bstx

If you happen to be in SF, the Vallejo/Mare Island ferry should be a relatively convenient (and scenic) public transit option.

ARM architectural licensees develop their own microarchitectures that implement the ARM ISA spec, they do not license any particular microarchitecture from ARM (e.g. Cortex-A? IP cores). That includes Apple, Samsung, Nvidia and others.

No. It has become significantly easier for Europeans to immigrate since 2005. Labor certification alone now takes less than a year rather than multiple years pre-PERM. EB2 and EB3 priority dates are current. Europeans can expect to get a green card within a couple years (even accounting for the new mandatory Adjustment of Staus interviews for employment based applicants). It has become significantly harder for Indians where the green card backlog approaches a century (!) for EB2/EB3. The only thing that has become harder is getting an H-1B visa due to oversubscription.

It doesn't make sense for speculatively executed code to throw architecturally visible exceptions. The appropriate behavior would be to not perform speculative loads across protection domains (i.e. the behavior of AMD implementations).

Hypervisor.framework itself only deals with CPU virtualization and is agnostic to whatever device model is implemented in userspace.

Nope, VMs are demand paged like any other process.

Well, it's not just the userland. The XNU kernel is a hybrid of Mach (tasks, threads, IPC, VM), BSD (processes, users, devices, networking, system calls, VFS) and IOKit (driver framework)

Did you have a VirtualBox VM running before xhyve?

Any KVM vs. Xen PV benchmarks? The Xen (PV)HVM case isn't that interesting.

> You seem to be laboring under some serious misapprehensions here. QEMU is very much a virtualization solution. It calls itself an emulator because it's older than the name 'virtualization.'

QEMU is an emulator, it isn't by itself capable of doing virtualization. It can allow a hypervisor to do CPU (and some I/O) virtualization in lieu of QEMU doing binary translation/emulation. In that case QEMUs purpose is mainly I/O emulation. 'Virtualization' as a name and concept predates QEMU by several decades.

> KVM is merely the kernel infrastructure that enables ring-0 access for faster hardware access. For many years it was called qemu-kmod; to this day the actual process that you run is called 'kvm-qemu.'

kqemu and KVM are technologically unrelated.

> You are overstating the difference between 'virtualization' and 'emulation.' There really isn't one. What used to be called 'emulation' is nowadays called 'full-virtualization' as opposed to 'paravirtualization.' QEMU, KVM, and XenKVM are full-virt, while traditional Xen is paravirt.

Para-Virtualization is the concept where the guest OS is aware that it is being run on a hypervisor. For example using hypervisor hypercalls to create new address spaces rather than setting up hardware page tables. "Full" or hardware level virtualization happens when the guest OS does not need to be aware of this fact. Neither is related to emulation.

> KVM is not a hypervisor;

It is.

> libvirt is.

This one isn't.

> Xen blurs the line between virtualization platform and hypervisor because both pieces ship within the same project. Either libvirt or Xen can act as hypervisor for QEMU or KVM.

"hypervisor":"virtualization platform" ~ "kernel":"operating system". KVM and Xen can act as hypervisor for Qemu managed by libvirt.

They do that to some degree in Germany, which has a 3-tier secondary education system ought to group students based on their (perceived) academic aptitude. It has however been accused for being biased toward social status.

If you don't mind trailing whitespace block select might be an option (option + select on OS X / ctrl + alt + select on Linux). Or just use copy mode and yank the selection to the system clipboard.