firecracker
firecracker copied to clipboard
Secure and fast microVMs for serverless computing.
There have been several discussions about the way devices, device resources and device managers need to coexist in Firecracker. Here are some links: - https://github.com/firecracker-microvm/firecracker/pull/4487#discussion_r1552065520 - https://github.com/firecracker-microvm/firecracker/pull/4428#discussion_r1509748674 - https://github.com/firecracker-microvm/firecracker/pull/4428#discussion_r1516203612 -...
### Description Currently, the logic to create a `Vmm` object is scattered across `vmm/lib.rs` and `vmm.builder.rs` files and it is quite convoluted and some times difficult to follow. Moreover, there...
Currently Firecracker have an ability to create multiple irqs for MMIO devices. These irqs are store in the `MMIODeviceInfo` struct [here](https://github.com/firecracker-microvm/firecracker/blob/8f36d3a3a7d600915fd5989f4783edcd9e047ae9/src/vmm/src/device_manager/mmio.rs#L80). As per discussion [here](https://github.com/firecracker-microvm/firecracker/pull/4428#discussion_r1528162777) we would like to remove...
Currently when using anonymous memory for KVM guest RAM, the memory all remains mapped into the kernel direct map. We are looking at options to get KVM guest memory out...
# Describe the bug Hi, I was checking out if I could run firecracker-vm on the Volterra box (Windows Dev Kit 2023). It is booted into EL2, has /dev/kvm, and...
## Reason Currently Firecracker does not have a single `DeviceManger` but instead has `MMIODeviceManager` and `PortIODeviceManager` (only for x86). This design works for now, but with an addition of ACPI...
Derived from discussion here: https://github.com/firecracker-microvm/firecracker/pull/4492#discussion_r1526045051 After we move to Rust 1.77.0 (the current latest is 1.76.0 at the time of writing), [`core::mem::offset_of!()` macro](https://doc.rust-lang.org/nightly/src/core/mem/mod.rs.html#1396-1401) seems to become stable finally!! Currently, we...
Our setup of cache topology (at least on x86_64) is kinda dummy. At the end of the [Automated Performance TestingFor Virtualization with MMTests](https://fosdem.org/2020/schedule/event/testing_automated_performance_testing_virtualization/attachments/slides/3926/export/events/attachments/testing_automated_performance_testing_virtualization/slides/3926/Automated_Performance_Testing_for_Virtualization_with_MMTests.pdf) there are some slides with performance benchmarks...
Right now the MMDS functionality is backed into the virtio net implementation. We should find a way of separating this or making it pluggable so that we can consume Virtio...
## Changes vhost-net: create vhost based Net backend This patch adds a second backend to Net devices. It can be enabled with 'vhost' bool on the network interface config, like...