Re: [PATCH v9 4/4] vhost: Add a KConfig knob to enable IOCTL VHOST_FORK_FROM_OWNER

["Michael S. Tsirkin" <mst@redhat.com>]

On Fri, May 16, 2025 at 09:31:42AM +0800, Jason Wang wrote:
> On Thu, May 15, 2025 at 2:14 PM Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > On Wed, May 14, 2025 at 10:52:58AM +0800, Jason Wang wrote:
> > > On Tue, May 13, 2025 at 3:09 PM Michael S. Tsirkin <mst@redhat.com> wrote:
> > > >
> > > > On Tue, May 13, 2025 at 12:08:51PM +0800, Jason Wang wrote:
> > > > > On Wed, Apr 30, 2025 at 5:27 PM Michael S. Tsirkin <mst@redhat.com> wrote:
> > > > > >
> > > > > > On Wed, Apr 30, 2025 at 11:34:49AM +0800, Jason Wang wrote:
> > > > > > > On Tue, Apr 29, 2025 at 6:56 PM Michael S. Tsirkin <mst@redhat.com> wrote:
> > > > > > > >
> > > > > > > > On Tue, Apr 29, 2025 at 11:39:37AM +0800, Jason Wang wrote:
> > > > > > > > > On Mon, Apr 21, 2025 at 11:46 AM Jason Wang <jasowang@redhat.com> wrote:
> > > > > > > > > >
> > > > > > > > > > On Mon, Apr 21, 2025 at 11:45 AM Jason Wang <jasowang@redhat.com> wrote:
> > > > > > > > > > >
> > > > > > > > > > > On Mon, Apr 21, 2025 at 10:45 AM Cindy Lu <lulu@redhat.com> wrote:
> > > > > > > > > > > >
> > > > > > > > > > > > Introduce a new config knob `CONFIG_VHOST_ENABLE_FORK_OWNER_IOCTL`,
> > > > > > > > > > > > to control the availability of the `VHOST_FORK_FROM_OWNER` ioctl.
> > > > > > > > > > > > When CONFIG_VHOST_ENABLE_FORK_OWNER_IOCTL is set to n, the ioctl
> > > > > > > > > > > > is disabled, and any attempt to use it will result in failure.
> > > > > > > > > > >
> > > > > > > > > > > I think we need to describe why the default value was chosen to be false.
> > > > > > > > > > >
> > > > > > > > > > > What's more, should we document the implications here?
> > > > > > > > > > >
> > > > > > > > > > > inherit_owner was set to false: this means "legacy" userspace may
> > > > > > > > > >
> > > > > > > > > > I meant "true" actually.
> > > > > > > > >
> > > > > > > > > MIchael, I'd expect inherit_owner to be false. Otherwise legacy
> > > > > > > > > applications need to be modified in order to get the behaviour
> > > > > > > > > recovered which is an impossible taks.
> > > > > > > > >
> > > > > > > > > Any idea on this?
> > > > > > > > >
> > > > > > > > > Thanks
> > > > > >
> > > > > > So, let's say we had a modparam? Enough for this customer?
> > > > > > WDYT?
> > > > >
> > > > > Just to make sure I understand the proposal.
> > > > >
> > > > > Did you mean a module parameter like "inherit_owner_by_default"? I
> > > > > think it would be fine if we make it false by default.
> > > > >
> > > > > Thanks
> > > >
> > > > I think we should keep it true by default, changing the default
> > > > risks regressing what we already fixes.
> > >
> > > I think it's not a regression since it comes since the day vhost is
> > > introduced. To my understanding the real regression is the user space
> > > noticeable behaviour changes introduced by vhost thread.
> > >
> > > > The specific customer can
> > > > flip the modparam and be happy.
> > >
> > > If you stick to the false as default, I'm fine.
> > >
> > > Thanks
> >
> > That would be yet another behaviour change.
> 
> Back to the original behaviour.

yes but the original was also a bugfix.

> > I think one was enough, don't you think?
> 
> I think such kind of change is unavoidable if we want to fix the
> usersapce behaviour change.
> 
> Thanks

I feel it is too late to "fix". the new behaviour is generally ok, and I
feel the right thing so to give management control knobs do pick the
desired behaviour.
And really modparam is wrong here because different userspace
can have different requirements, and in ~10 years I want to see us
disable the legacy behaviour altogether.
But given your time constraints, a modparam knob as a quick workaround
for the specific customer is kind of not very terrible.

> >
> >
> > > >
> > > > > >
> > > > > > --
> > > > > > MST
> > > > > >
> > > >
> >