| Message ID | 20250305223852.85839-10-chia-yu.chang@nokia-bell-labs.com (mailing list archive) |
|---|---|
| State | Needs ACK |
| Delegated to: | Netdev Maintainers |
| Headers | show |
| Series | AccECN protocol preparation patch series | expand |
On 3/5/25 11:38 PM, chia-yu.chang@nokia-bell-labs.com wrote: > From: Ilpo Järvinen <ij@kernel.org> > > There are important differences in how the CWR field behaves > in RFC3168 and AccECN. With AccECN, CWR flag is part of the > ACE counter and its changes are important so adjust the flags > changed mask accordingly. > > Also, if CWR is there, set the Accurate ECN GSO flag to avoid > corrupting CWR flag somewhere. > > Signed-off-by: Ilpo Järvinen <ij@kernel.org> > Signed-off-by: Chia-Yu Chang <chia-yu.chang@nokia-bell-labs.com> > --- > net/ipv4/tcp_offload.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c > index a4cea85288ff..ef12aee5deb4 100644 > --- a/net/ipv4/tcp_offload.c > +++ b/net/ipv4/tcp_offload.c > @@ -329,7 +329,7 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, > th2 = tcp_hdr(p); > flush = (__force int)(flags & TCP_FLAG_CWR); > flush |= (__force int)((flags ^ tcp_flag_word(th2)) & > - ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH)); > + ~(TCP_FLAG_FIN | TCP_FLAG_PSH)); > flush |= (__force int)(th->ack_seq ^ th2->ack_seq); > for (i = sizeof(*th); i < thlen; i += 4) > flush |= *(u32 *)((u8 *)th + i) ^ > @@ -405,7 +405,7 @@ void tcp_gro_complete(struct sk_buff *skb) > shinfo->gso_segs = NAPI_GRO_CB(skb)->count; > > if (th->cwr) > - shinfo->gso_type |= SKB_GSO_TCP_ECN; > + shinfo->gso_type |= SKB_GSO_TCP_ACCECN; > } > EXPORT_SYMBOL(tcp_gro_complete); To recap: when an host implementing the above will receive a GSO_TCP_ECN train transmitted by a RFC3168 endpoint, it will re-assemble it in 2 packets: a GSO one with !th->cwr and a non GSO one with th->cwr set. When receiving a GSO train with constant CWR set on all the wire packets, it will assemble it in a single SKB_GSO_TCP_ACCECN packet. I think should work correctly. Side note: the SKB_GSO_TCP_ACCECN flag is required: NETIF_F_TSO_ECN enabled driver will likely unconditionally apply RFC3168-like TSO to any GSO packet carrying the CWR flag, regardless of the skb gso_type. @Eric: are you ok with this change? Thanks, Paolo
diff --git a/net/ipv4/tcp_offload.c b/net/ipv4/tcp_offload.c index a4cea85288ff..ef12aee5deb4 100644 --- a/net/ipv4/tcp_offload.c +++ b/net/ipv4/tcp_offload.c @@ -329,7 +329,7 @@ struct sk_buff *tcp_gro_receive(struct list_head *head, struct sk_buff *skb, th2 = tcp_hdr(p); flush = (__force int)(flags & TCP_FLAG_CWR); flush |= (__force int)((flags ^ tcp_flag_word(th2)) & - ~(TCP_FLAG_CWR | TCP_FLAG_FIN | TCP_FLAG_PSH)); + ~(TCP_FLAG_FIN | TCP_FLAG_PSH)); flush |= (__force int)(th->ack_seq ^ th2->ack_seq); for (i = sizeof(*th); i < thlen; i += 4) flush |= *(u32 *)((u8 *)th + i) ^ @@ -405,7 +405,7 @@ void tcp_gro_complete(struct sk_buff *skb) shinfo->gso_segs = NAPI_GRO_CB(skb)->count; if (th->cwr) - shinfo->gso_type |= SKB_GSO_TCP_ECN; + shinfo->gso_type |= SKB_GSO_TCP_ACCECN; } EXPORT_SYMBOL(tcp_gro_complete);