From: Julia Lawall <[EMAIL PROTECTED]>
pci_get_class implicitly does a pci_dev_put on its second argument, so
pci_dev_put is only needed if there is a break out of the loop.
The semantic match detecting this problem is as follows:
// <smpl>
@@
expression dev;
expression E;
@@
* pci_dev_put(dev)
... when != dev = E
(
* pci_get_device(...,dev)
|
* pci_get_device_reverse(...,dev)
|
* pci_get_subsys(...,dev)
|
* pci_get_class(...,dev)
)
// </smpl>
Signed-off-by: Julia Lawall <[EMAIL PROTECTED]>
---
diff -up a/drivers/char/agp/amd-k7-agp.c b/drivers/char/agp/amd-k7-agp.c
--- a/drivers/char/agp/amd-k7-agp.c 2007-11-15 07:33:31.000000000 +0100
+++ b/drivers/char/agp/amd-k7-agp.c 2007-11-15 07:40:48.000000000 +0100
@@ -436,10 +436,6 @@ static int __devinit agp_amdk7_probe(str
return -ENODEV;
}
cap_ptr = pci_find_capability(gfxcard, PCI_CAP_ID_AGP);
- if (!cap_ptr) {
- pci_dev_put(gfxcard);
- continue;
- }
}
/* With so many variants of NVidia cards, it's simpler just
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
