[PATCH 4.9 17/78] Revert "net: use lib/percpu_counter API for fragmentation mem accounting"

[Greg Kroah-Hartman <gregkh@linuxfoundation.org>]

4.9-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Jesper Dangaard Brouer <brouer@redhat.com>


[ Upstream commit fb452a1aa3fd4034d7999e309c5466ff2d7005aa ]

This reverts commit 6d7b857d541ecd1d9bd997c97242d4ef94b19de2.

There is a bug in fragmentation codes use of the percpu_counter API,
that can cause issues on systems with many CPUs.

The frag_mem_limit() just reads the global counter (fbc->count),
without considering other CPUs can have upto batch size (130K) that
haven't been subtracted yet.  Due to the 3MBytes lower thresh limit,
this become dangerous at >=24 CPUs (3*1024*1024/130000=24).

The correct API usage would be to use __percpu_counter_compare() which
does the right thing, and takes into account the number of (online)
CPUs and batch size, to account for this and call __percpu_counter_sum()
when needed.

We choose to revert the use of the lib/percpu_counter API for frag
memory accounting for several reasons:

1) On systems with CPUs > 24, the heavier fully locked
   __percpu_counter_sum() is always invoked, which will be more
   expensive than the atomic_t that is reverted to.

Given systems with more than 24 CPUs are becoming common this doesn't
seem like a good option.  To mitigate this, the batch size could be
decreased and thresh be increased.

2) The add_frag_mem_limit+sub_frag_mem_limit pairs happen on the RX
   CPU, before SKBs are pushed into sockets on remote CPUs.  Given
   NICs can only hash on L2 part of the IP-header, the NIC-RXq's will
   likely be limited.  Thus, a fair chance that atomic add+dec happen
   on the same CPU.

Revert note that commit 1d6119baf061 ("net: fix percpu memory leaks")
removed init_frag_mem_limit() and instead use inet_frags_init_net().
After this revert, inet_frags_uninit_net() becomes empty.

Fixes: 6d7b857d541e ("net: use lib/percpu_counter API for fragmentation mem accounting")
Fixes: 1d6119baf061 ("net: fix percpu memory leaks")
Signed-off-by: Jesper Dangaard Brouer <brouer@redhat.com>
Acked-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 include/net/inet_frag.h  |   36 +++++++++---------------------------
 net/ipv4/inet_fragment.c |    4 +---
 2 files changed, 10 insertions(+), 30 deletions(-)

--- a/include/net/inet_frag.h
+++ b/include/net/inet_frag.h
@@ -1,14 +1,9 @@
 #ifndef __NET_FRAG_H__
 #define __NET_FRAG_H__
 
-#include <linux/percpu_counter.h>
-
 struct netns_frags {
-	/* The percpu_counter "mem" need to be cacheline aligned.
-	 *  mem.count must not share cacheline with other writers
-	 */
-	struct percpu_counter   mem ____cacheline_aligned_in_smp;
-
+	/* Keep atomic mem on separate cachelines in structs that include it */
+	atomic_t		mem ____cacheline_aligned_in_smp;
 	/* sysctls */
 	int			timeout;
 	int			high_thresh;
@@ -110,11 +105,11 @@ void inet_frags_fini(struct inet_frags *
 
 static inline int inet_frags_init_net(struct netns_frags *nf)
 {
-	return percpu_counter_init(&nf->mem, 0, GFP_KERNEL);
+	atomic_set(&nf->mem, 0);
+	return 0;
 }
 static inline void inet_frags_uninit_net(struct netns_frags *nf)
 {
-	percpu_counter_destroy(&nf->mem);
 }
 
 void inet_frags_exit_net(struct netns_frags *nf, struct inet_frags *f);
@@ -140,37 +135,24 @@ static inline bool inet_frag_evicting(st
 
 /* Memory Tracking Functions. */
 
-/* The default percpu_counter batch size is not big enough to scale to
- * fragmentation mem acct sizes.
- * The mem size of a 64K fragment is approx:
- *  (44 fragments * 2944 truesize) + frag_queue struct(200) = 129736 bytes
- */
-static unsigned int frag_percpu_counter_batch = 130000;
-
 static inline int frag_mem_limit(struct netns_frags *nf)
 {
-	return percpu_counter_read(&nf->mem);
+	return atomic_read(&nf->mem);
 }
 
 static inline void sub_frag_mem_limit(struct netns_frags *nf, int i)
 {
-	__percpu_counter_add(&nf->mem, -i, frag_percpu_counter_batch);
+	atomic_sub(i, &nf->mem);
 }
 
 static inline void add_frag_mem_limit(struct netns_frags *nf, int i)
 {
-	__percpu_counter_add(&nf->mem, i, frag_percpu_counter_batch);
+	atomic_add(i, &nf->mem);
 }
 
-static inline unsigned int sum_frag_mem_limit(struct netns_frags *nf)
+static inline int sum_frag_mem_limit(struct netns_frags *nf)
 {
-	unsigned int res;
-
-	local_bh_disable();
-	res = percpu_counter_sum_positive(&nf->mem);
-	local_bh_enable();
-
-	return res;
+	return atomic_read(&nf->mem);
 }
 
 /* RFC 3168 support :
--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -234,10 +234,8 @@ evict_again:
 	cond_resched();
 
 	if (read_seqretry(&f->rnd_seqlock, seq) ||
-	    percpu_counter_sum(&nf->mem))
+	    sum_frag_mem_limit(nf))
 		goto evict_again;
-
-	percpu_counter_destroy(&nf->mem);
 }
 EXPORT_SYMBOL(inet_frags_exit_net);