From ddd588b5dd55f14320379961e47683db4e4c1d90 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 22 Mar 2011 16:30:46 -0700 Subject: oom: suppress nodes that are not allowed from meminfo on oom kill The oom killer is extremely verbose for machines with a large number of cpus and/or nodes. This verbosity can often be harmful if it causes other important messages to be scrolled from the kernel log and incurs a signicant time delay, specifically for kernels with CONFIG_NODES_SHIFT > 8. This patch causes only memory information to be displayed for nodes that are allowed by current's cpuset when dumping the VM state. Information for all other nodes is irrelevant to the oom condition; we don't care if there's an abundance of memory elsewhere if we can't access it. This only affects the behavior of dumping memory information when an oom is triggered. Other dumps, such as for sysrq+m, still display the unfiltered form when using the existing show_mem() interface. Additionally, the per-cpu pageset statistics are extremely verbose in oom killer output, so it is now suppressed. This removes nodes_weight(current->mems_allowed) * (1 + nr_cpus) lines from the oom killer output. Callers may use __show_mem(SHOW_MEM_FILTER_NODES) to filter disallowed nodes. Signed-off-by: David Rientjes Cc: Mel Gorman Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 34 +++++++++++++++++++++++++++++++++- 1 file changed, 33 insertions(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 7945247..36be3ba 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2411,19 +2411,42 @@ void si_meminfo_node(struct sysinfo *val, int nid) } #endif +/* + * Determine whether the zone's node should be displayed or not, depending on + * whether SHOW_MEM_FILTER_NODES was passed to __show_free_areas(). + */ +static bool skip_free_areas_zone(unsigned int flags, const struct zone *zone) +{ + bool ret = false; + + if (!(flags & SHOW_MEM_FILTER_NODES)) + goto out; + + get_mems_allowed(); + ret = !node_isset(zone->zone_pgdat->node_id, + cpuset_current_mems_allowed); + put_mems_allowed(); +out: + return ret; +} + #define K(x) ((x) << (PAGE_SHIFT-10)) /* * Show free area list (used inside shift_scroll-lock stuff) * We also calculate the percentage fragmentation. We do this by counting the * memory on each free list with the exception of the first item on the list. + * Suppresses nodes that are not allowed by current's cpuset if + * SHOW_MEM_FILTER_NODES is passed. */ -void show_free_areas(void) +void __show_free_areas(unsigned int filter) { int cpu; struct zone *zone; for_each_populated_zone(zone) { + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s per-cpu:\n", zone->name); @@ -2465,6 +2488,8 @@ void show_free_areas(void) for_each_populated_zone(zone) { int i; + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s" " free:%lukB" @@ -2532,6 +2557,8 @@ void show_free_areas(void) for_each_populated_zone(zone) { unsigned long nr[MAX_ORDER], flags, order, total = 0; + if (skip_free_areas_zone(filter, zone)) + continue; show_node(zone); printk("%s: ", zone->name); @@ -2551,6 +2578,11 @@ void show_free_areas(void) show_swap_cache_info(); } +void show_free_areas(void) +{ + __show_free_areas(0); +} + static void zoneref_set_zone(struct zone *zone, struct zoneref *zoneref) { zoneref->zone = zone; -- cgit v1.1 From 29423e77c06cee7d4e335ef4a7cbd949da978c91 Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 22 Mar 2011 16:30:47 -0700 Subject: oom: suppress show_mem() for many nodes in irq context on page alloc failure When a page allocation failure occurs, show_mem() is called to dump the state of the VM so users may understand what happened to get into that condition. This output, however, can be extremely verbose. In irq context, it may result in significant delays that incur NMI watchdog timeouts when the machine is large (we use CONFIG_NODES_SHIFT > 8 here to define a "large" machine since the length of the show_mem() output is proportional to the number of possible nodes). This patch suppresses the show_mem() call in irq context when the kernel has CONFIG_NODES_SHIFT > 8. Signed-off-by: David Rientjes Cc: Mel Gorman Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 17 ++++++++++++++++- 1 file changed, 16 insertions(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 36be3ba..2aaafe8 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1714,6 +1714,20 @@ try_next_zone: return page; } +/* + * Large machines with many possible nodes should not always dump per-node + * meminfo in irq context. + */ +static inline bool should_suppress_show_mem(void) +{ + bool ret = false; + +#if NODES_SHIFT > 8 + ret = in_interrupt(); +#endif + return ret; +} + static inline int should_alloc_retry(gfp_t gfp_mask, unsigned int order, unsigned long pages_reclaimed) @@ -2161,7 +2175,8 @@ nopage: " order:%d, mode:0x%x\n", current->comm, order, gfp_mask); dump_stack(); - show_mem(); + if (!should_suppress_show_mem()) + show_mem(); } return page; got_pg: -- cgit v1.1 From cbf978bfb12d7deca97d7333f65eda0381a072de Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 22 Mar 2011 16:30:48 -0700 Subject: oom: suppress nodes that are not allowed from meminfo on page alloc failure Displaying extremely verbose meminfo for all nodes on the system is overkill for page allocation failures when the context restricts that allocation to only a subset of nodes. We don't particularly care about the state of all nodes when some are not allowed in the current context, they can have an abundance of memory but we can't allocate from that part of memory. This patch suppresses disallowed nodes from the meminfo dump on a page allocation failure if the context requires it. Signed-off-by: David Rientjes Cc: Mel Gorman Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 19 ++++++++++++++++--- 1 file changed, 16 insertions(+), 3 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2aaafe8..36a168e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2171,12 +2171,25 @@ rebalance: nopage: if (!(gfp_mask & __GFP_NOWARN) && printk_ratelimit()) { - printk(KERN_WARNING "%s: page allocation failure." - " order:%d, mode:0x%x\n", + unsigned int filter = SHOW_MEM_FILTER_NODES; + + /* + * This documents exceptions given to allocations in certain + * contexts that are allowed to allocate outside current's set + * of allowed nodes. + */ + if (!(gfp_mask & __GFP_NOMEMALLOC)) + if (test_thread_flag(TIF_MEMDIE) || + (current->flags & (PF_MEMALLOC | PF_EXITING))) + filter &= ~SHOW_MEM_FILTER_NODES; + if (in_interrupt() || !wait) + filter &= ~SHOW_MEM_FILTER_NODES; + + pr_warning("%s: page allocation failure. order:%d, mode:0x%x\n", current->comm, order, gfp_mask); dump_stack(); if (!should_suppress_show_mem()) - show_mem(); + __show_mem(filter); } return page; got_pg: -- cgit v1.1 From 1d16871d8c96deadc5f9753b6b096074f2cbcbe1 Mon Sep 17 00:00:00 2001 From: Namhyung Kim Date: Tue, 22 Mar 2011 16:32:45 -0700 Subject: mm: batch-free pcp list if possible free_pcppages_bulk() frees pages from pcp lists in a round-robin fashion by keeping batch_free counter. But it doesn't need to spin if there is only one non-empty list. This can be checked by batch_free == MIGRATE_PCPTYPES. [akpm@linux-foundation.org: fix comment] Signed-off-by: Namhyung Kim Acked-by: Johannes Weiner Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 36a168e..426056a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -614,6 +614,10 @@ static void free_pcppages_bulk(struct zone *zone, int count, list = &pcp->lists[migratetype]; } while (list_empty(list)); + /* This is the only non-empty list. Free them all. */ + if (batch_free == MIGRATE_PCPTYPES) + batch_free = to_free; + do { page = list_entry(list->prev, struct page, lru); /* must delete as __free_one_page list manipulates */ -- cgit v1.1 From 11bc82d67d1150767901bca54a24466621d763d7 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 22 Mar 2011 16:33:11 -0700 Subject: mm: compaction: Use async migration for __GFP_NO_KSWAPD and enforce no writeback __GFP_NO_KSWAPD allocations are usually very expensive and not mandatory to succeed as they have graceful fallback. Waiting for I/O in those, tends to be overkill in terms of latencies, so we can reduce their latency by disabling sync migrate. Unfortunately, even with async migration it's still possible for the process to be blocked waiting for a request slot (e.g. get_request_wait in the block layer) when ->writepage is called. To prevent __GFP_NO_KSWAPD blocking, this patch prevents ->writepage being called on dirty page cache for asynchronous migration. Addresses https://bugzilla.kernel.org/show_bug.cgi?id=31142 [mel@csn.ul.ie: Avoid writebacks for NFS, retry locked pages, use bool] Signed-off-by: Andrea Arcangeli Signed-off-by: Mel Gorman Cc: Arthur Marsh Cc: Clemens Ladisch Cc: Johannes Weiner Cc: KAMEZAWA Hiroyuki Cc: Minchan Kim Reported-by: Alex Villacis Lasso Tested-by: Alex Villacis Lasso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 426056a..6d0032b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2103,7 +2103,7 @@ rebalance: sync_migration); if (page) goto got_pg; - sync_migration = true; + sync_migration = !(gfp_mask & __GFP_NO_KSWAPD); /* Try direct reclaim and then allocating */ page = __alloc_pages_direct_reclaim(gfp_mask, order, -- cgit v1.1 From 78afd5612deb8268bafc8b6507d72341d5ed9aac Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 22 Mar 2011 16:33:12 -0700 Subject: mm: add __GFP_OTHER_NODE flag Add a new __GFP_OTHER_NODE flag to tell the low level numa statistics in zone_statistics() that an allocation is on behalf of another thread. This way the local and remote counters can be still correct, even when background daemons like khugepaged are changing memory mappings. This only affects the accounting, but I think it's worth doing that right to avoid confusing users. I first tried to just pass down the right node, but this required a lot of changes to pass down this parameter and at least one addition of a 10th argument to a 9 argument function. Using the flag is a lot less intrusive. Open: should be also used for migration? [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Andi Kleen Cc: Andrea Arcangeli Reviewed-by: KAMEZAWA Hiroyuki Cc: Johannes Weiner Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 6d0032b..136a547 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1337,7 +1337,7 @@ again: } __count_zone_vm_events(PGALLOC, zone, 1 << order); - zone_statistics(preferred_zone, zone); + zone_statistics(preferred_zone, zone, gfp_flags); local_irq_restore(flags); VM_BUG_ON(bad_range(zone, page)); -- cgit v1.1 From 84be48d84a53044e13aa8816aab201ab5480815d Mon Sep 17 00:00:00 2001 From: "Kirill A. Shutemov" Date: Tue, 22 Mar 2011 16:33:41 -0700 Subject: mm/page_alloc.c: use list_move() instead of list_del()/list_add() combination Signed-off-by: Kirill A. Shutemov Cc: Mel Gorman Cc: Rik van Riel Cc: KAMEZAWA Hiroyuki Reviewed-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 136a547..3a58221 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -867,9 +867,8 @@ static int move_freepages(struct zone *zone, } order = page_order(page); - list_del(&page->lru); - list_add(&page->lru, - &zone->free_area[order].free_list[migratetype]); + list_move(&page->lru, + &zone->free_area[order].free_list[migratetype]); page += 1 << order; pages_moved += 1 << order; } -- cgit v1.1 From f212ad7cf9c73f8a7fa160e223dcb3f074441a72 Mon Sep 17 00:00:00 2001 From: Daisuke Nishimura Date: Wed, 23 Mar 2011 16:42:25 -0700 Subject: memcg: add memcg sanity checks at allocating and freeing pages Add checks at allocating or freeing a page whether the page is used (iow, charged) from the view point of memcg. This check may be useful in debugging a problem and we did similar checks before the commit 52d4b9ac(memcg: allocate all page_cgroup at boot). This patch adds some overheads at allocating or freeing memory, so it's enabled only when CONFIG_DEBUG_VM is enabled. Signed-off-by: Daisuke Nishimura Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Cc: Balbir Singh Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 3a58221..8e5726a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -53,6 +53,7 @@ #include #include #include +#include #include #include @@ -565,7 +566,8 @@ static inline int free_pages_check(struct page *page) if (unlikely(page_mapcount(page) | (page->mapping != NULL) | (atomic_read(&page->_count) != 0) | - (page->flags & PAGE_FLAGS_CHECK_AT_FREE))) { + (page->flags & PAGE_FLAGS_CHECK_AT_FREE) | + (mem_cgroup_bad_page_check(page)))) { bad_page(page); return 1; } @@ -754,7 +756,8 @@ static inline int check_new_page(struct page *page) if (unlikely(page_mapcount(page) | (page->mapping != NULL) | (atomic_read(&page->_count) != 0) | - (page->flags & PAGE_FLAGS_CHECK_AT_PREP))) { + (page->flags & PAGE_FLAGS_CHECK_AT_PREP) | + (mem_cgroup_bad_page_check(page)))) { bad_page(page); return 1; } @@ -5684,4 +5687,5 @@ void dump_page(struct page *page) page, atomic_read(&page->_count), page_mapcount(page), page->mapping, page->index); dump_page_flags(page->flags); + mem_cgroup_print_bad_page(page); } -- cgit v1.1 From b2b755b5f10eb32fbdc73a9907c07006b17f714b Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Thu, 24 Mar 2011 15:18:15 -0700 Subject: lib, arch: add filter argument to show_mem and fix private implementations Commit ddd588b5dd55 ("oom: suppress nodes that are not allowed from meminfo on oom kill") moved lib/show_mem.o out of lib/lib.a, which resulted in build warnings on all architectures that implement their own versions of show_mem(): lib/lib.a(show_mem.o): In function `show_mem': show_mem.c:(.text+0x1f4): multiple definition of `show_mem' arch/sparc/mm/built-in.o:(.text+0xd70): first defined here The fix is to remove __show_mem() and add its argument to show_mem() in all implementations to prevent this breakage. Architectures that implement their own show_mem() actually don't do anything with the argument yet, but they could be made to filter nodes that aren't allowed in the current context in the future just like the generic implementation. Reported-by: Stephen Rothwell Reported-by: James Bottomley Suggested-by: Andrew Morton Signed-off-by: David Rientjes Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 8e5726a..d6e7ba7 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2195,7 +2195,7 @@ nopage: current->comm, order, gfp_mask); dump_stack(); if (!should_suppress_show_mem()) - __show_mem(filter); + show_mem(filter); } return page; got_pg: -- cgit v1.1 From 25985edcedea6396277003854657b5f3cb31a628 Mon Sep 17 00:00:00 2001 From: Lucas De Marchi Date: Wed, 30 Mar 2011 22:57:33 -0300 Subject: Fix common misspellings Fixes generated by 'codespell' and manually reviewed. Signed-off-by: Lucas De Marchi --- mm/page_alloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index d6e7ba7..2747f5e5a 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -942,7 +942,7 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype) * If breaking a large block of pages, move all free * pages to the preferred allocation list. If falling * back for a reclaimable kernel allocation, be more - * agressive about taking ownership of free pages + * aggressive about taking ownership of free pages */ if (unlikely(current_order >= (pageblock_order >> 1)) || start_migratetype == MIGRATE_RECLAIMABLE || @@ -3926,7 +3926,7 @@ static void __init find_usable_zone_for_movable(void) /* * The zone ranges provided by the architecture do not include ZONE_MOVABLE - * because it is sized independant of architecture. Unlike the other zones, + * because it is sized independent of architecture. Unlike the other zones, * the starting point for ZONE_MOVABLE is not fixed. It may be different * in each node depending on the size of each node and how evenly kernelcore * is distributed. This helper function adjusts the zone ranges -- cgit v1.1 From 9f6ae448bfc6cdf40279f43bb0b4fd159edc4e0a Mon Sep 17 00:00:00 2001 From: Paul Mundt Date: Thu, 14 Apr 2011 15:21:57 -0700 Subject: mm/page_alloc.c: silence build_all_zonelists() section mismatch The memory hotplug case involves calling to build_all_zonelists() which in turns calls in to setup_zone_pageset(). The latter is marked __meminit while build_all_zonelists() itself has no particular annotation. build_all_zonelists() is only handed a non-NULL pointer in the case of memory hotplug through an existing __meminit path, so the setup_zone_pageset() reference is always safe. The options as such are either to flag build_all_zonelists() as __ref (as per __build_all_zonelists()), or to simply discard the __meminit annotation from setup_zone_pageset(). Signed-off-by: Paul Mundt Acked-by: Mel Gorman Cc: KAMEZAWA Hiroyuki Acked-by: Randy Dunlap Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm/page_alloc.c') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 2747f5e5a..9f8a97b 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3176,7 +3176,7 @@ static __init_refok int __build_all_zonelists(void *data) * Called with zonelists_mutex held always * unless system_state == SYSTEM_BOOTING. */ -void build_all_zonelists(void *data) +void __ref build_all_zonelists(void *data) { set_zonelist_order(); -- cgit v1.1