Hi all,
As you've hopefully noticed we've moved next sub-group meeting to next
Thursday. This was to stagger the meetings out for those needing to
attend the other virtualsiation sub-groups so they don't all appear in
the same week.
Hopefully you will have already seen Julianus' notes:
https://groups.google.com/u/0/a/corecollective.dev/g/virtualization/c/EgDZwPPKXcM
We gathered interests from the attendees but we also noted there is more
potential interest from elsewhere in members orgs. I've left the survey
open if you want to forward it to those people:
https://docs.google.com/forms/d/e/1FAIpQLScgYRQz0oIudLaT3_GqfhYY48soB0_NDRKpEbZvc7k5Sr5upQ/viewform?usp=dialog
For the time being people seemed happy for QEMU to be the main focus of
our work.
Broadly there was interest in:
- additional ARM feature emulation
- co-simulation
- tcg plugins
- validation of hypervisors
- virtio devices
I propose we leave discussion of virtio device emulation to the virtio
group. We can happily discuss QEMU (and other) implementations there.
I felt as though I rushed through the current Linaro roadmap so I wonder
if it would be worth returning to that and discussing if there are any
gaps in the emulation support?
The single binary work is now mostly with QC but would it be useful to
look at what the timelines would look like and thinking about next steps
or is that too early?
I feel for co-simulation there is more research needed and syncing up
with other groups who already have this as a topic to look at.
In the maintainer sub-group we did discuss a number of areas in QEMU that
do need maintainer love but the point was raised the list was not really
Arm specific. I'm going to reach out to the upstream to see if there is
scope for trying to arrange funding for maintainer work there.
So what do we want to talk about?
--
Alex Bennée
Virtualisation Tech Lead @ Linaro