[PATCH v4 1/3] acpi : cpu hot-remove returns error when cpu_down() fails

[PATCH v4 1/3] acpi : cpu hot-remove returns error when cpu_down() fails

[Yasuaki Ishimatsu]

Even if cpu_down() fails, acpi_processor_remove() continues to remove the cpu.
But in this case, it should return error number since some process may run on
the cpu. If the cpu has a running process and the cpu is turned the power off,
the system may not work well.

Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/processor_driver.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-08 09:23:56.000000000 +0900
+++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 15:11:06.135541317 +0900
@@ -610,7 +610,7 @@ err_free_pr:
 static int acpi_processor_remove(struct acpi_device *device, int type)
 {
 	struct acpi_processor *pr = NULL;
-
+	int ret;

 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
@@ -621,8 +621,9 @@ static int acpi_processor_remove(struct
 		goto free;

 	if (type == ACPI_BUS_REMOVAL_EJECT) {
-		if (acpi_processor_handle_eject(pr))
-			return -EINVAL;
+		ret = acpi_processor_handle_eject(pr);
+		if (ret)
+			return ret;
 	}

 	acpi_processor_power_exit(pr, device);
@@ -841,12 +842,17 @@ static acpi_status acpi_processor_hotadd

 static int acpi_processor_handle_eject(struct acpi_processor *pr)
 {
-	if (cpu_online(pr->id))
-		cpu_down(pr->id);
+	int ret = 0;
+
+	if (cpu_online(pr->id)) {
+		ret = cpu_down(pr->id);
+		if (ret)
+			return ret;
+	}

 	arch_unregister_cpu(pr->id);
 	acpi_unmap_lsapic(pr->id);
-	return (0);
+	return ret;
 }
 #else
 static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)

[PATCH v4 1/3] acpi : cpu hot-remove returns error when cpu_down() fails

[Yasuaki Ishimatsu]

Even if cpu_down() fails, acpi_processor_remove() continues to remove the cpu.
But in this case, it should return error number since some process may run on
the cpu. If the cpu has a running process and the cpu is turned the power off,
the system may not work well.

Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/processor_driver.c |   18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-08 09:23:56.000000000 +0900
+++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 15:11:06.135541317 +0900
@@ -610,7 +610,7 @@ err_free_pr:
 static int acpi_processor_remove(struct acpi_device *device, int type)
 {
 	struct acpi_processor *pr = NULL;
-
+	int ret;

 	if (!device || !acpi_driver_data(device))
 		return -EINVAL;
@@ -621,8 +621,9 @@ static int acpi_processor_remove(struct
 		goto free;

 	if (type == ACPI_BUS_REMOVAL_EJECT) {
-		if (acpi_processor_handle_eject(pr))
-			return -EINVAL;
+		ret = acpi_processor_handle_eject(pr);
+		if (ret)
+			return ret;
 	}

 	acpi_processor_power_exit(pr, device);
@@ -841,12 +842,17 @@ static acpi_status acpi_processor_hotadd

 static int acpi_processor_handle_eject(struct acpi_processor *pr)
 {
-	if (cpu_online(pr->id))
-		cpu_down(pr->id);
+	int ret = 0;
+
+	if (cpu_online(pr->id)) {
+		ret = cpu_down(pr->id);
+		if (ret)
+			return ret;
+	}

 	arch_unregister_cpu(pr->id);
 	acpi_unmap_lsapic(pr->id);
-	return (0);
+	return ret;
 }
 #else
 static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)

[PATCH v4 2/3] acpi : prevent cpu from becoming online

[Yasuaki Ishimatsu]

Even if acpi_processor_handle_eject() offlines cpu, there is a chance
to online the cpu after that. So the patch closes the window by using
get/put_online_cpus().

Why does the patch change _cpu_up() logic?

The patch cares the race of hot-remove cpu and _cpu_up(). If the patch
does not change it, there is the following race.

hot-remove cpu                         |  _cpu_up()
------------------------------------- ------------------------------------
call acpi_processor_handle_eject()     |
     call cpu_down()                   |
     call get_online_cpus()            |
                                       | call cpu_hotplug_begin() and stop here
     call arch_unregister_cpu()        |
     call acpi_unmap_lsapic()          |
     call put_online_cpus()            |
                                       | start and continue _cpu_up()
     return acpi_processor_remove()    |
continue hot-remove the cpu            |

So _cpu_up() can continue to itself. And hot-remove cpu can also continue
itself. If the patch changes _cpu_up() logic, the race disappears as below:

hot-remove cpu                         | _cpu_up()
-----------------------------------------------------------------------
call acpi_processor_handle_eject()     |
     call cpu_down()                   |
     call get_online_cpus()            |
                                       | call cpu_hotplug_begin() and stop here
     call arch_unregister_cpu()        |
     call acpi_unmap_lsapic()          |
          cpu's cpu_present is set     |
          to false by set_cpu_present()|
     call put_online_cpus()            |
                                       | start _cpu_up()
                                       | check cpu_present() and return -EINVAL
     return acpi_processor_remove()    |
continue hot-remove the cpu            |

Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/processor_driver.c |   14 ++++++++++++++
 kernel/cpu.c                    |    8 +++++---
 2 files changed, 19 insertions(+), 3 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-13 17:31:37.799130100 +0900
+++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 17:39:47.727006338 +0900
@@ -850,8 +850,22 @@ static int acpi_processor_handle_eject(s
 			return ret;
 	}

+	get_online_cpus();
+	/*
+	 * The cpu might become online again at this point. So we check whether
+	 * the cpu has been onlined or not. If the cpu became online, it means
+	 * that someone wants to use the cpu. So acpi_processor_handle_eject()
+	 * returns -EAGAIN.
+	 */
+	if (unlikely(cpu_online(pr->id))) {
+		put_online_cpus();
+		pr_warn("Failed to remove CPU %d, ", pr->id);
+		pr_warn("because other task brought the CPU back online\n");
+		return -EAGAIN;
+	}
 	arch_unregister_cpu(pr->id);
 	acpi_unmap_lsapic(pr->id);
+	put_online_cpus();
 	return ret;
 }
 #else
Index: linux-3.5-rc6/kernel/cpu.c
===================================================================
--- linux-3.5-rc6.orig/kernel/cpu.c	2012-07-13 17:31:37.800130087 +0900
+++ linux-3.5-rc6/kernel/cpu.c	2012-07-13 17:31:39.661106874 +0900
@@ -343,11 +343,13 @@ static int __cpuinit _cpu_up(unsigned in
 	unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0;
 	struct task_struct *idle;

-	if (cpu_online(cpu) || !cpu_present(cpu))
-		return -EINVAL;
-
 	cpu_hotplug_begin();

+	if (cpu_online(cpu) || !cpu_present(cpu)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	idle = idle_thread_get(cpu);
 	if (IS_ERR(idle)) {
 		ret = PTR_ERR(idle);

[PATCH v4 2/3] acpi : prevent cpu from becoming online

[Yasuaki Ishimatsu]

Even if acpi_processor_handle_eject() offlines cpu, there is a chance
to online the cpu after that. So the patch closes the window by using
get/put_online_cpus().

Why does the patch change _cpu_up() logic?

The patch cares the race of hot-remove cpu and _cpu_up(). If the patch
does not change it, there is the following race.

hot-remove cpu                         |  _cpu_up()
------------------------------------- ------------------------------------
call acpi_processor_handle_eject()     |
     call cpu_down()                   |
     call get_online_cpus()            |
                                       | call cpu_hotplug_begin() and stop here
     call arch_unregister_cpu()        |
     call acpi_unmap_lsapic()          |
     call put_online_cpus()            |
                                       | start and continue _cpu_up()
     return acpi_processor_remove()    |
continue hot-remove the cpu            |

So _cpu_up() can continue to itself. And hot-remove cpu can also continue
itself. If the patch changes _cpu_up() logic, the race disappears as below:

hot-remove cpu                         | _cpu_up()
-----------------------------------------------------------------------
call acpi_processor_handle_eject()     |
     call cpu_down()                   |
     call get_online_cpus()            |
                                       | call cpu_hotplug_begin() and stop here
     call arch_unregister_cpu()        |
     call acpi_unmap_lsapic()          |
          cpu's cpu_present is set     |
          to false by set_cpu_present()|
     call put_online_cpus()            |
                                       | start _cpu_up()
                                       | check cpu_present() and return -EINVAL
     return acpi_processor_remove()    |
continue hot-remove the cpu            |

Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/processor_driver.c |   14 ++++++++++++++
 kernel/cpu.c                    |    8 +++++---
 2 files changed, 19 insertions(+), 3 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-13 17:31:37.799130100 +0900
+++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 17:39:47.727006338 +0900
@@ -850,8 +850,22 @@ static int acpi_processor_handle_eject(s
 			return ret;
 	}

+	get_online_cpus();
+	/*
+	 * The cpu might become online again at this point. So we check whether
+	 * the cpu has been onlined or not. If the cpu became online, it means
+	 * that someone wants to use the cpu. So acpi_processor_handle_eject()
+	 * returns -EAGAIN.
+	 */
+	if (unlikely(cpu_online(pr->id))) {
+		put_online_cpus();
+		pr_warn("Failed to remove CPU %d, ", pr->id);
+		pr_warn("because other task brought the CPU back online\n");
+		return -EAGAIN;
+	}
 	arch_unregister_cpu(pr->id);
 	acpi_unmap_lsapic(pr->id);
+	put_online_cpus();
 	return ret;
 }
 #else
Index: linux-3.5-rc6/kernel/cpu.c
===================================================================
--- linux-3.5-rc6.orig/kernel/cpu.c	2012-07-13 17:31:37.800130087 +0900
+++ linux-3.5-rc6/kernel/cpu.c	2012-07-13 17:31:39.661106874 +0900
@@ -343,11 +343,13 @@ static int __cpuinit _cpu_up(unsigned in
 	unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0;
 	struct task_struct *idle;

-	if (cpu_online(cpu) || !cpu_present(cpu))
-		return -EINVAL;
-
 	cpu_hotplug_begin();

+	if (cpu_online(cpu) || !cpu_present(cpu)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
 	idle = idle_thread_get(cpu);
 	if (IS_ERR(idle)) {
 		ret = PTR_ERR(idle);

[PATCH v4 3/3] acpi : acpi_bus_trim() stops removing devices when failing to remove the device

[Yasuaki Ishimatsu]

acpi_bus_trim() stops removing devices, when acpi_bus_remove() return error
number. But acpi_bus_remove() cannot return error number correctly.
acpi_bus_remove() only return -EINVAL, when dev argument is NULL. Thus even if
device cannot be removed correctly, acpi_bus_trim() ignores and continues to
remove devices. acpi_bus_hot_remove_device() uses acpi_bus_trim() for removing
devices. Therefore acpi_bus_hot_remove_device() can send "_EJ0" to firmware,
even if the device is running on the system. In this case, the system cannot
work well. So acpi_bus_trim() should check whether device was removed or not
correctly. The patch adds error check into some functions to remove the device.

device_release_driver() can return error value by the patch. But the change
does not impact other caller function excluding acpi_bus_trim(), since all
of them does not check return value of device_releae_driver().

Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/scan.c    |   15 ++++++++++++---
 drivers/base/dd.c      |   22 +++++++++++++++++-----
 include/linux/device.h |    2 +-
 3 files changed, 30 insertions(+), 9 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/scan.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/scan.c	2012-07-13 15:10:46.136790418 +0900
+++ linux-3.5-rc6/drivers/acpi/scan.c	2012-07-13 15:12:41.364349387 +0900
@@ -425,12 +425,17 @@ static int acpi_device_remove(struct dev
 {
 	struct acpi_device *acpi_dev = to_acpi_device(dev);
 	struct acpi_driver *acpi_drv = acpi_dev->driver;
+	int ret;

 	if (acpi_drv) {
 		if (acpi_drv->ops.notify)
 			acpi_device_remove_notify_handler(acpi_dev);
-		if (acpi_drv->ops.remove)
-			acpi_drv->ops.remove(acpi_dev, acpi_dev->removal_type);
+		if (acpi_drv->ops.remove) {
+			ret = acpi_drv->ops.remove(acpi_dev,
+						   acpi_dev->removal_type);
+			if (ret)
+				return ret;
+		}
 	}
 	acpi_dev->driver = NULL;
 	acpi_dev->driver_data = NULL;
@@ -1208,11 +1213,15 @@ static int acpi_device_set_context(struc

 static int acpi_bus_remove(struct acpi_device *dev, int rmdevice)
 {
+	int ret;
+
 	if (!dev)
 		return -EINVAL;

 	dev->removal_type = ACPI_BUS_REMOVAL_EJECT;
-	device_release_driver(&dev->dev);
+	ret = device_release_driver(&dev->dev);
+	if (ret)
+		return ret;

 	if (!rmdevice)
 		return 0;
Index: linux-3.5-rc6/drivers/base/dd.c
===================================================================
--- linux-3.5-rc6.orig/drivers/base/dd.c	2012-07-13 15:10:46.136790418 +0900
+++ linux-3.5-rc6/drivers/base/dd.c	2012-07-13 15:14:13.895193383 +0900
@@ -464,9 +464,10 @@ EXPORT_SYMBOL_GPL(driver_attach);
  * __device_release_driver() must be called with @dev lock held.
  * When called for a USB interface, @dev->parent lock must be held as well.
  */
-static void __device_release_driver(struct device *dev)
+static int __device_release_driver(struct device *dev)
 {
 	struct device_driver *drv;
+	int ret = 0;

 	drv = dev->driver;
 	if (drv) {
@@ -482,9 +483,11 @@ static void __device_release_driver(stru
 		pm_runtime_put_sync(dev);

 		if (dev->bus && dev->bus->remove)
-			dev->bus->remove(dev);
+			ret = dev->bus->remove(dev);
 		else if (drv->remove)
-			drv->remove(dev);
+			ret = drv->remove(dev);
+		if (ret)
+			goto rollback;
 		devres_release_all(dev);
 		dev->driver = NULL;
 		klist_remove(&dev->p->knode_driver);
@@ -494,6 +497,12 @@ static void __device_release_driver(stru
 						     dev);

 	}
+
+	return ret;
+
+rollback:
+	driver_sysfs_add(dev);
+	return ret;
 }

 /**
@@ -503,16 +512,19 @@ static void __device_release_driver(stru
  * Manually detach device from driver.
  * When called for a USB interface, @dev->parent lock must be held.
  */
-void device_release_driver(struct device *dev)
+int device_release_driver(struct device *dev)
 {
+	int ret;
 	/*
 	 * If anyone calls device_release_driver() recursively from
 	 * within their ->remove callback for the same device, they
 	 * will deadlock right here.
 	 */
 	device_lock(dev);
-	__device_release_driver(dev);
+	ret = __device_release_driver(dev);
 	device_unlock(dev);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(device_release_driver);

Index: linux-3.5-rc6/include/linux/device.h
===================================================================
--- linux-3.5-rc6.orig/include/linux/device.h	2012-07-13 15:10:46.137790406 +0900
+++ linux-3.5-rc6/include/linux/device.h	2012-07-13 15:12:41.371349562 +0900
@@ -827,7 +827,7 @@ static inline void *dev_get_platdata(con
  * for information on use.
  */
 extern int __must_check device_bind_driver(struct device *dev);
-extern void device_release_driver(struct device *dev);
+extern int device_release_driver(struct device *dev);
 extern int  __must_check device_attach(struct device *dev);
 extern int __must_check driver_attach(struct device_driver *drv);
 extern int __must_check device_reprobe(struct device *dev);

[PATCH v4 3/3] acpi : acpi_bus_trim() stops removing devices when failing to remove the device

[Yasuaki Ishimatsu]

acpi_bus_trim() stops removing devices, when acpi_bus_remove() return error
number. But acpi_bus_remove() cannot return error number correctly.
acpi_bus_remove() only return -EINVAL, when dev argument is NULL. Thus even if
device cannot be removed correctly, acpi_bus_trim() ignores and continues to
remove devices. acpi_bus_hot_remove_device() uses acpi_bus_trim() for removing
devices. Therefore acpi_bus_hot_remove_device() can send "_EJ0" to firmware,
even if the device is running on the system. In this case, the system cannot
work well. So acpi_bus_trim() should check whether device was removed or not
correctly. The patch adds error check into some functions to remove the device.

device_release_driver() can return error value by the patch. But the change
does not impact other caller function excluding acpi_bus_trim(), since all
of them does not check return value of device_releae_driver().

Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

---
 drivers/acpi/scan.c    |   15 ++++++++++++---
 drivers/base/dd.c      |   22 +++++++++++++++++-----
 include/linux/device.h |    2 +-
 3 files changed, 30 insertions(+), 9 deletions(-)

Index: linux-3.5-rc6/drivers/acpi/scan.c
===================================================================
--- linux-3.5-rc6.orig/drivers/acpi/scan.c	2012-07-13 15:10:46.136790418 +0900
+++ linux-3.5-rc6/drivers/acpi/scan.c	2012-07-13 15:12:41.364349387 +0900
@@ -425,12 +425,17 @@ static int acpi_device_remove(struct dev
 {
 	struct acpi_device *acpi_dev = to_acpi_device(dev);
 	struct acpi_driver *acpi_drv = acpi_dev->driver;
+	int ret;

 	if (acpi_drv) {
 		if (acpi_drv->ops.notify)
 			acpi_device_remove_notify_handler(acpi_dev);
-		if (acpi_drv->ops.remove)
-			acpi_drv->ops.remove(acpi_dev, acpi_dev->removal_type);
+		if (acpi_drv->ops.remove) {
+			ret = acpi_drv->ops.remove(acpi_dev,
+						   acpi_dev->removal_type);
+			if (ret)
+				return ret;
+		}
 	}
 	acpi_dev->driver = NULL;
 	acpi_dev->driver_data = NULL;
@@ -1208,11 +1213,15 @@ static int acpi_device_set_context(struc

 static int acpi_bus_remove(struct acpi_device *dev, int rmdevice)
 {
+	int ret;
+
 	if (!dev)
 		return -EINVAL;

 	dev->removal_type = ACPI_BUS_REMOVAL_EJECT;
-	device_release_driver(&dev->dev);
+	ret = device_release_driver(&dev->dev);
+	if (ret)
+		return ret;

 	if (!rmdevice)
 		return 0;
Index: linux-3.5-rc6/drivers/base/dd.c
===================================================================
--- linux-3.5-rc6.orig/drivers/base/dd.c	2012-07-13 15:10:46.136790418 +0900
+++ linux-3.5-rc6/drivers/base/dd.c	2012-07-13 15:14:13.895193383 +0900
@@ -464,9 +464,10 @@ EXPORT_SYMBOL_GPL(driver_attach);
  * __device_release_driver() must be called with @dev lock held.
  * When called for a USB interface, @dev->parent lock must be held as well.
  */
-static void __device_release_driver(struct device *dev)
+static int __device_release_driver(struct device *dev)
 {
 	struct device_driver *drv;
+	int ret = 0;

 	drv = dev->driver;
 	if (drv) {
@@ -482,9 +483,11 @@ static void __device_release_driver(stru
 		pm_runtime_put_sync(dev);

 		if (dev->bus && dev->bus->remove)
-			dev->bus->remove(dev);
+			ret = dev->bus->remove(dev);
 		else if (drv->remove)
-			drv->remove(dev);
+			ret = drv->remove(dev);
+		if (ret)
+			goto rollback;
 		devres_release_all(dev);
 		dev->driver = NULL;
 		klist_remove(&dev->p->knode_driver);
@@ -494,6 +497,12 @@ static void __device_release_driver(stru
 						     dev);

 	}
+
+	return ret;
+
+rollback:
+	driver_sysfs_add(dev);
+	return ret;
 }

 /**
@@ -503,16 +512,19 @@ static void __device_release_driver(stru
  * Manually detach device from driver.
  * When called for a USB interface, @dev->parent lock must be held.
  */
-void device_release_driver(struct device *dev)
+int device_release_driver(struct device *dev)
 {
+	int ret;
 	/*
 	 * If anyone calls device_release_driver() recursively from
 	 * within their ->remove callback for the same device, they
 	 * will deadlock right here.
 	 */
 	device_lock(dev);
-	__device_release_driver(dev);
+	ret = __device_release_driver(dev);
 	device_unlock(dev);
+
+	return ret;
 }
 EXPORT_SYMBOL_GPL(device_release_driver);

Index: linux-3.5-rc6/include/linux/device.h
===================================================================
--- linux-3.5-rc6.orig/include/linux/device.h	2012-07-13 15:10:46.137790406 +0900
+++ linux-3.5-rc6/include/linux/device.h	2012-07-13 15:12:41.371349562 +0900
@@ -827,7 +827,7 @@ static inline void *dev_get_platdata(con
  * for information on use.
  */
 extern int __must_check device_bind_driver(struct device *dev);
-extern void device_release_driver(struct device *dev);
+extern int device_release_driver(struct device *dev);
 extern int  __must_check device_attach(struct device *dev);
 extern int __must_check driver_attach(struct device_driver *drv);
 extern int __must_check device_reprobe(struct device *dev);

Re: [PATCH v4 1/3] acpi : cpu hot-remove returns error when cpu_down() fails

[Toshi Kani]

On Fri, 2012-07-13 at 17:48 +0900, Yasuaki Ishimatsu wrote:
> Even if cpu_down() fails, acpi_processor_remove() continues to remove the cpu.
> But in this case, it should return error number since some process may run on
> the cpu. If the cpu has a running process and the cpu is turned the power off,
> the system may not work well.
> 
> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
> Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>

Thanks for the update.  It looks good.

Reviewed-by: Toshi Kani <toshi.kani@hp.com>


> 
> ---
>  drivers/acpi/processor_driver.c |   18 ++++++++++++------
>  1 file changed, 12 insertions(+), 6 deletions(-)
> 
> Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
> ===================================================================
> --- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-08 09:23:56.000000000 +0900
> +++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 15:11:06.135541317 +0900
> @@ -610,7 +610,7 @@ err_free_pr:
>  static int acpi_processor_remove(struct acpi_device *device, int type)
>  {
>  	struct acpi_processor *pr = NULL;
> -
> +	int ret;
> 
>  	if (!device || !acpi_driver_data(device))
>  		return -EINVAL;
> @@ -621,8 +621,9 @@ static int acpi_processor_remove(struct
>  		goto free;
> 
>  	if (type == ACPI_BUS_REMOVAL_EJECT) {
> -		if (acpi_processor_handle_eject(pr))
> -			return -EINVAL;
> +		ret = acpi_processor_handle_eject(pr);
> +		if (ret)
> +			return ret;
>  	}
> 
>  	acpi_processor_power_exit(pr, device);
> @@ -841,12 +842,17 @@ static acpi_status acpi_processor_hotadd
> 
>  static int acpi_processor_handle_eject(struct acpi_processor *pr)
>  {
> -	if (cpu_online(pr->id))
> -		cpu_down(pr->id);
> +	int ret = 0;
> +
> +	if (cpu_online(pr->id)) {
> +		ret = cpu_down(pr->id);
> +		if (ret)
> +			return ret;
> +	}
> 
>  	arch_unregister_cpu(pr->id);
>  	acpi_unmap_lsapic(pr->id);
> -	return (0);
> +	return ret;
>  }
>  #else
>  static acpi_status acpi_processor_hotadd_init(struct acpi_processor *pr)
> 

Re: [PATCH v4 2/3] acpi : prevent cpu from becoming online

[Toshi Kani]

On Fri, 2012-07-13 at 17:51 +0900, Yasuaki Ishimatsu wrote:
> Even if acpi_processor_handle_eject() offlines cpu, there is a chance
> to online the cpu after that. So the patch closes the window by using
> get/put_online_cpus().
> 
> Why does the patch change _cpu_up() logic?
> 
> The patch cares the race of hot-remove cpu and _cpu_up(). If the patch
> does not change it, there is the following race.
> 
> hot-remove cpu                         |  _cpu_up()
> ------------------------------------- ------------------------------------
> call acpi_processor_handle_eject()     |
>      call cpu_down()                   |
>      call get_online_cpus()            |
>                                        | call cpu_hotplug_begin() and stop here
>      call arch_unregister_cpu()        |
>      call acpi_unmap_lsapic()          |
>      call put_online_cpus()            |
>                                        | start and continue _cpu_up()
>      return acpi_processor_remove()    |
> continue hot-remove the cpu            |
> 
> So _cpu_up() can continue to itself. And hot-remove cpu can also continue
> itself. If the patch changes _cpu_up() logic, the race disappears as below:
> 
> hot-remove cpu                         | _cpu_up()
> -----------------------------------------------------------------------
> call acpi_processor_handle_eject()     |
>      call cpu_down()                   |
>      call get_online_cpus()            |
>                                        | call cpu_hotplug_begin() and stop here
>      call arch_unregister_cpu()        |
>      call acpi_unmap_lsapic()          |
>           cpu's cpu_present is set     |
>           to false by set_cpu_present()|
>      call put_online_cpus()            |
>                                        | start _cpu_up()
>                                        | check cpu_present() and return -EINVAL
>      return acpi_processor_remove()    |
> continue hot-remove the cpu            |
> 
> Reviewed-by: Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com>
> Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
> 
> ---
>  drivers/acpi/processor_driver.c |   14 ++++++++++++++
>  kernel/cpu.c                    |    8 +++++---
>  2 files changed, 19 insertions(+), 3 deletions(-)
> 
> Index: linux-3.5-rc6/drivers/acpi/processor_driver.c
> ===================================================================
> --- linux-3.5-rc6.orig/drivers/acpi/processor_driver.c	2012-07-13 17:31:37.799130100 +0900
> +++ linux-3.5-rc6/drivers/acpi/processor_driver.c	2012-07-13 17:39:47.727006338 +0900
> @@ -850,8 +850,22 @@ static int acpi_processor_handle_eject(s
>  			return ret;
>  	}
> 
> +	get_online_cpus();
> +	/*
> +	 * The cpu might become online again at this point. So we check whether
> +	 * the cpu has been onlined or not. If the cpu became online, it means
> +	 * that someone wants to use the cpu. So acpi_processor_handle_eject()
> +	 * returns -EAGAIN.
> +	 */
> +	if (unlikely(cpu_online(pr->id))) {
> +		put_online_cpus();
> +		pr_warn("Failed to remove CPU %d, ", pr->id);
> +		pr_warn("because other task brought the CPU back online\n");

Just a nit-pick, but you can simply do it like:

		pr_warn("Failed to remove CPU %d, because other task "
			"brought the CPU back online\n", pr->id);

Other than that,

Reviewed-by: Toshi Kani <toshi.kani@hp.com>

Thanks,
-Toshi

> +		return -EAGAIN;
> +	}
>  	arch_unregister_cpu(pr->id);
>  	acpi_unmap_lsapic(pr->id);
> +	put_online_cpus();
>  	return ret;
>  }
>  #else
> Index: linux-3.5-rc6/kernel/cpu.c
> ===================================================================
> --- linux-3.5-rc6.orig/kernel/cpu.c	2012-07-13 17:31:37.800130087 +0900
> +++ linux-3.5-rc6/kernel/cpu.c	2012-07-13 17:31:39.661106874 +0900
> @@ -343,11 +343,13 @@ static int __cpuinit _cpu_up(unsigned in
>  	unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0;
>  	struct task_struct *idle;
> 
> -	if (cpu_online(cpu) || !cpu_present(cpu))
> -		return -EINVAL;
> -
>  	cpu_hotplug_begin();
> 
> +	if (cpu_online(cpu) || !cpu_present(cpu)) {
> +		ret = -EINVAL;
> +		goto out;
> +	}
> +
>  	idle = idle_thread_get(cpu);
>  	if (IS_ERR(idle)) {
>  		ret = PTR_ERR(idle);
> 

Re: [PATCH v4 3/3] acpi : acpi_bus_trim() stops removing devices when failing to remove the device

[Toshi Kani]

On Fri, 2012-07-13 at 17:53 +0900, Yasuaki Ishimatsu wrote:
> acpi_bus_trim() stops removing devices, when acpi_bus_remove() return error
> number. But acpi_bus_remove() cannot return error number correctly.
> acpi_bus_remove() only return -EINVAL, when dev argument is NULL. Thus even if
> device cannot be removed correctly, acpi_bus_trim() ignores and continues to
> remove devices. acpi_bus_hot_remove_device() uses acpi_bus_trim() for removing
> devices. Therefore acpi_bus_hot_remove_device() can send "_EJ0" to firmware,
> even if the device is running on the system. In this case, the system cannot
> work well. So acpi_bus_trim() should check whether device was removed or not
> correctly. The patch adds error check into some functions to remove the device.
> 
> device_release_driver() can return error value by the patch. But the change
> does not impact other caller function excluding acpi_bus_trim(), since all
> of them does not check return value of device_releae_driver().

I think potential risk here is that __device_release_driver() now
performs rollback in case of error from driver's remove interface.  I
agree with doing the rollback, but this leads to a different situation
if the caller does not check error from device_release_driver() and
proceeds the operation.  So, we will need to make sure that:
 - Other driver's remove interfaces do not fail (or very unlikely to
fail), or
 - If other driver's remove interfaces failed, their end results are no
worse than today.

Thanks,
-Toshi


> Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
> 
> ---
>  drivers/acpi/scan.c    |   15 ++++++++++++---
>  drivers/base/dd.c      |   22 +++++++++++++++++-----
>  include/linux/device.h |    2 +-
>  3 files changed, 30 insertions(+), 9 deletions(-)
> 
> Index: linux-3.5-rc6/drivers/acpi/scan.c
> ===================================================================
> --- linux-3.5-rc6.orig/drivers/acpi/scan.c	2012-07-13 15:10:46.136790418 +0900
> +++ linux-3.5-rc6/drivers/acpi/scan.c	2012-07-13 15:12:41.364349387 +0900
> @@ -425,12 +425,17 @@ static int acpi_device_remove(struct dev
>  {
>  	struct acpi_device *acpi_dev = to_acpi_device(dev);
>  	struct acpi_driver *acpi_drv = acpi_dev->driver;
> +	int ret;
> 
>  	if (acpi_drv) {
>  		if (acpi_drv->ops.notify)
>  			acpi_device_remove_notify_handler(acpi_dev);
> -		if (acpi_drv->ops.remove)
> -			acpi_drv->ops.remove(acpi_dev, acpi_dev->removal_type);
> +		if (acpi_drv->ops.remove) {
> +			ret = acpi_drv->ops.remove(acpi_dev,
> +						   acpi_dev->removal_type);
> +			if (ret)
> +				return ret;
> +		}
>  	}
>  	acpi_dev->driver = NULL;
>  	acpi_dev->driver_data = NULL;
> @@ -1208,11 +1213,15 @@ static int acpi_device_set_context(struc
> 
>  static int acpi_bus_remove(struct acpi_device *dev, int rmdevice)
>  {
> +	int ret;
> +
>  	if (!dev)
>  		return -EINVAL;
> 
>  	dev->removal_type = ACPI_BUS_REMOVAL_EJECT;
> -	device_release_driver(&dev->dev);
> +	ret = device_release_driver(&dev->dev);
> +	if (ret)
> +		return ret;
> 
>  	if (!rmdevice)
>  		return 0;
> Index: linux-3.5-rc6/drivers/base/dd.c
> ===================================================================
> --- linux-3.5-rc6.orig/drivers/base/dd.c	2012-07-13 15:10:46.136790418 +0900
> +++ linux-3.5-rc6/drivers/base/dd.c	2012-07-13 15:14:13.895193383 +0900
> @@ -464,9 +464,10 @@ EXPORT_SYMBOL_GPL(driver_attach);
>   * __device_release_driver() must be called with @dev lock held.
>   * When called for a USB interface, @dev->parent lock must be held as well.
>   */
> -static void __device_release_driver(struct device *dev)
> +static int __device_release_driver(struct device *dev)
>  {
>  	struct device_driver *drv;
> +	int ret = 0;
> 
>  	drv = dev->driver;
>  	if (drv) {
> @@ -482,9 +483,11 @@ static void __device_release_driver(stru
>  		pm_runtime_put_sync(dev);
> 
>  		if (dev->bus && dev->bus->remove)
> -			dev->bus->remove(dev);
> +			ret = dev->bus->remove(dev);
>  		else if (drv->remove)
> -			drv->remove(dev);
> +			ret = drv->remove(dev);
> +		if (ret)
> +			goto rollback;
>  		devres_release_all(dev);
>  		dev->driver = NULL;
>  		klist_remove(&dev->p->knode_driver);
> @@ -494,6 +497,12 @@ static void __device_release_driver(stru
>  						     dev);
> 
>  	}
> +
> +	return ret;
> +
> +rollback:
> +	driver_sysfs_add(dev);
> +	return ret;
>  }
> 
>  /**
> @@ -503,16 +512,19 @@ static void __device_release_driver(stru
>   * Manually detach device from driver.
>   * When called for a USB interface, @dev->parent lock must be held.
>   */
> -void device_release_driver(struct device *dev)
> +int device_release_driver(struct device *dev)
>  {
> +	int ret;
>  	/*
>  	 * If anyone calls device_release_driver() recursively from
>  	 * within their ->remove callback for the same device, they
>  	 * will deadlock right here.
>  	 */
>  	device_lock(dev);
> -	__device_release_driver(dev);
> +	ret = __device_release_driver(dev);
>  	device_unlock(dev);
> +
> +	return ret;
>  }
>  EXPORT_SYMBOL_GPL(device_release_driver);
> 
> Index: linux-3.5-rc6/include/linux/device.h
> ===================================================================
> --- linux-3.5-rc6.orig/include/linux/device.h	2012-07-13 15:10:46.137790406 +0900
> +++ linux-3.5-rc6/include/linux/device.h	2012-07-13 15:12:41.371349562 +0900
> @@ -827,7 +827,7 @@ static inline void *dev_get_platdata(con
>   * for information on use.
>   */
>  extern int __must_check device_bind_driver(struct device *dev);
> -extern void device_release_driver(struct device *dev);
> +extern int device_release_driver(struct device *dev);
>  extern int  __must_check device_attach(struct device *dev);
>  extern int __must_check driver_attach(struct device_driver *drv);
>  extern int __must_check device_reprobe(struct device *dev);
>