very interesting work! I've been exploring a different idea on the side, using SPDK+libvfio-user [0] to emulate PCIe devices inside QEMU, which doesn't require a kernel module but it's a bit less flexible than this approach.
Highly interesting! I kinda wanted not to rely on QEMU as a default "end" for the emulation (As in, I want the end user to be able to choose whatever userspace shim/transport layer/thing they want), but for some of my tests I did forward accesses to QEMU itself (And worked wonders). Thanks for that link! Super cool stuff!
cakehonolulu|1 month ago