The patch titled Subject: hugetlb: allow to free gigantic pages regardless of the configuration has been removed from the -mm tree. Its filename was hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration.patch This patch was dropped because an updated version will be merged ------------------------------------------------------ From: Alexandre Ghiti <alex@xxxxxxxx> Subject: hugetlb: allow to free gigantic pages regardless of the configuration On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but that support gigantic pages, boottime reserved gigantic pages can not be freed at all. This patchs simply enables the possibility to hand back those pages to memory allocator. This commit then renames gigantic_page_supported and ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values being false does not mean that the system cannot use gigantic pages: it just means that runtime allocation of gigantic pages is not supported, one can still allocate boottime gigantic pages if the architecture supports it. Link: http://lkml.kernel.org/r/20190117183953.5990-1-aghiti@xxxxxxxxx Signed-off-by: Alexandre Ghiti <alex@xxxxxxxx> Reviewed-by: Mike Kravetz <mike.kravetz@xxxxxxxxxx> Acked-by: Will Deacon <will.deacon@xxxxxxx> [arm64] Acked-by: Vlastimil Babka <vbabka@xxxxxxx> Cc: Catalin Marinas <catalin.marinas@xxxxxxx> Cc: Benjamin Herrenschmidt <benh@xxxxxxxxxxxxxxxxxxx> Cc: Paul Mackerras <paulus@xxxxxxxxx> Cc: Michael Ellerman <mpe@xxxxxxxxxxxxxx> Cc: Martin Schwidefsky <schwidefsky@xxxxxxxxxx> Cc: Heiko Carstens <heiko.carstens@xxxxxxxxxx> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> Cc: Ingo Molnar <mingo@xxxxxxxxxx> Cc: Borislav Petkov <bp@xxxxxxxxx> Cc: "H . Peter Anvin" <hpa@xxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- arch/arm64/Kconfig | 2 arch/arm64/include/asm/hugetlb.h | 7 +- arch/powerpc/include/asm/book3s/64/hugetlb.h | 4 - arch/powerpc/platforms/Kconfig.cputype | 2 arch/s390/Kconfig | 2 arch/s390/include/asm/hugetlb.h | 7 +- arch/x86/Kconfig | 2 arch/x86/include/asm/hugetlb.h | 7 +- fs/Kconfig | 2 include/linux/gfp.h | 2 mm/hugetlb.c | 43 +++++++++-------- mm/page_alloc.c | 4 - 12 files changed, 48 insertions(+), 36 deletions(-) --- a/arch/arm64/include/asm/hugetlb.h~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/arm64/include/asm/hugetlb.h @@ -70,8 +70,11 @@ extern void set_huge_swap_pte_at(struct #include <asm-generic/hugetlb.h> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE -static inline bool gigantic_page_supported(void) { return true; } +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION +static inline bool gigantic_page_runtime_allocation_supported(void) +{ + return true; +} #endif #endif /* __ASM_HUGETLB_H */ --- a/arch/arm64/Kconfig~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/arm64/Kconfig @@ -18,7 +18,7 @@ config ARM64 select ARCH_HAS_FAST_MULTIPLIER select ARCH_HAS_FORTIFY_SOURCE select ARCH_HAS_GCOV_PROFILE_ALL - select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA + select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA select ARCH_HAS_KCOV select ARCH_HAS_MEMBARRIER_SYNC_CORE select ARCH_HAS_PTE_SPECIAL --- a/arch/powerpc/include/asm/book3s/64/hugetlb.h~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/powerpc/include/asm/book3s/64/hugetlb.h @@ -36,8 +36,8 @@ static inline int hstate_get_psize(struc } } -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE -static inline bool gigantic_page_supported(void) +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION +static inline bool gigantic_page_runtime_allocation_supported(void) { /* * We used gigantic page reservation with hypervisor assist in some case. --- a/arch/powerpc/platforms/Kconfig.cputype~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/powerpc/platforms/Kconfig.cputype @@ -325,7 +325,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK config PPC_RADIX_MMU bool "Radix MMU Support" depends on PPC_BOOK3S_64 - select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA + select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA default y help Enable support for the Power ISA 3.0 Radix style MMU. Currently this --- a/arch/s390/include/asm/hugetlb.h~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/s390/include/asm/hugetlb.h @@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_ return pte_modify(pte, newprot); } -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE -static inline bool gigantic_page_supported(void) { return true; } +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION +static inline bool gigantic_page_runtime_allocation_supported(void) +{ + return true; +} #endif #endif /* _ASM_S390_HUGETLB_H */ --- a/arch/s390/Kconfig~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/s390/Kconfig @@ -69,7 +69,7 @@ config S390 select ARCH_HAS_ELF_RANDOMIZE select ARCH_HAS_FORTIFY_SOURCE select ARCH_HAS_GCOV_PROFILE_ALL - select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA + select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA select ARCH_HAS_KCOV select ARCH_HAS_PTE_SPECIAL select ARCH_HAS_SET_MEMORY --- a/arch/x86/include/asm/hugetlb.h~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/x86/include/asm/hugetlb.h @@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_f { } -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE -static inline bool gigantic_page_supported(void) { return true; } +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION +static inline bool gigantic_page_runtime_allocation_supported(void) +{ + return true; +} #endif #endif /* _ASM_X86_HUGETLB_H */ --- a/arch/x86/Kconfig~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/arch/x86/Kconfig @@ -21,7 +21,7 @@ config X86_64 def_bool y depends on 64BIT # Options that are inherently 64-bit kernel only: - select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA + select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA select ARCH_SUPPORTS_INT128 select ARCH_USE_CMPXCHG_LOCKREF select HAVE_ARCH_SOFT_DIRTY --- a/fs/Kconfig~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/fs/Kconfig @@ -214,7 +214,7 @@ config HUGETLB_PAGE config MEMFD_CREATE def_bool TMPFS || HUGETLBFS -config ARCH_HAS_GIGANTIC_PAGE +config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION bool source "fs/configfs/Kconfig" --- a/include/linux/gfp.h~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/include/linux/gfp.h @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage( /* The below functions must be run on a range from a single zone. */ extern int alloc_contig_range(unsigned long start, unsigned long end, unsigned migratetype, gfp_t gfp_mask); -extern void free_contig_range(unsigned long pfn, unsigned nr_pages); #endif +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages); #ifdef CONFIG_CMA /* CMA stuff */ --- a/mm/hugetlb.c~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/mm/hugetlb.c @@ -1036,7 +1036,6 @@ static int hstate_next_node_to_free(stru ((node = hstate_next_node_to_free(hs, mask)) || 1); \ nr_nodes--) -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE static void destroy_compound_gigantic_page(struct page *page, unsigned int order) { @@ -1059,6 +1058,7 @@ static void free_gigantic_page(struct pa free_contig_range(page_to_pfn(page), 1 << order); } +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION static int __alloc_gigantic_page(unsigned long start_pfn, unsigned long nr_pages, gfp_t gfp_mask) { @@ -1144,22 +1144,19 @@ static struct page *alloc_gigantic_page( static void prep_new_huge_page(struct hstate *h, struct page *page, int nid); static void prep_compound_gigantic_page(struct page *page, unsigned int order); -#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */ -static inline bool gigantic_page_supported(void) { return false; } +#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */ +static inline bool gigantic_page_runtime_allocation_supported(void) +{ + return false; +} static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask, int nid, nodemask_t *nodemask) { return NULL; } -static inline void free_gigantic_page(struct page *page, unsigned int order) { } -static inline void destroy_compound_gigantic_page(struct page *page, - unsigned int order) { } #endif static void update_and_free_page(struct hstate *h, struct page *page) { int i; - if (hstate_is_gigantic(h) && !gigantic_page_supported()) - return; - h->nr_huge_pages--; h->nr_huge_pages_node[page_to_nid(page)]--; for (i = 0; i < pages_per_huge_page(h); i++) { @@ -2277,13 +2274,20 @@ found: } #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages) -static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count, +static int set_max_huge_pages(struct hstate *h, unsigned long count, nodemask_t *nodes_allowed) { unsigned long min_count, ret; - if (hstate_is_gigantic(h) && !gigantic_page_supported()) - return h->max_huge_pages; + if (hstate_is_gigantic(h) && + !gigantic_page_runtime_allocation_supported()) { + spin_lock(&hugetlb_lock); + if (count > persistent_huge_pages(h)) { + spin_unlock(&hugetlb_lock); + return -EINVAL; + } + goto decrease_pool; + } /* * Increase the pool size @@ -2323,6 +2327,7 @@ static unsigned long set_max_huge_pages( goto out; } +decrease_pool: /* * Decrease the pool size * First return free pages to the buddy allocator (being careful @@ -2351,9 +2356,10 @@ static unsigned long set_max_huge_pages( break; } out: - ret = persistent_huge_pages(h); + h->max_huge_pages = persistent_huge_pages(h); spin_unlock(&hugetlb_lock); - return ret; + + return 0; } #define HSTATE_ATTR_RO(_name) \ @@ -2405,11 +2411,6 @@ static ssize_t __nr_hugepages_store_comm int err; NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY); - if (hstate_is_gigantic(h) && !gigantic_page_supported()) { - err = -EINVAL; - goto out; - } - if (nid == NUMA_NO_NODE) { /* * global hstate attribute @@ -2429,7 +2430,9 @@ static ssize_t __nr_hugepages_store_comm } else nodes_allowed = &node_states[N_MEMORY]; - h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed); + err = set_max_huge_pages(h, count, nodes_allowed); + if (err) + goto out; if (nodes_allowed != &node_states[N_MEMORY]) NODEMASK_FREE(nodes_allowed); --- a/mm/page_alloc.c~hugetlb-allow-to-free-gigantic-pages-regardless-of-the-configuration +++ a/mm/page_alloc.c @@ -8319,8 +8319,9 @@ done: pfn_max_align_up(end), migratetype); return ret; } +#endif -void free_contig_range(unsigned long pfn, unsigned nr_pages) +void free_contig_range(unsigned long pfn, unsigned int nr_pages) { unsigned int count = 0; @@ -8332,7 +8333,6 @@ void free_contig_range(unsigned long pfn } WARN(count != 0, "%d pages are still in use!\n", count); } -#endif #ifdef CONFIG_MEMORY_HOTPLUG /* _ Patches currently in -mm which might be from alex@xxxxxxxx are