awesome! I was very concerned that the new fib table optimizations, including RCU, and the routing/switching hardware offload stuff entering the Linux kernel, was going to be dicy without atomic updates implemented at higher levels. And I always hated losing packets for any reason.
Is there some sort of regression or test suite that could drive very high levels of route changes into the kernel (while under massive forwarding load?) that could be tried with this? _______________________________________________ Quagga-dev mailing list [email protected] https://lists.quagga.net/mailman/listinfo/quagga-dev
