> I started working on a VM format that stores signed machine state into an OCI container repository, using the hypervisor migration support of KVM/QEMU.
This report says multikernel is about 2.5x faster than VM hypervisor switching?
> Though this is not safe yet if ever, VM migrations are probably another way to sandbox and deploy en masse
How does this work with multikernel? Can you clone the kernel and it's machine state to multiple nodes, flush their netfilter tables, roll the random seed, and set a node hostname and IP and go from a machine state image?
For anyone who was confused (I certainly was!), mklinux is not MkLinux.
Oh. m(ulti)k(ernel) vs m(icro)k(ernel) and neither means "make".
Can you run several MkLinux kernels (one concurrently with a Darwin kernel) at once?
https://github.com/multikernel/linux https://multikernel.io/
From https://news.ycombinator.com/item?id=49239751#49244282 re: agent sandboxing methods :
> I started working on a VM format that stores signed machine state into an OCI container repository, using the hypervisor migration support of KVM/QEMU.
This report says multikernel is about 2.5x faster than VM hypervisor switching?
> Though this is not safe yet if ever, VM migrations are probably another way to sandbox and deploy en masse
How does this work with multikernel? Can you clone the kernel and it's machine state to multiple nodes, flush their netfilter tables, roll the random seed, and set a node hostname and IP and go from a machine state image?
Does mklinux multikernel flush the CPU caches at KHO (Kexec HandOver) when confidential computing is used, and what about the Landauer limit?