[PATCH 2/5] Dynamic ID addition doesn't need get_driver()

STALE5298d

4 messages, 4 authors, 2012-02-01 · open the first message on its own page

[PATCH 2/5] Dynamic ID addition doesn't need get_driver()

From: Alan Stern <stern@rowland.harvard.edu>
Date: 2012-01-24 18:34:45

As part of the removal of get_driver()/put_driver(), this patch
(as1511) changes all the places that add dynamic IDs for drivers.
Since these additions are done by writing to the drivers' sysfs
attribute files, and the attributes are removed when the drivers are
unregistered, there is no reason to take an extra reference to the
drivers.

The one exception is the pci-stub driver, which calls pci_add_dynid()
as part of its registration.  But again, there's no reason to take an
extra reference here, because the driver can't be unloaded while it is
being registered.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: Jiri Kosina <redacted>
CC: Jesse Barnes <redacted>
CC: Dominik Brodowski <linux@dominikbrodowski.net>

---

 drivers/hid/hid-core.c    |    6 +-----
 drivers/pci/pci-driver.c  |    2 --
 drivers/pcmcia/ds.c       |    5 +----
 drivers/usb/core/driver.c |    5 +----
 4 files changed, 3 insertions(+), 15 deletions(-)

Index: usb-3.3/drivers/hid/hid-core.c
===================================================================
--- usb-3.3.orig/drivers/hid/hid-core.c
+++ usb-3.3/drivers/hid/hid-core.c
@@ -1619,11 +1619,7 @@ static ssize_t store_new_id(struct devic
 	list_add_tail(&dynid->list, &hdrv->dyn_list);
 	spin_unlock(&hdrv->dyn_lock);
 
-	ret = 0;
-	if (get_driver(&hdrv->driver)) {
-		ret = driver_attach(&hdrv->driver);
-		put_driver(&hdrv->driver);
-	}
+	ret = driver_attach(&hdrv->driver);
 
 	return ret ? : count;
 }
Index: usb-3.3/drivers/pci/pci-driver.c
===================================================================
--- usb-3.3.orig/drivers/pci/pci-driver.c
+++ usb-3.3/drivers/pci/pci-driver.c
@@ -72,9 +72,7 @@ int pci_add_dynid(struct pci_driver *drv
 	list_add_tail(&dynid->node, &drv->dynids.list);
 	spin_unlock(&drv->dynids.lock);
 
-	get_driver(&drv->driver);
 	retval = driver_attach(&drv->driver);
-	put_driver(&drv->driver);
 
 	return retval;
 }
Index: usb-3.3/drivers/pcmcia/ds.c
===================================================================
--- usb-3.3.orig/drivers/pcmcia/ds.c
+++ usb-3.3/drivers/pcmcia/ds.c
@@ -127,10 +127,7 @@ pcmcia_store_new_id(struct device_driver
 	list_add_tail(&dynid->node, &pdrv->dynids.list);
 	mutex_unlock(&pdrv->dynids.lock);
 
-	if (get_driver(&pdrv->drv)) {
-		retval = driver_attach(&pdrv->drv);
-		put_driver(&pdrv->drv);
-	}
+	retval = driver_attach(&pdrv->drv);
 
 	if (retval)
 		return retval;
Index: usb-3.3/drivers/usb/core/driver.c
===================================================================
--- usb-3.3.orig/drivers/usb/core/driver.c
+++ usb-3.3/drivers/usb/core/driver.c
@@ -71,10 +71,7 @@ ssize_t usb_store_new_id(struct usb_dyni
 	list_add_tail(&dynid->node, &dynids->list);
 	spin_unlock(&dynids->lock);
 
-	if (get_driver(driver)) {
-		retval = driver_attach(driver);
-		put_driver(driver);
-	}
+	retval = driver_attach(driver);
 
 	if (retval)
 		return retval;

Re: [PATCH 2/5] Dynamic ID addition doesn't need get_driver()

From: Dominik Brodowski <linux@dominikbrodowski.net>
Date: 2012-01-25 19:56:09

On Tue, Jan 24, 2012 at 01:34:41PM -0500, Alan Stern wrote:
As part of the removal of get_driver()/put_driver(), this patch
(as1511) changes all the places that add dynamic IDs for drivers.
Since these additions are done by writing to the drivers' sysfs
attribute files, and the attributes are removed when the drivers are
unregistered, there is no reason to take an extra reference to the
drivers.

The one exception is the pci-stub driver, which calls pci_add_dynid()
as part of its registration.  But again, there's no reason to take an
extra reference here, because the driver can't be unloaded while it is
being registered.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: Jiri Kosina <redacted>
CC: Jesse Barnes <redacted>
CC: Dominik Brodowski <linux@dominikbrodowski.net>
For the PCMCIA part: Acked-by: Dominik Brodowski [off-list ref]

Thanks,
	Dominik

Re: [PATCH 2/5] Dynamic ID addition doesn't need get_driver()

From: Jesse Barnes <hidden>
Date: 2012-01-27 18:40:19

On Tue, 24 Jan 2012 13:34:41 -0500 (EST)
Alan Stern [off-list ref] wrote:
As part of the removal of get_driver()/put_driver(), this patch
(as1511) changes all the places that add dynamic IDs for drivers.
Since these additions are done by writing to the drivers' sysfs
attribute files, and the attributes are removed when the drivers are
unregistered, there is no reason to take an extra reference to the
drivers.

The one exception is the pci-stub driver, which calls pci_add_dynid()
as part of its registration.  But again, there's no reason to take an
extra reference here, because the driver can't be unloaded while it is
being registered.
Looks fine, you can add my ack too.

Thanks,
-- 
Jesse Barnes, Intel Open Source Technology Center

Re: [PATCH 2/5] Dynamic ID addition doesn't need get_driver()

From: Jiri Kosina <hidden>
Date: 2012-02-01 13:15:49

On Tue, 24 Jan 2012, Alan Stern wrote:
As part of the removal of get_driver()/put_driver(), this patch
(as1511) changes all the places that add dynamic IDs for drivers.
Since these additions are done by writing to the drivers' sysfs
attribute files, and the attributes are removed when the drivers are
unregistered, there is no reason to take an extra reference to the
drivers.

The one exception is the pci-stub driver, which calls pci_add_dynid()
as part of its registration.  But again, there's no reason to take an
extra reference here, because the driver can't be unloaded while it is
being registered.

Signed-off-by: Alan Stern <stern@rowland.harvard.edu>
CC: Dmitry Torokhov <dmitry.torokhov@gmail.com>
CC: Jiri Kosina <redacted>
CC: Jesse Barnes <redacted>
CC: Dominik Brodowski <linux@dominikbrodowski.net>
Acked-by: Jiri Kosina <redacted>

for the drivers/hid part. Thanks,

-- 
Jiri Kosina
SUSE Labs
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help