From f95ba941d1bee594d536cdcbf879a0865381b903 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Tue, 25 Jan 2011 15:07:11 -0800 Subject: mm/pgtable-generic.c: fix CONFIG_SWAP=n build mips (and sparc32): In file included from arch/mips/include/asm/tlb.h:21, from mm/pgtable-generic.c:9: include/asm-generic/tlb.h: In function `tlb_flush_mmu': include/asm-generic/tlb.h:76: error: implicit declaration of function `release_pages' include/asm-generic/tlb.h: In function `tlb_remove_page': include/asm-generic/tlb.h:105: error: implicit declaration of function `page_cache_release' free_pages_and_swap_cache() and free_page_and_swap_cache() are macros which call release_pages() and page_cache_release(). The obvious fix is to include pagemap.h in swap.h, where those macros are defined. But that breaks sparc for weird reasons. So fix it within mm/pgtable-generic.c instead. Reported-by: Yoichi Yuasa Cc: Geert Uytterhoeven Acked-by: Sam Ravnborg Cc: Sergei Shtylyov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/pgtable-generic.c | 1 + 1 file changed, 1 insertion(+) (limited to 'mm') diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c index 0369f5b3ba1b..eb663fb533e0 100644 --- a/mm/pgtable-generic.c +++ b/mm/pgtable-generic.c @@ -6,6 +6,7 @@ * Copyright (C) 2010 Linus Torvalds */ +#include #include #include -- cgit v1.2.3 From f33261d75b88f55a08e6a9648cef73509979bfba Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 25 Jan 2011 15:07:20 -0800 Subject: mm: fix deferred congestion timeout if preferred zone is not allowed Before 0e093d99763e ("writeback: do not sleep on the congestion queue if there are no congested BDIs or if significant congestion is not being encountered in the current zone"), preferred_zone was only used for NUMA statistics, to determine the zoneidx from which to allocate from given the type requested, and whether to utilize memory compaction. wait_iff_congested(), though, uses preferred_zone to determine if the congestion wait should be deferred because its dirty pages are backed by a congested bdi. This incorrectly defers the timeout and busy loops in the page allocator with various cond_resched() calls if preferred_zone is not allowed in the current context, usually consuming 100% of a cpu. This patch ensures preferred_zone is an allowed zone in the fastpath depending on whether current is constrained by its cpuset or nodes in its mempolicy (when the nodemask passed is non-NULL). This is correct since the fastpath allocation always passes ALLOC_CPUSET when trying to allocate memory. In the slowpath, this patch resets preferred_zone to the first zone of the allowed type when the allocation is not constrained by current's cpuset, i.e. it does not pass ALLOC_CPUSET. This patch also ensures preferred_zone is from the set of allowed nodes when called from within direct reclaim since allocations are always constrained by cpusets in this context (it is blockable). Both of these uses of cpuset_current_mems_allowed are protected by get_mems_allowed(). Signed-off-by: David Rientjes Cc: Mel Gorman Cc: Johannes Weiner Cc: Minchan Kim Cc: Wu Fengguang Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Acked-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 12 +++++++++++- mm/vmscan.c | 3 ++- 2 files changed, 13 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 90c1439549fd..f4967910c967 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2034,6 +2034,14 @@ restart: */ alloc_flags = gfp_to_alloc_flags(gfp_mask); + /* + * Find the true preferred zone if the allocation is unconstrained by + * cpusets. + */ + if (!(alloc_flags & ALLOC_CPUSET) && !nodemask) + first_zones_zonelist(zonelist, high_zoneidx, NULL, + &preferred_zone); + /* This is the last chance, in general, before the goto nopage. */ page = get_page_from_freelist(gfp_mask, nodemask, order, zonelist, high_zoneidx, alloc_flags & ~ALLOC_NO_WATERMARKS, @@ -2192,7 +2200,9 @@ __alloc_pages_nodemask(gfp_t gfp_mask, unsigned int order, get_mems_allowed(); /* The preferred zone is used for statistics later */ - first_zones_zonelist(zonelist, high_zoneidx, nodemask, &preferred_zone); + first_zones_zonelist(zonelist, high_zoneidx, + nodemask ? : &cpuset_current_mems_allowed, + &preferred_zone); if (!preferred_zone) { put_mems_allowed(); return NULL; diff --git a/mm/vmscan.c b/mm/vmscan.c index f5d90dedebba..148c6e630df2 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -2083,7 +2083,8 @@ static unsigned long do_try_to_free_pages(struct zonelist *zonelist, struct zone *preferred_zone; first_zones_zonelist(zonelist, gfp_zone(sc->gfp_mask), - NULL, &preferred_zone); + &cpuset_current_mems_allowed, + &preferred_zone); wait_iff_congested(preferred_zone, BLK_RW_ASYNC, HZ/10); } } -- cgit v1.2.3 From 2ff754fa8f416e82327f2d8f1354a033b66286df Mon Sep 17 00:00:00 2001 From: David Rientjes Date: Tue, 25 Jan 2011 15:07:23 -0800 Subject: mm: clear pages_scanned only if draining a pcp adds pages to the buddy allocator Commit 0e093d99763e ("writeback: do not sleep on the congestion queue if there are no congested BDIs or if significant congestion is not being encountered in the current zone") uncovered a livelock in the page allocator that resulted in tasks infinitely looping trying to find memory and kswapd running at 100% cpu. The issue occurs because drain_all_pages() is called immediately following direct reclaim when no memory is freed and try_to_free_pages() returns non-zero because all zones in the zonelist do not have their all_unreclaimable flag set. When draining the per-cpu pagesets back to the buddy allocator for each zone, the zone->pages_scanned counter is cleared to avoid erroneously setting zone->all_unreclaimable later. The problem is that no pages may actually be drained and, thus, the unreclaimable logic never fails direct reclaim so the oom killer may be invoked. This apparently only manifested after wait_iff_congested() was introduced and the zone was full of anonymous memory that would not congest the backing store. The page allocator would infinitely loop if there were no other tasks waiting to be scheduled and clear zone->pages_scanned because of drain_all_pages() as the result of this change before kswapd could scan enough pages to trigger the reclaim logic. Additionally, with every loop of the page allocator and in the reclaim path, kswapd would be kicked and would end up running at 100% cpu. In this scenario, current and kswapd are all running continuously with kswapd incrementing zone->pages_scanned and current clearing it. The problem is even more pronounced when current swaps some of its memory to swap cache and the reclaimable logic then considers all active anonymous memory in the all_unreclaimable logic, which requires a much higher zone->pages_scanned value for try_to_free_pages() to return zero that is never attainable in this scenario. Before wait_iff_congested(), the page allocator would incur an unconditional timeout and allow kswapd to elevate zone->pages_scanned to a level that the oom killer would be called the next time it loops. The fix is to only attempt to drain pcp pages if there is actually a quantity to be drained. The unconditional clearing of zone->pages_scanned in free_pcppages_bulk() need not be changed since other callers already ensure that draining will occur. This patch ensures that free_pcppages_bulk() will actually free memory before calling into it from drain_all_pages() so zone->pages_scanned is only cleared if appropriate. Signed-off-by: David Rientjes Cc: Mel Gorman Reviewed-by: Johannes Weiner Cc: Minchan Kim Cc: Wu Fengguang Cc: KAMEZAWA Hiroyuki Cc: KOSAKI Motohiro Reviewed-by: Rik van Riel Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/page_alloc.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/page_alloc.c b/mm/page_alloc.c index f4967910c967..a873e61e312e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1088,8 +1088,10 @@ static void drain_pages(unsigned int cpu) pset = per_cpu_ptr(zone->pageset, cpu); pcp = &pset->pcp; - free_pcppages_bulk(zone, pcp->count, pcp); - pcp->count = 0; + if (pcp->count) { + free_pcppages_bulk(zone, pcp->count, pcp); + pcp->count = 0; + } local_irq_restore(flags); } } -- cgit v1.2.3 From 8dba474f034c322d96ada39cb20cac711d80dcb2 Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Tue, 25 Jan 2011 15:07:24 -0800 Subject: mm/memcontrol.c: fix uninitialized variable use in mem_cgroup_move_parent() In mm/memcontrol.c::mem_cgroup_move_parent() there's a path that jumps to the 'put_back' label ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false, charge); if (ret || !parent) goto put_back; where we'll if (charge > PAGE_SIZE) compound_unlock_irqrestore(page, flags); but, we have not assigned anything to 'flags' at this point, nor have we called 'compound_lock_irqsave()' (which is what sets 'flags'). The 'put_back' label should be moved below the call to compound_unlock_irqrestore() as per this patch. Signed-off-by: Jesper Juhl Cc: Balbir Singh Cc: Daisuke Nishimura Cc: KAMEZAWA Hiroyuki Cc: Pavel Emelianov Cc: "Kirill A. Shutemov" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index db76ef726293..4fcf47a62550 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2292,9 +2292,10 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc, ret = mem_cgroup_move_account(pc, child, parent, true, charge); if (ret) mem_cgroup_cancel_charge(parent, charge); -put_back: + if (charge > PAGE_SIZE) compound_unlock_irqrestore(page, flags); +put_back: putback_lru_page(page); put: put_page(page); -- cgit v1.2.3 From 33a938774fdb9933e9c77504b035f4f87c0859df Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 25 Jan 2011 15:07:25 -0800 Subject: mm: compaction: don't depend on HUGETLB_PAGE Commit 5d6892407 ("thp: select CONFIG_COMPACTION if TRANSPARENT_HUGEPAGE enabled") causes this warning during the configuration process: warning: (TRANSPARENT_HUGEPAGE) selects COMPACTION which has unmet direct dependencies (EXPERIMENTAL && HUGETLB_PAGE && MMU) COMPACTION doesn't depend on HUGETLB_PAGE, it doesn't depend on THP either, it is also useful for regular alloc_pages(order > 0) including the very kernel stack during fork (THREAD_ORDER = 1). It's always better to enable COMPACTION. The warning should be an error because we would end up with MIGRATION not selected, and COMPACTION wouldn't work without migration (despite it seems to build with an inline migrate_pages returning -ENOSYS). I'd also like to remove EXPERIMENTAL: compaction has been in the kernel for some releases (for full safety the default remains disabled which I think is enough). Signed-off-by: Andrea Arcangeli Reported-by: Luca Tettamanti Tested-by: Luca Tettamanti Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/Kconfig b/mm/Kconfig index 3ad483bdf505..e9c0c61f2ddd 100644 --- a/mm/Kconfig +++ b/mm/Kconfig @@ -179,7 +179,7 @@ config SPLIT_PTLOCK_CPUS config COMPACTION bool "Allow for memory compaction" select MIGRATION - depends on EXPERIMENTAL && HUGETLB_PAGE && MMU + depends on MMU help Allows the compaction of memory for the allocation of huge pages. -- cgit v1.2.3 From 28bd65781c848d95ba6a7f58b5c4b8265a804ec6 Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 25 Jan 2011 15:07:26 -0800 Subject: mm: migration: clarify migrate_pages() comment Callers of migrate_pages should putback_lru_pages to return pages isolated to LRU or free list. Now comment is rather confusing. It says caller always have to call it. It is more clear to point out that the caller has to call it if migrate_pages's return value isn't zero. Signed-off-by: Minchan Kim Cc: Christoph Lameter Cc: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/migrate.c b/mm/migrate.c index 46fe8cc13d67..9f29a3b7aac2 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -888,7 +888,7 @@ out: * are movable anymore because to has become empty * or no retryable pages exist anymore. * Caller should call putback_lru_pages to return pages to the LRU - * or free list. + * or free list only if ret != 0. * * Return: Number of pages not migrated or error code. */ -- cgit v1.2.3 From 01c88e2d6b7330c0cc5867fe2297e7d826e1337d Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 25 Jan 2011 15:07:27 -0800 Subject: memcg: fix account leak at failure of memsw acconting Commit 4b53433468 ("memcg: clean up try_charge main loop") removes a cancel of charge at case: memory charge-> success. mem+swap charge-> failure. This leaks usage of memory. Fix it. Signed-off-by: KAMEZAWA Hiroyuki Reviewed-by: Johannes Weiner Acked-by: Daisuke Nishimura Cc: Balbir Singh Cc: [2.6.36+] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 1 + 1 file changed, 1 insertion(+) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 4fcf47a62550..1eb1a04f874c 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1832,6 +1832,7 @@ static int __mem_cgroup_do_charge(struct mem_cgroup *mem, gfp_t gfp_mask, if (likely(!ret)) return CHARGE_OK; + res_counter_uncharge(&mem->res, csize); mem_over_limit = mem_cgroup_from_res_counter(fail_res, memsw); flags |= MEM_CGROUP_RECLAIM_NOSWAP; } else -- cgit v1.2.3 From 3d37c4a9199920964ffdfaec6335d93b9dcf9ca5 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 25 Jan 2011 15:07:28 -0800 Subject: memcg: bugfix check mem_cgroup_disabled() at split fixup mem_cgroup_disabled() should be checked at splitting. If disabled, no heavy work is necesary. Signed-off-by: KAMEZAWA Hiroyuki Acked-by: Daisuke Nishimura Reviewed-by: Johannes Weiner Cc: Balbir Singh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 1eb1a04f874c..8ab1d42664fb 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2145,6 +2145,8 @@ void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail) struct page_cgroup *tail_pc = lookup_page_cgroup(tail); unsigned long flags; + if (mem_cgroup_disabled()) + return; /* * We have no races with charge/uncharge but will have races with * page state accounting. -- cgit v1.2.3 From 52dbb9050936fd33ceb45f10529dbc992507c058 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 25 Jan 2011 15:07:29 -0800 Subject: memcg: fix race at move_parent around compound_order() A fix up mem_cgroup_move_parent() which use compound_order() in asynchronous manner. This compound_order() may return unknown value because we don't take lock. Use PageTransHuge() and HPAGE_SIZE instead of it. Also clean up for mem_cgroup_move_parent(). - remove unnecessary initialization of local variable. - rename charge_size -> page_size - remove unnecessary (wrong) comment. - added a comment about THP. Note: Current design take compound_page_lock() in caller of move_account(). This should be revisited when we implement direct move_task of hugepage without splitting. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: KAMEZAWA Hiroyuki Reviewed-by: Johannes Weiner Acked-by: Daisuke Nishimura Cc: Balbir Singh Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 25 ++++++++++++++++--------- 1 file changed, 16 insertions(+), 9 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 8ab1d42664fb..3878cfe399dc 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2236,7 +2236,12 @@ static int mem_cgroup_move_account(struct page_cgroup *pc, { int ret = -EINVAL; unsigned long flags; - + /* + * The page is isolated from LRU. So, collapse function + * will not handle this page. But page splitting can happen. + * Do this check under compound_page_lock(). The caller should + * hold it. + */ if ((charge_size > PAGE_SIZE) && !PageTransHuge(pc->page)) return -EBUSY; @@ -2268,7 +2273,7 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc, struct cgroup *cg = child->css.cgroup; struct cgroup *pcg = cg->parent; struct mem_cgroup *parent; - int charge = PAGE_SIZE; + int page_size = PAGE_SIZE; unsigned long flags; int ret; @@ -2281,22 +2286,24 @@ static int mem_cgroup_move_parent(struct page_cgroup *pc, goto out; if (isolate_lru_page(page)) goto put; - /* The page is isolated from LRU and we have no race with splitting */ - charge = PAGE_SIZE << compound_order(page); + + if (PageTransHuge(page)) + page_size = HPAGE_SIZE; parent = mem_cgroup_from_cont(pcg); - ret = __mem_cgroup_try_charge(NULL, gfp_mask, &parent, false, charge); + ret = __mem_cgroup_try_charge(NULL, gfp_mask, + &parent, false, page_size); if (ret || !parent) goto put_back; - if (charge > PAGE_SIZE) + if (page_size > PAGE_SIZE) flags = compound_lock_irqsave(page); - ret = mem_cgroup_move_account(pc, child, parent, true, charge); + ret = mem_cgroup_move_account(pc, child, parent, true, page_size); if (ret) - mem_cgroup_cancel_charge(parent, charge); + mem_cgroup_cancel_charge(parent, page_size); - if (charge > PAGE_SIZE) + if (page_size > PAGE_SIZE) compound_unlock_irqrestore(page, flags); put_back: putback_lru_page(page); -- cgit v1.2.3 From 0a08739e81671de2cb690774937fe510c000b27f Mon Sep 17 00:00:00 2001 From: Jesper Juhl Date: Sat, 30 Oct 2010 23:43:05 +0200 Subject: kmemleak: remove memset by using kzalloc We don't need to memset if we just use kzalloc() rather than kmalloc() in kmemleak_test_init(). Signed-off-by: Jesper Juhl Reviewed-by: Minchan Kim Signed-off-by: Catalin Marinas --- mm/kmemleak-test.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'mm') diff --git a/mm/kmemleak-test.c b/mm/kmemleak-test.c index 177a5169bbde..ff0d9779cec8 100644 --- a/mm/kmemleak-test.c +++ b/mm/kmemleak-test.c @@ -75,13 +75,11 @@ static int __init kmemleak_test_init(void) * after the module is removed. */ for (i = 0; i < 10; i++) { - elem = kmalloc(sizeof(*elem), GFP_KERNEL); - pr_info("kmemleak: kmalloc(sizeof(*elem)) = %p\n", elem); + elem = kzalloc(sizeof(*elem), GFP_KERNEL); + pr_info("kmemleak: kzalloc(sizeof(*elem)) = %p\n", elem); if (!elem) return -ENOMEM; - memset(elem, 0, sizeof(*elem)); INIT_LIST_HEAD(&elem->list); - list_add_tail(&elem->list, &test_list); } -- cgit v1.2.3 From 6ae4bd1f0bc479984f30061b5e5116060c24a267 Mon Sep 17 00:00:00 2001 From: Catalin Marinas Date: Thu, 27 Jan 2011 10:30:26 +0000 Subject: kmemleak: Allow kmemleak metadata allocations to fail MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit This patch adds __GFP_NORETRY and __GFP_NOMEMALLOC flags to the kmemleak metadata allocations so that it has a smaller effect on the users of the kernel slab allocator. Since kmemleak allocations can now fail more often, this patch also reduces the verbosity by passing __GFP_NOWARN and not dumping the stack trace when a kmemleak allocation fails. Signed-off-by: Catalin Marinas Reported-by: Toralf Förster Acked-by: Pekka Enberg Acked-by: David Rientjes Cc: Ted Ts'o --- mm/kmemleak.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'mm') diff --git a/mm/kmemleak.c b/mm/kmemleak.c index bd9bc214091b..84225f3b7190 100644 --- a/mm/kmemleak.c +++ b/mm/kmemleak.c @@ -113,7 +113,9 @@ #define BYTES_PER_POINTER sizeof(void *) /* GFP bitmask for kmemleak internal allocations */ -#define GFP_KMEMLEAK_MASK (GFP_KERNEL | GFP_ATOMIC) +#define gfp_kmemleak_mask(gfp) (((gfp) & (GFP_KERNEL | GFP_ATOMIC)) | \ + __GFP_NORETRY | __GFP_NOMEMALLOC | \ + __GFP_NOWARN) /* scanning area inside a memory block */ struct kmemleak_scan_area { @@ -511,9 +513,10 @@ static struct kmemleak_object *create_object(unsigned long ptr, size_t size, struct kmemleak_object *object; struct prio_tree_node *node; - object = kmem_cache_alloc(object_cache, gfp & GFP_KMEMLEAK_MASK); + object = kmem_cache_alloc(object_cache, gfp_kmemleak_mask(gfp)); if (!object) { - kmemleak_stop("Cannot allocate a kmemleak_object structure\n"); + pr_warning("Cannot allocate a kmemleak_object structure\n"); + kmemleak_disable(); return NULL; } @@ -734,9 +737,9 @@ static void add_scan_area(unsigned long ptr, size_t size, gfp_t gfp) return; } - area = kmem_cache_alloc(scan_area_cache, gfp & GFP_KMEMLEAK_MASK); + area = kmem_cache_alloc(scan_area_cache, gfp_kmemleak_mask(gfp)); if (!area) { - kmemleak_warn("Cannot allocate a scan area\n"); + pr_warning("Cannot allocate a scan area\n"); goto out; } -- cgit v1.2.3 From fdf4c587a793ba87935e38e7f25a9540bc9a7b95 Mon Sep 17 00:00:00 2001 From: Michel Lespinasse Date: Mon, 31 Jan 2011 17:03:41 -0800 Subject: mlock: operate on any regions with protection != PROT_NONE As Tao Ma noticed, change 5ecfda0 breaks blktrace. This is because blktrace mmaps a file with PROT_WRITE permissions but without PROT_READ, so my attempt to not unnecessarity break COW during mlock ended up causing mlock to fail with a permission problem. I am proposing to let mlock ignore vma protection in all cases except PROT_NONE. In particular, mlock should not fail for PROT_WRITE regions (as in the blktrace case, which broke at 5ecfda0) or for PROT_EXEC regions (which seem to me like they were always broken). Signed-off-by: Michel Lespinasse Acked-by: Rik van Riel Signed-off-by: Linus Torvalds --- mm/mlock.c | 7 +++++++ 1 file changed, 7 insertions(+) (limited to 'mm') diff --git a/mm/mlock.c b/mm/mlock.c index 13e81ee8be9d..c3924c7f00be 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -178,6 +178,13 @@ static long __mlock_vma_pages_range(struct vm_area_struct *vma, if ((vma->vm_flags & (VM_WRITE | VM_SHARED)) == VM_WRITE) gup_flags |= FOLL_WRITE; + /* + * We want mlock to succeed for regions that have any permissions + * other than PROT_NONE. + */ + if (vma->vm_flags & (VM_READ | VM_WRITE | VM_EXEC)) + gup_flags |= FOLL_FORCE; + if (vma->vm_flags & VM_LOCKED) gup_flags |= FOLL_MLOCK; -- cgit v1.2.3 From fceda1bf498677501befc7da72fd2e4de7f18466 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Tue, 1 Feb 2011 15:52:30 -0800 Subject: memsw: handle swapaccount kernel parameter correctly __setup based kernel command line parameters handlers which are handled in obsolete_checksetup are provided with the parameter value including = (more precisely everything right after the parameter name). This means that the current implementation of swapaccount[=1|0] doesn't work at all because if there is a value for the parameter then we are testing for "0" resp. "1" but we are getting "=0" resp. "=1" and if there is no parameter value we are getting an empty string rather than NULL. The original noswapccount parameter, which doesn't care about the value, works correctly. Signed-off-by: Michal Hocko Acked-by: KAMEZAWA Hiroyuki Cc: Daisuke Nishimura Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 3878cfe399dc..44f9f9c89f0c 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -5024,9 +5024,9 @@ struct cgroup_subsys mem_cgroup_subsys = { static int __init enable_swap_account(char *s) { /* consider enabled if no parameter or 1 is given */ - if (!s || !strcmp(s, "1")) + if (!(*s) || !strcmp(s, "=1")) really_do_swap_account = 1; - else if (!strcmp(s, "0")) + else if (!strcmp(s, "=0")) really_do_swap_account = 0; return 1; } @@ -5034,7 +5034,7 @@ __setup("swapaccount", enable_swap_account); static int __init disable_swap_account(char *s) { - enable_swap_account("0"); + enable_swap_account("=0"); return 1; } __setup("noswapaccount", disable_swap_account); -- cgit v1.2.3 From 552b372ba9db85751e7db2998f07cca2e51f5865 Mon Sep 17 00:00:00 2001 From: Michal Hocko Date: Tue, 1 Feb 2011 15:52:31 -0800 Subject: memsw: deprecate noswapaccount kernel parameter and schedule it for removal noswapaccount couldn't be used to control memsw for both on/off cases so we have added swapaccount[=0|1] parameter. This way we can turn the feature in two ways noswapaccount resp. swapaccount=0. We have kept the original noswapaccount but I think we should remove it after some time as it just makes more command line parameters without any advantages and also the code to handle parameters is uglier if we want both parameters. Signed-off-by: Michal Hocko Requested-by: KAMEZAWA Hiroyuki Acked-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 1 + 1 file changed, 1 insertion(+) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 44f9f9c89f0c..79abb1fd39d2 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -5034,6 +5034,7 @@ __setup("swapaccount", enable_swap_account); static int __init disable_swap_account(char *s) { + printk_once("noswapaccount is deprecated and will be removed in 2.6.40. Use swapaccount=0 instead\n"); enable_swap_account("=0"); return 1; } -- cgit v1.2.3 From 57fc4a5ee322cde96c33f101d3c2d3b79011c05c Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 1 Feb 2011 15:52:32 -0800 Subject: mm: when migrate_pages returns 0, all pages must have been released In some cases migrate_pages could return zero while still leaving a few pages in the pagelist (and some caller wouldn't notice it has to call putback_lru_pages after commit cf608ac19c9 ("mm: compaction: fix COMPACTPAGEFAILED counting")). Add one missing putback_lru_pages not added by commit cf608ac19c95 ("mm: compaction: fix COMPACTPAGEFAILED counting"). Signed-off-by: Andrea Arcangeli Signed-off-by: Minchan Kim Reviewed-by: Minchan Kim Cc: Christoph Lameter Acked-by: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 1 + mm/migrate.c | 3 +-- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 548fbd70f026..75398b0bfede 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1419,6 +1419,7 @@ int soft_offline_page(struct page *page, int flags) ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0, true); if (ret) { + putback_lru_pages(&pagelist); pr_info("soft offline: %#lx: migration failed %d, type %lx\n", pfn, ret, page->flags); if (ret > 0) diff --git a/mm/migrate.c b/mm/migrate.c index 9f29a3b7aac2..155a2e9a8059 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -772,6 +772,7 @@ uncharge: unlock: unlock_page(page); +move_newpage: if (rc != -EAGAIN) { /* * A page that has been migrated has all references @@ -785,8 +786,6 @@ unlock: putback_lru_page(page); } -move_newpage: - /* * Move the new page to the LRU. If migration was not successful * then this will free the page. -- cgit v1.2.3 From 48db54ee2f41e8ae2faf330b55db34a9fffb5b3c Mon Sep 17 00:00:00 2001 From: Minchan Kim Date: Tue, 1 Feb 2011 15:52:33 -0800 Subject: mm/migration: fix page corruption during hugepage migration If migrate_huge_page by memory-failure fails , it calls put_page in itself to decrease page reference and caller of migrate_huge_page also calls putback_lru_pages. It can do double free of page so it can make page corruption on page holder. In addtion, clean of pages on caller is consistent behavior with migrate_pages by cf608ac19c ("mm: compaction: fix COMPACTPAGEFAILED counting"). Signed-off-by: Minchan Kim Cc: Andrea Arcangeli Cc: Christoph Lameter Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 5 ++++- mm/migrate.c | 4 ---- 2 files changed, 4 insertions(+), 5 deletions(-) (limited to 'mm') diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 75398b0bfede..237aaa488f4e 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -1295,7 +1295,10 @@ static int soft_offline_huge_page(struct page *page, int flags) ret = migrate_huge_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL, 0, true); if (ret) { - putback_lru_pages(&pagelist); + struct page *page1, *page2; + list_for_each_entry_safe(page1, page2, &pagelist, lru) + put_page(page1); + pr_debug("soft offline: %#lx: migration failed %d, type %lx\n", pfn, ret, page->flags); if (ret > 0) diff --git a/mm/migrate.c b/mm/migrate.c index 155a2e9a8059..766115253807 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -980,10 +980,6 @@ int migrate_huge_pages(struct list_head *from, } rc = 0; out: - - list_for_each_entry_safe(page, page2, from, lru) - put_page(page); - if (rc) return rc; -- cgit v1.2.3 From efeda7a41e09efce506a68c3549b60b16dd7dedd Mon Sep 17 00:00:00 2001 From: Jin Dongming Date: Tue, 1 Feb 2011 15:52:39 -0800 Subject: thp: fix splitting of hwpoisoned hugepages The poisoned THP is now split with split_huge_page() in collect_procs_anon(). If kmalloc() is failed in collect_procs(), split_huge_page() could not be called. And the work after split_huge_page() for collecting the processes using poisoned page will not be done, too. So the processes using the poisoned page could not be killed. The condition becomes worse when CONFIG_DEBUG_VM == "Y". Because the poisoned THP could not be split, system panic will be caused by VM_BUG_ON(PageTransHuge(page)) in try_to_unmap(). This patch does: 1. move split_huge_page() to the place before collect_procs(). This can be sure the failure of splitting THP is caused by itself. 2. when splitting THP is failed, stop the operations after it. This can avoid unexpected system panic or non sense works. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Jin Dongming Reviewed-by: Hidetoshi Seto Cc: Andrea Arcangeli Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 30 ++++++++++++++++++++++++++++-- 1 file changed, 28 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 237aaa488f4e..1e9c30b241c3 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -386,8 +386,6 @@ static void collect_procs_anon(struct page *page, struct list_head *to_kill, struct task_struct *tsk; struct anon_vma *av; - if (!PageHuge(page) && unlikely(split_huge_page(page))) - return; read_lock(&tasklist_lock); av = page_lock_anon_vma(page); if (av == NULL) /* Not actually mapped anymore */ @@ -896,6 +894,34 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, } } + if (PageTransHuge(hpage)) { + /* + * Verify that this isn't a hugetlbfs head page, the check for + * PageAnon is just for avoid tripping a split_huge_page + * internal debug check, as split_huge_page refuses to deal with + * anything that isn't an anon page. PageAnon can't go away fro + * under us because we hold a refcount on the hpage, without a + * refcount on the hpage. split_huge_page can't be safely called + * in the first place, having a refcount on the tail isn't + * enough * to be safe. + */ + if (!PageHuge(hpage) && PageAnon(hpage)) { + if (unlikely(split_huge_page(hpage))) { + /* + * FIXME: if splitting THP is failed, it is + * better to stop the following operation rather + * than causing panic by unmapping. System might + * survive if the page is freed later. + */ + printk(KERN_INFO + "MCE %#lx: failed to split THP\n", pfn); + + BUG_ON(!PageHWPoison(p)); + return SWAP_FAIL; + } + } + } + /* * First collect all the processes that have the page * mapped in dirty form. This has to be done before try_to_unmap, -- cgit v1.2.3 From a6d30dddae4648837be5a0c0cb2c0ae9ad0377db Mon Sep 17 00:00:00 2001 From: Jin Dongming Date: Tue, 1 Feb 2011 15:52:40 -0800 Subject: thp: fix the wrong reported address of hwpoisoned hugepages When the tail page of THP is poisoned, the head page will be poisoned too. And the wrong address, address of head page, will be sent with sigbus always. So when the poisoned page is used by Guest OS which is running on KVM, after the address changing(hva->gpa) by qemu, the unexpected process on Guest OS will be killed by sigbus. What we expected is that the process using the poisoned tail page could be killed on Guest OS, but not that the process using the healthy head page is killed. Since it is not good to poison the healthy page, avoid poisoning other than the page which is really poisoned. (While we poison all pages in a huge page in case of hugetlb, we can do this for THP thanks to split_huge_page().) Here we fix two parts: 1. Isolate the poisoned page only to make sure the reported address is the address of poisoned page. 2. make the poisoned page work as the poisoned regular page. [akpm@linux-foundation.org: fix spello in comment] Signed-off-by: Jin Dongming Reviewed-by: Hidetoshi Seto Cc: Andrea Arcangeli Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/huge_memory.c | 7 ++++++- mm/memory-failure.c | 27 ++++++++++++++++++++++----- 2 files changed, 28 insertions(+), 6 deletions(-) (limited to 'mm') diff --git a/mm/huge_memory.c b/mm/huge_memory.c index e187454d82f6..b6c1ce3c53b5 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1162,7 +1162,12 @@ static void __split_huge_page_refcount(struct page *page) /* after clearing PageTail the gup refcount can be released */ smp_mb(); - page_tail->flags &= ~PAGE_FLAGS_CHECK_AT_PREP; + /* + * retain hwpoison flag of the poisoned tail page: + * fix for the unsuitable process killed on Guest Machine(KVM) + * by the memory-failure. + */ + page_tail->flags &= ~PAGE_FLAGS_CHECK_AT_PREP | __PG_HWPOISON; page_tail->flags |= (page->flags & ((1L << PG_referenced) | (1L << PG_swapbacked) | diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 1e9c30b241c3..04158d6f44d4 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -854,6 +854,7 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, int ret; int kill = 1; struct page *hpage = compound_head(p); + struct page *ppage; if (PageReserved(p) || PageSlab(p)) return SWAP_SUCCESS; @@ -894,6 +895,14 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, } } + /* + * ppage: poisoned page + * if p is regular page(4k page) + * ppage == real poisoned page; + * else p is hugetlb or THP, ppage == head page. + */ + ppage = hpage; + if (PageTransHuge(hpage)) { /* * Verify that this isn't a hugetlbfs head page, the check for @@ -919,6 +928,8 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, BUG_ON(!PageHWPoison(p)); return SWAP_FAIL; } + /* THP is split, so ppage should be the real poisoned page. */ + ppage = p; } } @@ -931,12 +942,18 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, * there's nothing that can be done. */ if (kill) - collect_procs(hpage, &tokill); + collect_procs(ppage, &tokill); - ret = try_to_unmap(hpage, ttu); + if (hpage != ppage) + lock_page_nosync(ppage); + + ret = try_to_unmap(ppage, ttu); if (ret != SWAP_SUCCESS) printk(KERN_ERR "MCE %#lx: failed to unmap page (mapcount=%d)\n", - pfn, page_mapcount(hpage)); + pfn, page_mapcount(ppage)); + + if (hpage != ppage) + unlock_page(ppage); /* * Now that the dirty bit has been propagated to the @@ -947,7 +964,7 @@ static int hwpoison_user_mappings(struct page *p, unsigned long pfn, * use a more force-full uncatchable kill to prevent * any accesses to the poisoned memory. */ - kill_procs_ao(&tokill, !!PageDirty(hpage), trapno, + kill_procs_ao(&tokill, !!PageDirty(ppage), trapno, ret != SWAP_SUCCESS, p, pfn); return ret; @@ -1090,7 +1107,7 @@ int __memory_failure(unsigned long pfn, int trapno, int flags) * For error on the tail page, we should set PG_hwpoison * on the head page to show that the hugepage is hwpoisoned */ - if (PageTail(p) && TestSetPageHWPoison(hpage)) { + if (PageHuge(p) && PageTail(p) && TestSetPageHWPoison(hpage)) { action_result(pfn, "hugepage already hardware poisoned", IGNORED); unlock_page(hpage); -- cgit v1.2.3 From af241a083404acda7ba3690e5b7697949d729fcc Mon Sep 17 00:00:00 2001 From: Jin Dongming Date: Tue, 1 Feb 2011 15:52:41 -0800 Subject: thp: fix unsuitable behavior for hwpoisoned tail page When a tail page of THP is poisoned, memory-failure will do nothing except setting PG_hwpoison, while the expected behavior is that the process, who is using the poisoned tail page, should be killed. The above problem is caused by lru check of the poisoned tail page of THP. Because PG_lru flag is only set on the head page of THP, the check always consider the poisoned tail page as NON lru page. So the lru check for the tail page of THP should be avoided, as like as hugetlb. This patch adds !PageTransCompound() before lru check for THP, because of the check (!PageHuge() && !PageTransCompound()) the whole branch could be optimized away at build time when both hugetlbfs and THP are set with "N" (or in archs not supporting either of those). [akpm@linux-foundation.org: fix unrelated typo in shake_page() comment] Signed-off-by: Jin Dongming Reviewed-by: Hidetoshi Seto Cc: Andrea Arcangeli Cc: Andi Kleen Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory-failure.c | 31 +++++++++++++++++-------------- 1 file changed, 17 insertions(+), 14 deletions(-) (limited to 'mm') diff --git a/mm/memory-failure.c b/mm/memory-failure.c index 04158d6f44d4..0207c2f6f8bd 100644 --- a/mm/memory-failure.c +++ b/mm/memory-failure.c @@ -233,8 +233,8 @@ void shake_page(struct page *p, int access) } /* - * Only all shrink_slab here (which would also - * shrink other caches) if access is not potentially fatal. + * Only call shrink_slab here (which would also shrink other caches) if + * access is not potentially fatal. */ if (access) { int nr; @@ -1065,19 +1065,22 @@ int __memory_failure(unsigned long pfn, int trapno, int flags) * The check (unnecessarily) ignores LRU pages being isolated and * walked by the page reclaim code, however that's not a big loss. */ - if (!PageLRU(p) && !PageHuge(p)) - shake_page(p, 0); - if (!PageLRU(p) && !PageHuge(p)) { - /* - * shake_page could have turned it free. - */ - if (is_free_buddy_page(p)) { - action_result(pfn, "free buddy, 2nd try", DELAYED); - return 0; + if (!PageHuge(p) && !PageTransCompound(p)) { + if (!PageLRU(p)) + shake_page(p, 0); + if (!PageLRU(p)) { + /* + * shake_page could have turned it free. + */ + if (is_free_buddy_page(p)) { + action_result(pfn, "free buddy, 2nd try", + DELAYED); + return 0; + } + action_result(pfn, "non LRU", IGNORED); + put_page(p); + return -EBUSY; } - action_result(pfn, "non LRU", IGNORED); - put_page(p); - return -EBUSY; } /* -- cgit v1.2.3 From 9221edb7120e2dc3ae90f1c58514979f7ba40e46 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 1 Feb 2011 15:52:42 -0800 Subject: memcg: prevent endless loop when charging huge pages The charging code can encounter a charge size that is bigger than a regular page in two situations: one is a batched charge to fill the per-cpu stocks, the other is a huge page charge. This code is distributed over two functions, however, and only the outer one is aware of huge pages. In case the charging fails, the inner function will tell the outer function to retry if the charge size is bigger than regular pages--assuming batched charging is the only case. And the outer function will retry forever charging a huge page. This patch makes sure the inner function can distinguish between batch charging and a single huge page charge. It will only signal another attempt if batch charging failed, and go into regular reclaim when it is called on behalf of a huge page. Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Reviewed-by: Minchan Kim Cc: Balbir Singh Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 79abb1fd39d2..50eb50e100fd 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1837,8 +1837,15 @@ static int __mem_cgroup_do_charge(struct mem_cgroup *mem, gfp_t gfp_mask, flags |= MEM_CGROUP_RECLAIM_NOSWAP; } else mem_over_limit = mem_cgroup_from_res_counter(fail_res, res); - - if (csize > PAGE_SIZE) /* change csize and retry */ + /* + * csize can be either a huge page (HPAGE_SIZE), a batch of + * regular pages (CHARGE_SIZE), or a single regular page + * (PAGE_SIZE). + * + * Never reclaim on behalf of optional batching, retry with a + * single page instead. + */ + if (csize == CHARGE_SIZE) return CHARGE_RETRY; if (!(gfp_mask & __GFP_WAIT)) -- cgit v1.2.3 From 19942822df65ee4a47c2e6d6d70cace1b7f01710 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 1 Feb 2011 15:52:43 -0800 Subject: memcg: prevent endless loop when charging huge pages to near-limit group If reclaim after a failed charging was unsuccessful, the limits are checked again, just in case they settled by means of other tasks. This is all fine as long as every charge is of size PAGE_SIZE, because in that case, being below the limit means having at least PAGE_SIZE bytes available. But with transparent huge pages, we may end up in an endless loop where charging and reclaim fail, but we keep going because the limits are not yet exceeded, although not allowing for a huge page. Fix this up by explicitely checking for enough room, not just whether we are within limits. Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Reviewed-by: Minchan Kim Cc: Balbir Singh Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 35 ++++++++++++++++++++++++++++------- 1 file changed, 28 insertions(+), 7 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 50eb50e100fd..0e81eb5f0aea 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -1111,6 +1111,23 @@ static bool mem_cgroup_check_under_limit(struct mem_cgroup *mem) return false; } +/** + * mem_cgroup_check_margin - check if the memory cgroup allows charging + * @mem: memory cgroup to check + * @bytes: the number of bytes the caller intends to charge + * + * Returns a boolean value on whether @mem can be charged @bytes or + * whether this would exceed the limit. + */ +static bool mem_cgroup_check_margin(struct mem_cgroup *mem, unsigned long bytes) +{ + if (!res_counter_check_margin(&mem->res, bytes)) + return false; + if (do_swap_account && !res_counter_check_margin(&mem->memsw, bytes)) + return false; + return true; +} + static unsigned int get_swappiness(struct mem_cgroup *memcg) { struct cgroup *cgrp = memcg->css.cgroup; @@ -1852,15 +1869,19 @@ static int __mem_cgroup_do_charge(struct mem_cgroup *mem, gfp_t gfp_mask, return CHARGE_WOULDBLOCK; ret = mem_cgroup_hierarchical_reclaim(mem_over_limit, NULL, - gfp_mask, flags); + gfp_mask, flags); + if (mem_cgroup_check_margin(mem_over_limit, csize)) + return CHARGE_RETRY; /* - * try_to_free_mem_cgroup_pages() might not give us a full - * picture of reclaim. Some pages are reclaimed and might be - * moved to swap cache or just unmapped from the cgroup. - * Check the limit again to see if the reclaim reduced the - * current usage of the cgroup before giving up + * Even though the limit is exceeded at this point, reclaim + * may have been able to free some pages. Retry the charge + * before killing the task. + * + * Only for regular pages, though: huge pages are rather + * unlikely to succeed so close to the limit, and we fall back + * to regular pages anyway in case of failure. */ - if (ret || mem_cgroup_check_under_limit(mem_over_limit)) + if (csize == PAGE_SIZE && ret) return CHARGE_RETRY; /* -- cgit v1.2.3 From 8493ae439f7038b502df1d687e61dde54c27ca92 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Tue, 1 Feb 2011 15:52:44 -0800 Subject: memcg: never OOM when charging huge pages Huge page coverage should obviously have less priority than the continued execution of a process. Never kill a process when charging it a huge page fails. Instead, give up after the first failed reclaim attempt and fall back to regular pages. Signed-off-by: Johannes Weiner Acked-by: KAMEZAWA Hiroyuki Reviewed-by: Minchan Kim Cc: Balbir Singh Cc: Daisuke Nishimura Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 0e81eb5f0aea..fc75f34ba609 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -2351,13 +2351,19 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm, gfp_t gfp_mask, enum charge_type ctype) { struct mem_cgroup *mem = NULL; + int page_size = PAGE_SIZE; struct page_cgroup *pc; + bool oom = true; int ret; - int page_size = PAGE_SIZE; if (PageTransHuge(page)) { page_size <<= compound_order(page); VM_BUG_ON(!PageTransHuge(page)); + /* + * Never OOM-kill a process for a huge page. The + * fault handler will fall back to regular pages. + */ + oom = false; } pc = lookup_page_cgroup(page); @@ -2366,7 +2372,7 @@ static int mem_cgroup_charge_common(struct page *page, struct mm_struct *mm, return 0; prefetchw(pc); - ret = __mem_cgroup_try_charge(mm, gfp_mask, &mem, true, page_size); + ret = __mem_cgroup_try_charge(mm, gfp_mask, &mem, oom, page_size); if (ret || !mem) return ret; -- cgit v1.2.3 From 3751d60430fe4c26460a5ca8ad8672d32f93bcb1 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Tue, 1 Feb 2011 15:52:45 -0800 Subject: memcg: fix event counting breakage from recent THP update Changes in e401f1761 ("memcg: modify accounting function for supporting THP better") adds nr_pages to support multiple page size in memory_cgroup_charge_statistics. But counting the number of event nees abs(nr_pages) for increasing counters. This patch fixes event counting. Signed-off-by: KAMEZAWA Hiroyuki Reviewed-by: Johannes Weiner Cc: Daisuke Nishimura Cc: Balbir Singh Cc: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memcontrol.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/memcontrol.c b/mm/memcontrol.c index fc75f34ba609..da53a252b259 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -612,8 +612,10 @@ static void mem_cgroup_charge_statistics(struct mem_cgroup *mem, /* pagein of a big page is an event. So, ignore page size */ if (nr_pages > 0) __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_PGPGIN_COUNT]); - else + else { __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_PGPGOUT_COUNT]); + nr_pages = -nr_pages; /* for event */ + } __this_cpu_add(mem->stat->count[MEM_CGROUP_EVENTS], nr_pages); -- cgit v1.2.3 From e6d2e2b2b1e1455df16d68a78f4a3874c7b3ad20 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Thu, 10 Feb 2011 15:01:30 -0800 Subject: memblock: don't adjust size in memblock_find_base() While applying patch to use memblock to find aperture for 64bit x86. Ingo found system with 1g + force_iommu > No AGP bridge found > Node 0: aperture @ 38000000 size 32 MB > Aperture pointing to e820 RAM. Ignoring. > Your BIOS doesn't leave a aperture memory hole > Please enable the IOMMU option in the BIOS setup > This costs you 64 MB of RAM > Cannot allocate aperture memory hole (0,65536K) the corresponding code: addr = memblock_find_in_range(0, 1ULL<<32, aper_size, 512ULL<<20); if (addr == MEMBLOCK_ERROR || addr + aper_size > 0xffffffff) { printk(KERN_ERR "Cannot allocate aperture memory hole (%lx,%uK)\n", addr, aper_size>>10); return 0; } memblock_x86_reserve_range(addr, addr + aper_size, "aperture64") fails because memblock core code align the size with 512M. That could make size way too big. So don't align the size in that case. actually __memblock_alloc_base, the another caller already align that before calling that function. BTW. x86 does not use __memblock_alloc_base... Signed-off-by: Yinghai Lu Cc: Ingo Molnar Cc: David Miller Cc: "H. Peter Anvin" Cc: Benjamin Herrenschmidt Cc: Dave Airlie Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memblock.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'mm') diff --git a/mm/memblock.c b/mm/memblock.c index bdba245d8afd..4618fda975a0 100644 --- a/mm/memblock.c +++ b/mm/memblock.c @@ -137,8 +137,6 @@ static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size, BUG_ON(0 == size); - size = memblock_align_up(size, align); - /* Pump up max_addr */ if (end == MEMBLOCK_ALLOC_ACCESSIBLE) end = memblock.current_limit; -- cgit v1.2.3 From e15f8c01af924e611bc7be1e45449c4a74e5dfdd Mon Sep 17 00:00:00 2001 From: Michel Lespinasse Date: Thu, 10 Feb 2011 15:01:32 -0800 Subject: mlock: fix race when munlocking pages in do_wp_page() vmscan can lazily find pages that are mapped within VM_LOCKED vmas, and set the PageMlocked bit on these pages, transfering them onto the unevictable list. When do_wp_page() breaks COW within a VM_LOCKED vma, it may need to clear PageMlocked on the old page and set it on the new page instead. This change fixes an issue where do_wp_page() was clearing PageMlocked on the old page while the pte was still pointing to it (as well as rmap). Therefore, we were not protected against vmscan immediately transfering the old page back onto the unevictable list. This could cause pages to get stranded there forever. I propose to move the corresponding code to the end of do_wp_page(), after the pte (and rmap) have been pointed to the new page. Additionally, we can use munlock_vma_page() instead of clear_page_mlock(), so that the old page stays mlocked if there are still other VM_LOCKED vmas mapping it. Signed-off-by: Michel Lespinasse Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Hugh Dickins Cc: Rik van Riel Cc: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory.c | 26 ++++++++++++-------------- 1 file changed, 12 insertions(+), 14 deletions(-) (limited to 'mm') diff --git a/mm/memory.c b/mm/memory.c index 31250faff390..32df03cf13a5 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2219,7 +2219,6 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, &ptl); if (!pte_same(*page_table, orig_pte)) { unlock_page(old_page); - page_cache_release(old_page); goto unlock; } page_cache_release(old_page); @@ -2289,7 +2288,6 @@ static int do_wp_page(struct mm_struct *mm, struct vm_area_struct *vma, &ptl); if (!pte_same(*page_table, orig_pte)) { unlock_page(old_page); - page_cache_release(old_page); goto unlock; } @@ -2367,16 +2365,6 @@ gotten: } __SetPageUptodate(new_page); - /* - * Don't let another task, with possibly unlocked vma, - * keep the mlocked page. - */ - if ((vma->vm_flags & VM_LOCKED) && old_page) { - lock_page(old_page); /* for LRU manipulation */ - clear_page_mlock(old_page); - unlock_page(old_page); - } - if (mem_cgroup_newpage_charge(new_page, mm, GFP_KERNEL)) goto oom_free_new; @@ -2444,10 +2432,20 @@ gotten: if (new_page) page_cache_release(new_page); - if (old_page) - page_cache_release(old_page); unlock: pte_unmap_unlock(page_table, ptl); + if (old_page) { + /* + * Don't let another task, with possibly unlocked vma, + * keep the mlocked page. + */ + if ((ret & VM_FAULT_WRITE) && (vma->vm_flags & VM_LOCKED)) { + lock_page(old_page); /* LRU manipulation */ + munlock_vma_page(old_page); + unlock_page(old_page); + } + page_cache_release(old_page); + } return ret; oom_free_new: page_cache_release(new_page); -- cgit v1.2.3 From 419d8c96dbfa558f00e623023917d0a5afc46129 Mon Sep 17 00:00:00 2001 From: Michel Lespinasse Date: Thu, 10 Feb 2011 15:01:33 -0800 Subject: mlock: do not munlock pages in __do_fault() If the page is going to be written to, __do_page needs to break COW. However, the old page (before breaking COW) was never mapped mapped into the current pte (__do_fault is only called when the pte is not present), so vmscan can't have marked the old page as PageMlocked due to being mapped in __do_fault's VMA. Therefore, __do_fault() does not need to worry about clearing PageMlocked() on the old page. Signed-off-by: Michel Lespinasse Reviewed-by: KAMEZAWA Hiroyuki Acked-by: Hugh Dickins Cc: Rik van Riel Cc: Andrea Arcangeli Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/memory.c | 6 ------ 1 file changed, 6 deletions(-) (limited to 'mm') diff --git a/mm/memory.c b/mm/memory.c index 32df03cf13a5..8e8c18324863 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -3051,12 +3051,6 @@ static int __do_fault(struct mm_struct *mm, struct vm_area_struct *vma, goto out; } charged = 1; - /* - * Don't let another task, with possibly unlocked vma, - * keep the mlocked page. - */ - if (vma->vm_flags & VM_LOCKED) - clear_page_mlock(vmf.page); copy_user_highpage(page, vmf.page, address, vma); __SetPageUptodate(page); } else { -- cgit v1.2.3 From f0fdc5e8e6f579310458aef43d1610a0bb5e81a4 Mon Sep 17 00:00:00 2001 From: Johannes Weiner Date: Thu, 10 Feb 2011 15:01:34 -0800 Subject: vmscan: fix zone shrinking exit when scan work is done Commit 3e7d34497067 ("mm: vmscan: reclaim order-0 and use compaction instead of lumpy reclaim") introduced an indefinite loop in shrink_zone(). It meant to break out of this loop when no pages had been reclaimed and not a single page was even scanned. The way it would detect the latter is by taking a snapshot of sc->nr_scanned at the beginning of the function and comparing it against the new sc->nr_scanned after the scan loop. But it would re-iterate without updating that snapshot, looping forever if sc->nr_scanned changed at least once since shrink_zone() was invoked. This is not the sole condition that would exit that loop, but it requires other processes to change the zone state, as the reclaimer that is stuck obviously can not anymore. This is only happening for higher-order allocations, where reclaim is run back to back with compaction. Signed-off-by: Johannes Weiner Reported-by: Michal Hocko Tested-by: Kent Overstreet Reported-by: Kent Overstreet Acked-by: Mel Gorman Cc: Andrea Arcangeli Cc: Rik van Riel Reviewed-by: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/vmscan.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'mm') diff --git a/mm/vmscan.c b/mm/vmscan.c index 148c6e630df2..17497d0cd8b9 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1882,12 +1882,12 @@ static void shrink_zone(int priority, struct zone *zone, unsigned long nr[NR_LRU_LISTS]; unsigned long nr_to_scan; enum lru_list l; - unsigned long nr_reclaimed; + unsigned long nr_reclaimed, nr_scanned; unsigned long nr_to_reclaim = sc->nr_to_reclaim; - unsigned long nr_scanned = sc->nr_scanned; restart: nr_reclaimed = 0; + nr_scanned = sc->nr_scanned; get_scan_count(zone, sc, nr, priority); while (nr[LRU_INACTIVE_ANON] || nr[LRU_ACTIVE_FILE] || -- cgit v1.2.3 From 678ff896a37afdbca292c7846ec895463aed35a5 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Thu, 10 Feb 2011 15:01:36 -0800 Subject: memcg: fix leak of accounting at failure path of hugepage collapsing mem_cgroup_uncharge_page() should be called in all failure cases after mem_cgroup_charge_newpage() is called in huge_memory.c::collapse_huge_page() [ 4209.076861] BUG: Bad page state in process khugepaged pfn:1e9800 [ 4209.077601] page:ffffea0006b14000 count:0 mapcount:0 mapping: (null) index:0x2800 [ 4209.078674] page flags: 0x40000000004000(head) [ 4209.079294] pc:ffff880214a30000 pc->flags:2146246697418756 pc->mem_cgroup:ffffc9000177a000 [ 4209.082177] (/A) [ 4209.082500] Pid: 31, comm: khugepaged Not tainted 2.6.38-rc3-mm1 #1 [ 4209.083412] Call Trace: [ 4209.083678] [] ? bad_page+0xe4/0x140 [ 4209.084240] [] ? free_pages_prepare+0xd6/0x120 [ 4209.084837] [] ? rwsem_down_failed_common+0xbd/0x150 [ 4209.085509] [] ? __free_pages_ok+0x32/0xe0 [ 4209.086110] [] ? free_compound_page+0x1b/0x20 [ 4209.086699] [] ? __put_compound_page+0x1c/0x30 [ 4209.087333] [] ? put_compound_page+0x4d/0x200 [ 4209.087935] [] ? put_page+0x45/0x50 [ 4209.097361] [] ? khugepaged+0x9e9/0x1430 [ 4209.098364] [] ? autoremove_wake_function+0x0/0x40 [ 4209.099121] [] ? khugepaged+0x0/0x1430 [ 4209.099780] [] ? kthread+0x96/0xa0 [ 4209.100452] [] ? kernel_thread_helper+0x4/0x10 [ 4209.101214] [] ? kthread+0x0/0xa0 [ 4209.101842] [] ? kernel_thread_helper+0x0/0x10 Signed-off-by: KAMEZAWA Hiroyuki Acked-by: Daisuke Nishimura Reviewed-by: Johannes Weiner Cc: Andrea Arcangeli Reviewed-by: Minchan Kim Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/huge_memory.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'mm') diff --git a/mm/huge_memory.c b/mm/huge_memory.c index b6c1ce3c53b5..e62ddb8f24b6 100644 --- a/mm/huge_memory.c +++ b/mm/huge_memory.c @@ -1852,7 +1852,6 @@ static void collapse_huge_page(struct mm_struct *mm, set_pmd_at(mm, address, pmd, _pmd); spin_unlock(&mm->page_table_lock); anon_vma_unlock(vma->anon_vma); - mem_cgroup_uncharge_page(new_page); goto out; } @@ -1898,6 +1897,7 @@ out_up_write: return; out: + mem_cgroup_uncharge_page(new_page); #ifdef CONFIG_NUMA put_page(new_page); #endif -- cgit v1.2.3 From a7d6e4ecdb7648478ddec76d30d87d03d6e22b31 Mon Sep 17 00:00:00 2001 From: Andrea Arcangeli Date: Tue, 15 Feb 2011 19:02:45 +0100 Subject: thp: prevent hugepages during args/env copying into the user stack Transparent hugepages can only be created if rmap is fully functional. So we must prevent hugepages to be created while is_vma_temporary_stack() is true. This also optmizes away some harmless but unnecessary setting of khugepaged_scan.address and it switches some BUG_ON to VM_BUG_ON. Signed-off-by: Andrea Arcangeli Acked-by: Rik van Riel Signed-off-by: Linus Torvalds --- mm/huge_memory.c | 35 ++++++++++++++++-------------