From d00285884c0892bb1310df96bce6056e9ce9b9d9 Mon Sep 17 00:00:00 2001 From: Wanpeng Li Date: Fri, 22 Mar 2013 15:04:40 -0700 Subject: mm/hugetlb: fix total hugetlbfs pages count when using memory overcommit accouting hugetlb_total_pages is used for overcommit calculations but the current implementation considers only the default hugetlb page size (which is either the first defined hugepage size or the one specified by default_hugepagesz kernel boot parameter). If the system is configured for more than one hugepage size, which is possible since commit a137e1cc6d6e ("hugetlbfs: per mount huge page sizes") then the overcommit estimation done by __vm_enough_memory() (resp. shown by meminfo_proc_show) is not precise - there is an impression of more available/allowed memory. This can lead to an unexpected ENOMEM/EFAULT resp. SIGSEGV when memory is accounted. Testcase: boot: hugepagesz=1G hugepages=1 the default overcommit ratio is 50 before patch: egrep 'CommitLimit' /proc/meminfo CommitLimit: 55434168 kB after patch: egrep 'CommitLimit' /proc/meminfo CommitLimit: 54909880 kB [akpm@linux-foundation.org: coding-style tweak] Signed-off-by: Wanpeng Li Acked-by: Michal Hocko Cc: "Aneesh Kumar K.V" Cc: Hillf Danton Cc: KAMEZAWA Hiroyuki Cc: [3.0+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/hugetlb.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 0a0be33bb199..ca9a7c6d7e97 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -2124,8 +2124,12 @@ int hugetlb_report_node_meminfo(int nid, char *buf) /* Return the number pages of memory we physically have, in PAGE_SIZE units. */ unsigned long hugetlb_total_pages(void) { - struct hstate *h = &default_hstate; - return h->nr_huge_pages * pages_per_huge_page(h); + struct hstate *h; + unsigned long nr_total_pages = 0; + + for_each_hstate(h) + nr_total_pages += h->nr_huge_pages * pages_per_huge_page(h); + return nr_total_pages; } static int hugetlb_acct_memory(struct hstate *h, long delta) -- cgit v1.2.1 From ca4b3f302c90de5e516296e581c31c80125cd24b Mon Sep 17 00:00:00 2001 From: Jianguo Wu Date: Fri, 22 Mar 2013 15:04:50 -0700 Subject: mm/hotplug: only free wait_table if it's allocated by vmalloc zone->wait_table may be allocated from bootmem, it can not be freed. Signed-off-by: Jianguo Wu Reviewed-by: Tang Chen Cc: Tang Chen Cc: Jiang Liu Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory_hotplug.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c index 9597eec8239d..ee3765760818 100644 --- a/mm/memory_hotplug.c +++ b/mm/memory_hotplug.c @@ -1779,7 +1779,11 @@ void try_offline_node(int nid) for (i = 0; i < MAX_NR_ZONES; i++) { struct zone *zone = pgdat->node_zones + i; - if (zone->wait_table) + /* + * wait_table may be allocated from boot memory, + * here only free if it's allocated by vmalloc. + */ + if (is_vmalloc_addr(zone->wait_table)) vfree(zone->wait_table); } -- cgit v1.2.1 From 09a9f1d27892255cfb9c91203f19476765e2d8d1 Mon Sep 17 00:00:00 2001 From: Michel Lespinasse Date: Thu, 28 Mar 2013 16:26:23 -0700 Subject: Revert "mm: introduce VM_POPULATE flag to better deal with racy userspace programs" This reverts commit 186930500985 ("mm: introduce VM_POPULATE flag to better deal with racy userspace programs"). VM_POPULATE only has any effect when userspace plays racy games with vmas by trying to unmap and remap memory regions that mmap or mlock are operating on. Also, the only effect of VM_POPULATE when userspace plays such games is that it avoids populating new memory regions that get remapped into the address range that was being operated on by the original mmap or mlock calls. Let's remove VM_POPULATE as there isn't any strong argument to mandate a new vm_flag. Signed-off-by: Michel Lespinasse Signed-off-by: Hugh Dickins Signed-off-by: Linus Torvalds --- mm/fremap.c | 12 ++---------- mm/mlock.c | 11 +++++------ mm/mmap.c | 4 +++- 3 files changed, 10 insertions(+), 17 deletions(-) (limited to 'mm') diff --git a/mm/fremap.c b/mm/fremap.c index 4723ac8d2fc2..87da3590c61e 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -204,10 +204,8 @@ get_write_lock: unsigned long addr; struct file *file = get_file(vma->vm_file); - vm_flags = vma->vm_flags; - if (!(flags & MAP_NONBLOCK)) - vm_flags |= VM_POPULATE; - addr = mmap_region(file, start, size, vm_flags, pgoff); + addr = mmap_region(file, start, size, + vma->vm_flags, pgoff); fput(file); if (IS_ERR_VALUE(addr)) { err = addr; @@ -226,12 +224,6 @@ get_write_lock: mutex_unlock(&mapping->i_mmap_mutex); } - if (!(flags & MAP_NONBLOCK) && !(vma->vm_flags & VM_POPULATE)) { - if (!has_write_lock) - goto get_write_lock; - vma->vm_flags |= VM_POPULATE; - } - if (vma->vm_flags & VM_LOCKED) { /* * drop PG_Mlocked flag for over-mapped range diff --git a/mm/mlock.c b/mm/mlock.c index 1c5e33fce639..79b7cf7d1bca 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -358,7 +358,7 @@ static int do_mlock(unsigned long start, size_t len, int on) newflags = vma->vm_flags & ~VM_LOCKED; if (on) - newflags |= VM_LOCKED | VM_POPULATE; + newflags |= VM_LOCKED; tmp = vma->vm_end; if (tmp > end) @@ -418,8 +418,7 @@ int __mm_populate(unsigned long start, unsigned long len, int ignore_errors) * range with the first VMA. Also, skip undesirable VMA types. */ nend = min(end, vma->vm_end); - if ((vma->vm_flags & (VM_IO | VM_PFNMAP | VM_POPULATE)) != - VM_POPULATE) + if (vma->vm_flags & (VM_IO | VM_PFNMAP)) continue; if (nstart < vma->vm_start) nstart = vma->vm_start; @@ -492,9 +491,9 @@ static int do_mlockall(int flags) struct vm_area_struct * vma, * prev = NULL; if (flags & MCL_FUTURE) - current->mm->def_flags |= VM_LOCKED | VM_POPULATE; + current->mm->def_flags |= VM_LOCKED; else - current->mm->def_flags &= ~(VM_LOCKED | VM_POPULATE); + current->mm->def_flags &= ~VM_LOCKED; if (flags == MCL_FUTURE) goto out; @@ -503,7 +502,7 @@ static int do_mlockall(int flags) newflags = vma->vm_flags & ~VM_LOCKED; if (flags & MCL_CURRENT) - newflags |= VM_LOCKED | VM_POPULATE; + newflags |= VM_LOCKED; /* Ignore errors */ mlock_fixup(vma, &prev, vma->vm_start, vma->vm_end, newflags); diff --git a/mm/mmap.c b/mm/mmap.c index 2664a47cec93..6466699b16cb 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -1306,7 +1306,9 @@ unsigned long do_mmap_pgoff(struct file *file, unsigned long addr, } addr = mmap_region(file, addr, len, vm_flags, pgoff); - if (!IS_ERR_VALUE(addr) && (vm_flags & VM_POPULATE)) + if (!IS_ERR_VALUE(addr) && + ((vm_flags & VM_LOCKED) || + (flags & (MAP_POPULATE | MAP_NONBLOCK)) == MAP_POPULATE)) *populate = len; return addr; } -- cgit v1.2.1 From 181387da2d64c3129e5b5186c4dd388bc5041d53 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 1 Apr 2013 19:08:06 -0700 Subject: writeback: remove unused bdi_pending_list There's no user left. Remove it. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Fengguang Wu --- mm/backing-dev.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 41733c5dc820..657569b3fcf6 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -31,13 +31,11 @@ EXPORT_SYMBOL_GPL(noop_backing_dev_info); static struct class *bdi_class; /* - * bdi_lock protects updates to bdi_list and bdi_pending_list, as well as - * reader side protection for bdi_pending_list. bdi_list has RCU reader side + * bdi_lock protects updates to bdi_list. bdi_list has RCU reader side * locking. */ DEFINE_SPINLOCK(bdi_lock); LIST_HEAD(bdi_list); -LIST_HEAD(bdi_pending_list); void bdi_lock_two(struct bdi_writeback *wb1, struct bdi_writeback *wb2) { -- cgit v1.2.1 From 839a8e8660b6777e7fe4e80af1a048aebe2b5977 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 1 Apr 2013 19:08:06 -0700 Subject: writeback: replace custom worker pool implementation with unbound workqueue Writeback implements its own worker pool - each bdi can be associated with a worker thread which is created and destroyed dynamically. The worker thread for the default bdi is always present and serves as the "forker" thread which forks off worker threads for other bdis. there's no reason for writeback to implement its own worker pool when using unbound workqueue instead is much simpler and more efficient. This patch replaces custom worker pool implementation in writeback with an unbound workqueue. The conversion isn't too complicated but the followings are worth mentioning. * bdi_writeback->last_active, task and wakeup_timer are removed. delayed_work ->dwork is added instead. Explicit timer handling is no longer necessary. Everything works by either queueing / modding / flushing / canceling the delayed_work item. * bdi_writeback_thread() becomes bdi_writeback_workfn() which runs off bdi_writeback->dwork. On each execution, it processes bdi->work_list and reschedules itself if there are more things to do. The function also handles low-mem condition, which used to be handled by the forker thread. If the function is running off a rescuer thread, it only writes out limited number of pages so that the rescuer can serve other bdis too. This preserves the flusher creation failure behavior of the forker thread. * INIT_LIST_HEAD(&bdi->bdi_list) is used to tell bdi_writeback_workfn() about on-going bdi unregistration so that it always drains work_list even if it's running off the rescuer. Note that the original code was broken in this regard. Under memory pressure, a bdi could finish unregistration with non-empty work_list. * The default bdi is no longer special. It now is treated the same as any other bdi and bdi_cap_flush_forker() is removed. * BDI_pending is no longer used. Removed. * Some tracepoints become non-applicable. The following TPs are removed - writeback_nothread, writeback_wake_thread, writeback_wake_forker_thread, writeback_thread_start, writeback_thread_stop. Everything, including devices coming and going away and rescuer operation under simulated memory pressure, seems to work fine in my test setup. Signed-off-by: Tejun Heo Reviewed-by: Jan Kara Cc: Jens Axboe Cc: Fengguang Wu Cc: Jeff Moyer --- mm/backing-dev.c | 255 ++++++------------------------------------------------- 1 file changed, 28 insertions(+), 227 deletions(-) (limited to 'mm') diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 657569b3fcf6..2857d4f6bca4 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -37,6 +37,9 @@ static struct class *bdi_class; DEFINE_SPINLOCK(bdi_lock); LIST_HEAD(bdi_list); +/* bdi_wq serves all asynchronous writeback tasks */ +struct workqueue_struct *bdi_wq; + void bdi_lock_two(struct bdi_writeback *wb1, struct bdi_writeback *wb2) { if (wb1 < wb2) { @@ -255,6 +258,11 @@ static int __init default_bdi_init(void) { int err; + bdi_wq = alloc_workqueue("writeback", WQ_MEM_RECLAIM | WQ_FREEZABLE | + WQ_UNBOUND, 0); + if (!bdi_wq) + return -ENOMEM; + err = bdi_init(&default_backing_dev_info); if (!err) bdi_register(&default_backing_dev_info, NULL, "default"); @@ -269,26 +277,6 @@ int bdi_has_dirty_io(struct backing_dev_info *bdi) return wb_has_dirty_io(&bdi->wb); } -static void wakeup_timer_fn(unsigned long data) -{ - struct backing_dev_info *bdi = (struct backing_dev_info *)data; - - spin_lock_bh(&bdi->wb_lock); - if (bdi->wb.task) { - trace_writeback_wake_thread(bdi); - wake_up_process(bdi->wb.task); - } else if (bdi->dev) { - /* - * When bdi tasks are inactive for long time, they are killed. - * In this case we have to wake-up the forker thread which - * should create and run the bdi thread. - */ - trace_writeback_wake_forker_thread(bdi); - wake_up_process(default_backing_dev_info.wb.task); - } - spin_unlock_bh(&bdi->wb_lock); -} - /* * This function is used when the first inode for this bdi is marked dirty. It * wakes-up the corresponding bdi thread which should then take care of the @@ -305,176 +293,7 @@ void bdi_wakeup_thread_delayed(struct backing_dev_info *bdi) unsigned long timeout; timeout = msecs_to_jiffies(dirty_writeback_interval * 10); - mod_timer(&bdi->wb.wakeup_timer, jiffies + timeout); -} - -/* - * Calculate the longest interval (jiffies) bdi threads are allowed to be - * inactive. - */ -static unsigned long bdi_longest_inactive(void) -{ - unsigned long interval; - - interval = msecs_to_jiffies(dirty_writeback_interval * 10); - return max(5UL * 60 * HZ, interval); -} - -/* - * Clear pending bit and wakeup anybody waiting for flusher thread creation or - * shutdown - */ -static void bdi_clear_pending(struct backing_dev_info *bdi) -{ - clear_bit(BDI_pending, &bdi->state); - smp_mb__after_clear_bit(); - wake_up_bit(&bdi->state, BDI_pending); -} - -static int bdi_forker_thread(void *ptr) -{ - struct bdi_writeback *me = ptr; - - current->flags |= PF_SWAPWRITE; - set_freezable(); - - /* - * Our parent may run at a different priority, just set us to normal - */ - set_user_nice(current, 0); - - for (;;) { - struct task_struct *task = NULL; - struct backing_dev_info *bdi; - enum { - NO_ACTION, /* Nothing to do */ - FORK_THREAD, /* Fork bdi thread */ - KILL_THREAD, /* Kill inactive bdi thread */ - } action = NO_ACTION; - - /* - * Temporary measure, we want to make sure we don't see - * dirty data on the default backing_dev_info - */ - if (wb_has_dirty_io(me) || !list_empty(&me->bdi->work_list)) { - del_timer(&me->wakeup_timer); - wb_do_writeback(me, 0); - } - - spin_lock_bh(&bdi_lock); - /* - * In the following loop we are going to check whether we have - * some work to do without any synchronization with tasks - * waking us up to do work for them. Set the task state here - * so that we don't miss wakeups after verifying conditions. - */ - set_current_state(TASK_INTERRUPTIBLE); - - list_for_each_entry(bdi, &bdi_list, bdi_list) { - bool have_dirty_io; - - if (!bdi_cap_writeback_dirty(bdi) || - bdi_cap_flush_forker(bdi)) - continue; - - WARN(!test_bit(BDI_registered, &bdi->state), - "bdi %p/%s is not registered!\n", bdi, bdi->name); - - have_dirty_io = !list_empty(&bdi->work_list) || - wb_has_dirty_io(&bdi->wb); - - /* - * If the bdi has work to do, but the thread does not - * exist - create it. - */ - if (!bdi->wb.task && have_dirty_io) { - /* - * Set the pending bit - if someone will try to - * unregister this bdi - it'll wait on this bit. - */ - set_bit(BDI_pending, &bdi->state); - action = FORK_THREAD; - break; - } - - spin_lock(&bdi->wb_lock); - - /* - * If there is no work to do and the bdi thread was - * inactive long enough - kill it. The wb_lock is taken - * to make sure no-one adds more work to this bdi and - * wakes the bdi thread up. - */ - if (bdi->wb.task && !have_dirty_io && - time_after(jiffies, bdi->wb.last_active + - bdi_longest_inactive())) { - task = bdi->wb.task; - bdi->wb.task = NULL; - spin_unlock(&bdi->wb_lock); - set_bit(BDI_pending, &bdi->state); - action = KILL_THREAD; - break; - } - spin_unlock(&bdi->wb_lock); - } - spin_unlock_bh(&bdi_lock); - - /* Keep working if default bdi still has things to do */ - if (!list_empty(&me->bdi->work_list)) - __set_current_state(TASK_RUNNING); - - switch (action) { - case FORK_THREAD: - __set_current_state(TASK_RUNNING); - task = kthread_create(bdi_writeback_thread, &bdi->wb, - "flush-%s", dev_name(bdi->dev)); - if (IS_ERR(task)) { - /* - * If thread creation fails, force writeout of - * the bdi from the thread. Hopefully 1024 is - * large enough for efficient IO. - */ - writeback_inodes_wb(&bdi->wb, 1024, - WB_REASON_FORKER_THREAD); - } else { - /* - * The spinlock makes sure we do not lose - * wake-ups when racing with 'bdi_queue_work()'. - * And as soon as the bdi thread is visible, we - * can start it. - */ - spin_lock_bh(&bdi->wb_lock); - bdi->wb.task = task; - spin_unlock_bh(&bdi->wb_lock); - wake_up_process(task); - } - bdi_clear_pending(bdi); - break; - - case KILL_THREAD: - __set_current_state(TASK_RUNNING); - kthread_stop(task); - bdi_clear_pending(bdi); - break; - - case NO_ACTION: - if (!wb_has_dirty_io(me) || !dirty_writeback_interval) - /* - * There are no dirty data. The only thing we - * should now care about is checking for - * inactive bdi threads and killing them. Thus, - * let's sleep for longer time, save energy and - * be friendly for battery-driven devices. - */ - schedule_timeout(bdi_longest_inactive()); - else - schedule_timeout(msecs_to_jiffies(dirty_writeback_interval * 10)); - try_to_freeze(); - break; - } - } - - return 0; + mod_delayed_work(bdi_wq, &bdi->wb.dwork, timeout); } /* @@ -487,6 +306,9 @@ static void bdi_remove_from_list(struct backing_dev_info *bdi) spin_unlock_bh(&bdi_lock); synchronize_rcu_expedited(); + + /* bdi_list is now unused, clear it to mark @bdi dying */ + INIT_LIST_HEAD(&bdi->bdi_list); } int bdi_register(struct backing_dev_info *bdi, struct device *parent, @@ -506,20 +328,6 @@ int bdi_register(struct backing_dev_info *bdi, struct device *parent, bdi->dev = dev; - /* - * Just start the forker thread for our default backing_dev_info, - * and add other bdi's to the list. They will get a thread created - * on-demand when they need it. - */ - if (bdi_cap_flush_forker(bdi)) { - struct bdi_writeback *wb = &bdi->wb; - - wb->task = kthread_run(bdi_forker_thread, wb, "bdi-%s", - dev_name(dev)); - if (IS_ERR(wb->task)) - return PTR_ERR(wb->task); - } - bdi_debug_register(bdi, dev_name(dev)); set_bit(BDI_registered, &bdi->state); @@ -543,8 +351,6 @@ EXPORT_SYMBOL(bdi_register_dev); */ static void bdi_wb_shutdown(struct backing_dev_info *bdi) { - struct task_struct *task; - if (!bdi_cap_writeback_dirty(bdi)) return; @@ -554,22 +360,20 @@ static void bdi_wb_shutdown(struct backing_dev_info *bdi) bdi_remove_from_list(bdi); /* - * If setup is pending, wait for that to complete first + * Drain work list and shutdown the delayed_work. At this point, + * @bdi->bdi_list is empty telling bdi_Writeback_workfn() that @bdi + * is dying and its work_list needs to be drained no matter what. */ - wait_on_bit(&bdi->state, BDI_pending, bdi_sched_wait, - TASK_UNINTERRUPTIBLE); + mod_delayed_work(bdi_wq, &bdi->wb.dwork, 0); + flush_delayed_work(&bdi->wb.dwork); + WARN_ON(!list_empty(&bdi->work_list)); /* - * Finally, kill the kernel thread. We don't need to be RCU - * safe anymore, since the bdi is gone from visibility. + * This shouldn't be necessary unless @bdi for some reason has + * unflushed dirty IO after work_list is drained. Do it anyway + * just in case. */ - spin_lock_bh(&bdi->wb_lock); - task = bdi->wb.task; - bdi->wb.task = NULL; - spin_unlock_bh(&bdi->wb_lock); - - if (task) - kthread_stop(task); + cancel_delayed_work_sync(&bdi->wb.dwork); } /* @@ -595,10 +399,8 @@ void bdi_unregister(struct backing_dev_info *bdi) bdi_set_min_ratio(bdi, 0); trace_writeback_bdi_unregister(bdi); bdi_prune_sb(bdi); - del_timer_sync(&bdi->wb.wakeup_timer); - if (!bdi_cap_flush_forker(bdi)) - bdi_wb_shutdown(bdi); + bdi_wb_shutdown(bdi); bdi_debug_unregister(bdi); spin_lock_bh(&bdi->wb_lock); @@ -620,7 +422,7 @@ static void bdi_wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi) INIT_LIST_HEAD(&wb->b_io); INIT_LIST_HEAD(&wb->b_more_io); spin_lock_init(&wb->list_lock); - setup_timer(&wb->wakeup_timer, wakeup_timer_fn, (unsigned long)bdi); + INIT_DELAYED_WORK(&wb->dwork, bdi_writeback_workfn); } /* @@ -693,12 +495,11 @@ void bdi_destroy(struct backing_dev_info *bdi) bdi_unregister(bdi); /* - * If bdi_unregister() had already been called earlier, the - * wakeup_timer could still be armed because bdi_prune_sb() - * can race with the bdi_wakeup_thread_delayed() calls from - * __mark_inode_dirty(). + * If bdi_unregister() had already been called earlier, the dwork + * could still be pending because bdi_prune_sb() can race with the + * bdi_wakeup_thread_delayed() calls from __mark_inode_dirty(). */ - del_timer_sync(&bdi->wb.wakeup_timer); + cancel_delayed_work_sync(&bdi->wb.dwork); for (i = 0; i < NR_BDI_STAT_ITEMS; i++) percpu_counter_destroy(&bdi->bdi_stat[i]); -- cgit v1.2.1 From b5c872ddb7083c7909fb76a170c3807e04564bb3 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 1 Apr 2013 19:08:06 -0700 Subject: writeback: expose the bdi_wq workqueue There are cases where userland wants to tweak the priority and affinity of writeback flushers. Expose bdi_wq to userland by setting WQ_SYSFS. It appears under /sys/bus/workqueue/devices/writeback/ and allows adjusting maximum concurrency level, cpumask and nice level. Signed-off-by: Tejun Heo Cc: Jens Axboe Cc: Fengguang Wu Cc: Jeff Moyer Cc: Kay Sievers Cc: Greg Kroah-Hartman --- mm/backing-dev.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/backing-dev.c b/mm/backing-dev.c index 2857d4f6bca4..502517492258 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -259,7 +259,7 @@ static int __init default_bdi_init(void) int err; bdi_wq = alloc_workqueue("writeback", WQ_MEM_RECLAIM | WQ_FREEZABLE | - WQ_UNBOUND, 0); + WQ_UNBOUND | WQ_SYSFS, 0); if (!bdi_wq) return -ENOMEM; -- cgit v1.2.1