[PATCH net-next v6 0/3] udp: refactor memory accounting

STALE3591d

Revision v6 of 2 in this series.

4 messages, 2 authors, 2016-10-21 · open the first message on its own page

[PATCH net-next v6 0/3] udp: refactor memory accounting

From: Paolo Abeni <pabeni@redhat.com>
Date: 2016-10-21 11:55:57

This patch series refactor the udp memory accounting, replacing the
generic implementation with a custom one, in order to remove the needs for
locking the socket on the enqueue and dequeue operations. The socket backlog
usage is dropped, as well.

The first patch factor out pieces of some queue and memory management
socket helpers, so that they can later be used by the udp memory accounting
functions.
The second patch adds the memory account helpers, without using them.
The third patch replacse the old rx memory accounting path for udp over ipv4 and
udp over ipv6. In kernel UDP users are updated, as well.

The memory accounting schema is described in detail in the individual patch
commit message.

The performance gain depends on the specific scenario; with few flows (and
little contention in the original code) the differences are in the noise range,
while with several flows contending the same socket, the measured speed-up
is relevant (e.g. even over 100% in case of extreme contention)

Many thanks to Eric Dumazet for the reiterated reviews and suggestions.

v5 -> v6:
 - do not orphan the skb on enqueue, skb_steal_sock() already did
   the work for us

v4 -> v5:
 - use the receive queue spin lock to protect the memory accounting
 - several minor clean-up

v3 -> v4:
 - simplified the locking schema, always use a plain spinlock

v2 -> v3:
 - do not set the now unsed backlog_rcv callback

v1 -> v2:
 - changed slighly the memory accounting schema, we now perform lazy reclaim
 - fixed forward_alloc updating issue
 - fixed memory counter integer overflows

Paolo Abeni (3):
  net/socket: factor out helpers for memory and queue manipulation
  udp: implement memory accounting helpers
  udp: use it's own memory accounting schema

 include/net/sock.h    |   4 ++
 include/net/udp.h     |   4 ++
 net/core/datagram.c   |  36 +++++++-----
 net/core/sock.c       |  64 +++++++++++++++-------
 net/ipv4/udp.c        | 149 ++++++++++++++++++++++++++++++++++++++------------
 net/ipv6/udp.c        |  34 +++---------
 net/sunrpc/svcsock.c  |  20 +++++--
 net/sunrpc/xprtsock.c |   2 +-
 8 files changed, 214 insertions(+), 99 deletions(-)

-- 
1.8.3.1

[PATCH net-next v6 1/3] net/socket: factor out helpers for memory and queue manipulation

From: Paolo Abeni <pabeni@redhat.com>
Date: 2016-10-21 11:55:59

Basic sock operations that udp code can use with its own
memory accounting schema. No functional change is introduced
in the existing APIs.

v4 -> v5:
  - avoid whitespace changes

v2 -> v4:
  - avoid exporting __sock_enqueue_skb

v1 -> v2:
  - avoid export sock_rmem_free

Acked-by: Hannes Frederic Sowa <redacted>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/sock.h  |  4 ++++
 net/core/datagram.c | 36 ++++++++++++++++++------------
 net/core/sock.c     | 64 +++++++++++++++++++++++++++++++++++++----------------
 3 files changed, 71 insertions(+), 33 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index ebf75db..2764895 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h

@@ -1274,7 +1274,9 @@ static inline struct inode *SOCK_INODE(struct socket *socket)
 /*
  * Functions for memory accounting
  */
+int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind);
 int __sk_mem_schedule(struct sock *sk, int size, int kind);
+void __sk_mem_reduce_allocated(struct sock *sk, int amount);
 void __sk_mem_reclaim(struct sock *sk, int amount);
 
 #define SK_MEM_QUANTUM ((int)PAGE_SIZE)

@@ -1950,6 +1952,8 @@ void sk_reset_timer(struct sock *sk, struct timer_list *timer,
 
 void sk_stop_timer(struct sock *sk, struct timer_list *timer);
 
+int __sk_queue_drop_skb(struct sock *sk, struct sk_buff *skb,
+			unsigned int flags);
 int __sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);
 int sock_queue_rcv_skb(struct sock *sk, struct sk_buff *skb);

diff --git a/net/core/datagram.c b/net/core/datagram.c
index b7de71f..bfb973a 100644
--- a/net/core/datagram.c
+++ b/net/core/datagram.c

@@ -323,6 +323,27 @@ void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len)
 }
 EXPORT_SYMBOL(__skb_free_datagram_locked);
 
+int __sk_queue_drop_skb(struct sock *sk, struct sk_buff *skb,
+			unsigned int flags)
+{
+	int err = 0;
+
+	if (flags & MSG_PEEK) {
+		err = -ENOENT;
+		spin_lock_bh(&sk->sk_receive_queue.lock);
+		if (skb == skb_peek(&sk->sk_receive_queue)) {
+			__skb_unlink(skb, &sk->sk_receive_queue);
+			atomic_dec(&skb->users);
+			err = 0;
+		}
+		spin_unlock_bh(&sk->sk_receive_queue.lock);
+	}
+
+	atomic_inc(&sk->sk_drops);
+	return err;
+}
+EXPORT_SYMBOL(__sk_queue_drop_skb);
+
 /**
  *	skb_kill_datagram - Free a datagram skbuff forcibly
  *	@sk: socket

@@ -346,23 +367,10 @@ void __skb_free_datagram_locked(struct sock *sk, struct sk_buff *skb, int len)
 
 int skb_kill_datagram(struct sock *sk, struct sk_buff *skb, unsigned int flags)
 {
-	int err = 0;
-
-	if (flags & MSG_PEEK) {
-		err = -ENOENT;
-		spin_lock_bh(&sk->sk_receive_queue.lock);
-		if (skb == skb_peek(&sk->sk_receive_queue)) {
-			__skb_unlink(skb, &sk->sk_receive_queue);
-			atomic_dec(&skb->users);
-			err = 0;
-		}
-		spin_unlock_bh(&sk->sk_receive_queue.lock);
-	}
+	int err = __sk_queue_drop_skb(sk, skb, flags);
 
 	kfree_skb(skb);
-	atomic_inc(&sk->sk_drops);
 	sk_mem_reclaim_partial(sk);
-
 	return err;
 }
 EXPORT_SYMBOL(skb_kill_datagram);

diff --git a/net/core/sock.c b/net/core/sock.c
index c73e28f..d8e4532e 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c

@@ -2091,24 +2091,18 @@ int sk_wait_data(struct sock *sk, long *timeo, const struct sk_buff *skb)
 EXPORT_SYMBOL(sk_wait_data);
 
 /**
- *	__sk_mem_schedule - increase sk_forward_alloc and memory_allocated
+ *	__sk_mem_raise_allocated - increase memory_allocated
  *	@sk: socket
  *	@size: memory size to allocate
+ *	@amt: pages to allocate
  *	@kind: allocation type
  *
- *	If kind is SK_MEM_SEND, it means wmem allocation. Otherwise it means
- *	rmem allocation. This function assumes that protocols which have
- *	memory_pressure use sk_wmem_queued as write buffer accounting.
+ *	Similar to __sk_mem_schedule(), but does not update sk_forward_alloc
  */
-int __sk_mem_schedule(struct sock *sk, int size, int kind)
+int __sk_mem_raise_allocated(struct sock *sk, int size, int amt, int kind)
 {
 	struct proto *prot = sk->sk_prot;
-	int amt = sk_mem_pages(size);
-	long allocated;
-
-	sk->sk_forward_alloc += amt * SK_MEM_QUANTUM;
-
-	allocated = sk_memory_allocated_add(sk, amt);
+	long allocated = sk_memory_allocated_add(sk, amt);
 
 	if (mem_cgroup_sockets_enabled && sk->sk_memcg &&
 	    !mem_cgroup_charge_skmem(sk->sk_memcg, amt))

@@ -2169,9 +2163,6 @@ int __sk_mem_schedule(struct sock *sk, int size, int kind)
 
 	trace_sock_exceed_buf_limit(sk, prot, allocated);
 
-	/* Alas. Undo changes. */
-	sk->sk_forward_alloc -= amt * SK_MEM_QUANTUM;
-
 	sk_memory_allocated_sub(sk, amt);
 
 	if (mem_cgroup_sockets_enabled && sk->sk_memcg)

@@ -2179,18 +2170,40 @@ int __sk_mem_schedule(struct sock *sk, int size, int kind)
 
 	return 0;
 }
+EXPORT_SYMBOL(__sk_mem_raise_allocated);
+
+/**
+ *	__sk_mem_schedule - increase sk_forward_alloc and memory_allocated
+ *	@sk: socket
+ *	@size: memory size to allocate
+ *	@kind: allocation type
+ *
+ *	If kind is SK_MEM_SEND, it means wmem allocation. Otherwise it means
+ *	rmem allocation. This function assumes that protocols which have
+ *	memory_pressure use sk_wmem_queued as write buffer accounting.
+ */
+int __sk_mem_schedule(struct sock *sk, int size, int kind)
+{
+	int ret, amt = sk_mem_pages(size);
+
+	sk->sk_forward_alloc += amt << SK_MEM_QUANTUM_SHIFT;
+	ret = __sk_mem_raise_allocated(sk, size, amt, kind);
+	if (!ret)
+		sk->sk_forward_alloc -= amt << SK_MEM_QUANTUM_SHIFT;
+	return ret;
+}
 EXPORT_SYMBOL(__sk_mem_schedule);
 
 /**
- *	__sk_mem_reclaim - reclaim memory_allocated
+ *	__sk_mem_reduce_allocated - reclaim memory_allocated
  *	@sk: socket
- *	@amount: number of bytes (rounded down to a SK_MEM_QUANTUM multiple)
+ *	@amount: number of quanta
+ *
+ *	Similar to __sk_mem_reclaim(), but does not update sk_forward_alloc
  */
-void __sk_mem_reclaim(struct sock *sk, int amount)
+void __sk_mem_reduce_allocated(struct sock *sk, int amount)
 {
-	amount >>= SK_MEM_QUANTUM_SHIFT;
 	sk_memory_allocated_sub(sk, amount);
-	sk->sk_forward_alloc -= amount << SK_MEM_QUANTUM_SHIFT;
 
 	if (mem_cgroup_sockets_enabled && sk->sk_memcg)
 		mem_cgroup_uncharge_skmem(sk->sk_memcg, amount);

@@ -2199,6 +2212,19 @@ void __sk_mem_reclaim(struct sock *sk, int amount)
 	    (sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)))
 		sk_leave_memory_pressure(sk);
 }
+EXPORT_SYMBOL(__sk_mem_reduce_allocated);
+
+/**
+ *	__sk_mem_reclaim - reclaim sk_forward_alloc and memory_allocated
+ *	@sk: socket
+ *	@amount: number of bytes (rounded down to a SK_MEM_QUANTUM multiple)
+ */
+void __sk_mem_reclaim(struct sock *sk, int amount)
+{
+	amount >>= SK_MEM_QUANTUM_SHIFT;
+	sk->sk_forward_alloc -= amount << SK_MEM_QUANTUM_SHIFT;
+	__sk_mem_reduce_allocated(sk, amount);
+}
 EXPORT_SYMBOL(__sk_mem_reclaim);
 
 int sk_set_peek_off(struct sock *sk, int val)

-- 
1.8.3.1

[PATCH net-next v6 2/3] udp: implement memory accounting helpers

From: Paolo Abeni <pabeni@redhat.com>
Date: 2016-10-21 11:56:01

Avoid using the generic helpers.
Use the receive queue spin lock to protect the memory
accounting operation, both on enqueue and on dequeue.

On dequeue perform partial memory reclaiming, trying to
leave a quantum of forward allocated memory.

On enqueue use a custom helper, to allow some optimizations:
- use a plain spin_lock() variant instead of the slightly
  costly spin_lock_irqsave(),
- avoid dst_force check, since the calling code has already
  dropped the skb dst
- avoid orphaning the skb, since skb_steal_sock() already did
  the work for us

The above needs custom memory reclaiming on shutdown, provided
by the udp_destruct_sock().

v5 -> v6:
  - don't orphan the skb on enqueue

v4 -> v5:
  - replace the mem_lock with the receive queue spin lock
  - ensure that the bh is always allowed to enqueue at least
    a skb, even if sk_rcvbuf is exceeded

v3 -> v4:
  - reworked memory accunting, simplifying the schema
  - provide an helper for both memory scheduling and enqueuing

v1 -> v2:
  - use a udp specific destrctor to perform memory reclaiming
  - remove a couple of helpers, unneeded after the above cleanup
  - do not reclaim memory on dequeue if not under memory
    pressure
  - reworked the fwd accounting schema to avoid potential
    integer overflow

Acked-by: Hannes Frederic Sowa <redacted>
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 include/net/udp.h |   4 +++
 net/ipv4/udp.c    | 106 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 110 insertions(+)

diff --git a/include/net/udp.h b/include/net/udp.h
index ea53a87..18f1e6b 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h

@@ -246,6 +246,9 @@ static inline __be16 udp_flow_src_port(struct net *net, struct sk_buff *skb,
 }
 
 /* net/ipv4/udp.c */
+void skb_consume_udp(struct sock *sk, struct sk_buff *skb, int len);
+int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb);
+
 void udp_v4_early_demux(struct sk_buff *skb);
 int udp_get_port(struct sock *sk, unsigned short snum,
 		 int (*saddr_cmp)(const struct sock *,

@@ -258,6 +261,7 @@ int udp_get_port(struct sock *sk, unsigned short snum,
 void udp4_hwcsum(struct sk_buff *skb, __be32 src, __be32 dst);
 int udp_rcv(struct sk_buff *skb);
 int udp_ioctl(struct sock *sk, int cmd, unsigned long arg);
+int udp_init_sock(struct sock *sk);
 int udp_disconnect(struct sock *sk, int flags);
 unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait);
 struct sk_buff *skb_udp_tunnel_segment(struct sk_buff *skb,

diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 7d96dc2..e137e1d 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c

@@ -1172,6 +1172,112 @@ int udp_sendpage(struct sock *sk, struct page *page, int offset,
 	return ret;
 }
 
+static void udp_rmem_release(struct sock *sk, int size, int partial)
+{
+	int amt;
+
+	atomic_sub(size, &sk->sk_rmem_alloc);
+
+	spin_lock_bh(&sk->sk_receive_queue.lock);
+	sk->sk_forward_alloc += size;
+	amt = (sk->sk_forward_alloc - partial) & ~(SK_MEM_QUANTUM - 1);
+	sk->sk_forward_alloc -= amt;
+	spin_unlock_bh(&sk->sk_receive_queue.lock);
+
+	if (amt)
+		__sk_mem_reduce_allocated(sk, amt >> SK_MEM_QUANTUM_SHIFT);
+}
+
+static void udp_rmem_free(struct sk_buff *skb)
+{
+	udp_rmem_release(skb->sk, skb->truesize, 1);
+}
+
+int __udp_enqueue_schedule_skb(struct sock *sk, struct sk_buff *skb)
+{
+	struct sk_buff_head *list = &sk->sk_receive_queue;
+	int rmem, delta, amt, err = -ENOMEM;
+	int size = skb->truesize;
+
+	/* try to avoid the costly atomic add/sub pair when the receive
+	 * queue is full; always allow at least a packet
+	 */
+	rmem = atomic_read(&sk->sk_rmem_alloc);
+	if (rmem && (rmem + size > sk->sk_rcvbuf))
+		goto drop;
+
+	/* we drop only if the receive buf is full and the receive
+	 * queue contains some other skb
+	 */
+	rmem = atomic_add_return(size, &sk->sk_rmem_alloc);
+	if ((rmem > sk->sk_rcvbuf) && (rmem > size))
+		goto uncharge_drop;
+
+	spin_lock(&list->lock);
+	if (size >= sk->sk_forward_alloc) {
+		amt = sk_mem_pages(size);
+		delta = amt << SK_MEM_QUANTUM_SHIFT;
+		if (!__sk_mem_raise_allocated(sk, delta, amt, SK_MEM_RECV)) {
+			err = -ENOBUFS;
+			spin_unlock(&list->lock);
+			goto uncharge_drop;
+		}
+
+		sk->sk_forward_alloc += delta;
+	}
+
+	sk->sk_forward_alloc -= size;
+
+	/* the skb owner in now the udp socket */
+	skb->sk = sk;
+	skb->destructor = udp_rmem_free;
+	skb->dev = NULL;
+	sock_skb_set_dropcount(sk, skb);
+
+	__skb_queue_tail(list, skb);
+	spin_unlock(&list->lock);
+
+	if (!sock_flag(sk, SOCK_DEAD))
+		sk->sk_data_ready(sk);
+
+	return 0;
+
+uncharge_drop:
+	atomic_sub(skb->truesize, &sk->sk_rmem_alloc);
+
+drop:
+	atomic_inc(&sk->sk_drops);
+	return err;
+}
+EXPORT_SYMBOL_GPL(__udp_enqueue_schedule_skb);
+
+static void udp_destruct_sock(struct sock *sk)
+{
+	/* reclaim completely the forward allocated memory */
+	__skb_queue_purge(&sk->sk_receive_queue);
+	udp_rmem_release(sk, 0, 0);
+	inet_sock_destruct(sk);
+}
+
+int udp_init_sock(struct sock *sk)
+{
+	sk->sk_destruct = udp_destruct_sock;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(udp_init_sock);
+
+void skb_consume_udp(struct sock *sk, struct sk_buff *skb, int len)
+{
+	if (unlikely(READ_ONCE(sk->sk_peek_off) >= 0)) {
+		bool slow = lock_sock_fast(sk);
+
+		sk_peek_offset_bwd(sk, len);
+		unlock_sock_fast(sk, slow);
+	}
+	consume_skb(skb);
+}
+EXPORT_SYMBOL_GPL(skb_consume_udp);
+
 /**
  *	first_packet_length	- return length of first packet in receive queue
  *	@sk: socket

-- 
1.8.3.1

Re: [PATCH net-next v6 2/3] udp: implement memory accounting helpers

From: Eric Dumazet <hidden>
Date: 2016-10-21 21:13:45

On Fri, 2016-10-21 at 13:55 +0200, Paolo Abeni wrote:

Avoid using the generic helpers.
Use the receive queue spin lock to protect the memory
accounting operation, both on enqueue and on dequeue.

On dequeue perform partial memory reclaiming, trying to
leave a quantum of forward allocated memory.

On enqueue use a custom helper, to allow some optimizations:
- use a plain spin_lock() variant instead of the slightly
  costly spin_lock_irqsave(),
- avoid dst_force check, since the calling code has already
  dropped the skb dst
- avoid orphaning the skb, since skb_steal_sock() already did
  the work for us

The above needs custom memory reclaiming on shutdown, provided
by the udp_destruct_sock().

Acked-by: Eric Dumazet <edumazet@google.com>

`h`	back out one level
`j`	next message in thread
`k`	previous message in thread
`l`	drill in
`Esc`	close help / fold thread tree
`?`	toggle this help