The patch titled Subject: mm: x86: add CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG has been added to the -mm tree. Its filename is mm-x86-add-config_arch_has_nonleaf_pmd_young.patch This patch should soon appear at https://ozlabs.org/~akpm/mmots/broken-out/mm-x86-add-config_arch_has_nonleaf_pmd_young.patch and later at https://ozlabs.org/~akpm/mmotm/broken-out/mm-x86-add-config_arch_has_nonleaf_pmd_young.patch 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 and is updated there every 3-4 working days ------------------------------------------------------ From: Yu Zhao <yuzhao@xxxxxxxxxx> Subject: mm: x86: add CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG Some architectures support the accessed bit in non-leaf PMD entries, e.g., x86 sets the accessed bit in a non-leaf PMD entry when using it as part of linear address translation [1]. Page table walkers that clear the accessed bit may use this capability to reduce their search space. Note that: 1. Although an inline function is preferable, this capability is added as a configuration option for consistency with the existing macros. 2. Due to the little interest in other varieties, this capability was only tested on Intel and AMD CPUs. [1]: Intel 64 and IA-32 Architectures Software Developer's Manual Volume 3 (June 2021), section 4.8 Link: https://lkml.kernel.org/r/20220407031525.2368067-3-yuzhao@xxxxxxxxxx Signed-off-by: Yu Zhao <yuzhao@xxxxxxxxxx> Reviewed-by: Barry Song <baohua@xxxxxxxxxx> Acked-by: Brian Geffon <bgeffon@xxxxxxxxxx> Acked-by: Jan Alexander Steffens (heftig) <heftig@xxxxxxxxxxxxx> Acked-by: Oleksandr Natalenko <oleksandr@xxxxxxxxxxxxxx> Acked-by: Steven Barrett <steven@xxxxxxxxxxxx> Acked-by: Suleiman Souhlal <suleiman@xxxxxxxxxx> Tested-by: Daniel Byrne <djbyrne@xxxxxxx> Tested-by: Donald Carr <d@xxxxxxxxxxxxxxx> Tested-by: Holger Hoffstätte <holger@xxxxxxxxxxxxxxxxxxxxxx> Tested-by: Konstantin Kharlamov <Hi-Angel@xxxxxxxxx> Tested-by: Shuang Zhai <szhai2@xxxxxxxxxxxxxxxx> Tested-by: Sofia Trinh <sofia.trinh@edi.works> Tested-by: Vaibhav Jain <vaibhav@xxxxxxxxxxxxx> Cc: Johannes Weiner <hannes@xxxxxxxxxxx> Cc: Mel Gorman <mgorman@xxxxxxxxxxxxxxxxxxx> Cc: Will Deacon <will@xxxxxxxxxx> Cc: Yang Shi <shy828301@xxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- --- a/arch/Kconfig~mm-x86-add-config_arch_has_nonleaf_pmd_young +++ a/arch/Kconfig @@ -1378,6 +1378,15 @@ config DYNAMIC_SIGFRAME config HAVE_ARCH_NODE_DEV_GROUP bool +config ARCH_HAS_NONLEAF_PMD_YOUNG + bool + depends on PGTABLE_LEVELS > 2 + help + Architectures that select this option are capable of setting the + accessed bit in non-leaf PMD entries when using them as part of linear + address translations. Page table walkers that clear the accessed bit + may use this capability to reduce their search space. + source "kernel/gcov/Kconfig" source "scripts/gcc-plugins/Kconfig" --- a/arch/x86/include/asm/pgtable.h~mm-x86-add-config_arch_has_nonleaf_pmd_young +++ a/arch/x86/include/asm/pgtable.h @@ -815,7 +815,8 @@ static inline unsigned long pmd_page_vad static inline int pmd_bad(pmd_t pmd) { - return (pmd_flags(pmd) & ~_PAGE_USER) != _KERNPG_TABLE; + return (pmd_flags(pmd) & ~(_PAGE_USER | _PAGE_ACCESSED)) != + (_KERNPG_TABLE & ~_PAGE_ACCESSED); } static inline unsigned long pages_to_mb(unsigned long npg) --- a/arch/x86/Kconfig~mm-x86-add-config_arch_has_nonleaf_pmd_young +++ a/arch/x86/Kconfig @@ -85,6 +85,7 @@ config X86 select ARCH_HAS_PMEM_API if X86_64 select ARCH_HAS_PTE_DEVMAP if X86_64 select ARCH_HAS_PTE_SPECIAL + select ARCH_HAS_NONLEAF_PMD_YOUNG select ARCH_HAS_UACCESS_FLUSHCACHE if X86_64 select ARCH_HAS_COPY_MC if X86_64 select ARCH_HAS_SET_MEMORY --- a/arch/x86/mm/pgtable.c~mm-x86-add-config_arch_has_nonleaf_pmd_young +++ a/arch/x86/mm/pgtable.c @@ -550,7 +550,7 @@ int ptep_test_and_clear_young(struct vm_ return ret; } -#ifdef CONFIG_TRANSPARENT_HUGEPAGE +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG) int pmdp_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pmd_t *pmdp) { @@ -562,6 +562,9 @@ int pmdp_test_and_clear_young(struct vm_ return ret; } +#endif + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE int pudp_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pud_t *pudp) { --- a/include/linux/pgtable.h~mm-x86-add-config_arch_has_nonleaf_pmd_young +++ a/include/linux/pgtable.h @@ -212,7 +212,7 @@ static inline int ptep_test_and_clear_yo #endif #ifndef __HAVE_ARCH_PMDP_TEST_AND_CLEAR_YOUNG -#ifdef CONFIG_TRANSPARENT_HUGEPAGE +#if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG) static inline int pmdp_test_and_clear_young(struct vm_area_struct *vma, unsigned long address, pmd_t *pmdp) @@ -233,7 +233,7 @@ static inline int pmdp_test_and_clear_yo BUILD_BUG(); return 0; } -#endif /* CONFIG_TRANSPARENT_HUGEPAGE */ +#endif /* CONFIG_TRANSPARENT_HUGEPAGE || CONFIG_ARCH_HAS_NONLEAF_PMD_YOUNG */ #endif #ifndef __HAVE_ARCH_PTEP_CLEAR_YOUNG_FLUSH _ Patches currently in -mm which might be from yuzhao@xxxxxxxxxx are mm-x86-arm64-add-arch_has_hw_pte_young.patch mm-x86-add-config_arch_has_nonleaf_pmd_young.patch mm-vmscanc-refactor-shrink_node.patch revert-include-linux-mm_inlineh-fold-__update_lru_size-into-its-sole-caller.patch mm-multi-gen-lru-groundwork.patch mm-multi-gen-lru-minimal-implementation.patch mm-multi-gen-lru-exploit-locality-in-rmap.patch mm-multi-gen-lru-support-page-table-walks.patch mm-multi-gen-lru-optimize-multiple-memcgs.patch mm-multi-gen-lru-kill-switch.patch mm-multi-gen-lru-thrashing-prevention.patch mm-multi-gen-lru-debugfs-interface.patch mm-multi-gen-lru-admin-guide.patch mm-multi-gen-lru-design-doc.patch