Merge branch 'ndo_xdp_xmit-cleanup'

Jesper Dangaard Brouer says:

====================
As I mentioned in merge commit 10f678683e4 ("Merge branch 'xdp_xmit-bulking'")
I plan to change the API for ndo_xdp_xmit once more, by adding a flags
argument, which is done in this patchset.

I know it is late in the cycle (currently at rc7), but it would be
nice to avoid changing NDOs over several kernel releases, as it is
annoying to vendors and distro backporters, but it is not strictly
UAPI so it is allowed (according to Alexei).

The end-goal is getting rid of the ndo_xdp_flush operation, as it will
make it possible for drivers to implement a TXQ synchronization mechanism
that is not necessarily derived from the CPU id (smp_processor_id).

This patchset removes all callers of the ndo_xdp_flush operation, but
it doesn't take the last step of removing it from all drivers.  This
can be done later, or I can update the patchset on request.

Micro-benchmarks only show a very small performance improvement, for
map-redirect around ~2 ns, and for non-map redirect ~7 ns.  I've not
benchmarked this with CONFIG_RETPOLINE, but the performance benefit
should be more visible given we end-up removing an indirect call.

---
V2: Updated based on feedback from Song Liu <songliubraving@fb.com>
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

9 files changed, 72 insertions, 35 deletions

diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.c b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
index 9b698c5acd05..5f01e4ce9c92 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.c
@@ -3670,11 +3670,13 @@ netdev_tx_t i40e_lan_xmit_frame(struct sk_buff *skb, struct net_device *netdev)
  * For error cases, a negative errno code is returned and no-frames
  * are transmitted (caller must handle freeing frames).
  **/
-int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames)
+int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+                  u32 flags)
 {
         struct i40e_netdev_priv *np = netdev_priv(dev);
         unsigned int queue_index = smp_processor_id();
         struct i40e_vsi *vsi = np->vsi;
+        struct i40e_ring *xdp_ring;
         int drops = 0;
         int i;
 
@@ -3684,17 +3686,25 @@ int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames)
         if (!i40e_enabled_xdp_vsi(vsi) || queue_index >= vsi->num_queue_pairs)
                 return -ENXIO;
 
+        if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+                return -EINVAL;
+
+        xdp_ring = vsi->xdp_rings[queue_index];
+
         for (i = 0; i < n; i++) {
                 struct xdp_frame *xdpf = frames[i];
                 int err;
 
-                err = i40e_xmit_xdp_ring(xdpf, vsi->xdp_rings[queue_index]);
+                err = i40e_xmit_xdp_ring(xdpf, xdp_ring);
                 if (err != I40E_XDP_TX) {
                         xdp_return_frame_rx_napi(xdpf);
                         drops++;
                 }
         }
 
+        if (unlikely(flags & XDP_XMIT_FLUSH))
+                i40e_xdp_ring_update_tail(xdp_ring);
+
         return n - drops;
 }
 
diff --git a/drivers/net/ethernet/intel/i40e/i40e_txrx.h b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
index eb8804b3d7b6..820f76db251b 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_txrx.h
+++ b/drivers/net/ethernet/intel/i40e/i40e_txrx.h
@@ -487,7 +487,8 @@ u32 i40e_get_tx_pending(struct i40e_ring *ring, bool in_sw);
 void i40e_detect_recover_hung(struct i40e_vsi *vsi);
 int __i40e_maybe_stop_tx(struct i40e_ring *tx_ring, int size);
 bool __i40e_chk_linearize(struct sk_buff *skb);
-int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames);
+int i40e_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames,
+                  u32 flags);
 void i40e_xdp_flush(struct net_device *dev);
 
 /**
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
index 031d65c4178d..4fd77c9067f2 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_main.c
@@ -10022,8 +10022,17 @@ static int ixgbe_xdp(struct net_device *dev, struct netdev_bpf *xdp)
         }
 }
 
+static void ixgbe_xdp_ring_update_tail(struct ixgbe_ring *ring)
+{
+        /* Force memory writes to complete before letting h/w know there
+         * are new descriptors to fetch.
+         */
+        wmb();
+        writel(ring->next_to_use, ring->tail);
+}
+
 static int ixgbe_xdp_xmit(struct net_device *dev, int n,
-                          struct xdp_frame **frames)
+                          struct xdp_frame **frames, u32 flags)
 {
         struct ixgbe_adapter *adapter = netdev_priv(dev);
         struct ixgbe_ring *ring;
@@ -10033,6 +10042,9 @@ static int ixgbe_xdp_xmit(struct net_device *dev, int n,
         if (unlikely(test_bit(__IXGBE_DOWN, &adapter->state)))
                 return -ENETDOWN;
 
+        if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+                return -EINVAL;
+
         /* During program transitions its possible adapter->xdp_prog is assigned
          * but ring has not been configured yet. In this case simply abort xmit.
          */
@@ -10051,6 +10063,9 @@ static int ixgbe_xdp_xmit(struct net_device *dev, int n,
                 }
         }
 
+        if (unlikely(flags & XDP_XMIT_FLUSH))
+                ixgbe_xdp_ring_update_tail(ring);
+
         return n - drops;
 }
 
@@ -10069,11 +10084,7 @@ static void ixgbe_xdp_flush(struct net_device *dev)
         if (unlikely(!ring))
                 return;
 
-        /* Force memory writes to complete before letting h/w know there
-         * are new descriptors to fetch.
-         */
-        wmb();
-        writel(ring->next_to_use, ring->tail);
+        ixgbe_xdp_ring_update_tail(ring);
 
         return;
 }
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 2265d2ccea47..d82a05fb0594 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -1285,7 +1285,16 @@ static const struct net_device_ops tun_netdev_ops = {
         .ndo_get_stats64        = tun_net_get_stats64,
 };
 
-static int tun_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames)
+static void __tun_xdp_flush_tfile(struct tun_file *tfile)
+{
+        /* Notify and wake up reader process */
+        if (tfile->flags & TUN_FASYNC)
+                kill_fasync(&tfile->fasync, SIGIO, POLL_IN);
+        tfile->socket.sk->sk_data_ready(tfile->socket.sk);
+}
+
+static int tun_xdp_xmit(struct net_device *dev, int n,
+                        struct xdp_frame **frames, u32 flags)
 {
         struct tun_struct *tun = netdev_priv(dev);
         struct tun_file *tfile;
@@ -1294,6 +1303,9 @@ static int tun_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames
         int cnt = n;
         int i;
 
+        if (unlikely(flags & ~XDP_XMIT_FLAGS_MASK))
+                return -EINVAL;
+
         rcu_read_lock();
 
         numqueues = READ_ONCE(tun->numqueues);
@@ -1321,6 +1333,9 @@ static int tun_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames
         }
         spin_unlock(&tfile->tx_ring.producer_lock);
 
+        if (flags & XDP_XMIT_FLUSH)
+                __tun_xdp_flush_tfile(tfile);
+
         rcu_read_unlock();
         return cnt - drops;