@@ -70,60 +70,13 @@ First Byte = 0x03 (X25_IFACE_PARAMS)
LAPB parameters. To be defined.
+Requirements for the device driver
+----------------------------------
-Possible Problems
-=================
-
-(Henner Eisen, 2000-10-28)
-
-The X.25 packet layer protocol depends on a reliable datalink service.
-The LAPB protocol provides such reliable service. But this reliability
-is not preserved by the Linux network device driver interface:
-
-- With Linux 2.4.x (and above) SMP kernels, packet ordering is not
- preserved. Even if a device driver calls netif_rx(skb1) and later
- netif_rx(skb2), skb2 might be delivered to the network layer
- earlier that skb1.
-- Data passed upstream by means of netif_rx() might be dropped by the
- kernel if the backlog queue is congested.
-
-The X.25 packet layer protocol will detect this and reset the virtual
-call in question. But many upper layer protocols are not designed to
-handle such N-Reset events gracefully. And frequent N-Reset events
-will always degrade performance.
-
-Thus, driver authors should make netif_rx() as reliable as possible:
-
-SMP re-ordering will not occur if the driver's interrupt handler is
-always executed on the same CPU. Thus,
-
-- Driver authors should use irq affinity for the interrupt handler.
-
-The probability of packet loss due to backlog congestion can be
-reduced by the following measures or a combination thereof:
-
-(1) Drivers for kernel versions 2.4.x and above should always check the
- return value of netif_rx(). If it returns NET_RX_DROP, the
- driver's LAPB protocol must not confirm reception of the frame
- to the peer.
- This will reliably suppress packet loss. The LAPB protocol will
- automatically cause the peer to re-transmit the dropped packet
- later.
- The lapb module interface was modified to support this. Its
- data_indication() method should now transparently pass the
- netif_rx() return value to the (lapb module) caller.
-(2) Drivers for kernel versions 2.2.x should always check the global
- variable netdev_dropping when a new frame is received. The driver
- should only call netif_rx() if netdev_dropping is zero. Otherwise
- the driver should not confirm delivery of the frame and drop it.
- Alternatively, the driver can queue the frame internally and call
- netif_rx() later when netif_dropping is 0 again. In that case, delivery
- confirmation should also be deferred such that the internal queue
- cannot grow to much.
- This will not reliably avoid packet loss, but the probability
- of packet loss in netif_rx() path will be significantly reduced.
-(3) Additionally, driver authors might consider to support
- CONFIG_NET_HW_FLOWCONTROL. This allows the driver to be woken up
- when a previously congested backlog queue becomes empty again.
- The driver could uses this for flow-controlling the peer by means
- of the LAPB protocol's flow-control service.
+Packets should not be reordered or dropped when delivering between the
+Packet Layer and the device driver.
+
+To avoid packets from being reordered or dropped when delivering from
+the device driver to the Packet Layer, the device driver should not
+call "netif_rx" to deliver the received packets. Instead, it should
+call "netif_receive_skb_core" from softirq context to deliver them.
@@ -25,6 +25,8 @@ struct x25_state {
x25_hdlc_proto settings;
bool up;
spinlock_t up_lock; /* Protects "up" */
+ struct sk_buff_head rx_queue;
+ struct tasklet_struct rx_tasklet;
};
static int x25_ioctl(struct net_device *dev, struct ifreq *ifr);
@@ -34,10 +36,22 @@ static struct x25_state *state(hdlc_device *hdlc)
return hdlc->state;
}
+static void x25_rx_queue_kick(struct tasklet_struct *t)
+{
+ struct x25_state *x25st = from_tasklet(x25st, t, rx_tasklet);
+ struct sk_buff *skb = skb_dequeue(&x25st->rx_queue);
+
+ while (skb) {
+ netif_receive_skb_core(skb);
+ skb = skb_dequeue(&x25st->rx_queue);
+ }
+}
+
/* These functions are callbacks called by LAPB layer */
static void x25_connect_disconnect(struct net_device *dev, int reason, int code)
{
+ struct x25_state *x25st = state(dev_to_hdlc(dev));
struct sk_buff *skb;
unsigned char *ptr;
@@ -50,7 +64,9 @@ static void x25_connect_disconnect(struct net_device *dev, int reason, int code)
*ptr = code;
skb->protocol = x25_type_trans(skb, dev);
- netif_rx(skb);
+
+ skb_queue_tail(&x25st->rx_queue, skb);
+ tasklet_schedule(&x25st->rx_tasklet);
}
@@ -71,6 +87,7 @@ static void x25_disconnected(struct net_device *dev, int reason)
static int x25_data_indication(struct net_device *dev, struct sk_buff *skb)
{
+ struct x25_state *x25st = state(dev_to_hdlc(dev));
unsigned char *ptr;
if (skb_cow(skb, 1)) {
@@ -84,7 +101,10 @@ static int x25_data_indication(struct net_device *dev, struct sk_buff *skb)
*ptr = X25_IFACE_DATA;
skb->protocol = x25_type_trans(skb, dev);
- return netif_rx(skb);
+
+ skb_queue_tail(&x25st->rx_queue, skb);
+ tasklet_schedule(&x25st->rx_tasklet);
+ return NET_RX_SUCCESS;
}
@@ -223,6 +243,7 @@ static void x25_close(struct net_device *dev)
spin_unlock_bh(&x25st->up_lock);
lapb_unregister(dev);
+ tasklet_kill(&x25st->rx_tasklet);
}
@@ -338,6 +359,8 @@ static int x25_ioctl(struct net_device *dev, struct ifreq *ifr)
memcpy(&state(hdlc)->settings, &new_settings, size);
state(hdlc)->up = false;
spin_lock_init(&state(hdlc)->up_lock);
+ skb_queue_head_init(&state(hdlc)->rx_queue);
+ tasklet_setup(&state(hdlc)->rx_tasklet, x25_rx_queue_kick);
/* There's no header_ops so hard_header_len should be 0. */
dev->hard_header_len = 0;
@@ -53,6 +53,8 @@ struct lapbethdev {
struct net_device *axdev; /* lapbeth device (lapb#) */
bool up;
spinlock_t up_lock; /* Protects "up" */
+ struct sk_buff_head rx_queue;
+ struct napi_struct napi;
};
static LIST_HEAD(lapbeth_devices);
@@ -83,6 +85,25 @@ static __inline__ int dev_is_ethdev(struct net_device *dev)
/* ------------------------------------------------------------------------ */
+static int lapbeth_napi_poll(struct napi_struct *napi, int budget)
+{
+ struct lapbethdev *lapbeth = container_of(napi, struct lapbethdev, napi);
+ struct sk_buff *skb;
+ int processed = 0;
+
+ for (; processed < budget; ++processed) {
+ skb = skb_dequeue(&lapbeth->rx_queue);
+ if (!skb)
+ break;
+ netif_receive_skb_core(skb);
+ }
+
+ if (processed < budget)
+ napi_complete(napi);
+
+ return processed;
+}
+
/*
* Receive a LAPB frame via an ethernet interface.
*/
@@ -135,6 +156,7 @@ static int lapbeth_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
static int lapbeth_data_indication(struct net_device *dev, struct sk_buff *skb)
{
+ struct lapbethdev *lapbeth = netdev_priv(dev);
unsigned char *ptr;
if (skb_cow(skb, 1)) {
@@ -148,7 +170,10 @@ static int lapbeth_data_indication(struct net_device *dev, struct sk_buff *skb)
*ptr = X25_IFACE_DATA;
skb->protocol = x25_type_trans(skb, dev);
- return netif_rx(skb);
+
+ skb_queue_tail(&lapbeth->rx_queue, skb);
+ napi_schedule(&lapbeth->napi);
+ return NET_RX_SUCCESS;
}
/*
@@ -233,6 +258,7 @@ static void lapbeth_data_transmit(struct net_device *ndev, struct sk_buff *skb)
static void lapbeth_connected(struct net_device *dev, int reason)
{
+ struct lapbethdev *lapbeth = netdev_priv(dev);
unsigned char *ptr;
struct sk_buff *skb = dev_alloc_skb(1);
@@ -245,11 +271,14 @@ static void lapbeth_connected(struct net_device *dev, int reason)
*ptr = X25_IFACE_CONNECT;
skb->protocol = x25_type_trans(skb, dev);
- netif_rx(skb);
+
+ skb_queue_tail(&lapbeth->rx_queue, skb);
+ napi_schedule(&lapbeth->napi);
}
static void lapbeth_disconnected(struct net_device *dev, int reason)
{
+ struct lapbethdev *lapbeth = netdev_priv(dev);
unsigned char *ptr;
struct sk_buff *skb = dev_alloc_skb(1);
@@ -262,7 +291,9 @@ static void lapbeth_disconnected(struct net_device *dev, int reason)
*ptr = X25_IFACE_DISCONNECT;
skb->protocol = x25_type_trans(skb, dev);
- netif_rx(skb);
+
+ skb_queue_tail(&lapbeth->rx_queue, skb);
+ napi_schedule(&lapbeth->napi);
}
/*
@@ -293,6 +324,8 @@ static int lapbeth_open(struct net_device *dev)
struct lapbethdev *lapbeth = netdev_priv(dev);
int err;
+ napi_enable(&lapbeth->napi);
+
if ((err = lapb_register(dev, &lapbeth_callbacks)) != LAPB_OK) {
pr_err("lapb_register error: %d\n", err);
return -ENODEV;
@@ -317,6 +350,8 @@ static int lapbeth_close(struct net_device *dev)
if ((err = lapb_unregister(dev)) != LAPB_OK)
pr_err("lapb_unregister error: %d\n", err);
+ napi_disable(&lapbeth->napi);
+
return 0;
}
@@ -374,6 +409,9 @@ static int lapbeth_new_device(struct net_device *dev)
lapbeth->up = false;
spin_lock_init(&lapbeth->up_lock);
+ skb_queue_head_init(&lapbeth->rx_queue);
+ netif_napi_add(ndev, &lapbeth->napi, lapbeth_napi_poll, 16);
+
rc = -EIO;
if (register_netdevice(ndev))
goto fail;