summaryrefslogtreecommitdiff
path: root/drivers/net/veth.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/net/veth.c')
-rw-r--r--drivers/net/veth.c1190
1 files changed, 854 insertions, 336 deletions
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index b594f03eeddb..14e6f2a2fb77 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -17,6 +17,7 @@
#include <net/rtnetlink.h>
#include <net/dst.h>
+#include <net/netdev_lock.h>
#include <net/xfrm.h>
#include <net/xdp.h>
#include <linux/veth.h>
@@ -26,6 +27,8 @@
#include <linux/ptr_ring.h>
#include <linux/bpf_trace.h>
#include <linux/net_tstamp.h>
+#include <linux/skbuff_ref.h>
+#include <net/page_pool/helpers.h>
#define DRV_NAME "veth"
#define DRV_VERSION "1.0"
@@ -35,6 +38,7 @@
#define VETH_XDP_HEADROOM (XDP_PACKET_HEADROOM + NET_IP_ALIGN)
#define VETH_XDP_TX_BULK_SIZE 16
+#define VETH_XDP_BATCH 16
struct veth_stats {
u64 rx_drops;
@@ -56,6 +60,7 @@ struct veth_rq_stats {
struct veth_rq {
struct napi_struct xdp_napi;
+ struct napi_struct __rcu *napi; /* points to xdp_napi when the latter is initialized */
struct net_device *dev;
struct bpf_prog __rcu *xdp_prog;
struct xdp_mem_info xdp_mem;
@@ -63,6 +68,7 @@ struct veth_rq {
bool rx_notify_masked;
struct ptr_ring xdp_ring;
struct xdp_rxq_info xdp_rxq;
+ struct page_pool *page_pool;
};
struct veth_priv {
@@ -114,6 +120,11 @@ static struct {
{ "peer_ifindex" },
};
+struct veth_xdp_buff {
+ struct xdp_buff xdp;
+ struct sk_buff *skb;
+};
+
static int veth_get_link_ksettings(struct net_device *dev,
struct ethtool_link_ksettings *cmd)
{
@@ -126,35 +137,30 @@ static int veth_get_link_ksettings(struct net_device *dev,
static void veth_get_drvinfo(struct net_device *dev, struct ethtool_drvinfo *info)
{
- strlcpy(info->driver, DRV_NAME, sizeof(info->driver));
- strlcpy(info->version, DRV_VERSION, sizeof(info->version));
+ strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+ strscpy(info->version, DRV_VERSION, sizeof(info->version));
}
static void veth_get_strings(struct net_device *dev, u32 stringset, u8 *buf)
{
- char *p = (char *)buf;
+ u8 *p = buf;
int i, j;
switch(stringset) {
case ETH_SS_STATS:
memcpy(p, &ethtool_stats_keys, sizeof(ethtool_stats_keys));
p += sizeof(ethtool_stats_keys);
- for (i = 0; i < dev->real_num_rx_queues; i++) {
- for (j = 0; j < VETH_RQ_STATS_LEN; j++) {
- snprintf(p, ETH_GSTRING_LEN,
- "rx_queue_%u_%.18s",
- i, veth_rq_stats_desc[j].desc);
- p += ETH_GSTRING_LEN;
- }
- }
- for (i = 0; i < dev->real_num_tx_queues; i++) {
- for (j = 0; j < VETH_TQ_STATS_LEN; j++) {
- snprintf(p, ETH_GSTRING_LEN,
- "tx_queue_%u_%.18s",
- i, veth_tq_stats_desc[j].desc);
- p += ETH_GSTRING_LEN;
- }
- }
+ for (i = 0; i < dev->real_num_rx_queues; i++)
+ for (j = 0; j < VETH_RQ_STATS_LEN; j++)
+ ethtool_sprintf(&p, "rx_queue_%u_%.18s",
+ i, veth_rq_stats_desc[j].desc);
+
+ for (i = 0; i < dev->real_num_tx_queues; i++)
+ for (j = 0; j < VETH_TQ_STATS_LEN; j++)
+ ethtool_sprintf(&p, "tx_queue_%u_%.18s",
+ i, veth_tq_stats_desc[j].desc);
+
+ page_pool_ethtool_stats_get_strings(p);
break;
}
}
@@ -165,18 +171,35 @@ static int veth_get_sset_count(struct net_device *dev, int sset)
case ETH_SS_STATS:
return ARRAY_SIZE(ethtool_stats_keys) +
VETH_RQ_STATS_LEN * dev->real_num_rx_queues +
- VETH_TQ_STATS_LEN * dev->real_num_tx_queues;
+ VETH_TQ_STATS_LEN * dev->real_num_tx_queues +
+ page_pool_ethtool_stats_get_count();
default:
return -EOPNOTSUPP;
}
}
+static void veth_get_page_pool_stats(struct net_device *dev, u64 *data)
+{
+#ifdef CONFIG_PAGE_POOL_STATS
+ struct veth_priv *priv = netdev_priv(dev);
+ struct page_pool_stats pp_stats = {};
+ int i;
+
+ for (i = 0; i < dev->real_num_rx_queues; i++) {
+ if (!priv->rq[i].page_pool)
+ continue;
+ page_pool_get_stats(priv->rq[i].page_pool, &pp_stats);
+ }
+ page_pool_ethtool_stats_get(data, &pp_stats);
+#endif /* CONFIG_PAGE_POOL_STATS */
+}
+
static void veth_get_ethtool_stats(struct net_device *dev,
struct ethtool_stats *stats, u64 *data)
{
struct veth_priv *rcv_priv, *priv = netdev_priv(dev);
struct net_device *peer = rtnl_dereference(priv->peer);
- int i, j, idx;
+ int i, j, idx, pp_idx;
data[0] = peer ? peer->ifindex : 0;
idx = 1;
@@ -187,17 +210,18 @@ static void veth_get_ethtool_stats(struct net_device *dev,
size_t offset;
do {
- start = u64_stats_fetch_begin_irq(&rq_stats->syncp);
+ start = u64_stats_fetch_begin(&rq_stats->syncp);
for (j = 0; j < VETH_RQ_STATS_LEN; j++) {
offset = veth_rq_stats_desc[j].offset;
data[idx + j] = *(u64 *)(stats_base + offset);
}
- } while (u64_stats_fetch_retry_irq(&rq_stats->syncp, start));
+ } while (u64_stats_fetch_retry(&rq_stats->syncp, start));
idx += VETH_RQ_STATS_LEN;
}
+ pp_idx = idx;
if (!peer)
- return;
+ goto page_pool_stats;
rcv_priv = netdev_priv(peer);
for (i = 0; i < peer->real_num_rx_queues; i++) {
@@ -208,15 +232,31 @@ static void veth_get_ethtool_stats(struct net_device *dev,
tx_idx += (i % dev->real_num_tx_queues) * VETH_TQ_STATS_LEN;
do {
- start = u64_stats_fetch_begin_irq(&rq_stats->syncp);
+ start = u64_stats_fetch_begin(&rq_stats->syncp);
for (j = 0; j < VETH_TQ_STATS_LEN; j++) {
offset = veth_tq_stats_desc[j].offset;
data[tx_idx + j] += *(u64 *)(base + offset);
}
- } while (u64_stats_fetch_retry_irq(&rq_stats->syncp, start));
+ } while (u64_stats_fetch_retry(&rq_stats->syncp, start));
}
+ pp_idx = idx + dev->real_num_tx_queues * VETH_TQ_STATS_LEN;
+
+page_pool_stats:
+ veth_get_page_pool_stats(dev, &data[pp_idx]);
+}
+
+static void veth_get_channels(struct net_device *dev,
+ struct ethtool_channels *channels)
+{
+ channels->tx_count = dev->real_num_tx_queues;
+ channels->rx_count = dev->real_num_rx_queues;
+ channels->max_tx = dev->num_tx_queues;
+ channels->max_rx = dev->num_rx_queues;
}
+static int veth_set_channels(struct net_device *dev,
+ struct ethtool_channels *ch);
+
static const struct ethtool_ops veth_ethtool_ops = {
.get_drvinfo = veth_get_drvinfo,
.get_link = ethtool_op_get_link,
@@ -225,6 +265,8 @@ static const struct ethtool_ops veth_ethtool_ops = {
.get_ethtool_stats = veth_get_ethtool_stats,
.get_link_ksettings = veth_get_link_ksettings,
.get_ts_info = ethtool_op_get_ts_info,
+ .get_channels = veth_get_channels,
+ .set_channels = veth_set_channels,
};
/* general routines */
@@ -234,14 +276,14 @@ static bool veth_is_xdp_frame(void *ptr)
return (unsigned long)ptr & VETH_XDP_FLAG;
}
-static void *veth_ptr_to_xdp(void *ptr)
+static struct xdp_frame *veth_ptr_to_xdp(void *ptr)
{
return (void *)((unsigned long)ptr & ~VETH_XDP_FLAG);
}
-static void *veth_xdp_to_ptr(void *ptr)
+static void *veth_xdp_to_ptr(struct xdp_frame *xdp)
{
- return (void *)((unsigned long)ptr | VETH_XDP_FLAG);
+ return (void *)((unsigned long)xdp | VETH_XDP_FLAG);
}
static void veth_ptr_free(void *ptr)
@@ -256,20 +298,19 @@ static void __veth_xdp_flush(struct veth_rq *rq)
{
/* Write ptr_ring before reading rx_notify_masked */
smp_mb();
- if (!rq->rx_notify_masked) {
- rq->rx_notify_masked = true;
- napi_schedule(&rq->xdp_napi);
+ if (!READ_ONCE(rq->rx_notify_masked) &&
+ napi_schedule_prep(&rq->xdp_napi)) {
+ WRITE_ONCE(rq->rx_notify_masked, true);
+ __napi_schedule(&rq->xdp_napi);
}
}
static int veth_xdp_rx(struct veth_rq *rq, struct sk_buff *skb)
{
- if (unlikely(ptr_ring_produce(&rq->xdp_ring, skb))) {
- dev_kfree_skb_any(skb);
- return NET_RX_DROP;
- }
+ if (unlikely(ptr_ring_produce(&rq->xdp_ring, skb)))
+ return NETDEV_TX_BUSY; /* signal qdisc layer */
- return NET_RX_SUCCESS;
+ return NET_RX_SUCCESS; /* same as NETDEV_TX_OK */
}
static int veth_forward_skb(struct net_device *dev, struct sk_buff *skb,
@@ -277,21 +318,41 @@ static int veth_forward_skb(struct net_device *dev, struct sk_buff *skb,
{
return __dev_forward_skb(dev, skb) ?: xdp ?
veth_xdp_rx(rq, skb) :
- netif_rx(skb);
+ __netif_rx(skb);
+}
+
+/* return true if the specified skb has chances of GRO aggregation
+ * Don't strive for accuracy, but try to avoid GRO overhead in the most
+ * common scenarios.
+ * When XDP is enabled, all traffic is considered eligible, as the xmit
+ * device has TSO off.
+ * When TSO is enabled on the xmit device, we are likely interested only
+ * in UDP aggregation, explicitly check for that if the skb is suspected
+ * - the sock_wfree destructor is used by UDP, ICMP and XDP sockets -
+ * to belong to locally generated UDP traffic.
+ */
+static bool veth_skb_is_eligible_for_gro(const struct net_device *dev,
+ const struct net_device *rcv,
+ const struct sk_buff *skb)
+{
+ return !(dev->features & NETIF_F_ALL_TSO) ||
+ (skb->destructor == sock_wfree &&
+ rcv->features & (NETIF_F_GRO_FRAGLIST | NETIF_F_GRO_UDP_FWD));
}
static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
{
struct veth_priv *rcv_priv, *priv = netdev_priv(dev);
struct veth_rq *rq = NULL;
+ struct netdev_queue *txq;
struct net_device *rcv;
int length = skb->len;
- bool rcv_xdp = false;
- int rxq;
+ bool use_napi = false;
+ int ret, rxq;
rcu_read_lock();
rcv = rcu_dereference(priv->peer);
- if (unlikely(!rcv)) {
+ if (unlikely(!rcv) || !pskb_may_pull(skb, ETH_HLEN)) {
kfree_skb(skb);
goto drop;
}
@@ -300,34 +361,56 @@ static netdev_tx_t veth_xmit(struct sk_buff *skb, struct net_device *dev)
rxq = skb_get_queue_mapping(skb);
if (rxq < rcv->real_num_rx_queues) {
rq = &rcv_priv->rq[rxq];
- rcv_xdp = rcu_access_pointer(rq->xdp_prog);
- if (rcv_xdp)
- skb_record_rx_queue(skb, rxq);
+
+ /* The napi pointer is available when an XDP program is
+ * attached or when GRO is enabled
+ * Don't bother with napi/GRO if the skb can't be aggregated
+ */
+ use_napi = rcu_access_pointer(rq->napi) &&
+ veth_skb_is_eligible_for_gro(dev, rcv, skb);
}
skb_tx_timestamp(skb);
- if (likely(veth_forward_skb(rcv, skb, rq, rcv_xdp) == NET_RX_SUCCESS)) {
- if (!rcv_xdp)
- dev_lstats_add(dev, length);
- } else {
+
+ ret = veth_forward_skb(rcv, skb, rq, use_napi);
+ switch (ret) {
+ case NET_RX_SUCCESS: /* same as NETDEV_TX_OK */
+ if (!use_napi)
+ dev_sw_netstats_tx_add(dev, 1, length);
+ else
+ __veth_xdp_flush(rq);
+ break;
+ case NETDEV_TX_BUSY:
+ /* If a qdisc is attached to our virtual device, returning
+ * NETDEV_TX_BUSY is allowed.
+ */
+ txq = netdev_get_tx_queue(dev, rxq);
+
+ if (qdisc_txq_has_no_queue(txq)) {
+ dev_kfree_skb_any(skb);
+ goto drop;
+ }
+ /* Restore Eth hdr pulled by dev_forward_skb/eth_type_trans */
+ __skb_push(skb, ETH_HLEN);
+ netif_tx_stop_queue(txq);
+ /* Makes sure NAPI peer consumer runs. Consumer is responsible
+ * for starting txq again, until then ndo_start_xmit (this
+ * function) will not be invoked by the netstack again.
+ */
+ __veth_xdp_flush(rq);
+ break;
+ case NET_RX_DROP: /* same as NET_XMIT_DROP */
drop:
atomic64_inc(&priv->dropped);
+ ret = NET_XMIT_DROP;
+ break;
+ default:
+ net_crit_ratelimited("%s(%s): Invalid return code(%d)",
+ __func__, dev->name, ret);
}
-
- if (rcv_xdp)
- __veth_xdp_flush(rq);
-
rcu_read_unlock();
- return NETDEV_TX_OK;
-}
-
-static u64 veth_stats_tx(struct net_device *dev, u64 *packets, u64 *bytes)
-{
- struct veth_priv *priv = netdev_priv(dev);
-
- dev_lstats_read(dev, packets, bytes);
- return atomic64_read(&priv->dropped);
+ return ret;
}
static void veth_stats_rx(struct veth_stats *result, struct net_device *dev)
@@ -346,13 +429,13 @@ static void veth_stats_rx(struct veth_stats *result, struct net_device *dev)
unsigned int start;
do {
- start = u64_stats_fetch_begin_irq(&stats->syncp);
+ start = u64_stats_fetch_begin(&stats->syncp);
peer_tq_xdp_xmit_err = stats->vs.peer_tq_xdp_xmit_err;
xdp_tx_err = stats->vs.xdp_tx_err;
packets = stats->vs.xdp_packets;
bytes = stats->vs.xdp_bytes;
drops = stats->vs.rx_drops;
- } while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+ } while (u64_stats_fetch_retry(&stats->syncp, start));
result->peer_tq_xdp_xmit_err += peer_tq_xdp_xmit_err;
result->xdp_tx_err += xdp_tx_err;
result->xdp_packets += packets;
@@ -367,24 +450,24 @@ static void veth_get_stats64(struct net_device *dev,
struct veth_priv *priv = netdev_priv(dev);
struct net_device *peer;
struct veth_stats rx;
- u64 packets, bytes;
- tot->tx_dropped = veth_stats_tx(dev, &packets, &bytes);
- tot->tx_bytes = bytes;
- tot->tx_packets = packets;
+ tot->tx_dropped = atomic64_read(&priv->dropped);
+ dev_fetch_sw_netstats(tot, dev->tstats);
veth_stats_rx(&rx, dev);
tot->tx_dropped += rx.xdp_tx_err;
tot->rx_dropped = rx.rx_drops + rx.peer_tq_xdp_xmit_err;
- tot->rx_bytes = rx.xdp_bytes;
- tot->rx_packets = rx.xdp_packets;
+ tot->rx_bytes += rx.xdp_bytes;
+ tot->rx_packets += rx.xdp_packets;
rcu_read_lock();
peer = rcu_dereference(priv->peer);
if (peer) {
- veth_stats_tx(peer, &packets, &bytes);
- tot->rx_bytes += bytes;
- tot->rx_packets += packets;
+ struct rtnl_link_stats64 tot_peer = {};
+
+ dev_fetch_sw_netstats(&tot_peer, peer->tstats);
+ tot->rx_bytes += tot_peer.tx_bytes;
+ tot->rx_packets += tot_peer.tx_packets;
veth_stats_rx(&rx, peer);
tot->tx_dropped += rx.peer_tq_xdp_xmit_err;
@@ -400,24 +483,17 @@ static void veth_set_multicast_list(struct net_device *dev)
{
}
-static struct sk_buff *veth_build_skb(void *head, int headroom, int len,
- int buflen)
+static int veth_select_rxq(struct net_device *dev)
{
- struct sk_buff *skb;
-
- skb = build_skb(head, buflen);
- if (!skb)
- return NULL;
-
- skb_reserve(skb, headroom);
- skb_put(skb, len);
-
- return skb;
+ return smp_processor_id() % dev->real_num_rx_queues;
}
-static int veth_select_rxq(struct net_device *dev)
+static struct net_device *veth_peer_dev(struct net_device *dev)
{
- return smp_processor_id() % dev->real_num_rx_queues;
+ struct veth_priv *priv = netdev_priv(dev);
+
+ /* Callers must be under RCU read side. */
+ return rcu_dereference(priv->peer);
}
static int veth_xdp_xmit(struct net_device *dev, int n,
@@ -425,7 +501,7 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
u32 flags, bool ndo_xmit)
{
struct veth_priv *rcv_priv, *priv = netdev_priv(dev);
- int i, ret = -ENXIO, drops = 0;
+ int i, ret = -ENXIO, nxmit = 0;
struct net_device *rcv;
unsigned int max_len;
struct veth_rq *rq;
@@ -440,11 +516,10 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
rcv_priv = netdev_priv(rcv);
rq = &rcv_priv->rq[veth_select_rxq(rcv)];
- /* Non-NULL xdp_prog ensures that xdp_ring is initialized on receive
- * side. This means an XDP program is loaded on the peer and the peer
- * device is up.
+ /* The napi pointer is set if NAPI is enabled, which ensures that
+ * xdp_ring is initialized on receive side and the peer device is up.
*/
- if (!rcu_access_pointer(rq->xdp_prog))
+ if (!rcu_access_pointer(rq->napi))
goto out;
max_len = rcv->mtu + rcv->hard_header_len + VLAN_HLEN;
@@ -454,22 +529,21 @@ static int veth_xdp_xmit(struct net_device *dev, int n,
struct xdp_frame *frame = frames[i];
void *ptr = veth_xdp_to_ptr(frame);
- if (unlikely(frame->len > max_len ||
- __ptr_ring_produce(&rq->xdp_ring, ptr))) {
- xdp_return_frame_rx_napi(frame);
- drops++;
- }
+ if (unlikely(xdp_get_frame_len(frame) > max_len ||
+ __ptr_ring_produce(&rq->xdp_ring, ptr)))
+ break;
+ nxmit++;
}
spin_unlock(&rq->xdp_ring.producer_lock);
if (flags & XDP_XMIT_FLUSH)
__veth_xdp_flush(rq);
- ret = n - drops;
+ ret = nxmit;
if (ndo_xmit) {
u64_stats_update_begin(&rq->stats.syncp);
- rq->stats.vs.peer_tq_xdp_xmit += n - drops;
- rq->stats.vs.peer_tq_xdp_xmit_err += drops;
+ rq->stats.vs.peer_tq_xdp_xmit += nxmit;
+ rq->stats.vs.peer_tq_xdp_xmit_err += n - nxmit;
u64_stats_update_end(&rq->stats.syncp);
}
@@ -496,20 +570,23 @@ static int veth_ndo_xdp_xmit(struct net_device *dev, int n,
static void veth_xdp_flush_bq(struct veth_rq *rq, struct veth_xdp_tx_bq *bq)
{
- int sent, i, err = 0;
+ int sent, i, err = 0, drops;
sent = veth_xdp_xmit(rq->dev, bq->count, bq->q, 0, false);
if (sent < 0) {
err = sent;
sent = 0;
- for (i = 0; i < bq->count; i++)
- xdp_return_frame(bq->q[i]);
}
- trace_xdp_bulk_tx(rq->dev, sent, bq->count - sent, err);
+
+ for (i = sent; unlikely(i < bq->count); i++)
+ xdp_return_frame(bq->q[i]);
+
+ drops = bq->count - sent;
+ trace_xdp_bulk_tx(rq->dev, sent, drops, err);
u64_stats_update_begin(&rq->stats.syncp);
rq->stats.vs.xdp_tx += sent;
- rq->stats.vs.xdp_tx_err += bq->count - sent;
+ rq->stats.vs.xdp_tx_err += drops;
u64_stats_update_end(&rq->stats.syncp);
bq->count = 0;
@@ -554,41 +631,36 @@ static int veth_xdp_tx(struct veth_rq *rq, struct xdp_buff *xdp,
return 0;
}
-static struct sk_buff *veth_xdp_rcv_one(struct veth_rq *rq,
- struct xdp_frame *frame,
- struct veth_xdp_tx_bq *bq,
- struct veth_stats *stats)
+static struct xdp_frame *veth_xdp_rcv_one(struct veth_rq *rq,
+ struct xdp_frame *frame,
+ struct veth_xdp_tx_bq *bq,
+ struct veth_stats *stats)
{
- void *hard_start = frame->data - frame->headroom;
- int len = frame->len, delta = 0;
struct xdp_frame orig_frame;
struct bpf_prog *xdp_prog;
- unsigned int headroom;
- struct sk_buff *skb;
-
- /* bpf_xdp_adjust_head() assures BPF cannot access xdp_frame area */
- hard_start -= sizeof(struct xdp_frame);
rcu_read_lock();
xdp_prog = rcu_dereference(rq->xdp_prog);
if (likely(xdp_prog)) {
- struct xdp_buff xdp;
+ struct veth_xdp_buff vxbuf;
+ struct xdp_buff *xdp = &vxbuf.xdp;
u32 act;
- xdp_convert_frame_to_buff(frame, &xdp);
- xdp.rxq = &rq->xdp_rxq;
+ xdp_convert_frame_to_buff(frame, xdp);
+ xdp->rxq = &rq->xdp_rxq;
+ vxbuf.skb = NULL;
- act = bpf_prog_run_xdp(xdp_prog, &xdp);
+ act = bpf_prog_run_xdp(xdp_prog, xdp);
switch (act) {
case XDP_PASS:
- delta = frame->data - xdp.data;
- len = xdp.data_end - xdp.data;
+ if (xdp_update_frame_from_buff(xdp, frame))
+ goto err_xdp;
break;
case XDP_TX:
orig_frame = *frame;
- xdp.rxq->mem = frame->mem;
- if (unlikely(veth_xdp_tx(rq, &xdp, bq) < 0)) {
+ xdp->rxq->mem.type = frame->mem_type;
+ if (unlikely(veth_xdp_tx(rq, xdp, bq) < 0)) {
trace_xdp_exception(rq->dev, xdp_prog, act);
frame = &orig_frame;
stats->rx_drops++;
@@ -599,8 +671,8 @@ static struct sk_buff *veth_xdp_rcv_one(struct veth_rq *rq,
goto xdp_xmit;
case XDP_REDIRECT:
orig_frame = *frame;
- xdp.rxq->mem = frame->mem;
- if (xdp_do_redirect(rq->dev, &xdp, xdp_prog)) {
+ xdp->rxq->mem.type = frame->mem_type;
+ if (xdp_do_redirect(rq->dev, xdp, xdp_prog)) {
frame = &orig_frame;
stats->rx_drops++;
goto err_xdp;
@@ -609,11 +681,11 @@ static struct sk_buff *veth_xdp_rcv_one(struct veth_rq *rq,
rcu_read_unlock();
goto xdp_xmit;
default:
- bpf_warn_invalid_xdp_action(act);
- /* fall through */
+ bpf_warn_invalid_xdp_action(rq->dev, xdp_prog, act);
+ fallthrough;
case XDP_ABORTED:
trace_xdp_exception(rq->dev, xdp_prog, act);
- /* fall through */
+ fallthrough;
case XDP_DROP:
stats->xdp_drops++;
goto err_xdp;
@@ -621,19 +693,7 @@ static struct sk_buff *veth_xdp_rcv_one(struct veth_rq *rq,
}
rcu_read_unlock();
- headroom = sizeof(struct xdp_frame) + frame->headroom - delta;
- skb = veth_build_skb(hard_start, headroom, len, frame->frame_sz);
- if (!skb) {
- xdp_return_frame(frame);
- stats->rx_drops++;
- goto err;
- }
-
- xdp_release_frame(frame);
- xdp_scrub_frame(frame);
- skb->protocol = eth_type_trans(skb, rq->dev);
-err:
- return skb;
+ return frame;
err_xdp:
rcu_read_unlock();
xdp_return_frame(frame);
@@ -641,18 +701,101 @@ xdp_xmit:
return NULL;
}
+/* frames array contains VETH_XDP_BATCH at most */
+static void veth_xdp_rcv_bulk_skb(struct veth_rq *rq, void **frames,
+ int n_xdpf, struct veth_xdp_tx_bq *bq,
+ struct veth_stats *stats)
+{
+ void *skbs[VETH_XDP_BATCH];
+ int i;
+
+ if (unlikely(!napi_skb_cache_get_bulk(skbs, n_xdpf))) {
+ for (i = 0; i < n_xdpf; i++)
+ xdp_return_frame(frames[i]);
+ stats->rx_drops += n_xdpf;
+
+ return;
+ }
+
+ for (i = 0; i < n_xdpf; i++) {
+ struct sk_buff *skb = skbs[i];
+
+ skb = __xdp_build_skb_from_frame(frames[i], skb,
+ rq->dev);
+ if (!skb) {
+ xdp_return_frame(frames[i]);
+ stats->rx_drops++;
+ continue;
+ }
+ napi_gro_receive(&rq->xdp_napi, skb);
+ }
+}
+
+static void veth_xdp_get(struct xdp_buff *xdp)
+{
+ struct skb_shared_info *sinfo = xdp_get_shared_info_from_buff(xdp);
+ int i;
+
+ get_page(virt_to_page(xdp->data));
+ if (likely(!xdp_buff_has_frags(xdp)))
+ return;
+
+ for (i = 0; i < sinfo->nr_frags; i++)
+ __skb_frag_ref(&sinfo->frags[i]);
+}
+
+static int veth_convert_skb_to_xdp_buff(struct veth_rq *rq,
+ struct xdp_buff *xdp,
+ struct sk_buff **pskb)
+{
+ struct sk_buff *skb = *pskb;
+ u32 frame_sz;
+
+ if (skb_shared(skb) || skb_head_is_locked(skb) ||
+ skb_shinfo(skb)->nr_frags ||
+ skb_headroom(skb) < XDP_PACKET_HEADROOM) {
+ if (skb_pp_cow_data(rq->page_pool, pskb, XDP_PACKET_HEADROOM))
+ goto drop;
+
+ skb = *pskb;
+ }
+
+ /* SKB "head" area always have tailroom for skb_shared_info */
+ frame_sz = skb_end_pointer(skb) - skb->head;
+ frame_sz += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+ xdp_init_buff(xdp, frame_sz, &rq->xdp_rxq);
+ xdp_prepare_buff(xdp, skb->head, skb_headroom(skb),
+ skb_headlen(skb), true);
+
+ if (skb_is_nonlinear(skb)) {
+ skb_shinfo(skb)->xdp_frags_size = skb->data_len;
+ xdp_buff_set_frags_flag(xdp);
+ } else {
+ xdp_buff_clear_frags_flag(xdp);
+ }
+ *pskb = skb;
+
+ return 0;
+drop:
+ consume_skb(skb);
+ *pskb = NULL;
+
+ return -ENOMEM;
+}
+
static struct sk_buff *veth_xdp_rcv_skb(struct veth_rq *rq,
struct sk_buff *skb,
struct veth_xdp_tx_bq *bq,
struct veth_stats *stats)
{
- u32 pktlen, headroom, act, metalen;
void *orig_data, *orig_data_end;
struct bpf_prog *xdp_prog;
- int mac_len, delta, off;
- struct xdp_buff xdp;
+ struct veth_xdp_buff vxbuf;
+ struct xdp_buff *xdp = &vxbuf.xdp;
+ u32 act, metalen;
+ int off;
- skb_orphan(skb);
+ skb_prepare_for_gro(skb);
rcu_read_lock();
xdp_prog = rcu_dereference(rq->xdp_prog);
@@ -661,70 +804,24 @@ static struct sk_buff *veth_xdp_rcv_skb(struct veth_rq *rq,
goto out;
}
- mac_len = skb->data - skb_mac_header(skb);
- pktlen = skb->len + mac_len;
- headroom = skb_headroom(skb) - mac_len;
-
- if (skb_shared(skb) || skb_head_is_locked(skb) ||
- skb_is_nonlinear(skb) || headroom < XDP_PACKET_HEADROOM) {
- struct sk_buff *nskb;
- int size, head_off;
- void *head, *start;
- struct page *page;
-
- size = SKB_DATA_ALIGN(VETH_XDP_HEADROOM + pktlen) +
- SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
- if (size > PAGE_SIZE)
- goto drop;
-
- page = alloc_page(GFP_ATOMIC | __GFP_NOWARN);
- if (!page)
- goto drop;
-
- head = page_address(page);
- start = head + VETH_XDP_HEADROOM;
- if (skb_copy_bits(skb, -mac_len, start, pktlen)) {
- page_frag_free(head);
- goto drop;
- }
-
- nskb = veth_build_skb(head, VETH_XDP_HEADROOM + mac_len,
- skb->len, PAGE_SIZE);
- if (!nskb) {
- page_frag_free(head);
- goto drop;
- }
-
- skb_copy_header(nskb, skb);
- head_off = skb_headroom(nskb) - skb_headroom(skb);
- skb_headers_offset_update(nskb, head_off);
- consume_skb(skb);
- skb = nskb;
- }
-
- xdp.data_hard_start = skb->head;
- xdp.data = skb_mac_header(skb);
- xdp.data_end = xdp.data + pktlen;
- xdp.data_meta = xdp.data;
- xdp.rxq = &rq->xdp_rxq;
-
- /* SKB "head" area always have tailroom for skb_shared_info */
- xdp.frame_sz = (void *)skb_end_pointer(skb) - xdp.data_hard_start;
- xdp.frame_sz += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+ __skb_push(skb, skb->data - skb_mac_header(skb));
+ if (veth_convert_skb_to_xdp_buff(rq, xdp, &skb))
+ goto drop;
+ vxbuf.skb = skb;
- orig_data = xdp.data;
- orig_data_end = xdp.data_end;
+ orig_data = xdp->data;
+ orig_data_end = xdp->data_end;
- act = bpf_prog_run_xdp(xdp_prog, &xdp);
+ act = bpf_prog_run_xdp(xdp_prog, xdp);
switch (act) {
case XDP_PASS:
break;
case XDP_TX:
- get_page(virt_to_page(xdp.data));
+ veth_xdp_get(xdp);
consume_skb(skb);
- xdp.rxq->mem = rq->xdp_mem;
- if (unlikely(veth_xdp_tx(rq, &xdp, bq) < 0)) {
+ xdp->rxq->mem = rq->xdp_mem;
+ if (unlikely(veth_xdp_tx(rq, xdp, bq) < 0)) {
trace_xdp_exception(rq->dev, xdp_prog, act);
stats->rx_drops++;
goto err_xdp;
@@ -733,10 +830,10 @@ static struct sk_buff *veth_xdp_rcv_skb(struct veth_rq *rq,
rcu_read_unlock();
goto xdp_xmit;
case XDP_REDIRECT:
- get_page(virt_to_page(xdp.data));
+ veth_xdp_get(xdp);
consume_skb(skb);
- xdp.rxq->mem = rq->xdp_mem;
- if (xdp_do_redirect(rq->dev, &xdp, xdp_prog)) {
+ xdp->rxq->mem = rq->xdp_mem;
+ if (xdp_do_redirect(rq->dev, xdp, xdp_prog)) {
stats->rx_drops++;
goto err_xdp;
}
@@ -744,11 +841,11 @@ static struct sk_buff *veth_xdp_rcv_skb(struct veth_rq *rq,
rcu_read_unlock();
goto xdp_xmit;
default:
- bpf_warn_invalid_xdp_action(act);
- /* fall through */
+ bpf_warn_invalid_xdp_action(rq->dev, xdp_prog, act);
+ fallthrough;
case XDP_ABORTED:
trace_xdp_exception(rq->dev, xdp_prog, act);
- /* fall through */
+ fallthrough;
case XDP_DROP:
stats->xdp_drops++;
goto xdp_drop;
@@ -756,21 +853,30 @@ static struct sk_buff *veth_xdp_rcv_skb(struct veth_rq *rq,
rcu_read_unlock();
/* check if bpf_xdp_adjust_head was used */
- delta = orig_data - xdp.data;
- off = mac_len + delta;
+ off = orig_data - xdp->data;
if (off > 0)
__skb_push(skb, off);
else if (off < 0)
__skb_pull(skb, -off);
- skb->mac_header -= delta;
+
+ skb_reset_mac_header(skb);
/* check if bpf_xdp_adjust_tail was used */
- off = xdp.data_end - orig_data_end;
+ off = xdp->data_end - orig_data_end;
if (off != 0)
__skb_put(skb, off); /* positive on grow, negative on shrink */
+
+ /* XDP frag metadata (e.g. nr_frags) are updated in eBPF helpers
+ * (e.g. bpf_xdp_adjust_tail), we need to update data_len here.
+ */
+ if (xdp_buff_has_frags(xdp))
+ skb->data_len = skb_shinfo(skb)->xdp_frags_size;
+ else
+ skb->data_len = 0;
+
skb->protocol = eth_type_trans(skb, rq->dev);
- metalen = xdp.data - xdp.data_meta;
+ metalen = xdp->data - xdp->data_meta;
if (metalen)
skb_metadata_set(skb, metalen);
out:
@@ -783,7 +889,7 @@ xdp_drop:
return NULL;
err_xdp:
rcu_read_unlock();
- page_frag_free(xdp.data);
+ xdp_return_buff(xdp);
xdp_xmit:
return NULL;
}
@@ -792,32 +898,49 @@ static int veth_xdp_rcv(struct veth_rq *rq, int budget,
struct veth_xdp_tx_bq *bq,
struct veth_stats *stats)
{
- int i, done = 0;
+ int i, done = 0, n_xdpf = 0;
+ void *xdpf[VETH_XDP_BATCH];
for (i = 0; i < budget; i++) {
void *ptr = __ptr_ring_consume(&rq->xdp_ring);
- struct sk_buff *skb;
if (!ptr)
break;
if (veth_is_xdp_frame(ptr)) {
+ /* ndo_xdp_xmit */
struct xdp_frame *frame = veth_ptr_to_xdp(ptr);
- stats->xdp_bytes += frame->len;
- skb = veth_xdp_rcv_one(rq, frame, bq, stats);
+ stats->xdp_bytes += xdp_get_frame_len(frame);
+ frame = veth_xdp_rcv_one(rq, frame, bq, stats);
+ if (frame) {
+ /* XDP_PASS */
+ xdpf[n_xdpf++] = frame;
+ if (n_xdpf == VETH_XDP_BATCH) {
+ veth_xdp_rcv_bulk_skb(rq, xdpf, n_xdpf,
+ bq, stats);
+ n_xdpf = 0;
+ }
+ }
} else {
- skb = ptr;
+ /* ndo_start_xmit */
+ struct sk_buff *skb = ptr;
+
stats->xdp_bytes += skb->len;
skb = veth_xdp_rcv_skb(rq, skb, bq, stats);
+ if (skb) {
+ if (skb_shared(skb) || skb_unclone(skb, GFP_ATOMIC))
+ netif_receive_skb(skb);
+ else
+ napi_gro_receive(&rq->xdp_napi, skb);
+ }
}
-
- if (skb)
- napi_gro_receive(&rq->xdp_napi, skb);
-
done++;
}
+ if (n_xdpf)
+ veth_xdp_rcv_bulk_skb(rq, xdpf, n_xdpf, bq, stats);
+
u64_stats_update_begin(&rq->stats.syncp);
rq->stats.vs.xdp_redirect += stats->xdp_redirect;
rq->stats.vs.xdp_bytes += stats->xdp_bytes;
@@ -833,39 +956,82 @@ static int veth_poll(struct napi_struct *napi, int budget)
{
struct veth_rq *rq =
container_of(napi, struct veth_rq, xdp_napi);
+ struct veth_priv *priv = netdev_priv(rq->dev);
+ int queue_idx = rq->xdp_rxq.queue_index;
+ struct netdev_queue *peer_txq;
struct veth_stats stats = {};
+ struct net_device *peer_dev;
struct veth_xdp_tx_bq bq;
int done;
bq.count = 0;
+ /* NAPI functions as RCU section */
+ peer_dev = rcu_dereference_check(priv->peer, rcu_read_lock_bh_held());
+ peer_txq = peer_dev ? netdev_get_tx_queue(peer_dev, queue_idx) : NULL;
+
xdp_set_return_frame_no_direct();
done = veth_xdp_rcv(rq, budget, &bq, &stats);
+ if (stats.xdp_redirect > 0)
+ xdp_do_flush();
+ if (stats.xdp_tx > 0)
+ veth_xdp_flush(rq, &bq);
+ xdp_clear_return_frame_no_direct();
+
if (done < budget && napi_complete_done(napi, done)) {
/* Write rx_notify_masked before reading ptr_ring */
smp_store_mb(rq->rx_notify_masked, false);
if (unlikely(!__ptr_ring_empty(&rq->xdp_ring))) {
- rq->rx_notify_masked = true;
- napi_schedule(&rq->xdp_napi);
+ if (napi_schedule_prep(&rq->xdp_napi)) {
+ WRITE_ONCE(rq->rx_notify_masked, true);
+ __napi_schedule(&rq->xdp_napi);
+ }
}
}
- if (stats.xdp_tx > 0)
- veth_xdp_flush(rq, &bq);
- if (stats.xdp_redirect > 0)
- xdp_do_flush();
- xdp_clear_return_frame_no_direct();
+ /* Release backpressure per NAPI poll */
+ smp_rmb(); /* Paired with netif_tx_stop_queue set_bit */
+ if (peer_txq && netif_tx_queue_stopped(peer_txq)) {
+ txq_trans_cond_update(peer_txq);
+ netif_tx_wake_queue(peer_txq);
+ }
return done;
}
-static int veth_napi_add(struct net_device *dev)
+static int veth_create_page_pool(struct veth_rq *rq)
+{
+ struct page_pool_params pp_params = {
+ .order = 0,
+ .pool_size = VETH_RING_SIZE,
+ .nid = NUMA_NO_NODE,
+ .dev = &rq->dev->dev,
+ };
+
+ rq->page_pool = page_pool_create(&pp_params);
+ if (IS_ERR(rq->page_pool)) {
+ int err = PTR_ERR(rq->page_pool);
+
+ rq->page_pool = NULL;
+ return err;
+ }
+
+ return 0;
+}
+
+static int __veth_napi_enable_range(struct net_device *dev, int start, int end)
{
struct veth_priv *priv = netdev_priv(dev);
int err, i;
- for (i = 0; i < dev->real_num_rx_queues; i++) {
+ for (i = start; i < end; i++) {
+ err = veth_create_page_pool(&priv->rq[i]);
+ if (err)
+ goto err_page_pool;
+ }
+
+ for (i = start; i < end; i++) {
struct veth_rq *rq = &priv->rq[i];
err = ptr_ring_init(&rq->xdp_ring, VETH_RING_SIZE, GFP_KERNEL);
@@ -873,84 +1039,155 @@ static int veth_napi_add(struct net_device *dev)
goto err_xdp_ring;
}
- for (i = 0; i < dev->real_num_rx_queues; i++) {
+ for (i = start; i < end; i++) {
struct veth_rq *rq = &priv->rq[i];
- netif_napi_add(dev, &rq->xdp_napi, veth_poll, NAPI_POLL_WEIGHT);
napi_enable(&rq->xdp_napi);
+ rcu_assign_pointer(priv->rq[i].napi, &priv->rq[i].xdp_napi);
}
return 0;
+
err_xdp_ring:
- for (i--; i >= 0; i--)
+ for (i--; i >= start; i--)
ptr_ring_cleanup(&priv->rq[i].xdp_ring, veth_ptr_free);
+ i = end;
+err_page_pool:
+ for (i--; i >= start; i--) {
+ page_pool_destroy(priv->rq[i].page_pool);
+ priv->rq[i].page_pool = NULL;
+ }
return err;
}
-static void veth_napi_del(struct net_device *dev)
+static int __veth_napi_enable(struct net_device *dev)
+{
+ return __veth_napi_enable_range(dev, 0, dev->real_num_rx_queues);
+}
+
+static void veth_napi_del_range(struct net_device *dev, int start, int end)
{
struct veth_priv *priv = netdev_priv(dev);
int i;
- for (i = 0; i < dev->real_num_rx_queues; i++) {
+ for (i = start; i < end; i++) {
struct veth_rq *rq = &priv->rq[i];
+ rcu_assign_pointer(priv->rq[i].napi, NULL);
napi_disable(&rq->xdp_napi);
- napi_hash_del(&rq->xdp_napi);
+ __netif_napi_del(&rq->xdp_napi);
}
synchronize_net();
- for (i = 0; i < dev->real_num_rx_queues; i++) {
+ for (i = start; i < end; i++) {
struct veth_rq *rq = &priv->rq[i];
- netif_napi_del(&rq->xdp_napi);
rq->rx_notify_masked = false;
ptr_ring_cleanup(&rq->xdp_ring, veth_ptr_free);
}
+
+ for (i = start; i < end; i++) {
+ page_pool_destroy(priv->rq[i].page_pool);
+ priv->rq[i].page_pool = NULL;
+ }
}
-static int veth_enable_xdp(struct net_device *dev)
+static void veth_napi_del(struct net_device *dev)
{
- struct veth_priv *priv = netdev_priv(dev);
- int err, i;
-
- if (!xdp_rxq_info_is_reg(&priv->rq[0].xdp_rxq)) {
- for (i = 0; i < dev->real_num_rx_queues; i++) {
- struct veth_rq *rq = &priv->rq[i];
+ veth_napi_del_range(dev, 0, dev->real_num_rx_queues);
+}
- err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i);
- if (err < 0)
- goto err_rxq_reg;
+static bool veth_gro_requested(const struct net_device *dev)
+{
+ return !!(dev->wanted_features & NETIF_F_GRO);
+}
- err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq,
- MEM_TYPE_PAGE_SHARED,
- NULL);
- if (err < 0)
- goto err_reg_mem;
+static int veth_enable_xdp_range(struct net_device *dev, int start, int end,
+ bool napi_already_on)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ int err, i;
- /* Save original mem info as it can be overwritten */
- rq->xdp_mem = rq->xdp_rxq.mem;
- }
+ for (i = start; i < end; i++) {
+ struct veth_rq *rq = &priv->rq[i];
- err = veth_napi_add(dev);
- if (err)
+ if (!napi_already_on)
+ netif_napi_add(dev, &rq->xdp_napi, veth_poll);
+ err = xdp_rxq_info_reg(&rq->xdp_rxq, dev, i, rq->xdp_napi.napi_id);
+ if (err < 0)
goto err_rxq_reg;
- }
- for (i = 0; i < dev->real_num_rx_queues; i++)
- rcu_assign_pointer(priv->rq[i].xdp_prog, priv->_xdp_prog);
+ err = xdp_rxq_info_reg_mem_model(&rq->xdp_rxq,
+ MEM_TYPE_PAGE_SHARED,
+ NULL);
+ if (err < 0)
+ goto err_reg_mem;
+ /* Save original mem info as it can be overwritten */
+ rq->xdp_mem = rq->xdp_rxq.mem;
+ }
return 0;
+
err_reg_mem:
xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
err_rxq_reg:
- for (i--; i >= 0; i--)
- xdp_rxq_info_unreg(&priv->rq[i].xdp_rxq);
+ for (i--; i >= start; i--) {
+ struct veth_rq *rq = &priv->rq[i];
+
+ xdp_rxq_info_unreg(&rq->xdp_rxq);
+ if (!napi_already_on)
+ netif_napi_del(&rq->xdp_napi);
+ }
return err;
}
+static void veth_disable_xdp_range(struct net_device *dev, int start, int end,
+ bool delete_napi)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ int i;
+
+ for (i = start; i < end; i++) {
+ struct veth_rq *rq = &priv->rq[i];
+
+ rq->xdp_rxq.mem = rq->xdp_mem;
+ xdp_rxq_info_unreg(&rq->xdp_rxq);
+
+ if (delete_napi)
+ netif_napi_del(&rq->xdp_napi);
+ }
+}
+
+static int veth_enable_xdp(struct net_device *dev)
+{
+ bool napi_already_on = veth_gro_requested(dev) && (dev->flags & IFF_UP);
+ struct veth_priv *priv = netdev_priv(dev);
+ int err, i;
+
+ if (!xdp_rxq_info_is_reg(&priv->rq[0].xdp_rxq)) {
+ err = veth_enable_xdp_range(dev, 0, dev->real_num_rx_queues, napi_already_on);
+ if (err)
+ return err;
+
+ if (!napi_already_on) {
+ err = __veth_napi_enable(dev);
+ if (err) {
+ veth_disable_xdp_range(dev, 0, dev->real_num_rx_queues, true);
+ return err;
+ }
+ }
+ }
+
+ for (i = 0; i < dev->real_num_rx_queues; i++) {
+ rcu_assign_pointer(priv->rq[i].xdp_prog, priv->_xdp_prog);
+ rcu_assign_pointer(priv->rq[i].napi, &priv->rq[i].xdp_napi);
+ }
+
+ return 0;
+}
+
static void veth_disable_xdp(struct net_device *dev)
{
struct veth_priv *priv = netdev_priv(dev);
@@ -958,13 +1195,183 @@ static void veth_disable_xdp(struct net_device *dev)
for (i = 0; i < dev->real_num_rx_queues; i++)
rcu_assign_pointer(priv->rq[i].xdp_prog, NULL);
- veth_napi_del(dev);
- for (i = 0; i < dev->real_num_rx_queues; i++) {
+
+ if (!netif_running(dev) || !veth_gro_requested(dev))
+ veth_napi_del(dev);
+
+ veth_disable_xdp_range(dev, 0, dev->real_num_rx_queues, false);
+}
+
+static int veth_napi_enable_range(struct net_device *dev, int start, int end)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ int err, i;
+
+ for (i = start; i < end; i++) {
struct veth_rq *rq = &priv->rq[i];
- rq->xdp_rxq.mem = rq->xdp_mem;
- xdp_rxq_info_unreg(&rq->xdp_rxq);
+ netif_napi_add(dev, &rq->xdp_napi, veth_poll);
}
+
+ err = __veth_napi_enable_range(dev, start, end);
+ if (err) {
+ for (i = start; i < end; i++) {
+ struct veth_rq *rq = &priv->rq[i];
+
+ netif_napi_del(&rq->xdp_napi);
+ }
+ return err;
+ }
+ return err;
+}
+
+static int veth_napi_enable(struct net_device *dev)
+{
+ return veth_napi_enable_range(dev, 0, dev->real_num_rx_queues);
+}
+
+static void veth_disable_range_safe(struct net_device *dev, int start, int end)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+
+ if (start >= end)
+ return;
+
+ if (priv->_xdp_prog) {
+ veth_napi_del_range(dev, start, end);
+ veth_disable_xdp_range(dev, start, end, false);
+ } else if (veth_gro_requested(dev)) {
+ veth_napi_del_range(dev, start, end);
+ }
+}
+
+static int veth_enable_range_safe(struct net_device *dev, int start, int end)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ int err;
+
+ if (start >= end)
+ return 0;
+
+ if (priv->_xdp_prog) {
+ /* these channels are freshly initialized, napi is not on there even
+ * when GRO is requeste
+ */
+ err = veth_enable_xdp_range(dev, start, end, false);
+ if (err)
+ return err;
+
+ err = __veth_napi_enable_range(dev, start, end);
+ if (err) {
+ /* on error always delete the newly added napis */
+ veth_disable_xdp_range(dev, start, end, true);
+ return err;
+ }
+ } else if (veth_gro_requested(dev)) {
+ return veth_napi_enable_range(dev, start, end);
+ }
+ return 0;
+}
+
+static void veth_set_xdp_features(struct net_device *dev)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ struct net_device *peer;
+
+ peer = rtnl_dereference(priv->peer);
+ if (peer && peer->real_num_tx_queues <= dev->real_num_rx_queues) {
+ struct veth_priv *priv_peer = netdev_priv(peer);
+ xdp_features_t val = NETDEV_XDP_ACT_BASIC |
+ NETDEV_XDP_ACT_REDIRECT |
+ NETDEV_XDP_ACT_RX_SG;
+
+ if (priv_peer->_xdp_prog || veth_gro_requested(peer))
+ val |= NETDEV_XDP_ACT_NDO_XMIT |
+ NETDEV_XDP_ACT_NDO_XMIT_SG;
+ xdp_set_features_flag(dev, val);
+ } else {
+ xdp_clear_features_flag(dev);
+ }
+}
+
+static int veth_set_channels(struct net_device *dev,
+ struct ethtool_channels *ch)
+{
+ struct veth_priv *priv = netdev_priv(dev);
+ unsigned int old_rx_count, new_rx_count;
+ struct veth_priv *peer_priv;
+ struct net_device *peer;
+ int err;
+
+ /* sanity check. Upper bounds are already enforced by the caller */
+ if (!ch->rx_count || !ch->tx_count)
+ return -EINVAL;
+
+ /* avoid braking XDP, if that is enabled */
+ peer = rtnl_dereference(priv->peer);
+ peer_priv = peer ? netdev_priv(peer) : NULL;
+ if (priv->_xdp_prog && peer && ch->rx_count < peer->real_num_tx_queues)
+ return -EINVAL;
+
+ if (peer && peer_priv && peer_priv->_xdp_prog && ch->tx_count > peer->real_num_rx_queues)
+ return -EINVAL;
+
+ old_rx_count = dev->real_num_rx_queues;
+ new_rx_count = ch->rx_count;
+ if (netif_running(dev)) {
+ /* turn device off */
+ netif_carrier_off(dev);
+ if (peer)
+ netif_carrier_off(peer);
+
+ /* try to allocate new resources, as needed*/
+ err = veth_enable_range_safe(dev, old_rx_count, new_rx_count);
+ if (err)
+ goto out;
+ }
+
+ err = netif_set_real_num_rx_queues(dev, ch->rx_count);
+ if (err)
+ goto revert;
+
+ err = netif_set_real_num_tx_queues(dev, ch->tx_count);
+ if (err) {
+ int err2 = netif_set_real_num_rx_queues(dev, old_rx_count);
+
+ /* this error condition could happen only if rx and tx change
+ * in opposite directions (e.g. tx nr raises, rx nr decreases)
+ * and we can't do anything to fully restore the original
+ * status
+ */
+ if (err2)
+ pr_warn("Can't restore rx queues config %d -> %d %d",
+ new_rx_count, old_rx_count, err2);
+ else
+ goto revert;
+ }
+
+out:
+ if (netif_running(dev)) {
+ /* note that we need to swap the arguments WRT the enable part
+ * to identify the range we have to disable
+ */
+ veth_disable_range_safe(dev, new_rx_count, old_rx_count);
+ netif_carrier_on(dev);
+ if (peer)
+ netif_carrier_on(peer);
+ }
+
+ /* update XDP supported features */
+ veth_set_xdp_features(dev);
+ if (peer)
+ veth_set_xdp_features(peer);
+
+ return err;
+
+revert:
+ new_rx_count = old_rx_count;
+ old_rx_count = ch->rx_count;
+ goto out;
}
static int veth_open(struct net_device *dev)
@@ -980,6 +1387,10 @@ static int veth_open(struct net_device *dev)
err = veth_enable_xdp(dev);
if (err)
return err;
+ } else if (veth_gro_requested(dev)) {
+ err = veth_napi_enable(dev);
+ if (err)
+ return err;
}
if (peer->flags & IFF_UP) {
@@ -987,6 +1398,8 @@ static int veth_open(struct net_device *dev)
netif_carrier_on(peer);
}
+ veth_set_xdp_features(dev);
+
return 0;
}
@@ -1001,6 +1414,8 @@ static int veth_close(struct net_device *dev)
if (priv->_xdp_prog)
veth_disable_xdp(dev);
+ else if (veth_gro_requested(dev))
+ veth_napi_del(dev);
return 0;
}
@@ -1015,7 +1430,8 @@ static int veth_alloc_queues(struct net_device *dev)
struct veth_priv *priv = netdev_priv(dev);
int i;
- priv->rq = kcalloc(dev->num_rx_queues, sizeof(*priv->rq), GFP_KERNEL);
+ priv->rq = kvcalloc(dev->num_rx_queues, sizeof(*priv->rq),
+ GFP_KERNEL_ACCOUNT | __GFP_RETRY_MAYFAIL);
if (!priv->rq)
return -ENOMEM;
@@ -1031,30 +1447,18 @@ static void veth_free_queues(struct net_device *dev)
{
struct veth_priv *priv = netdev_priv(dev);
- kfree(priv->rq);
+ kvfree(priv->rq);
}
static int veth_dev_init(struct net_device *dev)
{
- int err;
-
- dev->lstats = netdev_alloc_pcpu_stats(struct pcpu_lstats);
- if (!dev->lstats)
- return -ENOMEM;
-
- err = veth_alloc_queues(dev);
- if (err) {
- free_percpu(dev->lstats);
- return err;
- }
-
- return 0;
+ netdev_lockdep_set_classes(dev);
+ return veth_alloc_queues(dev);
}
static void veth_dev_free(struct net_device *dev)
{
veth_free_queues(dev);
- free_percpu(dev->lstats);
}
#ifdef CONFIG_NET_POLL_CONTROLLER
@@ -1080,7 +1484,7 @@ static int veth_get_iflink(const struct net_device *dev)
rcu_read_lock();
peer = rcu_dereference(priv->peer);
- iflink = peer ? peer->ifindex : 0;
+ iflink = peer ? READ_ONCE(peer->ifindex) : 0;
rcu_read_unlock();
return iflink;
@@ -1103,6 +1507,33 @@ static netdev_features_t veth_fix_features(struct net_device *dev,
return features;
}
+static int veth_set_features(struct net_device *dev,
+ netdev_features_t features)
+{
+ netdev_features_t changed = features ^ dev->features;
+ struct veth_priv *priv = netdev_priv(dev);
+ struct net_device *peer;
+ int err;
+
+ if (!(changed & NETIF_F_GRO) || !(dev->flags & IFF_UP) || priv->_xdp_prog)
+ return 0;
+
+ peer = rtnl_dereference(priv->peer);
+ if (features & NETIF_F_GRO) {
+ err = veth_napi_enable(dev);
+ if (err)
+ return err;
+
+ if (peer)
+ xdp_features_set_redirect_target(peer, true);
+ } else {
+ if (peer)
+ xdp_features_clear_redirect_target(peer);
+ veth_napi_del(dev);
+ }
+ return 0;
+}
+
static void veth_set_rx_headroom(struct net_device *dev, int new_hr)
{
struct veth_priv *peer_priv, *priv = netdev_priv(dev);
@@ -1146,9 +1577,14 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog,
goto err;
}
- max_mtu = PAGE_SIZE - VETH_XDP_HEADROOM -
- peer->hard_header_len -
- SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
+ max_mtu = SKB_WITH_OVERHEAD(PAGE_SIZE - VETH_XDP_HEADROOM) -
+ peer->hard_header_len;
+ /* Allow increasing the max_mtu if the program supports
+ * XDP fragments.
+ */
+ if (prog->aux->xdp_has_frags)
+ max_mtu += PAGE_SIZE * MAX_SKB_FRAGS;
+
if (peer->mtu > max_mtu) {
NL_SET_ERR_MSG_MOD(extack, "Peer MTU is too large to set XDP");
err = -ERANGE;
@@ -1173,10 +1609,15 @@ static int veth_xdp_set(struct net_device *dev, struct bpf_prog *prog,
peer->hw_features &= ~NETIF_F_GSO_SOFTWARE;
peer->max_mtu = max_mtu;
}
+
+ xdp_features_set_redirect_target(peer, true);
}
if (old_prog) {
if (!prog) {
+ if (peer && !veth_gro_requested(dev))
+ xdp_features_clear_redirect_target(peer);
+
if (dev->flags & IFF_UP)
veth_disable_xdp(dev);
@@ -1198,31 +1639,60 @@ err:
return err;
}
-static u32 veth_xdp_query(struct net_device *dev)
-{
- struct veth_priv *priv = netdev_priv(dev);
- const struct bpf_prog *xdp_prog;
-
- xdp_prog = priv->_xdp_prog;
- if (xdp_prog)
- return xdp_prog->aux->id;
-
- return 0;
-}
-
static int veth_xdp(struct net_device *dev, struct netdev_bpf *xdp)
{
switch (xdp->command) {
case XDP_SETUP_PROG:
return veth_xdp_set(dev, xdp->prog, xdp->extack);
- case XDP_QUERY_PROG:
- xdp->prog_id = veth_xdp_query(dev);
- return 0;
default:
return -EINVAL;
}
}
+static int veth_xdp_rx_timestamp(const struct xdp_md *ctx, u64 *timestamp)
+{
+ struct veth_xdp_buff *_ctx = (void *)ctx;
+
+ if (!_ctx->skb)
+ return -ENODATA;
+
+ *timestamp = skb_hwtstamps(_ctx->skb)->hwtstamp;
+ return 0;
+}
+
+static int veth_xdp_rx_hash(const struct xdp_md *ctx, u32 *hash,
+ enum xdp_rss_hash_type *rss_type)
+{
+ struct veth_xdp_buff *_ctx = (void *)ctx;
+ struct sk_buff *skb = _ctx->skb;
+
+ if (!skb)
+ return -ENODATA;
+
+ *hash = skb_get_hash(skb);
+ *rss_type = skb->l4_hash ? XDP_RSS_TYPE_L4_ANY : XDP_RSS_TYPE_NONE;
+
+ return 0;
+}
+
+static int veth_xdp_rx_vlan_tag(const struct xdp_md *ctx, __be16 *vlan_proto,
+ u16 *vlan_tci)
+{
+ const struct veth_xdp_buff *_ctx = (void *)ctx;
+ const struct sk_buff *skb = _ctx->skb;
+ int err;
+
+ if (!skb)
+ return -ENODATA;
+
+ err = __vlan_hwaccel_get_tag(skb, vlan_tci);
+ if (err)
+ return err;
+
+ *vlan_proto = skb->vlan_proto;
+ return err;
+}
+
static const struct net_device_ops veth_netdev_ops = {
.ndo_init = veth_dev_init,
.ndo_open = veth_open,
@@ -1236,10 +1706,18 @@ static const struct net_device_ops veth_netdev_ops = {
#endif
.ndo_get_iflink = veth_get_iflink,
.ndo_fix_features = veth_fix_features,
+ .ndo_set_features = veth_set_features,
.ndo_features_check = passthru_features_check,
.ndo_set_rx_headroom = veth_set_rx_headroom,
.ndo_bpf = veth_xdp,
.ndo_xdp_xmit = veth_ndo_xdp_xmit,
+ .ndo_get_peer_dev = veth_peer_dev,
+};
+
+static const struct xdp_metadata_ops veth_xdp_metadata_ops = {
+ .xmo_rx_timestamp = veth_xdp_rx_timestamp,
+ .xmo_rx_hash = veth_xdp_rx_hash,
+ .xmo_rx_vlan_tag = veth_xdp_rx_vlan_tag,
};
#define VETH_FEATURES (NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HW_CSUM | \
@@ -1256,10 +1734,12 @@ static void veth_setup(struct net_device *dev)
dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
dev->priv_flags |= IFF_NO_QUEUE;
dev->priv_flags |= IFF_PHONY_HEADROOM;
+ dev->priv_flags |= IFF_DISABLE_NETPOLL;
+ dev->lltx = true;
dev->netdev_ops = &veth_netdev_ops;
+ dev->xdp_metadata_ops = &veth_xdp_metadata_ops;
dev->ethtool_ops = &veth_ethtool_ops;
- dev->features |= NETIF_F_LLTX;
dev->features |= VETH_FEATURES;
dev->vlan_features = dev->features &
~(NETIF_F_HW_VLAN_CTAG_TX |
@@ -1268,11 +1748,13 @@ static void veth_setup(struct net_device *dev)
NETIF_F_HW_VLAN_STAG_RX);
dev->needs_free_netdev = true;
dev->priv_destructor = veth_dev_free;
+ dev->pcpu_stat_type = NETDEV_PCPU_STAT_TSTATS;
dev->max_mtu = ETH_MAX_MTU;
dev->hw_features = VETH_FEATURES;
dev->hw_enc_features = VETH_FEATURES;
dev->mpls_features = NETIF_F_HW_CSUM | NETIF_F_GSO_SOFTWARE;
+ netif_set_tso_max_size(dev, GSO_MAX_SIZE);
}
/*
@@ -1297,10 +1779,37 @@ static int veth_validate(struct nlattr *tb[], struct nlattr *data[],
static struct rtnl_link_ops veth_link_ops;
-static int veth_newlink(struct net *src_net, struct net_device *dev,
- struct nlattr *tb[], struct nlattr *data[],
+static void veth_disable_gro(struct net_device *dev)
+{
+ dev->features &= ~NETIF_F_GRO;
+ dev->wanted_features &= ~NETIF_F_GRO;
+ netdev_update_features(dev);
+}
+
+static int veth_init_queues(struct net_device *dev, struct nlattr *tb[])
+{
+ int err;
+
+ if (!tb[IFLA_NUM_TX_QUEUES] && dev->num_tx_queues > 1) {
+ err = netif_set_real_num_tx_queues(dev, 1);
+ if (err)
+ return err;
+ }
+ if (!tb[IFLA_NUM_RX_QUEUES] && dev->num_rx_queues > 1) {
+ err = netif_set_real_num_rx_queues(dev, 1);
+ if (err)
+ return err;
+ }
+ return 0;
+}
+
+static int veth_newlink(struct net_device *dev,
+ struct rtnl_newlink_params *params,
struct netlink_ext_ack *extack)
{
+ struct net *peer_net = rtnl_newlink_peer_net(params);
+ struct nlattr **data = params->data;
+ struct nlattr **tb = params->tb;
int err;
struct net_device *peer;
struct veth_priv *priv;
@@ -1308,27 +1817,15 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
struct nlattr *peer_tb[IFLA_MAX + 1], **tbp;
unsigned char name_assign_type;
struct ifinfomsg *ifmp;
- struct net *net;
/*
* create and register peer first
*/
- if (data != NULL && data[VETH_INFO_PEER] != NULL) {
- struct nlattr *nla_peer;
+ if (data && data[VETH_INFO_PEER]) {
+ struct nlattr *nla_peer = data[VETH_INFO_PEER];
- nla_peer = data[VETH_INFO_PEER];
ifmp = nla_data(nla_peer);
- err = rtnl_nla_parse_ifla(peer_tb,
- nla_data(nla_peer) + sizeof(struct ifinfomsg),
- nla_len(nla_peer) - sizeof(struct ifinfomsg),
- NULL);
- if (err < 0)
- return err;
-
- err = veth_validate(peer_tb, NULL, extack);
- if (err < 0)
- return err;
-
+ rtnl_nla_parse_ifinfomsg(peer_tb, nla_peer, extack);
tbp = peer_tb;
} else {
ifmp = NULL;
@@ -1336,23 +1833,17 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
}
if (ifmp && tbp[IFLA_IFNAME]) {
- nla_strlcpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
+ nla_strscpy(ifname, tbp[IFLA_IFNAME], IFNAMSIZ);
name_assign_type = NET_NAME_USER;
} else {
snprintf(ifname, IFNAMSIZ, DRV_NAME "%%d");
name_assign_type = NET_NAME_ENUM;
}
- net = rtnl_link_get_net(src_net, tbp);
- if (IS_ERR(net))
- return PTR_ERR(net);
-
- peer = rtnl_create_link(net, ifname, name_assign_type,
+ peer = rtnl_create_link(peer_net, ifname, name_assign_type,
&veth_link_ops, tbp, extack);
- if (IS_ERR(peer)) {
- put_net(net);
+ if (IS_ERR(peer))
return PTR_ERR(peer);
- }
if (!ifmp || !tbp[IFLA_ADDRESS])
eth_hw_addr_random(peer);
@@ -1360,18 +1851,19 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
if (ifmp && (dev->ifindex != 0))
peer->ifindex = ifmp->ifi_index;
- peer->gso_max_size = dev->gso_max_size;
- peer->gso_max_segs = dev->gso_max_segs;
+ netif_inherit_tso_max(peer, dev);
err = register_netdevice(peer);
- put_net(net);
- net = NULL;
if (err < 0)
goto err_register_peer;
+ /* keep GRO disabled by default to be consistent with the established
+ * veth behavior
+ */
+ veth_disable_gro(peer);
netif_carrier_off(peer);
- err = rtnl_configure_link(peer, ifmp);
+ err = rtnl_configure_link(peer, ifmp, 0, NULL);
if (err < 0)
goto err_configure_peer;
@@ -1386,7 +1878,7 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
eth_hw_addr_random(dev);
if (tb[IFLA_IFNAME])
- nla_strlcpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
+ nla_strscpy(dev->name, tb[IFLA_IFNAME], IFNAMSIZ);
else
snprintf(dev->name, IFNAMSIZ, DRV_NAME "%%d");
@@ -1402,12 +1894,25 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
priv = netdev_priv(dev);
rcu_assign_pointer(priv->peer, peer);
+ err = veth_init_queues(dev, tb);
+ if (err)
+ goto err_queues;
priv = netdev_priv(peer);
rcu_assign_pointer(priv->peer, dev);
+ err = veth_init_queues(peer, tb);
+ if (err)
+ goto err_queues;
+
+ veth_disable_gro(dev);
+ /* update XDP supported features */
+ veth_set_xdp_features(dev);
+ veth_set_xdp_features(peer);
return 0;
+err_queues:
+ unregister_netdevice(dev);
err_register_dev:
/* nothing to do */
err_configure_peer:
@@ -1453,6 +1958,16 @@ static struct net *veth_get_link_net(const struct net_device *dev)
return peer ? dev_net(peer) : dev_net(dev);
}
+static unsigned int veth_get_num_queues(void)
+{
+ /* enforce the same queue limit as rtnl_create_link */
+ int queues = num_possible_cpus();
+
+ if (queues > 4096)
+ queues = 4096;
+ return queues;
+}
+
static struct rtnl_link_ops veth_link_ops = {
.kind = DRV_NAME,
.priv_size = sizeof(struct veth_priv),
@@ -1461,8 +1976,11 @@ static struct rtnl_link_ops veth_link_ops = {
.newlink = veth_newlink,
.dellink = veth_dellink,
.policy = veth_policy,
+ .peer_type = VETH_INFO_PEER,
.maxtype = VETH_INFO_MAX,
.get_link_net = veth_get_link_net,
+ .get_num_tx_queues = veth_get_num_queues,
+ .get_num_rx_queues = veth_get_num_queues,
};
/*
generated by cgit (git 2.34.1) at 2025-12-25 14:33:30 +0000