Hello, I'm interested in using netsniff_ng as a capture backend, in order to perform some on-the-fly processing of captured packets.
One thing that is puzzling me is the lack of any memory barriers in the netsniff_ng capture loop (receive_to_xmit()). How is memory access synchronized between the program and the kernel thread? In other example usages of PACKET_RX_RING I have seen, the authors insert a memory barrier to ensure that the write to a consumed packet is visible to the kernel. I do not see this in netsniff_ng's kernel_may_pull_from_rx(). Any enlightenment greatly appreciated. Regards, Mark -- You received this message because you are subscribed to the Google Groups "netsniff-ng" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
