1 files changed, 51 insertions, 45 deletions

diff --git a/net/core/flow.c b/net/core/flow.c
index f67dcbfe54ef..990703b8863b 100644
--- a/net/core/flow.c
+++ b/net/core/flow.c
@@ -53,8 +53,7 @@ struct flow_flush_info {
 
 struct flow_cache {
         u32                             hash_shift;
-        unsigned long                   order;
-        struct flow_cache_percpu        *percpu;
+        struct flow_cache_percpu __percpu *percpu;
         struct notifier_block           hotcpu_notifier;
         int                             low_watermark;
         int                             high_watermark;
@@ -64,7 +63,7 @@ struct flow_cache {
 atomic_t flow_cache_genid = ATOMIC_INIT(0);
 EXPORT_SYMBOL(flow_cache_genid);
 static struct flow_cache flow_cache_global;
-static struct kmem_cache *flow_cachep;
+static struct kmem_cache *flow_cachep __read_mostly;
 
 static DEFINE_SPINLOCK(flow_cache_gc_lock);
 static LIST_HEAD(flow_cache_gc_list);
@@ -173,35 +172,31 @@ static void flow_new_hash_rnd(struct flow_cache *fc,
 
 static u32 flow_hash_code(struct flow_cache *fc,
                           struct flow_cache_percpu *fcp,
-                          struct flowi *key)
+                          const struct flowi *key)
 {
-        u32 *k = (u32 *) key;
+        const u32 *k = (const u32 *) key;
 
-        return (jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd)
-                & (flow_cache_hash_size(fc) - 1));
+        return jhash2(k, (sizeof(*key) / sizeof(u32)), fcp->hash_rnd)
+                & (flow_cache_hash_size(fc) - 1);
 }
 
-#if (BITS_PER_LONG == 64)
-typedef u64 flow_compare_t;
-#else
-typedef u32 flow_compare_t;
-#endif
+typedef unsigned long flow_compare_t;
 
 /* I hear what you're saying, use memcmp.  But memcmp cannot make
  * important assumptions that we can here, such as alignment and
  * constant size.
  */
-static int flow_key_compare(struct flowi *key1, struct flowi *key2)
+static int flow_key_compare(const struct flowi *key1, const struct flowi *key2)
 {
-        flow_compare_t *k1, *k1_lim, *k2;
+        const flow_compare_t *k1, *k1_lim, *k2;
         const int n_elem = sizeof(struct flowi) / sizeof(flow_compare_t);
 
         BUILD_BUG_ON(sizeof(struct flowi) % sizeof(flow_compare_t));
 
-        k1 = (flow_compare_t *) key1;
+        k1 = (const flow_compare_t *) key1;
         k1_lim = k1 + n_elem;
 
-        k2 = (flow_compare_t *) key2;
+        k2 = (const flow_compare_t *) key2;
 
         do {
                 if (*k1++ != *k2++)
@@ -212,7 +207,7 @@ static int flow_key_compare(struct flowi *key1, struct flowi *key2)
 }
 
 struct flow_cache_object *
-flow_cache_lookup(struct net *net, struct flowi *key, u16 family, u8 dir,
+flow_cache_lookup(struct net *net, const struct flowi *key, u16 family, u8 dir,
                   flow_resolve_t resolver, void *ctx)
 {
         struct flow_cache *fc = &flow_cache_global;
@@ -357,62 +352,73 @@ void flow_cache_flush(void)
         put_online_cpus();
 }
 
-static void __init flow_cache_cpu_prepare(struct flow_cache *fc,
-                                          struct flow_cache_percpu *fcp)
+static int __cpuinit flow_cache_cpu_prepare(struct flow_cache *fc, int cpu)
 {
-        fcp->hash_table = (struct hlist_head *)
-                __get_free_pages(GFP_KERNEL|__GFP_ZERO, fc->order);
-        if (!fcp->hash_table)
-                panic("NET: failed to allocate flow cache order %lu\n", fc->order);
+        struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu);
+        size_t sz = sizeof(struct hlist_head) * flow_cache_hash_size(fc);
 
-        fcp->hash_rnd_recalc = 1;
-        fcp->hash_count = 0;
-        tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0);
+        if (!fcp->hash_table) {
+                fcp->hash_table = kzalloc_node(sz, GFP_KERNEL, cpu_to_node(cpu));
+                if (!fcp->hash_table) {
+                        pr_err("NET: failed to allocate flow cache sz %zu\n", sz);
+                        return -ENOMEM;
+                }
+                fcp->hash_rnd_recalc = 1;
+                fcp->hash_count = 0;
+                tasklet_init(&fcp->flush_tasklet, flow_cache_flush_tasklet, 0);
+        }
+        return 0;
 }
 
-static int flow_cache_cpu(struct notifier_block *nfb,
+static int __cpuinit flow_cache_cpu(struct notifier_block *nfb,
                           unsigned long action,
                           void *hcpu)
 {
         struct flow_cache *fc = container_of(nfb, struct flow_cache, hotcpu_notifier);
-        int cpu = (unsigned long) hcpu;
+        int res, cpu = (unsigned long) hcpu;
         struct flow_cache_percpu *fcp = per_cpu_ptr(fc->percpu, cpu);
 
-        if (action == CPU_DEAD || action == CPU_DEAD_FROZEN)
+        switch (action) {
+        case CPU_UP_PREPARE:
+        case CPU_UP_PREPARE_FROZEN:
+                res = flow_cache_cpu_prepare(fc, cpu);
+                if (res)
+                        return notifier_from_errno(res);
+                break;
+        case CPU_DEAD:
+        case CPU_DEAD_FROZEN:
                 __flow_cache_shrink(fc, fcp, 0);
+                break;
+        }
         return NOTIFY_OK;
 }
 
-static int flow_cache_init(struct flow_cache *fc)
+static int __init flow_cache_init(struct flow_cache *fc)
 {
-        unsigned long order;
         int i;
 
         fc->hash_shift = 10;
         fc->low_watermark = 2 * flow_cache_hash_size(fc);
         fc->high_watermark = 4 * flow_cache_hash_size(fc);
 
-        for (order = 0;
-             (PAGE_SIZE << order) <
-                     (sizeof(struct hlist_head)*flow_cache_hash_size(fc));
-             order++)
-                /* NOTHING */;
-        fc->order = order;
         fc->percpu = alloc_percpu(struct flow_cache_percpu);
+        if (!fc->percpu)
+                return -ENOMEM;
 
-        setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd,
-                    (unsigned long) fc);
-        fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD;
-        add_timer(&fc->rnd_timer);
-
-        for_each_possible_cpu(i)
-                flow_cache_cpu_prepare(fc, per_cpu_ptr(fc->percpu, i));
-
+        for_each_online_cpu(i) {
+                if (flow_cache_cpu_prepare(fc, i))
+                        return -ENOMEM;
+        }
         fc->hotcpu_notifier = (struct notifier_block){
                 .notifier_call = flow_cache_cpu,
         };
         register_hotcpu_notifier(&fc->hotcpu_notifier);
 
+        setup_timer(&fc->rnd_timer, flow_cache_new_hashrnd,
+                    (unsigned long) fc);
+        fc->rnd_timer.expires = jiffies + FLOW_HASH_RND_PERIOD;
+        add_timer(&fc->rnd_timer);
+
         return 0;
 }