From 1eede070a59e1cc73da51e1aaa00d9ab86572cfc Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Tue, 20 May 2008 23:00:01 +0200 Subject: Introduce new top level suspend and hibernation callbacks Introduce 'struct pm_ops' and 'struct pm_ext_ops' ('ext' meaning 'extended') representing suspend and hibernation operations for bus types, device classes, device types and device drivers. Modify the PM core to use 'struct pm_ops' and 'struct pm_ext_ops' objects, if defined, instead of the ->suspend(), ->resume(), ->suspend_late(), and ->resume_early() callbacks (the old callbacks will be considered as legacy and gradually phased out). The main purpose of doing this is to separate suspend (aka S2RAM and standby) callbacks from hibernation callbacks in such a way that the new callbacks won't take arguments and the semantics of each of them will be clearly specified. This has been requested for multiple times by many people, including Linus himself, and the reason is that within the current scheme if ->resume() is called, for example, it's difficult to say why it's been called (ie. is it a resume from RAM or from hibernation or a suspend/hibernation failure etc.?). The second purpose is to make the suspend/hibernation callbacks more flexible so that device drivers can handle more than they can within the current scheme. For example, some drivers may need to prevent new children of the device from being registered before their ->suspend() callbacks are executed or they may want to carry out some operations requiring the availability of some other devices, not directly bound via the parent-child relationship, in order to prepare for the execution of ->suspend(), etc. Ultimately, we'd like to stop using the freezing of tasks for suspend and therefore the drivers' suspend/hibernation code will have to take care of the handling of the user space during suspend/hibernation. That, in turn, would be difficult within the current scheme, without the new ->prepare() and ->complete() callbacks. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Jesse Barnes --- kernel/power/disk.c | 22 +++++++++++++++------- kernel/power/main.c | 6 ++++-- 2 files changed, 19 insertions(+), 9 deletions(-) (limited to 'kernel/power') diff --git a/kernel/power/disk.c b/kernel/power/disk.c index 14a656c..d416be0 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -193,6 +193,7 @@ static int create_image(int platform_mode) if (error) return error; + device_pm_lock(); local_irq_disable(); /* At this point, device_suspend() has been called, but *not* * device_power_down(). We *must* call device_power_down() now. @@ -224,9 +225,11 @@ static int create_image(int platform_mode) /* NOTE: device_power_up() is just a resume() for devices * that suspended with irqs off ... no overall powerup. */ - device_power_up(); + device_power_up(in_suspend ? + (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); Enable_irqs: local_irq_enable(); + device_pm_unlock(); return error; } @@ -280,7 +283,8 @@ int hibernation_snapshot(int platform_mode) Finish: platform_finish(platform_mode); Resume_devices: - device_resume(); + device_resume(in_suspend ? + (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); Resume_console: resume_console(); Close: @@ -300,8 +304,9 @@ static int resume_target_kernel(void) { int error; + device_pm_lock(); local_irq_disable(); - error = device_power_down(PMSG_PRETHAW); + error = device_power_down(PMSG_QUIESCE); if (error) { printk(KERN_ERR "PM: Some devices failed to power down, " "aborting resume\n"); @@ -329,9 +334,10 @@ static int resume_target_kernel(void) swsusp_free(); restore_processor_state(); touch_softlockup_watchdog(); - device_power_up(); + device_power_up(PMSG_RECOVER); Enable_irqs: local_irq_enable(); + device_pm_unlock(); return error; } @@ -350,7 +356,7 @@ int hibernation_restore(int platform_mode) pm_prepare_console(); suspend_console(); - error = device_suspend(PMSG_PRETHAW); + error = device_suspend(PMSG_QUIESCE); if (error) goto Finish; @@ -362,7 +368,7 @@ int hibernation_restore(int platform_mode) enable_nonboot_cpus(); } platform_restore_cleanup(platform_mode); - device_resume(); + device_resume(PMSG_RECOVER); Finish: resume_console(); pm_restore_console(); @@ -403,6 +409,7 @@ int hibernation_platform_enter(void) if (error) goto Finish; + device_pm_lock(); local_irq_disable(); error = device_power_down(PMSG_HIBERNATE); if (!error) { @@ -411,6 +418,7 @@ int hibernation_platform_enter(void) while (1); } local_irq_enable(); + device_pm_unlock(); /* * We don't need to reenable the nonboot CPUs or resume consoles, since @@ -419,7 +427,7 @@ int hibernation_platform_enter(void) Finish: hibernation_ops->finish(); Resume_devices: - device_resume(); + device_resume(PMSG_RESTORE); Resume_console: resume_console(); Close: diff --git a/kernel/power/main.c b/kernel/power/main.c index 6a6d5eb..d023b6b 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -228,6 +228,7 @@ static int suspend_enter(suspend_state_t state) { int error = 0; + device_pm_lock(); arch_suspend_disable_irqs(); BUG_ON(!irqs_disabled()); @@ -239,10 +240,11 @@ static int suspend_enter(suspend_state_t state) if (!suspend_test(TEST_CORE)) error = suspend_ops->enter(state); - device_power_up(); + device_power_up(PMSG_RESUME); Done: arch_suspend_enable_irqs(); BUG_ON(irqs_disabled()); + device_pm_unlock(); return error; } @@ -291,7 +293,7 @@ int suspend_devices_and_enter(suspend_state_t state) if (suspend_ops->finish) suspend_ops->finish(); Resume_devices: - device_resume(); + device_resume(PMSG_RESUME); Resume_console: resume_console(); Close: -- cgit v1.1 From d8f3de0d2412bb91639cfefc5b3c79dbf3812212 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Thu, 12 Jun 2008 23:24:06 +0200 Subject: Suspend-related patches for 2.6.27 ACPI PM: Add possibility to change suspend sequence There are some systems out there that don't work correctly with our current suspend/hibernation code ordering. Provide a workaround for these systems allowing them to pass 'acpi_sleep=old_ordering' in the kernel command line so that it will use the pre-ACPI 2.0 ("old") suspend code ordering. Unfortunately, this requires us to add a platform hook to the resuming of devices for recovering the platform in case one of the device drivers' .suspend() routines returns error code. Namely, ACPI 1.0 specifies that _PTS should be called before suspending devices, but _WAK still should be called before resuming them in order to undo the changes made by _PTS. However, if there is an error during suspending devices, they are automatically resumed without returning control to the PM core, so the _WAK has to be called from within device_resume() in that cases. The patch also reorders and refactors the ACPI suspend/hibernation code to avoid duplication as far as reasonably possible. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Jesse Barnes --- kernel/power/disk.c | 28 ++++++++++++++++++++++------ kernel/power/main.c | 10 +++++++--- 2 files changed, 29 insertions(+), 9 deletions(-) (limited to 'kernel/power') diff --git a/kernel/power/disk.c b/kernel/power/disk.c index d416be0..f011e08 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -180,6 +180,17 @@ static void platform_restore_cleanup(int platform_mode) } /** + * platform_recover - recover the platform from a failure to suspend + * devices. + */ + +static void platform_recover(int platform_mode) +{ + if (platform_mode && hibernation_ops && hibernation_ops->recover) + hibernation_ops->recover(); +} + +/** * create_image - freeze devices that need to be frozen with interrupts * off, create the hibernation image and thaw those devices. Control * reappears in this routine after a restore. @@ -258,10 +269,10 @@ int hibernation_snapshot(int platform_mode) suspend_console(); error = device_suspend(PMSG_FREEZE); if (error) - goto Resume_console; + goto Recover_platform; if (hibernation_test(TEST_DEVICES)) - goto Resume_devices; + goto Recover_platform; error = platform_pre_snapshot(platform_mode); if (error || hibernation_test(TEST_PLATFORM)) @@ -285,11 +296,14 @@ int hibernation_snapshot(int platform_mode) Resume_devices: device_resume(in_suspend ? (error ? PMSG_RECOVER : PMSG_THAW) : PMSG_RESTORE); - Resume_console: resume_console(); Close: platform_end(platform_mode); return error; + + Recover_platform: + platform_recover(platform_mode); + goto Resume_devices; } /** @@ -398,8 +412,11 @@ int hibernation_platform_enter(void) suspend_console(); error = device_suspend(PMSG_HIBERNATE); - if (error) - goto Resume_console; + if (error) { + if (hibernation_ops->recover) + hibernation_ops->recover(); + goto Resume_devices; + } error = hibernation_ops->prepare(); if (error) @@ -428,7 +445,6 @@ int hibernation_platform_enter(void) hibernation_ops->finish(); Resume_devices: device_resume(PMSG_RESTORE); - Resume_console: resume_console(); Close: hibernation_ops->end(); diff --git a/kernel/power/main.c b/kernel/power/main.c index d023b6b..3398f46 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -269,11 +269,11 @@ int suspend_devices_and_enter(suspend_state_t state) error = device_suspend(PMSG_SUSPEND); if (error) { printk(KERN_ERR "PM: Some devices failed to suspend\n"); - goto Resume_console; + goto Recover_platform; } if (suspend_test(TEST_DEVICES)) - goto Resume_devices; + goto Recover_platform; if (suspend_ops->prepare) { error = suspend_ops->prepare(); @@ -294,12 +294,16 @@ int suspend_devices_and_enter(suspend_state_t state) suspend_ops->finish(); Resume_devices: device_resume(PMSG_RESUME); - Resume_console: resume_console(); Close: if (suspend_ops->end) suspend_ops->end(); return error; + + Recover_platform: + if (suspend_ops->recover) + suspend_ops->recover(); + goto Resume_devices; } /** -- cgit v1.1 From ebb12db51f6c13b30752fcf506baad4c617b153c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 11 Jun 2008 22:04:29 +0200 Subject: Freezer: Introduce PF_FREEZER_NOSIG The freezer currently attempts to distinguish kernel threads from user space tasks by checking if their mm pointer is unset and it does not send fake signals to kernel threads. However, there are kernel threads, mostly related to networking, that behave like user space tasks and may want to be sent a fake signal to be frozen. Introduce the new process flag PF_FREEZER_NOSIG that will be set by default for all kernel threads and make the freezer only send fake signals to the tasks having PF_FREEZER_NOSIG unset. Provide the set_freezable_with_signal() function to be called by the kernel threads that want to be sent a fake signal for freezing. This patch should not change the freezer's observable behavior. Signed-off-by: Rafael J. Wysocki Signed-off-by: Andi Kleen Acked-by: Pavel Machek Signed-off-by: Len Brown --- kernel/power/process.c | 97 ++++++++++++++++++++++---------------------------- 1 file changed, 42 insertions(+), 55 deletions(-) (limited to 'kernel/power') diff --git a/kernel/power/process.c b/kernel/power/process.c index f1d0b34..5fb8765 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -19,9 +19,6 @@ */ #define TIMEOUT (20 * HZ) -#define FREEZER_KERNEL_THREADS 0 -#define FREEZER_USER_SPACE 1 - static inline int freezeable(struct task_struct * p) { if ((p == current) || @@ -84,63 +81,53 @@ static void fake_signal_wake_up(struct task_struct *p) spin_unlock_irqrestore(&p->sighand->siglock, flags); } -static int has_mm(struct task_struct *p) +static inline bool should_send_signal(struct task_struct *p) { - return (p->mm && !(p->flags & PF_BORROWED_MM)); + return !(p->flags & PF_FREEZER_NOSIG); } /** * freeze_task - send a freeze request to given task * @p: task to send the request to - * @with_mm_only: if set, the request will only be sent if the task has its - * own mm - * Return value: 0, if @with_mm_only is set and the task has no mm of its - * own or the task is frozen, 1, otherwise + * @sig_only: if set, the request will only be sent if the task has the + * PF_FREEZER_NOSIG flag unset + * Return value: 'false', if @sig_only is set and the task has + * PF_FREEZER_NOSIG set or the task is frozen, 'true', otherwise * - * The freeze request is sent by seting the tasks's TIF_FREEZE flag and + * The freeze request is sent by setting the tasks's TIF_FREEZE flag and * either sending a fake signal to it or waking it up, depending on whether - * or not it has its own mm (ie. it is a user land task). If @with_mm_only - * is set and the task has no mm of its own (ie. it is a kernel thread), - * its TIF_FREEZE flag should not be set. - * - * The task_lock() is necessary to prevent races with exit_mm() or - * use_mm()/unuse_mm() from occuring. + * or not it has PF_FREEZER_NOSIG set. If @sig_only is set and the task + * has PF_FREEZER_NOSIG set (ie. it is a typical kernel thread), its + * TIF_FREEZE flag will not be set. */ -static int freeze_task(struct task_struct *p, int with_mm_only) +static bool freeze_task(struct task_struct *p, bool sig_only) { - int ret = 1; + /* + * We first check if the task is freezing and next if it has already + * been frozen to avoid the race with frozen_process() which first marks + * the task as frozen and next clears its TIF_FREEZE. + */ + if (!freezing(p)) { + rmb(); + if (frozen(p)) + return false; - task_lock(p); - if (freezing(p)) { - if (has_mm(p)) { - if (!signal_pending(p)) - fake_signal_wake_up(p); - } else { - if (with_mm_only) - ret = 0; - else - wake_up_state(p, TASK_INTERRUPTIBLE); - } + if (!sig_only || should_send_signal(p)) + set_freeze_flag(p); + else + return false; + } + + if (should_send_signal(p)) { + if (!signal_pending(p)) + fake_signal_wake_up(p); + } else if (sig_only) { + return false; } else { - rmb(); - if (frozen(p)) { - ret = 0; - } else { - if (has_mm(p)) { - set_freeze_flag(p); - fake_signal_wake_up(p); - } else { - if (with_mm_only) { - ret = 0; - } else { - set_freeze_flag(p); - wake_up_state(p, TASK_INTERRUPTIBLE); - } - } - } + wake_up_state(p, TASK_INTERRUPTIBLE); } - task_unlock(p); - return ret; + + return true; } static void cancel_freezing(struct task_struct *p) @@ -156,7 +143,7 @@ static void cancel_freezing(struct task_struct *p) } } -static int try_to_freeze_tasks(int freeze_user_space) +static int try_to_freeze_tasks(bool sig_only) { struct task_struct *g, *p; unsigned long end_time; @@ -175,7 +162,7 @@ static int try_to_freeze_tasks(int freeze_user_space) if (frozen(p) || !freezeable(p)) continue; - if (!freeze_task(p, freeze_user_space)) + if (!freeze_task(p, sig_only)) continue; /* @@ -235,13 +222,13 @@ int freeze_processes(void) int error; printk("Freezing user space processes ... "); - error = try_to_freeze_tasks(FREEZER_USER_SPACE); + error = try_to_freeze_tasks(true); if (error) goto Exit; printk("done.\n"); printk("Freezing remaining freezable tasks ... "); - error = try_to_freeze_tasks(FREEZER_KERNEL_THREADS); + error = try_to_freeze_tasks(false); if (error) goto Exit; printk("done."); @@ -251,7 +238,7 @@ int freeze_processes(void) return error; } -static void thaw_tasks(int thaw_user_space) +static void thaw_tasks(bool nosig_only) { struct task_struct *g, *p; @@ -260,7 +247,7 @@ static void thaw_tasks(int thaw_user_space) if (!freezeable(p)) continue; - if (!p->mm == thaw_user_space) + if (nosig_only && should_send_signal(p)) continue; thaw_process(p); @@ -271,8 +258,8 @@ static void thaw_tasks(int thaw_user_space) void thaw_processes(void) { printk("Restarting tasks ... "); - thaw_tasks(FREEZER_KERNEL_THREADS); - thaw_tasks(FREEZER_USER_SPACE); + thaw_tasks(true); + thaw_tasks(false); schedule(); printk("done.\n"); } -- cgit v1.1 From 52d11025dba32bed696eaee1822b26529e764770 Mon Sep 17 00:00:00 2001 From: Alan Cox Date: Wed, 11 Jun 2008 22:07:52 +0200 Subject: snapshot: Push BKL down into ioctl handlers Push BKL down into ioctl handlers - snapshot device. Signed-off-by: Alan Cox Signed-off-by: Rafael J. Wysocki Signed-off-by: Len Brown Signed-off-by: Andi Kleen --- kernel/power/user.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'kernel/power') diff --git a/kernel/power/user.c b/kernel/power/user.c index f5512cb..658262b 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -23,6 +23,7 @@ #include #include #include +#include #include @@ -164,8 +165,8 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, return res; } -static int snapshot_ioctl(struct inode *inode, struct file *filp, - unsigned int cmd, unsigned long arg) +static long snapshot_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) { int error = 0; struct snapshot_data *data; @@ -181,6 +182,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, data = filp->private_data; + lock_kernel(); + switch (cmd) { case SNAPSHOT_FREEZE: @@ -389,7 +392,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -ENOTTY; } - + unlock_kernel(); return error; } @@ -399,7 +402,7 @@ static const struct file_operations snapshot_fops = { .read = snapshot_read, .write = snapshot_write, .llseek = no_llseek, - .ioctl = snapshot_ioctl, + .unlocked_ioctl = snapshot_ioctl, }; static struct miscdevice snapshot_device = { -- cgit v1.1 From 25f2f3daadaf0768a61d02ee3ed3d9a21e9dc46c Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 11 Jun 2008 22:09:45 +0200 Subject: snapshot: Use pm_mutex for mutual exclusion We can avoid taking the BKL in snapshot_ioctl() if pm_mutex is used to prevent the ioctls from being executed concurrently. In addition, although it is only possible to open /dev/snapshot once, the task which has done that may spawn a child that will inherit the open descriptor, so in theory they can call snapshot_write(), snapshot_read() and snapshot_release() concurrently. pm_mutex can also be used for mutual exclusion in such cases. Signed-off-by: Rafael J. Wysocki Signed-off-by: Andi Kleen Acked-by: Pavel Machek Signed-off-by: Len Brown --- kernel/power/user.c | 68 +++++++++++++++++++++++++++++++++-------------------- 1 file changed, 42 insertions(+), 26 deletions(-) (limited to 'kernel/power') diff --git a/kernel/power/user.c b/kernel/power/user.c index 658262b..a6332a3 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -70,16 +70,22 @@ static int snapshot_open(struct inode *inode, struct file *filp) struct snapshot_data *data; int error; - if (!atomic_add_unless(&snapshot_device_available, -1, 0)) - return -EBUSY; + mutex_lock(&pm_mutex); + + if (!atomic_add_unless(&snapshot_device_available, -1, 0)) { + error = -EBUSY; + goto Unlock; + } if ((filp->f_flags & O_ACCMODE) == O_RDWR) { atomic_inc(&snapshot_device_available); - return -ENOSYS; + error = -ENOSYS; + goto Unlock; } if(create_basic_memory_bitmaps()) { atomic_inc(&snapshot_device_available); - return -ENOMEM; + error = -ENOMEM; + goto Unlock; } nonseekable_open(inode, filp); data = &snapshot_state; @@ -99,33 +105,36 @@ static int snapshot_open(struct inode *inode, struct file *filp) if (error) pm_notifier_call_chain(PM_POST_HIBERNATION); } - if (error) { + if (error) atomic_inc(&snapshot_device_available); - return error; - } data->frozen = 0; data->ready = 0; data->platform_support = 0; - return 0; + Unlock: + mutex_unlock(&pm_mutex); + + return error; } static int snapshot_release(struct inode *inode, struct file *filp) { struct snapshot_data *data; + mutex_lock(&pm_mutex); + swsusp_free(); free_basic_memory_bitmaps(); data = filp->private_data; free_all_swap_pages(data->swap); - if (data->frozen) { - mutex_lock(&pm_mutex); + if (data->frozen) thaw_processes(); - mutex_unlock(&pm_mutex); - } pm_notifier_call_chain(data->mode == O_WRONLY ? PM_POST_HIBERNATION : PM_POST_RESTORE); atomic_inc(&snapshot_device_available); + + mutex_unlock(&pm_mutex); + return 0; } @@ -135,9 +144,13 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf, struct snapshot_data *data; ssize_t res; + mutex_lock(&pm_mutex); + data = filp->private_data; - if (!data->ready) - return -ENODATA; + if (!data->ready) { + res = -ENODATA; + goto Unlock; + } res = snapshot_read_next(&data->handle, count); if (res > 0) { if (copy_to_user(buf, data_of(data->handle), res)) @@ -145,6 +158,10 @@ static ssize_t snapshot_read(struct file *filp, char __user *buf, else *offp = data->handle.offset; } + + Unlock: + mutex_unlock(&pm_mutex); + return res; } @@ -154,6 +171,8 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, struct snapshot_data *data; ssize_t res; + mutex_lock(&pm_mutex); + data = filp->private_data; res = snapshot_write_next(&data->handle, count); if (res > 0) { @@ -162,6 +181,9 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, else *offp = data->handle.offset; } + + mutex_unlock(&pm_mutex); + return res; } @@ -180,16 +202,16 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, if (!capable(CAP_SYS_ADMIN)) return -EPERM; - data = filp->private_data; + if (!mutex_trylock(&pm_mutex)) + return -EBUSY; - lock_kernel(); + data = filp->private_data; switch (cmd) { case SNAPSHOT_FREEZE: if (data->frozen) break; - mutex_lock(&pm_mutex); printk("Syncing filesystems ... "); sys_sync(); printk("done.\n"); @@ -197,7 +219,6 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, error = freeze_processes(); if (error) thaw_processes(); - mutex_unlock(&pm_mutex); if (!error) data->frozen = 1; break; @@ -205,9 +226,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, case SNAPSHOT_UNFREEZE: if (!data->frozen || data->ready) break; - mutex_lock(&pm_mutex); thaw_processes(); - mutex_unlock(&pm_mutex); data->frozen = 0; break; @@ -310,16 +329,11 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, error = -EPERM; break; } - if (!mutex_trylock(&pm_mutex)) { - error = -EBUSY; - break; - } /* * Tasks are frozen and the notifiers have been called with * PM_HIBERNATION_PREPARE */ error = suspend_devices_and_enter(PM_SUSPEND_MEM); - mutex_unlock(&pm_mutex); break; case SNAPSHOT_PLATFORM_SUPPORT: @@ -392,7 +406,9 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, error = -ENOTTY; } - unlock_kernel(); + + mutex_unlock(&pm_mutex); + return error; } -- cgit v1.1