tcp: do not set queue_mapping on SYNACK
At the time of commit fff32699 ("tcp: reflect SYN queue_mapping into SYNACK packets") we had little ways to cope with SYN floods. We no longer need to reflect incoming skb queue mappings, and instead can pick a TX queue based on cpu cooking the SYNACK, with normal XPS affinities. Note that all SYNACK retransmits were picking TX queue 0, this no longer is a win given that SYNACK rtx are now distributed on all cpus. Signed-off-by:Eric Dumazet <edumazet@google.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
Showing
- include/net/tcp.h 1 addition, 1 deletioninclude/net/tcp.h
- net/ipv4/ip_output.c 0 additions, 1 deletionnet/ipv4/ip_output.c
- net/ipv4/tcp_input.c 2 additions, 2 deletionsnet/ipv4/tcp_input.c
- net/ipv4/tcp_ipv4.c 0 additions, 2 deletionsnet/ipv4/tcp_ipv4.c
- net/ipv4/tcp_output.c 1 addition, 1 deletionnet/ipv4/tcp_output.c
- net/ipv6/tcp_ipv6.c 0 additions, 2 deletionsnet/ipv6/tcp_ipv6.c
Loading
Please register or sign in to comment