[PATCH] driver core: Fix device link deferred probe

[PATCH] driver core: Fix device link deferred probe

[Adrian Hunter]

A device probe deferred because of a device link is never probed again
because it is not added to the deferred_probe_pending_list. Add it, taking
care of the race with driver_deferred_probe_trigger().

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
---
 drivers/base/dd.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index ad44b40fe284..e4a5bef93344 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -350,6 +350,15 @@ int device_bind_driver(struct device *dev)
 static atomic_t probe_count = ATOMIC_INIT(0);
 static DECLARE_WAIT_QUEUE_HEAD(probe_waitqueue);
 
+static void driver_deferred_probe_add_trigger(struct device *dev,
+					      int local_trigger_count)
+{
+	driver_deferred_probe_add(dev);
+	/* Did a trigger occur while probing? Need to re-trigger if yes */
+	if (local_trigger_count != atomic_read(&deferred_trigger_count))
+		driver_deferred_probe_trigger();
+}
+
 static int really_probe(struct device *dev, struct device_driver *drv)
 {
 	int ret = -EPROBE_DEFER;
@@ -369,6 +378,8 @@ static int really_probe(struct device *dev, struct device_driver *drv)
 	}
 
 	ret = device_links_check_suppliers(dev);
+	if (ret == -EPROBE_DEFER)
+		driver_deferred_probe_add_trigger(dev, local_trigger_count);
 	if (ret)
 		return ret;
 
@@ -469,10 +480,7 @@ static int really_probe(struct device *dev, struct device_driver *drv)
 	case -EPROBE_DEFER:
 		/* Driver requested deferred probing */
 		dev_dbg(dev, "Driver %s requests probe deferral\n", drv->name);
-		driver_deferred_probe_add(dev);
-		/* Did a trigger occur while probing? Need to re-trigger if yes */
-		if (local_trigger_count != atomic_read(&deferred_trigger_count))
-			driver_deferred_probe_trigger();
+		driver_deferred_probe_add_trigger(dev, local_trigger_count);
 		break;
 	case -ENODEV:
 	case -ENXIO:
-- 
1.9.1

Re: [PATCH] driver core: Fix device link deferred probe

[Adrian Hunter]

On 02/11/17 11:33, Greg Kroah-Hartman wrote:
> On Thu, Nov 02, 2017 at 11:22:53AM +0200, Adrian Hunter wrote:
>> A device probe deferred because of a device link is never probed again
>> because it is not added to the deferred_probe_pending_list. Add it, taking
>> care of the race with driver_deferred_probe_trigger().
> 
> What hardware is triggering this?  Is this a new thing, or something
> that has always happened?

I just found it trying to use device links in new code.  As far as I know it
is not a problem that anyone else is experiencing yet.

Re: [PATCH] driver core: Fix device link deferred probe

[Greg Kroah-Hartman]

On Thu, Nov 02, 2017 at 11:22:53AM +0200, Adrian Hunter wrote:
> A device probe deferred because of a device link is never probed again
> because it is not added to the deferred_probe_pending_list. Add it, taking
> care of the race with driver_deferred_probe_trigger().

What hardware is triggering this?  Is this a new thing, or something
that has always happened?

thanks,

greg k-h

Re: [PATCH] driver core: Fix device link deferred probe

[Rafael J. Wysocki]

On Thursday, November 2, 2017 10:22:53 AM CET Adrian Hunter wrote:
> A device probe deferred because of a device link is never probed again
> because it is not added to the deferred_probe_pending_list. Add it, taking
> care of the race with driver_deferred_probe_trigger().
> 
> Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
> ---
>  drivers/base/dd.c | 16 ++++++++++++----
>  1 file changed, 12 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/base/dd.c b/drivers/base/dd.c
> index ad44b40fe284..e4a5bef93344 100644
> --- a/drivers/base/dd.c
> +++ b/drivers/base/dd.c
> @@ -350,6 +350,15 @@ int device_bind_driver(struct device *dev)
>  static atomic_t probe_count = ATOMIC_INIT(0);
>  static DECLARE_WAIT_QUEUE_HEAD(probe_waitqueue);
>  
> +static void driver_deferred_probe_add_trigger(struct device *dev,
> +					      int local_trigger_count)
> +{
> +	driver_deferred_probe_add(dev);
> +	/* Did a trigger occur while probing? Need to re-trigger if yes */
> +	if (local_trigger_count != atomic_read(&deferred_trigger_count))
> +		driver_deferred_probe_trigger();
> +}
> +
>  static int really_probe(struct device *dev, struct device_driver *drv)
>  {
>  	int ret = -EPROBE_DEFER;
> @@ -369,6 +378,8 @@ static int really_probe(struct device *dev, struct device_driver *drv)
>  	}
>  
>  	ret = device_links_check_suppliers(dev);
> +	if (ret == -EPROBE_DEFER)
> +		driver_deferred_probe_add_trigger(dev, local_trigger_count);
>  	if (ret)
>  		return ret;
>  
> @@ -469,10 +480,7 @@ static int really_probe(struct device *dev, struct device_driver *drv)
>  	case -EPROBE_DEFER:
>  		/* Driver requested deferred probing */
>  		dev_dbg(dev, "Driver %s requests probe deferral\n", drv->name);
> -		driver_deferred_probe_add(dev);
> -		/* Did a trigger occur while probing? Need to re-trigger if yes */
> -		if (local_trigger_count != atomic_read(&deferred_trigger_count))
> -			driver_deferred_probe_trigger();
> +		driver_deferred_probe_add_trigger(dev, local_trigger_count);
>  		break;
>  	case -ENODEV:
>  	case -ENXIO:
> 

Looks OK to me, thanks.

Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>