Mesa's Venus Vulkan driver has made cross-device functionality optional in order to enable QEMU support for this open-source driver for virtualized environments.
Virtualization News Archives
576 Virtualization open-source and Linux related news articles on Phoronix since 2006.
Cloud Hypervisor 39 was released on Saturday for this cloud-focused, Rust-based VMM started by Intel and now a multi-vendor Linux Foundation project.
QEMU 9.0 is out tonight as the latest feature release for this prominent component to the open-source Linux virtualization stack.
For confidential computing "CoCo" virtual machines where the VM host is assumed to be un-trusted and aims to be as isolated as possible, RdRand hardware random number generator instructions are one of the limited sources of entropy for guest VMs. Right now RdRand can fail and the CoCo guest VMs will continue to boot albeit with limited or no entropy to see the VM's random number generation. But being merged today as part of x86 fixes for Linux 6.9 is now requiring seeding RNG with RdRand for CoCo environments otherwise a kernel panic.
Patches posted today by an Intel engineer allow for importing scanout buffers from other devices with the VirtIO DRM driver that is used in the virtualization space. The importing of scanout buffers from other devices/drivers can allow for more efficient use by avoiding excess copies.
The Kernel-based Virtual Machine changes for Linux 6.9 continue to enhance the capabilities of the open-source Linux virtualization software stack.
Ahead of next month's Ubuntu 24.04 LTS release, Canonical has released LXD 5.21 as the newest feature update to this container and VM manager. LXD 5.21 now ships with a production-grade graphical user interface by default, brings AMD SEV support for memory encryption of VMs on EPYC CPUs, object storage support, and other features.
Announced one month ago by Cyberus Technology was an open-source KVM back-end for VirtualBox. This work by Cyberus allows for using the KVM hypervisor with VirtualBox as opposed to its custom kernel module maintained by Oracle. That KVM back-end has now been extended to support SR-IOV graphics virtualization.
Ant Group and Alibaba have proposed PVM, the Pagetable Virtual Machine, as a new virtualization framework built upon the Kernel-based Virtual Machine (KVM) hypervisor. PVM does not require hardware-assisted virtualization while working with KVM-enabled software like Kata Containers.
Cloud Hypervisor 38 rolled out this week as the newest version of this Rust-based VMM that was started by Intel while now a Linux Foundation hosted project being worked on by multiple different organizations.
Cyberus Technology announced today the open-source release of a KVM back-end developed for VirtualBox. This new back-end allows the VirtualBox VMM to run virtual machines utilizing the Linux KVM hypervisor instead of the custom kernel module relied upon by the standard Oracle VM VirtualBox software.
Stemming from work done at Amazon Web Services (AWS) for better handling hypervisor live updates, a "request for comments" patch series was sent out on the Linux kernel mailing list for Pkernfs. The Pkernfs proposal was first detailed publicly by AWS last year and is for persisting guest memory and kernel/device state safely across Kexec.
Libvirt as the virtualization API/toolkit developed by Red Hat for managing virtualization on Linux is out today with its v10.0 release.
The KVM virtualization changes for the in-development Linux 6.8 kernel is quite heavy on the feature side.
With the Linux 6.7 kernel there is now KVM virtualization support for LoongArch CPUs. With the upcoming Linux 6.8 kernel cycle, that KVM support is being further extended for these RISC-V-inspired and MIPS64-derived Chinese processors.
QEMU 8.2 has been released as the newest update for this open-source processor emulator that plays an important role within the open-source Linux virtualization stack.
A new Long-Term Support version of Cloud Hypervisor was released this week, which is the open-source project originally started by Intel as a cloud-focused and Rust-written VMM that now has wide industry backing including from multiple other CPU vendors.
AMD Secure Encrypted Virtualization (SEV) and Intel Trust Domain Extensions (TDX) are intended to help provide better security for virtual machines and are key elements to both companies investments around confidential computing. It turns out they have a common enemy in their VM security goals: x86 32-bit software.
As part of AWS Nitro Enclaves, coming for the Linux 6.8 kernel in the new year is a Nitro Secure Module driver.
The Linux Foundation today announced the release of the Xen Project Hypervisor 4.18 for this open-source hypervisor that serves as an alternative to Linux KVM use.
Distrobox 1.6 released on Sunday for this open-source project that makes it easy to launch any Linux distribution inside your terminal. Distrobox builds upon Podman and Docker to allow creating containers of the Linux distribution of your choice and for that to integrate nicely with the host environment. With succeeding releases, Distrobox has built up quite an arsenal of features.
The IOMMU changes for Linux 6.7 aren't particularly noteworthy besides adding SMMUv2 support for the Qualcomm SDM670 and SM7150 SoCs. But the IOMMU updates also take the kernel one step away from supporting Shared Virtual Addressing (SVA) on AMD platforms in the near future.
Cloud Hypervisor 36 has been released as this Rust-written VMM that started out as an Intel open-source project that since was folded into the Linux Foundation umbrella with support from Microsoft, Arm, and other vendors. More recently even the likes of AMD and Ampere Computing have been onboard with this cloud and security focused virtualization hypervisor.
The Kernel-based Virtual Machine (KVM) changes were sent out on Thursday for the Linux 6.7 merge window.
While confidential computing is a hot area right now, there's been a limited amount of cross-vendor cooperation with AMD having their own route with Secure Encrypted Virtualization (SEV) and Intel designing the Trusted Domain Extensions (TDX) that is still available in limited form. As one improvement coming with Linux 6.7, "configfs-tsm" has been submitted for pulling as a cross-vendor solution for confidential computing attestation reports.
Both AMD and Intel engineers have experienced a lengthy journey getting their latest virtualization security features into the mainline Linux kernel -- and one that is still ongoing.
Back in August Linux Containers forked the LXD project as Incus following Canonical's decision to take in control of LXD. Released this weekend was the first formal release for the Incus software.
The past several Linux kernel cycles has seen Loongson engineers working on implementing more kernel features for LoongArch, the MIPS-derived and RISC-V-inspired architecture out of China for domestic PCs. The performance of LoongArch CPUs so far still aren't competitive to x86_64 or Arm hardware but at least slowly are becoming more practical with more features being wired up. The latest milestone is the Kernel-based Virtual Machine (KVM) virtualization support appearing ready ahead of Linux 6.7.
Cloud Hypervisor 35 was released on Thursday for this open-source, Rust-based VMM that was originally started by Intel software engineers before evolving into a multi-vendor initiative for secure and cloud focused workloads.
Merged on Thursday were a batch of KVM virtualization changes for the Linux 6.6 kernel.
Microsoft continues improving the Hyper-V support within the Linux kernel for benefiting Linux guest VMs running within this hypervisor on Windows. With Linux 6.6 the Hyper-V code adds support for SEV-SNP secure guests on the AMD EPYC side while over on the Intel Xeon Scalable Sapphire Rapids side is initial support for Trust Domain Extensions (TDX) protected guests.
QEMU 8.1 is now available as the latest feature update to this important piece of the open-source Linux virtualization stack.
Cloud Hypervisor -- the Rust-written open-source VMM that was started by Intel while having evolved into a Linux Foundation project with backing from multiple organizations -- is out with the Cloud-Hypervisor 34 release.
There is a phenomenon where running a multi-threaded workload inside a virtual machine (VM) with Simultaneously Multi-Threading (SMT / Intel Hyper Threading) that a sibling thread could find itself busy while the CPU core is idle. A new Linux CFS patch series aims to make the scheduler better adapt to the QEMU topology.
As was anticipated, UEFI Unaccepted Memory support was successfully merged for Linux 6.5 to support this standard that's important with the likes of Intel Trusted Domain Extensions (TDX) and AMD Secure Encrypted Virtualization Secure Nested Page (SEV-SNP) backed virtual machines.
The Kernel-based Virtual Machine (KVM) changes were submitted today that are ready for introduction in the Linux 6.5 kernel.
The Virtual Function I/O "VFIO" changes were merged last week for the ongoing Linux 6.5 kernel merge window. This IOMMU/device agnostic framework has added an AMD CDX driver this cycle along with other improvements for this subsystem that is important to the Linux virtualization stack.
Cloud Hypervisor has advanced quite nicely in the half-year it's been around since Intel software engineers began writing this Rust-based cloud-focused virtualization hypervisor. This VMM project has since become more independent and regularly receiving code contributions from the likes of Arm, Microsoft, and Tencent while also gaining the support of companies like AMD and Ampere. On Thursday marked the release of Cloud Hypervisor 33.
Google engineers have been working on Linux patches to improve the guest VM performance when the host encounters memory pressure or have over-committed too many guests. Similar patches already are used on Chrome OS and Google has been working to upstream the functionality under the mainline Linux kernel and have now provided some reference benchmark results.
For over two years Intel open-source engineers have been working on preparing the Linux kernel for UEFI unaccepted memory support and it looks like that will cross the finish line with the upcoming Linux 6.5 cycle.
Queued via drm-misc-next and now on its way to DRM-Next ahead of the upcoming Linux 6.5 cycle is support in the VirtIO GPU DRM driver for the synchronization object user-space API. This is important for Vulkan and is working with the Venus/Virgl driver context as well as Intel/AMDGPU/Freedreno native contexts too.
The Virgl driver within Mesa for allowing open-source OpenGL support within virtualized environments in conjunction with the Virglrenderer is now capable of exposing OpenGL 4.6.
An updated set of patches were posted on Friday that seem to improve the Linux guest VM performance when the host is under heavy memory pressure.
Released this week from the Red Hat Summit is Podman Desktop 1.0 as a container management tool akin to Docker Desktop.
Adding to the list of features coming with QEMU 8.1 is now having a native PipeWire audio back-end.
Near the start of 2022 engineers out of the Qualcomm Innovation Center posted Linux driver patches for their Gunyah hypervisor. Gunyah is an open-source type-1 hypervisor developed by Qualcomm with an emphasis on security and other features. More than one year later the Gunyah drivers have yet to be upstreamed into the mainline Linux kernel but work on them persists.
QEMU 8.0 is out today as the newest feature release for this processor emulator that plays an important role in the open-source Linux virtualization stack.
Cloud Hypervisor 31 was released this week as the Rust-written VMM started by Intel that runs atop Linux KVM and Microsoft MSHV while these days is a Linux Foundation project receiving regular contributions from not only Intel and Microsoft but also Tencent, Arm, and other players.
Google engineers on Thursday posted initial "request for comments" patches on their KVM-CPUFreq driver that is part of their effort to improve the dynamic voltage and frequency scaling behavior and task placement within KVM-based virtual machines. This effort is leading to big improvements in raw performance and performance-per-Watt for tasks running within Linux VMs.
The focus of this new effort isn't to immediately rewrite the Xen virtualization hypervisor in Rust but to begin gradually working toward rewriting some of the smaller Xen Project components in the Rust programming language and to see how everything pans out.
576 Virtualization news articles published on Phoronix.