xref: /cloud-hypervisor/release-notes.md (revision 93b77530c717a151c54c740b4a7f57995ebfaa35)
1*93b77530SSamuel Ortiz# v0.1.0
2*93b77530SSamuel Ortiz
3*93b77530SSamuel OrtizThis release has been tracked through the [0.1.0 project](https://github.com/intel/cloud-hypervisor/projects/1).
4*93b77530SSamuel Ortiz
5*93b77530SSamuel OrtizHighlights for `cloud-hypervisor` version 0.1.0 include:
6*93b77530SSamuel Ortiz
7*93b77530SSamuel Ortiz### Shared filesystem
8*93b77530SSamuel Ortiz
9*93b77530SSamuel OrtizWe added support for the [virtio-fs](https://virtio-fs.gitlab.io/) shared file
10*93b77530SSamuel Ortizsystem, allowing for an efficient and reliable way of sharing a filesystem
11*93b77530SSamuel Ortizbetween the host and the `cloud-hypervisor` guest.
12*93b77530SSamuel Ortiz
13*93b77530SSamuel OrtizSee our [filesystem sharing](https://github.com/intel/cloud-hypervisor/blob/master/docs/fs.md)
14*93b77530SSamuel Ortizdocumentation for more details on how to use virtio-fs with `cloud-hypervisor`.
15*93b77530SSamuel Ortiz
16*93b77530SSamuel Ortiz### Initial direct device assignment support
17*93b77530SSamuel Ortiz
18*93b77530SSamuel OrtizVFIO (Virtual Function I/O) is a kernel framework that exposes direct device
19*93b77530SSamuel Ortizaccess to userspace. `cloud-hypervisor` uses VFIO to directly assign host
20*93b77530SSamuel Ortizphysical devices into its guest.
21*93b77530SSamuel Ortiz
22*93b77530SSamuel OrtizSee our [VFIO](https://github.com/intel/cloud-hypervisor/blob/master/docs/vfio.md)
23*93b77530SSamuel Ortizdocumentation for more detail on how to directly assign host devices to
24*93b77530SSamuel Ortiz`cloud-hypervisor` guests.
25*93b77530SSamuel Ortiz
26*93b77530SSamuel Ortiz### Userspace IOAPIC
27*93b77530SSamuel Ortiz
28*93b77530SSamuel Ortiz`cloud-hypervisor` supports a so-called split IRQ chip implementation by
29*93b77530SSamuel Ortizimplementing support for the [IOAPIC](https://wiki.osdev.org/IOAPIC).
30*93b77530SSamuel OrtizBy moving part of the IRQ chip implementation from kernel space to user space,
31*93b77530SSamuel Ortizthe IRQ chip emulation does not always run in a fully privileged mode.
32*93b77530SSamuel Ortiz
33*93b77530SSamuel Ortiz### Virtual persistent memory
34*93b77530SSamuel Ortiz
35*93b77530SSamuel OrtizThe `virtio-pmem` implementation emulates a virtual persistent memory device
36*93b77530SSamuel Ortizthat `cloud-hypervisor` can e.g. boot from. Booting from a `virtio-pmem` device
37*93b77530SSamuel Ortizallows to bypass the guest page cache and improve the guest memory footprint.
38*93b77530SSamuel Ortiz
39*93b77530SSamuel Ortiz### Linux kernel bzImage
40*93b77530SSamuel Ortiz
41*93b77530SSamuel OrtizThe `cloud-hypervisor` linux kernel loader now supports direct kernel boot from
42*93b77530SSamuel Ortiz`bzImage` kernel images, which is usually the format that Linux distributions
43*93b77530SSamuel Ortizuse to ship their kernels. For example, this allows for booting from the host
44*93b77530SSamuel Ortizdistribution kernel image.
45*93b77530SSamuel Ortiz
46*93b77530SSamuel Ortiz### Console over virtio
47*93b77530SSamuel Ortiz
48*93b77530SSamuel Ortiz`cloud-hypervisor` now exposes a `virtio-console` device to the guest. Although
49*93b77530SSamuel Ortizusing this device as a guest console can potentially cut some early boot
50*93b77530SSamuel Ortizmessages, it can reduce the guest boot time and provides a complete console
51*93b77530SSamuel Ortizimplementation.
52*93b77530SSamuel Ortiz
53*93b77530SSamuel OrtizThe `virtio-console` device is enabled by default for the guest console.
54*93b77530SSamuel OrtizSwitching back to the legacy serial port is done by selecting
55*93b77530SSamuel Ortiz`--serial tty --console off` from the command line.
56*93b77530SSamuel Ortiz
57*93b77530SSamuel Ortiz### Unit testing
58*93b77530SSamuel Ortiz
59*93b77530SSamuel OrtizWe now run all unit tests from all our crates directly from our CI.
60*93b77530SSamuel Ortiz
61*93b77530SSamuel Ortiz### Integration tests parallelization
62*93b77530SSamuel Ortiz
63*93b77530SSamuel OrtizThe CI cycle run time has been significantly reduced by refactoring our
64*93b77530SSamuel Ortizintegration tests; allowing them to all be run in parallel.
65