Re: [RFC PATCH 1/2] driver core: add new dl device status DL_DEV_PROBE_FAILED

From: Wang, Dongsheng
Date: Thu Mar 29 2018 - 07:24:56 EST


On Thu, 2018-03-29 at 12:51 +0200, Rafael J. Wysocki wrote:
> On Thu, Mar 29, 2018 at 12:00 PM, Wang Dongsheng
> <dongsheng.wang@xxxxxxxxxxxxxxxx> wrote:
> > Currently the initialization state of device is DL_DEV_NO_DRIVER.
> > The problem is, after probe failure the state will also be set to
> > DL_DEV_NO_DRIVER as well. And the device is not linked, it has no
> > supplier or consumer. Thus adding a new state to distinguish
> > probe failure and not-probed-yet.
> >
> > Signed-off-by: Wang Dongsheng <dongsheng.wang@xxxxxxxxxxxxxxxx>
>
> I guess what you want is a cleanup after a failing probe, but after
> that the state really is "no driver" again, isn't it?
Yes, agree "no driver", device is never get a driver again.

But A depend on B successful probe. if B failed, A will never know:
1. B can't work.
2. B hasn't got a probe yet.

Like IOMMU. After SMMU successful probed, the driver add a resource
into "iommu_device_list". Master lookup the corresponding SMMU fwnode
from "iommu_device_list", after matched master will do probe. but if
the list is NULL master will get -EPROBE_DEFER, means SMMU device may
not probe yet, in fact SMMU may probe failed.

I try to use DL_DEV state to fix this issue, but NO_DRIVER does not
distinguish between the two cases.

Cheers,
-Dongsheng

>
> > ---
> > Documentation/driver-api/device_link.rst | 2 +-
> > drivers/base/base.h | 2 +-
> > drivers/base/core.c | 22
> > ++++++++++++++++++++--
> > drivers/base/dd.c | 2 +-
> > include/linux/device.h | 1 +
> > 5 files changed, 24 insertions(+), 5 deletions(-)
> >
> > diff --git a/Documentation/driver-api/device_link.rst
> > b/Documentation/driver-api/device_link.rst
> > index 70e328e..9054403 100644
> > --- a/Documentation/driver-api/device_link.rst
> > +++ b/Documentation/driver-api/device_link.rst
> > @@ -247,7 +247,7 @@ State machine
> > :c:func:`device_links_unbind_consumers()`.)
> >
> > * If the probe fails, links to suppliers revert back to
> > ``DL_STATE_AVAILABLE``.
> > - (Call to :c:func:`device_links_no_driver()` from
> > :c:func:`really_probe()`.)
> > + (Call to :c:func:`device_links_probe_failed()` from
> > :c:func:`really_probe()`.)
> >
> > * If the probe succeeds, links to suppliers progress to
> > ``DL_STATE_ACTIVE``.
> > (Call to :c:func:`device_links_driver_bound()` from
> > :c:func:`driver_bound()`.)
> > diff --git a/drivers/base/base.h b/drivers/base/base.h
> > index d800de6..f9931d9 100644
> > --- a/drivers/base/base.h
> > +++ b/drivers/base/base.h
> > @@ -158,6 +158,6 @@ extern void device_links_read_unlock(int idx);
> > extern int device_links_check_suppliers(struct device *dev);
> > extern void device_links_driver_bound(struct device *dev);
> > extern void device_links_driver_cleanup(struct device *dev);
> > -extern void device_links_no_driver(struct device *dev);
> > +extern void device_links_probe_failed(struct device *dev);
> > extern bool device_links_busy(struct device *dev);
> > extern void device_links_unbind_consumers(struct device *dev);
> > diff --git a/drivers/base/core.c b/drivers/base/core.c
> > index 5847364..31d4f68 100644
> > --- a/drivers/base/core.c
> > +++ b/drivers/base/core.c
> > @@ -452,10 +452,28 @@ static void __device_links_no_driver(struct
> > device *dev)
> > dev->links.status = DL_DEV_NO_DRIVER;
> > }
> >
> > -void device_links_no_driver(struct device *dev)
> > +static void __device_links_probe_failed(struct device *dev)
> > +{
> > + struct device_link *link, *ln;
> > +
> > + list_for_each_entry_safe_reverse(link, ln, &dev-
> > >links.suppliers,
> > + c_node) {
> > + if (link->flags & DL_FLAG_STATELESS)
> > + continue;
> > +
> > + if (link->flags & DL_FLAG_AUTOREMOVE)
> > + __device_link_del(link);
> > + else if (link->status != DL_STATE_SUPPLIER_UNBIND)
> > + WRITE_ONCE(link->status,
> > DL_STATE_AVAILABLE);
> > + }
> > +
> > + dev->links.status = DL_DEV_PROBE_FAILED;
> > +}
> > +
> > +void device_links_probe_failed(struct device *dev)
> > {
> > device_links_write_lock();
> > - __device_links_no_driver(dev);
> > + __device_links_probe_failed(dev);
> > device_links_write_unlock();
> > }
> >
> > diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> > index de6fd09..90d57e0 100644
> > --- a/drivers/base/dd.c
> > +++ b/drivers/base/dd.c
> > @@ -492,7 +492,7 @@ static int really_probe(struct device *dev,
> > struct device_driver *drv)
> > blocking_notifier_call_chain(&dev->bus->p-
> > >bus_notifier,
> > BUS_NOTIFY_DRIVER_NOT_
> > BOUND, dev);
> > pinctrl_bind_failed:
> > - device_links_no_driver(dev);
> > + device_links_probe_failed(dev);
> > devres_release_all(dev);
> > driver_sysfs_remove(dev);
> > dev->driver = NULL;
> > diff --git a/include/linux/device.h b/include/linux/device.h
> > index b093405..bf9630a 100644
> > --- a/include/linux/device.h
> > +++ b/include/linux/device.h
> > @@ -794,6 +794,7 @@ struct device_link {
> > enum dl_dev_state {
> > DL_DEV_NO_DRIVER = 0,
> > DL_DEV_PROBING,
> > + DL_DEV_PROBE_FAILED,
> > DL_DEV_DRIVER_BOUND,
> > DL_DEV_UNBINDING,
> > };
> > --
> > 2.7.4
> >
> > --
> > To unsubscribe from this list: send the line "unsubscribe linux-
> > acpi" in
> > the body of a message to majordomo@xxxxxxxxxxxxxxx
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-acpi"
> in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html