This looks very interesting. If this were included in the Linux kernel, would it read and utilize existing sysctl memory and qlen values as well as have it's own sysctl settings, or would all the settings be derived at module load in the modprobe parameters? Why I am asking is that I current disable TOE (TCP offload engine) on all my nic's as they have their own buffer and retry settings and ignore the OS network settings.
It's still in development at the moment. We'll see about the interface. But there are no plans to implement any segmentation offloads or TOE in Corundum, that will be left up to the network stack. However, scatter/gather DMA support is planned so that software GSO will work. Right now, most of the low-level twiddling is done from a user space app that directly accesses device registers. For a research device, that's fine, but that would obviously have to be improved for a commercial product.
