From: KOSAKI Motohiro Subject: mm: fix Committed_AS underflow on large NR_CPUS environment Patch-mainline: 2.6.30-rc5 Git-commit: 00a62ce91e554198ef28234c91c36f850f5a3bc9 References: bnc#505831 mm: fix Committed_AS underflow on large NR_CPUS environment The Committed_AS field can underflow in certain situations: > # while true; do cat /proc/meminfo | grep _AS; sleep 1; done | uniq -c > 1 Committed_AS: 18446744073709323392 kB > 11 Committed_AS: 18446744073709455488 kB > 6 Committed_AS: 35136 kB > 5 Committed_AS: 18446744073709454400 kB > 7 Committed_AS: 35904 kB > 3 Committed_AS: 18446744073709453248 kB > 2 Committed_AS: 34752 kB > 9 Committed_AS: 18446744073709453248 kB > 8 Committed_AS: 34752 kB > 3 Committed_AS: 18446744073709320960 kB > 7 Committed_AS: 18446744073709454080 kB > 3 Committed_AS: 18446744073709320960 kB > 5 Committed_AS: 18446744073709454080 kB > 6 Committed_AS: 18446744073709320960 kB Because NR_CPUS can be greater than 1000 and meminfo_proc_show() does not check for underflow. But NR_CPUS proportional isn't good calculation. In general, possibility of lock contention is proportional to the number of online cpus, not theorical maximum cpus (NR_CPUS). The current kernel has generic percpu-counter stuff. using it is right way. it makes code simplify and percpu_counter_read_positive() don't make underflow issue. Reported-by: Dave Hansen Signed-off-by: KOSAKI Motohiro Cc: Eric B Munson Cc: Mel Gorman Cc: Christoph Lameter Cc: [All kernel versions] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds Backported-by: Jeff Mahoney Signed-off-by: Jeff Mahoney --- fs/proc/proc_misc.c | 2 +- include/linux/mman.h | 9 +++------ kernel/fork.c | 2 ++ mm/mmap.c | 8 ++------ mm/nommu.c | 9 +++------ mm/swap.c | 46 ---------------------------------------------- 6 files changed, 11 insertions(+), 65 deletions(-) --- a/fs/proc/proc_misc.c +++ b/fs/proc/proc_misc.c @@ -145,7 +145,7 @@ static int meminfo_read_proc(char *page, #define K(x) ((x) << (PAGE_SHIFT - 10)) si_meminfo(&i); si_swapinfo(&i); - committed = atomic_long_read(&vm_committed_space); + committed = percpu_counter_read_positive(&vm_committed_as); allowed = ((totalram_pages - hugetlb_total_pages()) * sysctl_overcommit_ratio / 100) + total_swap_pages; --- a/include/linux/mman.h +++ b/include/linux/mman.h @@ -12,21 +12,18 @@ #ifdef __KERNEL__ #include +#include #include extern int sysctl_overcommit_memory; extern int sysctl_overcommit_ratio; -extern atomic_long_t vm_committed_space; +extern struct percpu_counter vm_committed_as; -#ifdef CONFIG_SMP -extern void vm_acct_memory(long pages); -#else static inline void vm_acct_memory(long pages) { - atomic_long_add(pages, &vm_committed_space); + percpu_counter_add(&vm_committed_as, pages); } -#endif static inline void vm_unacct_memory(long pages) { --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1442,6 +1442,8 @@ void __init proc_caches_init(void) mm_cachep = kmem_cache_create("mm_struct", sizeof(struct mm_struct), ARCH_MIN_MMSTRUCT_ALIGN, SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL); + if (percpu_counter_init(&vm_committed_as, 0)) + panic("Failed to allocate vm_committed_as"); } /* --- a/mm/mmap.c +++ b/mm/mmap.c @@ -84,7 +84,7 @@ 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; int heap_stack_gap __read_mostly = 1; /* @@ -178,11 +178,7 @@ int __vm_enough_memory(struct mm_struct leave 3% of the size of this process for other processes */ 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); --- a/mm/nommu.c +++ b/mm/nommu.c @@ -39,7 +39,7 @@ struct page *mem_map; unsigned long max_mapnr; unsigned long num_physpages; unsigned long askedalloc, realalloc; -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; @@ -1434,12 +1434,9 @@ int __vm_enough_memory(struct mm_struct leave 3% of the size of this process for other processes */ allowed -= current->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); --- a/mm/swap.c +++ b/mm/swap.c @@ -474,49 +474,6 @@ unsigned pagevec_lookup_tag(struct pagev 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 */ @@ -537,7 +494,4 @@ 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 }