[PATCH] Move pci_dev_put outside a spinlock

[Alan Stern <stern@rowland.harvard.edu>]

Greg:

This patch (as659) fixes a might_sleep problem in the PCI core, by moving 
a call to pci_dev_put() outside the scope of a spinlock.

Alan Stern



Signed-off-by: Alan Stern <stern@rowland.harvard.edu>

---

Maybe this has already been fixed; if so the fix isn't in your 
development tree for 2.6.16-rc4.

Index: usb-2.6/drivers/pci/search.c
===================================================================
--- usb-2.6.orig/drivers/pci/search.c
+++ usb-2.6/drivers/pci/search.c
@@ -246,9 +246,9 @@ pci_get_subsys(unsigned int vendor, unsi
 	}
 	dev = NULL;
 exit:
-	pci_dev_put(from);
 	dev = pci_dev_get(dev);
 	spin_unlock(&pci_bus_lock);
+	pci_dev_put(from);
 	return dev;
 }
 

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/