On 16.03.2018 21:31, Sowmini Varadhan wrote: > On (03/16/18 21:14), Kirill Tkhai wrote: >> >> I did the second version and sent you. Have you tried it? > > I tried it briefly, and it works for the handful of testcases > that I tried, but I still think its very werid to register > as both a device and a subsys, esp in the light of the > warning in net_namespace.h I'll send a patch updating comment in net_namespace.h to clarify the situation. > Thus I have to spend some time reviewing your patch, > and I cannot give you an answer in the next 5 minutes. No problem, 5 minutes response never required. Thanks for your review. >> Calling netdevice handler for every event is more disturbing, >> as number of events is several times bigger, than one more >> pernet exit method. > > So you are saying there are scaling constraints on subsystems > that register for netdevice handlers. The disturbing part > of that is that it does not scale. It does not scale. But if we do not work on this, nothing will change. My point about less numbers is that every namespace will call this new pernet->exit once. So, we will have 1 new function call per one namespace. Let's count numbers, if we have the same functionality implemented as netdev event. Every net namespace has loopback dev. This device will be registered and unregistered. The event handler will be called for many events during this actions: NETDEV_REGISTER, NETDEV_UNREGISTER, NETDEV_POST_INIT, etc, etc. So, if we use pernet exit, it leads us to signify less number of function calls. Just this. Kirill -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
