[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAMw=ZnQ79RRXGi11a3KO6Y2FcpVGL4adAT9yAKb2AK5Z2=1YSw@mail.gmail.com>
Date: Wed, 30 Oct 2024 00:38:43 +0000
From: Luca Boccassi <bluca@...ian.org>
To: Tejun Heo <tj@...nel.org>
Cc: Paolo Bonzini <pbonzini@...hat.com>, Roman Gushchin <roman.gushchin@...ux.dev>, kvm@...r.kernel.org,
cgroups@...r.kernel.org, Michal Koutný <mkoutny@...e.com>,
linux-kernel@...r.kernel.org
Subject: Re: cgroup2 freezer and kvm_vm_worker_thread()
On Wed, 30 Oct 2024 at 00:25, Tejun Heo <tj@...nel.org> wrote:
> If that can't be done, we can add a mechanism to ignore these tasks when
> determining when a cgroup is frozen provided that this doesn't affect the
> snapshotting (ie. when taking a snapshot of frozen cgroup, activities from
> the kthreads won't corrupt the snapshot).
What do these kvm kthreads actually do?
So one of the use cases in systemd is to freeze the entire user
session before suspend, so that the encryption key for the home
directory (homed) can be removed from memory (until after resume,
after the user has typed the password in the screen lock again),
without the risk of having programs trying to access the now
inaccessible files in the home directory. So if qemu is frozen, but
the kvm kthread is not and somehow lets the VM or part of it still run
and use files (e.g.: write to the disk image that is stored in the
homedir) that are now inaccessible, that would be problematic for this
use case.
Powered by blists - more mailing lists