Hi All,

On 2016-09-19 23:45, Tobias Jakobi wrote:
I did some tests with the new version today. Sadly the reboot/shutdown
issues are still present.

Thanks for the report. I've managed to reproduce this issue and it is again
caused by modifying device on devices_kset list before it will be finally
added by device_add(). I thought that the new patchset allows creating links
to a device, which has not been yet added to system device list.

What is your opinion? Should it be allowed to create a link to device, which
has not yet been added to system device list by device_add()? My code used to do that, because IOMMUs are configured from of_platform_device_create_pdata()
of_dma_configure() of_iommu_configure(), which happens before device_add().

To solve the reported corruption of devices_kset list, following change is

diff --git a/drivers/base/core.c b/drivers/base/core.c
index aa8196508db9..4542ba9f60d4 100644
--- a/drivers/base/core.c
+++ b/drivers/base/core.c
@@ -1039,11 +1039,18 @@ static void devices_kset_move_after(struct device *deva, struct device *devb)
 void devices_kset_move_last(struct device *dev)
+       struct device *d;
        if (!devices_kset)
pr_debug("devices_kset: Moving %s to end of list\n", dev_name(dev));
-       list_move_tail(&dev->kobj.entry, &devices_kset->list);
+       list_for_each_entry(d, &devices_kset->list, kobj.entry) {
+               if (d == dev) {
+ list_move_tail(&dev->kobj.entry, &devices_kset->list);
+                       break;
+               }
+       }

If you think that links can be created only to a device, which has been fully added to the system, I will register a bus notifier and create a link on consumers
device probe then.


Best regards
Marek Szyprowski, PhD
Samsung R&D Institute Poland

Reply via email to