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, 2 deletions

diff --git a/drivers/net/ixgb/ixgb_main.c b/drivers/net/ixgb/ixgb_main.c
index 6eb7f37a113b..9c897cf86b9f 100644
--- a/drivers/net/ixgb/ixgb_main.c
+++ b/drivers/net/ixgb/ixgb_main.c
@@ -1300,7 +1300,7 @@ ixgb_tx_map(struct ixgb_adapter *adapter, struct sk_buff *skb,
                 buffer_info->length = size;
                 WARN_ON(buffer_info->dma != 0);
                 buffer_info->time_stamp = jiffies;
-                buffer_info->dma = map[0] + offset;
+                buffer_info->dma = skb_shinfo(skb)->dma_head + offset;
                         pci_map_single(adapter->pdev,
                                 skb->data + offset,
                                 size,
@@ -1340,7 +1340,7 @@ ixgb_tx_map(struct ixgb_adapter *adapter, struct sk_buff *skb,
 
                         buffer_info->length = size;
                         buffer_info->time_stamp = jiffies;
-                        buffer_info->dma = map[f + 1] + offset;
+                        buffer_info->dma = map[f] + offset;
                         buffer_info->next_to_watch = 0;
 
                         len -= size;