Mailing List Archive

[PATCH v2 net-next v2 3/5] net: add code for TCP fraglist GRO
This implements fraglist GRO similar to how it's handled in UDP, however
no functional changes are added yet. The next change adds a heuristic for
using fraglist GRO instead of regular GRO.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
net/ipv4/tcp_offload.c | 22 ++++++++++++++++++++++
net/ipv6/tcpv6_offload.c | 9 +++++++++
2 files changed, 31 insertions(+)

diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
index e455f884190c..68157130c264 100644
--- a/net/ipv4/tcp_offload.c
+++ b/net/ipv4/tcp_offload.c
@@ -336,6 +336,19 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb)
flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq);
flush |= skb_cmp_decrypted(p, skb);

+ if (NAPI_GRO_CB(p)->is_flist) {
+ flush |= (__force int)(flags ^ tcp_flag_word(th2));
+ flush |= skb->ip_summed != p->ip_summed;
+ flush |= skb->csum_level != p->csum_level;
+ flush |= !pskb_may_pull(skb, skb_gro_offset(skb));
+ flush |= NAPI_GRO_CB(p)->count >= 64;
+
+ if (flush || skb_gro_receive_list(p, skb))
+ mss = 1;
+
+ goto out_check_final;
+ }
+
if (flush || skb_gro_receive(p, skb)) {
mss = 1;
goto out_check_final;
@@ -402,6 +415,15 @@ INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff)
const struct iphdr *iph = ip_hdr(skb);
struct tcphdr *th = tcp_hdr(skb);

+ if (NAPI_GRO_CB(skb)->is_flist) {
+ skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV4;
+ skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+
+ __skb_incr_checksum_unnecessary(skb);
+
+ return 0;
+ }
+
th->check = ~tcp_v4_check(skb->len - thoff, iph->saddr,
iph->daddr, 0);

diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c
index b3b8e1f6b92a..c97d55cf036f 100644
--- a/net/ipv6/tcpv6_offload.c
+++ b/net/ipv6/tcpv6_offload.c
@@ -32,6 +32,15 @@ INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff)
const struct ipv6hdr *iph = ipv6_hdr(skb);
struct tcphdr *th = tcp_hdr(skb);

+ if (NAPI_GRO_CB(skb)->is_flist) {
+ skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV6;
+ skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
+
+ __skb_incr_checksum_unnecessary(skb);
+
+ return 0;
+ }
+
th->check = ~tcp_v6_check(skb->len - thoff, &iph->saddr,
&iph->daddr, 0);
skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV6;
--
2.44.0
Re: [PATCH v2 net-next v2 3/5] net: add code for TCP fraglist GRO [ In reply to ]
Felix Fietkau wrote:
> This implements fraglist GRO similar to how it's handled in UDP, however
> no functional changes are added yet. The next change adds a heuristic for
> using fraglist GRO instead of regular GRO.
>
> Signed-off-by: Felix Fietkau <nbd@nbd.name>
> ---
> net/ipv4/tcp_offload.c | 22 ++++++++++++++++++++++
> net/ipv6/tcpv6_offload.c | 9 +++++++++
> 2 files changed, 31 insertions(+)
>
> diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
> index e455f884190c..68157130c264 100644
> --- a/net/ipv4/tcp_offload.c
> +++ b/net/ipv4/tcp_offload.c
> @@ -336,6 +336,19 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb)
> flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq);
> flush |= skb_cmp_decrypted(p, skb);
>
> + if (NAPI_GRO_CB(p)->is_flist) {
> + flush |= (__force int)(flags ^ tcp_flag_word(th2));

What is the purpose of this check, given the existing check above

flush |= (__force int)((flags ^ tcp_flag_word(th2)) &
~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH));

> + flush |= skb->ip_summed != p->ip_summed;
> + flush |= skb->csum_level != p->csum_level;
> + flush |= !pskb_may_pull(skb, skb_gro_offset(skb));
> + flush |= NAPI_GRO_CB(p)->count >= 64;
> +
> + if (flush || skb_gro_receive_list(p, skb))
> + mss = 1;
> +
> + goto out_check_final;
> + }
> +
> if (flush || skb_gro_receive(p, skb)) {
> mss = 1;
> goto out_check_final;
> @@ -402,6 +415,15 @@ INDIRECT_CALLABLE_SCOPE int tcp4_gro_complete(struct sk_buff *skb, int thoff)
> const struct iphdr *iph = ip_hdr(skb);
> struct tcphdr *th = tcp_hdr(skb);
>
> + if (NAPI_GRO_CB(skb)->is_flist) {
> + skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV4;
> + skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
> +
> + __skb_incr_checksum_unnecessary(skb);
> +
> + return 0;
> + }
> +
> th->check = ~tcp_v4_check(skb->len - thoff, iph->saddr,
> iph->daddr, 0);
>
> diff --git a/net/ipv6/tcpv6_offload.c b/net/ipv6/tcpv6_offload.c
> index b3b8e1f6b92a..c97d55cf036f 100644
> --- a/net/ipv6/tcpv6_offload.c
> +++ b/net/ipv6/tcpv6_offload.c
> @@ -32,6 +32,15 @@ INDIRECT_CALLABLE_SCOPE int tcp6_gro_complete(struct sk_buff *skb, int thoff)
> const struct ipv6hdr *iph = ipv6_hdr(skb);
> struct tcphdr *th = tcp_hdr(skb);
>
> + if (NAPI_GRO_CB(skb)->is_flist) {
> + skb_shinfo(skb)->gso_type |= SKB_GSO_FRAGLIST | SKB_GSO_TCPV6;
> + skb_shinfo(skb)->gso_segs = NAPI_GRO_CB(skb)->count;
> +
> + __skb_incr_checksum_unnecessary(skb);
> +
> + return 0;
> + }
> +
> th->check = ~tcp_v6_check(skb->len - thoff, &iph->saddr,
> &iph->daddr, 0);
> skb_shinfo(skb)->gso_type |= SKB_GSO_TCPV6;
> --
> 2.44.0
>
Re: [PATCH v2 net-next v2 3/5] net: add code for TCP fraglist GRO [ In reply to ]
On 25.04.24 17:18, Willem de Bruijn wrote:
> Felix Fietkau wrote:
>> This implements fraglist GRO similar to how it's handled in UDP, however
>> no functional changes are added yet. The next change adds a heuristic for
>> using fraglist GRO instead of regular GRO.
>>
>> Signed-off-by: Felix Fietkau <nbd@nbd.name>
>> ---
>> net/ipv4/tcp_offload.c | 22 ++++++++++++++++++++++
>> net/ipv6/tcpv6_offload.c | 9 +++++++++
>> 2 files changed, 31 insertions(+)
>>
>> diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c
>> index e455f884190c..68157130c264 100644
>> --- a/net/ipv4/tcp_offload.c
>> +++ b/net/ipv4/tcp_offload.c
>> @@ -336,6 +336,19 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb)
>> flush |= (ntohl(th2->seq) + skb_gro_len(p)) ^ ntohl(th->seq);
>> flush |= skb_cmp_decrypted(p, skb);
>>
>> + if (NAPI_GRO_CB(p)->is_flist) {
>> + flush |= (__force int)(flags ^ tcp_flag_word(th2));
>
> What is the purpose of this check, given the existing check above
>
> flush |= (__force int)((flags ^ tcp_flag_word(th2)) &
> ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH));

The fraglist codepath is not updating the TCP flags of the first packet,
so the extra check ensures it doesn't have to.

- Felix