Could your kernel module not just register its own qdisc? Why do you need to modify anything in sch_api.c for that? If your module had its own qdisc then you just need to configure devices to pass traffic through that qdisc (using tc), same as if your where using, for example, the htb qdisc. I would never use rtnetlink for in kernel networking. Anna > -----Original Message----- > From: linux-net-owner@xxxxxxxxxxxxxxx [mailto:linux-net- > owner@xxxxxxxxxxxxxxx] On Behalf Of Anirudh Badam > Sent: 10 July 2008 23:45 > To: linux-net@xxxxxxxxxxxxxxx > Subject: In Kernel Traffic Control > > Hello > I am porting a module from 2.6.11 to 2.6.25. This module needs > an in kernel support for traffic control. While the rtnetlink.c file > gives me hooks to do traffic controlling from user space, it is > inefficient to use netlink messages for traffic control inside a > kernel module. Besides, when I try to send netlink messages via > rtnl_unicast from a module it just blocks for ever! I have tried a lot > of hacks but all of them hit the road block of some data variable > being a static one or the structure definition being in a '.c' file. I > have tried mocking the doit and dumpit functions defined in sch_api.c > and sch_htb.c etc in net/sched but as I have said it hits the road > block of something being static or a data structure definition being > in a '.c' file. Is there is a route around this problem? > I am using the 2.6.25 kernel. > > Regards > Anirudh Badam > > -- > Like Marvin the paranoid android says, "Whats the point?"!!! > -- > To unsubscribe from this list: send the line "unsubscribe linux-net" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html -- To unsubscribe from this list: send the line "unsubscribe linux-net" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
