Skip to contentSkip to conversation

driver core: check bus->match without holding device lock(v2)

PATCH In review [email protected] 3 participants 4 messages Jan 21, 2009 – Apr 11, 2009 Driver core

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 driver core: check bus->match without holding device lock(v2)

From: Ming Lei <[email protected]>

This patch moves bus->match out from driver_probe_device and does not hold device lock to check the match between a device and a driver.

The idea has been verified by the commit 6cd495860901, which leads to a faster boot. But the commit 6cd495860901 has the following drawbacks: 1),only does the quick check in the path of __driver_attach->driver_probe_device, not in other paths; 2),for a matched device and driver, check the same match twice. It is a waste of cpu ,especially for some drivers with long device id table (eg. usb-storage driver).

This patch adds a helper of driver_match_device to check the match in all paths, and testes the match only once.

---
 drivers/base/base.h |    5 +++++
 drivers/base/bus.c  |    2 +-
 drivers/base/dd.c   |   19 +++++++------------
 3 files changed, 13 insertions(+), 13 deletions(-)
drivers/base/base.h +5 -0
index 0a5f055..e2599ba 100644
@@ -86,6 +86,11 @@ extern void bus_remove_driver(struct device_driver *drv);
 
 extern void driver_detach(struct device_driver *drv);
 extern int driver_probe_device(struct device_driver *drv, struct device *dev);
+static inline int driver_match_device(struct device_driver *drv,
+				      struct device *dev)
+{
+	return drv->bus->match && drv->bus->match(dev, drv);
+}
 
 extern void sysdev_shutdown(void);
 extern int sysdev_suspend(pm_message_t state);
drivers/base/bus.c +1 -1
index 83f32b8..8547b78 100644
@@ -198,7 +198,7 @@ static ssize_t driver_bind(struct device_driver *drv,
 	int err = -ENODEV;
 
 	dev = bus_find_device_by_name(bus, NULL, buf);
-	if (dev && dev->driver == NULL) {
+	if (dev && dev->driver == NULL && driver_match_device(drv, dev)) {
 		if (dev->parent)	/* Needed for USB */
 			down(&dev->parent->sem);
 		down(&dev->sem);
drivers/base/dd.c +7 -12
index 315bed8..9b721d3 100644
@@ -172,14 +172,8 @@ int driver_probe_done(void)
  * @drv: driver to bind a device to
  * @dev: device to try to bind to the driver
  *
- * First, we call the bus's match function, if one present, which should
- * compare the device IDs the driver supports with the device IDs of the
- * device. Note we don't do this ourselves because we don't know the
- * format of the ID structures, nor what is to be considered a match and
- * what is not.
- *
- * This function returns 1 if a match is found, -ENODEV if the device is
- * not registered, and 0 otherwise.
+ * This function returns -ENODEV if the device is not registered,
+ * 1 if the device is bound sucessfully and 0 otherwise.
  *
  * This function must be called with @dev->sem held.  When called for a
  * USB interface, @dev->parent->sem must be held as well.
@@ -190,21 +184,22 @@ int driver_probe_device(struct device_driver *drv, struct device *dev)
 
 	if (!device_is_registered(dev))
 		return -ENODEV;
-	if (drv->bus->match && !drv->bus->match(dev, drv))
-		goto done;
 
 	pr_debug("bus: '%s': %s: matched device %s with driver %s\n",
 		 drv->bus->name, __func__, dev_name(dev), drv->name);
 
 	ret = really_probe(dev, drv);
 
-done:
 	return ret;
 }
 
 static int __device_attach(struct device_driver *drv, void *data)
 {
 	struct device *dev = data;
+
+	if (!driver_match_device(drv, dev))
+		return 0;
+
 	return driver_probe_device(drv, dev);
 }
 
@@ -257,7 +252,7 @@ static int __driver_attach(struct device *dev, void *data)
 	 * is an error.
 	 */
 
-	if (drv->bus->match && !drv->bus->match(dev, drv))
+	if (!driver_match_device(drv, dev))
 		return 0;
 
 	if (dev->parent)	/* Needed for USB */
Signature · 1 line
1.6.0
Acked-by on v1 1/1
#1 lore raw
Quoted 17 lines · On Wed, 21 Jan 2009 23:27:47 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch moves bus->match out from driver_probe_device and does not hold device lock to check the match between a device and a driver.

The idea has been verified by the commit 6cd495860901, which leads to a faster boot. But the commit 6cd495860901 has the following drawbacks: 1),only does the quick check in the path of __driver_attach->driver_probe_device, not in other paths; 2),for a matched device and driver, check the same match twice. It is a waste of cpu ,especially for some drivers with long device id table (eg. usb-storage driver).

This patch adds a helper of driver_match_device to check the match in all paths, and testes the match only once.

Quoted 6 lines
---
 drivers/base/base.h |    5 +++++
 drivers/base/bus.c  |    2 +-
 drivers/base/dd.c   |   19 +++++++------------
 3 files changed, 13 insertions(+), 13 deletions(-)

The patch looks fine to me, and it seems to work well on my s390 LPAR.

#2 lore raw
In reply to [email protected]

Review

Quoted 5 lines · [email protected] writes:
(quoted diff) +5 -0
+static inline int driver_match_device(struct device_driver *drv,
+				      struct device *dev)
+{
+	return drv->bus->match && drv->bus->match(dev, drv);
+}

This logic is wrong. !(a && !b) <=> !a || b, not a && b

A bus that does not have a match function now never matches, whereas previously it would always match. This breaks aoa-soundbus.

Andreas.
---
driver core: Fix logic in driver_match_device

A bus that does not have a match function always matches. This fixes 49b420a13ff9 ("driver core: check bus->match without holding device lock").

---
 drivers/base/base.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/base/base.h +1 -1
index ddc9749..087e911 100644
@@ -115,7 +115,7 @@ extern int driver_probe_device(struct device_driver *drv, struct device *dev);
 static inline int driver_match_device(struct device_driver *drv,
 				      struct device *dev)
 {
-	return drv->bus->match && drv->bus->match(dev, drv);
+	return !drv->bus->match || drv->bus->match(dev, drv);
 }
 
 extern void sysdev_shutdown(void);
Signature · 3 lines
Andreas Schwab, [email protected]
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."
author
#3 lore raw

2009/4/10 Andreas Schwab <[email protected]>:

drivers/base/base.h extern int driver_probe_device(struct device_driver *drv, struct device *dev); · line 119

Quoted 11 lines
drivers/base/base.h +1 -1
index ddc9749..087e911 100644
@@ -115,7 +115,7 @@ extern int driver_probe_device(struct device_driver *drv, struct device *dev);
  static inline int driver_match_device(struct device_driver *drv,
                                      struct device *dev)
  {
-       return drv->bus->match && drv->bus->match(dev, drv);
+       return !drv->bus->match || drv->bus->match(dev, drv);
  }

A similar patch has been in greg's tree:

http://www.kernel.org/pub/linux/kernel/people/gregkh/gregkh-2.6/patches/driver-core.current/driver-core-fix-driver_match_device.patch

Thanks!

Signature · 1 line
Lei Ming