From a3d0a918502cc73af4f60da2cc4c5cac5573f183 Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 2 Feb 2016 16:57:08 -0800 Subject: thp: make split_queue per-node Andrea Arcangeli suggested to make split queue per-node to improve scalability. Let's do it. Signed-off-by: Kirill A. Shutemov Suggested-by: Andrea Arcangeli Reviewed-by: Andrea Arcangeli Cc: Hugh Dickins Cc: Dave Hansen Cc: Mel Gorman Cc: Rik van Riel Cc: Vlastimil Babka Cc: "Aneesh Kumar K.V" Cc: Johannes Weiner Cc: Michal Hocko Cc: Jerome Marchand Cc: Sasha Levin Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/mmzone.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 33bb1b19273e..7b6c2cfee390 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -682,6 +682,12 @@ typedef struct pglist_data { */ unsigned long first_deferred_pfn; #endif /* CONFIG_DEFERRED_STRUCT_PAGE_INIT */ + +#ifdef CONFIG_TRANSPARENT_HUGEPAGE + spinlock_t split_queue_lock; + struct list_head split_queue; + unsigned long split_queue_len; +#endif } pg_data_t; #define node_present_pages(nid) (NODE_DATA(nid)->node_present_pages) -- cgit v1.2.3 From 65376df582174ffcec9e6471bf5b0dd79ba05e4a Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 2 Feb 2016 16:57:29 -0800 Subject: proc: revert /proc//maps [stack:TID] annotation Commit b76437579d13 ("procfs: mark thread stack correctly in proc//maps") added [stack:TID] annotation to /proc//maps. Finding the task of a stack VMA requires walking the entire thread list, turning this into quadratic behavior: a thousand threads means a thousand stacks, so the rendering of /proc//maps needs to look at a million combinations. The cost is not in proportion to the usefulness as described in the patch. Drop the [stack:TID] annotation to make /proc//maps (and /proc//numa_maps) usable again for higher thread counts. The [stack] annotation inside /proc//task//maps is retained, as identifying the stack VMA there is an O(1) operation. Siddesh said: "The end users needed a way to identify thread stacks programmatically and there wasn't a way to do that. I'm afraid I no longer remember (or have access to the resources that would aid my memory since I changed employers) the details of their requirement. However, I did do this on my own time because I thought it was an interesting project for me and nobody really gave any feedback then as to its utility, so as far as I am concerned you could roll back the main thread maps information since the information is available in the thread-specific files" Signed-off-by: Johannes Weiner Cc: "Kirill A. Shutemov" Cc: Siddhesh Poyarekar Cc: Shaohua Li Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/proc.txt | 9 ++---- fs/proc/task_mmu.c | 66 +++++++++++++------------------------- fs/proc/task_nommu.c | 49 ++++++++++++---------------- include/linux/mm.h | 3 +- mm/util.c | 27 +--------------- 5 files changed, 48 insertions(+), 106 deletions(-) (limited to 'include') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index fde9fd06fa98..eaebf27539f5 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -356,7 +356,7 @@ address perms offset dev inode pathname a7cb1000-a7cb2000 ---p 00000000 00:00 0 a7cb2000-a7eb2000 rw-p 00000000 00:00 0 a7eb2000-a7eb3000 ---p 00000000 00:00 0 -a7eb3000-a7ed5000 rw-p 00000000 00:00 0 [stack:1001] +a7eb3000-a7ed5000 rw-p 00000000 00:00 0 a7ed5000-a8008000 r-xp 00000000 03:00 4222 /lib/libc.so.6 a8008000-a800a000 r--p 00133000 03:00 4222 /lib/libc.so.6 a800a000-a800b000 rw-p 00135000 03:00 4222 /lib/libc.so.6 @@ -388,7 +388,6 @@ is not associated with a file: [heap] = the heap of the program [stack] = the stack of the main process - [stack:1001] = the stack of the thread with tid 1001 [vdso] = the "virtual dynamic shared object", the kernel system call handler @@ -396,10 +395,8 @@ is not associated with a file: The /proc/PID/task/TID/maps is a view of the virtual memory from the viewpoint of the individual tasks of a process. In this file you will see a mapping marked -as [stack] if that task sees it as a stack. This is a key difference from the -content of /proc/PID/maps, where you will see all mappings that are being used -as stack by all of those tasks. Hence, for the example above, the task-level -map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this: +as [stack] if that task sees it as a stack. Hence, for the example above, the +task-level map, i.e. /proc/PID/task/TID/maps for thread 1001 will look like this: 08048000-08049000 r-xp 00000000 03:00 8312 /opt/test 08049000-0804a000 rw-p 00001000 03:00 8312 /opt/test diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 4a0c31f904a6..fa95ab2d3674 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -259,23 +259,29 @@ static int do_maps_open(struct inode *inode, struct file *file, sizeof(struct proc_maps_private)); } -static pid_t pid_of_stack(struct proc_maps_private *priv, - struct vm_area_struct *vma, bool is_pid) +/* + * Indicate if the VMA is a stack for the given task; for + * /proc/PID/maps that is the stack of the main task. + */ +static int is_stack(struct proc_maps_private *priv, + struct vm_area_struct *vma, int is_pid) { - struct inode *inode = priv->inode; - struct task_struct *task; - pid_t ret = 0; + int stack = 0; + + if (is_pid) { + stack = vma->vm_start <= vma->vm_mm->start_stack && + vma->vm_end >= vma->vm_mm->start_stack; + } else { + struct inode *inode = priv->inode; + struct task_struct *task; - rcu_read_lock(); - task = pid_task(proc_pid(inode), PIDTYPE_PID); - if (task) { - task = task_of_stack(task, vma, is_pid); + rcu_read_lock(); + task = pid_task(proc_pid(inode), PIDTYPE_PID); if (task) - ret = task_pid_nr_ns(task, inode->i_sb->s_fs_info); + stack = vma_is_stack_for_task(vma, task); + rcu_read_unlock(); } - rcu_read_unlock(); - - return ret; + return stack; } static void @@ -335,8 +341,6 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid) name = arch_vma_name(vma); if (!name) { - pid_t tid; - if (!mm) { name = "[vdso]"; goto done; @@ -348,21 +352,8 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid) goto done; } - tid = pid_of_stack(priv, vma, is_pid); - if (tid != 0) { - /* - * Thread stack in /proc/PID/task/TID/maps or - * the main process stack. - */ - if (!is_pid || (vma->vm_start <= mm->start_stack && - vma->vm_end >= mm->start_stack)) { - name = "[stack]"; - } else { - /* Thread stack in /proc/PID/maps */ - seq_pad(m, ' '); - seq_printf(m, "[stack:%d]", tid); - } - } + if (is_stack(priv, vma, is_pid)) + name = "[stack]"; } done: @@ -1618,19 +1609,8 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid) seq_file_path(m, file, "\n\t= "); } else if (vma->vm_start <= mm->brk && vma->vm_end >= mm->start_brk) { seq_puts(m, " heap"); - } else { - pid_t tid = pid_of_stack(proc_priv, vma, is_pid); - if (tid != 0) { - /* - * Thread stack in /proc/PID/task/TID/maps or - * the main process stack. - */ - if (!is_pid || (vma->vm_start <= mm->start_stack && - vma->vm_end >= mm->start_stack)) - seq_puts(m, " stack"); - else - seq_printf(m, " stack:%d", tid); - } + } else if (is_stack(proc_priv, vma, is_pid)) { + seq_puts(m, " stack"); } if (is_vm_hugetlb_page(vma)) diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index e0d64c92e4f6..faacb0c0d857 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -123,23 +123,26 @@ unsigned long task_statm(struct mm_struct *mm, return size; } -static pid_t pid_of_stack(struct proc_maps_private *priv, - struct vm_area_struct *vma, bool is_pid) +static int is_stack(struct proc_maps_private *priv, + struct vm_area_struct *vma, int is_pid) { - struct inode *inode = priv->inode; - struct task_struct *task; - pid_t ret = 0; - - rcu_read_lock(); - task = pid_task(proc_pid(inode), PIDTYPE_PID); - if (task) { - task = task_of_stack(task, vma, is_pid); + struct mm_struct *mm = vma->vm_mm; + int stack = 0; + + if (is_pid) { + stack = vma->vm_start <= mm->start_stack && + vma->vm_end >= mm->start_stack; + } else { + struct inode *inode = priv->inode; + struct task_struct *task; + + rcu_read_lock(); + task = pid_task(proc_pid(inode), PIDTYPE_PID); if (task) - ret = task_pid_nr_ns(task, inode->i_sb->s_fs_info); + stack = vma_is_stack_for_task(vma, task); + rcu_read_unlock(); } - rcu_read_unlock(); - - return ret; + return stack; } /* @@ -181,21 +184,9 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma, if (file) { seq_pad(m, ' '); seq_file_path(m, file, ""); - } else if (mm) { - pid_t tid = pid_of_stack(priv, vma, is_pid); - - if (tid != 0) { - seq_pad(m, ' '); - /* - * Thread stack in /proc/PID/task/TID/maps or - * the main process stack. - */ - if (!is_pid || (vma->vm_start <= mm->start_stack && - vma->vm_end >= mm->start_stack)) - seq_printf(m, "[stack]"); - else - seq_printf(m, "[stack:%d]", tid); - } + } else if (mm && is_stack(priv, vma, is_pid)) { + seq_pad(m, ' '); + seq_printf(m, "[stack]"); } seq_putc(m, '\n'); diff --git a/include/linux/mm.h b/include/linux/mm.h index f1cd22f2df1a..0b50d7848e3a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1341,8 +1341,7 @@ static inline int stack_guard_page_end(struct vm_area_struct *vma, !vma_growsup(vma->vm_next, addr); } -extern struct task_struct *task_of_stack(struct task_struct *task, - struct vm_area_struct *vma, bool in_group); +int vma_is_stack_for_task(struct vm_area_struct *vma, struct task_struct *t); extern unsigned long move_page_tables(struct vm_area_struct *vma, unsigned long old_addr, struct vm_area_struct *new_vma, diff --git a/mm/util.c b/mm/util.c index c108a6542d05..4fb14ca5a419 100644 --- a/mm/util.c +++ b/mm/util.c @@ -230,36 +230,11 @@ void __vma_link_list(struct mm_struct *mm, struct vm_area_struct *vma, } /* Check if the vma is being used as a stack by this task */ -static int vm_is_stack_for_task(struct task_struct *t, - struct vm_area_struct *vma) +int vma_is_stack_for_task(struct vm_area_struct *vma, struct task_struct *t) { return (vma->vm_start <= KSTK_ESP(t) && vma->vm_end >= KSTK_ESP(t)); } -/* - * Check if the vma is being used as a stack. - * If is_group is non-zero, check in the entire thread group or else - * just check in the current task. Returns the task_struct of the task - * that the vma is stack for. Must be called under rcu_read_lock(). - */ -struct task_struct *task_of_stack(struct task_struct *task, - struct vm_area_struct *vma, bool in_group) -{ - if (vm_is_stack_for_task(task, vma)) - return task; - - if (in_group) { - struct task_struct *t; - - for_each_thread(task, t) { - if (vm_is_stack_for_task(t, vma)) - return t; - } - } - - return NULL; -} - #if defined(CONFIG_MMU) && !defined(HAVE_ARCH_PICK_MMAP_LAYOUT) void arch_pick_mmap_layout(struct mm_struct *mm) { -- cgit v1.2.3 From c792e8240338e41eda4d06a3a71a7bb7af4e6156 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 2 Feb 2016 16:57:38 -0800 Subject: mm: memcontrol: drop superfluous entry in the per-memcg stats array MEM_CGROUP_STAT_NSTATS is just a delimiter for cgroup1 statistics, not an actual array entry. Reuse it for the first cgroup2 stat entry, like in the event array. Fixes: b2807f07f4f8 ("mm: memcontrol: add "sock" to cgroup2 memory.stat") Signed-off-by: Johannes Weiner Cc: Vladimir Davydov Cc: Michal Hocko Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/memcontrol.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include') diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h index 9ae48d4aeb5e..792c8981e633 100644 --- a/include/linux/memcontrol.h +++ b/include/linux/memcontrol.h @@ -51,7 +51,7 @@ enum mem_cgroup_stat_index { MEM_CGROUP_STAT_SWAP, /* # of pages, swapped out */ MEM_CGROUP_STAT_NSTATS, /* default hierarchy stats */ - MEMCG_SOCK, + MEMCG_SOCK = MEM_CGROUP_STAT_NSTATS, MEMCG_NR_STAT, }; -- cgit v1.2.3 From 30bdbb78009e67767983085e302bec6d97afc679 Mon Sep 17 00:00:00 2001 From: Konstantin Khlebnikov Date: Tue, 2 Feb 2016 16:57:46 -0800 Subject: mm: polish virtual memory accounting * add VM_STACK as alias for VM_GROWSUP/DOWN depending on architecture * always account VMAs with flag VM_STACK as stack (as it was before) * cleanup classifying helpers * update comments and documentation Signed-off-by: Konstantin Khlebnikov Tested-by: Sudip Mukherjee Cc: Cyrill Gorcunov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/filesystems/proc.txt | 4 ++-- include/linux/mm.h | 6 ++++-- include/linux/mm_types.h | 6 +++--- mm/internal.h | 23 +++++++++++++++++++---- 4 files changed, 28 insertions(+), 11 deletions(-) (limited to 'include') diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index eaebf27539f5..843b045b4069 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -240,8 +240,8 @@ Table 1-2: Contents of the status files (as of 4.1) RssFile size of resident file mappings RssShmem size of resident shmem memory (includes SysV shm, mapping of tmpfs and shared anonymous mappings) - VmData size of data, stack, and text segments - VmStk size of data, stack, and text segments + VmData size of private data segments + VmStk size of stack segments VmExe size of text segment VmLib size of shared library code VmPTE size of page table entries diff --git a/include/linux/mm.h b/include/linux/mm.h index 0b50d7848e3a..516e14944339 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -201,11 +201,13 @@ extern unsigned int kobjsize(const void *objp); #endif #ifdef CONFIG_STACK_GROWSUP -#define VM_STACK_FLAGS (VM_GROWSUP | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT) +#define VM_STACK VM_GROWSUP #else -#define VM_STACK_FLAGS (VM_GROWSDOWN | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT) +#define VM_STACK VM_GROWSDOWN #endif +#define VM_STACK_FLAGS (VM_STACK | VM_STACK_DEFAULT_FLAGS | VM_ACCOUNT) + /* * Special vmas that are non-mergable, non-mlock()able. * Note: mm/huge_memory.c VM_NO_THP depends on this definition. diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h index d3ebb9d21a53..624b78b848b8 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -424,9 +424,9 @@ struct mm_struct { unsigned long total_vm; /* Total pages mapped */ unsigned long locked_vm; /* Pages that have PG_mlocked set */ unsigned long pinned_vm; /* Refcount permanently increased */ - unsigned long data_vm; /* VM_WRITE & ~VM_SHARED/GROWSDOWN */ - unsigned long exec_vm; /* VM_EXEC & ~VM_WRITE */ - unsigned long stack_vm; /* VM_GROWSUP/DOWN */ + unsigned long data_vm; /* VM_WRITE & ~VM_SHARED & ~VM_STACK */ + unsigned long exec_vm; /* VM_EXEC & ~VM_WRITE & ~VM_STACK */ + unsigned long stack_vm; /* VM_STACK */ unsigned long def_flags; unsigned long start_code, end_code, start_data, end_data; unsigned long start_brk, brk, start_stack; diff --git a/mm/internal.h b/mm/internal.h index 6e976302ddd8..a38a21ebddb4 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -216,20 +216,35 @@ static inline bool is_cow_mapping(vm_flags_t flags) return (flags & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE; } +/* + * These three helpers classifies VMAs for virtual memory accounting. + */ + +/* + * Executable code area - executable, not writable, not stack + */ static inline bool is_exec_mapping(vm_flags_t flags) { - return (flags & (VM_EXEC | VM_WRITE)) == VM_EXEC; + return (flags & (VM_EXEC | VM_WRITE | VM_STACK)) == VM_EXEC; } +/* + * Stack area - atomatically grows in one direction + * + * VM_GROWSUP / VM_GROWSDOWN VMAs are always private anonymous: + * do_mmap() forbids all other combinations. + */ static inline bool is_stack_mapping(vm_flags_t flags) { - return (flags & (VM_STACK_FLAGS & (VM_GROWSUP | VM_GROWSDOWN))) != 0; + return (flags & VM_STACK) == VM_STACK; } +/* + * Data area - private, writable, not stack + */ static inline bool is_data_mapping(vm_flags_t flags) { - return (flags & ((VM_STACK_FLAGS & (VM_GROWSUP | VM_GROWSDOWN)) | - VM_WRITE | VM_SHARED)) == VM_WRITE; + return (flags & (VM_WRITE | VM_SHARED | VM_STACK)) == VM_WRITE; } /* mm/util.c */ -- cgit v1.2.3 From 46437f9a554fbe3e110580ca08ab703b59f2f95a Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Tue, 2 Feb 2016 16:57:52 -0800 Subject: radix-tree: fix race in gang lookup If the indirect_ptr bit is set on a slot, that indicates we need to redo the lookup. Introduce a new function radix_tree_iter_retry() which forces the loop to retry the lookup by setting 'slot' to NULL and turning the iterator back to point at the problematic entry. This is a pretty rare problem to hit at the moment; the lookup has to race with a grow of the radix tree from a height of 0. The consequences of hitting this race are that gang lookup could return a pointer to a radix_tree_node instead of a pointer to whatever the user had inserted in the tree. Fixes: cebbd29e1c2f ("radix-tree: rewrite gang lookup using iterator") Signed-off-by: Matthew Wilcox Cc: Hugh Dickins Cc: Ohad Ben-Cohen Cc: Konstantin Khlebnikov Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/radix-tree.h | 16 ++++++++++++++++ lib/radix-tree.c | 12 ++++++++++-- 2 files changed, 26 insertions(+), 2 deletions(-) (limited to 'include') diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h index 7c88ad156a29..00b17c526c1f 100644 --- a/include/linux/radix-tree.h +++ b/include/linux/radix-tree.h @@ -378,6 +378,22 @@ radix_tree_iter_init(struct radix_tree_iter *iter, unsigned long start) void **radix_tree_next_chunk(struct radix_tree_root *root, struct radix_tree_iter *iter, unsigned flags); +/** + * radix_tree_iter_retry - retry this chunk of the iteration + * @iter: iterator state + * + * If we iterate over a tree protected only by the RCU lock, a race + * against deletion or creation may result in seeing a slot for which + * radix_tree_deref_retry() returns true. If so, call this function + * and continue the iteration. + */ +static inline __must_check +void **radix_tree_iter_retry(struct radix_tree_iter *iter) +{ + iter->next_index = iter->index; + return NULL; +} + /** * radix_tree_chunk_size - get current chunk size * diff --git a/lib/radix-tree.c b/lib/radix-tree.c index fcf5d98574ce..6b79e9026e24 100644 --- a/lib/radix-tree.c +++ b/lib/radix-tree.c @@ -1019,9 +1019,13 @@ radix_tree_gang_lookup(struct radix_tree_root *root, void **results, return 0; radix_tree_for_each_slot(slot, root, &iter, first_index) { - results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot)); + results[ret] = rcu_dereference_raw(*slot); if (!results[ret]) continue; + if (radix_tree_is_indirect_ptr(results[ret])) { + slot = radix_tree_iter_retry(&iter); + continue; + } if (++ret == max_items) break; } @@ -1098,9 +1102,13 @@ radix_tree_gang_lookup_tag(struct radix_tree_root *root, void **results, return 0; radix_tree_for_each_tagged(slot, root, &iter, first_index, tag) { - results[ret] = indirect_to_ptr(rcu_dereference_raw(*slot)); + results[ret] = rcu_dereference_raw(*slot); if (!results[ret]) continue; + if (radix_tree_is_indirect_ptr(results[ret])) { + slot = radix_tree_iter_retry(&iter); + continue; + } if (++ret == max_items) break; } -- cgit v1.2.3