KVM Forum 2021
| title | author | length |
|---|---|---|
| Day 1 Keynote Sessions | Michael S. Tsirkin, Alex Bennée | 26m |
Charter expands virtio scope to cover hardware passthrough and nested setups and formally registers device‑IDs and feature bits with a compatibility guarantee… | ||
| Day 2 Keynote Sessions | Paolo Bonzini, Eric Ernst | 27m |
KVM sees about 1,400 commits from Oct 2020 to Sep 2021 with ~10% stable fixes and adds >10,000 lines of self-tests; x86 and ARM lead work. New page‑table MMU… | ||
| Dirty Quota-Based VM Live Migration Auto-Converge | Manish Mishra, Shivam Kumar | 22m |
Implements a kernel-level microstun that assigns per-vCPU delta quota windows and a shared common quota so dirty rate stays ≤ network_bandwidth/x (default… | ||
| New Qemu Backup Architecture and API | Vladimir Sementsov-Ogievskiy | 11m |
Claims a new QEMU backup architecture and API that implements a block-driver-based, point-in-time consistent backup path with bitmap-based incremental external… | ||
| KVM Dirty Page Tracking | Peter Xu | 25m |
Explains that KVM uses synchronous and asynchronous dirty tracking for snapshots and migration and reports concrete measurements: migrating a 3 TB guest with a… | ||
| Background Snapshots in QEMU: Towards Asynchronous Revert | Denis V. Lunev | 20m |
Prototype implements background snapshot in QEMU using Linux 5.7’s write-protect and userfaultfd-style page tracking, streaming a disk state while the VM runs… | ||
| Know your QEMU and KVM Test Frameworks | Thomas Huth | 22m |
Filesystem paths and commands: compile the KVM self-tests with make in tools/testing/self-tests/kvm and run binaries directly; the framework includes an… | ||
| Lessons Learned Building a Production Memory-Overcommit Solution | Florian Schmidt, Ivan Teterevkov | 21m |
A production control plane ties Linux memory management, cgroups and procfs to a hybrid strategy that first tries virtio balloon and only falls back to… | ||
| Qemu Performance Regression CI | Lukáš Doktor | 20m |
A CI pipeline runs QEMU system-level benchmarks with runperf driving tests and pbench producing machine-readable result.json files, stores about 2 GB per build… | ||
| Support for Fast and Reliable VMM Live Upgrades in Libvirt | Soham Ghosh, Prachatos Mitra | 24m |
Libvirt + QEMU implement a local live-migration that passes open memory backend FDs instead of copying RAM, adding a QMP command fetch-backend-fd and reusing… | ||
| HCK-CI: Enabling CI for Windows Guest Paravirtualized Drivers | Kostiantyn Kostiuk | 20m |
HCK-CI is a CI framework that runs WHQL/HLK driver certification inside VM snapshots, creating a controller plus client images, orchestrating networks and… | ||
| Kubevirt and the Cost of Containerizing VMs | Guoqing Li, Dario Faggioli | 19m |
Benchmarks run on a single server with 32 physical cores, Ubuntu host and guest images (OpenSUSE Leap 15.2) compare KVM (libvirt/QEMU built from source) and… | ||
| Protecting from MaliciousHypervisor Using AMD SEV-SNP | Brijesh Singh | 25m |
AMD SEV‑SNP aims to protect a VM from a malicious hypervisor by adding integrity and replay protection on top of SEV/SEV‑ES, enforced with a system‑wide… | ||
| Receive Side Scaling (RSS) with eBPF in QEMU and virtio-net | Yan Vugenfirer | 19m |
Implementation of RSS via eBPF inside QEMU for virtio-net, with a software proof‑of‑concept in QEMU and an eBPF program compiled into a skeleton and loaded via… | ||
| Status Update on TDX Support | Isaku Yamahata | 17m |
Status update on KVM TDX support covers concrete changes across the stack: Linux must load a special TDX module with large contiguous memory and support… | ||
| Towards High-availability for Virtio-fs | Jiachen Zhang, Yongji Xie | 21m |
Implements crash recovery, live upgrade, and live migration for virtio‑fs by adding a supervisor restart path, QEMU reconnection of the virtio-user socket… | ||
| TDX Live Migration | Wei Wang | 27m |
Implements live migration for Intel TDX by adding a microTD helper that establishes a secure migration key and uses TD module sim calls to export/import… | ||
| Hyperscale vDPA | Jason Wang | 23m |
Claims a path to scale vDPA to cloud densities of 10k–100k lightweight instances by moving control off the per-queue PCIe bar into a centralized management… | ||
| Unmapped Guest Memory | Yu Zhang | 21m |
Mapping entire guest memory into host user space breaks Intel TDX guarantees by exposing private pages that carry a 28‑bit MAC and TD owner bit whose… | ||
| VDUSE - vDPA Device in Userspace | Yongji Xie | 17m |
vDUSE implements vDPA device emulation entirely in userspace with the control path handled by the kernel driver and the data path by a userspace daemon that… | ||
| libkrun: More than a VMM, in Dynamic Library Form | Sergio Lopez Pascual | 15m |
libkrun is a dynamic library VMM that links into a runtime to carry KVM, firmware and a bundled minimal Linux kernel inside the process so namespace switches… | ||
| Hypervisor-less Virtio for Real-time and Safety | Maarten Koning | 22m |
A hypervisor-less virtio proof-of-concept implements virtio front ends in auxiliary real-time runtimes and a modified kvmtool daemon backend that maps shared… | ||
| Don't Peek Into my Container! | Alice Frosi, Christophe de Dinechin, Sergio Lopez Pascual | 33m |
Describes a working stack that combines Kubernetes + cri-o + KVM to run encrypted containers: workload mode boots a VM per container using libkiran and an… | ||
| vdpa-blk: Unified Hardware and Software Offload for virtio-blk | Stefano Garzarella | 18m |
Proposal builds a unified virtio-blk offload using the vDPA framework, reusing QEMU and kernel code so the same stack supports hardware accelerators and a… | ||
| Securing Linux VM boot with AMD SEV measurement | Dov Murik, Hubertus Franke | 19m |
Extends AMD SEV launch measurement by placing a hash table in initial guest memory (SHA‑256, 32‑byte hashes) that lists kernel, initrd and kernel command line… | ||
| rust-vmm: A Security Journey | Andreea Florescu | 20m |
rust-vmm provides modular Rust crates for virtualization and enforces security with cargo-audit scans, one-crate-per-component packaging, and documented unsafe… | ||
| Encrypted Virtual Machine Images for Confidential Computing | James Bottomley, Brijesh Singh | 20m |
Demonstrates an end-to-end method to boot an x86 encrypted VM where decryption happens inside the guest: a patched OVMF-with-GRUB image reserves a one-page… | ||
| Is QEMU too Complex, and What Can we do About It? | Paolo Bonzini | 25m |
QEMU is too complex because accidental complexity piles onto unavoidable domain complexity; the talk quantifies it: 120 command line options implemented in… | ||
| Secure Live Migration of Encrypted VMs | Tobin Feldman-Fitzthum, Dov Murik | 21m |
Implements a firmware migration handler that runs inside a mirror VM sharing the SEV encryption context so pages can be exported wrapped with a PSP transport… | ||
| Live Migrating VFIO, vhost-user, and vfio-user Devices | Stefan Hajnoczi | 23m |
QEMU today migrates devices by serializing runtime state and launching a full destination VM; vhost-user takeover stops the out‑of‑process device, QEMU copies… | ||
| Securing the Hypervisor with Control-Flow Integrity | Daniele Buono | 19m |
Adds clang's CFI and SafeStack to QEMU, with upstream patches that work but require clang and LTO for whole-binary jump tables. Implements SafeStack per… | ||
| VFIO User - Using VFIO as the IPC Protocol in Multi-process QEMU | John Johnson, Jagannathan Raman | 27m |
Project switches QEMU multi-process IPC from a custom proxy to VFIO encapsulated over a socket, using libvfio-user to reuse kernel ioctl semantics so existing… | ||
| Host & Guest Tracing in Virtualization: "To sync, or not to sync?" | Stefano De Venuto, Tzvetomir Stoyanov | 20m |
Implements host↔guest trace collection in trace-cmd 2.9 using vSockets (optionally FIFOs on KVM) and visualizes merged traces in KernelShark 2.0; synchronizes… | ||
| libvfio-user: Status Update | Thanos Makatos, John Levon | 19m |
libvfio-user is a C library that implements a socket protocol for external device servers, provides PCI config handling, and exposes APIs for… | ||
| Passthrough/Headless GPU Gets Ahead | Tina Zhang, Vivek Kasireddy | 20m |
Describes a working PV-based path to share guest scanout buffers from a passthrough GPU to a host display without CPU copies by exporting dma-bufs through a… | ||
| High Performance NVMe Offloading in SPDK Using the New vfio-user Protocol | Changpeng Liu, Benjamin Walker | 25m |
SPDK implements an nvme-over-fabrics transport that uses shared memory rings for data and a vfio-user unix domain socket for control, added as a standalone… | ||
| Analysis of AMD HW-assisted vIOMMU Implementation and Performance | Suravee Suthikulpanit, Wei Huang | 26m |
Prototype implements AMD hardware-assisted vIOMMU with QEMU amd_vio device model, host IOMMU driver changes and a guest v2 page-table path; code is pushed as… | ||
| Towards a More Efficiently Synchronization in KVM | Wanpeng Li | 10m |
Identifies three concrete KVM issues: busy‑waiting wastes host CPU when a vCPU is preempted and causes long waits for TLB shootdowns, so the patch boost… | ||
| How Hard Could it be to Flip a bit? KVM PV Feature Enablement up the Virtualization Stack | Vitaly Kuznetsov | 24m |
KVM PV features exist in kernel and qemu but enabling them by default is blocked because QEMU must remain compatible with the oldest supported kernel (linux… | ||
| QEMU Emulated NVMe - Lessons Learned and Future Work | Klaus Jensen | 23m |
QEMU implements an emulated NVMe controller that now supports NVMe v1.4, multiple namespaces (added in 2020), metadata, and namespace sharing; the original… | ||
| Support SDEI Virtualization and Asynchronous Page Fault for arm64 | Gavin Shan | 24m |
Adds arm64 support for asynchronous SDEI-based page-fault delivery using private SDEI event notifications via the SMC/SMCC calling convention, plus a shared… | ||
| Sharing IOMMU PageTables with TDP in KVM | Lu Baolu, Zhao Yan | 19m |
KVM shares the CPU TDP as IOMMU page tables to remove duplicated GPA→HPA mappings by adding three interfaces: request_sharing, pin/unpin and page‑fault… | ||
| Debugging Secured Windows OS guest using KVM/QEMU and Windbg | Marek Kędzierski | 20m |
Demonstrates kernel debugging of a secured Windows guest on KVM/QEMU using WinDbg by implementing an XDI-to-GDB bridge so WinDbg talks to QEMU instead of an… | ||
| The Traps of Using Hyper-V Features in KVM Environment | Liang Li | 19m |
KVM exposes many Hyper‑V features to improve Windows performance but enabling them can hurt real workloads: cloud gaming shows 35,000 IPIs/sec spikes and… | ||
| Mitigating Excessive Pause-Loop-Exiting in VM-Agnostic KVM | Kenta Ishiguro | 17m |
Identifies excessive Pause-Loop-Exiting (PLE) caused by continuous spinning—over 100 back-to-back PLEs make up >20% of events in some benchmarks—and implements… | ||
| KVM Memory Cost Optimization in Alibaba Cloud | Huaitong Han | 17m |
KVM Memory Cost Optimization in Alibaba Cloud - Huaitong Han, Alibaba Cloud | ||
| nothing matches | ||