diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2017-03-10 08:34:42 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2017-03-10 08:34:42 -0800 |
commit | 8fe3ccaed080a7804bc459c42e0419253973be92 (patch) | |
tree | d13a8bcb563cea7ab2a167dc504c7ae5ac887cbb /mm/kasan | |
parent | 9db61d6fd65f53eaee13fbb451fd761ce926dea9 (diff) | |
parent | 2378cd6181edd94748d699448823609977283b11 (diff) |
Merge branch 'akpm' (patches from Andrew)
Merge fixes from Andrew Morton:
"26 fixes"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (26 commits)
userfaultfd: remove wrong comment from userfaultfd_ctx_get()
fat: fix using uninitialized fields of fat_inode/fsinfo_inode
sh: cayman: IDE support fix
kasan: fix races in quarantine_remove_cache()
kasan: resched in quarantine_remove_cache()
mm: do not call mem_cgroup_free() from within mem_cgroup_alloc()
thp: fix another corner case of munlock() vs. THPs
rmap: fix NULL-pointer dereference on THP munlocking
mm/memblock.c: fix memblock_next_valid_pfn()
userfaultfd: selftest: vm: allow to build in vm/ directory
userfaultfd: non-cooperative: userfaultfd_remove revalidate vma in MADV_DONTNEED
userfaultfd: non-cooperative: fix fork fctx->new memleak
mm/cgroup: avoid panic when init with low memory
drivers/md/bcache/util.h: remove duplicate inclusion of blkdev.h
mm/vmstats: add thp_split_pud event for clarity
include/linux/fs.h: fix unsigned enum warning with gcc-4.2
userfaultfd: non-cooperative: release all ctx in dup_userfaultfd_complete
userfaultfd: non-cooperative: robustness check
userfaultfd: non-cooperative: rollback userfaultfd_exit
x86, mm: unify exit paths in gup_pte_range()
...
Diffstat (limited to 'mm/kasan')
-rw-r--r-- | mm/kasan/quarantine.c | 51 |
1 files changed, 44 insertions, 7 deletions
diff --git a/mm/kasan/quarantine.c b/mm/kasan/quarantine.c index 6f1ed1630873..3a8ddf8baf7d 100644 --- a/mm/kasan/quarantine.c +++ b/mm/kasan/quarantine.c @@ -25,6 +25,7 @@ #include <linux/printk.h> #include <linux/shrinker.h> #include <linux/slab.h> +#include <linux/srcu.h> #include <linux/string.h> #include <linux/types.h> @@ -103,6 +104,7 @@ static int quarantine_tail; /* Total size of all objects in global_quarantine across all batches. */ static unsigned long quarantine_size; static DEFINE_SPINLOCK(quarantine_lock); +DEFINE_STATIC_SRCU(remove_cache_srcu); /* Maximum size of the global queue. */ static unsigned long quarantine_max_size; @@ -173,17 +175,22 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache) struct qlist_head *q; struct qlist_head temp = QLIST_INIT; + /* + * Note: irq must be disabled until after we move the batch to the + * global quarantine. Otherwise quarantine_remove_cache() can miss + * some objects belonging to the cache if they are in our local temp + * list. quarantine_remove_cache() executes on_each_cpu() at the + * beginning which ensures that it either sees the objects in per-cpu + * lists or in the global quarantine. + */ local_irq_save(flags); q = this_cpu_ptr(&cpu_quarantine); qlist_put(q, &info->quarantine_link, cache->size); - if (unlikely(q->bytes > QUARANTINE_PERCPU_SIZE)) + if (unlikely(q->bytes > QUARANTINE_PERCPU_SIZE)) { qlist_move_all(q, &temp); - local_irq_restore(flags); - - if (unlikely(!qlist_empty(&temp))) { - spin_lock_irqsave(&quarantine_lock, flags); + spin_lock(&quarantine_lock); WRITE_ONCE(quarantine_size, quarantine_size + temp.bytes); qlist_move_all(&temp, &global_quarantine[quarantine_tail]); if (global_quarantine[quarantine_tail].bytes >= @@ -196,20 +203,33 @@ void quarantine_put(struct kasan_free_meta *info, struct kmem_cache *cache) if (new_tail != quarantine_head) quarantine_tail = new_tail; } - spin_unlock_irqrestore(&quarantine_lock, flags); + spin_unlock(&quarantine_lock); } + + local_irq_restore(flags); } void quarantine_reduce(void) { size_t total_size, new_quarantine_size, percpu_quarantines; unsigned long flags; + int srcu_idx; struct qlist_head to_free = QLIST_INIT; if (likely(READ_ONCE(quarantine_size) <= READ_ONCE(quarantine_max_size))) return; + /* + * srcu critical section ensures that quarantine_remove_cache() + * will not miss objects belonging to the cache while they are in our + * local to_free list. srcu is chosen because (1) it gives us private + * grace period domain that does not interfere with anything else, + * and (2) it allows synchronize_srcu() to return without waiting + * if there are no pending read critical sections (which is the + * expected case). + */ + srcu_idx = srcu_read_lock(&remove_cache_srcu); spin_lock_irqsave(&quarantine_lock, flags); /* @@ -237,6 +257,7 @@ void quarantine_reduce(void) spin_unlock_irqrestore(&quarantine_lock, flags); qlist_free_all(&to_free, NULL); + srcu_read_unlock(&remove_cache_srcu, srcu_idx); } static void qlist_move_cache(struct qlist_head *from, @@ -280,12 +301,28 @@ void quarantine_remove_cache(struct kmem_cache *cache) unsigned long flags, i; struct qlist_head to_free = QLIST_INIT; + /* + * Must be careful to not miss any objects that are being moved from + * per-cpu list to the global quarantine in quarantine_put(), + * nor objects being freed in quarantine_reduce(). on_each_cpu() + * achieves the first goal, while synchronize_srcu() achieves the + * second. + */ on_each_cpu(per_cpu_remove_cache, cache, 1); spin_lock_irqsave(&quarantine_lock, flags); - for (i = 0; i < QUARANTINE_BATCHES; i++) + for (i = 0; i < QUARANTINE_BATCHES; i++) { + if (qlist_empty(&global_quarantine[i])) + continue; qlist_move_cache(&global_quarantine[i], &to_free, cache); + /* Scanning whole quarantine can take a while. */ + spin_unlock_irqrestore(&quarantine_lock, flags); + cond_resched(); + spin_lock_irqsave(&quarantine_lock, flags); + } spin_unlock_irqrestore(&quarantine_lock, flags); qlist_free_all(&to_free, cache); + + synchronize_srcu(&remove_cache_srcu); } |