On 17.04.25 04:46, Chenyi Qiang wrote:
On 4/16/2025 7:58 PM, David Hildenbrand wrote:
Hi everybody,
our next guest_memfd upstream call is scheduled for Thursday,
2025-04-17 at 8:00 - 9:00am (GMT-07:00) Pacific Time - Vancouver.
We'll be using the following Google meet:
http://meet.google.com/wxp-wtju-jzw
The meeting notes can be found at [1], where we also link recordings and
collect current guest_memfd upstream proposals. If you want an google
calendar invitation that also covers all future meetings, just write me
a mail.
If nothing else comes up, let's talk about the next steps to get basic
mmap support [2] ready for upstream, to prepare for actual in-place
conversion, direct-map removal and much more.
In particular, let's talk about what "basic mmap support" is, and what
we can use it for without actual in-place conversion: IIUC "only shared
memory in guest_memfd" use cases and some cases of software-protected
VMs can use it.
Also, let's talk about the relationship/expectations between guest_memfd
and the user (mmap) address when it comes to KVM memory slots that have
a guest_memfd that supports "shared" memory.
To put something to discuss onto the agenda, reply to this mail or add
them to the "Topics/questions for next meeting(s)" section in the
meeting notes as a comment.
[1]
https://docs.google.com/document/d/1M6766BzdY1Lhk7LiR5IqVR8B8mG3cr-
cxTxOrAosPOk/edit?usp=sharing
[2] https://lore.kernel.org/all/20250318161823.4005529-1-
tabba@xxxxxxxxxx/T/#u
Hi David,
Hi,
If we have time, I'd like to discuss about my v4 posting of shared
device assignment support
(https://lore.kernel.org/qemu-devel/20250407074939.18657-1-chenyi.qiang@xxxxxxxxx/)
which introduces a new abstract parent class of RamDiscardManager, and a
new priority listener to apply to in-place conversion. Hope to get some
suggestion or confirmation if I'm in the correct direction.
yes we can discuss that (and it's on my todo list as well to review). I
suspect that it's mostly review that's missing at that point, and that
it is conceptually ok.
Interestingly, I might be looking into virtio-mem support for
confidential VMs at some point; I'll have to figure out how to allow for
more states then :)
--
Cheers,
David / dhildenb
