Re: [PATCH V9 06/17] monitor: refactor monitor_data_init() to pass ID

["Daniel P. Berrangé" <berrange@redhat.com>]

On Mon, Jun 29, 2026 at 04:31:56PM +0800, Zhang Chen wrote:
> On Fri, Jun 26, 2026 at 9:32 PM Daniel P. Berrangé <berrange@redhat.com> wrote:
> >
> > On Wed, Jun 24, 2026 at 03:08:40PM +0800, Zhang Chen wrote:
> > > Anonymous monitors (e.g., '-monitor stdio' or GDB stub) leave their 'id'
> > > field as NULL. To allow downstream subsystems (like iothread tracking) to
> > > identify these instances later, the initialization paths must propagate
> > > the context.
> > >
> > > Refactor monitor_data_init() to accept the 'id' parameters, and
> > > update all internal QMP/HMP paths to pass these identifiers.
> > >
> > > For anonymous monitors, 'mon->id' remains NULL at this stage to avoid
> > > namespace pollution, setting up infrastructure for subsequent patches.
> > >
> > > Signed-off-by: Zhang Chen <zhangckid@gmail.com>
> > > ---
> > >  chardev/char.c             | 2 +-
> > >  gdbstub/system.c           | 3 ++-
> > >  include/monitor/monitor.h  | 5 +++--
> > >  monitor/hmp.c              | 5 +++--
> > >  monitor/monitor-internal.h | 4 +++-
> > >  monitor/monitor.c          | 9 ++++++---
> > >  monitor/qmp-cmds.c         | 2 +-
> > >  monitor/qmp.c              | 5 +++--
> > >  stubs/monitor-internal.c   | 3 ++-
> > >  9 files changed, 24 insertions(+), 14 deletions(-)
> >
> > This patch will significantly conflict with my series that
> > creates QOM objects for the monitors. They will all be
> > guaranteed to have an "id" at that point though, which
> > addresses the problem you're trying to solve IIUC.
> >
> > https://lists.gnu.org/archive/html/qemu-devel/2026-06/msg06427.html
> >
> 
> Thank you for your updates.
> As Markus's help, we already know your series for create
> QOM for the monitors. So this series V7 based on the assumption of
> your patch merged. But for the comments:
> "Feels like this patch should not be merged as is."
> https://lists.gnu.org/archive/html/qemu-devel/2026-05/msg04214.html
> 
> I have to fix it by myself.
> I'm very happy this series based on your series, I can conveniently make
> the monitor as a regular QOM object like others in this series.
> If you think that OK, I will rebase my series on your QOM for monitor patches
> in next version.

FYI, my monitor hotplug series is merged, so you can simply
rebase your work onto current git HEAD.


With regards,
Daniel
-- 
|: https://berrange.com       ~~        https://hachyderm.io/@berrange :|
|: https://libvirt.org          ~~          https://entangle-photo.org :|
|: https://pixelfed.art/berrange   ~~    https://fstop138.berrange.com :|