Re: [ANNOUNCE] KVM Microconference at LPC 2024

From: Michael Roth
Date: Wed Apr 03 2024 - 13:53:48 EST


On Tue, Apr 02, 2024 at 12:06:52PM -0700, Sean Christopherson wrote:
> We are planning on submitting a CFP to host a second annual KVM Microconference
> at Linux Plumbers Conference 2024 (https://lpc.events/event/18). To help make
> our submission as strong as possible, please respond if you will likely attend,
> and/or have a potential topic that you would like to include in the proposal.
> The tentative submission is below.
>
> Note! This is extremely time sensitive, as the deadline for submitting is
> April 4th (yeah, we completely missed the initial announcement).
>
> Sorry for the super short notice. :-(
>
> P.S. The Cc list is very ad hoc, please forward at will.
>
> ===================
> KVM Microconference
> ===================
>
> KVM (Kernel-based Virtual Machine) enables the use of hardware features to
> improve the efficiency, performance, and security of virtual machines (VMs)
> created and managed by userspace. KVM was originally developed to accelerate
> VMs running a traditional kernel and operating system, in a world where the
> host kernel and userspace are part of the VM's trusted computing base (TCB).
>
> KVM has long since expanded to cover a wide (and growing) array of use cases,
> e.g. sandboxing untrusted workloads, deprivileging third party code, reducing
> the TCB of security sensitive workloads, etc. The expectations placed on KVM
> have also matured accordingly, e.g. functionality that once was "good enough"
> no longer meets the needs and demands of KVM users.
>
> The KVM Microconference will focus on how to evolve KVM and adjacent subsystems
> in order to satisfy new and upcoming requirements. Of particular interest is
> extending and enhancing guest_memfd, a guest-first memory API that was heavily
> discussed at the 2023 KVM Microconference, and merged in v6.8.
>
> Potential Topics:
> - Removing guest memory from the host kernel's direct map[1]
> - Mapping guest_memfd into host userspace[2]
> - Hugepage support for guest_memfd[3]
> - Eliminating "struct page" for guest_memfd

Another gmem proposal we were considering was:

- Scalability/Performance Analysis of guest_memfd

Mainly looking at things like points of contention during lazy acceptance for
large guests, page-conversion latency increases, impact of discard/realloc
(prealloc?) of gmem pages from userspace, etc.

Thanks,

Mike

> - Passthrough/mediated PMU virtualization[4]
> - Pagetable-based Virtual Machine (PVM)[5]
> - Optimizing/hardening KVM usage of GUP[6][7]
> - Defining KVM requirements for hardware vendors
> - Utilizing "fault" injection to increase test coverage of edge cases
>
> [1] https://lore.kernel.org/all/cc1bb8e9bc3e1ab637700a4d3defeec95b55060a.camel@xxxxxxxxxx
> [2] https://lore.kernel.org/all/20240222161047.402609-1-tabba@xxxxxxxxxx
> [3] https://lore.kernel.org/all/CABgObfa=DH7FySBviF63OS9sVog_wt-AqYgtUAGKqnY5Bizivw@xxxxxxxxxxxxxx
> [4] https://lore.kernel.org/all/20240126085444.324918-1-xiong.y.zhang@xxxxxxxxxxxxxxx
> [5] https://lore.kernel.org/all/20240226143630.33643-1-jiangshanlai@xxxxxxxxx
> [6] https://lore.kernel.org/all/CABgObfZCay5-zaZd9mCYGMeS106L055CxsdOWWvRTUk2TPYycg@xxxxxxxxxxxxxx
> [7] https://lore.kernel.org/all/20240320005024.3216282-1-seanjc@xxxxxxxxxx