aboutsummaryrefslogtreecommitdiff
path: root/mm
diff options
context:
space:
mode:
Diffstat (limited to 'mm')
-rw-r--r--mm/Kconfig28
-rw-r--r--mm/madvise.c8
-rw-r--r--mm/memcontrol.c38
-rw-r--r--mm/memory.c112
-rw-r--r--mm/mmap.c12
-rw-r--r--mm/nommu.c17
-rw-r--r--mm/oom_kill.c44
-rw-r--r--mm/page_alloc.c20
-rw-r--r--mm/shmem.c8
-rw-r--r--mm/swap.c46
-rw-r--r--mm/vmalloc.c1
-rw-r--r--mm/vmscan.c2
12 files changed, 190 insertions, 146 deletions
diff --git a/mm/Kconfig b/mm/Kconfig
index 57971d2ab84..c2b57d81e15 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -225,3 +225,31 @@ config HAVE_MLOCKED_PAGE_BIT
config MMU_NOTIFIER
bool
+
+config NOMMU_INITIAL_TRIM_EXCESS
+ int "Turn on mmap() excess space trimming before booting"
+ depends on !MMU
+ default 1
+ help
+ The NOMMU mmap() frequently needs to allocate large contiguous chunks
+ of memory on which to store mappings, but it can only ask the system
+ allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
+ more than it requires. To deal with this, mmap() is able to trim off
+ the excess and return it to the allocator.
+
+ If trimming is enabled, the excess is trimmed off and returned to the
+ system allocator, which can cause extra fragmentation, particularly
+ if there are a lot of transient processes.
+
+ If trimming is disabled, the excess is kept, but not used, which for
+ long-term mappings means that the space is wasted.
+
+ Trimming can be dynamically controlled through a sysctl option
+ (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
+ excess pages there must be before trimming should occur, or zero if
+ no trimming is to occur.
+
+ This option specifies the initial value of this option. The default
+ of 1 says that all excess pages should be trimmed.
+
+ See Documentation/nommu-mmap.txt for more information.
diff --git a/mm/madvise.c b/mm/madvise.c
index b9ce574827c..36d6ea2b634 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -112,6 +112,14 @@ static long madvise_willneed(struct vm_area_struct * vma,
if (!file)
return -EBADF;
+ /*
+ * Page cache readahead assumes page cache pages are order-0 which
+ * is not the case for hugetlbfs. Do not give a bad return value
+ * but ignore the advice.
+ */
+ if (vma->vm_flags & VM_HUGETLB)
+ return 0;
+
if (file->f_mapping->a_ops->get_xip_mem) {
/* no bad return value, but ignore advice */
return 0;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e44fb0fbb80..01c2d8f1468 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1024,9 +1024,7 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
return NULL;
pc = lookup_page_cgroup(page);
- /*
- * Used bit of swapcache is solid under page lock.
- */
+ lock_page_cgroup(pc);
if (PageCgroupUsed(pc)) {
mem = pc->mem_cgroup;
if (mem && !css_tryget(&mem->css))
@@ -1040,6 +1038,7 @@ static struct mem_cgroup *try_get_mem_cgroup_from_swapcache(struct page *page)
mem = NULL;
rcu_read_unlock();
}
+ unlock_page_cgroup(pc);
return mem;
}
@@ -1618,37 +1617,28 @@ void mem_cgroup_end_migration(struct mem_cgroup *mem,
}
/*
- * A call to try to shrink memory usage under specified resource controller.
- * This is typically used for page reclaiming for shmem for reducing side
- * effect of page allocation from shmem, which is used by some mem_cgroup.
+ * A call to try to shrink memory usage on charge failure at shmem's swapin.
+ * Calling hierarchical_reclaim is not enough because we should update
+ * last_oom_jiffies to prevent pagefault_out_of_memory from invoking global OOM.
+ * Moreover considering hierarchy, we should reclaim from the mem_over_limit,
+ * not from the memcg which this page would be charged to.
+ * try_charge_swapin does all of these works properly.
*/
-int mem_cgroup_shrink_usage(struct page *page,
+int mem_cgroup_shmem_charge_fallback(struct page *page,
struct mm_struct *mm,
gfp_t gfp_mask)
{
struct mem_cgroup *mem = NULL;
- int progress = 0;
- int retry = MEM_CGROUP_RECLAIM_RETRIES;
+ int ret;
if (mem_cgroup_disabled())
return 0;
- if (page)
- mem = try_get_mem_cgroup_from_swapcache(page);
- if (!mem && mm)
- mem = try_get_mem_cgroup_from_mm(mm);
- if (unlikely(!mem))
- return 0;
- do {
- progress = mem_cgroup_hierarchical_reclaim(mem,
- gfp_mask, true, false);
- progress += mem_cgroup_check_under_limit(mem);
- } while (!progress && --retry);
+ ret = mem_cgroup_try_charge_swapin(mm, page, gfp_mask, &mem);
+ if (!ret)
+ mem_cgroup_cancel_charge_swapin(mem); /* it does !mem check */
- css_put(&mem->css);
- if (!retry)
- return -ENOMEM;
- return 0;
+ return ret;
}
static DEFINE_MUTEX(set_limit_mutex);
diff --git a/mm/memory.c b/mm/memory.c
index cf6873e91c6..4126dd16778 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1971,6 +1971,15 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
ret = tmp;
goto unwritable_page;
}
+ if (unlikely(!(tmp & VM_FAULT_LOCKED))) {
+ lock_page(old_page);
+ if (!old_page->mapping) {
+ ret = 0; /* retry the fault */
+ unlock_page(old_page);
+ goto unwritable_page;
+ }
+ } else
+ VM_BUG_ON(!PageLocked(old_page));
/*
* Since we dropped the lock we need to revalidate
@@ -1980,9 +1989,11 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma,
*/
page_table = pte_offset_map_lock(mm, pmd, address,
&ptl);
- page_cache_release(old_page);
- if (!pte_same(*page_table, orig_pte))
+ if (!pte_same(*page_table, orig_pte)) {
+ unlock_page(old_page);
+ page_cache_release(old_page);
goto unlock;
+ }
page_mkwrite = 1;
}
@@ -2094,9 +2105,6 @@ gotten:
unlock:
pte_unmap_unlock(page_table, ptl);
if (dirty_page) {
- if (vma->vm_file)
- file_update_time(vma->vm_file);
-
/*
* Yes, Virginia, this is actually required to prevent a race
* with clear_page_dirty_for_io() from clearing the page dirty
@@ -2105,16 +2113,41 @@ unlock:
*
* do_no_page is protected similarly.
*/
- wait_on_page_locked(dirty_page);
- set_page_dirty_balance(dirty_page, page_mkwrite);
+ if (!page_mkwrite) {
+ wait_on_page_locked(dirty_page);
+ set_page_dirty_balance(dirty_page, page_mkwrite);
+ }
put_page(dirty_page);
+ if (page_mkwrite) {
+ struct address_space *mapping = dirty_page->mapping;
+
+ set_page_dirty(dirty_page);
+ unlock_page(dirty_page);
+ page_cache_release(dirty_page);
+ if (mapping) {
+ /*
+ * Some device drivers do not set page.mapping
+ * but still dirty their pages
+ */
+ balance_dirty_pages_ratelimited(mapping);
+ }
+ }
+
+ /* file_update_time outside page_lock */
+ if (vma->vm_file)
+ file_update_time(vma->vm_file);
}
return ret;
oom_free_new:
page_cache_release(new_page);
oom:
- if (old_page)
+ if (old_page) {
+ if (page_mkwrite) {
+ unlock_page(old_page);
+ page_cache_release(old_page);
+ }
page_cache_release(old_page);
+ }
return VM_FAULT_OOM;
unwritable_page:
@@ -2458,8 +2491,7 @@ static int do_swap_page(struct mm_struct *mm, struct vm_area_struct *vma,
if (mem_cgroup_try_charge_swapin(mm, page, GFP_KERNEL, &ptr)) {
ret = VM_FAULT_OOM;
- unlock_page(page);
- goto out;
+ goto out_page;
}
/*
@@ -2521,6 +2553,7 @@ out:
out_nomap:
mem_cgroup_cancel_charge_swapin(ptr);
pte_unmap_unlock(page_table, ptl);
+out_page:
unlock_page(page);
page_cache_release(page);
return ret;
@@ -2664,27 +2697,22 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
int tmp;
unlock_page(page);
- vmf.flags |= FAULT_FLAG_MKWRITE;
+ vmf.flags = FAULT_FLAG_WRITE|FAULT_FLAG_MKWRITE;
tmp = vma->vm_ops->page_mkwrite(vma, &vmf);
if (unlikely(tmp &
(VM_FAULT_ERROR | VM_FAULT_NOPAGE))) {
ret = tmp;
- anon = 1; /* no anon but release vmf.page */
- goto out_unlocked;
- }
- lock_page(page);
- /*
- * XXX: this is not quite right (racy vs
- * invalidate) to unlock and relock the page
- * like this, however a better fix requires
- * reworking page_mkwrite locking API, which
- * is better done later.
- */
- if (!page->mapping) {
- ret = 0;
- anon = 1; /* no anon but release vmf.page */
- goto out;
+ goto unwritable_page;
}
+ if (unlikely(!(tmp & VM_FAULT_LOCKED))) {
+ lock_page(page);
+ if (!page->mapping) {
+ ret = 0; /* retry the fault */
+ unlock_page(page);
+ goto unwritable_page;
+ }
+ } else
+ VM_BUG_ON(!PageLocked(page));
page_mkwrite = 1;
}
}
@@ -2736,19 +2764,35 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma,
pte_unmap_unlock(page_table, ptl);
out:
- unlock_page(vmf.page);
-out_unlocked:
- if (anon)
- page_cache_release(vmf.page);
- else if (dirty_page) {
- if (vma->vm_file)
- file_update_time(vma->vm_file);
+ if (dirty_page) {
+ struct address_space *mapping = page->mapping;
- set_page_dirty_balance(dirty_page, page_mkwrite);
+ if (set_page_dirty(dirty_page))
+ page_mkwrite = 1;
+ unlock_page(dirty_page);
put_page(dirty_page);
+ if (page_mkwrite && mapping) {
+ /*
+ * Some device drivers do not set page.mapping but still
+ * dirty their pages
+ */
+ balance_dirty_pages_ratelimited(mapping);
+ }
+
+ /* file_update_time outside page_lock */
+ if (vma->vm_file)
+ file_update_time(vma->vm_file);
+ } else {
+ unlock_page(vmf.page);
+ if (anon)
+ page_cache_release(vmf.page);
}
return ret;
+
+unwritable_page:
+ page_cache_release(page);
+ return ret;
}
static int do_linear_fault(struct mm_struct *mm, struct vm_area_struct *vma,
diff --git a/mm/mmap.c b/mm/mmap.c
index 3303d1ba8e8..6b7b1a95944 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -85,7 +85,7 @@ EXPORT_SYMBOL(vm_get_page_prot);
int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
int sysctl_overcommit_ratio = 50; /* default is 50% */
int sysctl_max_map_count __read_mostly = DEFAULT_MAX_MAP_COUNT;
-atomic_long_t vm_committed_space = ATOMIC_LONG_INIT(0);
+struct percpu_counter vm_committed_as;
/*
* Check that a process has enough memory to allocate a new virtual
@@ -179,11 +179,7 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
if (mm)
allowed -= mm->total_vm / 32;
- /*
- * cast `allowed' as a signed long because vm_committed_space
- * sometimes has a negative value
- */
- if (atomic_long_read(&vm_committed_space) < (long)allowed)
+ if (percpu_counter_read_positive(&vm_committed_as) < allowed)
return 0;
error:
vm_unacct_memory(pages);
@@ -2481,4 +2477,8 @@ void mm_drop_all_locks(struct mm_struct *mm)
*/
void __init mmap_init(void)
{
+ int ret;
+
+ ret = percpu_counter_init(&vm_committed_as, 0);
+ VM_BUG_ON(ret);
}
diff --git a/mm/nommu.c b/mm/nommu.c
index 72eda4aee2c..b571ef70742 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -62,11 +62,11 @@ void *high_memory;
struct page *mem_map;
unsigned long max_mapnr;
unsigned long num_physpages;
-atomic_long_t vm_committed_space = ATOMIC_LONG_INIT(0);
+struct percpu_counter vm_committed_as;
int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
int sysctl_overcommit_ratio = 50; /* default is 50% */
int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
-int sysctl_nr_trim_pages = 1; /* page trimming behaviour */
+int sysctl_nr_trim_pages = CONFIG_NOMMU_INITIAL_TRIM_EXCESS;
int heap_stack_gap = 0;
atomic_long_t mmap_pages_allocated;
@@ -463,6 +463,10 @@ SYSCALL_DEFINE1(brk, unsigned long, brk)
*/
void __init mmap_init(void)
{
+ int ret;
+
+ ret = percpu_counter_init(&vm_committed_as, 0);
+ VM_BUG_ON(ret);
vm_region_jar = KMEM_CACHE(vm_region, SLAB_PANIC);
}
@@ -511,8 +515,6 @@ static void add_nommu_region(struct vm_region *region)
validate_nommu_regions();
- BUG_ON(region->vm_start & ~PAGE_MASK);
-
parent = NULL;
p = &nommu_region_tree.rb_node;
while (*p) {
@@ -1847,12 +1849,9 @@ int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
if (mm)
allowed -= mm->total_vm / 32;
- /*
- * cast `allowed' as a signed long because vm_committed_space
- * sometimes has a negative value
- */
- if (atomic_long_read(&vm_committed_space) < (long)allowed)
+ if (percpu_counter_read_positive(&vm_committed_as) < allowed)
return 0;
+
error:
vm_unacct_memory(pages);
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 2f3166e308d..92bcf1db16b 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -514,34 +514,32 @@ void clear_zonelist_oom(struct zonelist *zonelist, gfp_t gfp_mask)
*/
static void __out_of_memory(gfp_t gfp_mask, int order)
{
- if (sysctl_oom_kill_allocating_task) {
- oom_kill_process(current, gfp_mask, order, 0, NULL,
- "Out of memory (oom_kill_allocating_task)");
-
- } else {
- unsigned long points;
- struct task_struct *p;
-
-retry:
- /*
- * Rambo mode: Shoot down a process and hope it solves whatever
- * issues we may have.
- */
- p = select_bad_process(&points, NULL);
+ struct task_struct *p;
+ unsigned long points;
- if (PTR_ERR(p) == -1UL)
+ if (sysctl_oom_kill_allocating_task)
+ if (!oom_kill_process(current, gfp_mask, order, 0, NULL,
+ "Out of memory (oom_kill_allocating_task)"))
return;
+retry:
+ /*
+ * Rambo mode: Shoot down a process and hope it solves whatever
+ * issues we may have.
+ */
+ p = select_bad_process(&points, NULL);
- /* Found nothing?!?! Either we hang forever, or we panic. */
- if (!p) {
- read_unlock(&tasklist_lock);
- panic("Out of memory and no killable processes...\n");
- }
+ if (PTR_ERR(p) == -1UL)
+ return;
- if (oom_kill_process(p, gfp_mask, order, points, NULL,
- "Out of memory"))
- goto retry;
+ /* Found nothing?!?! Either we hang forever, or we panic. */
+ if (!p) {
+ read_unlock(&tasklist_lock);
+ panic("Out of memory and no killable processes...\n");
}
+
+ if (oom_kill_process(p, gfp_mask, order, points, NULL,
+ "Out of memory"))
+ goto retry;
}
/*
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index e2f26991fff..fe753ecf2aa 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -2681,6 +2681,7 @@ static void __meminit zone_init_free_lists(struct zone *zone)
static int zone_batchsize(struct zone *zone)
{
+#ifdef CONFIG_MMU
int batch;
/*
@@ -2706,9 +2707,26 @@ static int zone_batchsize(struct zone *zone)
* of pages of one half of the possible page colors
* and the other with pages of the other colors.
*/
- batch = (1 << (fls(batch + batch/2)-1)) - 1;
+ batch = rounddown_pow_of_two(batch + batch/2) - 1;
return batch;
+
+#else
+ /* The deferral and batching of frees should be suppressed under NOMMU
+ * conditions.
+ *
+ * The problem is that NOMMU needs to be able to allocate large chunks
+ * of contiguous memory as there's no hardware page translation to
+ * assemble apparent contiguous memory from discontiguous pages.
+ *
+ * Queueing large contiguous runs of pages for batching, however,
+ * causes the pages to actually be freed in smaller chunks. As there
+ * can be a significant delay between the individual batches being
+ * recycled, this leads to the once large chunks of space being
+ * fragmented and becoming unavailable for high-order allocations.
+ */
+ return 0;
+#endif
}
static void setup_pageset(struct per_cpu_pageset *p, unsigned long batch)
diff --git a/mm/shmem.c b/mm/shmem.c
index f9cb20ebb99..b25f95ce3db 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -1340,8 +1340,12 @@ repeat:
shmem_swp_unmap(entry);
spin_unlock(&info->lock);
if (error == -ENOMEM) {
- /* allow reclaim from this memory cgroup */
- error = mem_cgroup_shrink_usage(swappage,
+ /*
+ * reclaim from proper memory cgroup and
+ * call memcg's OOM if needed.
+ */
+ error = mem_cgroup_shmem_charge_fallback(
+ swappage,
current->mm,
gfp);
if (error) {
diff --git a/mm/swap.c b/mm/swap.c
index bede23ce64e..cb29ae5d33a 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -491,49 +491,6 @@ unsigned pagevec_lookup_tag(struct pagevec *pvec, struct address_space *mapping,
EXPORT_SYMBOL(pagevec_lookup_tag);
-#ifdef CONFIG_SMP
-/*
- * We tolerate a little inaccuracy to avoid ping-ponging the counter between
- * CPUs
- */
-#define ACCT_THRESHOLD max(16, NR_CPUS * 2)
-
-static DEFINE_PER_CPU(long, committed_space);
-
-void vm_acct_memory(long pages)
-{
- long *local;
-
- preempt_disable();
- local = &__get_cpu_var(committed_space);
- *local += pages;
- if (*local > ACCT_THRESHOLD || *local < -ACCT_THRESHOLD) {
- atomic_long_add(*local, &vm_committed_space);
- *local = 0;
- }
- preempt_enable();
-}
-
-#ifdef CONFIG_HOTPLUG_CPU
-
-/* Drop the CPU's cached committed space back into the central pool. */
-static int cpu_swap_callback(struct notifier_block *nfb,
- unsigned long action,
- void *hcpu)
-{
- long *committed;
-
- committed = &per_cpu(committed_space, (long)hcpu);
- if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
- atomic_long_add(*committed, &vm_committed_space);
- *committed = 0;
- drain_cpu_pagevecs((long)hcpu);
- }
- return NOTIFY_OK;
-}
-#endif /* CONFIG_HOTPLUG_CPU */
-#endif /* CONFIG_SMP */
-
/*
* Perform any setup for the swap system
*/
@@ -554,7 +511,4 @@ void __init swap_setup(void)
* Right now other parts of the system means that we
* _really_ don't want to cluster much more
*/
-#ifdef CONFIG_HOTPLUG_CPU
- hotcpu_notifier(cpu_swap_callback, 0);
-#endif
}
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index fab19876b4d..083716ea38c 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -402,6 +402,7 @@ overflow:
printk(KERN_WARNING
"vmap allocation for size %lu failed: "
"use vmalloc=<size> to increase size.\n", size);
+ kfree(va);
return ERR_PTR(-EBUSY);
}
diff --git a/mm/vmscan.c b/mm/vmscan.c
index eac9577941f..5fa3eda1f03 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -1471,7 +1471,7 @@ static void shrink_zone(int priority, struct zone *zone,
for_each_evictable_lru(l) {
int file = is_file_lru(l);
- int scan;
+ unsigned long scan;
scan = zone_nr_pages(zone, sc, l);
if (priority) {