[PATCH 6/8] i3c: master: Defer new-device registration out of DAA caller context

[Frank Li]

On Fri, May 22, 2026 at 07:52:17AM +0300, Adrian Hunter wrote:
> On 21/05/2026 21:32, Frank Li wrote:
> > On Fri, May 15, 2026 at 07:42:20PM +0300, Adrian Hunter wrote:
> >> On 13/05/2026 22:03, Frank Li wrote:
> >>> On Wed, May 13, 2026 at 08:45:55AM +0300, Adrian Hunter wrote:
> >>>> On 12/05/2026 19:39, Frank Li wrote:
> >>>>> On Tue, May 12, 2026 at 03:17:30PM +0300, Adrian Hunter wrote:
> >>>>>> Master drivers may invoke i3c_master_do_daa_ext() during resume to
> >>>>>> re-run Dynamic Address Assignment.  As well as assigning addresses to
> >>>>>> any newly arrived devices, this restores the dynamic address of devices
> >>>>>> that lost it across system suspend, so it has to run as part of the
> >>>>>> controller's resume path.
> >>>>>>
> >>>>>> A side effect of i3c_master_do_daa_ext() today is that it also
> >>>>>> registers any newly discovered I3C devices with the driver model
> >>>>>> inline, via i3c_master_register_new_i3c_devs().  Doing that from the
> >>>>>> resume path is problematic: a hot-join-capable device may join the bus
> >>>>>> during this same DAA, and registering it immediately would push driver
> >>>>>> model work (probing, sysfs, etc.) into the controller's resume context,
> >>>>>> where the rest of the system is not yet fully resumed and the
> >>>>>> controller driver is still partway through its own resume sequence.
> >>>>>>
> >>>>>> Decouple discovery from registration: add a reg_work work item to
> >>>>>> struct i3c_master_controller and have i3c_master_do_daa_ext() queue it
> >>>>>> on master->wq (the freezable workqueue) instead of calling
> >>>>>> i3c_master_register_new_i3c_devs() directly.  The worker performs the
> >>>>>> registration only when the controller is not shutting_down, and is
> >>>>>> cancelled alongside hj_work in i3c_master_shutdown().  Because wq is
> >>>>>> freezable, any newly observed devices end up being registered after
> >>>>>> the system has finished resuming.
> >>>>>>
> >>>>>> i3c_master_register() also routes its initial post-bus-init registration
> >>>>>> through reg_work, using flush_work() to keep probe-time behavior
> >>>>>> synchronous.  This keeps a single registration code path and ensures the
> >>>>>> worker is the only writer of desc->dev.
> >>>>>
> >>>>> why not direct use hj_work?
> >>>>
> >>>> i3c_master_register_new_i3c_devs() use of desc->dev is racy, so
> >>>> i3c_master_register_new_i3c_devs() must not be allowed to race
> >>>> with itself.  Having it only ever run via reg_work achieves that.
> >>>
> >>> Sorry, I have not understand these, Can provide some detail?
> >>
> >> >From i3c_master_register_new_i3c_devs():
> >>
> >> 	i3c_bus_for_each_i3cdev(&master->bus, desc) {
> >> 		if (desc->dev || !desc->info.dyn_addr || desc == master->this)
> >> 			continue;
> >>
> >> 		desc->dev = kzalloc_obj(*desc->dev);
> >> 		...
> >> 		ret = device_register(&desc->dev->dev);
> >>
> >> This is done under the shared i3c_bus_normaluse_lock(), so there can
> >
> > i3c_bus_normaluse_lock() may is wrong, suppose it should be
> > i3c_bus_maintenance_lock(), register new devices change i3c bus's
> > hierarchical structure.
>
> If device_register() probes the device and the probe tries to
> access the device, won't it deadlock if i3c_bus_maintenance_lock()
> is held.

Okay,

Reviewed-by: Frank Li <Frank.Li at nxp.com>

>
> >
> > Frank
> >
> >> be 2 or more instances of i3c_master_register_new_i3c_devs() running
> >> at the same time.  They might all see desc->dev is NULL and then all
> >> of them try to initialize and register a dev for the same I3C device.
> >>
>