diff options
| author | Michael Ellerman <michael@ellerman.id.au> | 2007-09-14 15:33:13 +1000 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@suse.de> | 2007-10-12 15:03:17 -0700 |
| commit | a56bc69a182f501582557af7fad5bc882b1c856c (patch) | |
| tree | 052d752634d36fc4997c7224cd9a546497c19490 /drivers/pci | |
| parent | 5b1ea82fa245d4be6409cdba7e522ecb0b966b3c (diff) | |
| download | kernel_samsung_smdk4412-a56bc69a182f501582557af7fad5bc882b1c856c.zip kernel_samsung_smdk4412-a56bc69a182f501582557af7fad5bc882b1c856c.tar.gz kernel_samsung_smdk4412-a56bc69a182f501582557af7fad5bc882b1c856c.tar.bz2 | |
PCI: Fix incorrect argument order to list_add_tail() in PCI dynamic ID code
The code for dynamically assigning new ids to PCI drivers,
store_new_id(), calls list_add_tail() with the list head and new node
arguments in reversed order.
The result is that every new id written essentially overwrites the
previous list of ids.
Caught with the help of Rusty's "horribly bad" list_node patch:
http://lkml.org/lkml/2007/6/10/10
Signed-off-by: Michael Ellerman <michael@ellerman.id.au>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Diffstat (limited to 'drivers/pci')
| -rw-r--r-- | drivers/pci/pci-driver.c | 3 |
1 files changed, 1 insertions, 2 deletions
diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 004bc24..c43eced 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -54,7 +54,6 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count) if (!dynid) return -ENOMEM; - INIT_LIST_HEAD(&dynid->node); dynid->id.vendor = vendor; dynid->id.device = device; dynid->id.subvendor = subvendor; @@ -65,7 +64,7 @@ store_new_id(struct device_driver *driver, const char *buf, size_t count) driver_data : 0UL; spin_lock(&pdrv->dynids.lock); - list_add_tail(&pdrv->dynids.list, &dynid->node); + list_add_tail(&dynid->node, &pdrv->dynids.list); spin_unlock(&pdrv->dynids.lock); if (get_driver(&pdrv->driver)) { |