Skip to contentSkip to conversation

driver core: remove polling for driver_probe_done(v3)

PATCH In review [email protected] 4 participants 10 messages Jan 30, 2009 – Feb 17, 2009

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 driver core: remove polling for driver_probe_done(v3)

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Removing polling in fs initialization may lead to a faster boot.

---
 drivers/base/dd.c      |   12 ++++++++++++
 include/linux/device.h |    1 +
 init/do_mounts.c       |    4 ++--
 init/do_mounts_md.c    |    3 +--
 4 files changed, 16 insertions(+), 4 deletions(-)
drivers/base/dd.c +12 -0
index 9b721d3..8f9f6fb 100644
@@ -168,6 +168,18 @@ int driver_probe_done(void)
 }
 
 /**
+ * driver_wait_probe_done
+ * Wait until the probe sequence is finished.
+ *
+ */
+void driver_wait_probe_done(void)
+{
+	pr_debug("%s: probe_count = %d\n", __func__,
+		 atomic_read(&probe_count));
+	wait_event(probe_waitqueue, atomic_read(&probe_count) == 0);
+}
+
+/**
  * driver_probe_device - attempt to bind device & driver together
  * @drv: driver to bind a device to
  * @dev: device to try to bind to the driver
include/linux/device.h +1 -0
index 45e5b19..c53cdbf 100644
@@ -147,6 +147,7 @@ extern void put_driver(struct device_driver *drv);
 extern struct device_driver *driver_find(const char *name,
 					 struct bus_type *bus);
 extern int driver_probe_done(void);
+extern void driver_wait_probe_done(void);
 
 /* sysfs interface for exporting driver attributes */
 
init/do_mounts.c +2 -2
index 708105e..9cb1985 100644
@@ -371,8 +371,8 @@ void __init prepare_namespace(void)
 	}
 
 	/* wait for the known devices to complete their probing */
-	while (driver_probe_done() != 0)
-		msleep(100);
+	driver_wait_probe_done();
+
 	async_synchronize_full();
 
 	md_run_setup();
init/do_mounts_md.c +1 -2
index ff95e31..63b77f0 100644
@@ -281,8 +281,7 @@ static void __init autodetect_raid(void)
 	 */
 	printk(KERN_INFO "md: Waiting for all devices to be available before autodetect\n");
 	printk(KERN_INFO "md: If you don't use raid, use raid=noautodetect\n");
-	while (driver_probe_done() < 0)
-		msleep(100);
+	driver_wait_probe_done();
 	fd = sys_open("/dev/md0", 0, 0);
 	if (fd >= 0) {
 		sys_ioctl(fd, RAID_AUTORUN, raid_autopart);
Signature · 1 line
1.6.0
#1 lore raw
Quoted 14 lines · On Fri, 30 Jan 2009 20:48:35 +0800, [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Removing polling in fs initialization may lead to a faster boot.

Quoted 7 lines
---
 drivers/base/dd.c      |   12 ++++++++++++
 include/linux/device.h |    1 +
 init/do_mounts.c       |    4 ++--
 init/do_mounts_md.c    |    3 +--
 4 files changed, 16 insertions(+), 4 deletions(-)

Looks good and also survived a quick test on my s390 LPAR.

Reviewed-by on v1 1/1
#2 lore raw
In reply to [email protected]
Quoted 11 lines · On Fri, 30 Jan 2009 20:48:35 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Quoted 4 lines

Removing polling in fs initialization may lead to a faster boot.

ok this one will work fine ;)

thanks for fixing this up
(it will clash with the patch that moves this to one place, but the same cleanup can be done there)

Signature · 3 lines
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org
#3 lore raw
Quoted 19 lines · Arjan van de Ven
Quoted 11 lines · On Fri, 30 Jan 2009 20:48:35 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Quoted 4 lines

Removing polling in fs initialization may lead to a faster boot.

Quoted 7 lines

ok this one will work fine ;)

thanks for fixing this up
(it will clash with the patch that moves this to one place, but the same cleanup can be done there)

What patch is that? Should i be taking this? Or are they in some other tree?

thanks,

greg k-h

#4 lore raw
Quoted 22 lines · On Wed, 4 Feb 2009 15:08:51 -0800 Greg KH <[email protected]> wrote:
Quoted 21 lines · Arjan van de Ven
Quoted 11 lines · On Fri, 30 Jan 2009 20:48:35 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Quoted 4 lines

Removing polling in fs initialization may lead to a faster boot.

ok this one will work fine ;)

Quoted 8 lines
Quoted 5 lines

thanks for fixing this up
(it will clash with the patch that moves this to one place, but the same cleanup can be done there)

What patch is that? Should i be taking this? Or are they in some other tree?

it's the suspend-resume regression fixes that either you already have or that raphael sent to you

Quoted 4 lines

thanks,

greg k-h

Signature · 3 lines
Arjan van de Ven 	Intel Open Source Technology Centre
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org
#5 lore raw
Quoted 31 lines · Arjan van de Ven
Quoted 28 lines · On Wed, 4 Feb 2009 15:08:51 -0800 Greg KH <[email protected]> wrote:
Quoted 26 lines · Arjan van de Ven
Quoted 11 lines · On Fri, 30 Jan 2009 20:48:35 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Quoted 4 lines

Removing polling in fs initialization may lead to a faster boot.

ok this one will work fine ;)

thanks for fixing this up
(it will clash with the patch that moves this to one place, but the same cleanup can be done there)

Quoted 5 lines
Quoted 2 lines

What patch is that? Should i be taking this? Or are they in some other tree?

it's the suspend-resume regression fixes that either you already have or that raphael sent to you

Hm, ok, I don't see any left in my tree, or queue, so perhaps someone might want to resend them to me?

thanks,

greg k-h

author
#6 lore raw

Hi, Greg
Would you mind queuing this patch into your tree?
Thanks!

2009/2/6 Greg KH <[email protected]>:

Quoted 45 lines · Arjan van de Ven
Quoted 31 lines
Quoted 28 lines · On Wed, 4 Feb 2009 15:08:51 -0800 Greg KH <[email protected]> wrote:
Quoted 26 lines · Arjan van de Ven
Quoted 11 lines · On Fri, 30 Jan 2009 20:48:35 +0800 [email protected] wrote:

From: Ming Lei <[email protected]>

This patch adds a function : driver_wait_probe_done, which waits on condition of probing done to replace polling for driver_probe_done in fs initialization.

There is no better way to avoid polling for driver_probe_done _and_ existence of the root device, so we does not replace the driver_probe_done with driver_wait_probe_done in such special case.

Quoted 4 lines

Removing polling in fs initialization may lead to a faster boot.

ok this one will work fine ;)

thanks for fixing this up
(it will clash with the patch that moves this to one place, but the same cleanup can be done there)

Quoted 5 lines
Quoted 2 lines

What patch is that? Should i be taking this? Or are they in some other tree?

it's the suspend-resume regression fixes that either you already have or that raphael sent to you

Hm, ok, I don't see any left in my tree, or queue, so perhaps someone might want to resend them to me?

thanks,

greg k-h

Signature · 1 line
Lei Ming
#7 lore raw
Quoted 2 lines · Ming Lei

Hi, Greg
Would you mind queuing this patch into your tree?

What patch? Did anyone resend it to me like I asked?

confused,

greg k-h

author
#8 lore raw

2009/2/17 Greg KH <[email protected]>:

Quoted 5 lines · Ming Lei
Quoted 2 lines

Hi, Greg
Would you mind queuing this patch into your tree?

What patch? Did anyone resend it to me like I asked?

I mean the patch in the mail list,with the following subject:

[PATCH] driver core: remove polling for driver_probe_done(v3)

Thanks!

Quoted 5 lines

confused,

greg k-h

Signature · 1 line
Lei Ming
#9 lore raw
Quoted 10 lines · Ming Lei

2009/2/17 Greg KH <[email protected]>:

Quoted 5 lines · Ming Lei
Quoted 2 lines

Hi, Greg
Would you mind queuing this patch into your tree?

What patch? Did anyone resend it to me like I asked?

I mean the patch in the mail list,with the following subject:

[PATCH] driver core: remove polling for driver_probe_done(v3)

Care to resend it?