diff options
Diffstat (limited to 'drivers/base/core.c')
-rw-r--r-- | drivers/base/core.c | 29 |
1 files changed, 28 insertions, 1 deletions
diff --git a/drivers/base/core.c b/drivers/base/core.c index 080e9ca11017..9cd3b5cfcc42 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c | |||
@@ -1551,7 +1551,34 @@ EXPORT_SYMBOL_GPL(device_destroy); | |||
1551 | * on the same device to ensure that new_name is valid and | 1551 | * on the same device to ensure that new_name is valid and |
1552 | * won't conflict with other devices. | 1552 | * won't conflict with other devices. |
1553 | * | 1553 | * |
1554 | * "Never use this function, bad things will happen" - gregkh | 1554 | * Note: Don't call this function. Currently, the networking layer calls this |
1555 | * function, but that will change. The following text from Kay Sievers offers | ||
1556 | * some insight: | ||
1557 | * | ||
1558 | * Renaming devices is racy at many levels, symlinks and other stuff are not | ||
1559 | * replaced atomically, and you get a "move" uevent, but it's not easy to | ||
1560 | * connect the event to the old and new device. Device nodes are not renamed at | ||
1561 | * all, there isn't even support for that in the kernel now. | ||
1562 | * | ||
1563 | * In the meantime, during renaming, your target name might be taken by another | ||
1564 | * driver, creating conflicts. Or the old name is taken directly after you | ||
1565 | * renamed it -- then you get events for the same DEVPATH, before you even see | ||
1566 | * the "move" event. It's just a mess, and nothing new should ever rely on | ||
1567 | * kernel device renaming. Besides that, it's not even implemented now for | ||
1568 | * other things than (driver-core wise very simple) network devices. | ||
1569 | * | ||
1570 | * We are currently about to change network renaming in udev to completely | ||
1571 | * disallow renaming of devices in the same namespace as the kernel uses, | ||
1572 | * because we can't solve the problems properly, that arise with swapping names | ||
1573 | * of multiple interfaces without races. Means, renaming of eth[0-9]* will only | ||
1574 | * be allowed to some other name than eth[0-9]*, for the aforementioned | ||
1575 | * reasons. | ||
1576 | * | ||
1577 | * Make up a "real" name in the driver before you register anything, or add | ||
1578 | * some other attributes for userspace to find the device, or use udev to add | ||
1579 | * symlinks -- but never rename kernel devices later, it's a complete mess. We | ||
1580 | * don't even want to get into that and try to implement the missing pieces in | ||
1581 | * the core. We really have other pieces to fix in the driver core mess. :) | ||
1555 | */ | 1582 | */ |
1556 | int device_rename(struct device *dev, const char *new_name) | 1583 | int device_rename(struct device *dev, const char *new_name) |
1557 | { | 1584 | { |