net: skb_shared_info optimization

skb_dma_unmap() is quite expensive for small packets,
because we use two different cache lines from skb_shared_info.

One to access nr_frags, one to access dma_maps[0]

Instead of dma_maps being an array of MAX_SKB_FRAGS + 1 elements,
let dma_head alone in a new dma_head field, close to nr_frags,
to reduce cache lines misses.

Tested on my dev machine (bnx2 & tg3 adapters), nice speedup !

Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 2 insertions, 3 deletions

diff --git a/drivers/net/igbvf/netdev.c b/drivers/net/igbvf/netdev.c
index 5f7ba1a4990b..22aadb7884fa 100644
--- a/drivers/net/igbvf/netdev.c
+++ b/drivers/net/igbvf/netdev.c
@@ -2119,8 +2119,7 @@ static inline int igbvf_tx_map_adv(struct igbvf_adapter *adapter,
         /* set time_stamp *before* dma to help avoid a possible race */
         buffer_info->time_stamp = jiffies;
         buffer_info->next_to_watch = i;
-        buffer_info->dma = map[count];
-        count++;
+        buffer_info->dma = skb_shinfo(skb)->dma_head;
 
         for (f = 0; f < skb_shinfo(skb)->nr_frags; f++) {
                 struct skb_frag_struct *frag;
@@ -2144,7 +2143,7 @@ static inline int igbvf_tx_map_adv(struct igbvf_adapter *adapter,
         tx_ring->buffer_info[i].skb = skb;
         tx_ring->buffer_info[first].next_to_watch = i;
 
-        return count;
+        return count + 1;
 }
 
 static inline void igbvf_tx_queue_adv(struct igbvf_adapter *adapter,