1 This target is an extension of the QUEUE target. As opposed to QUEUE, it allows 2 you to put a packet into any specific queue, identified by its 16-bit queue 3 number. 4 It can only be used with Kernel versions 2.6.14 or later, since it requires 5 the 6 .B 7 nfnetlink_queue 8 kernel support. The \fBqueue-balance\fP option was added in Linux 2.6.31, 9 \fBqueue-bypass\fP in 2.6.39. 10 .TP 11 \fB\-\-queue\-num\fP \fIvalue\fP 12 This specifies the QUEUE number to use. Valid queue numbers are 0 to 65535. The default value is 0. 13 .PP 14 .TP 15 \fB\-\-queue\-balance\fP \fIvalue\fP\fB:\fP\fIvalue\fP 16 This specifies a range of queues to use. Packets are then balanced across the given queues. 17 This is useful for multicore systems: start multiple instances of the userspace program on 18 queues x, x+1, .. x+n and use "\-\-queue\-balance \fIx\fP\fB:\fP\fIx+n\fP". 19 Packets belonging to the same connection are put into the same nfqueue. 20 .PP 21 .TP 22 \fB\-\-queue\-bypass\fP 23 By default, if no userspace program is listening on an NFQUEUE, then all packets that are to be queued 24 are dropped. When this option is used, the NFQUEUE rule behaves like ACCEPT instead, and the packet 25 will move on to the next table. 26 .PP 27 .TP 28 \fB\-\-queue\-cpu-fanout\fP 29 Available starting Linux kernel 3.10. When used together with 30 \fB--queue-balance\fP this will use the CPU ID as an index to map packets to 31 the queues. The idea is that you can improve performance if there's a queue 32 per CPU. This requires \fB--queue-balance\fP to be specified. 33
