[v3,2/2] gro: optimise redundant parsing of packets
Commit Message
Currently the IPv6 extension headers are parsed twice: first in
ipv6_gro_receive, and then again in ipv6_gro_complete.
By using the new ->transport_proto field, and also storing the size of the
network header, we can avoid parsing extension headers a second time in
ipv6_gro_complete (which saves multiple memory dereferences and conditional
checks inside ipv6_exthdrs_len for a varying amount of extension headers in
IPv6 packets).
The implementation had to handle both inner and outer layers in case of
encapsulation (as they can't use the same field). I've applied a similar
optimisation to Ethernet.
Performance tests for TCP stream over IPv6 with a varying amount of
extension headers demonstrate throughput improvement of ~0.7%.
In addition, I fixed a potential future problem:
- The call to skb_set_inner_network_header at the beginning of
ipv6_gro_complete calculates inner_network_header based on skb->data by
calling skb_set_inner_network_header, and setting it to point to the
beginning of the ip header.
- If a packet is going to be handled by BIG TCP, the following code block
is going to shift the packet header, and skb->data is going to be
changed as well.
When the two flows are combined, inner_network_header will point to the
wrong place - which might happen if encapsulation of BIG TCP will be
supported in the future.
The fix is to place the whole encapsulation branch after the BIG TCP code
block. This way, if encapsulation of BIG TCP will be supported,
inner_network_header will still be calculated with the correct value of
skb->data.
Also, by arranging the code that way, the optimisation does not
add an additional branch.
Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
---
include/net/gro.h | 9 +++++++++
net/ethernet/eth.c | 14 +++++++++++---
net/ipv6/ip6_offload.c | 20 +++++++++++++++-----
3 files changed, 35 insertions(+), 8 deletions(-)
Comments
> >
> > Currently the IPv6 extension headers are parsed twice: first in
> > ipv6_gro_receive, and then again in ipv6_gro_complete.
> >
> > By using the new ->transport_proto field, and also storing the size of the
> > network header, we can avoid parsing extension headers a second time in
> > ipv6_gro_complete (which saves multiple memory dereferences and conditional
> > checks inside ipv6_exthdrs_len for a varying amount of extension headers in
> > IPv6 packets).
> >
> > The implementation had to handle both inner and outer layers in case of
> > encapsulation (as they can't use the same field). I've applied a similar
> > optimisation to Ethernet.
> >
> > Performance tests for TCP stream over IPv6 with a varying amount of
> > extension headers demonstrate throughput improvement of ~0.7%.
> >
> > In addition, I fixed a potential future problem:
>
> I would remove all this block.
>
> We fix current problems, not future hypothetical ones.
>
I agree, I did it primarily to avoid an additional branch (the logic
remains exactly the same). I'll remove this part from the commit message.
> > - The call to skb_set_inner_network_header at the beginning of
> > ipv6_gro_complete calculates inner_network_header based on skb->data by
> > calling skb_set_inner_network_header, and setting it to point to the
> > beginning of the ip header.
> > - If a packet is going to be handled by BIG TCP, the following code block
> > is going to shift the packet header, and skb->data is going to be
> > changed as well.
> >
> > When the two flows are combined, inner_network_header will point to the
> > wrong place - which might happen if encapsulation of BIG TCP will be
> > supported in the future.
> >
> > The fix is to place the whole encapsulation branch after the BIG TCP code
> > block. This way, if encapsulation of BIG TCP will be supported,
> > inner_network_header will still be calculated with the correct value of
> > skb->data.
>
> We do not support encapsulated BIG TCP yet.
> We will do this later, and whoever does it will make sure to also support GRO.
>
> > Also, by arranging the code that way, the optimisation does not
> > add an additional branch.
> >
> > Signed-off-by: Richard Gobert <richardbgobert@gmail.com>
> > ---
> >
>
> Can you give us a good explanation of why extension headers are used exactly ?
>
> I am not sure we want to add code to GRO for something that 99.99% of
> us do not use.
IMO, some common use cases that will benefit from this patch are:
- Parsing of BIG TCP packets which include a hbh ext hdr.
- dstopts and routing ext hdrs that are used for Mobile IPv6 features.
Generally, when a packet includes ext hdrs we will avoid the recalculation
of the ext hdrs len. When there are no ext hdrs, we will not call the
ipv6_exthdrs_len function so the performance isn't negatively impacted
(potentially even saving some opcodes in ipv6_exthdrs_len).
@@ -86,6 +86,15 @@ struct napi_gro_cb {
/* used to support CHECKSUM_COMPLETE for tunneling protocols */
__wsum csum;
+
+ /* Used in ipv6_gro_receive() */
+ u16 network_len;
+
+ /* Used in eth_gro_receive() */
+ __be16 network_proto;
+
+ /* Used in ipv6_gro_receive() */
+ u8 transport_proto;
};
#define NAPI_GRO_CB(skb) ((struct napi_gro_cb *)(skb)->cb)
@@ -439,6 +439,9 @@ struct sk_buff *eth_gro_receive(struct list_head *head, struct sk_buff *skb)
goto out;
}
+ if (!NAPI_GRO_CB(skb)->encap_mark)
+ NAPI_GRO_CB(skb)->network_proto = type;
+
skb_gro_pull(skb, sizeof(*eh));
skb_gro_postpull_rcsum(skb, eh, sizeof(*eh));
@@ -455,13 +458,18 @@ EXPORT_SYMBOL(eth_gro_receive);
int eth_gro_complete(struct sk_buff *skb, int nhoff)
{
- struct ethhdr *eh = (struct ethhdr *)(skb->data + nhoff);
- __be16 type = eh->h_proto;
struct packet_offload *ptype;
+ struct ethhdr *eh;
int err = -ENOSYS;
+ __be16 type;
- if (skb->encapsulation)
+ if (skb->encapsulation) {
+ eh = (struct ethhdr *)(skb->data + nhoff);
skb_set_inner_mac_header(skb, nhoff);
+ type = eh->h_proto;
+ } else {
+ type = NAPI_GRO_CB(skb)->network_proto;
+ }
ptype = gro_find_complete_by_type(type);
if (ptype != NULL)
@@ -232,6 +232,11 @@ INDIRECT_CALLABLE_SCOPE struct sk_buff *ipv6_gro_receive(struct list_head *head,
flush--;
nlen = skb_network_header_len(skb);
+ if (!NAPI_GRO_CB(skb)->encap_mark) {
+ NAPI_GRO_CB(skb)->transport_proto = proto;
+ NAPI_GRO_CB(skb)->network_len = nlen;
+ }
+
list_for_each_entry(p, head, list) {
const struct ipv6hdr *iph2;
__be32 first_word; /* <Version:4><Traffic_Class:8><Flow_Label:20> */
@@ -324,10 +329,6 @@ INDIRECT_CALLABLE_SCOPE int ipv6_gro_complete(struct sk_buff *skb, int nhoff)
int err = -ENOSYS;
u32 payload_len;
- if (skb->encapsulation) {
- skb_set_inner_protocol(skb, cpu_to_be16(ETH_P_IPV6));
- skb_set_inner_network_header(skb, nhoff);
- }
payload_len = skb->len - nhoff - sizeof(*iph);
if (unlikely(payload_len > IPV6_MAXPLEN)) {
@@ -341,6 +342,7 @@ INDIRECT_CALLABLE_SCOPE int ipv6_gro_complete(struct sk_buff *skb, int nhoff)
skb->len += hoplen;
skb->mac_header -= hoplen;
skb->network_header -= hoplen;
+ NAPI_GRO_CB(skb)->network_len += hoplen;
iph = (struct ipv6hdr *)(skb->data + nhoff);
hop_jumbo = (struct hop_jumbo_hdr *)(iph + 1);
@@ -358,7 +360,15 @@ INDIRECT_CALLABLE_SCOPE int ipv6_gro_complete(struct sk_buff *skb, int nhoff)
iph->payload_len = htons(payload_len);
}
- nhoff += sizeof(*iph) + ipv6_exthdrs_len(iph, &ops);
+ if (skb->encapsulation) {
+ skb_set_inner_protocol(skb, cpu_to_be16(ETH_P_IPV6));
+ skb_set_inner_network_header(skb, nhoff);
+ nhoff += sizeof(*iph) + ipv6_exthdrs_len(iph, &ops);
+ } else {
+ ops = rcu_dereference(inet6_offloads[NAPI_GRO_CB(skb)->transport_proto]);
+ nhoff += NAPI_GRO_CB(skb)->network_len;
+ }
+
if (WARN_ON(!ops || !ops->callbacks.gro_complete))
goto out;