iv/linux
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

net: minor optimization in __alloc_skb()

Browse Source 
TCP allocates 'fast clones' skbs for packets in tx queues.

Currently, __alloc_skb() initializes the companion fclone
field to SKB_FCLONE_CLONE, and leaves other fields untouched.

It makes sense to defer this init much later in skb_clone(),
because all fclone fields are copied and hot in cpu caches
at that time.

This removes one cache line miss in __alloc_skb(), cost seen
on an host with 256 cpus all competing on memory accesses.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This commit is contained in:
Eric Dumazet 2022-07-07 19:18:46 +00:00 committed by David S. Miller
parent 8e15145792
commit c2dd4059dc
1 changed files with 1 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
net/core/skbuff.c
View File

@ -454,8 +454,6 @@ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
skb->fclone = SKB_FCLONE_ORIG; skb->fclone = SKB_FCLONE_ORIG;
refcount_set(&fclones->fclone_ref, 1); refcount_set(&fclones->fclone_ref, 1);
fclones->skb2.fclone = SKB_FCLONE_CLONE;
} }
return skb; return skb;
@ -1513,6 +1511,7 @@ struct sk_buff *skb_clone(struct sk_buff *skb, gfp_t gfp_mask)
refcount_read(&fclones->fclone_ref) == 1) { refcount_read(&fclones->fclone_ref) == 1) {
n = &fclones->skb2; n = &fclones->skb2;
refcount_set(&fclones->fclone_ref, 2); refcount_set(&fclones->fclone_ref, 2);
n->fclone = SKB_FCLONE_CLONE;
} else { } else {
if (skb_pfmemalloc(skb)) if (skb_pfmemalloc(skb))
gfp_mask |= __GFP_MEMALLOC; gfp_mask |= __GFP_MEMALLOC;