From b4e98d9ac775907cc53fb08fcb6776deb7694e30 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Wed, 15 Nov 2017 17:35:33 -0800 Subject: mm: account pud page tables On a machine with 5-level paging support a process can allocate significant amount of memory and stay unnoticed by oom-killer and memory cgroup. The trick is to allocate a lot of PUD page tables. We don't account PUD page tables, only PMD and PTE. We already addressed the same issue for PMD page tables, see commit dc6c9a35b66b ("mm: account pmd page tables to the process"). Introduction of 5-level paging brings the same issue for PUD page tables. The patch expands accounting to PUD level. [kirill.shutemov@linux.intel.com: s/pmd_t/pud_t/] Link: http://lkml.kernel.org/r/20171004074305.x35eh5u7ybbt5kar@black.fi.intel.com [heiko.carstens@de.ibm.com: s390/mm: fix pud table accounting] Link: http://lkml.kernel.org/r/20171103090551.18231-1-heiko.carstens@de.ibm.com Link: http://lkml.kernel.org/r/20171002080427.3320-1-kirill.shutemov@linux.intel.com Signed-off-by: Kirill A. Shutemov Signed-off-by: Heiko Carstens Acked-by: Rik van Riel Acked-by: Michal Hocko Cc: Vlastimil Babka Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/hugetlbpage.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/powerpc') diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c index 1571a498a33f..a9b9083c5e49 100644 --- a/arch/powerpc/mm/hugetlbpage.c +++ b/arch/powerpc/mm/hugetlbpage.c @@ -433,6 +433,7 @@ static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd, pud = pud_offset(pgd, start); pgd_clear(pgd); pud_free_tlb(tlb, pud, start); + mm_dec_nr_puds(tlb->mm); } /* -- cgit v1.2.3-70-g09d2 From 75f296d93bcebcfe375884ddac79e30263a31766 Mon Sep 17 00:00:00 2001 From: "Levin, Alexander (Sasha Levin)" Date: Wed, 15 Nov 2017 17:35:54 -0800 Subject: kmemcheck: stop using GFP_NOTRACK and SLAB_NOTRACK Convert all allocations that used a NOTRACK flag to stop using it. Link: http://lkml.kernel.org/r/20171007030159.22241-3-alexander.levin@verizon.com Signed-off-by: Sasha Levin Cc: Alexander Potapenko Cc: Eric W. Biederman Cc: Michal Hocko Cc: Pekka Enberg Cc: Steven Rostedt Cc: Tim Hansen Cc: Vegard Nossum Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/arm/include/asm/pgalloc.h | 2 +- arch/arm64/include/asm/pgalloc.h | 2 +- arch/powerpc/include/asm/pgalloc.h | 2 +- arch/sh/kernel/dwarf.c | 4 ++-- arch/sh/kernel/process.c | 2 +- arch/sparc/mm/init_64.c | 4 ++-- arch/unicore32/include/asm/pgalloc.h | 2 +- arch/x86/kernel/espfix_64.c | 2 +- arch/x86/mm/init.c | 3 +-- arch/x86/mm/init_64.c | 2 +- arch/x86/mm/pageattr.c | 10 +++++----- arch/x86/mm/pgtable.c | 2 +- arch/x86/platform/efi/efi_64.c | 2 +- crypto/xor.c | 7 +------ include/linux/thread_info.h | 5 ++--- init/do_mounts.c | 3 +-- kernel/fork.c | 12 ++++++------ kernel/signal.c | 3 +-- mm/kmemcheck.c | 2 +- mm/slab.c | 2 +- mm/slab.h | 5 ++--- mm/slab_common.c | 2 +- mm/slub.c | 4 +--- 23 files changed, 36 insertions(+), 48 deletions(-) (limited to 'arch/powerpc') diff --git a/arch/arm/include/asm/pgalloc.h b/arch/arm/include/asm/pgalloc.h index b2902a5cd780..2d7344f0e208 100644 --- a/arch/arm/include/asm/pgalloc.h +++ b/arch/arm/include/asm/pgalloc.h @@ -57,7 +57,7 @@ static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd) extern pgd_t *pgd_alloc(struct mm_struct *mm); extern void pgd_free(struct mm_struct *mm, pgd_t *pgd); -#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL | __GFP_ZERO) static inline void clean_pte_table(pte_t *pte) { diff --git a/arch/arm64/include/asm/pgalloc.h b/arch/arm64/include/asm/pgalloc.h index d25f4f137c2a..5ca6a573a701 100644 --- a/arch/arm64/include/asm/pgalloc.h +++ b/arch/arm64/include/asm/pgalloc.h @@ -26,7 +26,7 @@ #define check_pgt_cache() do { } while (0) -#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL | __GFP_ZERO) #define PGD_SIZE (PTRS_PER_PGD * sizeof(pgd_t)) #if CONFIG_PGTABLE_LEVELS > 2 diff --git a/arch/powerpc/include/asm/pgalloc.h b/arch/powerpc/include/asm/pgalloc.h index a14203c005f1..e11f03007b57 100644 --- a/arch/powerpc/include/asm/pgalloc.h +++ b/arch/powerpc/include/asm/pgalloc.h @@ -18,7 +18,7 @@ static inline gfp_t pgtable_gfp_flags(struct mm_struct *mm, gfp_t gfp) } #endif /* MODULE */ -#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL | __GFP_ZERO) #ifdef CONFIG_PPC_BOOK3S #include diff --git a/arch/sh/kernel/dwarf.c b/arch/sh/kernel/dwarf.c index e1d751ae2498..1a2526676a87 100644 --- a/arch/sh/kernel/dwarf.c +++ b/arch/sh/kernel/dwarf.c @@ -1172,11 +1172,11 @@ static int __init dwarf_unwinder_init(void) dwarf_frame_cachep = kmem_cache_create("dwarf_frames", sizeof(struct dwarf_frame), 0, - SLAB_PANIC | SLAB_HWCACHE_ALIGN | SLAB_NOTRACK, NULL); + SLAB_PANIC | SLAB_HWCACHE_ALIGN, NULL); dwarf_reg_cachep = kmem_cache_create("dwarf_regs", sizeof(struct dwarf_reg), 0, - SLAB_PANIC | SLAB_HWCACHE_ALIGN | SLAB_NOTRACK, NULL); + SLAB_PANIC | SLAB_HWCACHE_ALIGN, NULL); dwarf_frame_pool = mempool_create_slab_pool(DWARF_FRAME_MIN_REQ, dwarf_frame_cachep); diff --git a/arch/sh/kernel/process.c b/arch/sh/kernel/process.c index b2d9963d5978..68b1a67533ce 100644 --- a/arch/sh/kernel/process.c +++ b/arch/sh/kernel/process.c @@ -59,7 +59,7 @@ void arch_task_cache_init(void) task_xstate_cachep = kmem_cache_create("task_xstate", xstate_size, __alignof__(union thread_xstate), - SLAB_PANIC | SLAB_NOTRACK, NULL); + SLAB_PANIC, NULL); } #ifdef CONFIG_SH_FPU_EMU diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 61bdc1270d19..2de22d703076 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -2927,7 +2927,7 @@ void __flush_tlb_all(void) pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { - struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO); + struct page *page = alloc_page(GFP_KERNEL | __GFP_ZERO); pte_t *pte = NULL; if (page) @@ -2939,7 +2939,7 @@ pte_t *pte_alloc_one_kernel(struct mm_struct *mm, pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) { - struct page *page = alloc_page(GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO); + struct page *page = alloc_page(GFP_KERNEL | __GFP_ZERO); if (!page) return NULL; if (!pgtable_page_ctor(page)) { diff --git a/arch/unicore32/include/asm/pgalloc.h b/arch/unicore32/include/asm/pgalloc.h index 26775793c204..f0fdb268f8f2 100644 --- a/arch/unicore32/include/asm/pgalloc.h +++ b/arch/unicore32/include/asm/pgalloc.h @@ -28,7 +28,7 @@ extern void free_pgd_slow(struct mm_struct *mm, pgd_t *pgd); #define pgd_alloc(mm) get_pgd_slow(mm) #define pgd_free(mm, pgd) free_pgd_slow(mm, pgd) -#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL | __GFP_ZERO) /* * Allocate one PTE table. diff --git a/arch/x86/kernel/espfix_64.c b/arch/x86/kernel/espfix_64.c index 7d7715dde901..e5ec3cafa72e 100644 --- a/arch/x86/kernel/espfix_64.c +++ b/arch/x86/kernel/espfix_64.c @@ -57,7 +57,7 @@ # error "Need more virtual address space for the ESPFIX hack" #endif -#define PGALLOC_GFP (GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL | __GFP_ZERO) /* This contains the *bottom* address of the espfix stack */ DEFINE_PER_CPU_READ_MOSTLY(unsigned long, espfix_stack); diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index a22c2b95e513..ef94620ceb8a 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -92,8 +92,7 @@ __ref void *alloc_low_pages(unsigned int num) unsigned int order; order = get_order((unsigned long)num << PAGE_SHIFT); - return (void *)__get_free_pages(GFP_ATOMIC | __GFP_NOTRACK | - __GFP_ZERO, order); + return (void *)__get_free_pages(GFP_ATOMIC | __GFP_ZERO, order); } if ((pgt_buf_end + num) > pgt_buf_top || !can_use_brk_pgt) { diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index adcea90a2046..5fa3a58b5d78 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -184,7 +184,7 @@ static __ref void *spp_getpage(void) void *ptr; if (after_bootmem) - ptr = (void *) get_zeroed_page(GFP_ATOMIC | __GFP_NOTRACK); + ptr = (void *) get_zeroed_page(GFP_ATOMIC); else ptr = alloc_bootmem_pages(PAGE_SIZE); diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 3fe68483463c..85cf12219dea 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -753,7 +753,7 @@ static int split_large_page(struct cpa_data *cpa, pte_t *kpte, if (!debug_pagealloc_enabled()) spin_unlock(&cpa_lock); - base = alloc_pages(GFP_KERNEL | __GFP_NOTRACK, 0); + base = alloc_pages(GFP_KERNEL, 0); if (!debug_pagealloc_enabled()) spin_lock(&cpa_lock); if (!base) @@ -904,7 +904,7 @@ static void unmap_pud_range(p4d_t *p4d, unsigned long start, unsigned long end) static int alloc_pte_page(pmd_t *pmd) { - pte_t *pte = (pte_t *)get_zeroed_page(GFP_KERNEL | __GFP_NOTRACK); + pte_t *pte = (pte_t *)get_zeroed_page(GFP_KERNEL); if (!pte) return -1; @@ -914,7 +914,7 @@ static int alloc_pte_page(pmd_t *pmd) static int alloc_pmd_page(pud_t *pud) { - pmd_t *pmd = (pmd_t *)get_zeroed_page(GFP_KERNEL | __GFP_NOTRACK); + pmd_t *pmd = (pmd_t *)get_zeroed_page(GFP_KERNEL); if (!pmd) return -1; @@ -1120,7 +1120,7 @@ static int populate_pgd(struct cpa_data *cpa, unsigned long addr) pgd_entry = cpa->pgd + pgd_index(addr); if (pgd_none(*pgd_entry)) { - p4d = (p4d_t *)get_zeroed_page(GFP_KERNEL | __GFP_NOTRACK); + p4d = (p4d_t *)get_zeroed_page(GFP_KERNEL); if (!p4d) return -1; @@ -1132,7 +1132,7 @@ static int populate_pgd(struct cpa_data *cpa, unsigned long addr) */ p4d = p4d_offset(pgd_entry, addr); if (p4d_none(*p4d)) { - pud = (pud_t *)get_zeroed_page(GFP_KERNEL | __GFP_NOTRACK); + pud = (pud_t *)get_zeroed_page(GFP_KERNEL); if (!pud) return -1; diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 17ebc5a978cc..96d456a94b03 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -7,7 +7,7 @@ #include #include -#define PGALLOC_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK | __GFP_ZERO) +#define PGALLOC_GFP (GFP_KERNEL_ACCOUNT | __GFP_ZERO) #ifdef CONFIG_HIGHPTE #define PGALLOC_USER_GFP __GFP_HIGHMEM diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index 9e4ee5b04b2d..6a151ce70e86 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -207,7 +207,7 @@ int __init efi_alloc_page_tables(void) if (efi_enabled(EFI_OLD_MEMMAP)) return 0; - gfp_mask = GFP_KERNEL | __GFP_NOTRACK | __GFP_ZERO; + gfp_mask = GFP_KERNEL | __GFP_ZERO; efi_pgd = (pgd_t *)__get_free_page(gfp_mask); if (!efi_pgd) return -ENOMEM; diff --git a/crypto/xor.c b/crypto/xor.c index 263af9fb45ea..bce9fe7af40a 100644 --- a/crypto/xor.c +++ b/crypto/xor.c @@ -122,12 +122,7 @@ calibrate_xor_blocks(void) goto out; } - /* - * Note: Since the memory is not actually used for _anything_ but to - * test the XOR speed, we don't really want kmemcheck to warn about - * reading uninitialized bytes here. - */ - b1 = (void *) __get_free_pages(GFP_KERNEL | __GFP_NOTRACK, 2); + b1 = (void *) __get_free_pages(GFP_KERNEL, 2); if (!b1) { printk(KERN_WARNING "xor: Yikes! No memory available.\n"); return -ENOMEM; diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h index 4bcdf00c110f..34f053a150a9 100644 --- a/include/linux/thread_info.h +++ b/include/linux/thread_info.h @@ -44,10 +44,9 @@ enum { #endif #if IS_ENABLED(CONFIG_DEBUG_STACK_USAGE) || IS_ENABLED(CONFIG_DEBUG_KMEMLEAK) -# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK | \ - __GFP_ZERO) +# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_ZERO) #else -# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT | __GFP_NOTRACK) +# define THREADINFO_GFP (GFP_KERNEL_ACCOUNT) #endif /* diff --git a/init/do_mounts.c b/init/do_mounts.c index f6d4dd764a52..7cf4f6dafd5f 100644 --- a/init/do_mounts.c +++ b/init/do_mounts.c @@ -380,8 +380,7 @@ static int __init do_mount_root(char *name, char *fs, int flags, void *data) void __init mount_block_root(char *name, int flags) { - struct page *page = alloc_page(GFP_KERNEL | - __GFP_NOTRACK_FALSE_POSITIVE); + struct page *page = alloc_page(GFP_KERNEL); char *fs_names = page_address(page); char *p; #ifdef CONFIG_BLOCK diff --git a/kernel/fork.c b/kernel/fork.c index 006dc5899a1a..4e55eedba8d6 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -469,7 +469,7 @@ void __init fork_init(void) /* create a slab on which task_structs can be allocated */ task_struct_cachep = kmem_cache_create("task_struct", arch_task_struct_size, align, - SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, NULL); + SLAB_PANIC|SLAB_ACCOUNT, NULL); #endif /* do the arch specific task caches init */ @@ -2205,18 +2205,18 @@ void __init proc_caches_init(void) sighand_cachep = kmem_cache_create("sighand_cache", sizeof(struct sighand_struct), 0, SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_TYPESAFE_BY_RCU| - SLAB_NOTRACK|SLAB_ACCOUNT, sighand_ctor); + SLAB_ACCOUNT, sighand_ctor); signal_cachep = kmem_cache_create("signal_cache", sizeof(struct signal_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT, NULL); files_cachep = kmem_cache_create("files_cache", sizeof(struct files_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT, NULL); fs_cachep = kmem_cache_create("fs_cache", sizeof(struct fs_struct), 0, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT, NULL); /* * FIXME! The "sizeof(struct mm_struct)" currently includes the @@ -2227,7 +2227,7 @@ void __init proc_caches_init(void) */ mm_cachep = kmem_cache_create("mm_struct", sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN, - SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_NOTRACK|SLAB_ACCOUNT, + SLAB_HWCACHE_ALIGN|SLAB_PANIC|SLAB_ACCOUNT, NULL); vm_area_cachep = KMEM_CACHE(vm_area_struct, SLAB_PANIC|SLAB_ACCOUNT); mmap_init(); diff --git a/kernel/signal.c b/kernel/signal.c index 8dcd8825b2de..aa1fb9f905db 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1036,8 +1036,7 @@ static int __send_signal(int sig, struct siginfo *info, struct task_struct *t, else override_rlimit = 0; - q = __sigqueue_alloc(sig, t, GFP_ATOMIC | __GFP_NOTRACK_FALSE_POSITIVE, - override_rlimit); + q = __sigqueue_alloc(sig, t, GFP_ATOMIC, override_rlimit); if (q) { list_add_tail(&q->list, &pending->list); switch ((unsigned long) info) { diff --git a/mm/kmemcheck.c b/mm/kmemcheck.c index 800d64b854ea..b3a4d61d341c 100644 --- a/mm/kmemcheck.c +++ b/mm/kmemcheck.c @@ -18,7 +18,7 @@ void kmemcheck_alloc_shadow(struct page *page, int order, gfp_t flags, int node) * With kmemcheck enabled, we need to allocate a memory area for the * shadow bits as well. */ - shadow = alloc_pages_node(node, flags | __GFP_NOTRACK, order); + shadow = alloc_pages_node(node, flags, order); if (!shadow) { if (printk_ratelimit()) pr_err("kmemcheck: failed to allocate shadow bitmap\n"); diff --git a/mm/slab.c b/mm/slab.c index c84365e9a591..183e996dde5f 100644 --- a/mm/slab.c +++ b/mm/slab.c @@ -1410,7 +1410,7 @@ static struct page *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, flags |= cachep->allocflags; - page = __alloc_pages_node(nodeid, flags | __GFP_NOTRACK, cachep->gfporder); + page = __alloc_pages_node(nodeid, flags, cachep->gfporder); if (!page) { slab_out_of_memory(cachep, flags, nodeid); return NULL; diff --git a/mm/slab.h b/mm/slab.h index e60a3d1d8f6f..ad657ffa44e5 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -141,10 +141,10 @@ static inline slab_flags_t kmem_cache_flags(unsigned long object_size, #if defined(CONFIG_SLAB) #define SLAB_CACHE_FLAGS (SLAB_MEM_SPREAD | SLAB_NOLEAKTRACE | \ SLAB_RECLAIM_ACCOUNT | SLAB_TEMPORARY | \ - SLAB_NOTRACK | SLAB_ACCOUNT) + SLAB_ACCOUNT) #elif defined(CONFIG_SLUB) #define SLAB_CACHE_FLAGS (SLAB_NOLEAKTRACE | SLAB_RECLAIM_ACCOUNT | \ - SLAB_TEMPORARY | SLAB_NOTRACK | SLAB_ACCOUNT) + SLAB_TEMPORARY | SLAB_ACCOUNT) #else #define SLAB_CACHE_FLAGS (0) #endif @@ -163,7 +163,6 @@ static inline slab_flags_t kmem_cache_flags(unsigned long object_size, SLAB_NOLEAKTRACE | \ SLAB_RECLAIM_ACCOUNT | \ SLAB_TEMPORARY | \ - SLAB_NOTRACK | \ SLAB_ACCOUNT) int __kmem_cache_shutdown(struct kmem_cache *); diff --git a/mm/slab_common.c b/mm/slab_common.c index 175e86637afd..c8cb36774ba1 100644 --- a/mm/slab_common.c +++ b/mm/slab_common.c @@ -44,7 +44,7 @@ static DECLARE_WORK(slab_caches_to_rcu_destroy_work, SLAB_FAILSLAB | SLAB_KASAN) #define SLAB_MERGE_SAME (SLAB_RECLAIM_ACCOUNT | SLAB_CACHE_DMA | \ - SLAB_NOTRACK | SLAB_ACCOUNT) + SLAB_ACCOUNT) /* * Merge control. If this is set then no merging of slab caches will occur. diff --git a/mm/slub.c b/mm/slub.c index ac3b50b9abec..91aa99b4b836 100644 --- a/mm/slub.c +++ b/mm/slub.c @@ -1436,8 +1436,6 @@ static inline struct page *alloc_slab_page(struct kmem_cache *s, struct page *page; int order = oo_order(oo); - flags |= __GFP_NOTRACK; - if (node == NUMA_NO_NODE) page = alloc_pages(flags, order); else @@ -3774,7 +3772,7 @@ static void *kmalloc_large_node(size_t size, gfp_t flags, int node) struct page *page; void *ptr = NULL; - flags |= __GFP_COMP | __GFP_NOTRACK; + flags |= __GFP_COMP; page = alloc_pages_node(node, flags, get_order(size)); if (page) ptr = page_address(page); -- cgit v1.2.3-70-g09d2 From 2d4894b5d2ae0fe1725ea7abd57b33bfbbe45492 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Wed, 15 Nov 2017 17:37:59 -0800 Subject: mm: remove cold parameter from free_hot_cold_page* Most callers users of free_hot_cold_page claim the pages being released are cache hot. The exception is the page reclaim paths where it is likely that enough pages will be freed in the near future that the per-cpu lists are going to be recycled and the cache hotness information is lost. As no one really cares about the hotness of pages being released to the allocator, just ditch the parameter. The APIs are renamed to indicate that it's no longer about hot/cold pages. It should also be less confusing as there are subtle differences between them. __free_pages drops a reference and frees a page when the refcount reaches zero. free_hot_cold_page handled pages whose refcount was already zero which is non-obvious from the name. free_unref_page should be more obvious. No performance impact is expected as the overhead is marginal. The parameter is removed simply because it is a bit stupid to have a useless parameter copied everywhere. [mgorman@techsingularity.net: add pages to head, not tail] Link: http://lkml.kernel.org/r/20171019154321.qtpzaeftoyyw4iey@techsingularity.net Link: http://lkml.kernel.org/r/20171018075952.10627-8-mgorman@techsingularity.net Signed-off-by: Mel Gorman Acked-by: Vlastimil Babka Cc: Andi Kleen Cc: Dave Chinner Cc: Dave Hansen Cc: Jan Kara Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/mm/mmu_context_book3s64.c | 2 +- arch/powerpc/mm/pgtable_64.c | 2 +- arch/sparc/mm/init_64.c | 2 +- arch/tile/mm/homecache.c | 2 +- include/linux/gfp.h | 4 ++-- include/trace/events/kmem.h | 11 ++++------- mm/page_alloc.c | 29 ++++++++++++----------------- mm/rmap.c | 2 +- mm/swap.c | 4 ++-- mm/vmscan.c | 6 +++--- 10 files changed, 28 insertions(+), 36 deletions(-) (limited to 'arch/powerpc') diff --git a/arch/powerpc/mm/mmu_context_book3s64.c b/arch/powerpc/mm/mmu_context_book3s64.c index 05e15386d4cb..a7e998158f37 100644 --- a/arch/powerpc/mm/mmu_context_book3s64.c +++ b/arch/powerpc/mm/mmu_context_book3s64.c @@ -200,7 +200,7 @@ static void destroy_pagetable_page(struct mm_struct *mm) /* We allow PTE_FRAG_NR fragments from a PTE page */ if (page_ref_sub_and_test(page, PTE_FRAG_NR - count)) { pgtable_page_dtor(page); - free_hot_cold_page(page, 0); + free_unref_page(page); } } diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c index ac0717a90ca6..1ec3aee43624 100644 --- a/arch/powerpc/mm/pgtable_64.c +++ b/arch/powerpc/mm/pgtable_64.c @@ -404,7 +404,7 @@ void pte_fragment_free(unsigned long *table, int kernel) if (put_page_testzero(page)) { if (!kernel) pgtable_page_dtor(page); - free_hot_cold_page(page, 0); + free_unref_page(page); } } diff --git a/arch/sparc/mm/init_64.c b/arch/sparc/mm/init_64.c index 051f73401793..55ba62957e64 100644 --- a/arch/sparc/mm/init_64.c +++ b/arch/sparc/mm/init_64.c @@ -2939,7 +2939,7 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, if (!page) return NULL; if (!pgtable_page_ctor(page)) { - free_hot_cold_page(page, 0); + free_unref_page(page); return NULL; } return (pte_t *) page_address(page); diff --git a/arch/tile/mm/homecache.c b/arch/tile/mm/homecache.c index b51cc28acd0a..4432f31e8479 100644 --- a/arch/tile/mm/homecache.c +++ b/arch/tile/mm/homecache.c @@ -409,7 +409,7 @@ void __homecache_free_pages(struct page *page, unsigned int order) if (put_page_testzero(page)) { homecache_change_page_home(page, order, PAGE_HOME_HASH); if (order == 0) { - free_hot_cold_page(page, false); + free_unref_page(page); } else { init_page_count(page); __free_pages(page, order); diff --git a/include/linux/gfp.h b/include/linux/gfp.h index b041f94678de..f7e62d9096fe 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -530,8 +530,8 @@ void * __meminit alloc_pages_exact_nid(int nid, size_t size, gfp_t gfp_mask); extern void __free_pages(struct page *page, unsigned int order); extern void free_pages(unsigned long addr, unsigned int order); -extern void free_hot_cold_page(struct page *page, bool cold); -extern void free_hot_cold_page_list(struct list_head *list, bool cold); +extern void free_unref_page(struct page *page); +extern void free_unref_page_list(struct list_head *list); struct page_frag_cache; extern void __page_frag_cache_drain(struct page *page, unsigned int count); diff --git a/include/trace/events/kmem.h b/include/trace/events/kmem.h index 285feeadac39..eb57e3037deb 100644 --- a/include/trace/events/kmem.h +++ b/include/trace/events/kmem.h @@ -172,24 +172,21 @@ TRACE_EVENT(mm_page_free, TRACE_EVENT(mm_page_free_batched, - TP_PROTO(struct page *page, int cold), + TP_PROTO(struct page *page), - TP_ARGS(page, cold), + TP_ARGS(page), TP_STRUCT__entry( __field( unsigned long, pfn ) - __field( int, cold ) ), TP_fast_assign( __entry->pfn = page_to_pfn(page); - __entry->cold = cold; ), - TP_printk("page=%p pfn=%lu order=0 cold=%d", + TP_printk("page=%p pfn=%lu order=0", pfn_to_page(__entry->pfn), - __entry->pfn, - __entry->cold) + __entry->pfn) ); TRACE_EVENT(mm_page_alloc, diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 6a3c4a1d513f..f265d37b3152 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2611,7 +2611,7 @@ void mark_free_pages(struct zone *zone) } #endif /* CONFIG_PM */ -static bool free_hot_cold_page_prepare(struct page *page, unsigned long pfn) +static bool free_unref_page_prepare(struct page *page, unsigned long pfn) { int migratetype; @@ -2623,8 +2623,7 @@ static bool free_hot_cold_page_prepare(struct page *page, unsigned long pfn) return true; } -static void free_hot_cold_page_commit(struct page *page, unsigned long pfn, - bool cold) +static void free_unref_page_commit(struct page *page, unsigned long pfn) { struct zone *zone = page_zone(page); struct per_cpu_pages *pcp; @@ -2649,10 +2648,7 @@ static void free_hot_cold_page_commit(struct page *page, unsigned long pfn, } pcp = &this_cpu_ptr(zone->pageset)->pcp; - if (!cold) - list_add(&page->lru, &pcp->lists[migratetype]); - else - list_add_tail(&page->lru, &pcp->lists[migratetype]); + list_add(&page->lru, &pcp->lists[migratetype]); pcp->count++; if (pcp->count >= pcp->high) { unsigned long batch = READ_ONCE(pcp->batch); @@ -2663,25 +2659,24 @@ static void free_hot_cold_page_commit(struct page *page, unsigned long pfn, /* * Free a 0-order page - * cold == true ? free a cold page : free a hot page */ -void free_hot_cold_page(struct page *page, bool cold) +void free_unref_page(struct page *page) { unsigned long flags; unsigned long pfn = page_to_pfn(page); - if (!free_hot_cold_page_prepare(page, pfn)) + if (!free_unref_page_prepare(page, pfn)) return; local_irq_save(flags); - free_hot_cold_page_commit(page, pfn, cold); + free_unref_page_commit(page, pfn); local_irq_restore(flags); } /* * Free a list of 0-order pages */ -void free_hot_cold_page_list(struct list_head *list, bool cold) +void free_unref_page_list(struct list_head *list) { struct page *page, *next; unsigned long flags, pfn; @@ -2689,7 +2684,7 @@ void free_hot_cold_page_list(struct list_head *list, bool cold) /* Prepare pages for freeing */ list_for_each_entry_safe(page, next, list, lru) { pfn = page_to_pfn(page); - if (!free_hot_cold_page_prepare(page, pfn)) + if (!free_unref_page_prepare(page, pfn)) list_del(&page->lru); set_page_private(page, pfn); } @@ -2699,8 +2694,8 @@ void free_hot_cold_page_list(struct list_head *list, bool cold) unsigned long pfn = page_private(page); set_page_private(page, 0); - trace_mm_page_free_batched(page, cold); - free_hot_cold_page_commit(page, pfn, cold); + trace_mm_page_free_batched(page); + free_unref_page_commit(page, pfn); } local_irq_restore(flags); } @@ -4301,7 +4296,7 @@ void __free_pages(struct page *page, unsigned int order) { if (put_page_testzero(page)) { if (order == 0) - free_hot_cold_page(page, false); + free_unref_page(page); else __free_pages_ok(page, order); } @@ -4359,7 +4354,7 @@ void __page_frag_cache_drain(struct page *page, unsigned int count) unsigned int order = compound_order(page); if (order == 0) - free_hot_cold_page(page, false); + free_unref_page(page); else __free_pages_ok(page, order); } diff --git a/mm/rmap.c b/mm/rmap.c index 6b5a0f219ac0..47db27f8049e 100644 --- a/mm/rmap.c +++ b/mm/rmap.c @@ -1321,7 +1321,7 @@ void page_remove_rmap(struct page *page, bool compound) * It would be tidy to reset the PageAnon mapping here, * but that might overwrite a racing page_add_anon_rmap * which increments mapcount after us but sets mapping - * before us: so leave the reset to free_hot_cold_page, + * before us: so leave the reset to free_unref_page, * and remember that it's only reliable while mapped. * Leaving it set also helps swapoff to reinstate ptes * faster for those pages still in swapcache. diff --git a/mm/swap.c b/mm/swap.c index 29cf75f1a860..b480279c760c 100644 --- a/mm/swap.c +++ b/mm/swap.c @@ -76,7 +76,7 @@ static void __page_cache_release(struct page *page) static void __put_single_page(struct page *page) { __page_cache_release(page); - free_hot_cold_page(page, false); + free_unref_page(page); } static void __put_compound_page(struct page *page) @@ -817,7 +817,7 @@ void release_pages(struct page **pages, int nr) spin_unlock_irqrestore(&locked_pgdat->lru_lock, flags); mem_cgroup_uncharge_list(&pages_to_free); - free_hot_cold_page_list(&pages_to_free, 0); + free_unref_page_list(&pages_to_free); } EXPORT_SYMBOL(release_pages); diff --git a/mm/vmscan.c b/mm/vmscan.c index 2852b8c5a917..c02c850ea349 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1349,7 +1349,7 @@ keep: mem_cgroup_uncharge_list(&free_pages); try_to_unmap_flush(); - free_hot_cold_page_list(&free_pages, true); + free_unref_page_list(&free_pages); list_splice(&ret_pages, page_list); count_vm_events(PGACTIVATE, pgactivate); @@ -1824,7 +1824,7 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec, spin_unlock_irq(&pgdat->lru_lock); mem_cgroup_uncharge_list(&page_list); - free_hot_cold_page_list(&page_list, true); + free_unref_page_list(&page_list); /* * If reclaim is isolating dirty pages under writeback, it implies @@ -2063,7 +2063,7 @@ static void shrink_active_list(unsigned long nr_to_scan, spin_unlock_irq(&pgdat->lru_lock); mem_cgroup_uncharge_list(&l_hold); - free_hot_cold_page_list(&l_hold, true); + free_unref_page_list(&l_hold); trace_mm_vmscan_lru_shrink_active(pgdat->node_id, nr_taken, nr_activate, nr_deactivate, nr_rotated, sc->priority, file); } -- cgit v1.2.3-70-g09d2