On Mar 15, 2012, at 8:30 PM, Ben Pfaff wrote:
> +#ifndef SO_RCVBUFFORCE
> +#define SO_RCVBUFFORCE 33
> +#endif
It seems like on some platforms, this is defined to be other values. For
example, this appears to be the definition on SPARC:
#define SO_RCVBUFFORCE 0x100b
> @@ -122,6 +126,13 @@ nl_sock_create(int protocol, struct nl_sock **sockp)
> sock->protocol = protocol;
> sock->dump = NULL;
>
> + rcvbuf = 1024 * 1024;
> + if (setsockopt(sock->fd, SOL_SOCKET, SO_RCVBUFFORCE,
> + &rcvbuf, sizeof rcvbuf)) {
> + VLOG_WARN_RL(&rl, "setting %d-byte socket receive buffer failed
> (%s)",
> + rcvbuf, strerror(errno));
> + }
It appears that this increase will happen for anything created with
nl_sock_create(), of which there are a few things now. Would it make sense to
make this an optional argument to nl_sock_create()? (Not necessarily right at
this moment, of course.)
Thanks for the quick turnaround on this!
--Justin
_______________________________________________
dev mailing list
[email protected]
http://openvswitch.org/mailman/listinfo/dev
