On Mon, Jun 27, 2022 at 01:02:49PM -0700, Stephen Boyd wrote: > Quoting Pavan Kondeti (2022-06-20 01:54:15) > > +Felipe, Bjorn > > > > On Thu, Jun 16, 2022 at 10:15:49AM -0700, Matthias Kaehlcke wrote: > > > On Thu, Jun 16, 2022 at 02:41:10PM +0530, Pavan Kondeti wrote: > > > > > > Good point! It doesn't really ensure that the child is probed (actually it > > > won't be probed and DL_FLAG_AUTOPROBE_CONSUMER doesn't make sense here), it > > > could happen that dwc3_qcom_probe() is deferred multiple times, but eventually > > > the PHYs should be ready and dwc3_probe() be invoked through > > > of_platform_populate(). > > > > This is a generic problem i.e if a parent can only proceed after the child > > devices are bounded (i.e probed successfully), how to ensure this behavior > > from the parent's probe? Since we can't block the parent probe (async probe is > > not the default behavior), we have to identify the condition that the children > > are deferring probe, so that parent also can do that. > > > > Can we add a API in drivers core to tell if a device probe is deferred or > > not? This can be done by testing list_empty(&dev->p->deferred_probe) under > > deferred_probe_mutex mutex. The parent can return EPROBE_DEFER based on this > > API return value. > > > > Another alternative would be explicitly checking if the child device suppliers > > are ready or not before adding child device. That would require decoupling > > of_platform_populate() to creating devices and adding devices. > > > > Note that this problem is not just limited to suppliers not ready. if the > > dwc3-qcom is made asynchronous probe, then its child also probed > > asynchronously and there is no guarantee that child would be probed by the > > time of_platform_populate() is returned. The bus notifier might come handy > > in this case. The parent can register for this notifier and waiting for > > the children device's BUS_NOTIFY_BOUND_DRIVER/BUS_NOTIFY_DRIVER_NOT_BOUND > > notifications. This would also work in our case, if we move to > > of_platform_populate() outside the probe(). > > > > Would like to hear other people thoughts on this. > > > > I'm not following very closely but it sounds like a problem that may be > solved by using the component driver code (see > include/linux/component.h). That would let you move anything that needs > to be done once the child devices probe to the aggregate driver 'bind' > function (see struct component_master_ops::bind). Thanks Stephen for letting us know about the component device framework. IIUC, - dwc3-qcom (parent of the dwc3 core) registers as a component master by calling component_master_add_with_match() before calling of_platform_populate(). The match callback could be as simple as comparing the device against our child device. - The dwc3 core (child) at the end of its probe can add as a component by calling component_add(). - The above triggers the component_master_ops::bind callback implemented in dwc3-qcom driver which signals that we are good to go. - The dwc-qcom can call component_bind_all() to finish the formality i.e telling the dwc3 core that we are good to go. Is my understanding correct? This is what we are looking for i.e a way for the child device(s) to signal the parent when the former is bounded. Also what happens when the child device probe fails for any reason. i.e component_add() would never be called so the master driver i.e dwc3-qcom would wait indefinitely. May be it needs to implement a timeout or runtime suspend etc should take care of keeping the resoures in suspend state. Thanks, Pavan
