Filter TCP connections to SO_REUSEPORT_LB listen sockets by NUMA domain

In order to efficiently serve web traffic on a NUMA
machine, one must avoid as many NUMA domain crossings as
possible. With SO_REUSEPORT_LB, a number of workers can share a
listen socket. However, even if a worker sets affinity to a core
or set of cores on a NUMA domain, it will receive connections
associated with all NUMA domains in the system. This will lead to
cross-domain traffic when the server writes to the socket or
calls sendfile(), and memory is allocated on the server's local
NUMA node, but transmitted on the NUMA node associated with the
TCP connection. Similarly, when the server reads from the socket,
he will likely be reading memory allocated on the NUMA domain
associated with the TCP connection.

This change provides a new socket ioctl, TCP_REUSPORT_LB_NUMA. A
server can now tell the kernel to filter traffic so that only
incoming connections associated with the desired NUMA domain are
given to the server. (Of course, in the case where there are no
servers sharing the listen socket on some domain, then as a
fallback, traffic will be hashed as normal to all servers sharing
the listen socket regardless of domain). This allows a server to
deal only with traffic that is local to its NUMA domain, and
avoids cross-domain traffic in most cases.

This patch, and a corresponding small patch to nginx to use
TCP_REUSPORT_LB_NUMA allows us to serve 190Gb/s of kTLS encrypted
https media content from dual-socket Xeons with only 13% (as
measured by pcm.x) cross domain traffic on the memory controller.

Reviewed by:	jhb, bz (earlier version), bcr (man page)
Tested by: gonzo
Sponsored by:	Netfix
Differential Revision:	https://reviews.freebsd.org/D21636

1 files changed, 4 insertions, 0 deletions

diff --git a/newlib/libc/sys/rtems/include/netinet/tcp.h b/newlib/libc/sys/rtems/include/netinet/tcp.h
index faf142959..0b71bd465 100644
--- a/newlib/libc/sys/rtems/include/netinet/tcp.h
+++ b/newlib/libc/sys/rtems/include/netinet/tcp.h
@@ -196,6 +196,7 @@ struct tcphdr {
 #define	TCP_PCAP_IN	4096	/* number of input packets to keep */
 #define TCP_FUNCTION_BLK 8192	/* Set the tcp function pointers to the specified stack */
 /* Options for Rack and BBR */
+#define	TCP_REUSPORT_LB_NUMA   1026	/* set listen socket numa domain */
 #define TCP_RACK_MBUF_QUEUE   1050 /* Do we allow mbuf queuing if supported */
 #define TCP_RACK_PROP	      1051 /* RACK proportional rate reduction (bool) */
 #define TCP_RACK_TLP_REDUCE   1052 /* RACK TLP cwnd reduction (bool) */
@@ -406,4 +407,7 @@ struct tcp_function_set {
 #define	VOI_TCP_GPUT_ND		8 /* Goodput normalised delta */
 #define	VOI_TCP_ACKLEN		9 /* Average ACKed bytes per ACK */
 
+#define TCP_REUSPORT_LB_NUMA_NODOM	(-2) /* remove numa binding */
+#define TCP_REUSPORT_LB_NUMA_CURDOM	(-1) /* bind to current domain */
+
 #endif /* !_NETINET_TCP_H_ */