Commit b4727e69 authored by Sebastian Siewior's avatar Sebastian Siewior Committed by David S. Miller
Browse files

net/cpsw: redo rx skb allocation in rx path

In case that we run into OOM during the allocation of the new rx-skb we
don't get one and we have one skb less than we used to have. If this
continues to happen then we end up with no rx-skbs at all.
This patch changes the following:
- if we fail to allocate the new skb, then we treat the currently
  completed skb as the new one and so drop the currently received data.
- instead of testing multiple times if the device is gone we rely one
  the status field which is set to -ENOSYS in case the channel is going
  down and incomplete requests are purged.
  cpdma_chan_stop() removes most of the packages with -ENOSYS. The
  currently active packet which is removed has the "tear down" bit set.
  So if that bit is set, we send ENOSYS as well otherwise we pass the
  status bits which are required to figure out which of the two possible
  just finished.
Acked-by: default avatarMugunthan V N <>
Signed-off-by: default avatarSebastian Andrzej Siewior <>
Signed-off-by: default avatarDavid S. Miller <>
parent aef614e1
......@@ -468,43 +468,36 @@ void cpsw_tx_handler(void *token, int len, int status)
void cpsw_rx_handler(void *token, int len, int status)
struct sk_buff *skb = token;
struct sk_buff *new_skb;
struct net_device *ndev = skb->dev;
struct cpsw_priv *priv = netdev_priv(ndev);
int ret = 0;
cpsw_dual_emac_src_port_detect(status, priv, ndev, skb);
/* free and bail if we are shutting down */
if (unlikely(!netif_running(ndev)) ||
unlikely(!netif_carrier_ok(ndev))) {
if (unlikely(status < 0)) {
/* the interface is going down, skbs are purged */
if (likely(status >= 0)) {
new_skb = netdev_alloc_skb_ip_align(ndev, priv->rx_packet_max);
if (new_skb) {
skb_put(skb, len);
cpts_rx_timestamp(priv->cpts, skb);
skb->protocol = eth_type_trans(skb, ndev);
priv->stats.rx_bytes += len;
skb = NULL;
if (unlikely(!netif_running(ndev))) {
if (skb)
} else {
new_skb = skb;
if (likely(!skb)) {
skb = netdev_alloc_skb_ip_align(ndev, priv->rx_packet_max);
if (WARN_ON(!skb))
ret = cpdma_chan_submit(priv->rxch, skb, skb->data,
skb_tailroom(skb), 0);
WARN_ON(ret < 0);
ret = cpdma_chan_submit(priv->rxch, new_skb, new_skb->data,
skb_tailroom(new_skb), 0);
if (WARN_ON(ret < 0))
static irqreturn_t cpsw_interrupt(int irq, void *dev_id)
......@@ -776,6 +776,7 @@ static int __cpdma_chan_process(struct cpdma_chan *chan)
struct cpdma_ctlr *ctlr = chan->ctlr;
struct cpdma_desc __iomem *desc;
int status, outlen;
int cb_status = 0;
struct cpdma_desc_pool *pool = ctlr->pool;
dma_addr_t desc_dma;
unsigned long flags;
......@@ -811,8 +812,12 @@ static int __cpdma_chan_process(struct cpdma_chan *chan)
spin_unlock_irqrestore(&chan->lock, flags);
if (unlikely(status & CPDMA_DESC_TD_COMPLETE))
cb_status = -ENOSYS;
cb_status = status;
__cpdma_chan_free(chan, desc, outlen, status);
__cpdma_chan_free(chan, desc, outlen, cb_status);
return status;
Supports Markdown
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment