Well, it's not uncommon for apparently useful patch sets to be posted to lkml and just be ignored, even if they come from large companies and are posted many times. Given that kernel development seems to suffer from throughput bottlenecks, it makes sense that researchers ignore it.