From 5877231f646bbd6d1d545e7af83aaa6e6b746013 Mon Sep 17 00:00:00 2001 From: "Aneesh Kumar K.V" Date: Fri, 6 Dec 2013 00:08:22 +0530 Subject: mm: Move change_prot_numa outside CONFIG_ARCH_USES_NUMA_PROT_NONE change_prot_numa should work even if _PAGE_NUMA != _PAGE_PROTNONE. On archs like ppc64 that don't use _PAGE_PROTNONE and also have a separate page table outside linux pagetable, we just need to make sure that when calling change_prot_numa we flush the hardware page table entry so that next page access result in a numa fault. We still need to make sure we use the numa faulting logic only when CONFIG_NUMA_BALANCING is set. This implies the migrate-on-fault (Lazy migration) via mbind will only work if CONFIG_NUMA_BALANCING is set. Signed-off-by: Aneesh Kumar K.V Reviewed-by: Rik van Riel Acked-by: Mel Gorman Signed-off-by: Benjamin Herrenschmidt --- include/linux/mm.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 1cedd000cf29..a7b4e310bf42 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1842,7 +1842,7 @@ static inline pgprot_t vm_get_page_prot(unsigned long vm_flags) } #endif -#ifdef CONFIG_ARCH_USES_NUMA_PROT_NONE +#ifdef CONFIG_NUMA_BALANCING unsigned long change_prot_numa(struct vm_area_struct *vma, unsigned long start, unsigned long end); #endif -- cgit v1.2.3 From f92f455f67fef27929e6043499414605b0c94872 Mon Sep 17 00:00:00 2001 From: Geert Uytterhoeven Date: Tue, 21 Jan 2014 15:48:47 -0800 Subject: mm: Make {,set}page_address() static inline if WANT_PAGE_VIRTUAL {,set}page_address() are macros if WANT_PAGE_VIRTUAL. If !WANT_PAGE_VIRTUAL, they're plain C functions. If someone calls them with a void *, this pointer is auto-converted to struct page * if !WANT_PAGE_VIRTUAL, but causes a build failure on architectures using WANT_PAGE_VIRTUAL (arc, m68k and sparc64): drivers/md/bcache/bset.c: In function `__btree_sort': drivers/md/bcache/bset.c:1190: warning: dereferencing `void *' pointer drivers/md/bcache/bset.c:1190: error: request for member `virtual' in something not a structure or union Convert them to static inline functions to fix this. There are already plenty of users of struct page members inside , so there's no reason to keep them as macros. Signed-off-by: Geert Uytterhoeven Acked-by: Michael S. Tsirkin Tested-by: Guenter Roeck Tested-by: David Rientjes Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 35527173cf50..9fac6dd69b11 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -846,11 +846,14 @@ static __always_inline void *lowmem_page_address(const struct page *page) #endif #if defined(WANT_PAGE_VIRTUAL) -#define page_address(page) ((page)->virtual) -#define set_page_address(page, address) \ - do { \ - (page)->virtual = (address); \ - } while(0) +static inline void *page_address(const struct page *page) +{ + return page->virtual; +} +static inline void set_page_address(struct page *page, void *address) +{ + page->virtual = address; +} #define page_address_init() do { } while(0) #endif -- cgit v1.2.3 From 44518d2b32646e37b4b7a0813bbbe98dc21c7f8f Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 21 Jan 2014 15:48:54 -0800 Subject: mm: tail page refcounting optimization for slab and hugetlbfs This skips the _mapcount mangling for slab and hugetlbfs pages. The main trouble in doing this is to guarantee that PageSlab and PageHeadHuge remains constant for all get_page/put_page run on the tail of slab or hugetlbfs compound pages. Otherwise if they're set during get_page but not set during put_page, the _mapcount of the tail page would underflow. PageHeadHuge will remain true until the compound page is released and enters the buddy allocator so it won't risk to change even if the tail page is the last reference left on the page. PG_slab instead is cleared before the slab frees the head page with put_page, so if the tail pin is released after the slab freed the page, we would have a problem. But in the slab case the tail pin cannot be the last reference left on the page. This is because the slab code is free to reuse the compound page after a kfree/kmem_cache_free without having to check if there's any tail pin left. In turn all tail pins must be always released while the head is still pinned by the slab code and so we know PG_slab will be still set too. Signed-off-by: Andrea Arcangeli Reviewed-by: Khalid Aziz Cc: Pravin Shelar Cc: Greg Kroah-Hartman Cc: Ben Hutchings Cc: Christoph Lameter Cc: Johannes Weiner Cc: Mel Gorman Cc: Rik van Riel Cc: Andi Kleen Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 32 +++++++++++++++++++++++++++++++- 1 file changed, 31 insertions(+), 1 deletion(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 9fac6dd69b11..f95c71b7c1fd 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -414,15 +414,45 @@ static inline int page_count(struct page *page) return atomic_read(&compound_head(page)->_count); } +#ifdef CONFIG_HUGETLB_PAGE +extern int PageHeadHuge(struct page *page_head); +#else /* CONFIG_HUGETLB_PAGE */ +static inline int PageHeadHuge(struct page *page_head) +{ + return 0; +} +#endif /* CONFIG_HUGETLB_PAGE */ + +static inline bool __compound_tail_refcounted(struct page *page) +{ + return !PageSlab(page) && !PageHeadHuge(page); +} + +/* + * This takes a head page as parameter and tells if the + * tail page reference counting can be skipped. + * + * For this to be safe, PageSlab and PageHeadHuge must remain true on + * any given page where they return true here, until all tail pins + * have been released. + */ +static inline bool compound_tail_refcounted(struct page *page) +{ + VM_BUG_ON(!PageHead(page)); + return __compound_tail_refcounted(page); +} + static inline void get_huge_page_tail(struct page *page) { /* * __split_huge_page_refcount() cannot run * from under us. + * In turn no need of compound_trans_head here. */ VM_BUG_ON(page_mapcount(page) < 0); VM_BUG_ON(atomic_read(&page->_count) != 0); - atomic_inc(&page->_mapcount); + if (compound_tail_refcounted(compound_head(page))) + atomic_inc(&page->_mapcount); } extern bool __get_page_tail(struct page *page); -- cgit v1.2.3 From 5eaf1a9e233d61438377f57facb167f8208ba9fd Mon Sep 17 00:00:00 2001 From: Oleg Nesterov Date: Tue, 21 Jan 2014 15:49:04 -0800 Subject: mm: thp: turn compound_head() into BUG_ON(!PageTail) in get_huge_page_tail() get_huge_page_tail()->compound_head() looks confusing. Every caller must check PageTail(page), otherwise atomic_inc(&page->_mapcount) is simply wrong if this page is compound-trans-head. Signed-off-by: Oleg Nesterov Cc: Thomas Gleixner Cc: Dave Jones Cc: Darren Hart Cc: Peter Zijlstra Cc: Mel Gorman Acked-by: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 7 +++---- 1 file changed, 3 insertions(+), 4 deletions(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index f95c71b7c1fd..58202c26c559 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -445,13 +445,12 @@ static inline bool compound_tail_refcounted(struct page *page) static inline void get_huge_page_tail(struct page *page) { /* - * __split_huge_page_refcount() cannot run - * from under us. - * In turn no need of compound_trans_head here. + * __split_huge_page_refcount() cannot run from under us. */ + VM_BUG_ON(!PageTail(page)); VM_BUG_ON(page_mapcount(page) < 0); VM_BUG_ON(atomic_read(&page->_count) != 0); - if (compound_tail_refcounted(compound_head(page))) + if (compound_tail_refcounted(page->first_page)) atomic_inc(&page->_mapcount); } -- cgit v1.2.3 From b35f1819acd9243a3ff7ad25b1fa8bd6bfe80fb2 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 21 Jan 2014 15:49:07 -0800 Subject: mm: create a separate slab for page->ptl allocation If DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC are enabled spinlock_t on x86_64 is 72 bytes. For page->ptl they will be allocated from kmalloc-96 slab, so we loose 24 on each. An average system can easily allocate few tens thousands of page->ptl and overhead is significant. Let's create a separate slab for page->ptl allocation to solve this. To make sure that it really works this time, some numbers from my test machine (just booted, no load): Before: # grep '^\(kmalloc-96\|page->ptl\)' /proc/slabinfo kmalloc-96 31987 32190 128 30 1 : tunables 120 60 8 : slabdata 1073 1073 92 After: # grep '^\(kmalloc-96\|page->ptl\)' /proc/slabinfo page->ptl 27516 28143 72 53 1 : tunables 120 60 8 : slabdata 531 531 9 kmalloc-96 3853 5280 128 30 1 : tunables 120 60 8 : slabdata 176 176 0 Note that the patch is useful not only for debug case, but also for PREEMPT_RT, where spinlock_t is always bloated. Signed-off-by: Kirill A. Shutemov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 58202c26c559..fc4415256ec3 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1350,6 +1350,7 @@ static inline pmd_t *pmd_alloc(struct mm_struct *mm, pud_t *pud, unsigned long a #if USE_SPLIT_PTE_PTLOCKS #if ALLOC_SPLIT_PTLOCKS +void __init ptlock_cache_init(void); extern bool ptlock_alloc(struct page *page); extern void ptlock_free(struct page *page); @@ -1358,6 +1359,10 @@ static inline spinlock_t *ptlock_ptr(struct page *page) return page->ptl; } #else /* ALLOC_SPLIT_PTLOCKS */ +static inline void ptlock_cache_init(void) +{ +} + static inline bool ptlock_alloc(struct page *page) { return true; @@ -1410,10 +1415,17 @@ static inline spinlock_t *pte_lockptr(struct mm_struct *mm, pmd_t *pmd) { return &mm->page_table_lock; } +static inline void ptlock_cache_init(void) {} static inline bool ptlock_init(struct page *page) { return true; } static inline void pte_lock_deinit(struct page *page) {} #endif /* USE_SPLIT_PTE_PTLOCKS */ +static inline void pgtable_init(void) +{ + ptlock_cache_init(); + pgtable_cache_init(); +} + static inline bool pgtable_page_ctor(struct page *page) { inc_zone_page_state(page, NR_PAGETABLE); -- cgit v1.2.3 From aec6a8889a98a0cd58357cd0937a25189908f191 Mon Sep 17 00:00:00 2001 From: Mel Gorman Date: Tue, 21 Jan 2014 15:49:13 -0800 Subject: mm, show_mem: remove SHOW_MEM_FILTER_PAGE_COUNT Commit 4b59e6c47309 ("mm, show_mem: suppress page counts in non-blockable contexts") introduced SHOW_MEM_FILTER_PAGE_COUNT to suppress PFN walks on large memory machines. Commit c78e93630d15 ("mm: do not walk all of system memory during show_mem") avoided a PFN walk in the generic show_mem helper which removes the requirement for SHOW_MEM_FILTER_PAGE_COUNT in that case. This patch removes PFN walkers from the arch-specific implementations that report on a per-node or per-zone granularity. ARM and unicore32 still do a PFN walk as they report memory usage on each bank which is a much finer granularity where the debugging information may still be of use. As the remaining arches doing PFN walks have relatively small amounts of memory, this patch simply removes SHOW_MEM_FILTER_PAGE_COUNT. [akpm@linux-foundation.org: fix parisc] Signed-off-by: Mel Gorman Acked-by: David Rientjes Cc: Tony Luck Cc: Russell King Cc: James Bottomley Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 1 - 1 file changed, 1 deletion(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index fc4415256ec3..4c0c01afc19b 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1016,7 +1016,6 @@ extern void pagefault_out_of_memory(void); * various contexts. */ #define SHOW_MEM_FILTER_NODES (0x0001u) /* disallowed nodes */ -#define SHOW_MEM_FILTER_PAGE_COUNT (0x0002u) /* page type count */ extern void show_free_areas(unsigned int flags); extern bool skip_free_areas_node(unsigned int flags, int nid); -- cgit v1.2.3 From 49f0ce5f92321cdcf741e35f385669a421013cb7 Mon Sep 17 00:00:00 2001 From: Jerome Marchand Date: Tue, 21 Jan 2014 15:49:14 -0800 Subject: mm: add overcommit_kbytes sysctl variable Some applications that run on HPC clusters are designed around the availability of RAM and the overcommit ratio is fine tuned to get the maximum usage of memory without swapping. With growing memory, the 1%-of-all-RAM grain provided by overcommit_ratio has become too coarse for these workload (on a 2TB machine it represents no less than 20GB). This patch adds the new overcommit_kbytes sysctl variable that allow a much finer grain. [akpm@linux-foundation.org: coding-style fixes] [akpm@linux-foundation.org: fix nommu build] Signed-off-by: Jerome Marchand Cc: Dave Hansen Cc: Alan Cox Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 9 +++++++++ 1 file changed, 9 insertions(+) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 4c0c01afc19b..a512dd836931 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -57,6 +57,15 @@ extern int sysctl_legacy_va_layout; extern unsigned long sysctl_user_reserve_kbytes; extern unsigned long sysctl_admin_reserve_kbytes; +extern int sysctl_overcommit_memory; +extern int sysctl_overcommit_ratio; +extern unsigned long sysctl_overcommit_kbytes; + +extern int overcommit_ratio_handler(struct ctl_table *, int, void __user *, + size_t *, loff_t *); +extern int overcommit_kbytes_handler(struct ctl_table *, int, void __user *, + size_t *, loff_t *); + #define nth_page(page,n) pfn_to_page(page_to_pfn((page)) + (n)) /* to align the pointer to the (next) page boundary */ -- cgit v1.2.3 From f0b791a34cb3cffd2bbc3ca4365c9b719fa2c9f3 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Thu, 23 Jan 2014 15:52:49 -0800 Subject: mm: print more details for bad_page() bad_page() is cool in that it prints out a bunch of data about the page. But, I can never remember which page flags are good and which are bad, or whether ->index or ->mapping is required to be NULL. This patch allows bad/dump_page() callers to specify a string about why they are dumping the page and adds explanation strings to a number of places. It also adds a 'bad_flags' argument to bad_page(), which it then dumps out separately from the flags which are actually set. This way, the messages will show specifically why the page was bad, *specifically* which flags it is complaining about, if it was a page flag combination which was the problem. [akpm@linux-foundation.org: switch to pr_alert] Signed-off-by: Dave Hansen Reviewed-by: Christoph Lameter Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index a512dd836931..03bbcb84d96e 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -2029,7 +2029,9 @@ extern void shake_page(struct page *p, int access); extern atomic_long_t num_poisoned_pages; extern int soft_offline_page(struct page *page, int flags); -extern void dump_page(struct page *page); +extern void dump_page(struct page *page, char *reason); +extern void dump_page_badflags(struct page *page, char *reason, + unsigned long badflags); #if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) extern void clear_huge_page(struct page *page, -- cgit v1.2.3 From 309381feaee564281c3d9e90fbca8963bb7428ad Mon Sep 17 00:00:00 2001 From: Sasha Levin Date: Thu, 23 Jan 2014 15:52:54 -0800 Subject: mm: dump page when hitting a VM_BUG_ON using VM_BUG_ON_PAGE Most of the VM_BUG_ON assertions are performed on a page. Usually, when one of these assertions fails we'll get a BUG_ON with a call stack and the registers. I've recently noticed based on the requests to add a small piece of code that dumps the page to various VM_BUG_ON sites that the page dump is quite useful to people debugging issues in mm. This patch adds a VM_BUG_ON_PAGE(cond, page) which beyond doing what VM_BUG_ON() does, also dumps the page before executing the actual BUG_ON. [akpm@linux-foundation.org: fix up includes] Signed-off-by: Sasha Levin Cc: "Kirill A. Shutemov" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mm.h | 29 +++++++++++++---------------- 1 file changed, 13 insertions(+), 16 deletions(-) (limited to 'include/linux/mm.h') diff --git a/include/linux/mm.h b/include/linux/mm.h index 03bbcb84d96e..d9992fc128ca 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -5,6 +5,7 @@ #ifdef __KERNEL__ +#include #include #include #include @@ -303,7 +304,7 @@ static inline int get_freepage_migratetype(struct page *page) */ static inline int put_page_testzero(struct page *page) { - VM_BUG_ON(atomic_read(&page->_count) == 0); + VM_BUG_ON_PAGE(atomic_read(&page->_count) == 0, page); return atomic_dec_and_test(&page->_count); } @@ -364,7 +365,7 @@ static inline int is_vmalloc_or_module_addr(const void *x) static inline void compound_lock(struct page *page) { #ifdef CONFIG_TRANSPARENT_HUGEPAGE - VM_BUG_ON(PageSlab(page)); + VM_BUG_ON_PAGE(PageSlab(page), page); bit_spin_lock(PG_compound_lock, &page->flags); #endif } @@ -372,7 +373,7 @@ static inline void compound_lock(struct page *page) static inline void compound_unlock(struct page *page) { #ifdef CONFIG_TRANSPARENT_HUGEPAGE - VM_BUG_ON(PageSlab(page)); + VM_BUG_ON_PAGE(PageSlab(page), page); bit_spin_unlock(PG_compound_lock, &page->flags); #endif } @@ -447,7 +448,7 @@ static inline bool __compound_tail_refcounted(struct page *page) */ static inline bool compound_tail_refcounted(struct page *page) { - VM_BUG_ON(!PageHead(page)); + VM_BUG_ON_PAGE(!PageHead(page), page); return __compound_tail_refcounted(page); } @@ -456,9 +457,9 @@ static inline void get_huge_page_tail(struct page *page) /* * __split_huge_page_refcount() cannot run from under us. */ - VM_BUG_ON(!PageTail(page)); - VM_BUG_ON(page_mapcount(page) < 0); - VM_BUG_ON(atomic_read(&page->_count) != 0); + VM_BUG_ON_PAGE(!PageTail(page), page); + VM_BUG_ON_PAGE(page_mapcount(page) < 0, page); + VM_BUG_ON_PAGE(atomic_read(&page->_count) != 0, page); if (compound_tail_refcounted(page->first_page)) atomic_inc(&page->_mapcount); } @@ -474,7 +475,7 @@ static inline void get_page(struct page *page) * Getting a normal page or the head of a compound page * requires to already have an elevated page->_count. */ - VM_BUG_ON(atomic_read(&page->_count) <= 0); + VM_BUG_ON_PAGE(atomic_read(&page->_count) <= 0, page); atomic_inc(&page->_count); } @@ -511,13 +512,13 @@ static inline int PageBuddy(struct page *page) static inline void __SetPageBuddy(struct page *page) { - VM_BUG_ON(atomic_read(&page->_mapcount) != -1); + VM_BUG_ON_PAGE(atomic_read(&page->_mapcount) != -1, page); atomic_set(&page->_mapcount, PAGE_BUDDY_MAPCOUNT_VALUE); } static inline void __ClearPageBuddy(struct page *page) { - VM_BUG_ON(!PageBuddy(page)); + VM_BUG_ON_PAGE(!PageBuddy(page), page); atomic_set(&page->_mapcount, -1); } @@ -1401,7 +1402,7 @@ static inline bool ptlock_init(struct page *page) * slab code uses page->slab_cache and page->first_page (for tail * pages), which share storage with page->ptl. */ - VM_BUG_ON(*(unsigned long *)&page->ptl); + VM_BUG_ON_PAGE(*(unsigned long *)&page->ptl, page); if (!ptlock_alloc(page)) return false; spin_lock_init(ptlock_ptr(page)); @@ -1492,7 +1493,7 @@ static inline bool pgtable_pmd_page_ctor(struct page *page) static inline void pgtable_pmd_page_dtor(struct page *page) { #ifdef CONFIG_TRANSPARENT_HUGEPAGE - VM_BUG_ON(page->pmd_huge_pte); + VM_BUG_ON_PAGE(page->pmd_huge_pte, page); #endif ptlock_free(page); } @@ -2029,10 +2030,6 @@ extern void shake_page(struct page *p, int access); extern atomic_long_t num_poisoned_pages; extern int soft_offline_page(struct page *page, int flags); -extern void dump_page(struct page *page, char *reason); -extern void dump_page_badflags(struct page *page, char *reason, - unsigned long badflags); - #if defined(CONFIG_TRANSPARENT_HUGEPAGE) || defined(CONFIG_HUGETLBFS) extern void clear_huge_page(struct page *page, unsigned long addr, -- cgit v1.2.3