summaryrefslogtreecommitdiff
path: root/mm
diff options
context:
space:
mode:
authorIngo Molnar <mingo@elte.hu>2009-05-18 10:15:09 +0200
committerIngo Molnar <mingo@elte.hu>2009-05-18 10:15:35 +0200
commit1079cac0f4eb7d968395378b1625979d4c818dd6 (patch)
treed261058d95d27fe7250511d5fad69db30dc17731 /mm
parent5872144f64b34a5942f6b4acedc90b02de72c58b (diff)
parent1406de8e11eb043681297adf86d6892ff8efc27a (diff)
Merge commit 'v2.6.30-rc6' into tracing/core
Merge reason: we were on an -rc4 base, sync up to -rc6 Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'mm')
-rw-r--r--mm/Kconfig28
-rw-r--r--mm/madvise.c8
-rw-r--r--mm/nommu.c4
-rw-r--r--mm/oom_kill.c44
-rw-r--r--mm/page_alloc.c20
-rw-r--r--mm/pdflush.c31
-rw-r--r--mm/vmalloc.c1
7 files changed, 82 insertions, 54 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 36d6ea2b634..b9ce574827c 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -112,14 +112,6 @@ 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/nommu.c b/mm/nommu.c
index 809998aa7b5..b571ef70742 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
@@ -66,7 +66,7 @@ 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;
@@ -515,8 +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) {
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/pdflush.c b/mm/pdflush.c
index f2caf96993f..235ac440c44 100644
--- a/mm/pdflush.c
+++ b/mm/pdflush.c
@@ -58,14 +58,6 @@ static DEFINE_SPINLOCK(pdflush_lock);
int nr_pdflush_threads = 0;
/*
- * The max/min number of pdflush threads. R/W by sysctl at
- * /proc/sys/vm/nr_pdflush_threads_max/min
- */
-int nr_pdflush_threads_max __read_mostly = MAX_PDFLUSH_THREADS;
-int nr_pdflush_threads_min __read_mostly = MIN_PDFLUSH_THREADS;
-
-
-/*
* The time at which the pdflush thread pool last went empty
*/
static unsigned long last_empty_jifs;
@@ -76,7 +68,7 @@ static unsigned long last_empty_jifs;
* Thread pool management algorithm:
*
* - The minimum and maximum number of pdflush instances are bound
- * by nr_pdflush_threads_min and nr_pdflush_threads_max.
+ * by MIN_PDFLUSH_THREADS and MAX_PDFLUSH_THREADS.
*
* - If there have been no idle pdflush instances for 1 second, create
* a new one.
@@ -142,13 +134,14 @@ static int __pdflush(struct pdflush_work *my_work)
* To throttle creation, we reset last_empty_jifs.
*/
if (time_after(jiffies, last_empty_jifs + 1 * HZ)) {
- if (list_empty(&pdflush_list) &&
- nr_pdflush_threads < nr_pdflush_threads_max) {
- last_empty_jifs = jiffies;
- nr_pdflush_threads++;
- spin_unlock_irq(&pdflush_lock);
- start_one_pdflush_thread();
- spin_lock_irq(&pdflush_lock);
+ if (list_empty(&pdflush_list)) {
+ if (nr_pdflush_threads < MAX_PDFLUSH_THREADS) {
+ last_empty_jifs = jiffies;
+ nr_pdflush_threads++;
+ spin_unlock_irq(&pdflush_lock);
+ start_one_pdflush_thread();
+ spin_lock_irq(&pdflush_lock);
+ }
}
}
@@ -160,7 +153,7 @@ static int __pdflush(struct pdflush_work *my_work)
*/
if (list_empty(&pdflush_list))
continue;
- if (nr_pdflush_threads <= nr_pdflush_threads_min)
+ if (nr_pdflush_threads <= MIN_PDFLUSH_THREADS)
continue;
pdf = list_entry(pdflush_list.prev, struct pdflush_work, list);
if (time_after(jiffies, pdf->when_i_went_to_sleep + 1 * HZ)) {
@@ -266,9 +259,9 @@ static int __init pdflush_init(void)
* Pre-set nr_pdflush_threads... If we fail to create,
* the count will be decremented.
*/
- nr_pdflush_threads = nr_pdflush_threads_min;
+ nr_pdflush_threads = MIN_PDFLUSH_THREADS;
- for (i = 0; i < nr_pdflush_threads_min; i++)
+ for (i = 0; i < MIN_PDFLUSH_THREADS; i++)
start_one_pdflush_thread();
return 0;
}
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);
}