reclaimed. So, change pte_mkwrite_novma() to clear the PTE_RDONLY bit only if the PTE_DIRTY bit is set to make it possible to make the PTE writable and clean. The current behavior was introduced in commit 73e86cb03cf2 ("arm64: Move PTE_RDONLY bit handling out of set_pte_at()"). Before that, pte_mkwrite() only sets the PTE_WRITE bit, while set_pte_at() only clears the PTE_RDONLY bit if both the PTE_WRITE and the PTE_DIRTY bits are set. To test the performance impact of the patch, on an arm64 server machine, run 16 redis-server processes on socket 1 and 16 memtier_benchmark processes on socket 0 with mostly get transactions (that is, redis-server will mostly read memory only). The memory footprint of redis-server is larger than the available memory, so swap out/in will be triggered. Test results show that the patch can avoid most swapping out because the pages are mostly clean. And the benchmark throughput improves ~23.9% in the test. Fixes: 73e86cb03cf2 ("arm64: Move PTE_RDONLY bit handling out of set_pte_at()") Signed-off-by: Huang Ying Cc: Will Deacon Cc: Anshuman Khandual Cc: Ryan Roberts Cc: Gavin Shan Cc: Ard Biesheuvel Cc: Matthew Wilcox (Oracle) Cc: Yicong Yang Cc: linux-arm-kernel@lists.infradead.org Cc: linux-kernel@vger.kernel.org Reviewed-by: Catalin Marinas Signed-off-by: Catalin Marinas --- arch/arm64/include/asm/pgtable.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h index aa89c2e67ebc..0944e296dd4a 100644 --- a/arch/arm64/include/asm/pgtable.h +++ b/arch/arm64/include/asm/pgtable.h @@ -293,7 +293,8 @@ static inline pmd_t set_pmd_bit(pmd_t pmd, pgprot_t prot) static inline pte_t pte_mkwrite_novma(pte_t pte) { pte = set_pte_bit(pte, __pgprot(PTE_WRITE)); - pte = clear_pte_bit(pte, __pgprot(PTE_RDONLY)); + if (pte_sw_dirty(pte)) + pte = clear_pte_bit(pte, __pgprot(PTE_RDONLY)); return pte; } From b98c94eed4a975e0c80b7e90a649a46967376f58 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Wed, 22 Oct 2025 11:09:14 +0100 Subject: [PATCH 2/2] arm64: mte: Do not warn if the page is already tagged in copy_highpage() The arm64 copy_highpage() assumes that the destination page is newly allocated and not MTE-tagged (PG_mte_tagged unset) and warns accordingly. However, following commit 060913999d7a ("mm: migrate: support poisoned recover from migrate folio"), folio_mc_copy() is called before __folio_migrate_mapping(). If the latter fails (-EAGAIN), the copy will be done again to the same destination page. Since copy_highpage() already set the PG_mte_tagged flag, this second copy will warn. Replace the WARN_ON_ONCE(page already tagged) in the arm64 copy_highpage() with a comment. Reported-by: syzbot+d1974fc28545a3e6218b@syzkaller.appspotmail.com Link: https://lore.kernel.org/r/68dda1ae.a00a0220.102ee.0065.GAE@google.com Reviewed-by: David Hildenbrand Cc: Will Deacon Cc: Kefeng Wang Cc: stable@vger.kernel.org # 6.12.x Reviewed-by: Yang Shi Signed-off-by: Catalin Marinas --- arch/arm64/mm/copypage.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/arch/arm64/mm/copypage.c b/arch/arm64/mm/copypage.c index a86c897017df..cd5912ba617b 100644 --- a/arch/arm64/mm/copypage.c +++ b/arch/arm64/mm/copypage.c @@ -35,7 +35,7 @@ void copy_highpage(struct page *to, struct page *from) from != folio_page(src, 0)) return; - WARN_ON_ONCE(!folio_try_hugetlb_mte_tagging(dst)); + folio_try_hugetlb_mte_tagging(dst); /* * Populate tags for all subpages. @@ -51,8 +51,13 @@ void copy_highpage(struct page *to, struct page *from) } folio_set_hugetlb_mte_tagged(dst); } else if (page_mte_tagged(from)) { - /* It's a new page, shouldn't have been tagged yet */ - WARN_ON_ONCE(!try_page_mte_tagging(to)); + /* + * Most of the time it's a new page that shouldn't have been + * tagged yet. However, folio migration can end up reusing the + * same page without untagging it. Ignore the warning if the + * page is already tagged. + */ + try_page_mte_tagging(to); mte_copy_page_tags(kto, kfrom); set_page_mte_tagged(to);