From 143aa5c53bd3895d42d7c08753fe58293988a69d Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 12 May 2008 21:21:13 +0200 Subject: cpu: change some globals to statics in drivers/base/cpu.c v2 This patch makes the following needlessly global code static: - attr_online_map - attr_possible_map - attr_present_map - cpu_state_attr [v2] Signed-off-by: Adrian Bunk Signed-off-by: Mike Travis Signed-off-by: Ingo Molnar --- drivers/base/cpu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index e38dfed41d8..5000402ae09 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -119,14 +119,14 @@ static ssize_t print_cpus_##type(struct sysdev_class *class, char *buf) \ { \ return print_cpus_map(buf, &cpu_##type##_map); \ } \ -struct sysdev_class_attribute attr_##type##_map = \ +static struct sysdev_class_attribute attr_##type##_map = \ _SYSDEV_CLASS_ATTR(type, 0444, print_cpus_##type, NULL) print_cpus_func(online); print_cpus_func(possible); print_cpus_func(present); -struct sysdev_class_attribute *cpu_state_attr[] = { +static struct sysdev_class_attribute *cpu_state_attr[] = { &attr_online_map, &attr_possible_map, &attr_present_map, -- cgit v1.2.3-70-g09d2 From 63687a528c39a67c1a213cdffa09feb0e6af9dbe Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 12 May 2008 15:44:41 +0200 Subject: x86: move tracedata to RODATA .. allowing it to be write-protected just as other read-only data under CONFIG_DEBUG_RODATA. Signed-off-by: Jan Beulich Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- arch/x86/kernel/vmlinux_32.lds.S | 7 ------- arch/x86/kernel/vmlinux_64.lds.S | 7 ------- drivers/base/power/trace.c | 2 +- include/asm-generic/vmlinux.lds.h | 14 ++++++++++++++ include/asm-x86/resume-trace.h | 2 +- include/linux/resume-trace.h | 2 +- 6 files changed, 17 insertions(+), 17 deletions(-) (limited to 'drivers/base') diff --git a/arch/x86/kernel/vmlinux_32.lds.S b/arch/x86/kernel/vmlinux_32.lds.S index ce5ed083a1e..2674f579627 100644 --- a/arch/x86/kernel/vmlinux_32.lds.S +++ b/arch/x86/kernel/vmlinux_32.lds.S @@ -60,13 +60,6 @@ SECTIONS BUG_TABLE :text - . = ALIGN(4); - .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { - __tracedata_start = .; - *(.tracedata) - __tracedata_end = .; - } - RODATA /* writeable */ diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S index fad3674b06a..687041bfbae 100644 --- a/arch/x86/kernel/vmlinux_64.lds.S +++ b/arch/x86/kernel/vmlinux_64.lds.S @@ -53,13 +53,6 @@ SECTIONS RODATA - . = ALIGN(4); - .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { - __tracedata_start = .; - *(.tracedata) - __tracedata_end = .; - } - . = ALIGN(PAGE_SIZE); /* Align data segment to page size boundary */ /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c index 2b4b392dcbc..87a7f1d0257 100644 --- a/drivers/base/power/trace.c +++ b/drivers/base/power/trace.c @@ -153,7 +153,7 @@ EXPORT_SYMBOL(set_trace_device); * it's not any guarantee, but it's a high _likelihood_ that * the match is valid). */ -void generate_resume_trace(void *tracedata, unsigned int user) +void generate_resume_trace(const void *tracedata, unsigned int user) { unsigned short lineno = *(unsigned short *)tracedata; const char *file = *(const char **)(tracedata + 2); diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index f054778e916..f1992dc5c42 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -93,6 +93,8 @@ VMLINUX_SYMBOL(__end_rio_route_ops) = .; \ } \ \ + TRACEDATA \ + \ /* Kernel symbol table: Normal symbols */ \ __ksymtab : AT(ADDR(__ksymtab) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start___ksymtab) = .; \ @@ -318,6 +320,18 @@ __stop___bug_table = .; \ } +#ifdef CONFIG_PM_TRACE +#define TRACEDATA \ + . = ALIGN(4); \ + .tracedata : AT(ADDR(.tracedata) - LOAD_OFFSET) { \ + __tracedata_start = .; \ + *(.tracedata) \ + __tracedata_end = .; \ + } +#else +#define TRACEDATA +#endif + #define NOTES \ .notes : AT(ADDR(.notes) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start_notes) = .; \ diff --git a/include/asm-x86/resume-trace.h b/include/asm-x86/resume-trace.h index 2557514d7ef..8d9f0b41ee8 100644 --- a/include/asm-x86/resume-trace.h +++ b/include/asm-x86/resume-trace.h @@ -6,7 +6,7 @@ #define TRACE_RESUME(user) \ do { \ if (pm_trace_enabled) { \ - void *tracedata; \ + const void *tracedata; \ asm volatile(_ASM_MOV_UL " $1f,%0\n" \ ".section .tracedata,\"a\"\n" \ "1:\t.word %c1\n\t" \ diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h index f3f4f28c696..c9ba2fdf807 100644 --- a/include/linux/resume-trace.h +++ b/include/linux/resume-trace.h @@ -8,7 +8,7 @@ extern int pm_trace_enabled; struct device; extern void set_trace_device(struct device *); -extern void generate_resume_trace(void *tracedata, unsigned int user); +extern void generate_resume_trace(const void *tracedata, unsigned int user); #define TRACE_DEVICE(dev) do { \ if (pm_trace_enabled) \ -- cgit v1.2.3-70-g09d2 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 --- arch/x86/kernel/apm_32.c | 8 +- drivers/base/power/main.c | 675 +++++++++++++++++++++++++++++++++++---------- drivers/base/power/power.h | 2 +- drivers/base/power/trace.c | 4 +- include/linux/device.h | 9 + include/linux/pm.h | 314 +++++++++++++++++++-- kernel/power/disk.c | 22 +- kernel/power/main.c | 6 +- 8 files changed, 845 insertions(+), 195 deletions(-) (limited to 'drivers/base') diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index bf9290e2901..c1735f61a2c 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1211,9 +1211,9 @@ static int suspend(int vetoable) if (err != APM_SUCCESS) apm_error("suspend", err); err = (err == APM_SUCCESS) ? 0 : -EIO; - device_power_up(); + device_power_up(PMSG_RESUME); local_irq_enable(); - device_resume(); + device_resume(PMSG_RESUME); queue_event(APM_NORMAL_RESUME, NULL); spin_lock(&user_list_lock); for (as = user_list; as != NULL; as = as->next) { @@ -1238,7 +1238,7 @@ static void standby(void) apm_error("standby", err); local_irq_disable(); - device_power_up(); + device_power_up(PMSG_RESUME); local_irq_enable(); } @@ -1324,7 +1324,7 @@ static void check_events(void) ignore_bounce = 1; if ((event != APM_NORMAL_RESUME) || (ignore_normal_resume == 0)) { - device_resume(); + device_resume(PMSG_RESUME); queue_event(event, NULL); } ignore_normal_resume = 0; diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 45cc3d9eacb..d571204aaff 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -12,11 +12,9 @@ * and add it to the list of power-controlled devices. sysfs entries for * controlling device power management will also be added. * - * A different set of lists than the global subsystem list are used to - * keep track of power info because we use different lists to hold - * devices based on what stage of the power management process they - * are in. The power domain dependencies may also differ from the - * ancestral dependencies that the subsystem list maintains. + * A separate list is used for keeping track of power info, because the power + * domain dependencies may differ from the ancestral dependencies that the + * subsystem list maintains. */ #include @@ -30,31 +28,40 @@ #include "power.h" /* - * The entries in the dpm_active list are in a depth first order, simply + * The entries in the dpm_list list are in a depth first order, simply * because children are guaranteed to be discovered after parents, and * are inserted at the back of the list on discovery. * - * All the other lists are kept in the same order, for consistency. - * However the lists aren't always traversed in the same order. - * Semaphores must be acquired from the top (i.e., front) down - * and released in the opposite order. Devices must be suspended - * from the bottom (i.e., end) up and resumed in the opposite order. - * That way no parent will be suspended while it still has an active - * child. - * * Since device_pm_add() may be called with a device semaphore held, * we must never try to acquire a device semaphore while holding * dpm_list_mutex. */ -LIST_HEAD(dpm_active); -static LIST_HEAD(dpm_off); -static LIST_HEAD(dpm_off_irq); +LIST_HEAD(dpm_list); static DEFINE_MUTEX(dpm_list_mtx); -/* 'true' if all devices have been suspended, protected by dpm_list_mtx */ -static bool all_sleeping; +/* + * Set once the preparation of devices for a PM transition has started, reset + * before starting to resume devices. Protected by dpm_list_mtx. + */ +static bool transition_started; + +/** + * device_pm_lock - lock the list of active devices used by the PM core + */ +void device_pm_lock(void) +{ + mutex_lock(&dpm_list_mtx); +} + +/** + * device_pm_unlock - unlock the list of active devices used by the PM core + */ +void device_pm_unlock(void) +{ + mutex_unlock(&dpm_list_mtx); +} /** * device_pm_add - add a device to the list of active devices @@ -68,17 +75,25 @@ int device_pm_add(struct device *dev) dev->bus ? dev->bus->name : "No Bus", kobject_name(&dev->kobj)); mutex_lock(&dpm_list_mtx); - if ((dev->parent && dev->parent->power.sleeping) || all_sleeping) { - if (dev->parent->power.sleeping) - dev_warn(dev, "parent %s is sleeping\n", + if (dev->parent) { + if (dev->parent->power.status >= DPM_SUSPENDING) { + dev_warn(dev, "parent %s is sleeping, will not add\n", dev->parent->bus_id); - else - dev_warn(dev, "all devices are sleeping\n"); + WARN_ON(true); + } + } else if (transition_started) { + /* + * We refuse to register parentless devices while a PM + * transition is in progress in order to avoid leaving them + * unhandled down the road + */ WARN_ON(true); } error = dpm_sysfs_add(dev); - if (!error) - list_add_tail(&dev->power.entry, &dpm_active); + if (!error) { + dev->power.status = DPM_ON; + list_add_tail(&dev->power.entry, &dpm_list); + } mutex_unlock(&dpm_list_mtx); return error; } @@ -100,73 +115,243 @@ void device_pm_remove(struct device *dev) mutex_unlock(&dpm_list_mtx); } +/** + * pm_op - execute the PM operation appropiate for given PM event + * @dev: Device. + * @ops: PM operations to choose from. + * @state: PM transition of the system being carried out. + */ +static int pm_op(struct device *dev, struct pm_ops *ops, pm_message_t state) +{ + int error = 0; + + switch (state.event) { +#ifdef CONFIG_SUSPEND + case PM_EVENT_SUSPEND: + if (ops->suspend) { + error = ops->suspend(dev); + suspend_report_result(ops->suspend, error); + } + break; + case PM_EVENT_RESUME: + if (ops->resume) { + error = ops->resume(dev); + suspend_report_result(ops->resume, error); + } + break; +#endif /* CONFIG_SUSPEND */ +#ifdef CONFIG_HIBERNATION + case PM_EVENT_FREEZE: + case PM_EVENT_QUIESCE: + if (ops->freeze) { + error = ops->freeze(dev); + suspend_report_result(ops->freeze, error); + } + break; + case PM_EVENT_HIBERNATE: + if (ops->poweroff) { + error = ops->poweroff(dev); + suspend_report_result(ops->poweroff, error); + } + break; + case PM_EVENT_THAW: + case PM_EVENT_RECOVER: + if (ops->thaw) { + error = ops->thaw(dev); + suspend_report_result(ops->thaw, error); + } + break; + case PM_EVENT_RESTORE: + if (ops->restore) { + error = ops->restore(dev); + suspend_report_result(ops->restore, error); + } + break; +#endif /* CONFIG_HIBERNATION */ + default: + error = -EINVAL; + } + return error; +} + +/** + * pm_noirq_op - execute the PM operation appropiate for given PM event + * @dev: Device. + * @ops: PM operations to choose from. + * @state: PM transition of the system being carried out. + * + * The operation is executed with interrupts disabled by the only remaining + * functional CPU in the system. + */ +static int pm_noirq_op(struct device *dev, struct pm_ext_ops *ops, + pm_message_t state) +{ + int error = 0; + + switch (state.event) { +#ifdef CONFIG_SUSPEND + case PM_EVENT_SUSPEND: + if (ops->suspend_noirq) { + error = ops->suspend_noirq(dev); + suspend_report_result(ops->suspend_noirq, error); + } + break; + case PM_EVENT_RESUME: + if (ops->resume_noirq) { + error = ops->resume_noirq(dev); + suspend_report_result(ops->resume_noirq, error); + } + break; +#endif /* CONFIG_SUSPEND */ +#ifdef CONFIG_HIBERNATION + case PM_EVENT_FREEZE: + case PM_EVENT_QUIESCE: + if (ops->freeze_noirq) { + error = ops->freeze_noirq(dev); + suspend_report_result(ops->freeze_noirq, error); + } + break; + case PM_EVENT_HIBERNATE: + if (ops->poweroff_noirq) { + error = ops->poweroff_noirq(dev); + suspend_report_result(ops->poweroff_noirq, error); + } + break; + case PM_EVENT_THAW: + case PM_EVENT_RECOVER: + if (ops->thaw_noirq) { + error = ops->thaw_noirq(dev); + suspend_report_result(ops->thaw_noirq, error); + } + break; + case PM_EVENT_RESTORE: + if (ops->restore_noirq) { + error = ops->restore_noirq(dev); + suspend_report_result(ops->restore_noirq, error); + } + break; +#endif /* CONFIG_HIBERNATION */ + default: + error = -EINVAL; + } + return error; +} + +static char *pm_verb(int event) +{ + switch (event) { + case PM_EVENT_SUSPEND: + return "suspend"; + case PM_EVENT_RESUME: + return "resume"; + case PM_EVENT_FREEZE: + return "freeze"; + case PM_EVENT_QUIESCE: + return "quiesce"; + case PM_EVENT_HIBERNATE: + return "hibernate"; + case PM_EVENT_THAW: + return "thaw"; + case PM_EVENT_RESTORE: + return "restore"; + case PM_EVENT_RECOVER: + return "recover"; + default: + return "(unknown PM event)"; + } +} + +static void pm_dev_dbg(struct device *dev, pm_message_t state, char *info) +{ + dev_dbg(dev, "%s%s%s\n", info, pm_verb(state.event), + ((state.event & PM_EVENT_SLEEP) && device_may_wakeup(dev)) ? + ", may wakeup" : ""); +} + +static void pm_dev_err(struct device *dev, pm_message_t state, char *info, + int error) +{ + printk(KERN_ERR "PM: Device %s failed to %s%s: error %d\n", + kobject_name(&dev->kobj), pm_verb(state.event), info, error); +} + /*------------------------- Resume routines -------------------------*/ /** - * resume_device_early - Power on one device (early resume). + * resume_device_noirq - Power on one device (early resume). * @dev: Device. + * @state: PM transition of the system being carried out. * * Must be called with interrupts disabled. */ -static int resume_device_early(struct device *dev) +static int resume_device_noirq(struct device *dev, pm_message_t state) { int error = 0; TRACE_DEVICE(dev); TRACE_RESUME(0); - if (dev->bus && dev->bus->resume_early) { - dev_dbg(dev, "EARLY resume\n"); + if (!dev->bus) + goto End; + + if (dev->bus->pm) { + pm_dev_dbg(dev, state, "EARLY "); + error = pm_noirq_op(dev, dev->bus->pm, state); + } else if (dev->bus->resume_early) { + pm_dev_dbg(dev, state, "legacy EARLY "); error = dev->bus->resume_early(dev); } - + End: TRACE_RESUME(error); return error; } /** * dpm_power_up - Power on all regular (non-sysdev) devices. + * @state: PM transition of the system being carried out. * - * Walk the dpm_off_irq list and power each device up. This - * is used for devices that required they be powered down with - * interrupts disabled. As devices are powered on, they are moved - * to the dpm_off list. + * Execute the appropriate "noirq resume" callback for all devices marked + * as DPM_OFF_IRQ. * * Must be called with interrupts disabled and only one CPU running. */ -static void dpm_power_up(void) +static void dpm_power_up(pm_message_t state) { + struct device *dev; - while (!list_empty(&dpm_off_irq)) { - struct list_head *entry = dpm_off_irq.next; - struct device *dev = to_device(entry); + list_for_each_entry(dev, &dpm_list, power.entry) + if (dev->power.status > DPM_OFF) { + int error; - list_move_tail(entry, &dpm_off); - resume_device_early(dev); - } + dev->power.status = DPM_OFF; + error = resume_device_noirq(dev, state); + if (error) + pm_dev_err(dev, state, " early", error); + } } /** * device_power_up - Turn on all devices that need special attention. + * @state: PM transition of the system being carried out. * * Power on system devices, then devices that required we shut them down * with interrupts disabled. * * Must be called with interrupts disabled. */ -void device_power_up(void) +void device_power_up(pm_message_t state) { sysdev_resume(); - dpm_power_up(); + dpm_power_up(state); } EXPORT_SYMBOL_GPL(device_power_up); /** * resume_device - Restore state for one device. * @dev: Device. - * + * @state: PM transition of the system being carried out. */ -static int resume_device(struct device *dev) +static int resume_device(struct device *dev, pm_message_t state) { int error = 0; @@ -175,21 +360,40 @@ static int resume_device(struct device *dev) down(&dev->sem); - if (dev->bus && dev->bus->resume) { - dev_dbg(dev,"resuming\n"); - error = dev->bus->resume(dev); + if (dev->bus) { + if (dev->bus->pm) { + pm_dev_dbg(dev, state, ""); + error = pm_op(dev, &dev->bus->pm->base, state); + } else if (dev->bus->resume) { + pm_dev_dbg(dev, state, "legacy "); + error = dev->bus->resume(dev); + } + if (error) + goto End; } - if (!error && dev->type && dev->type->resume) { - dev_dbg(dev,"resuming\n"); - error = dev->type->resume(dev); + if (dev->type) { + if (dev->type->pm) { + pm_dev_dbg(dev, state, "type "); + error = pm_op(dev, dev->type->pm, state); + } else if (dev->type->resume) { + pm_dev_dbg(dev, state, "legacy type "); + error = dev->type->resume(dev); + } + if (error) + goto End; } - if (!error && dev->class && dev->class->resume) { - dev_dbg(dev,"class resume\n"); - error = dev->class->resume(dev); + if (dev->class) { + if (dev->class->pm) { + pm_dev_dbg(dev, state, "class "); + error = pm_op(dev, dev->class->pm, state); + } else if (dev->class->resume) { + pm_dev_dbg(dev, state, "legacy class "); + error = dev->class->resume(dev); + } } - + End: up(&dev->sem); TRACE_RESUME(error); @@ -198,78 +402,161 @@ static int resume_device(struct device *dev) /** * dpm_resume - Resume every device. + * @state: PM transition of the system being carried out. * - * Resume the devices that have either not gone through - * the late suspend, or that did go through it but also - * went through the early resume. + * Execute the appropriate "resume" callback for all devices the status of + * which indicates that they are inactive. + */ +static void dpm_resume(pm_message_t state) +{ + struct list_head list; + + INIT_LIST_HEAD(&list); + mutex_lock(&dpm_list_mtx); + transition_started = false; + while (!list_empty(&dpm_list)) { + struct device *dev = to_device(dpm_list.next); + + get_device(dev); + if (dev->power.status >= DPM_OFF) { + int error; + + dev->power.status = DPM_RESUMING; + mutex_unlock(&dpm_list_mtx); + + error = resume_device(dev, state); + + mutex_lock(&dpm_list_mtx); + if (error) + pm_dev_err(dev, state, "", error); + } else if (dev->power.status == DPM_SUSPENDING) { + /* Allow new children of the device to be registered */ + dev->power.status = DPM_RESUMING; + } + if (!list_empty(&dev->power.entry)) + list_move_tail(&dev->power.entry, &list); + put_device(dev); + } + list_splice(&list, &dpm_list); + mutex_unlock(&dpm_list_mtx); +} + +/** + * complete_device - Complete a PM transition for given device + * @dev: Device. + * @state: PM transition of the system being carried out. + */ +static void complete_device(struct device *dev, pm_message_t state) +{ + down(&dev->sem); + + if (dev->class && dev->class->pm && dev->class->pm->complete) { + pm_dev_dbg(dev, state, "completing class "); + dev->class->pm->complete(dev); + } + + if (dev->type && dev->type->pm && dev->type->pm->complete) { + pm_dev_dbg(dev, state, "completing type "); + dev->type->pm->complete(dev); + } + + if (dev->bus && dev->bus->pm && dev->bus->pm->base.complete) { + pm_dev_dbg(dev, state, "completing "); + dev->bus->pm->base.complete(dev); + } + + up(&dev->sem); +} + +/** + * dpm_complete - Complete a PM transition for all devices. + * @state: PM transition of the system being carried out. * - * Take devices from the dpm_off_list, resume them, - * and put them on the dpm_locked list. + * Execute the ->complete() callbacks for all devices that are not marked + * as DPM_ON. */ -static void dpm_resume(void) +static void dpm_complete(pm_message_t state) { + struct list_head list; + + INIT_LIST_HEAD(&list); mutex_lock(&dpm_list_mtx); - all_sleeping = false; - while(!list_empty(&dpm_off)) { - struct list_head *entry = dpm_off.next; - struct device *dev = to_device(entry); + while (!list_empty(&dpm_list)) { + struct device *dev = to_device(dpm_list.prev); - list_move_tail(entry, &dpm_active); - dev->power.sleeping = false; - mutex_unlock(&dpm_list_mtx); - resume_device(dev); - mutex_lock(&dpm_list_mtx); + get_device(dev); + if (dev->power.status > DPM_ON) { + dev->power.status = DPM_ON; + mutex_unlock(&dpm_list_mtx); + + complete_device(dev, state); + + mutex_lock(&dpm_list_mtx); + } + if (!list_empty(&dev->power.entry)) + list_move(&dev->power.entry, &list); + put_device(dev); } + list_splice(&list, &dpm_list); mutex_unlock(&dpm_list_mtx); } /** * device_resume - Restore state of each device in system. + * @state: PM transition of the system being carried out. * * Resume all the devices, unlock them all, and allow new * devices to be registered once again. */ -void device_resume(void) +void device_resume(pm_message_t state) { might_sleep(); - dpm_resume(); + dpm_resume(state); + dpm_complete(state); } EXPORT_SYMBOL_GPL(device_resume); /*------------------------- Suspend routines -------------------------*/ -static inline char *suspend_verb(u32 event) +/** + * resume_event - return a PM message representing the resume event + * corresponding to given sleep state. + * @sleep_state: PM message representing a sleep state. + */ +static pm_message_t resume_event(pm_message_t sleep_state) { - switch (event) { - case PM_EVENT_SUSPEND: return "suspend"; - case PM_EVENT_FREEZE: return "freeze"; - case PM_EVENT_PRETHAW: return "prethaw"; - default: return "(unknown suspend event)"; + switch (sleep_state.event) { + case PM_EVENT_SUSPEND: + return PMSG_RESUME; + case PM_EVENT_FREEZE: + case PM_EVENT_QUIESCE: + return PMSG_RECOVER; + case PM_EVENT_HIBERNATE: + return PMSG_RESTORE; } -} - -static void -suspend_device_dbg(struct device *dev, pm_message_t state, char *info) -{ - dev_dbg(dev, "%s%s%s\n", info, suspend_verb(state.event), - ((state.event == PM_EVENT_SUSPEND) && device_may_wakeup(dev)) ? - ", may wakeup" : ""); + return PMSG_ON; } /** - * suspend_device_late - Shut down one device (late suspend). + * suspend_device_noirq - Shut down one device (late suspend). * @dev: Device. - * @state: Power state device is entering. + * @state: PM transition of the system being carried out. * * This is called with interrupts off and only a single CPU running. */ -static int suspend_device_late(struct device *dev, pm_message_t state) +static int suspend_device_noirq(struct device *dev, pm_message_t state) { int error = 0; - if (dev->bus && dev->bus->suspend_late) { - suspend_device_dbg(dev, state, "LATE "); + if (!dev->bus) + return 0; + + if (dev->bus->pm) { + pm_dev_dbg(dev, state, "LATE "); + error = pm_noirq_op(dev, dev->bus->pm, state); + } else if (dev->bus->suspend_late) { + pm_dev_dbg(dev, state, "legacy LATE "); error = dev->bus->suspend_late(dev, state); suspend_report_result(dev->bus->suspend_late, error); } @@ -278,37 +565,30 @@ static int suspend_device_late(struct device *dev, pm_message_t state) /** * device_power_down - Shut down special devices. - * @state: Power state to enter. + * @state: PM transition of the system being carried out. * - * Power down devices that require interrupts to be disabled - * and move them from the dpm_off list to the dpm_off_irq list. + * Power down devices that require interrupts to be disabled. * Then power down system devices. * * Must be called with interrupts disabled and only one CPU running. */ int device_power_down(pm_message_t state) { + struct device *dev; int error = 0; - while (!list_empty(&dpm_off)) { - struct list_head *entry = dpm_off.prev; - struct device *dev = to_device(entry); - - error = suspend_device_late(dev, state); + list_for_each_entry_reverse(dev, &dpm_list, power.entry) { + error = suspend_device_noirq(dev, state); if (error) { - printk(KERN_ERR "Could not power down device %s: " - "error %d\n", - kobject_name(&dev->kobj), error); + pm_dev_err(dev, state, " late", error); break; } - if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_off_irq); + dev->power.status = DPM_OFF_IRQ; } - if (!error) error = sysdev_suspend(state); if (error) - dpm_power_up(); + dpm_power_up(resume_event(state)); return error; } EXPORT_SYMBOL_GPL(device_power_down); @@ -316,7 +596,7 @@ EXPORT_SYMBOL_GPL(device_power_down); /** * suspend_device - Save state of one device. * @dev: Device. - * @state: Power state device is entering. + * @state: PM transition of the system being carried out. */ static int suspend_device(struct device *dev, pm_message_t state) { @@ -324,24 +604,43 @@ static int suspend_device(struct device *dev, pm_message_t state) down(&dev->sem); - if (dev->class && dev->class->suspend) { - suspend_device_dbg(dev, state, "class "); - error = dev->class->suspend(dev, state); - suspend_report_result(dev->class->suspend, error); + if (dev->class) { + if (dev->class->pm) { + pm_dev_dbg(dev, state, "class "); + error = pm_op(dev, dev->class->pm, state); + } else if (dev->class->suspend) { + pm_dev_dbg(dev, state, "legacy class "); + error = dev->class->suspend(dev, state); + suspend_report_result(dev->class->suspend, error); + } + if (error) + goto End; } - if (!error && dev->type && dev->type->suspend) { - suspend_device_dbg(dev, state, "type "); - error = dev->type->suspend(dev, state); - suspend_report_result(dev->type->suspend, error); + if (dev->type) { + if (dev->type->pm) { + pm_dev_dbg(dev, state, "type "); + error = pm_op(dev, dev->type->pm, state); + } else if (dev->type->suspend) { + pm_dev_dbg(dev, state, "legacy type "); + error = dev->type->suspend(dev, state); + suspend_report_result(dev->type->suspend, error); + } + if (error) + goto End; } - if (!error && dev->bus && dev->bus->suspend) { - suspend_device_dbg(dev, state, ""); - error = dev->bus->suspend(dev, state); - suspend_report_result(dev->bus->suspend, error); + if (dev->bus) { + if (dev->bus->pm) { + pm_dev_dbg(dev, state, ""); + error = pm_op(dev, &dev->bus->pm->base, state); + } else if (dev->bus->suspend) { + pm_dev_dbg(dev, state, "legacy "); + error = dev->bus->suspend(dev, state); + suspend_report_result(dev->bus->suspend, error); + } } - + End: up(&dev->sem); return error; @@ -349,67 +648,141 @@ static int suspend_device(struct device *dev, pm_message_t state) /** * dpm_suspend - Suspend every device. - * @state: Power state to put each device in. + * @state: PM transition of the system being carried out. * - * Walk the dpm_locked list. Suspend each device and move it - * to the dpm_off list. - * - * (For historical reasons, if it returns -EAGAIN, that used to mean - * that the device would be called again with interrupts disabled. - * These days, we use the "suspend_late()" callback for that, so we - * print a warning and consider it an error). + * Execute the appropriate "suspend" callbacks for all devices. */ static int dpm_suspend(pm_message_t state) { + struct list_head list; int error = 0; + INIT_LIST_HEAD(&list); mutex_lock(&dpm_list_mtx); - while (!list_empty(&dpm_active)) { - struct list_head *entry = dpm_active.prev; - struct device *dev = to_device(entry); - - WARN_ON(dev->parent && dev->parent->power.sleeping); + while (!list_empty(&dpm_list)) { + struct device *dev = to_device(dpm_list.prev); - dev->power.sleeping = true; + get_device(dev); mutex_unlock(&dpm_list_mtx); + error = suspend_device(dev, state); + mutex_lock(&dpm_list_mtx); if (error) { - printk(KERN_ERR "Could not suspend device %s: " - "error %d%s\n", - kobject_name(&dev->kobj), - error, - (error == -EAGAIN ? - " (please convert to suspend_late)" : - "")); - dev->power.sleeping = false; + pm_dev_err(dev, state, "", error); + put_device(dev); break; } + dev->power.status = DPM_OFF; if (!list_empty(&dev->power.entry)) - list_move(&dev->power.entry, &dpm_off); + list_move(&dev->power.entry, &list); + put_device(dev); } - if (!error) - all_sleeping = true; + list_splice(&list, dpm_list.prev); mutex_unlock(&dpm_list_mtx); + return error; +} + +/** + * prepare_device - Execute the ->prepare() callback(s) for given device. + * @dev: Device. + * @state: PM transition of the system being carried out. + */ +static int prepare_device(struct device *dev, pm_message_t state) +{ + int error = 0; + + down(&dev->sem); + + if (dev->bus && dev->bus->pm && dev->bus->pm->base.prepare) { + pm_dev_dbg(dev, state, "preparing "); + error = dev->bus->pm->base.prepare(dev); + suspend_report_result(dev->bus->pm->base.prepare, error); + if (error) + goto End; + } + + if (dev->type && dev->type->pm && dev->type->pm->prepare) { + pm_dev_dbg(dev, state, "preparing type "); + error = dev->type->pm->prepare(dev); + suspend_report_result(dev->type->pm->prepare, error); + if (error) + goto End; + } + + if (dev->class && dev->class->pm && dev->class->pm->prepare) { + pm_dev_dbg(dev, state, "preparing class "); + error = dev->class->pm->prepare(dev); + suspend_report_result(dev->class->pm->prepare, error); + } + End: + up(&dev->sem); + + return error; +} +/** + * dpm_prepare - Prepare all devices for a PM transition. + * @state: PM transition of the system being carried out. + * + * Execute the ->prepare() callback for all devices. + */ +static int dpm_prepare(pm_message_t state) +{ + struct list_head list; + int error = 0; + + INIT_LIST_HEAD(&list); + mutex_lock(&dpm_list_mtx); + transition_started = true; + while (!list_empty(&dpm_list)) { + struct device *dev = to_device(dpm_list.next); + + get_device(dev); + dev->power.status = DPM_PREPARING; + mutex_unlock(&dpm_list_mtx); + + error = prepare_device(dev, state); + + mutex_lock(&dpm_list_mtx); + if (error) { + dev->power.status = DPM_ON; + if (error == -EAGAIN) { + put_device(dev); + continue; + } + printk(KERN_ERR "PM: Failed to prepare device %s " + "for power transition: error %d\n", + kobject_name(&dev->kobj), error); + put_device(dev); + break; + } + dev->power.status = DPM_SUSPENDING; + if (!list_empty(&dev->power.entry)) + list_move_tail(&dev->power.entry, &list); + put_device(dev); + } + list_splice(&list, &dpm_list); + mutex_unlock(&dpm_list_mtx); return error; } /** * device_suspend - Save state and stop all devices in system. - * @state: new power management state + * @state: PM transition of the system being carried out. * - * Prevent new devices from being registered, then lock all devices - * and suspend them. + * Prepare and suspend all devices. */ int device_suspend(pm_message_t state) { int error; might_sleep(); - error = dpm_suspend(state); + error = dpm_prepare(state); + if (!error) + error = dpm_suspend(state); if (error) - device_resume(); + device_resume(resume_event(state)); return error; } EXPORT_SYMBOL_GPL(device_suspend); diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index a6894f2a4b9..a3252c0e288 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -4,7 +4,7 @@ * main.c */ -extern struct list_head dpm_active; /* The active device list */ +extern struct list_head dpm_list; /* The active device list */ static inline struct device *to_device(struct list_head *entry) { diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c index 2b4b392dcbc..8c1e656b5f8 100644 --- a/drivers/base/power/trace.c +++ b/drivers/base/power/trace.c @@ -188,9 +188,9 @@ static int show_file_hash(unsigned int value) static int show_dev_hash(unsigned int value) { int match = 0; - struct list_head * entry = dpm_active.prev; + struct list_head *entry = dpm_list.prev; - while (entry != &dpm_active) { + while (entry != &dpm_list) { struct device * dev = to_device(entry); unsigned int hash = hash_string(DEVSEED, dev->bus_id, DEVHASH); if (hash == value) { diff --git a/include/linux/device.h b/include/linux/device.h index 6a2d04c011b..f71a78d123a 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -68,6 +68,8 @@ struct bus_type { int (*resume_early)(struct device *dev); int (*resume)(struct device *dev); + struct pm_ext_ops *pm; + struct bus_type_private *p; }; @@ -131,6 +133,8 @@ struct device_driver { int (*resume) (struct device *dev); struct attribute_group **groups; + struct pm_ops *pm; + struct driver_private *p; }; @@ -197,6 +201,8 @@ struct class { int (*suspend)(struct device *dev, pm_message_t state); int (*resume)(struct device *dev); + + struct pm_ops *pm; }; extern int __must_check class_register(struct class *class); @@ -248,8 +254,11 @@ struct device_type { struct attribute_group **groups; int (*uevent)(struct device *dev, struct kobj_uevent_env *env); void (*release)(struct device *dev); + int (*suspend)(struct device *dev, pm_message_t state); int (*resume)(struct device *dev); + + struct pm_ops *pm; }; /* interface for exporting device attributes */ diff --git a/include/linux/pm.h b/include/linux/pm.h index 39a7ee859b6..4ad9de94449 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -112,7 +112,9 @@ typedef struct pm_message { int event; } pm_message_t; -/* +/** + * struct pm_ops - device PM callbacks + * * Several driver power state transitions are externally visible, affecting * the state of pending I/O queues and (for drivers that touch hardware) * interrupts, wakeups, DMA, and other hardware state. There may also be @@ -120,6 +122,284 @@ typedef struct pm_message { * to the rest of the driver stack (such as a driver that's ON gating off * clocks which are not in active use). * + * The externally visible transitions are handled with the help of the following + * callbacks included in this structure: + * + * @prepare: Prepare the device for the upcoming transition, but do NOT change + * its hardware state. Prevent new children of the device from being + * registered after @prepare() returns (the driver's subsystem and + * generally the rest of the kernel is supposed to prevent new calls to the + * probe method from being made too once @prepare() has succeeded). If + * @prepare() detects a situation it cannot handle (e.g. registration of a + * child already in progress), it may return -EAGAIN, so that the PM core + * can execute it once again (e.g. after the new child has been registered) + * to recover from the race condition. This method is executed for all + * kinds of suspend transitions and is followed by one of the suspend + * callbacks: @suspend(), @freeze(), or @poweroff(). + * The PM core executes @prepare() for all devices before starting to + * execute suspend callbacks for any of them, so drivers may assume all of + * the other devices to be present and functional while @prepare() is being + * executed. In particular, it is safe to make GFP_KERNEL memory + * allocations from within @prepare(). However, drivers may NOT assume + * anything about the availability of the user space at that time and it + * is not correct to request firmware from within @prepare() (it's too + * late to do that). [To work around this limitation, drivers may + * register suspend and hibernation notifiers that are executed before the + * freezing of tasks.] + * + * @complete: Undo the changes made by @prepare(). This method is executed for + * all kinds of resume transitions, following one of the resume callbacks: + * @resume(), @thaw(), @restore(). Also called if the state transition + * fails before the driver's suspend callback (@suspend(), @freeze(), + * @poweroff()) can be executed (e.g. if the suspend callback fails for one + * of the other devices that the PM core has unsuccessfully attempted to + * suspend earlier). + * The PM core executes @complete() after it has executed the appropriate + * resume callback for all devices. + * + * @suspend: Executed before putting the system into a sleep state in which the + * contents of main memory are preserved. Quiesce the device, put it into + * a low power state appropriate for the upcoming system state (such as + * PCI_D3hot), and enable wakeup events as appropriate. + * + * @resume: Executed after waking the system up from a sleep state in which the + * contents of main memory were preserved. Put the device into the + * appropriate state, according to the information saved in memory by the + * preceding @suspend(). The driver starts working again, responding to + * hardware events and software requests. The hardware may have gone + * through a power-off reset, or it may have maintained state from the + * previous suspend() which the driver may rely on while resuming. On most + * platforms, there are no restrictions on availability of resources like + * clocks during @resume(). + * + * @freeze: Hibernation-specific, executed before creating a hibernation image. + * Quiesce operations so that a consistent image can be created, but do NOT + * otherwise put the device into a low power device state and do NOT emit + * system wakeup events. Save in main memory the device settings to be + * used by @restore() during the subsequent resume from hibernation or by + * the subsequent @thaw(), if the creation of the image or the restoration + * of main memory contents from it fails. + * + * @thaw: Hibernation-specific, executed after creating a hibernation image OR + * if the creation of the image fails. Also executed after a failing + * attempt to restore the contents of main memory from such an image. + * Undo the changes made by the preceding @freeze(), so the device can be + * operated in the same way as immediately before the call to @freeze(). + * + * @poweroff: Hibernation-specific, executed after saving a hibernation image. + * Quiesce the device, put it into a low power state appropriate for the + * upcoming system state (such as PCI_D3hot), and enable wakeup events as + * appropriate. + * + * @restore: Hibernation-specific, executed after restoring the contents of main + * memory from a hibernation image. Driver starts working again, + * responding to hardware events and software requests. Drivers may NOT + * make ANY assumptions about the hardware state right prior to @restore(). + * On most platforms, there are no restrictions on availability of + * resources like clocks during @restore(). + * + * All of the above callbacks, except for @complete(), return error codes. + * However, the error codes returned by the resume operations, @resume(), + * @thaw(), and @restore(), do not cause the PM core to abort the resume + * transition during which they are returned. The error codes returned in + * that cases are only printed by the PM core to the system logs for debugging + * purposes. Still, it is recommended that drivers only return error codes + * from their resume methods in case of an unrecoverable failure (i.e. when the + * device being handled refuses to resume and becomes unusable) to allow us to + * modify the PM core in the future, so that it can avoid attempting to handle + * devices that failed to resume and their children. + * + * It is allowed to unregister devices while the above callbacks are being + * executed. However, it is not allowed to unregister a device from within any + * of its own callbacks. + */ + +struct pm_ops { + int (*prepare)(struct device *dev); + void (*complete)(struct device *dev); + int (*suspend)(struct device *dev); + int (*resume)(struct device *dev); + int (*freeze)(struct device *dev); + int (*thaw)(struct device *dev); + int (*poweroff)(struct device *dev); + int (*restore)(struct device *dev); +}; + +/** + * struct pm_ext_ops - extended device PM callbacks + * + * Some devices require certain operations related to suspend and hibernation + * to be carried out with interrupts disabled. Thus, 'struct pm_ext_ops' below + * is defined, adding callbacks to be executed with interrupts disabled to + * 'struct pm_ops'. + * + * The following callbacks included in 'struct pm_ext_ops' are executed with + * the nonboot CPUs switched off and with interrupts disabled on the only + * functional CPU. They also are executed with the PM core list of devices + * locked, so they must NOT unregister any devices. + * + * @suspend_noirq: Complete the operations of ->suspend() by carrying out any + * actions required for suspending the device that need interrupts to be + * disabled + * + * @resume_noirq: Prepare for the execution of ->resume() by carrying out any + * actions required for resuming the device that need interrupts to be + * disabled + * + * @freeze_noirq: Complete the operations of ->freeze() by carrying out any + * actions required for freezing the device that need interrupts to be + * disabled + * + * @thaw_noirq: Prepare for the execution of ->thaw() by carrying out any + * actions required for thawing the device that need interrupts to be + * disabled + * + * @poweroff_noirq: Complete the operations of ->poweroff() by carrying out any + * actions required for handling the device that need interrupts to be + * disabled + * + * @restore_noirq: Prepare for the execution of ->restore() by carrying out any + * actions required for restoring the operations of the device that need + * interrupts to be disabled + * + * All of the above callbacks return error codes, but the error codes returned + * by the resume operations, @resume_noirq(), @thaw_noirq(), and + * @restore_noirq(), do not cause the PM core to abort the resume transition + * during which they are returned. The error codes returned in that cases are + * only printed by the PM core to the system logs for debugging purposes. + * Still, as stated above, it is recommended that drivers only return error + * codes from their resume methods if the device being handled fails to resume + * and is not usable any more. + */ + +struct pm_ext_ops { + struct pm_ops base; + int (*suspend_noirq)(struct device *dev); + int (*resume_noirq)(struct device *dev); + int (*freeze_noirq)(struct device *dev); + int (*thaw_noirq)(struct device *dev); + int (*poweroff_noirq)(struct device *dev); + int (*restore_noirq)(struct device *dev); +}; + +/** + * PM_EVENT_ messages + * + * The following PM_EVENT_ messages are defined for the internal use of the PM + * core, in order to provide a mechanism allowing the high level suspend and + * hibernation code to convey the necessary information to the device PM core + * code: + * + * ON No transition. + * + * FREEZE System is going to hibernate, call ->prepare() and ->freeze() + * for all devices. + * + * SUSPEND System is going to suspend, call ->prepare() and ->suspend() + * for all devices. + * + * HIBERNATE Hibernation image has been saved, call ->prepare() and + * ->poweroff() for all devices. + * + * QUIESCE Contents of main memory are going to be restored from a (loaded) + * hibernation image, call ->prepare() and ->freeze() for all + * devices. + * + * RESUME System is resuming, call ->resume() and ->complete() for all + * devices. + * + * THAW Hibernation image has been created, call ->thaw() and + * ->complete() for all devices. + * + * RESTORE Contents of main memory have been restored from a hibernation + * image, call ->restore() and ->complete() for all devices. + * + * RECOVER Creation of a hibernation image or restoration of the main + * memory contents from a hibernation image has failed, call + * ->thaw() and ->complete() for all devices. + */ + +#define PM_EVENT_ON 0x0000 +#define PM_EVENT_FREEZE 0x0001 +#define PM_EVENT_SUSPEND 0x0002 +#define PM_EVENT_HIBERNATE 0x0004 +#define PM_EVENT_QUIESCE 0x0008 +#define PM_EVENT_RESUME 0x0010 +#define PM_EVENT_THAW 0x0020 +#define PM_EVENT_RESTORE 0x0040 +#define PM_EVENT_RECOVER 0x0080 + +#define PM_EVENT_SLEEP (PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE) + +#define PMSG_FREEZE ((struct pm_message){ .event = PM_EVENT_FREEZE, }) +#define PMSG_QUIESCE ((struct pm_message){ .event = PM_EVENT_QUIESCE, }) +#define PMSG_SUSPEND ((struct pm_message){ .event = PM_EVENT_SUSPEND, }) +#define PMSG_HIBERNATE ((struct pm_message){ .event = PM_EVENT_HIBERNATE, }) +#define PMSG_RESUME ((struct pm_message){ .event = PM_EVENT_RESUME, }) +#define PMSG_THAW ((struct pm_message){ .event = PM_EVENT_THAW, }) +#define PMSG_RESTORE ((struct pm_message){ .event = PM_EVENT_RESTORE, }) +#define PMSG_RECOVER ((struct pm_message){ .event = PM_EVENT_RECOVER, }) +#define PMSG_ON ((struct pm_message){ .event = PM_EVENT_ON, }) + +/** + * Device power management states + * + * These state labels are used internally by the PM core to indicate the current + * status of a device with respect to the PM core operations. + * + * DPM_ON Device is regarded as operational. Set this way + * initially and when ->complete() is about to be called. + * Also set when ->prepare() fails. + * + * DPM_PREPARING Device is going to be prepared for a PM transition. Set + * when ->prepare() is about to be called. + * + * DPM_RESUMING Device is going to be resumed. Set when ->resume(), + * ->thaw(), or ->restore() is about to be called. + * + * DPM_SUSPENDING Device has been prepared for a power transition. Set + * when ->prepare() has just succeeded. + * + * DPM_OFF Device is regarded as inactive. Set immediately after + * ->suspend(), ->freeze(), or ->poweroff() has succeeded. + * Also set when ->resume()_noirq, ->thaw_noirq(), or + * ->restore_noirq() is about to be called. + * + * DPM_OFF_IRQ Device is in a "deep sleep". Set immediately after + * ->suspend_noirq(), ->freeze_noirq(), or + * ->poweroff_noirq() has just succeeded. + */ + +enum dpm_state { + DPM_INVALID, + DPM_ON, + DPM_PREPARING, + DPM_RESUMING, + DPM_SUSPENDING, + DPM_OFF, + DPM_OFF_IRQ, +}; + +struct dev_pm_info { + pm_message_t power_state; + unsigned can_wakeup:1; + unsigned should_wakeup:1; + enum dpm_state status; /* Owned by the PM core */ +#ifdef CONFIG_PM_SLEEP + struct list_head entry; +#endif +}; + +/* + * The PM_EVENT_ messages are also used by drivers implementing the legacy + * suspend framework, based on the ->suspend() and ->resume() callbacks common + * for suspend and hibernation transitions, according to the rules below. + */ + +/* Necessary, because several drivers use PM_EVENT_PRETHAW */ +#define PM_EVENT_PRETHAW PM_EVENT_QUIESCE + +/* * One transition is triggered by resume(), after a suspend() call; the * message is implicit: * @@ -164,35 +444,13 @@ typedef struct pm_message { * or from system low-power states such as standby or suspend-to-RAM. */ -#define PM_EVENT_ON 0 -#define PM_EVENT_FREEZE 1 -#define PM_EVENT_SUSPEND 2 -#define PM_EVENT_HIBERNATE 4 -#define PM_EVENT_PRETHAW 8 - -#define PM_EVENT_SLEEP (PM_EVENT_SUSPEND | PM_EVENT_HIBERNATE) - -#define PMSG_FREEZE ((struct pm_message){ .event = PM_EVENT_FREEZE, }) -#define PMSG_PRETHAW ((struct pm_message){ .event = PM_EVENT_PRETHAW, }) -#define PMSG_SUSPEND ((struct pm_message){ .event = PM_EVENT_SUSPEND, }) -#define PMSG_HIBERNATE ((struct pm_message){ .event = PM_EVENT_HIBERNATE, }) -#define PMSG_ON ((struct pm_message){ .event = PM_EVENT_ON, }) - -struct dev_pm_info { - pm_message_t power_state; - unsigned can_wakeup:1; - unsigned should_wakeup:1; - bool sleeping:1; /* Owned by the PM core */ -#ifdef CONFIG_PM_SLEEP - struct list_head entry; -#endif -}; +#ifdef CONFIG_PM_SLEEP +extern void device_pm_lock(void); +extern void device_power_up(pm_message_t state); +extern void device_resume(pm_message_t state); +extern void device_pm_unlock(void); extern int device_power_down(pm_message_t state); -extern void device_power_up(void); -extern void device_resume(void); - -#ifdef CONFIG_PM_SLEEP extern int device_suspend(pm_message_t state); extern int device_prepare_suspend(pm_message_t state); diff --git a/kernel/power/disk.c b/kernel/power/disk.c index 14a656cdc65..d416be0efa8 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 6a6d5eb3524..d023b6b584e 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 v1.2.3-70-g09d2 From 25e18499e08cb097cbbfeab5de25d094d5312ee5 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Wed, 21 May 2008 01:40:43 +0200 Subject: Implement new suspend and hibernation callbacks for platform busses Implement new suspend and hibernation callbacks for the platform bus type. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Greg KH Signed-off-by: Jesse Barnes --- drivers/base/platform.c | 296 ++++++++++++++++++++++++++++++++++++++-- include/linux/platform_device.h | 1 + 2 files changed, 289 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/platform.c b/drivers/base/platform.c index 911ec600fe7..3f940393d6c 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -453,6 +453,8 @@ int platform_driver_register(struct platform_driver *drv) drv->driver.suspend = platform_drv_suspend; if (drv->resume) drv->driver.resume = platform_drv_resume; + if (drv->pm) + drv->driver.pm = &drv->pm->base; return driver_register(&drv->driver); } EXPORT_SYMBOL_GPL(platform_driver_register); @@ -560,7 +562,9 @@ static int platform_match(struct device *dev, struct device_driver *drv) return (strncmp(pdev->name, drv->name, BUS_ID_SIZE) == 0); } -static int platform_suspend(struct device *dev, pm_message_t mesg) +#ifdef CONFIG_PM_SLEEP + +static int platform_legacy_suspend(struct device *dev, pm_message_t mesg) { int ret = 0; @@ -570,7 +574,7 @@ static int platform_suspend(struct device *dev, pm_message_t mesg) return ret; } -static int platform_suspend_late(struct device *dev, pm_message_t mesg) +static int platform_legacy_suspend_late(struct device *dev, pm_message_t mesg) { struct platform_driver *drv = to_platform_driver(dev->driver); struct platform_device *pdev; @@ -583,7 +587,7 @@ static int platform_suspend_late(struct device *dev, pm_message_t mesg) return ret; } -static int platform_resume_early(struct device *dev) +static int platform_legacy_resume_early(struct device *dev) { struct platform_driver *drv = to_platform_driver(dev->driver); struct platform_device *pdev; @@ -596,7 +600,7 @@ static int platform_resume_early(struct device *dev) return ret; } -static int platform_resume(struct device *dev) +static int platform_legacy_resume(struct device *dev) { int ret = 0; @@ -606,15 +610,291 @@ static int platform_resume(struct device *dev) return ret; } +static int platform_pm_prepare(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm && drv->pm->prepare) + ret = drv->pm->prepare(dev); + + return ret; +} + +static void platform_pm_complete(struct device *dev) +{ + struct device_driver *drv = dev->driver; + + if (drv && drv->pm && drv->pm->complete) + drv->pm->complete(dev); +} + +#ifdef CONFIG_SUSPEND + +static int platform_pm_suspend(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm) { + if (drv->pm->suspend) + ret = drv->pm->suspend(dev); + } else { + ret = platform_legacy_suspend(dev, PMSG_SUSPEND); + } + + return ret; +} + +static int platform_pm_suspend_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->suspend_noirq) + ret = pdrv->pm->suspend_noirq(dev); + } else { + ret = platform_legacy_suspend_late(dev, PMSG_SUSPEND); + } + + return ret; +} + +static int platform_pm_resume(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm) { + if (drv->pm->resume) + ret = drv->pm->resume(dev); + } else { + ret = platform_legacy_resume(dev); + } + + return ret; +} + +static int platform_pm_resume_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->resume_noirq) + ret = pdrv->pm->resume_noirq(dev); + } else { + ret = platform_legacy_resume_early(dev); + } + + return ret; +} + +#else /* !CONFIG_SUSPEND */ + +#define platform_pm_suspend NULL +#define platform_pm_resume NULL +#define platform_pm_suspend_noirq NULL +#define platform_pm_resume_noirq NULL + +#endif /* !CONFIG_SUSPEND */ + +#ifdef CONFIG_HIBERNATION + +static int platform_pm_freeze(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (!drv) + return 0; + + if (drv->pm) { + if (drv->pm->freeze) + ret = drv->pm->freeze(dev); + } else { + ret = platform_legacy_suspend(dev, PMSG_FREEZE); + } + + return ret; +} + +static int platform_pm_freeze_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->freeze_noirq) + ret = pdrv->pm->freeze_noirq(dev); + } else { + ret = platform_legacy_suspend_late(dev, PMSG_FREEZE); + } + + return ret; +} + +static int platform_pm_thaw(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm) { + if (drv->pm->thaw) + ret = drv->pm->thaw(dev); + } else { + ret = platform_legacy_resume(dev); + } + + return ret; +} + +static int platform_pm_thaw_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->thaw_noirq) + ret = pdrv->pm->thaw_noirq(dev); + } else { + ret = platform_legacy_resume_early(dev); + } + + return ret; +} + +static int platform_pm_poweroff(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm) { + if (drv->pm->poweroff) + ret = drv->pm->poweroff(dev); + } else { + ret = platform_legacy_suspend(dev, PMSG_HIBERNATE); + } + + return ret; +} + +static int platform_pm_poweroff_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->poweroff_noirq) + ret = pdrv->pm->poweroff_noirq(dev); + } else { + ret = platform_legacy_suspend_late(dev, PMSG_HIBERNATE); + } + + return ret; +} + +static int platform_pm_restore(struct device *dev) +{ + struct device_driver *drv = dev->driver; + int ret = 0; + + if (drv && drv->pm) { + if (drv->pm->restore) + ret = drv->pm->restore(dev); + } else { + ret = platform_legacy_resume(dev); + } + + return ret; +} + +static int platform_pm_restore_noirq(struct device *dev) +{ + struct platform_driver *pdrv; + int ret = 0; + + if (!dev->driver) + return 0; + + pdrv = to_platform_driver(dev->driver); + if (pdrv->pm) { + if (pdrv->pm->restore_noirq) + ret = pdrv->pm->restore_noirq(dev); + } else { + ret = platform_legacy_resume_early(dev); + } + + return ret; +} + +#else /* !CONFIG_HIBERNATION */ + +#define platform_pm_freeze NULL +#define platform_pm_thaw NULL +#define platform_pm_poweroff NULL +#define platform_pm_restore NULL +#define platform_pm_freeze_noirq NULL +#define platform_pm_thaw_noirq NULL +#define platform_pm_poweroff_noirq NULL +#define platform_pm_restore_noirq NULL + +#endif /* !CONFIG_HIBERNATION */ + +struct pm_ext_ops platform_pm_ops = { + .base = { + .prepare = platform_pm_prepare, + .complete = platform_pm_complete, + .suspend = platform_pm_suspend, + .resume = platform_pm_resume, + .freeze = platform_pm_freeze, + .thaw = platform_pm_thaw, + .poweroff = platform_pm_poweroff, + .restore = platform_pm_restore, + }, + .suspend_noirq = platform_pm_suspend_noirq, + .resume_noirq = platform_pm_resume_noirq, + .freeze_noirq = platform_pm_freeze_noirq, + .thaw_noirq = platform_pm_thaw_noirq, + .poweroff_noirq = platform_pm_poweroff_noirq, + .restore_noirq = platform_pm_restore_noirq, +}; + +#define PLATFORM_PM_OPS_PTR &platform_pm_ops + +#else /* !CONFIG_PM_SLEEP */ + +#define PLATFORM_PM_OPS_PTR NULL + +#endif /* !CONFIG_PM_SLEEP */ + struct bus_type platform_bus_type = { .name = "platform", .dev_attrs = platform_dev_attrs, .match = platform_match, .uevent = platform_uevent, - .suspend = platform_suspend, - .suspend_late = platform_suspend_late, - .resume_early = platform_resume_early, - .resume = platform_resume, + .pm = PLATFORM_PM_OPS_PTR, }; EXPORT_SYMBOL_GPL(platform_bus_type); diff --git a/include/linux/platform_device.h b/include/linux/platform_device.h index 3261681c82a..95ac21ab3a0 100644 --- a/include/linux/platform_device.h +++ b/include/linux/platform_device.h @@ -53,6 +53,7 @@ struct platform_driver { int (*suspend_late)(struct platform_device *, pm_message_t state); int (*resume_early)(struct platform_device *); int (*resume)(struct platform_device *); + struct pm_ext_ops *pm; struct device_driver driver; }; -- cgit v1.2.3-70-g09d2 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 --- Documentation/kernel-parameters.txt | 6 +- arch/x86/kernel/acpi/sleep.c | 2 + drivers/acpi/sleep/main.c | 276 +++++++++++++++++++++--------------- drivers/base/power/main.c | 2 - include/linux/acpi.h | 3 + include/linux/suspend.h | 14 +- kernel/power/disk.c | 28 +++- kernel/power/main.c | 10 +- 8 files changed, 215 insertions(+), 126 deletions(-) (limited to 'drivers/base') diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index 9cf7b34f2db..18d793ea0dd 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -147,10 +147,14 @@ and is between 256 and 4096 characters. It is defined in the file default: 0 acpi_sleep= [HW,ACPI] Sleep options - Format: { s3_bios, s3_mode, s3_beep } + Format: { s3_bios, s3_mode, s3_beep, old_ordering } See Documentation/power/video.txt for s3_bios and s3_mode. s3_beep is for debugging; it makes the PC's speaker beep as soon as the kernel's real-mode entry point is called. + old_ordering causes the ACPI 1.0 ordering of the _PTS + control method, wrt putting devices into low power + states, to be enforced (the ACPI 2.0 ordering of _PTS is + used by default). acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode Format: { level | edge | high | low } diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index afc25ee9964..882e970032d 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -124,6 +124,8 @@ static int __init acpi_sleep_setup(char *str) acpi_realmode_flags |= 2; if (strncmp(str, "s3_beep", 7) == 0) acpi_realmode_flags |= 4; + if (strncmp(str, "old_ordering", 12) == 0) + acpi_old_suspend_ordering(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 0f2caea2fc8..4addf8ad50a 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -24,10 +24,6 @@ u8 sleep_states[ACPI_S_STATE_COUNT]; -#ifdef CONFIG_PM_SLEEP -static u32 acpi_target_sleep_state = ACPI_STATE_S0; -#endif - static int acpi_sleep_prepare(u32 acpi_state) { #ifdef CONFIG_ACPI_SLEEP @@ -50,9 +46,96 @@ static int acpi_sleep_prepare(u32 acpi_state) return 0; } -#ifdef CONFIG_SUSPEND -static struct platform_suspend_ops acpi_suspend_ops; +#ifdef CONFIG_PM_SLEEP +static u32 acpi_target_sleep_state = ACPI_STATE_S0; + +/* + * ACPI 1.0 wants us to execute _PTS before suspending devices, so we allow the + * user to request that behavior by using the 'acpi_old_suspend_ordering' + * kernel command line option that causes the following variable to be set. + */ +static bool old_suspend_ordering; + +void __init acpi_old_suspend_ordering(void) +{ + old_suspend_ordering = true; +} + +/** + * acpi_pm_disable_gpes - Disable the GPEs. + */ +static int acpi_pm_disable_gpes(void) +{ + acpi_hw_disable_all_gpes(); + return 0; +} + +/** + * __acpi_pm_prepare - Prepare the platform to enter the target state. + * + * If necessary, set the firmware waking vector and do arch-specific + * nastiness to get the wakeup code to the waking vector. + */ +static int __acpi_pm_prepare(void) +{ + int error = acpi_sleep_prepare(acpi_target_sleep_state); + + if (error) + acpi_target_sleep_state = ACPI_STATE_S0; + return error; +} + +/** + * acpi_pm_prepare - Prepare the platform to enter the target sleep + * state and disable the GPEs. + */ +static int acpi_pm_prepare(void) +{ + int error = __acpi_pm_prepare(); + + if (!error) + acpi_hw_disable_all_gpes(); + return error; +} + +/** + * acpi_pm_finish - Instruct the platform to leave a sleep state. + * + * This is called after we wake back up (or if entering the sleep state + * failed). + */ +static void acpi_pm_finish(void) +{ + u32 acpi_state = acpi_target_sleep_state; + + if (acpi_state == ACPI_STATE_S0) + return; + printk(KERN_INFO PREFIX "Waking up from system sleep state S%d\n", + acpi_state); + acpi_disable_wakeup_device(acpi_state); + acpi_leave_sleep_state(acpi_state); + + /* reset firmware waking vector */ + acpi_set_firmware_waking_vector((acpi_physical_address) 0); + + acpi_target_sleep_state = ACPI_STATE_S0; +} + +/** + * acpi_pm_end - Finish up suspend sequence. + */ +static void acpi_pm_end(void) +{ + /* + * This is necessary in case acpi_pm_finish() is not called during a + * failing transition to a sleep state. + */ + acpi_target_sleep_state = ACPI_STATE_S0; +} +#endif /* CONFIG_PM_SLEEP */ + +#ifdef CONFIG_SUSPEND extern void do_suspend_lowlevel(void); static u32 acpi_suspend_states[] = { @@ -66,7 +149,6 @@ static u32 acpi_suspend_states[] = { * acpi_suspend_begin - Set the target system sleep state to the state * associated with given @pm_state, if supported. */ - static int acpi_suspend_begin(suspend_state_t pm_state) { u32 acpi_state = acpi_suspend_states[pm_state]; @@ -82,25 +164,6 @@ static int acpi_suspend_begin(suspend_state_t pm_state) return error; } -/** - * acpi_suspend_prepare - Do preliminary suspend work. - * - * If necessary, set the firmware waking vector and do arch-specific - * nastiness to get the wakeup code to the waking vector. - */ - -static int acpi_suspend_prepare(void) -{ - int error = acpi_sleep_prepare(acpi_target_sleep_state); - - if (error) { - acpi_target_sleep_state = ACPI_STATE_S0; - return error; - } - - return ACPI_SUCCESS(acpi_hw_disable_all_gpes()) ? 0 : -EFAULT; -} - /** * acpi_suspend_enter - Actually enter a sleep state. * @pm_state: ignored @@ -109,7 +172,6 @@ static int acpi_suspend_prepare(void) * assembly, which in turn call acpi_enter_sleep_state(). * It's unfortunate, but it works. Please fix if you're feeling frisky. */ - static int acpi_suspend_enter(suspend_state_t pm_state) { acpi_status status = AE_OK; @@ -166,39 +228,6 @@ static int acpi_suspend_enter(suspend_state_t pm_state) return ACPI_SUCCESS(status) ? 0 : -EFAULT; } -/** - * acpi_suspend_finish - Instruct the platform to leave a sleep state. - * - * This is called after we wake back up (or if entering the sleep state - * failed). - */ - -static void acpi_suspend_finish(void) -{ - u32 acpi_state = acpi_target_sleep_state; - - acpi_disable_wakeup_device(acpi_state); - acpi_leave_sleep_state(acpi_state); - - /* reset firmware waking vector */ - acpi_set_firmware_waking_vector((acpi_physical_address) 0); - - acpi_target_sleep_state = ACPI_STATE_S0; -} - -/** - * acpi_suspend_end - Finish up suspend sequence. - */ - -static void acpi_suspend_end(void) -{ - /* - * This is necessary in case acpi_suspend_finish() is not called during a - * failing transition to a sleep state. - */ - acpi_target_sleep_state = ACPI_STATE_S0; -} - static int acpi_suspend_state_valid(suspend_state_t pm_state) { u32 acpi_state; @@ -218,10 +247,39 @@ static int acpi_suspend_state_valid(suspend_state_t pm_state) static struct platform_suspend_ops acpi_suspend_ops = { .valid = acpi_suspend_state_valid, .begin = acpi_suspend_begin, - .prepare = acpi_suspend_prepare, + .prepare = acpi_pm_prepare, + .enter = acpi_suspend_enter, + .finish = acpi_pm_finish, + .end = acpi_pm_end, +}; + +/** + * acpi_suspend_begin_old - Set the target system sleep state to the + * state associated with given @pm_state, if supported, and + * execute the _PTS control method. This function is used if the + * pre-ACPI 2.0 suspend ordering has been requested. + */ +static int acpi_suspend_begin_old(suspend_state_t pm_state) +{ + int error = acpi_suspend_begin(pm_state); + + if (!error) + error = __acpi_pm_prepare(); + return error; +} + +/* + * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has + * been requested. + */ +static struct platform_suspend_ops acpi_suspend_ops_old = { + .valid = acpi_suspend_state_valid, + .begin = acpi_suspend_begin_old, + .prepare = acpi_pm_disable_gpes, .enter = acpi_suspend_enter, - .finish = acpi_suspend_finish, - .end = acpi_suspend_end, + .finish = acpi_pm_finish, + .end = acpi_pm_end, + .recover = acpi_pm_finish, }; #endif /* CONFIG_SUSPEND */ @@ -229,22 +287,9 @@ static struct platform_suspend_ops acpi_suspend_ops = { static int acpi_hibernation_begin(void) { acpi_target_sleep_state = ACPI_STATE_S4; - return 0; } -static int acpi_hibernation_prepare(void) -{ - int error = acpi_sleep_prepare(ACPI_STATE_S4); - - if (error) { - acpi_target_sleep_state = ACPI_STATE_S0; - return error; - } - - return ACPI_SUCCESS(acpi_hw_disable_all_gpes()) ? 0 : -EFAULT; -} - static int acpi_hibernation_enter(void) { acpi_status status = AE_OK; @@ -274,52 +319,55 @@ static void acpi_hibernation_leave(void) acpi_leave_sleep_state_prep(ACPI_STATE_S4); } -static void acpi_hibernation_finish(void) +static void acpi_pm_enable_gpes(void) { - acpi_disable_wakeup_device(ACPI_STATE_S4); - acpi_leave_sleep_state(ACPI_STATE_S4); - - /* reset firmware waking vector */ - acpi_set_firmware_waking_vector((acpi_physical_address) 0); - - acpi_target_sleep_state = ACPI_STATE_S0; + acpi_hw_enable_all_runtime_gpes(); } -static void acpi_hibernation_end(void) -{ - /* - * This is necessary in case acpi_hibernation_finish() is not called - * during a failing transition to the sleep state. - */ - acpi_target_sleep_state = ACPI_STATE_S0; -} +static struct platform_hibernation_ops acpi_hibernation_ops = { + .begin = acpi_hibernation_begin, + .end = acpi_pm_end, + .pre_snapshot = acpi_pm_prepare, + .finish = acpi_pm_finish, + .prepare = acpi_pm_prepare, + .enter = acpi_hibernation_enter, + .leave = acpi_hibernation_leave, + .pre_restore = acpi_pm_disable_gpes, + .restore_cleanup = acpi_pm_enable_gpes, +}; -static int acpi_hibernation_pre_restore(void) +/** + * acpi_hibernation_begin_old - Set the target system sleep state to + * ACPI_STATE_S4 and execute the _PTS control method. This + * function is used if the pre-ACPI 2.0 suspend ordering has been + * requested. + */ +static int acpi_hibernation_begin_old(void) { - acpi_status status; - - status = acpi_hw_disable_all_gpes(); - - return ACPI_SUCCESS(status) ? 0 : -EFAULT; -} + int error = acpi_sleep_prepare(ACPI_STATE_S4); -static void acpi_hibernation_restore_cleanup(void) -{ - acpi_hw_enable_all_runtime_gpes(); + if (!error) + acpi_target_sleep_state = ACPI_STATE_S4; + return error; } -static struct platform_hibernation_ops acpi_hibernation_ops = { - .begin = acpi_hibernation_begin, - .end = acpi_hibernation_end, - .pre_snapshot = acpi_hibernation_prepare, - .finish = acpi_hibernation_finish, - .prepare = acpi_hibernation_prepare, +/* + * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has + * been requested. + */ +static struct platform_hibernation_ops acpi_hibernation_ops_old = { + .begin = acpi_hibernation_begin_old, + .end = acpi_pm_end, + .pre_snapshot = acpi_pm_disable_gpes, + .finish = acpi_pm_finish, + .prepare = acpi_pm_disable_gpes, .enter = acpi_hibernation_enter, .leave = acpi_hibernation_leave, - .pre_restore = acpi_hibernation_pre_restore, - .restore_cleanup = acpi_hibernation_restore_cleanup, + .pre_restore = acpi_pm_disable_gpes, + .restore_cleanup = acpi_pm_enable_gpes, + .recover = acpi_pm_finish, }; -#endif /* CONFIG_HIBERNATION */ +#endif /* CONFIG_HIBERNATION */ int acpi_suspend(u32 acpi_state) { @@ -461,13 +509,15 @@ int __init acpi_sleep_init(void) } } - suspend_set_ops(&acpi_suspend_ops); + suspend_set_ops(old_suspend_ordering ? + &acpi_suspend_ops_old : &acpi_suspend_ops); #endif #ifdef CONFIG_HIBERNATION status = acpi_get_sleep_type_data(ACPI_STATE_S4, &type_a, &type_b); if (ACPI_SUCCESS(status)) { - hibernation_set_ops(&acpi_hibernation_ops); + hibernation_set_ops(old_suspend_ordering ? + &acpi_hibernation_ops_old : &acpi_hibernation_ops); sleep_states[ACPI_STATE_S4] = 1; printk(" S4"); } diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index d571204aaff..3250c5257b7 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -781,8 +781,6 @@ int device_suspend(pm_message_t state) error = dpm_prepare(state); if (!error) error = dpm_suspend(state); - if (error) - device_resume(resume_event(state)); return error; } EXPORT_SYMBOL_GPL(device_suspend); diff --git a/include/linux/acpi.h b/include/linux/acpi.h index 41f7ce7edd7..33adcf91ef4 100644 --- a/include/linux/acpi.h +++ b/include/linux/acpi.h @@ -234,6 +234,9 @@ int acpi_check_region(resource_size_t start, resource_size_t n, int acpi_check_mem_region(resource_size_t start, resource_size_t n, const char *name); +#ifdef CONFIG_PM_SLEEP +void __init acpi_old_suspend_ordering(void); +#endif /* CONFIG_PM_SLEEP */ #else /* CONFIG_ACPI */ static inline int early_acpi_boot_init(void) diff --git a/include/linux/suspend.h b/include/linux/suspend.h index a6977423baf..e8e69159af7 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -86,6 +86,11 @@ typedef int __bitwise suspend_state_t; * that implement @begin(), but platforms implementing @begin() should * also provide a @end() which cleans up transitions aborted before * @enter(). + * + * @recover: Recover the platform from a suspend failure. + * Called by the PM core if the suspending of devices fails. + * This callback is optional and should only be implemented by platforms + * which require special recovery actions in that situation. */ struct platform_suspend_ops { int (*valid)(suspend_state_t state); @@ -94,6 +99,7 @@ struct platform_suspend_ops { int (*enter)(suspend_state_t state); void (*finish)(void); void (*end)(void); + void (*recover)(void); }; #ifdef CONFIG_SUSPEND @@ -149,7 +155,7 @@ extern void mark_free_pages(struct zone *zone); * The methods in this structure allow a platform to carry out special * operations required by it during a hibernation transition. * - * All the methods below must be implemented. + * All the methods below, except for @recover(), must be implemented. * * @begin: Tell the platform driver that we're starting hibernation. * Called right after shrinking memory and before freezing devices. @@ -189,6 +195,11 @@ extern void mark_free_pages(struct zone *zone); * @restore_cleanup: Clean up after a failing image restoration. * Called right after the nonboot CPUs have been enabled and before * thawing devices (runs with IRQs on). + * + * @recover: Recover the platform from a failure to suspend devices. + * Called by the PM core if the suspending of devices during hibernation + * fails. This callback is optional and should only be implemented by + * platforms which require special recovery actions in that situation. */ struct platform_hibernation_ops { int (*begin)(void); @@ -200,6 +211,7 @@ struct platform_hibernation_ops { void (*leave)(void); int (*pre_restore)(void); void (*restore_cleanup)(void); + void (*recover)(void); }; #ifdef CONFIG_HIBERNATION diff --git a/kernel/power/disk.c b/kernel/power/disk.c index d416be0efa8..f011e0870b5 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -179,6 +179,17 @@ static void platform_restore_cleanup(int platform_mode) hibernation_ops->restore_cleanup(); } +/** + * 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 @@ -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 d023b6b584e..3398f4651aa 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 v1.2.3-70-g09d2 From c50cbb05a05cf1f9ca3592272eff053c847727d8 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Wed, 4 Jun 2008 21:47:29 -0700 Subject: cpu topology: always define CPU topology information This can result in an empty topology directory in sysfs, and requires in-kernel users to protect all uses with #ifdef - see . The documentation of CPU topology specifies what the defaults should be if only partial information is available from the hardware. So we can provide these defaults as a fallback. This patch: - Adds default definitions of the 4 topology macros to - Changes drivers/base/topology.c to use the topology macros unconditionally and to cope with definitions that aren't lvalues - Updates documentation accordingly [ From: Andrew Morton - fold now-duplicated code - fix layout ] Signed-off-by: Ben Hutchings Cc: Vegard Nossum Cc: Nick Piggin Cc: Chandra Seetharaman Cc: Suresh Siddha Cc: Mike Travis Cc: Christoph Lameter Cc: John Hawkes Cc: Zhang, Yanmin Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- Documentation/cputopology.txt | 26 +++++++++----------------- drivers/base/topology.c | 38 ++++++++++---------------------------- include/linux/topology.h | 13 +++++++++++++ 3 files changed, 32 insertions(+), 45 deletions(-) (limited to 'drivers/base') diff --git a/Documentation/cputopology.txt b/Documentation/cputopology.txt index b61cb956402..bd699da2466 100644 --- a/Documentation/cputopology.txt +++ b/Documentation/cputopology.txt @@ -14,9 +14,8 @@ represent the thread siblings to cpu X in the same physical package; To implement it in an architecture-neutral way, a new source file, drivers/base/topology.c, is to export the 4 attributes. -If one architecture wants to support this feature, it just needs to -implement 4 defines, typically in file include/asm-XXX/topology.h. -The 4 defines are: +For an architecture to support this feature, it must define some of +these macros in include/asm-XXX/topology.h: #define topology_physical_package_id(cpu) #define topology_core_id(cpu) #define topology_thread_siblings(cpu) @@ -25,17 +24,10 @@ The 4 defines are: The type of **_id is int. The type of siblings is cpumask_t. -To be consistent on all architectures, the 4 attributes should have -default values if their values are unavailable. Below is the rule. -1) physical_package_id: If cpu has no physical package id, -1 is the -default value. -2) core_id: If cpu doesn't support multi-core, its core id is 0. -3) thread_siblings: Just include itself, if the cpu doesn't support -HT/multi-thread. -4) core_siblings: Just include itself, if the cpu doesn't support -multi-core and HT/Multi-thread. - -So be careful when declaring the 4 defines in include/asm-XXX/topology.h. - -If an attribute isn't defined on an architecture, it won't be exported. - +To be consistent on all architectures, include/linux/topology.h +provides default definitions for any of the above macros that are +not defined by include/asm-XXX/topology.h: +1) physical_package_id: -1 +2) core_id: 0 +3) thread_siblings: just the given CPU +4) core_siblings: just the given CPU diff --git a/drivers/base/topology.c b/drivers/base/topology.c index fdf4044d2e7..24d29a9fc25 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -59,60 +59,42 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) static inline ssize_t show_##name(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ - return show_cpumap(0, &(topology_##name(cpu)), buf); \ + cpumask_t siblings = topology_##name(cpu); \ + return show_cpumap(0, &siblings, buf); \ } #define define_siblings_show_list(name) \ static inline ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ - return show_cpumap(1, &(topology_##name(cpu)), buf); \ + cpumask_t siblings = topology_##name(cpu); \ + return show_cpumap(1, &siblings, buf); \ } #define define_siblings_show_func(name) \ define_siblings_show_map(name); define_siblings_show_list(name) -#ifdef topology_physical_package_id define_id_show_func(physical_package_id); define_one_ro(physical_package_id); -#define ref_physical_package_id_attr &attr_physical_package_id.attr, -#else -#define ref_physical_package_id_attr -#endif -#ifdef topology_core_id define_id_show_func(core_id); define_one_ro(core_id); -#define ref_core_id_attr &attr_core_id.attr, -#else -#define ref_core_id_attr -#endif -#ifdef topology_thread_siblings define_siblings_show_func(thread_siblings); define_one_ro(thread_siblings); define_one_ro(thread_siblings_list); -#define ref_thread_siblings_attr \ - &attr_thread_siblings.attr, &attr_thread_siblings_list.attr, -#else -#define ref_thread_siblings_attr -#endif -#ifdef topology_core_siblings define_siblings_show_func(core_siblings); define_one_ro(core_siblings); define_one_ro(core_siblings_list); -#define ref_core_siblings_attr \ - &attr_core_siblings.attr, &attr_core_siblings_list.attr, -#else -#define ref_core_siblings_attr -#endif static struct attribute *default_attrs[] = { - ref_physical_package_id_attr - ref_core_id_attr - ref_thread_siblings_attr - ref_core_siblings_attr + &attr_physical_package_id.attr, + &attr_core_id.attr, + &attr_thread_siblings.attr, + &attr_thread_siblings_list.attr, + &attr_core_siblings.attr, + &attr_core_siblings_list.attr, NULL }; diff --git a/include/linux/topology.h b/include/linux/topology.h index 24f3d2282e1..2158fc0d5a5 100644 --- a/include/linux/topology.h +++ b/include/linux/topology.h @@ -179,4 +179,17 @@ void arch_update_cpu_topology(void); #endif #endif /* CONFIG_NUMA */ +#ifndef topology_physical_package_id +#define topology_physical_package_id(cpu) ((void)(cpu), -1) +#endif +#ifndef topology_core_id +#define topology_core_id(cpu) ((void)(cpu), 0) +#endif +#ifndef topology_thread_siblings +#define topology_thread_siblings(cpu) cpumask_of_cpu(cpu) +#endif +#ifndef topology_core_siblings +#define topology_core_siblings(cpu) cpumask_of_cpu(cpu) +#endif + #endif /* _LINUX_TOPOLOGY_H */ -- cgit v1.2.3-70-g09d2 From 131b943ae643b1ad6febd67cdbb31d955706ecf4 Mon Sep 17 00:00:00 2001 From: Ben Hutchings Date: Thu, 5 Jun 2008 17:37:15 +0100 Subject: cputopology: always define CPU topology information, clean up simplify drivers/base/topology.c a bit. Signed-off-by: Ben Hutchings Signed-off-by: Ingo Molnar --- drivers/base/topology.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 24d29a9fc25..f0cb2701193 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -59,16 +59,14 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) static inline ssize_t show_##name(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ - cpumask_t siblings = topology_##name(cpu); \ - return show_cpumap(0, &siblings, buf); \ + return show_cpumap(0, &(topology_##name(cpu)), buf); \ } #define define_siblings_show_list(name) \ static inline ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ - cpumask_t siblings = topology_##name(cpu); \ - return show_cpumap(1, &siblings, buf); \ + return show_cpumap(1, &(topology_##name(cpu)), buf); \ } #define define_siblings_show_func(name) \ -- cgit v1.2.3-70-g09d2 From eb9d0fe40e313c0a74115ef456a2e43a6c8da72f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 7 Jul 2008 03:34:48 +0200 Subject: PCI ACPI: Rework PCI handling of wake-up * Introduce function acpi_pm_device_sleep_wake() for enabling and disabling the system wake-up capability of devices that are power manageable by ACPI. * Introduce function acpi_bus_can_wakeup() allowing other (dependent) subsystems to check if ACPI is able to enable the system wake-up capability of given device. * Introduce callback .sleep_wake() in struct pci_platform_pm_ops and for the ACPI PCI 'driver' make it use acpi_pm_device_sleep_wake(). * Introduce callback .can_wakeup() in struct pci_platform_pm_ops and for the ACPI 'driver' make it use acpi_bus_can_wakeup(). * Move the PME# handlig code out of pci_enable_wake() and split it into two functions, pci_pme_capable() and pci_pme_active(), allowing the caller to check if given device is capable of generating PME# from given power state and to enable/disable the device's PME# functionality, respectively. * Modify pci_enable_wake() to use the new ACPI callbacks and the new PME#-related functions. * Drop the generic .platform_enable_wakeup() callback that is not used any more. * Introduce device_set_wakeup_capable() that will set the power.can_wakeup flag of given device. * Rework PCI device PM initialization so that, if given device is capable of generating wake-up events, either natively through the PME# mechanism, or with the help of the platform, its power.can_wakeup flag is set and its power.should_wakeup flag is unset as appropriate. * Make ACPI set the power.can_wakeup flag for devices found to be wake-up capable by it. * Make the ACPI wake-up code enable/disable GPEs for devices that have the wakeup.flags.prepared flag set (which means that their wake-up power has been enabled). Signed-off-by: Rafael J. Wysocki Signed-off-by: Jesse Barnes --- drivers/acpi/bus.c | 11 +++ drivers/acpi/glue.c | 2 + drivers/acpi/sleep/main.c | 25 +++++++ drivers/acpi/sleep/wakeup.c | 11 +-- drivers/base/power/sysfs.c | 3 - drivers/pci/pci-acpi.c | 20 ++++++ drivers/pci/pci.c | 169 ++++++++++++++++++++++++++++++++------------ drivers/pci/pci.h | 8 +++ drivers/pci/probe.c | 47 +----------- include/acpi/acpi_bus.h | 2 + include/linux/pm_wakeup.h | 26 ++----- 11 files changed, 208 insertions(+), 116 deletions(-) (limited to 'drivers/base') diff --git a/drivers/acpi/bus.c b/drivers/acpi/bus.c index b9b69d9629b..fc1110d6a07 100644 --- a/drivers/acpi/bus.c +++ b/drivers/acpi/bus.c @@ -306,6 +306,17 @@ bool acpi_bus_power_manageable(acpi_handle handle) EXPORT_SYMBOL(acpi_bus_power_manageable); +bool acpi_bus_can_wakeup(acpi_handle handle) +{ + struct acpi_device *device; + int result; + + result = acpi_bus_get_device(handle, &device); + return result ? false : device->wakeup.flags.valid; +} + +EXPORT_SYMBOL(acpi_bus_can_wakeup); + /* -------------------------------------------------------------------------- Event Management -------------------------------------------------------------------------- */ diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 06f8634fe58..87c5d456e18 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -166,6 +166,8 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) "firmware_node"); ret = sysfs_create_link(&acpi_dev->dev.kobj, &dev->kobj, "physical_node"); + if (acpi_dev->wakeup.flags.valid) + device_set_wakeup_capable(dev, true); } return 0; diff --git a/drivers/acpi/sleep/main.c b/drivers/acpi/sleep/main.c index 4addf8ad50a..af7f4663dea 100644 --- a/drivers/acpi/sleep/main.c +++ b/drivers/acpi/sleep/main.c @@ -468,6 +468,31 @@ int acpi_pm_device_sleep_state(struct device *dev, int *d_min_p) *d_min_p = d_min; return d_max; } + +/** + * acpi_pm_device_sleep_wake - enable or disable the system wake-up + * capability of given device + * @dev: device to handle + * @enable: 'true' - enable, 'false' - disable the wake-up capability + */ +int acpi_pm_device_sleep_wake(struct device *dev, bool enable) +{ + acpi_handle handle; + struct acpi_device *adev; + + if (!device_may_wakeup(dev)) + return -EINVAL; + + handle = DEVICE_ACPI_HANDLE(dev); + if (!handle || ACPI_FAILURE(acpi_bus_get_device(handle, &adev))) { + printk(KERN_DEBUG "ACPI handle has no context!\n"); + return -ENODEV; + } + + return enable ? + acpi_enable_wakeup_device_power(adev, acpi_target_sleep_state) : + acpi_disable_wakeup_device_power(adev); +} #endif static void acpi_power_off_prepare(void) diff --git a/drivers/acpi/sleep/wakeup.c b/drivers/acpi/sleep/wakeup.c index 7422a221394..38655eb132d 100644 --- a/drivers/acpi/sleep/wakeup.c +++ b/drivers/acpi/sleep/wakeup.c @@ -66,13 +66,15 @@ void acpi_enable_wakeup_device(u8 sleep_state) list_for_each_safe(node, next, &acpi_wakeup_device_list) { struct acpi_device *dev = container_of(node, struct acpi_device, wakeup_list); + if (!dev->wakeup.flags.valid) continue; + /* If users want to disable run-wake GPE, * we only disable it for wake and leave it for runtime */ - if (!dev->wakeup.state.enabled || - sleep_state > (u32) dev->wakeup.sleep_state) { + if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared) + || sleep_state > (u32) dev->wakeup.sleep_state) { if (dev->wakeup.flags.run_wake) { spin_unlock(&acpi_device_lock); /* set_gpe_type will disable GPE, leave it like that */ @@ -110,8 +112,9 @@ void acpi_disable_wakeup_device(u8 sleep_state) if (!dev->wakeup.flags.valid) continue; - if (!dev->wakeup.state.enabled || - sleep_state > (u32) dev->wakeup.sleep_state) { + + if ((!dev->wakeup.state.enabled && !dev->wakeup.flags.prepared) + || sleep_state > (u32) dev->wakeup.sleep_state) { if (dev->wakeup.flags.run_wake) { spin_unlock(&acpi_device_lock); acpi_set_gpe_type(dev->wakeup.gpe_device, diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index d11f74b038d..596aeecfdff 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -6,9 +6,6 @@ #include #include "power.h" -int (*platform_enable_wakeup)(struct device *dev, int is_on); - - /* * wakeup - Report/change current wakeup option for device * diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c index 6bc0d8c870a..7764768b6a0 100644 --- a/drivers/pci/pci-acpi.c +++ b/drivers/pci/pci-acpi.c @@ -299,10 +299,30 @@ static int acpi_pci_set_power_state(struct pci_dev *dev, pci_power_t state) return error; } +static bool acpi_pci_can_wakeup(struct pci_dev *dev) +{ + acpi_handle handle = DEVICE_ACPI_HANDLE(&dev->dev); + + return handle ? acpi_bus_can_wakeup(handle) : false; +} + +static int acpi_pci_sleep_wake(struct pci_dev *dev, bool enable) +{ + int error = acpi_pm_device_sleep_wake(&dev->dev, enable); + + if (!error) + dev_printk(KERN_INFO, &dev->dev, + "wake-up capability %s by ACPI\n", + enable ? "enabled" : "disabled"); + return error; +} + static struct pci_platform_pm_ops acpi_pci_platform_pm = { .is_manageable = acpi_pci_power_manageable, .set_state = acpi_pci_set_power_state, .choose_state = acpi_pci_choose_state, + .can_wakeup = acpi_pci_can_wakeup, + .sleep_wake = acpi_pci_sleep_wake, }; /* ACPI bus type */ diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c index 20e28077b96..a6b1b6f96ab 100644 --- a/drivers/pci/pci.c +++ b/drivers/pci/pci.c @@ -380,7 +380,8 @@ static struct pci_platform_pm_ops *pci_platform_pm; int pci_set_platform_pm(struct pci_platform_pm_ops *ops) { - if (!ops->is_manageable || !ops->set_state || !ops->choose_state) + if (!ops->is_manageable || !ops->set_state || !ops->choose_state + || !ops->sleep_wake || !ops->can_wakeup) return -EINVAL; pci_platform_pm = ops; return 0; @@ -403,6 +404,17 @@ static inline pci_power_t platform_pci_choose_state(struct pci_dev *dev) pci_platform_pm->choose_state(dev) : PCI_POWER_ERROR; } +static inline bool platform_pci_can_wakeup(struct pci_dev *dev) +{ + return pci_platform_pm ? pci_platform_pm->can_wakeup(dev) : false; +} + +static inline int platform_pci_sleep_wake(struct pci_dev *dev, bool enable) +{ + return pci_platform_pm ? + pci_platform_pm->sleep_wake(dev, enable) : -ENODEV; +} + /** * pci_raw_set_power_state - Use PCI PM registers to set the power state of * given PCI device @@ -1035,6 +1047,56 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state) return pcibios_set_pcie_reset_state(dev, state); } +/** + * pci_pme_capable - check the capability of PCI device to generate PME# + * @dev: PCI device to handle. + * @pm: PCI PM capability offset of the device. + * @state: PCI state from which device will issue PME#. + */ +static bool pci_pme_capable(struct pci_dev *dev, int pm, pci_power_t state) +{ + u16 pmc; + + if (!pm) + return false; + + /* Check device's ability to generate PME# from given state */ + pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc); + + pmc &= PCI_PM_CAP_PME_MASK; + pmc >>= ffs(PCI_PM_CAP_PME_MASK) - 1; /* First bit of mask */ + + return !!(pmc & (1 << state)); +} + +/** + * pci_pme_active - enable or disable PCI device's PME# function + * @dev: PCI device to handle. + * @pm: PCI PM capability offset of the device. + * @enable: 'true' to enable PME# generation; 'false' to disable it. + * + * The caller must verify that the device is capable of generating PME# before + * calling this function with @enable equal to 'true'. + */ +static void pci_pme_active(struct pci_dev *dev, int pm, bool enable) +{ + u16 pmcsr; + + if (!pm) + return; + + pci_read_config_word(dev, pm + PCI_PM_CTRL, &pmcsr); + /* Clear PME_Status by writing 1 to it and enable PME# */ + pmcsr |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE; + if (!enable) + pmcsr &= ~PCI_PM_CTRL_PME_ENABLE; + + pci_write_config_word(dev, pm + PCI_PM_CTRL, pmcsr); + + dev_printk(KERN_INFO, &dev->dev, "PME# %s\n", + enable ? "enabled" : "disabled"); +} + /** * pci_enable_wake - enable PCI device as wakeup event source * @dev: PCI device affected @@ -1046,66 +1108,83 @@ int pci_set_pcie_reset_state(struct pci_dev *dev, enum pcie_reset_state state) * called automatically by this routine. * * Devices with legacy power management (no standard PCI PM capabilities) - * always require such platform hooks. Depending on the platform, devices - * supporting the standard PCI PME# signal may require such platform hooks; - * they always update bits in config space to allow PME# generation. + * always require such platform hooks. * - * -EIO is returned if the device can't ever be a wakeup event source. - * -EINVAL is returned if the device can't generate wakeup events from - * the specified PCI state. Returns zero if the operation is successful. + * RETURN VALUE: + * 0 is returned on success + * -EINVAL is returned if device is not supposed to wake up the system + * Error code depending on the platform is returned if both the platform and + * the native mechanism fail to enable the generation of wake-up events */ int pci_enable_wake(struct pci_dev *dev, pci_power_t state, int enable) { int pm; - int status; - u16 value; - - /* Note that drivers should verify device_may_wakeup(&dev->dev) - * before calling this function. Platform code should report - * errors when drivers try to enable wakeup on devices that - * can't issue wakeups, or on which wakeups were disabled by - * userspace updating the /sys/devices.../power/wakeup file. - */ + int error = 0; + bool pme_done = false; - status = call_platform_enable_wakeup(&dev->dev, enable); - - /* find PCI PM capability in list */ - pm = pci_find_capability(dev, PCI_CAP_ID_PM); + if (!device_may_wakeup(&dev->dev)) + return -EINVAL; - /* If device doesn't support PM Capabilities, but caller wants to - * disable wake events, it's a NOP. Otherwise fail unless the - * platform hooks handled this legacy device already. + /* + * According to "PCI System Architecture" 4th ed. by Tom Shanley & Don + * Anderson we should be doing PME# wake enable followed by ACPI wake + * enable. To disable wake-up we call the platform first, for symmetry. */ - if (!pm) - return enable ? status : 0; - /* Check device's ability to generate PME# */ - pci_read_config_word(dev,pm+PCI_PM_PMC,&value); + if (!enable && platform_pci_can_wakeup(dev)) + error = platform_pci_sleep_wake(dev, false); - value &= PCI_PM_CAP_PME_MASK; - value >>= ffs(PCI_PM_CAP_PME_MASK) - 1; /* First bit of mask */ - - /* Check if it can generate PME# from requested state. */ - if (!value || !(value & (1 << state))) { - /* if it can't, revert what the platform hook changed, - * always reporting the base "EINVAL, can't PME#" error - */ - if (enable) - call_platform_enable_wakeup(&dev->dev, 0); - return enable ? -EINVAL : 0; + pm = pci_find_capability(dev, PCI_CAP_ID_PM); + if (!enable || pci_pme_capable(dev, pm, state)) { + pci_pme_active(dev, pm, enable); + pme_done = true; } - pci_read_config_word(dev, pm + PCI_PM_CTRL, &value); + if (enable && platform_pci_can_wakeup(dev)) + error = platform_pci_sleep_wake(dev, true); - /* Clear PME_Status by writing 1 to it and enable PME# */ - value |= PCI_PM_CTRL_PME_STATUS | PCI_PM_CTRL_PME_ENABLE; + return pme_done ? 0 : error; +} - if (!enable) - value &= ~PCI_PM_CTRL_PME_ENABLE; +/** + * pci_pm_init - Initialize PM functions of given PCI device + * @dev: PCI device to handle. + */ +void pci_pm_init(struct pci_dev *dev) +{ + int pm; + u16 pmc; - pci_write_config_word(dev, pm + PCI_PM_CTRL, value); + /* find PCI PM capability in list */ + pm = pci_find_capability(dev, PCI_CAP_ID_PM); + if (!pm) + return; + /* Check device's ability to generate PME# */ + pci_read_config_word(dev, pm + PCI_PM_PMC, &pmc); - return 0; + if ((pmc & PCI_PM_CAP_VER_MASK) > 3) { + dev_err(&dev->dev, "unsupported PM cap regs version (%u)\n", + pmc & PCI_PM_CAP_VER_MASK); + return; + } + + if (pmc & PCI_PM_CAP_PME_MASK) { + dev_printk(KERN_INFO, &dev->dev, + "PME# supported from%s%s%s%s%s\n", + (pmc & PCI_PM_CAP_PME_D0) ? " D0" : "", + (pmc & PCI_PM_CAP_PME_D1) ? " D1" : "", + (pmc & PCI_PM_CAP_PME_D2) ? " D2" : "", + (pmc & PCI_PM_CAP_PME_D3) ? " D3hot" : "", + (pmc & PCI_PM_CAP_PME_D3cold) ? " D3cold" : ""); + /* + * Make device's PM flags reflect the wake-up capability, but + * let the user space enable it to wake up the system as needed. + */ + device_set_wakeup_capable(&dev->dev, true); + device_set_wakeup_enable(&dev->dev, false); + /* Disable the PME# generation functionality */ + pci_pme_active(dev, pm, false); + } } int diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h index 0cd2e719933..b08dfc9746a 100644 --- a/drivers/pci/pci.h +++ b/drivers/pci/pci.h @@ -17,6 +17,11 @@ extern void pci_cleanup_rom(struct pci_dev *dev); * platform; to be used during system-wide transitions from a * sleeping state to the working state and vice versa * + * @can_wakeup - returns 'true' if given device is capable of waking up the + * system from a sleeping state + * + * @sleep_wake - enables/disables the system wake up capability of given device + * * If given platform is generally capable of power managing PCI devices, all of * these callbacks are mandatory. */ @@ -24,9 +29,12 @@ struct pci_platform_pm_ops { bool (*is_manageable)(struct pci_dev *dev); int (*set_state)(struct pci_dev *dev, pci_power_t state); pci_power_t (*choose_state)(struct pci_dev *dev); + bool (*can_wakeup)(struct pci_dev *dev); + int (*sleep_wake)(struct pci_dev *dev, bool enable); }; extern int pci_set_platform_pm(struct pci_platform_pm_ops *ops); +extern void pci_pm_init(struct pci_dev *dev); extern int pci_user_read_config_byte(struct pci_dev *dev, int where, u8 *val); extern int pci_user_read_config_word(struct pci_dev *dev, int where, u16 *val); diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c index 2f0ae70710d..b1724cf31b6 100644 --- a/drivers/pci/probe.c +++ b/drivers/pci/probe.c @@ -860,49 +860,6 @@ int pci_cfg_space_size_ext(struct pci_dev *dev) return PCI_CFG_SPACE_SIZE; } -/** - * pci_disable_pme - Disable the PME function of PCI device - * @dev: PCI device affected - * -EINVAL is returned if PCI device doesn't support PME. - * Zero is returned if the PME is supported and can be disabled. - */ -static int pci_disable_pme(struct pci_dev *dev) -{ - int pm; - u16 value; - - /* find PCI PM capability in list */ - pm = pci_find_capability(dev, PCI_CAP_ID_PM); - - /* If device doesn't support PM Capabilities, it means that PME is - * not supported. - */ - if (!pm) - return -EINVAL; - /* Check device's ability to generate PME# */ - pci_read_config_word(dev, pm + PCI_PM_PMC, &value); - - value &= PCI_PM_CAP_PME_MASK; - /* Check if it can generate PME# */ - if (!value) { - /* - * If it is zero, it means that PME is still unsupported - * although there exists the PM capability. - */ - return -EINVAL; - } - - pci_read_config_word(dev, pm + PCI_PM_CTRL, &value); - - /* Clear PME_Status by writing 1 to it */ - value |= PCI_PM_CTRL_PME_STATUS ; - /* Disable PME enable bit */ - value &= ~PCI_PM_CTRL_PME_ENABLE; - pci_write_config_word(dev, pm + PCI_PM_CTRL, value); - - return 0; -} - int pci_cfg_space_size(struct pci_dev *dev) { int pos; @@ -1010,7 +967,6 @@ static struct pci_dev *pci_scan_device(struct pci_bus *bus, int devfn) } pci_vpd_pci22_init(dev); - pci_disable_pme(dev); return dev; } @@ -1031,6 +987,9 @@ void pci_device_add(struct pci_dev *dev, struct pci_bus *bus) /* Fix up broken headers */ pci_fixup_device(pci_fixup_header, dev); + /* Initialize power management of the device */ + pci_pm_init(dev); + /* * Add the device to our list of discovered devices * and the bus list for fixup functions, etc. diff --git a/include/acpi/acpi_bus.h b/include/acpi/acpi_bus.h index 071daf8db60..7ab5a611e43 100644 --- a/include/acpi/acpi_bus.h +++ b/include/acpi/acpi_bus.h @@ -337,6 +337,7 @@ int acpi_bus_get_status(struct acpi_device *device); int acpi_bus_get_power(acpi_handle handle, int *state); int acpi_bus_set_power(acpi_handle handle, int state); bool acpi_bus_power_manageable(acpi_handle handle); +bool acpi_bus_can_wakeup(acpi_handle handle); #ifdef CONFIG_ACPI_PROC_EVENT int acpi_bus_generate_proc_event(struct acpi_device *device, u8 type, int data); int acpi_bus_generate_proc_event4(const char *class, const char *bid, u8 type, int data); @@ -379,6 +380,7 @@ acpi_handle acpi_get_pci_rootbridge_handle(unsigned int, unsigned int); #ifdef CONFIG_PM_SLEEP int acpi_pm_device_sleep_state(struct device *, int *); +int acpi_pm_device_sleep_wake(struct device *, bool); #else /* !CONFIG_PM_SLEEP */ static inline int acpi_pm_device_sleep_state(struct device *d, int *p) { diff --git a/include/linux/pm_wakeup.h b/include/linux/pm_wakeup.h index f0d0b2cb8d2..3af0c8d05cd 100644 --- a/include/linux/pm_wakeup.h +++ b/include/linux/pm_wakeup.h @@ -35,6 +35,11 @@ static inline void device_init_wakeup(struct device *dev, int val) dev->power.can_wakeup = dev->power.should_wakeup = !!val; } +static inline void device_set_wakeup_capable(struct device *dev, int val) +{ + dev->power.can_wakeup = !!val; +} + static inline int device_can_wakeup(struct device *dev) { return dev->power.can_wakeup; @@ -47,21 +52,7 @@ static inline void device_set_wakeup_enable(struct device *dev, int val) static inline int device_may_wakeup(struct device *dev) { - return dev->power.can_wakeup & dev->power.should_wakeup; -} - -/* - * Platform hook to activate device wakeup capability, if that's not already - * handled by enable_irq_wake() etc. - * Returns zero on success, else negative errno - */ -extern int (*platform_enable_wakeup)(struct device *dev, int is_on); - -static inline int call_platform_enable_wakeup(struct device *dev, int is_on) -{ - if (platform_enable_wakeup) - return (*platform_enable_wakeup)(dev, is_on); - return 0; + return dev->power.can_wakeup && dev->power.should_wakeup; } #else /* !CONFIG_PM */ @@ -80,11 +71,6 @@ static inline int device_can_wakeup(struct device *dev) #define device_set_wakeup_enable(dev, val) do {} while (0) #define device_may_wakeup(dev) 0 -static inline int call_platform_enable_wakeup(struct device *dev, int is_on) -{ - return 0; -} - #endif /* !CONFIG_PM */ #endif /* _LINUX_PM_WAKEUP_H */ -- cgit v1.2.3-70-g09d2 From 23ca4bba3e20c6c3cb11c1bb0ab4770b724d39ac Mon Sep 17 00:00:00 2001 From: Mike Travis Date: Mon, 12 May 2008 21:21:12 +0200 Subject: x86: cleanup early per cpu variables/accesses v4 * Introduce a new PER_CPU macro called "EARLY_PER_CPU". This is used by some per_cpu variables that are initialized and accessed before there are per_cpu areas allocated. ["Early" in respect to per_cpu variables is "earlier than the per_cpu areas have been setup".] This patchset adds these new macros: DEFINE_EARLY_PER_CPU(_type, _name, _initvalue) EXPORT_EARLY_PER_CPU_SYMBOL(_name) DECLARE_EARLY_PER_CPU(_type, _name) early_per_cpu_ptr(_name) early_per_cpu_map(_name, _idx) early_per_cpu(_name, _cpu) The DEFINE macro defines the per_cpu variable as well as the early map and pointer. It also initializes the per_cpu variable and map elements to "_initvalue". The early_* macros provide access to the initial map (usually setup during system init) and the early pointer. This pointer is initialized to point to the early map but is then NULL'ed when the actual per_cpu areas are setup. After that the per_cpu variable is the correct access to the variable. The early_per_cpu() macro is not very efficient but does show how to access the variable if you have a function that can be called both "early" and "late". It tests the early ptr to be NULL, and if not then it's still valid. Otherwise, the per_cpu variable is used instead: #define early_per_cpu(_name, _cpu) \ (early_per_cpu_ptr(_name) ? \ early_per_cpu_ptr(_name)[_cpu] : \ per_cpu(_name, _cpu)) A better method is to actually check the pointer manually. In the case below, numa_set_node can be called both "early" and "late": void __cpuinit numa_set_node(int cpu, int node) { int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map); if (cpu_to_node_map) cpu_to_node_map[cpu] = node; else per_cpu(x86_cpu_to_node_map, cpu) = node; } * Add a flag "arch_provides_topology_pointers" that indicates pointers to topology cpumask_t maps are available. Otherwise, use the function returning the cpumask_t value. This is useful if cpumask_t set size is very large to avoid copying data on to/off of the stack. * The coverage of CONFIG_DEBUG_PER_CPU_MAPS has been increased while the non-debug case has been optimized a bit. * Remove an unreferenced compiler warning in drivers/base/topology.c * Clean up #ifdef in setup.c For inclusion into sched-devel/latest tree. Based on: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux-2.6.git + sched-devel/latest .../mingo/linux-2.6-sched-devel.git Signed-off-by: Mike Travis Signed-off-by: Ingo Molnar Signed-off-by: Thomas Gleixner --- arch/x86/Kconfig | 2 +- arch/x86/Kconfig.debug | 2 +- arch/x86/kernel/apic_32.c | 9 +-- arch/x86/kernel/apic_64.c | 11 ++-- arch/x86/kernel/setup.c | 96 ++++++++++++++++++++++++++---- arch/x86/kernel/setup_32.c | 24 -------- arch/x86/kernel/setup_64.c | 9 --- arch/x86/kernel/smpboot.c | 20 +------ arch/x86/mm/numa_64.c | 43 ++++---------- arch/x86/mm/srat_64.c | 2 +- drivers/base/topology.c | 25 +++++++- include/asm-x86/numa_64.h | 19 +++--- include/asm-x86/percpu.h | 46 +++++++++++++++ include/asm-x86/smp.h | 15 +---- include/asm-x86/topology.h | 143 ++++++++++++++++++++++++++------------------- 15 files changed, 270 insertions(+), 196 deletions(-) (limited to 'drivers/base') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 2e325521e5e..4469a0db1ae 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -121,7 +121,7 @@ config ARCH_HAS_CACHE_LINE_SIZE def_bool y config HAVE_SETUP_PER_CPU_AREA - def_bool X86_64 || (X86_SMP && !X86_VOYAGER) + def_bool X86_64_SMP || (X86_SMP && !X86_VOYAGER) config HAVE_CPUMASK_OF_CPU_MAP def_bool X86_64_SMP diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 18363374d51..24ca95a0ba5 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -60,7 +60,7 @@ config DEBUG_PAGEALLOC config DEBUG_PER_CPU_MAPS bool "Debug access to per_cpu maps" depends on DEBUG_KERNEL - depends on X86_64_SMP + depends on X86_SMP default n help Say Y to verify that the per_cpu map being accessed has diff --git a/arch/x86/kernel/apic_32.c b/arch/x86/kernel/apic_32.c index 4b99b1bdeb6..f17c1c1bc38 100644 --- a/arch/x86/kernel/apic_32.c +++ b/arch/x86/kernel/apic_32.c @@ -52,9 +52,6 @@ unsigned long mp_lapic_addr; -DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid); - /* * Knob to control our willingness to enable the local APIC. * @@ -1534,9 +1531,9 @@ void __cpuinit generic_processor_info(int apicid, int version) } #ifdef CONFIG_SMP /* are we being called early in kernel startup? */ - if (x86_cpu_to_apicid_early_ptr) { - u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; - u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + if (early_per_cpu_ptr(x86_cpu_to_apicid)) { + u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); + u16 *bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); cpu_to_apicid[cpu] = apicid; bios_cpu_apicid[cpu] = apicid; diff --git a/arch/x86/kernel/apic_64.c b/arch/x86/kernel/apic_64.c index 0633cfd0dc2..4fd21f7d698 100644 --- a/arch/x86/kernel/apic_64.c +++ b/arch/x86/kernel/apic_64.c @@ -87,9 +87,6 @@ static unsigned long apic_phys; unsigned long mp_lapic_addr; -DEFINE_PER_CPU(u16, x86_bios_cpu_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_bios_cpu_apicid); - unsigned int __cpuinitdata maxcpus = NR_CPUS; /* * Get the LAPIC version @@ -1091,9 +1088,9 @@ void __cpuinit generic_processor_info(int apicid, int version) cpu = 0; } /* are we being called early in kernel startup? */ - if (x86_cpu_to_apicid_early_ptr) { - u16 *cpu_to_apicid = x86_cpu_to_apicid_early_ptr; - u16 *bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + if (early_per_cpu_ptr(x86_cpu_to_apicid)) { + u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); + u16 *bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); cpu_to_apicid[cpu] = apicid; bios_cpu_apicid[cpu] = apicid; @@ -1269,7 +1266,7 @@ __cpuinit int apic_is_clustered_box(void) if ((boot_cpu_data.x86_vendor == X86_VENDOR_AMD) && !is_vsmp_box()) return 0; - bios_cpu_apicid = x86_bios_cpu_apicid_early_ptr; + bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); bitmap_zero(clustermap, NUM_APIC_CLUSTERS); for (i = 0; i < NR_CPUS; i++) { diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 6f80b852a19..03caa8e4351 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -19,13 +19,23 @@ unsigned disabled_cpus __cpuinitdata; unsigned int boot_cpu_physical_apicid = -1U; EXPORT_SYMBOL(boot_cpu_physical_apicid); -DEFINE_PER_CPU(u16, x86_cpu_to_apicid) = BAD_APICID; -EXPORT_PER_CPU_SYMBOL(x86_cpu_to_apicid); - /* Bitmask of physically existing CPUs */ physid_mask_t phys_cpu_present_map; #endif +/* map cpu index to physical APIC ID */ +DEFINE_EARLY_PER_CPU(u16, x86_cpu_to_apicid, BAD_APICID); +DEFINE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid, BAD_APICID); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_apicid); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_bios_cpu_apicid); + +#if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) +#define X86_64_NUMA 1 + +DEFINE_EARLY_PER_CPU(int, x86_cpu_to_node_map, NUMA_NO_NODE); +EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_node_map); +#endif + #if defined(CONFIG_HAVE_SETUP_PER_CPU_AREA) && defined(CONFIG_X86_SMP) /* * Copy data used in early init routines from the initial arrays to the @@ -37,20 +47,21 @@ static void __init setup_per_cpu_maps(void) int cpu; for_each_possible_cpu(cpu) { - per_cpu(x86_cpu_to_apicid, cpu) = x86_cpu_to_apicid_init[cpu]; + per_cpu(x86_cpu_to_apicid, cpu) = + early_per_cpu_map(x86_cpu_to_apicid, cpu); per_cpu(x86_bios_cpu_apicid, cpu) = - x86_bios_cpu_apicid_init[cpu]; -#ifdef CONFIG_NUMA + early_per_cpu_map(x86_bios_cpu_apicid, cpu); +#ifdef X86_64_NUMA per_cpu(x86_cpu_to_node_map, cpu) = - x86_cpu_to_node_map_init[cpu]; + early_per_cpu_map(x86_cpu_to_node_map, cpu); #endif } /* indicate the early static arrays will soon be gone */ - x86_cpu_to_apicid_early_ptr = NULL; - x86_bios_cpu_apicid_early_ptr = NULL; -#ifdef CONFIG_NUMA - x86_cpu_to_node_map_early_ptr = NULL; + early_per_cpu_ptr(x86_cpu_to_apicid) = NULL; + early_per_cpu_ptr(x86_bios_cpu_apicid) = NULL; +#ifdef X86_64_NUMA + early_per_cpu_ptr(x86_cpu_to_node_map) = NULL; #endif } @@ -109,7 +120,8 @@ void __init setup_per_cpu_areas(void) if (!node_online(node) || !NODE_DATA(node)) { ptr = alloc_bootmem_pages(size); printk(KERN_INFO - "cpu %d has no node or node-local memory\n", i); + "cpu %d has no node %d or node-local memory\n", + i, node); } else ptr = alloc_bootmem_pages_node(NODE_DATA(node), size); @@ -137,3 +149,63 @@ void __init setup_per_cpu_areas(void) } #endif + +#ifdef X86_64_NUMA +void __cpuinit numa_set_node(int cpu, int node) +{ + int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map); + + if (cpu_to_node_map) + cpu_to_node_map[cpu] = node; + + else if (per_cpu_offset(cpu)) + per_cpu(x86_cpu_to_node_map, cpu) = node; + + else + Dprintk(KERN_INFO "Setting node for non-present cpu %d\n", cpu); +} + +void __cpuinit numa_clear_node(int cpu) +{ + numa_set_node(cpu, NUMA_NO_NODE); +} + +void __cpuinit numa_add_cpu(int cpu) +{ + cpu_set(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]); +} + +void __cpuinit numa_remove_cpu(int cpu) +{ + cpu_clear(cpu, node_to_cpumask_map[cpu_to_node(cpu)]); +} +#endif /* CONFIG_NUMA */ + +#if defined(CONFIG_DEBUG_PER_CPU_MAPS) && defined(CONFIG_X86_64) + +int cpu_to_node(int cpu) +{ + if (early_per_cpu_ptr(x86_cpu_to_node_map)) { + printk(KERN_WARNING + "cpu_to_node(%d): usage too early!\n", cpu); + dump_stack(); + return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu]; + } + return per_cpu(x86_cpu_to_node_map, cpu); +} +EXPORT_SYMBOL(cpu_to_node); + +int early_cpu_to_node(int cpu) +{ + if (early_per_cpu_ptr(x86_cpu_to_node_map)) + return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu]; + + if (!per_cpu_offset(cpu)) { + printk(KERN_WARNING + "early_cpu_to_node(%d): no per_cpu area!\n", cpu); + dump_stack(); + return NUMA_NO_NODE; + } + return per_cpu(x86_cpu_to_node_map, cpu); +} +#endif diff --git a/arch/x86/kernel/setup_32.c b/arch/x86/kernel/setup_32.c index 5a2f8e06388..ccd5f5cdbbe 100644 --- a/arch/x86/kernel/setup_32.c +++ b/arch/x86/kernel/setup_32.c @@ -737,18 +737,6 @@ char * __init __attribute__((weak)) memory_setup(void) return machine_specific_memory_setup(); } -#ifdef CONFIG_NUMA -/* - * In the golden day, when everything among i386 and x86_64 will be - * integrated, this will not live here - */ -void *x86_cpu_to_node_map_early_ptr; -int x86_cpu_to_node_map_init[NR_CPUS] = { - [0 ... NR_CPUS-1] = NUMA_NO_NODE -}; -DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE; -#endif - /* * Determine if we were loaded by an EFI loader. If so, then we have also been * passed the efi memmap, systab, etc., so we should use these data structures @@ -887,18 +875,6 @@ void __init setup_arch(char **cmdline_p) io_delay_init(); -#ifdef CONFIG_X86_SMP - /* - * setup to use the early static init tables during kernel startup - * X86_SMP will exclude sub-arches that don't deal well with it. - */ - x86_cpu_to_apicid_early_ptr = (void *)x86_cpu_to_apicid_init; - x86_bios_cpu_apicid_early_ptr = (void *)x86_bios_cpu_apicid_init; -#ifdef CONFIG_NUMA - x86_cpu_to_node_map_early_ptr = (void *)x86_cpu_to_node_map_init; -#endif -#endif - #ifdef CONFIG_X86_GENERICARCH generic_apic_probe(); #endif diff --git a/arch/x86/kernel/setup_64.c b/arch/x86/kernel/setup_64.c index 6dff1286ad8..e8df64fad54 100644 --- a/arch/x86/kernel/setup_64.c +++ b/arch/x86/kernel/setup_64.c @@ -406,15 +406,6 @@ void __init setup_arch(char **cmdline_p) kvmclock_init(); #endif -#ifdef CONFIG_SMP - /* setup to use the early static init tables during kernel startup */ - x86_cpu_to_apicid_early_ptr = (void *)x86_cpu_to_apicid_init; - x86_bios_cpu_apicid_early_ptr = (void *)x86_bios_cpu_apicid_init; -#ifdef CONFIG_NUMA - x86_cpu_to_node_map_early_ptr = (void *)x86_cpu_to_node_map_init; -#endif -#endif - #ifdef CONFIG_ACPI /* * Initialize the ACPI boot-time table parser (gets the RSDP and SDT). diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 3e1cecedde4..036604d3dae 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -68,22 +68,6 @@ #include #include -/* - * FIXME: For x86_64, those are defined in other files. But moving them here, - * would make the setup areas dependent on smp, which is a loss. When we - * integrate apic between arches, we can probably do a better job, but - * right now, they'll stay here -- glommer - */ - -/* which logical CPU number maps to which CPU (physical APIC ID) */ -u16 x86_cpu_to_apicid_init[NR_CPUS] __initdata = - { [0 ... NR_CPUS-1] = BAD_APICID }; -void *x86_cpu_to_apicid_early_ptr; - -u16 x86_bios_cpu_apicid_init[NR_CPUS] __initdata - = { [0 ... NR_CPUS-1] = BAD_APICID }; -void *x86_bios_cpu_apicid_early_ptr; - #ifdef CONFIG_X86_32 u8 apicid_2_node[MAX_APICID]; static int low_mappings; @@ -992,7 +976,7 @@ do_rest: /* Try to put things back the way they were before ... */ unmap_cpu_to_logical_apicid(cpu); #ifdef CONFIG_X86_64 - clear_node_cpumask(cpu); /* was set by numa_add_cpu */ + numa_remove_cpu(cpu); /* was set by numa_add_cpu */ #endif cpu_clear(cpu, cpu_callout_map); /* was set by do_boot_cpu() */ cpu_clear(cpu, cpu_initialized); /* was set by cpu_init() */ @@ -1373,7 +1357,7 @@ static void __ref remove_cpu_from_maps(int cpu) cpu_clear(cpu, cpu_callin_map); /* was set by cpu_init() */ clear_bit(cpu, (unsigned long *)&cpu_initialized); - clear_node_cpumask(cpu); + numa_remove_cpu(cpu); #endif } diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index c5066d519e5..970f86775c4 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -31,16 +31,6 @@ bootmem_data_t plat_node_bdata[MAX_NUMNODES]; struct memnode memnode; -#ifdef CONFIG_SMP -int x86_cpu_to_node_map_init[NR_CPUS] = { - [0 ... NR_CPUS-1] = NUMA_NO_NODE -}; -void *x86_cpu_to_node_map_early_ptr; -EXPORT_SYMBOL(x86_cpu_to_node_map_early_ptr); -#endif -DEFINE_PER_CPU(int, x86_cpu_to_node_map) = NUMA_NO_NODE; -EXPORT_PER_CPU_SYMBOL(x86_cpu_to_node_map); - s16 apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = { [0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE }; @@ -577,24 +567,6 @@ void __init numa_initmem_init(unsigned long start_pfn, unsigned long end_pfn) setup_node_bootmem(0, start_pfn << PAGE_SHIFT, end_pfn << PAGE_SHIFT); } -__cpuinit void numa_add_cpu(int cpu) -{ - set_bit(cpu, - (unsigned long *)&node_to_cpumask_map[early_cpu_to_node(cpu)]); -} - -void __cpuinit numa_set_node(int cpu, int node) -{ - int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr; - - if(cpu_to_node_map) - cpu_to_node_map[cpu] = node; - else if(per_cpu_offset(cpu)) - per_cpu(x86_cpu_to_node_map, cpu) = node; - else - Dprintk(KERN_INFO "Setting node for non-present cpu %d\n", cpu); -} - unsigned long __init numa_free_all_bootmem(void) { unsigned long pages = 0; @@ -641,6 +613,7 @@ static __init int numa_setup(char *opt) } early_param("numa", numa_setup); +#ifdef CONFIG_NUMA /* * Setup early cpu_to_node. * @@ -652,14 +625,19 @@ early_param("numa", numa_setup); * is already initialized in a round robin manner at numa_init_array, * prior to this call, and this initialization is good enough * for the fake NUMA cases. + * + * Called before the per_cpu areas are setup. */ void __init init_cpu_to_node(void) { - int i; + int cpu; + u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid); - for (i = 0; i < NR_CPUS; i++) { + BUG_ON(cpu_to_apicid == NULL); + + for_each_possible_cpu(cpu) { int node; - u16 apicid = x86_cpu_to_apicid_init[i]; + u16 apicid = cpu_to_apicid[cpu]; if (apicid == BAD_APICID) continue; @@ -668,8 +646,9 @@ void __init init_cpu_to_node(void) continue; if (!node_online(node)) continue; - numa_set_node(i, node); + numa_set_node(cpu, node); } } +#endif diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index 99649dccad2..012220e31c9 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -376,7 +376,7 @@ int __init acpi_scan_nodes(unsigned long start, unsigned long end) if (node == NUMA_NO_NODE) continue; if (!node_isset(node, node_possible_map)) - numa_set_node(i, NUMA_NO_NODE); + numa_clear_node(i); } numa_init_array(); return 0; diff --git a/drivers/base/topology.c b/drivers/base/topology.c index fdf4044d2e7..1efe162e16d 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -40,6 +40,7 @@ static ssize_t show_##name(struct sys_device *dev, char *buf) \ return sprintf(buf, "%d\n", topology_##name(cpu)); \ } +#if defined(topology_thread_siblings) || defined(topology_core_siblings) static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) { ptrdiff_t len = PTR_ALIGN(buf + PAGE_SIZE - 1, PAGE_SIZE) - buf; @@ -54,21 +55,41 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) } return n; } +#endif +#ifdef arch_provides_topology_pointers #define define_siblings_show_map(name) \ -static inline ssize_t show_##name(struct sys_device *dev, char *buf) \ +static ssize_t show_##name(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ return show_cpumap(0, &(topology_##name(cpu)), buf); \ } #define define_siblings_show_list(name) \ -static inline ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ +static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ { \ unsigned int cpu = dev->id; \ return show_cpumap(1, &(topology_##name(cpu)), buf); \ } +#else +#define define_siblings_show_map(name) \ +static ssize_t show_##name(struct sys_device *dev, char *buf) \ +{ \ + unsigned int cpu = dev->id; \ + cpumask_t mask = topology_##name(cpu); \ + return show_cpumap(0, &mask, buf); \ +} + +#define define_siblings_show_list(name) \ +static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ +{ \ + unsigned int cpu = dev->id; \ + cpumask_t mask = topology_##name(cpu); \ + return show_cpumap(1, &mask, buf); \ +} +#endif + #define define_siblings_show_func(name) \ define_siblings_show_map(name); define_siblings_show_list(name) diff --git a/include/asm-x86/numa_64.h b/include/asm-x86/numa_64.h index 22e87c9f6a8..b510daf4f4d 100644 --- a/include/asm-x86/numa_64.h +++ b/include/asm-x86/numa_64.h @@ -14,11 +14,9 @@ extern int compute_hash_shift(struct bootnode *nodes, int numblks, #define ZONE_ALIGN (1UL << (MAX_ORDER+PAGE_SHIFT)) -extern void numa_add_cpu(int cpu); extern void numa_init_array(void); extern int numa_off; -extern void numa_set_node(int cpu, int node); extern void srat_reserve_add_area(int nodeid); extern int hotadd_percent; @@ -31,15 +29,16 @@ extern void setup_node_bootmem(int nodeid, unsigned long start, #ifdef CONFIG_NUMA extern void __init init_cpu_to_node(void); - -static inline void clear_node_cpumask(int cpu) -{ - clear_bit(cpu, (unsigned long *)&node_to_cpumask_map[cpu_to_node(cpu)]); -} - +extern void __cpuinit numa_set_node(int cpu, int node); +extern void __cpuinit numa_clear_node(int cpu); +extern void __cpuinit numa_add_cpu(int cpu); +extern void __cpuinit numa_remove_cpu(int cpu); #else -#define init_cpu_to_node() do {} while (0) -#define clear_node_cpumask(cpu) do {} while (0) +static inline void init_cpu_to_node(void) { } +static inline void numa_set_node(int cpu, int node) { } +static inline void numa_clear_node(int cpu) { } +static inline void numa_add_cpu(int cpu, int node) { } +static inline void numa_remove_cpu(int cpu) { } #endif #endif diff --git a/include/asm-x86/percpu.h b/include/asm-x86/percpu.h index 736fc3bb8e1..912a3a17b9d 100644 --- a/include/asm-x86/percpu.h +++ b/include/asm-x86/percpu.h @@ -143,4 +143,50 @@ do { \ #define x86_or_percpu(var, val) percpu_to_op("or", per_cpu__##var, val) #endif /* !__ASSEMBLY__ */ #endif /* !CONFIG_X86_64 */ + +#ifdef CONFIG_SMP + +/* + * Define the "EARLY_PER_CPU" macros. These are used for some per_cpu + * variables that are initialized and accessed before there are per_cpu + * areas allocated. + */ + +#define DEFINE_EARLY_PER_CPU(_type, _name, _initvalue) \ + DEFINE_PER_CPU(_type, _name) = _initvalue; \ + __typeof__(_type) _name##_early_map[NR_CPUS] __initdata = \ + { [0 ... NR_CPUS-1] = _initvalue }; \ + __typeof__(_type) *_name##_early_ptr = _name##_early_map + +#define EXPORT_EARLY_PER_CPU_SYMBOL(_name) \ + EXPORT_PER_CPU_SYMBOL(_name) + +#define DECLARE_EARLY_PER_CPU(_type, _name) \ + DECLARE_PER_CPU(_type, _name); \ + extern __typeof__(_type) *_name##_early_ptr; \ + extern __typeof__(_type) _name##_early_map[] + +#define early_per_cpu_ptr(_name) (_name##_early_ptr) +#define early_per_cpu_map(_name, _idx) (_name##_early_map[_idx]) +#define early_per_cpu(_name, _cpu) \ + (early_per_cpu_ptr(_name) ? \ + early_per_cpu_ptr(_name)[_cpu] : \ + per_cpu(_name, _cpu)) + +#else /* !CONFIG_SMP */ +#define DEFINE_EARLY_PER_CPU(_type, _name, _initvalue) \ + DEFINE_PER_CPU(_type, _name) = _initvalue + +#define EXPORT_EARLY_PER_CPU_SYMBOL(_name) \ + EXPORT_PER_CPU_SYMBOL(_name) + +#define DECLARE_EARLY_PER_CPU(_type, _name) \ + DECLARE_PER_CPU(_type, _name) + +#define early_per_cpu(_name, _cpu) per_cpu(_name, _cpu) +#define early_per_cpu_ptr(_name) NULL +/* no early_per_cpu_map() */ + +#endif /* !CONFIG_SMP */ + #endif /* _ASM_X86_PERCPU_H_ */ diff --git a/include/asm-x86/smp.h b/include/asm-x86/smp.h index 1ebaa5cd311..ec841639fb4 100644 --- a/include/asm-x86/smp.h +++ b/include/asm-x86/smp.h @@ -29,21 +29,12 @@ extern int smp_num_siblings; extern unsigned int num_processors; extern cpumask_t cpu_initialized; -#ifdef CONFIG_SMP -extern u16 x86_cpu_to_apicid_init[]; -extern u16 x86_bios_cpu_apicid_init[]; -extern void *x86_cpu_to_apicid_early_ptr; -extern void *x86_bios_cpu_apicid_early_ptr; -#else -#define x86_cpu_to_apicid_early_ptr NULL -#define x86_bios_cpu_apicid_early_ptr NULL -#endif - DECLARE_PER_CPU(cpumask_t, cpu_sibling_map); DECLARE_PER_CPU(cpumask_t, cpu_core_map); DECLARE_PER_CPU(u16, cpu_llc_id); -DECLARE_PER_CPU(u16, x86_cpu_to_apicid); -DECLARE_PER_CPU(u16, x86_bios_cpu_apicid); + +DECLARE_EARLY_PER_CPU(u16, x86_cpu_to_apicid); +DECLARE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid); /* Static state in head.S used to set up a CPU */ extern struct { diff --git a/include/asm-x86/topology.h b/include/asm-x86/topology.h index dcf3f8131d6..dac09cb66dc 100644 --- a/include/asm-x86/topology.h +++ b/include/asm-x86/topology.h @@ -35,87 +35,67 @@ # endif #endif +/* Node not present */ +#define NUMA_NO_NODE (-1) + #ifdef CONFIG_NUMA #include #include -/* Mappings between logical cpu number and node number */ #ifdef CONFIG_X86_32 -extern int cpu_to_node_map[]; -#else -/* Returns the number of the current Node. */ -#define numa_node_id() (early_cpu_to_node(raw_smp_processor_id())) -#endif - -DECLARE_PER_CPU(int, x86_cpu_to_node_map); - -#ifdef CONFIG_SMP -extern int x86_cpu_to_node_map_init[]; -extern void *x86_cpu_to_node_map_early_ptr; -#else -#define x86_cpu_to_node_map_early_ptr NULL -#endif +/* Mappings between node number and cpus on that node. */ extern cpumask_t node_to_cpumask_map[]; -#define NUMA_NO_NODE (-1) +/* Mappings between logical cpu number and node number */ +extern int cpu_to_node_map[]; /* Returns the number of the node containing CPU 'cpu' */ -#ifdef CONFIG_X86_32 -#define early_cpu_to_node(cpu) cpu_to_node(cpu) static inline int cpu_to_node(int cpu) { return cpu_to_node_map[cpu]; } +#define early_cpu_to_node(cpu) cpu_to_node(cpu) #else /* CONFIG_X86_64 */ -#ifdef CONFIG_SMP -static inline int early_cpu_to_node(int cpu) -{ - int *cpu_to_node_map = x86_cpu_to_node_map_early_ptr; - - if (cpu_to_node_map) - return cpu_to_node_map[cpu]; - else if (per_cpu_offset(cpu)) - return per_cpu(x86_cpu_to_node_map, cpu); - else - return NUMA_NO_NODE; -} -#else -#define early_cpu_to_node(cpu) cpu_to_node(cpu) -#endif +/* Mappings between node number and cpus on that node. */ +extern cpumask_t node_to_cpumask_map[]; + +/* Mappings between logical cpu number and node number */ +DECLARE_EARLY_PER_CPU(int, x86_cpu_to_node_map); + +/* Returns the number of the current Node. */ +#define numa_node_id() (per_cpu(x86_cpu_to_node_map, raw_smp_processor_id())) + +#ifdef CONFIG_DEBUG_PER_CPU_MAPS +extern int cpu_to_node(int cpu); +extern int early_cpu_to_node(int cpu); +extern cpumask_t *_node_to_cpumask_ptr(int node); +extern cpumask_t node_to_cpumask(int node); +#else /* !CONFIG_DEBUG_PER_CPU_MAPS */ + +/* Returns the number of the node containing CPU 'cpu' */ static inline int cpu_to_node(int cpu) { -#ifdef CONFIG_DEBUG_PER_CPU_MAPS - if (x86_cpu_to_node_map_early_ptr) { - printk("KERN_NOTICE cpu_to_node(%d): usage too early!\n", - (int)cpu); - dump_stack(); - return ((int *)x86_cpu_to_node_map_early_ptr)[cpu]; - } -#endif return per_cpu(x86_cpu_to_node_map, cpu); } -#ifdef CONFIG_NUMA - -/* Returns a pointer to the cpumask of CPUs on Node 'node'. */ -#define node_to_cpumask_ptr(v, node) \ - cpumask_t *v = &(node_to_cpumask_map[node]) - -#define node_to_cpumask_ptr_next(v, node) \ - v = &(node_to_cpumask_map[node]) -#endif +/* Same function but used if called before per_cpu areas are setup */ +static inline int early_cpu_to_node(int cpu) +{ + if (early_per_cpu_ptr(x86_cpu_to_node_map)) + return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu]; -#endif /* CONFIG_X86_64 */ + return per_cpu(x86_cpu_to_node_map, cpu); +} -/* - * Returns the number of the node containing Node 'node'. This - * architecture is flat, so it is a pretty simple function! - */ -#define parent_node(node) (node) +/* Returns a pointer to the cpumask of CPUs on Node 'node'. */ +static inline cpumask_t *_node_to_cpumask_ptr(int node) +{ + return &node_to_cpumask_map[node]; +} /* Returns a bitmask of CPUs on Node 'node'. */ static inline cpumask_t node_to_cpumask(int node) @@ -123,14 +103,29 @@ static inline cpumask_t node_to_cpumask(int node) return node_to_cpumask_map[node]; } +#endif /* !CONFIG_DEBUG_PER_CPU_MAPS */ +#endif /* CONFIG_X86_64 */ + +/* Replace default node_to_cpumask_ptr with optimized version */ +#define node_to_cpumask_ptr(v, node) \ + cpumask_t *v = _node_to_cpumask_ptr(node) + +#define node_to_cpumask_ptr_next(v, node) \ + v = _node_to_cpumask_ptr(node) + /* Returns the number of the first CPU on Node 'node'. */ static inline int node_to_first_cpu(int node) { - cpumask_t mask = node_to_cpumask(node); - - return first_cpu(mask); + node_to_cpumask_ptr(mask, node); + return first_cpu(*mask); } +/* + * Returns the number of the node containing Node 'node'. This + * architecture is flat, so it is a pretty simple function! + */ +#define parent_node(node) (node) + #define pcibus_to_node(bus) __pcibus_to_node(bus) #define pcibus_to_cpumask(bus) __pcibus_to_cpumask(bus) @@ -180,8 +175,31 @@ extern int __node_distance(int, int); #define node_distance(a, b) __node_distance(a, b) #endif -#else /* CONFIG_NUMA */ +#else /* !CONFIG_NUMA */ + +#define numa_node_id() 0 +#define cpu_to_node(cpu) 0 +#define early_cpu_to_node(cpu) 0 + +static inline cpumask_t *_node_to_cpumask_ptr(int node) +{ + return &cpu_online_map; +} +static inline cpumask_t node_to_cpumask(int node) +{ + return cpu_online_map; +} +static inline int node_to_first_cpu(int node) +{ + return first_cpu(cpu_online_map); +} + +/* Replace default node_to_cpumask_ptr with optimized version */ +#define node_to_cpumask_ptr(v, node) \ + cpumask_t *v = _node_to_cpumask_ptr(node) +#define node_to_cpumask_ptr_next(v, node) \ + v = _node_to_cpumask_ptr(node) #endif #include @@ -193,6 +211,9 @@ extern cpumask_t cpu_coregroup_map(int cpu); #define topology_core_id(cpu) (cpu_data(cpu).cpu_core_id) #define topology_core_siblings(cpu) (per_cpu(cpu_core_map, cpu)) #define topology_thread_siblings(cpu) (per_cpu(cpu_sibling_map, cpu)) + +/* indicates that pointers to the topology cpumask_t maps are valid */ +#define arch_provides_topology_pointers yes #endif static inline void arch_fix_phys_package_id(int num, u32 slot) @@ -220,4 +241,4 @@ static inline void set_mp_bus_to_node(int busnum, int node) } #endif -#endif +#endif /* _ASM_X86_TOPOLOGY_H */ -- cgit v1.2.3-70-g09d2 From b7a39bd0afc4021e8ad2b1189e884551e147427f Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 23 May 2008 18:38:49 +0100 Subject: firmware: make fw->data const In preparation for supporting firmware files linked into the static kernel, make fw->data const to ensure that users aren't modifying it (so that we can pass a pointer to the original in-kernel copy, rather than having to copy it). Signed-off-by: David Woodhouse --- drivers/base/firmware_class.c | 2 +- include/linux/firmware.h | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 9fd4a853414..264b3a2cd86 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -257,7 +257,7 @@ firmware_data_write(struct kobject *kobj, struct bin_attribute *bin_attr, if (retval) goto out; - memcpy(fw->data + offset, buffer, count); + memcpy((u8 *)fw->data + offset, buffer, count); fw->size = max_t(size_t, offset + count, fw->size); retval = count; diff --git a/include/linux/firmware.h b/include/linux/firmware.h index 6c7eff2ebad..88718d60153 100644 --- a/include/linux/firmware.h +++ b/include/linux/firmware.h @@ -8,7 +8,7 @@ struct firmware { size_t size; - u8 *data; + const u8 *data; }; struct device; -- cgit v1.2.3-70-g09d2 From 5658c769443d543728b6c5c673dffc2df8676317 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 23 May 2008 13:52:42 +0100 Subject: firmware: allow firmware files to be built into kernel image Some drivers have their own hacks to bypass the kernel's firmware loader and build their firmware into the kernel; this renders those unnecessary. Other drivers don't use the firmware loader at all, because they always want the firmware to be available. This allows them to start using the firmware loader. A third set of drivers already use the firmware loader, but can't be used without help from userspace, which sometimes requires an initrd. This allows them to work in a static kernel. Signed-off-by: David Woodhouse --- drivers/base/firmware_class.c | 33 +++++++++++++++++++++++++++++++-- include/asm-generic/vmlinux.lds.h | 7 +++++++ include/linux/firmware.h | 21 +++++++++++++++++++++ 3 files changed, 59 insertions(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 264b3a2cd86..b0be1d18fee 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -49,6 +49,14 @@ struct firmware_priv { struct timer_list timeout; }; +#ifdef CONFIG_FW_LOADER +extern struct builtin_fw __start_builtin_fw[]; +extern struct builtin_fw __end_builtin_fw[]; +#else /* Module case. Avoid ifdefs later; it'll all optimise out */ +static struct builtin_fw *__start_builtin_fw; +static struct builtin_fw *__end_builtin_fw; +#endif + static void fw_load_abort(struct firmware_priv *fw_priv) { @@ -391,13 +399,12 @@ _request_firmware(const struct firmware **firmware_p, const char *name, struct device *f_dev; struct firmware_priv *fw_priv; struct firmware *firmware; + struct builtin_fw *builtin; int retval; if (!firmware_p) return -EINVAL; - printk(KERN_INFO "firmware: requesting %s\n", name); - *firmware_p = firmware = kzalloc(sizeof(*firmware), GFP_KERNEL); if (!firmware) { printk(KERN_ERR "%s: kmalloc(struct firmware) failed\n", @@ -406,6 +413,20 @@ _request_firmware(const struct firmware **firmware_p, const char *name, goto out; } + for (builtin = __start_builtin_fw; builtin != __end_builtin_fw; + builtin++) { + if (strcmp(name, builtin->name)) + continue; + printk(KERN_INFO "firmware: using built-in firmware %s\n", + name); + firmware->size = builtin->size; + firmware->data = builtin->data; + return 0; + } + + if (uevent) + printk(KERN_INFO "firmware: requesting %s\n", name); + retval = fw_setup_device(firmware, &f_dev, name, device, uevent); if (retval) goto error_kfree_fw; @@ -473,8 +494,16 @@ request_firmware(const struct firmware **firmware_p, const char *name, void release_firmware(const struct firmware *fw) { + struct builtin_fw *builtin; + if (fw) { + for (builtin = __start_builtin_fw; builtin != __end_builtin_fw; + builtin++) { + if (fw->data == builtin->data) + goto free_fw; + } vfree(fw->data); + free_fw: kfree(fw); } } diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index f054778e916..8d71a40625f 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -86,6 +86,13 @@ VMLINUX_SYMBOL(__end_pci_fixups_resume) = .; \ } \ \ + /* Built-in firmware blobs */ \ + .builtin_fw : AT(ADDR(.builtin_fw) - LOAD_OFFSET) { \ + VMLINUX_SYMBOL(__start_builtin_fw) = .; \ + *(.builtin_fw) \ + VMLINUX_SYMBOL(__end_builtin_fw) = .; \ + } \ + \ /* RapidIO route ops */ \ .rio_route : AT(ADDR(.rio_route) - LOAD_OFFSET) { \ VMLINUX_SYMBOL(__start_rio_route_ops) = .; \ diff --git a/include/linux/firmware.h b/include/linux/firmware.h index 88718d60153..c8ecf5b2a20 100644 --- a/include/linux/firmware.h +++ b/include/linux/firmware.h @@ -1,7 +1,10 @@ #ifndef _LINUX_FIRMWARE_H #define _LINUX_FIRMWARE_H + #include #include +#include + #define FIRMWARE_NAME_MAX 30 #define FW_ACTION_NOHOTPLUG 0 #define FW_ACTION_HOTPLUG 1 @@ -13,6 +16,24 @@ struct firmware { struct device; +struct builtin_fw { + char *name; + void *data; + unsigned long size; +}; + +/* We have to play tricks here much like stringify() to get the + __COUNTER__ macro to be expanded as we want it */ +#define __fw_concat1(x, y) x##y +#define __fw_concat(x, y) __fw_concat1(x, y) + +#define DECLARE_BUILTIN_FIRMWARE(name, blob) \ + DECLARE_BUILTIN_FIRMWARE_SIZE(name, &(blob), sizeof(blob)) + +#define DECLARE_BUILTIN_FIRMWARE_SIZE(name, blob, size) \ + static const struct builtin_fw __fw_concat(__builtin_fw,__COUNTER__) \ + __used __section(.builtin_fw) = { name, blob, size } + #if defined(CONFIG_FW_LOADER) || (defined(CONFIG_FW_LOADER_MODULE) && defined(MODULE)) int request_firmware(const struct firmware **fw, const char *name, struct device *device); -- cgit v1.2.3-70-g09d2 From 4d2acfbfdf68257e846aaa355edd10fc35ba0feb Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Fri, 23 May 2008 13:58:12 +0100 Subject: firmware: Add CONFIG_EXTRA_FIRMWARE option This allows arbitrary firmware files to be included in the static kernel where the firmware loader can find them without requiring userspace to be alive. (Updated and CONFIG_EXTRA_FIRMWARE_DIR added with lots of help from Johannes Berg). Signed-off-by: David Woodhouse Signed-off-by: Johannes Berg --- Makefile | 2 +- drivers/base/Kconfig | 39 +++++++++++++++++++++++ firmware/Makefile | 88 ++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 128 insertions(+), 1 deletion(-) create mode 100644 firmware/Makefile (limited to 'drivers/base') diff --git a/Makefile b/Makefile index 6315424a00b..f398cffa6c0 100644 --- a/Makefile +++ b/Makefile @@ -450,7 +450,7 @@ scripts: scripts_basic include/config/auto.conf # Objects we will link into vmlinux / subdirs we need to visit init-y := init/ -drivers-y := drivers/ sound/ +drivers-y := drivers/ sound/ firmware/ net-y := net/ libs-y := lib/ core-y := usr/ diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index d7da109c24f..13cfcb435f7 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -34,6 +34,45 @@ config FW_LOADER require userspace firmware loading support, but a module built outside the kernel tree does. +config EXTRA_FIRMWARE + string "External firmware blobs to build into the kernel binary" + depends on FW_LOADER + help + This option allows firmware to be built into the kernel, for the + cases where the user either cannot or doesn't want to provide it from + userspace at runtime (for example, when the firmware in question is + required for accessing the boot device, and the user doesn't want to + use an initrd). + + This option is a string, and takes the (space-separated) names of the + firmware files -- the same names which appear in MODULE_FIRMWARE() + and request_firmware() in the source. These files should exist under + the directory specified by the EXTRA_FIRMWARE_DIR option, which is + by default the firmware/ subdirectory of the kernel source tree. + + So, for example, you might set CONFIG_EXTRA_FIRMWARE="usb8388.bin", + copy the usb8388.bin file into the firmware/ directory, and build the + kernel. Then any request_firmware("usb8388.bin") will be + satisfied internally without needing to call out to userspace. + + WARNING: If you include additional firmware files into your binary + kernel image which are not available under the terms of the GPL, + then it may be a violation of the GPL to distribute the resulting + image -- since it combines both GPL and non-GPL work. You should + consult a lawyer of your own before distributing such an image. + +config EXTRA_FIRMWARE_DIR + string "Firmware blobs root directory" + depends on EXTRA_FIRMWARE != "" + default "firmware" + help + This option controls the directory in which the kernel build system + looks for the firmware files listed in the EXTRA_FIRMWARE option. + The default is the firmware/ directory in the kernel source tree, + but by changing this option you can point it elsewhere, such as + the /lib/firmware/ directory or another separate directory + containing firmware files. + config DEBUG_DRIVER bool "Driver Core verbose debug messages" depends on DEBUG_KERNEL diff --git a/firmware/Makefile b/firmware/Makefile new file mode 100644 index 00000000000..e69461f9362 --- /dev/null +++ b/firmware/Makefile @@ -0,0 +1,88 @@ +# +# kbuild file for firmware/ +# + +# Create $(fwabs) from $(CONFIG_EXTRA_FIRMWARE_DIR) -- if it doesn't have a +# leading /, it's relative to $(srctree). +fwdir := $(subst ",,$(CONFIG_EXTRA_FIRMWARE_DIR)) +fwabs := $(addprefix $(srctree)/,$(filter-out /%,$(fwdir)))$(filter /%,$(fwdir)) + +fw-external-y := $(subst ",,$(CONFIG_EXTRA_FIRMWARE)) + +firmware-y := $(fw-external-y) $(fw-shipped-y) +firmware-dirs := $(sort $(patsubst %,$(objtree)/$(obj)/%/,$(dir $(firmware-y) $(fw-shipped-)))) + +quiet_cmd_mkdir = MKDIR $(patsubst $(objtree)/%,%,$@) + cmd_mkdir = mkdir -p $@ + +quiet_cmd_ihex = IHEX $@ + cmd_ihex = $(OBJCOPY) -Iihex -Obinary $< $@ + +quiet_cmd_fwbin = MK_FW $@ + cmd_fwbin = FWNAME="$(patsubst firmware/%.gen.S,%,$@)"; \ + FWSTR="$(subst /,_,$(subst .,_,$(subst -,_,$(patsubst \ + firmware/%.gen.S,%,$@))))"; \ + ASM_WORD=$(if $(CONFIG_64BIT),.quad,.long); \ + ASM_ALIGN=$(if $(CONFIG_64BIT),3,2); \ + PROGBITS=$(if $(CONFIG_ARM),%,@)progbits; \ + echo "/* Generated by firmware/Makefile */" > $@;\ + echo " .section .rodata" >>$@;\ + echo " .p2align $${ASM_ALIGN}" >>$@;\ + echo "_fw_$${FWSTR}_bin:" >>$@;\ + echo " .incbin \"$(2)\"" >>$@;\ + echo "_fw_end:" >>$@;\ + echo " .section .rodata.str,\"aMS\",$${PROGBITS},1" >>$@;\ + echo " .p2align $${ASM_ALIGN}" >>$@;\ + echo "_fw_$${FWSTR}_name:" >>$@;\ + echo " .string \"$$FWNAME\"" >>$@;\ + echo " .section .builtin_fw,\"a\",$${PROGBITS}" >>$@;\ + echo " .p2align $${ASM_ALIGN}" >>$@;\ + echo " $${ASM_WORD} _fw_$${FWSTR}_name" >>$@;\ + echo " $${ASM_WORD} _fw_$${FWSTR}_bin" >>$@;\ + echo " $${ASM_WORD} _fw_end - _fw_$${FWSTR}_bin" >>$@; + +# One of these files will change, or come into existence, whenever +# the configuration changes between 32-bit and 64-bit. The .S files +# need to change when that happens. +wordsize_deps := $(wildcard include/config/64bit.h include/config/32bit.h \ + include/config/ppc32.h include/config/ppc64.h \ + include/config/superh32.h include/config/superh64.h \ + include/config/x86_32.h include/config/x86_64.h) + +# Workaround for make < 3.81, where .SECONDEXPANSION doesn't work. +# It'll end up depending on these targets, so make them a PHONY rule which +# depends on _all_ the directories in $(firmware-dirs), and it'll work out OK. +PHONY += $(objtree)/$$(%) $(objtree)/$(obj)/$$(%) +$(objtree)/$$(%) $(objtree)/$(obj)/$$(%): $(firmware-dirs) + @true + +# For the $$(dir %) trick, where we need % to be expanded first. +.SECONDEXPANSION: + +$(patsubst %,$(obj)/%.gen.S, $(fw-shipped-y)): %: $(wordsize_deps) \ + | $(objtree)/$$(dir %) + $(call cmd,fwbin,$(patsubst %.gen.S,%,$@)) +$(patsubst %,$(obj)/%.gen.S, $(fw-external-y)): %: $(wordsize_deps) \ + include/config/builtin/firmware/dir.h | $(objtree)/$$(dir %) + $(call cmd,fwbin,$(fwabs)/$(patsubst $(obj)/%.gen.S,%,$@)) + +# The .o files depend on the binaries directly; the .S files don't. +$(patsubst %,$(obj)/%.gen.o, $(fw-shipped-y)): %.gen.o: % +$(patsubst %,$(obj)/%.gen.o, $(fw-external-y)): $(obj)/%.gen.o: $(fwdir)/% + +$(obj)/%: $(obj)/%.ihex | $(objtree)/$(obj)/$$(dir %) + $(call cmd,ihex) + +$(firmware-dirs): + $(call cmd,mkdir) + +obj-y := $(patsubst %,%.gen.o, $(firmware-y)) + +# Remove .S files and binaries created from ihex +# (during 'make clean' .config isn't included so they're all in $(fw-shipped-)) +targets := $(fw-shipped-) $(patsubst $(obj)/%,%, \ + $(shell find $(obj) -name \*.gen.S 2>/dev/null)) + +# Without this, built-in.o won't be created when it's empty, and the +# final vmlinux link will fail. +obj-n := dummy -- cgit v1.2.3-70-g09d2 From d172e7f5c67f2d41f453c7aa83d3bdb405ef8ba5 Mon Sep 17 00:00:00 2001 From: David Woodhouse Date: Wed, 25 Jun 2008 13:56:07 +0100 Subject: firmware: Add CONFIG_FIRMWARE_IN_KERNEL option. This will control whether we build firmware into the kernel image for _every_ driver which we convert to request_firmware(), to avoid a proliferation of 'CONFIG_XXX_FIRMWARE' options for each one. Default to 'y' for now, which is the wrong thing to do but people seem to be insisting on it and refusing to even review patches until it's done. And it does preserve the existing behaviour for built-in drivers. Signed-off-by: David Woodhouse --- drivers/base/Kconfig | 25 +++++++++++++++++++++++++ firmware/Makefile | 5 +++++ 2 files changed, 30 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index 13cfcb435f7..d47482fa1d2 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -34,6 +34,31 @@ config FW_LOADER require userspace firmware loading support, but a module built outside the kernel tree does. +config FIRMWARE_IN_KERNEL + bool "Include in-kernel firmware blobs in kernel binary" + depends on FW_LOADER + default y + help + The kernel source tree includes a number of firmware 'blobs' + which are used by various drivers. The recommended way to + use these is to run "make firmware_install" and to copy the + resulting binary files created in usr/lib/firmware directory + of the kernel tree to the /lib/firmware on your system so + that they can be loaded by userspace helpers on request. + + Enabling this option will build each required firmware blob + into the kernel directly, where request_firmware() will find + them without having to call out to userspace. This may be + useful if your root file system requires a device which uses + such firmware, and do not wish to use an initrd. + + This single option controls the inclusion of firmware for + every driver which usees request_firmare() and ships its + firmware in the kernel source tree, to avoid a proliferation + of 'Include firmware for xxx device' options. + + Say 'N' and let firmware be loaded from userspace. + config EXTRA_FIRMWARE string "External firmware blobs to build into the kernel binary" depends on FW_LOADER diff --git a/firmware/Makefile b/firmware/Makefile index e69461f9362..cc25f5600d5 100644 --- a/firmware/Makefile +++ b/firmware/Makefile @@ -9,6 +9,11 @@ fwabs := $(addprefix $(srctree)/,$(filter-out /%,$(fwdir)))$(filter /%,$(fwdir)) fw-external-y := $(subst ",,$(CONFIG_EXTRA_FIRMWARE)) +# If CONFIG_FIRMWARE_IN_KERNEL is not set, then don't include any firmware +ifneq ($(CONFIG_FIRMWARE_IN_KERNEL),y) +fw-shipped-y := +endif + firmware-y := $(fw-external-y) $(fw-shipped-y) firmware-dirs := $(sort $(patsubst %,$(objtree)/$(obj)/%/,$(dir $(firmware-y) $(fw-shipped-)))) -- cgit v1.2.3-70-g09d2 From e105b8bfc769b0545b6f0f395179d1e43cbee822 Mon Sep 17 00:00:00 2001 From: Dan Williams Date: Mon, 21 Apr 2008 10:51:07 -0700 Subject: sysfs: add /sys/dev/{char,block} to lookup sysfs path by major:minor Why?: There are occasions where userspace would like to access sysfs attributes for a device but it may not know how sysfs has named the device or the path. For example what is the sysfs path for /dev/disk/by-id/ata-ST3160827AS_5MT004CK? With this change a call to stat(2) returns the major:minor then userspace can see that /sys/dev/block/8:32 links to /sys/block/sdc. What are the alternatives?: 1/ Add an ioctl to return the path: Doable, but sysfs is meant to reduce the need to proliferate ioctl interfaces into the kernel, so this seems counter productive. 2/ Use udev to create these symlinks: Also doable, but it adds a udev dependency to utilities that might be running in a limited environment like an initramfs. 3/ Do a full-tree search of sysfs. [kay.sievers@vrfy.org: fix duplicate registrations] [kay.sievers@vrfy.org: cleanup suggestions] Cc: Neil Brown Cc: Tejun Heo Acked-by: Kay Sievers Reviewed-by: SL Baur Acked-by: Kay Sievers Acked-by: Mark Lord Acked-by: H. Peter Anvin Signed-off-by: Dan Williams Signed-off-by: Greg Kroah-Hartman --- Documentation/ABI/testing/sysfs-dev | 20 +++++++++ Documentation/filesystems/sysfs.txt | 6 +++ block/genhd.c | 5 ++- drivers/base/class.c | 4 ++ drivers/base/core.c | 83 ++++++++++++++++++++++++++++++++++++- drivers/usb/core/devio.c | 5 +++ include/linux/device.h | 3 ++ 7 files changed, 124 insertions(+), 2 deletions(-) create mode 100644 Documentation/ABI/testing/sysfs-dev (limited to 'drivers/base') diff --git a/Documentation/ABI/testing/sysfs-dev b/Documentation/ABI/testing/sysfs-dev new file mode 100644 index 00000000000..a9f2b8b0530 --- /dev/null +++ b/Documentation/ABI/testing/sysfs-dev @@ -0,0 +1,20 @@ +What: /sys/dev +Date: April 2008 +KernelVersion: 2.6.26 +Contact: Dan Williams +Description: The /sys/dev tree provides a method to look up the sysfs + path for a device using the information returned from + stat(2). There are two directories, 'block' and 'char', + beneath /sys/dev containing symbolic links with names of + the form ":". These links point to the + corresponding sysfs path for the given device. + + Example: + $ readlink /sys/dev/block/8:32 + ../../block/sdc + + Entries in /sys/dev/char and /sys/dev/block will be + dynamically created and destroyed as devices enter and + leave the system. + +Users: mdadm diff --git a/Documentation/filesystems/sysfs.txt b/Documentation/filesystems/sysfs.txt index 7f27b8f840d..9e9c348275a 100644 --- a/Documentation/filesystems/sysfs.txt +++ b/Documentation/filesystems/sysfs.txt @@ -248,6 +248,7 @@ The top level sysfs directory looks like: block/ bus/ class/ +dev/ devices/ firmware/ net/ @@ -274,6 +275,11 @@ fs/ contains a directory for some filesystems. Currently each filesystem wanting to export attributes must create its own hierarchy below fs/ (see ./fuse.txt for an example). +dev/ contains two directories char/ and block/. Inside these two +directories there are symlinks named :. These symlinks +point to the sysfs directory for the given device. /sys/dev provides a +quick way to lookup the sysfs interface for a device from the result of +a stat(2) operation. More information can driver-model specific features can be found in Documentation/driver-model/. diff --git a/block/genhd.c b/block/genhd.c index 9074f384b09..24e3fc9095f 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -370,7 +370,10 @@ static struct kobject *base_probe(dev_t devt, int *part, void *data) static int __init genhd_device_init(void) { - int error = class_register(&block_class); + int error; + + block_class.dev_kobj = sysfs_dev_block_kobj; + error = class_register(&block_class); if (unlikely(error)) return error; bdev_map = kobj_map_init(base_probe, &block_class_lock); diff --git a/drivers/base/class.c b/drivers/base/class.c index e085af0ff94..71ce3ff6bdf 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -148,6 +148,10 @@ int class_register(struct class *cls) if (error) return error; + /* set the default /sys/dev directory for devices of this class */ + if (!cls->dev_kobj) + cls->dev_kobj = sysfs_dev_char_kobj; + #if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK) /* let the block class directory show up in the root of sysfs */ if (cls != &block_class) diff --git a/drivers/base/core.c b/drivers/base/core.c index ee0a51a3a41..be9aba4dc2f 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -27,6 +27,9 @@ int (*platform_notify)(struct device *dev) = NULL; int (*platform_notify_remove)(struct device *dev) = NULL; +static struct kobject *dev_kobj; +struct kobject *sysfs_dev_char_kobj; +struct kobject *sysfs_dev_block_kobj; #ifdef CONFIG_BLOCK static inline int device_is_not_partition(struct device *dev) @@ -775,6 +778,54 @@ int dev_set_name(struct device *dev, const char *fmt, ...) } EXPORT_SYMBOL_GPL(dev_set_name); +/** + * device_to_dev_kobj - select a /sys/dev/ directory for the device + * @dev: device + * + * By default we select char/ for new entries. Setting class->dev_obj + * to NULL prevents an entry from being created. class->dev_kobj must + * be set (or cleared) before any devices are registered to the class + * otherwise device_create_sys_dev_entry() and + * device_remove_sys_dev_entry() will disagree about the the presence + * of the link. + */ +static struct kobject *device_to_dev_kobj(struct device *dev) +{ + struct kobject *kobj; + + if (dev->class) + kobj = dev->class->dev_kobj; + else + kobj = sysfs_dev_char_kobj; + + return kobj; +} + +static int device_create_sys_dev_entry(struct device *dev) +{ + struct kobject *kobj = device_to_dev_kobj(dev); + int error = 0; + char devt_str[15]; + + if (kobj) { + format_dev_t(devt_str, dev->devt); + error = sysfs_create_link(kobj, &dev->kobj, devt_str); + } + + return error; +} + +static void device_remove_sys_dev_entry(struct device *dev) +{ + struct kobject *kobj = device_to_dev_kobj(dev); + char devt_str[15]; + + if (kobj) { + format_dev_t(devt_str, dev->devt); + sysfs_remove_link(kobj, devt_str); + } +} + /** * device_add - add device to device hierarchy. * @dev: device. @@ -829,6 +880,10 @@ int device_add(struct device *dev) error = device_create_file(dev, &devt_attr); if (error) goto ueventattrError; + + error = device_create_sys_dev_entry(dev); + if (error) + goto devtattrError; } error = device_add_class_symlinks(dev); @@ -872,6 +927,9 @@ int device_add(struct device *dev) AttrsError: device_remove_class_symlinks(dev); SymlinkError: + if (MAJOR(dev->devt)) + device_remove_sys_dev_entry(dev); + devtattrError: if (MAJOR(dev->devt)) device_remove_file(dev, &devt_attr); ueventattrError: @@ -948,8 +1006,10 @@ void device_del(struct device *dev) device_pm_remove(dev); if (parent) klist_del(&dev->knode_parent); - if (MAJOR(dev->devt)) + if (MAJOR(dev->devt)) { + device_remove_sys_dev_entry(dev); device_remove_file(dev, &devt_attr); + } if (dev->class) { device_remove_class_symlinks(dev); @@ -1074,7 +1134,25 @@ int __init devices_init(void) devices_kset = kset_create_and_add("devices", &device_uevent_ops, NULL); if (!devices_kset) return -ENOMEM; + dev_kobj = kobject_create_and_add("dev", NULL); + if (!dev_kobj) + goto dev_kobj_err; + sysfs_dev_block_kobj = kobject_create_and_add("block", dev_kobj); + if (!sysfs_dev_block_kobj) + goto block_kobj_err; + sysfs_dev_char_kobj = kobject_create_and_add("char", dev_kobj); + if (!sysfs_dev_char_kobj) + goto char_kobj_err; + return 0; + + char_kobj_err: + kobject_put(sysfs_dev_block_kobj); + block_kobj_err: + kobject_put(dev_kobj); + dev_kobj_err: + kset_unregister(devices_kset); + return -ENOMEM; } EXPORT_SYMBOL_GPL(device_for_each_child); @@ -1447,4 +1525,7 @@ void device_shutdown(void) dev->driver->shutdown(dev); } } + kobject_put(sysfs_dev_char_kobj); + kobject_put(sysfs_dev_block_kobj); + kobject_put(dev_kobj); } diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index 54a350ccd03..6fbc8f5ab80 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -1792,6 +1792,11 @@ int __init usb_devio_init(void) usb_classdev_class = NULL; goto out; } + /* devices of this class shadow the major:minor of their parent + * device, so clear ->dev_kobj to prevent adding duplicate entries + * to /sys/dev + */ + usb_classdev_class->dev_kobj = NULL; usb_register_notify(&usbdev_nb); #endif diff --git a/include/linux/device.h b/include/linux/device.h index f71a78d123a..e49aa74f248 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -193,6 +193,7 @@ struct class { struct semaphore sem; /* locks children, devices, interfaces */ struct class_attribute *class_attrs; struct device_attribute *dev_attrs; + struct kobject *dev_kobj; int (*dev_uevent)(struct device *dev, struct kobj_uevent_env *env); @@ -205,6 +206,8 @@ struct class { struct pm_ops *pm; }; +extern struct kobject *sysfs_dev_block_kobj; +extern struct kobject *sysfs_dev_char_kobj; extern int __must_check class_register(struct class *class); extern void class_unregister(struct class *class); extern int class_for_each_device(struct class *class, void *data, -- cgit v1.2.3-70-g09d2 From ccea44fadca396b0f89aed5bdeb60e4abb212566 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 21 May 2008 12:52:33 -0700 Subject: driver core: remove device_create() There are no more users of this, and it is racy. Use device_create_drvdata() or device_create_vargs() instead. Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 34 ---------------------------------- include/linux/device.h | 3 --- 2 files changed, 37 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index be9aba4dc2f..4dc0d272f6d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1273,40 +1273,6 @@ struct device *device_create_drvdata(struct class *class, } EXPORT_SYMBOL_GPL(device_create_drvdata); -/** - * device_create - creates a device and registers it with sysfs - * @class: pointer to the struct class that this device should be registered to - * @parent: pointer to the parent struct device of this new device, if any - * @devt: the dev_t for the char device to be added - * @fmt: string for the device's name - * - * This function can be used by char device classes. A struct device - * will be created in sysfs, registered to the specified class. - * - * A "dev" file will be created, showing the dev_t for the device, if - * the dev_t is not 0,0. - * If a pointer to a parent struct device is passed in, the newly created - * struct device will be a child of that device in sysfs. - * The pointer to the struct device will be returned from the call. - * Any further sysfs files that might be required can be created using this - * pointer. - * - * Note: the struct class passed to this function must have previously - * been created with a call to class_create(). - */ -struct device *device_create(struct class *class, struct device *parent, - dev_t devt, const char *fmt, ...) -{ - va_list vargs; - struct device *dev; - - va_start(vargs, fmt); - dev = device_create_vargs(class, parent, devt, NULL, fmt, vargs); - va_end(vargs); - return dev; -} -EXPORT_SYMBOL_GPL(device_create); - static int __match_devt(struct device *dev, void *data) { dev_t *devt = data; diff --git a/include/linux/device.h b/include/linux/device.h index e49aa74f248..a3ef5a2d4fb 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -470,9 +470,6 @@ extern struct device *device_create_vargs(struct class *cls, void *drvdata, const char *fmt, va_list vargs); -extern struct device *device_create(struct class *cls, struct device *parent, - dev_t devt, const char *fmt, ...) - __attribute__((format(printf, 4, 5))); extern struct device *device_create_drvdata(struct class *cls, struct device *parent, dev_t devt, -- cgit v1.2.3-70-g09d2 From 4e10673944a5c386378ff9d692ae37e19993f9d5 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Mon, 21 Jul 2008 20:03:34 -0700 Subject: device create: convert device_create_drvdata to device_create Now that device_create() has been audited, rename things back to the original call to be sane. Keep the device_create_drvdata macro around to make merges easier. Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 11 ++++------- include/linux/device.h | 9 ++++----- 2 files changed, 8 insertions(+), 12 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index 4dc0d272f6d..9ae28aa709d 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1236,7 +1236,7 @@ error: EXPORT_SYMBOL_GPL(device_create_vargs); /** - * device_create_drvdata - creates a device and registers it with sysfs + * device_create - creates a device and registers it with sysfs * @class: pointer to the struct class that this device should be registered to * @parent: pointer to the parent struct device of this new device, if any * @devt: the dev_t for the char device to be added @@ -1257,11 +1257,8 @@ EXPORT_SYMBOL_GPL(device_create_vargs); * Note: the struct class passed to this function must have previously * been created with a call to class_create(). */ -struct device *device_create_drvdata(struct class *class, - struct device *parent, - dev_t devt, - void *drvdata, - const char *fmt, ...) +struct device *device_create(struct class *class, struct device *parent, + dev_t devt, void *drvdata, const char *fmt, ...) { va_list vargs; struct device *dev; @@ -1271,7 +1268,7 @@ struct device *device_create_drvdata(struct class *class, va_end(vargs); return dev; } -EXPORT_SYMBOL_GPL(device_create_drvdata); +EXPORT_SYMBOL_GPL(device_create); static int __match_devt(struct device *dev, void *data) { diff --git a/include/linux/device.h b/include/linux/device.h index a3ef5a2d4fb..de178712e02 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -470,12 +470,11 @@ extern struct device *device_create_vargs(struct class *cls, void *drvdata, const char *fmt, va_list vargs); -extern struct device *device_create_drvdata(struct class *cls, - struct device *parent, - dev_t devt, - void *drvdata, - const char *fmt, ...) +extern struct device *device_create(struct class *cls, struct device *parent, + dev_t devt, void *drvdata, + const char *fmt, ...) __attribute__((format(printf, 5, 6))); +#define device_create_drvdata device_create extern void device_destroy(struct class *cls, dev_t devt); /* -- cgit v1.2.3-70-g09d2 From 93562b537659fc0f63920fd4d9d24f54e434f4c4 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 22 May 2008 17:21:08 -0400 Subject: Driver Core: add ability for class_for_each_device to start in middle of list This mirrors the functionality that driver_for_each_device has as well. We add a start variable, and all callers of the function are fixed up at the same time. The block layer will be using this new functionality in a follow-on patch. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 21 +++++++++++++-------- drivers/i2c/i2c-core.c | 6 ++++-- drivers/ieee1394/nodemgr.c | 14 +++++++++----- drivers/power/apm_power.c | 2 +- drivers/power/power_supply_core.c | 4 ++-- include/linux/device.h | 3 ++- 6 files changed, 31 insertions(+), 19 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/class.c b/drivers/base/class.c index 71ce3ff6bdf..2eb7048003a 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -256,11 +256,14 @@ char *make_class_name(const char *name, struct kobject *kobj) /** * class_for_each_device - device iterator * @class: the class we're iterating + * @start: the device to start with in the list, if any. * @data: data for the callback * @fn: function to be called for each device * * Iterate over @class's list of devices, and call @fn for each, - * passing it @data. + * passing it @data. If @start is set, the list iteration will start + * there, otherwise if it is NULL, the iteration starts at the + * beginning of the list. * * We check the return of @fn each time. If it returns anything * other than 0, we break out and return that value. @@ -269,8 +272,8 @@ char *make_class_name(const char *name, struct kobject *kobj) * re-acquired in @fn, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ -int class_for_each_device(struct class *class, void *data, - int (*fn)(struct device *, void *)) +int class_for_each_device(struct class *class, struct device *start, + void *data, int (*fn)(struct device *, void *)) { struct device *dev; int error = 0; @@ -279,12 +282,14 @@ int class_for_each_device(struct class *class, void *data, return -EINVAL; down(&class->sem); list_for_each_entry(dev, &class->devices, node) { + if (start) { + if (start == dev) + start = NULL; + continue; + } dev = get_device(dev); - if (dev) { - error = fn(dev, data); - put_device(dev); - } else - error = -ENODEV; + error = fn(dev, data); + put_device(dev); if (error) break; } diff --git a/drivers/i2c/i2c-core.c b/drivers/i2c/i2c-core.c index 7608df83d6d..7bf38c41808 100644 --- a/drivers/i2c/i2c-core.c +++ b/drivers/i2c/i2c-core.c @@ -722,7 +722,8 @@ int i2c_register_driver(struct module *owner, struct i2c_driver *driver) INIT_LIST_HEAD(&driver->clients); /* Walk the adapters that are already present */ - class_for_each_device(&i2c_adapter_class, driver, __attach_adapter); + class_for_each_device(&i2c_adapter_class, NULL, driver, + __attach_adapter); mutex_unlock(&core_lock); return 0; @@ -782,7 +783,8 @@ void i2c_del_driver(struct i2c_driver *driver) { mutex_lock(&core_lock); - class_for_each_device(&i2c_adapter_class, driver, __detach_adapter); + class_for_each_device(&i2c_adapter_class, NULL, driver, + __detach_adapter); driver_unregister(&driver->driver); pr_debug("i2c-core: driver [%s] unregistered\n", driver->driver.name); diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c index 05710c7c122..47c0d85e0f3 100644 --- a/drivers/ieee1394/nodemgr.c +++ b/drivers/ieee1394/nodemgr.c @@ -1453,7 +1453,8 @@ static void nodemgr_suspend_ne(struct node_entry *ne) ne->in_limbo = 1; WARN_ON(device_create_file(&ne->device, &dev_attr_ne_in_limbo)); - class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_driver_suspend); + class_for_each_device(&nodemgr_ud_class, NULL, ne, + __nodemgr_driver_suspend); } @@ -1462,7 +1463,8 @@ static void nodemgr_resume_ne(struct node_entry *ne) ne->in_limbo = 0; device_remove_file(&ne->device, &dev_attr_ne_in_limbo); - class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_driver_resume); + class_for_each_device(&nodemgr_ud_class, NULL, ne, + __nodemgr_driver_resume); HPSB_DEBUG("Node resumed: ID:BUS[" NODE_BUS_FMT "] GUID[%016Lx]", NODE_BUS_ARGS(ne->host, ne->nodeid), (unsigned long long)ne->guid); } @@ -1498,7 +1500,8 @@ static int __nodemgr_update_pdrv(struct device *dev, void *data) static void nodemgr_update_pdrv(struct node_entry *ne) { - class_for_each_device(&nodemgr_ud_class, ne, __nodemgr_update_pdrv); + class_for_each_device(&nodemgr_ud_class, NULL, ne, + __nodemgr_update_pdrv); } @@ -1591,7 +1594,8 @@ static void nodemgr_node_probe(struct host_info *hi, int generation) * while probes are time-consuming. (Well, those probes need some * improvement...) */ - class_for_each_device(&nodemgr_ne_class, ¶m, __nodemgr_node_probe); + class_for_each_device(&nodemgr_ne_class, NULL, ¶m, + __nodemgr_node_probe); /* If we had a bus reset while we were scanning the bus, it is * possible that we did not probe all nodes. In that case, we @@ -1826,7 +1830,7 @@ int nodemgr_for_each_host(void *data, int (*cb)(struct hpsb_host *, void *)) hip.cb = cb; hip.data = data; - error = class_for_each_device(&hpsb_host_class, &hip, + error = class_for_each_device(&hpsb_host_class, NULL, &hip, __nodemgr_for_each_host); return error; diff --git a/drivers/power/apm_power.c b/drivers/power/apm_power.c index a4892275659..936bae560fa 100644 --- a/drivers/power/apm_power.c +++ b/drivers/power/apm_power.c @@ -78,7 +78,7 @@ static void find_main_battery(void) main_battery = NULL; bp.main = main_battery; - error = class_for_each_device(power_supply_class, &bp, + error = class_for_each_device(power_supply_class, NULL, &bp, __find_main_battery); if (error) { main_battery = bp.main; diff --git a/drivers/power/power_supply_core.c b/drivers/power/power_supply_core.c index af1633eb3b7..cb1ccb47292 100644 --- a/drivers/power/power_supply_core.c +++ b/drivers/power/power_supply_core.c @@ -41,7 +41,7 @@ static void power_supply_changed_work(struct work_struct *work) dev_dbg(psy->dev, "%s\n", __func__); - class_for_each_device(power_supply_class, psy, + class_for_each_device(power_supply_class, NULL, psy, __power_supply_changed_work); power_supply_update_leds(psy); @@ -79,7 +79,7 @@ int power_supply_am_i_supplied(struct power_supply *psy) { int error; - error = class_for_each_device(power_supply_class, psy, + error = class_for_each_device(power_supply_class, NULL, psy, __power_supply_am_i_supplied); dev_dbg(psy->dev, "%s %d\n", __func__, error); diff --git a/include/linux/device.h b/include/linux/device.h index de178712e02..6d5b351b29c 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -210,7 +210,8 @@ extern struct kobject *sysfs_dev_block_kobj; extern struct kobject *sysfs_dev_char_kobj; extern int __must_check class_register(struct class *class); extern void class_unregister(struct class *class); -extern int class_for_each_device(struct class *class, void *data, +extern int class_for_each_device(struct class *class, struct device *start, + void *data, int (*fn)(struct device *dev, void *data)); extern struct device *class_find_device(struct class *class, void *data, int (*match)(struct device *, void *)); -- cgit v1.2.3-70-g09d2 From 695794ae0c5bdd9bd06e35b118801e2e9be04f9e Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Thu, 22 May 2008 17:21:08 -0400 Subject: Driver Core: add ability for class_find_device to start in middle of list This mirrors the functionality that driver_find_device has as well. We add a start variable, and all callers of the function are fixed up at the same time. The block layer will be using this new functionality in a follow-on patch. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 22 +++++++++++++--------- drivers/base/core.c | 2 +- drivers/ieee1394/nodemgr.c | 9 ++++++--- drivers/rtc/interface.c | 2 +- drivers/scsi/hosts.c | 3 ++- drivers/scsi/scsi_transport_iscsi.c | 4 ++-- drivers/spi/spi.c | 2 +- include/linux/device.h | 3 ++- 8 files changed, 28 insertions(+), 19 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/class.c b/drivers/base/class.c index 2eb7048003a..3918d0e432d 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -302,6 +302,7 @@ EXPORT_SYMBOL_GPL(class_for_each_device); /** * class_find_device - device iterator for locating a particular device * @class: the class we're iterating + * @start: Device to begin with * @data: data for the match function * @match: function to check device * @@ -319,8 +320,9 @@ EXPORT_SYMBOL_GPL(class_for_each_device); * re-acquired in @match, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ -struct device *class_find_device(struct class *class, void *data, - int (*match)(struct device *, void *)) +struct device *class_find_device(struct class *class, struct device *start, + void *data, + int (*match)(struct device *, void *)) { struct device *dev; int found = 0; @@ -330,15 +332,17 @@ struct device *class_find_device(struct class *class, void *data, down(&class->sem); list_for_each_entry(dev, &class->devices, node) { + if (start) { + if (start == dev) + start = NULL; + continue; + } dev = get_device(dev); - if (dev) { - if (match(dev, data)) { - found = 1; - break; - } else - put_device(dev); - } else + if (match(dev, data)) { + found = 1; break; + } else + put_device(dev); } up(&class->sem); diff --git a/drivers/base/core.c b/drivers/base/core.c index 9ae28aa709d..9f05de6f80b 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1289,7 +1289,7 @@ void device_destroy(struct class *class, dev_t devt) { struct device *dev; - dev = class_find_device(class, &devt, __match_devt); + dev = class_find_device(class, NULL, &devt, __match_devt); if (dev) { put_device(dev); device_unregister(dev); diff --git a/drivers/ieee1394/nodemgr.c b/drivers/ieee1394/nodemgr.c index 47c0d85e0f3..994a21e5a0a 100644 --- a/drivers/ieee1394/nodemgr.c +++ b/drivers/ieee1394/nodemgr.c @@ -754,7 +754,8 @@ static void nodemgr_remove_uds(struct node_entry *ne) */ mutex_lock(&nodemgr_serialize_remove_uds); for (;;) { - dev = class_find_device(&nodemgr_ud_class, ne, __match_ne); + dev = class_find_device(&nodemgr_ud_class, NULL, ne, + __match_ne); if (!dev) break; ud = container_of(dev, struct unit_directory, unit_dev); @@ -901,7 +902,8 @@ static struct node_entry *find_entry_by_guid(u64 guid) struct device *dev; struct node_entry *ne; - dev = class_find_device(&nodemgr_ne_class, &guid, __match_ne_guid); + dev = class_find_device(&nodemgr_ne_class, NULL, &guid, + __match_ne_guid); if (!dev) return NULL; ne = container_of(dev, struct node_entry, node_dev); @@ -940,7 +942,8 @@ static struct node_entry *find_entry_by_nodeid(struct hpsb_host *host, param.host = host; param.nodeid = nodeid; - dev = class_find_device(&nodemgr_ne_class, ¶m, __match_ne_nodeid); + dev = class_find_device(&nodemgr_ne_class, NULL, ¶m, + __match_ne_nodeid); if (!dev) return NULL; ne = container_of(dev, struct node_entry, node_dev); diff --git a/drivers/rtc/interface.c b/drivers/rtc/interface.c index 58b7336640f..d397fa5f3a9 100644 --- a/drivers/rtc/interface.c +++ b/drivers/rtc/interface.c @@ -345,7 +345,7 @@ struct rtc_device *rtc_class_open(char *name) struct device *dev; struct rtc_device *rtc = NULL; - dev = class_find_device(rtc_class, name, __rtc_match); + dev = class_find_device(rtc_class, NULL, name, __rtc_match); if (dev) rtc = to_rtc_device(dev); diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 35cd892dce0..78dad28b70d 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -466,7 +466,8 @@ struct Scsi_Host *scsi_host_lookup(unsigned short hostnum) struct device *cdev; struct Scsi_Host *shost = ERR_PTR(-ENXIO); - cdev = class_find_device(&shost_class, &hostnum, __scsi_host_match); + cdev = class_find_device(&shost_class, NULL, &hostnum, + __scsi_host_match); if (cdev) { shost = scsi_host_get(class_to_shost(cdev)); put_device(cdev); diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index 3af7cbcc5c5..06748f318cd 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -170,7 +170,7 @@ iscsi_create_endpoint(int dd_size) int err; for (id = 1; id < ISCSI_MAX_EPID; id++) { - dev = class_find_device(&iscsi_endpoint_class, &id, + dev = class_find_device(&iscsi_endpoint_class, NULL, &id, iscsi_match_epid); if (!dev) break; @@ -222,7 +222,7 @@ struct iscsi_endpoint *iscsi_lookup_endpoint(u64 handle) struct iscsi_endpoint *ep; struct device *dev; - dev = class_find_device(&iscsi_endpoint_class, &handle, + dev = class_find_device(&iscsi_endpoint_class, NULL, &handle, iscsi_match_epid); if (!dev) return NULL; diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index 1ad12afc6ba..1771b2456bf 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -502,7 +502,7 @@ struct spi_master *spi_busnum_to_master(u16 bus_num) struct device *dev; struct spi_master *master = NULL; - dev = class_find_device(&spi_master_class, &bus_num, + dev = class_find_device(&spi_master_class, NULL, &bus_num, __spi_master_match); if (dev) master = container_of(dev, struct spi_master, dev); diff --git a/include/linux/device.h b/include/linux/device.h index 6d5b351b29c..c1f72984875 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -213,7 +213,8 @@ extern void class_unregister(struct class *class); extern int class_for_each_device(struct class *class, struct device *start, void *data, int (*fn)(struct device *dev, void *data)); -extern struct device *class_find_device(struct class *class, void *data, +extern struct device *class_find_device(struct class *class, + struct device *start, void *data, int (*match)(struct device *, void *)); struct class_attribute { -- cgit v1.2.3-70-g09d2 From 7c71448b8aa80123fc521563d5f7c63a099d97ab Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Tue, 22 Jan 2008 18:17:41 -0500 Subject: class: move driver core specific parts to a private structure This moves the portions of struct class that are dynamic (kobject and lock and lists) out of the main structure and into a dynamic, private, structure. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 27 ++++++++++++++++ drivers/base/class.c | 87 ++++++++++++++++++++++++++++---------------------- drivers/base/core.c | 57 +++++++++++++++++---------------- include/linux/device.h | 7 ++-- 4 files changed, 107 insertions(+), 71 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index 2c9ae43e221..0ec372a6776 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -36,6 +36,33 @@ struct driver_private { }; #define to_driver(obj) container_of(obj, struct driver_private, kobj) + +/** + * struct class_private - structure to hold the private to the driver core portions of the class structure. + * + * @subsys - the struct kset that defines this class. This is the main kobject + * @children - list of class_devices associated with this class + * @devices - list of devices associated with this class + * @interfaces - list of class_interfaces associated with this class + * @class_dirs - + * @sem - semaphore to protect the children, devices, and interfaces lists. + * @class - pointer back to the struct class that this structure is associated + * with. + * + * This structure is the one that is the actual kobject allowing struct + * class to be statically allocated safely. Nothing outside of the driver + * core should ever touch these fields. + */ +struct class_private { + struct kset subsys; + struct list_head devices; + struct list_head interfaces; + struct kset class_dirs; + struct semaphore sem; + struct class *class; +}; +#define to_class(obj) container_of(obj, struct class_private, subsys.kobj) + /* initialisation functions */ extern int devices_init(void); extern int buses_init(void); diff --git a/drivers/base/class.c b/drivers/base/class.c index 3918d0e432d..06f09c929a9 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -21,17 +21,16 @@ #include "base.h" #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) -#define to_class(obj) container_of(obj, struct class, subsys.kobj) static ssize_t class_attr_show(struct kobject *kobj, struct attribute *attr, char *buf) { struct class_attribute *class_attr = to_class_attr(attr); - struct class *dc = to_class(kobj); + struct class_private *cp = to_class(kobj); ssize_t ret = -EIO; if (class_attr->show) - ret = class_attr->show(dc, buf); + ret = class_attr->show(cp->class, buf); return ret; } @@ -39,17 +38,18 @@ static ssize_t class_attr_store(struct kobject *kobj, struct attribute *attr, const char *buf, size_t count) { struct class_attribute *class_attr = to_class_attr(attr); - struct class *dc = to_class(kobj); + struct class_private *cp = to_class(kobj); ssize_t ret = -EIO; if (class_attr->store) - ret = class_attr->store(dc, buf, count); + ret = class_attr->store(cp->class, buf, count); return ret; } static void class_release(struct kobject *kobj) { - struct class *class = to_class(kobj); + struct class_private *cp = to_class(kobj); + struct class *class = cp->class; pr_debug("class '%s': release.\n", class->name); @@ -78,7 +78,7 @@ int class_create_file(struct class *cls, const struct class_attribute *attr) { int error; if (cls) - error = sysfs_create_file(&cls->subsys.kobj, &attr->attr); + error = sysfs_create_file(&cls->p->subsys.kobj, &attr->attr); else error = -EINVAL; return error; @@ -87,21 +87,20 @@ int class_create_file(struct class *cls, const struct class_attribute *attr) void class_remove_file(struct class *cls, const struct class_attribute *attr) { if (cls) - sysfs_remove_file(&cls->subsys.kobj, &attr->attr); + sysfs_remove_file(&cls->p->subsys.kobj, &attr->attr); } static struct class *class_get(struct class *cls) { if (cls) - return container_of(kset_get(&cls->subsys), - struct class, subsys); - return NULL; + kset_get(&cls->p->subsys); + return cls; } static void class_put(struct class *cls) { if (cls) - kset_put(&cls->subsys); + kset_put(&cls->p->subsys); } static int add_class_attrs(struct class *cls) @@ -136,17 +135,23 @@ static void remove_class_attrs(struct class *cls) int class_register(struct class *cls) { + struct class_private *cp; int error; pr_debug("device class '%s': registering\n", cls->name); - INIT_LIST_HEAD(&cls->devices); - INIT_LIST_HEAD(&cls->interfaces); - kset_init(&cls->class_dirs); - init_MUTEX(&cls->sem); - error = kobject_set_name(&cls->subsys.kobj, "%s", cls->name); - if (error) + cp = kzalloc(sizeof(*cp), GFP_KERNEL); + if (!cp) + return -ENOMEM; + INIT_LIST_HEAD(&cp->devices); + INIT_LIST_HEAD(&cp->interfaces); + kset_init(&cp->class_dirs); + init_MUTEX(&cp->sem); + error = kobject_set_name(&cp->subsys.kobj, "%s", cls->name); + if (error) { + kfree(cp); return error; + } /* set the default /sys/dev directory for devices of this class */ if (!cls->dev_kobj) @@ -155,17 +160,21 @@ int class_register(struct class *cls) #if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK) /* let the block class directory show up in the root of sysfs */ if (cls != &block_class) - cls->subsys.kobj.kset = class_kset; + cp->subsys.kobj.kset = class_kset; #else - cls->subsys.kobj.kset = class_kset; + cp->subsys.kobj.kset = class_kset; #endif - cls->subsys.kobj.ktype = &class_ktype; + cp->subsys.kobj.ktype = &class_ktype; + cp->class = cls; + cls->p = cp; - error = kset_register(&cls->subsys); - if (!error) { - error = add_class_attrs(class_get(cls)); - class_put(cls); + error = kset_register(&cp->subsys); + if (error) { + kfree(cp); + return error; } + error = add_class_attrs(class_get(cls)); + class_put(cls); return error; } @@ -173,7 +182,7 @@ void class_unregister(struct class *cls) { pr_debug("device class '%s': unregistering\n", cls->name); remove_class_attrs(cls); - kset_unregister(&cls->subsys); + kset_unregister(&cls->p->subsys); } static void class_create_release(struct class *cls) @@ -280,8 +289,8 @@ int class_for_each_device(struct class *class, struct device *start, if (!class) return -EINVAL; - down(&class->sem); - list_for_each_entry(dev, &class->devices, node) { + down(&class->p->sem); + list_for_each_entry(dev, &class->p->devices, node) { if (start) { if (start == dev) start = NULL; @@ -293,7 +302,7 @@ int class_for_each_device(struct class *class, struct device *start, if (error) break; } - up(&class->sem); + up(&class->p->sem); return error; } @@ -330,8 +339,8 @@ struct device *class_find_device(struct class *class, struct device *start, if (!class) return NULL; - down(&class->sem); - list_for_each_entry(dev, &class->devices, node) { + down(&class->p->sem); + list_for_each_entry(dev, &class->p->devices, node) { if (start) { if (start == dev) start = NULL; @@ -344,7 +353,7 @@ struct device *class_find_device(struct class *class, struct device *start, } else put_device(dev); } - up(&class->sem); + up(&class->p->sem); return found ? dev : NULL; } @@ -362,13 +371,13 @@ int class_interface_register(struct class_interface *class_intf) if (!parent) return -EINVAL; - down(&parent->sem); - list_add_tail(&class_intf->node, &parent->interfaces); + down(&parent->p->sem); + list_add_tail(&class_intf->node, &parent->p->interfaces); if (class_intf->add_dev) { - list_for_each_entry(dev, &parent->devices, node) + list_for_each_entry(dev, &parent->p->devices, node) class_intf->add_dev(dev, class_intf); } - up(&parent->sem); + up(&parent->p->sem); return 0; } @@ -381,13 +390,13 @@ void class_interface_unregister(struct class_interface *class_intf) if (!parent) return; - down(&parent->sem); + down(&parent->p->sem); list_del_init(&class_intf->node); if (class_intf->remove_dev) { - list_for_each_entry(dev, &parent->devices, node) + list_for_each_entry(dev, &parent->p->devices, node) class_intf->remove_dev(dev, class_intf); } - up(&parent->sem); + up(&parent->p->sem); class_put(parent); } diff --git a/drivers/base/core.c b/drivers/base/core.c index 9f05de6f80b..64c150b5a88 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -551,7 +551,7 @@ static struct kobject *get_device_parent(struct device *dev, { /* class devices without a parent live in /sys/class// */ if (dev->class && (!parent || parent->class != dev->class)) - return &dev->class->subsys.kobj; + return &dev->class->p->subsys.kobj; /* all other devices keep their parent */ else if (parent) return &parent->kobj; @@ -597,13 +597,13 @@ static struct kobject *get_device_parent(struct device *dev, parent_kobj = &parent->kobj; /* find our class-directory at the parent and reference it */ - spin_lock(&dev->class->class_dirs.list_lock); - list_for_each_entry(k, &dev->class->class_dirs.list, entry) + spin_lock(&dev->class->p->class_dirs.list_lock); + list_for_each_entry(k, &dev->class->p->class_dirs.list, entry) if (k->parent == parent_kobj) { kobj = kobject_get(k); break; } - spin_unlock(&dev->class->class_dirs.list_lock); + spin_unlock(&dev->class->p->class_dirs.list_lock); if (kobj) return kobj; @@ -611,7 +611,7 @@ static struct kobject *get_device_parent(struct device *dev, k = kobject_create(); if (!k) return NULL; - k->kset = &dev->class->class_dirs; + k->kset = &dev->class->p->class_dirs; retval = kobject_add(k, parent_kobj, "%s", dev->class->name); if (retval < 0) { kobject_put(k); @@ -630,7 +630,7 @@ static void cleanup_glue_dir(struct device *dev, struct kobject *glue_dir) { /* see if we live in a "glue" directory */ if (!glue_dir || !dev->class || - glue_dir->kset != &dev->class->class_dirs) + glue_dir->kset != &dev->class->p->class_dirs) return; kobject_put(glue_dir); @@ -657,17 +657,17 @@ static int device_add_class_symlinks(struct device *dev) if (!dev->class) return 0; - error = sysfs_create_link(&dev->kobj, &dev->class->subsys.kobj, + error = sysfs_create_link(&dev->kobj, &dev->class->p->subsys.kobj, "subsystem"); if (error) goto out; #ifdef CONFIG_SYSFS_DEPRECATED /* stacked class devices need a symlink in the class directory */ - if (dev->kobj.parent != &dev->class->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->subsys.kobj && device_is_not_partition(dev)) { - error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj, - dev->bus_id); + error = sysfs_create_link(&dev->class->p->subsys.kobj, + &dev->kobj, dev->bus_id); if (error) goto out_subsys; } @@ -704,12 +704,12 @@ out_device: if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); out_busid: - if (dev->kobj.parent != &dev->class->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->subsys.kobj && device_is_not_partition(dev)) - sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); #else /* link in the class directory pointing to the device */ - error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj, + error = sysfs_create_link(&dev->class->p->subsys.kobj, &dev->kobj, dev->bus_id); if (error) goto out_subsys; @@ -723,7 +723,7 @@ out_busid: return 0; out_busid: - sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); #endif out_subsys: @@ -749,14 +749,14 @@ static void device_remove_class_symlinks(struct device *dev) sysfs_remove_link(&dev->kobj, "device"); } - if (dev->kobj.parent != &dev->class->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->subsys.kobj && device_is_not_partition(dev)) - sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); #else if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); - sysfs_remove_link(&dev->class->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); #endif sysfs_remove_link(&dev->kobj, "subsystem"); @@ -904,15 +904,16 @@ int device_add(struct device *dev) klist_add_tail(&dev->knode_parent, &parent->klist_children); if (dev->class) { - down(&dev->class->sem); + down(&dev->class->p->sem); /* tie the class to the device */ - list_add_tail(&dev->node, &dev->class->devices); + list_add_tail(&dev->node, &dev->class->p->devices); /* notify any interfaces that the device is here */ - list_for_each_entry(class_intf, &dev->class->interfaces, node) + list_for_each_entry(class_intf, &dev->class->p->interfaces, + node) if (class_intf->add_dev) class_intf->add_dev(dev, class_intf); - up(&dev->class->sem); + up(&dev->class->p->sem); } Done: put_device(dev); @@ -1013,14 +1014,15 @@ void device_del(struct device *dev) if (dev->class) { device_remove_class_symlinks(dev); - down(&dev->class->sem); + down(&dev->class->p->sem); /* notify any interfaces that the device is now gone */ - list_for_each_entry(class_intf, &dev->class->interfaces, node) + list_for_each_entry(class_intf, &dev->class->p->interfaces, + node) if (class_intf->remove_dev) class_intf->remove_dev(dev, class_intf); /* remove the device from the class list */ list_del_init(&dev->node); - up(&dev->class->sem); + up(&dev->class->p->sem); } device_remove_file(dev, &uevent_attr); device_remove_attrs(dev); @@ -1348,11 +1350,12 @@ int device_rename(struct device *dev, char *new_name) } #else if (dev->class) { - error = sysfs_create_link(&dev->class->subsys.kobj, &dev->kobj, - dev->bus_id); + error = sysfs_create_link(&dev->class->p->subsys.kobj, + &dev->kobj, dev->bus_id); if (error) goto out; - sysfs_remove_link(&dev->class->subsys.kobj, old_device_name); + sysfs_remove_link(&dev->class->p->subsys.kobj, + old_device_name); } #endif diff --git a/include/linux/device.h b/include/linux/device.h index c1f72984875..b0556082179 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -35,6 +35,7 @@ struct device; struct device_driver; struct driver_private; struct class; +struct class_private; struct bus_type; struct bus_type_private; @@ -186,11 +187,6 @@ struct class { const char *name; struct module *owner; - struct kset subsys; - struct list_head devices; - struct list_head interfaces; - struct kset class_dirs; - struct semaphore sem; /* locks children, devices, interfaces */ struct class_attribute *class_attrs; struct device_attribute *dev_attrs; struct kobject *dev_kobj; @@ -204,6 +200,7 @@ struct class { int (*resume)(struct device *dev); struct pm_ops *pm; + struct class_private *p; }; extern struct kobject *sysfs_dev_block_kobj; -- cgit v1.2.3-70-g09d2 From 97ae69fdbaa71a8f7dbc20bf10fb349d1759152f Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: rename "devices" to "class_devices" in internal class structure This renames the struct class "devices" field to be "class_devices" to make things easier when struct bus_type and struct class merge in the future. It also makes grepping for fields easier as well. Based on an idea from Kay. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 4 ++-- drivers/base/class.c | 10 +++++----- drivers/base/core.c | 2 +- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index 0ec372a6776..586c4ca7025 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -42,7 +42,7 @@ struct driver_private { * * @subsys - the struct kset that defines this class. This is the main kobject * @children - list of class_devices associated with this class - * @devices - list of devices associated with this class + * @class_devices - list of devices associated with this class * @interfaces - list of class_interfaces associated with this class * @class_dirs - * @sem - semaphore to protect the children, devices, and interfaces lists. @@ -55,7 +55,7 @@ struct driver_private { */ struct class_private { struct kset subsys; - struct list_head devices; + struct list_head class_devices; struct list_head interfaces; struct kset class_dirs; struct semaphore sem; diff --git a/drivers/base/class.c b/drivers/base/class.c index 06f09c929a9..9947560def6 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -143,7 +143,7 @@ int class_register(struct class *cls) cp = kzalloc(sizeof(*cp), GFP_KERNEL); if (!cp) return -ENOMEM; - INIT_LIST_HEAD(&cp->devices); + INIT_LIST_HEAD(&cp->class_devices); INIT_LIST_HEAD(&cp->interfaces); kset_init(&cp->class_dirs); init_MUTEX(&cp->sem); @@ -290,7 +290,7 @@ int class_for_each_device(struct class *class, struct device *start, if (!class) return -EINVAL; down(&class->p->sem); - list_for_each_entry(dev, &class->p->devices, node) { + list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) start = NULL; @@ -340,7 +340,7 @@ struct device *class_find_device(struct class *class, struct device *start, return NULL; down(&class->p->sem); - list_for_each_entry(dev, &class->p->devices, node) { + list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) start = NULL; @@ -374,7 +374,7 @@ int class_interface_register(struct class_interface *class_intf) down(&parent->p->sem); list_add_tail(&class_intf->node, &parent->p->interfaces); if (class_intf->add_dev) { - list_for_each_entry(dev, &parent->p->devices, node) + list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->add_dev(dev, class_intf); } up(&parent->p->sem); @@ -393,7 +393,7 @@ void class_interface_unregister(struct class_interface *class_intf) down(&parent->p->sem); list_del_init(&class_intf->node); if (class_intf->remove_dev) { - list_for_each_entry(dev, &parent->p->devices, node) + list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->remove_dev(dev, class_intf); } up(&parent->p->sem); diff --git a/drivers/base/core.c b/drivers/base/core.c index 64c150b5a88..52d1e71f2a4 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -906,7 +906,7 @@ int device_add(struct device *dev) if (dev->class) { down(&dev->class->p->sem); /* tie the class to the device */ - list_add_tail(&dev->node, &dev->class->p->devices); + list_add_tail(&dev->node, &dev->class->p->class_devices); /* notify any interfaces that the device is here */ list_for_each_entry(class_intf, &dev->class->p->interfaces, -- cgit v1.2.3-70-g09d2 From 184f1f779d5a2e62de4a0b34842ddf8546beca8f Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: rename "interfaces" to "class_interfaces" in internal class structure This renames the struct class "interfaces" field to be "class_interfaces" to make things easier when struct bus_type and struct class merge in the future. It also makes grepping for fields easier as well. Based on an idea from Kay. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 4 ++-- drivers/base/class.c | 4 ++-- drivers/base/core.c | 8 ++++---- 3 files changed, 8 insertions(+), 8 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index 586c4ca7025..670b95a1a51 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -43,7 +43,7 @@ struct driver_private { * @subsys - the struct kset that defines this class. This is the main kobject * @children - list of class_devices associated with this class * @class_devices - list of devices associated with this class - * @interfaces - list of class_interfaces associated with this class + * @class_interfaces - list of class_interfaces associated with this class * @class_dirs - * @sem - semaphore to protect the children, devices, and interfaces lists. * @class - pointer back to the struct class that this structure is associated @@ -56,7 +56,7 @@ struct driver_private { struct class_private { struct kset subsys; struct list_head class_devices; - struct list_head interfaces; + struct list_head class_interfaces; struct kset class_dirs; struct semaphore sem; struct class *class; diff --git a/drivers/base/class.c b/drivers/base/class.c index 9947560def6..48b518e66bf 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -144,7 +144,7 @@ int class_register(struct class *cls) if (!cp) return -ENOMEM; INIT_LIST_HEAD(&cp->class_devices); - INIT_LIST_HEAD(&cp->interfaces); + INIT_LIST_HEAD(&cp->class_interfaces); kset_init(&cp->class_dirs); init_MUTEX(&cp->sem); error = kobject_set_name(&cp->subsys.kobj, "%s", cls->name); @@ -372,7 +372,7 @@ int class_interface_register(struct class_interface *class_intf) return -EINVAL; down(&parent->p->sem); - list_add_tail(&class_intf->node, &parent->p->interfaces); + list_add_tail(&class_intf->node, &parent->p->class_interfaces); if (class_intf->add_dev) { list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->add_dev(dev, class_intf); diff --git a/drivers/base/core.c b/drivers/base/core.c index 52d1e71f2a4..76fdd41fea0 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -909,8 +909,8 @@ int device_add(struct device *dev) list_add_tail(&dev->node, &dev->class->p->class_devices); /* notify any interfaces that the device is here */ - list_for_each_entry(class_intf, &dev->class->p->interfaces, - node) + list_for_each_entry(class_intf, + &dev->class->p->class_interfaces, node) if (class_intf->add_dev) class_intf->add_dev(dev, class_intf); up(&dev->class->p->sem); @@ -1016,8 +1016,8 @@ void device_del(struct device *dev) down(&dev->class->p->sem); /* notify any interfaces that the device is now gone */ - list_for_each_entry(class_intf, &dev->class->p->interfaces, - node) + list_for_each_entry(class_intf, + &dev->class->p->class_interfaces, node) if (class_intf->remove_dev) class_intf->remove_dev(dev, class_intf); /* remove the device from the class list */ -- cgit v1.2.3-70-g09d2 From 1fbfee6c6dc0f4a4c587b6b163ee79643fc9aaa7 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: rename "subsys" to "class_subsys" in internal class structure This renames the struct class "subsys" field to be "class_subsys" to make things easier when struct bus_type and struct class merge in the future. It also makes grepping for fields easier as well. Based on an idea from Kay. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 7 ++++--- drivers/base/class.c | 23 ++++++++++++----------- drivers/base/core.c | 31 +++++++++++++++++-------------- 3 files changed, 33 insertions(+), 28 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index 670b95a1a51..a77d4bd2ea6 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -40,7 +40,7 @@ struct driver_private { /** * struct class_private - structure to hold the private to the driver core portions of the class structure. * - * @subsys - the struct kset that defines this class. This is the main kobject + * @class_subsys - the struct kset that defines this class. This is the main kobject * @children - list of class_devices associated with this class * @class_devices - list of devices associated with this class * @class_interfaces - list of class_interfaces associated with this class @@ -54,14 +54,15 @@ struct driver_private { * core should ever touch these fields. */ struct class_private { - struct kset subsys; + struct kset class_subsys; struct list_head class_devices; struct list_head class_interfaces; struct kset class_dirs; struct semaphore sem; struct class *class; }; -#define to_class(obj) container_of(obj, struct class_private, subsys.kobj) +#define to_class(obj) \ + container_of(obj, struct class_private, class_subsys.kobj) /* initialisation functions */ extern int devices_init(void); diff --git a/drivers/base/class.c b/drivers/base/class.c index 48b518e66bf..86778b86496 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -70,7 +70,7 @@ static struct kobj_type class_ktype = { .release = class_release, }; -/* Hotplug events for classes go to the class_obj subsys */ +/* Hotplug events for classes go to the class class_subsys */ static struct kset *class_kset; @@ -78,7 +78,8 @@ int class_create_file(struct class *cls, const struct class_attribute *attr) { int error; if (cls) - error = sysfs_create_file(&cls->p->subsys.kobj, &attr->attr); + error = sysfs_create_file(&cls->p->class_subsys.kobj, + &attr->attr); else error = -EINVAL; return error; @@ -87,20 +88,20 @@ int class_create_file(struct class *cls, const struct class_attribute *attr) void class_remove_file(struct class *cls, const struct class_attribute *attr) { if (cls) - sysfs_remove_file(&cls->p->subsys.kobj, &attr->attr); + sysfs_remove_file(&cls->p->class_subsys.kobj, &attr->attr); } static struct class *class_get(struct class *cls) { if (cls) - kset_get(&cls->p->subsys); + kset_get(&cls->p->class_subsys); return cls; } static void class_put(struct class *cls) { if (cls) - kset_put(&cls->p->subsys); + kset_put(&cls->p->class_subsys); } static int add_class_attrs(struct class *cls) @@ -147,7 +148,7 @@ int class_register(struct class *cls) INIT_LIST_HEAD(&cp->class_interfaces); kset_init(&cp->class_dirs); init_MUTEX(&cp->sem); - error = kobject_set_name(&cp->subsys.kobj, "%s", cls->name); + error = kobject_set_name(&cp->class_subsys.kobj, "%s", cls->name); if (error) { kfree(cp); return error; @@ -160,15 +161,15 @@ int class_register(struct class *cls) #if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK) /* let the block class directory show up in the root of sysfs */ if (cls != &block_class) - cp->subsys.kobj.kset = class_kset; + cp->class_subsys.kobj.kset = class_kset; #else - cp->subsys.kobj.kset = class_kset; + cp->class_subsys.kobj.kset = class_kset; #endif - cp->subsys.kobj.ktype = &class_ktype; + cp->class_subsys.kobj.ktype = &class_ktype; cp->class = cls; cls->p = cp; - error = kset_register(&cp->subsys); + error = kset_register(&cp->class_subsys); if (error) { kfree(cp); return error; @@ -182,7 +183,7 @@ void class_unregister(struct class *cls) { pr_debug("device class '%s': unregistering\n", cls->name); remove_class_attrs(cls); - kset_unregister(&cls->p->subsys); + kset_unregister(&cls->p->class_subsys); } static void class_create_release(struct class *cls) diff --git a/drivers/base/core.c b/drivers/base/core.c index 76fdd41fea0..6e1cff296d9 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -551,7 +551,7 @@ static struct kobject *get_device_parent(struct device *dev, { /* class devices without a parent live in /sys/class// */ if (dev->class && (!parent || parent->class != dev->class)) - return &dev->class->p->subsys.kobj; + return &dev->class->p->class_subsys.kobj; /* all other devices keep their parent */ else if (parent) return &parent->kobj; @@ -657,16 +657,17 @@ static int device_add_class_symlinks(struct device *dev) if (!dev->class) return 0; - error = sysfs_create_link(&dev->kobj, &dev->class->p->subsys.kobj, + error = sysfs_create_link(&dev->kobj, + &dev->class->p->class_subsys.kobj, "subsystem"); if (error) goto out; #ifdef CONFIG_SYSFS_DEPRECATED /* stacked class devices need a symlink in the class directory */ - if (dev->kobj.parent != &dev->class->p->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->class_subsys.kobj && device_is_not_partition(dev)) { - error = sysfs_create_link(&dev->class->p->subsys.kobj, + error = sysfs_create_link(&dev->class->p->class_subsys.kobj, &dev->kobj, dev->bus_id); if (error) goto out_subsys; @@ -704,13 +705,14 @@ out_device: if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); out_busid: - if (dev->kobj.parent != &dev->class->p->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->class_subsys.kobj && device_is_not_partition(dev)) - sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->class_subsys.kobj, + dev->bus_id); #else /* link in the class directory pointing to the device */ - error = sysfs_create_link(&dev->class->p->subsys.kobj, &dev->kobj, - dev->bus_id); + error = sysfs_create_link(&dev->class->p->class_subsys.kobj, + &dev->kobj, dev->bus_id); if (error) goto out_subsys; @@ -723,7 +725,7 @@ out_busid: return 0; out_busid: - sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev->bus_id); #endif out_subsys: @@ -749,14 +751,15 @@ static void device_remove_class_symlinks(struct device *dev) sysfs_remove_link(&dev->kobj, "device"); } - if (dev->kobj.parent != &dev->class->p->subsys.kobj && + if (dev->kobj.parent != &dev->class->p->class_subsys.kobj && device_is_not_partition(dev)) - sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->class_subsys.kobj, + dev->bus_id); #else if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); - sysfs_remove_link(&dev->class->p->subsys.kobj, dev->bus_id); + sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev->bus_id); #endif sysfs_remove_link(&dev->kobj, "subsystem"); @@ -1350,11 +1353,11 @@ int device_rename(struct device *dev, char *new_name) } #else if (dev->class) { - error = sysfs_create_link(&dev->class->p->subsys.kobj, + error = sysfs_create_link(&dev->class->p->class_subsys.kobj, &dev->kobj, dev->bus_id); if (error) goto out; - sysfs_remove_link(&dev->class->p->subsys.kobj, + sysfs_remove_link(&dev->class->p->class_subsys.kobj, old_device_name); } #endif -- cgit v1.2.3-70-g09d2 From d9a0157328507c5f563e16a583cd0a063854aebb Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: rename "sem" to "class_sem" in internal class structure This renames the struct class "sem" field to be "class_sem" to make things easier when struct bus_type and struct class merge in the future. It also makes grepping for fields easier as well. Based on an idea from Kay. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 4 ++-- drivers/base/class.c | 22 +++++++++++----------- drivers/base/core.c | 8 ++++---- 3 files changed, 17 insertions(+), 17 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index a77d4bd2ea6..4435732437a 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -45,7 +45,7 @@ struct driver_private { * @class_devices - list of devices associated with this class * @class_interfaces - list of class_interfaces associated with this class * @class_dirs - - * @sem - semaphore to protect the children, devices, and interfaces lists. + * @class_sem - semaphore to protect the children, devices, and interfaces lists. * @class - pointer back to the struct class that this structure is associated * with. * @@ -58,7 +58,7 @@ struct class_private { struct list_head class_devices; struct list_head class_interfaces; struct kset class_dirs; - struct semaphore sem; + struct semaphore class_sem; struct class *class; }; #define to_class(obj) \ diff --git a/drivers/base/class.c b/drivers/base/class.c index 86778b86496..d24d21114cc 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -147,7 +147,7 @@ int class_register(struct class *cls) INIT_LIST_HEAD(&cp->class_devices); INIT_LIST_HEAD(&cp->class_interfaces); kset_init(&cp->class_dirs); - init_MUTEX(&cp->sem); + init_MUTEX(&cp->class_sem); error = kobject_set_name(&cp->class_subsys.kobj, "%s", cls->name); if (error) { kfree(cp); @@ -278,7 +278,7 @@ char *make_class_name(const char *name, struct kobject *kobj) * We check the return of @fn each time. If it returns anything * other than 0, we break out and return that value. * - * Note, we hold class->sem in this function, so it can not be + * Note, we hold class->class_sem in this function, so it can not be * re-acquired in @fn, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ @@ -290,7 +290,7 @@ int class_for_each_device(struct class *class, struct device *start, if (!class) return -EINVAL; - down(&class->p->sem); + down(&class->p->class_sem); list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) @@ -303,7 +303,7 @@ int class_for_each_device(struct class *class, struct device *start, if (error) break; } - up(&class->p->sem); + up(&class->p->class_sem); return error; } @@ -326,7 +326,7 @@ EXPORT_SYMBOL_GPL(class_for_each_device); * * Note, you will need to drop the reference with put_device() after use. * - * We hold class->sem in this function, so it can not be + * We hold class->class_sem in this function, so it can not be * re-acquired in @match, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ @@ -340,7 +340,7 @@ struct device *class_find_device(struct class *class, struct device *start, if (!class) return NULL; - down(&class->p->sem); + down(&class->p->class_sem); list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) @@ -354,7 +354,7 @@ struct device *class_find_device(struct class *class, struct device *start, } else put_device(dev); } - up(&class->p->sem); + up(&class->p->class_sem); return found ? dev : NULL; } @@ -372,13 +372,13 @@ int class_interface_register(struct class_interface *class_intf) if (!parent) return -EINVAL; - down(&parent->p->sem); + down(&parent->p->class_sem); list_add_tail(&class_intf->node, &parent->p->class_interfaces); if (class_intf->add_dev) { list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->add_dev(dev, class_intf); } - up(&parent->p->sem); + up(&parent->p->class_sem); return 0; } @@ -391,13 +391,13 @@ void class_interface_unregister(struct class_interface *class_intf) if (!parent) return; - down(&parent->p->sem); + down(&parent->p->class_sem); list_del_init(&class_intf->node); if (class_intf->remove_dev) { list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->remove_dev(dev, class_intf); } - up(&parent->p->sem); + up(&parent->p->class_sem); class_put(parent); } diff --git a/drivers/base/core.c b/drivers/base/core.c index 6e1cff296d9..b90ae6f7be8 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -907,7 +907,7 @@ int device_add(struct device *dev) klist_add_tail(&dev->knode_parent, &parent->klist_children); if (dev->class) { - down(&dev->class->p->sem); + down(&dev->class->p->class_sem); /* tie the class to the device */ list_add_tail(&dev->node, &dev->class->p->class_devices); @@ -916,7 +916,7 @@ int device_add(struct device *dev) &dev->class->p->class_interfaces, node) if (class_intf->add_dev) class_intf->add_dev(dev, class_intf); - up(&dev->class->p->sem); + up(&dev->class->p->class_sem); } Done: put_device(dev); @@ -1017,7 +1017,7 @@ void device_del(struct device *dev) if (dev->class) { device_remove_class_symlinks(dev); - down(&dev->class->p->sem); + down(&dev->class->p->class_sem); /* notify any interfaces that the device is now gone */ list_for_each_entry(class_intf, &dev->class->p->class_interfaces, node) @@ -1025,7 +1025,7 @@ void device_del(struct device *dev) class_intf->remove_dev(dev, class_intf); /* remove the device from the class list */ list_del_init(&dev->node); - up(&dev->class->p->sem); + up(&dev->class->p->class_sem); } device_remove_file(dev, &uevent_attr); device_remove_attrs(dev); -- cgit v1.2.3-70-g09d2 From 1e41250863cd4de5ffa0678c405d001ca5b62796 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: fix docbook comments for class_private structure Removes a field that has been deleted, and adds a description fo the class_dirs field which was previously undocumented. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index 4435732437a..c035dc23266 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -41,10 +41,9 @@ struct driver_private { * struct class_private - structure to hold the private to the driver core portions of the class structure. * * @class_subsys - the struct kset that defines this class. This is the main kobject - * @children - list of class_devices associated with this class * @class_devices - list of devices associated with this class * @class_interfaces - list of class_interfaces associated with this class - * @class_dirs - + * @class_dirs - "glue" directory for virtual devices associated with this class * @class_sem - semaphore to protect the children, devices, and interfaces lists. * @class - pointer back to the struct class that this structure is associated * with. -- cgit v1.2.3-70-g09d2 From d2a3b9146e4f40c2e872d7567c996ef95083d802 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: add lockdep infrastructure This adds the infrastructure to properly handle lockdep issues when the internal class semaphore is changed to a mutex. Matthew wrote the original patch, and Greg fixed it up to work properly with the class_create() function. From: Matthew Wilcox Cc: Kay Sievers Cc: Dave Young Cc: Andrew Morton Cc: James Bottomley Cc: Peter Zijlstra Cc: Ingo Molnar Signed-off-by: Greg Kroah-Hartman --- drivers/base/class.c | 11 ++++++----- include/linux/device.h | 25 +++++++++++++++++++++++-- 2 files changed, 29 insertions(+), 7 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/class.c b/drivers/base/class.c index d24d21114cc..89000566690 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -134,7 +134,7 @@ static void remove_class_attrs(struct class *cls) } } -int class_register(struct class *cls) +int __class_register(struct class *cls, struct lock_class_key *key) { struct class_private *cp; int error; @@ -178,6 +178,7 @@ int class_register(struct class *cls) class_put(cls); return error; } +EXPORT_SYMBOL_GPL(__class_register); void class_unregister(struct class *cls) { @@ -203,7 +204,8 @@ static void class_create_release(struct class *cls) * Note, the pointer created here is to be destroyed when finished by * making a call to class_destroy(). */ -struct class *class_create(struct module *owner, const char *name) +struct class *__class_create(struct module *owner, const char *name, + struct lock_class_key *key) { struct class *cls; int retval; @@ -218,7 +220,7 @@ struct class *class_create(struct module *owner, const char *name) cls->owner = owner; cls->class_release = class_create_release; - retval = class_register(cls); + retval = __class_register(cls, key); if (retval) goto error; @@ -228,6 +230,7 @@ error: kfree(cls); return ERR_PTR(retval); } +EXPORT_SYMBOL_GPL(__class_create); /** * class_destroy - destroys a struct class structure @@ -412,9 +415,7 @@ int __init classes_init(void) EXPORT_SYMBOL_GPL(class_create_file); EXPORT_SYMBOL_GPL(class_remove_file); -EXPORT_SYMBOL_GPL(class_register); EXPORT_SYMBOL_GPL(class_unregister); -EXPORT_SYMBOL_GPL(class_create); EXPORT_SYMBOL_GPL(class_destroy); EXPORT_SYMBOL_GPL(class_interface_register); diff --git a/include/linux/device.h b/include/linux/device.h index b0556082179..0e1d24c2ed4 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -16,6 +16,7 @@ #include #include #include +#include #include #include #include @@ -205,8 +206,18 @@ struct class { extern struct kobject *sysfs_dev_block_kobj; extern struct kobject *sysfs_dev_char_kobj; -extern int __must_check class_register(struct class *class); +extern int __must_check __class_register(struct class *class, + struct lock_class_key *key); extern void class_unregister(struct class *class); + +/* This is a #define to keep the compiler from merging different + * instances of the __key variable */ +#define class_register(class) \ +({ \ + static struct lock_class_key __key; \ + __class_register(class, &__key); \ +}) + extern int class_for_each_device(struct class *class, struct device *start, void *data, int (*fn)(struct device *dev, void *data)); @@ -239,9 +250,19 @@ struct class_interface { extern int __must_check class_interface_register(struct class_interface *); extern void class_interface_unregister(struct class_interface *); -extern struct class *class_create(struct module *owner, const char *name); +extern struct class * __must_check __class_create(struct module *owner, + const char *name, + struct lock_class_key *key); extern void class_destroy(struct class *cls); +/* This is a #define to keep the compiler from merging different + * instances of the __key variable */ +#define class_create(owner, name) \ +({ \ + static struct lock_class_key __key; \ + __class_create(owner, name, &__key); \ +}) + /* * The type of device, "struct device" is embedded in. A class * or bus can contain devices of different types -- cgit v1.2.3-70-g09d2 From f75b1c60fc1e53c4004a79ea0be071aa3318cdcc Mon Sep 17 00:00:00 2001 From: Dave Young Date: Wed, 28 May 2008 09:28:39 -0700 Subject: class: change internal semaphore to a mutex Now that the lockdep infrastructure in the class core is in place, we should be able to properly change the internal class semaphore to be a mutex. David wrote the original patch, and Greg fixed it up to apply properly due to all of the recent changes in this area. From: Dave Young Cc: Matthew Wilcox Cc: Kay Sievers Cc: Andrew Morton Cc: James Bottomley Signed-off-by: Greg Kroah-Hartman --- drivers/base/base.h | 4 ++-- drivers/base/class.c | 23 ++++++++++++----------- drivers/base/core.c | 9 +++++---- 3 files changed, 19 insertions(+), 17 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/base.h b/drivers/base/base.h index c035dc23266..31dc0cd84af 100644 --- a/drivers/base/base.h +++ b/drivers/base/base.h @@ -44,7 +44,7 @@ struct driver_private { * @class_devices - list of devices associated with this class * @class_interfaces - list of class_interfaces associated with this class * @class_dirs - "glue" directory for virtual devices associated with this class - * @class_sem - semaphore to protect the children, devices, and interfaces lists. + * @class_mutex - mutex to protect the children, devices, and interfaces lists. * @class - pointer back to the struct class that this structure is associated * with. * @@ -57,7 +57,7 @@ struct class_private { struct list_head class_devices; struct list_head class_interfaces; struct kset class_dirs; - struct semaphore class_sem; + struct mutex class_mutex; struct class *class; }; #define to_class(obj) \ diff --git a/drivers/base/class.c b/drivers/base/class.c index 89000566690..839d27cecb3 100644 --- a/drivers/base/class.c +++ b/drivers/base/class.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "base.h" #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) @@ -147,7 +148,7 @@ int __class_register(struct class *cls, struct lock_class_key *key) INIT_LIST_HEAD(&cp->class_devices); INIT_LIST_HEAD(&cp->class_interfaces); kset_init(&cp->class_dirs); - init_MUTEX(&cp->class_sem); + __mutex_init(&cp->class_mutex, "struct class mutex", key); error = kobject_set_name(&cp->class_subsys.kobj, "%s", cls->name); if (error) { kfree(cp); @@ -281,7 +282,7 @@ char *make_class_name(const char *name, struct kobject *kobj) * We check the return of @fn each time. If it returns anything * other than 0, we break out and return that value. * - * Note, we hold class->class_sem in this function, so it can not be + * Note, we hold class->class_mutex in this function, so it can not be * re-acquired in @fn, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ @@ -293,7 +294,7 @@ int class_for_each_device(struct class *class, struct device *start, if (!class) return -EINVAL; - down(&class->p->class_sem); + mutex_lock(&class->p->class_mutex); list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) @@ -306,7 +307,7 @@ int class_for_each_device(struct class *class, struct device *start, if (error) break; } - up(&class->p->class_sem); + mutex_unlock(&class->p->class_mutex); return error; } @@ -329,7 +330,7 @@ EXPORT_SYMBOL_GPL(class_for_each_device); * * Note, you will need to drop the reference with put_device() after use. * - * We hold class->class_sem in this function, so it can not be + * We hold class->class_mutex in this function, so it can not be * re-acquired in @match, otherwise it will self-deadlocking. For * example, calls to add or remove class members would be verboten. */ @@ -343,7 +344,7 @@ struct device *class_find_device(struct class *class, struct device *start, if (!class) return NULL; - down(&class->p->class_sem); + mutex_lock(&class->p->class_mutex); list_for_each_entry(dev, &class->p->class_devices, node) { if (start) { if (start == dev) @@ -357,7 +358,7 @@ struct device *class_find_device(struct class *class, struct device *start, } else put_device(dev); } - up(&class->p->class_sem); + mutex_unlock(&class->p->class_mutex); return found ? dev : NULL; } @@ -375,13 +376,13 @@ int class_interface_register(struct class_interface *class_intf) if (!parent) return -EINVAL; - down(&parent->p->class_sem); + mutex_lock(&parent->p->class_mutex); list_add_tail(&class_intf->node, &parent->p->class_interfaces); if (class_intf->add_dev) { list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->add_dev(dev, class_intf); } - up(&parent->p->class_sem); + mutex_unlock(&parent->p->class_mutex); return 0; } @@ -394,13 +395,13 @@ void class_interface_unregister(struct class_interface *class_intf) if (!parent) return; - down(&parent->p->class_sem); + mutex_lock(&parent->p->class_mutex); list_del_init(&class_intf->node); if (class_intf->remove_dev) { list_for_each_entry(dev, &parent->p->class_devices, node) class_intf->remove_dev(dev, class_intf); } - up(&parent->p->class_sem); + mutex_unlock(&parent->p->class_mutex); class_put(parent); } diff --git a/drivers/base/core.c b/drivers/base/core.c index b90ae6f7be8..c05b1159023 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -21,6 +21,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -907,7 +908,7 @@ int device_add(struct device *dev) klist_add_tail(&dev->knode_parent, &parent->klist_children); if (dev->class) { - down(&dev->class->p->class_sem); + mutex_lock(&dev->class->p->class_mutex); /* tie the class to the device */ list_add_tail(&dev->node, &dev->class->p->class_devices); @@ -916,7 +917,7 @@ int device_add(struct device *dev) &dev->class->p->class_interfaces, node) if (class_intf->add_dev) class_intf->add_dev(dev, class_intf); - up(&dev->class->p->class_sem); + mutex_unlock(&dev->class->p->class_mutex); } Done: put_device(dev); @@ -1017,7 +1018,7 @@ void device_del(struct device *dev) if (dev->class) { device_remove_class_symlinks(dev); - down(&dev->class->p->class_sem); + mutex_lock(&dev->class->p->class_mutex); /* notify any interfaces that the device is now gone */ list_for_each_entry(class_intf, &dev->class->p->class_interfaces, node) @@ -1025,7 +1026,7 @@ void device_del(struct device *dev) class_intf->remove_dev(dev, class_intf); /* remove the device from the class list */ list_del_init(&dev->node); - up(&dev->class->p->class_sem); + mutex_unlock(&dev->class->p->class_mutex); } device_remove_file(dev, &uevent_attr); device_remove_attrs(dev); -- cgit v1.2.3-70-g09d2 From fc3a8828b139c24aade3f9d608775e36c248f8f5 Mon Sep 17 00:00:00 2001 From: Greg Kroah-Hartman Date: Fri, 2 May 2008 06:02:41 +0200 Subject: driver core: fix a lot of printk usages of bus_id We have the dev_printk() variants for this kind of thing, use them instead of directly trying to access the bus_id field of struct device. This is done in order to remove bus_id entirely. Cc: Kay Sievers Signed-off-by: Greg Kroah-Hartman --- arch/arm/common/dmabounce.c | 22 +++++++++------------- arch/arm/common/sa1111.c | 3 ++- arch/arm/kernel/ecard.c | 3 +-- arch/arm/mach-integrator/impd1.c | 3 +-- arch/powerpc/platforms/chrp/pci.c | 2 +- arch/x86/kernel/pci-dma.c | 6 ++---- arch/x86/kernel/pci-gart_64.c | 4 +--- drivers/acpi/fan.c | 10 +++++----- drivers/acpi/glue.c | 6 ++---- drivers/acpi/processor_core.c | 5 ++--- drivers/acpi/scan.c | 2 +- drivers/acpi/thermal.c | 4 ++-- drivers/acpi/video.c | 5 ++--- drivers/base/power/trace.c | 2 +- 14 files changed, 32 insertions(+), 45 deletions(-) (limited to 'drivers/base') diff --git a/arch/arm/common/dmabounce.c b/arch/arm/common/dmabounce.c index 2744673314b..dd294734260 100644 --- a/arch/arm/common/dmabounce.c +++ b/arch/arm/common/dmabounce.c @@ -554,9 +554,8 @@ dmabounce_register_dev(struct device *dev, unsigned long small_buffer_size, device_info = kmalloc(sizeof(struct dmabounce_device_info), GFP_ATOMIC); if (!device_info) { - printk(KERN_ERR - "Could not allocated dmabounce_device_info for %s", - dev->bus_id); + dev_err(dev, + "Could not allocated dmabounce_device_info\n"); return -ENOMEM; } @@ -594,8 +593,7 @@ dmabounce_register_dev(struct device *dev, unsigned long small_buffer_size, dev->archdata.dmabounce = device_info; - printk(KERN_INFO "dmabounce: registered device %s on %s bus\n", - dev->bus_id, dev->bus->name); + dev_info(dev, "dmabounce: registered device\n"); return 0; @@ -614,16 +612,15 @@ dmabounce_unregister_dev(struct device *dev) dev->archdata.dmabounce = NULL; if (!device_info) { - printk(KERN_WARNING - "%s: Never registered with dmabounce but attempting" \ - "to unregister!\n", dev->bus_id); + dev_warn(dev, + "Never registered with dmabounce but attempting" + "to unregister!\n"); return; } if (!list_empty(&device_info->safe_buffers)) { - printk(KERN_ERR - "%s: Removing from dmabounce with pending buffers!\n", - dev->bus_id); + dev_err(dev, + "Removing from dmabounce with pending buffers!\n"); BUG(); } @@ -639,8 +636,7 @@ dmabounce_unregister_dev(struct device *dev) kfree(device_info); - printk(KERN_INFO "dmabounce: device %s on %s bus unregistered\n", - dev->bus_id, dev->bus->name); + dev_info(dev, "dmabounce: device unregistered\n"); } diff --git a/arch/arm/common/sa1111.c b/arch/arm/common/sa1111.c index eb06d0b2cb7..2048ae0b155 100644 --- a/arch/arm/common/sa1111.c +++ b/arch/arm/common/sa1111.c @@ -593,7 +593,8 @@ sa1111_init_one_child(struct sa1111 *sachip, struct resource *parent, if (dev->dma_mask != 0xffffffffUL) { ret = dmabounce_register_dev(&dev->dev, 1024, 4096); if (ret) { - printk("SA1111: Failed to register %s with dmabounce", dev->dev.bus_id); + dev_err(&dev->dev, "SA1111: Failed to register" + " with dmabounce\n"); device_unregister(&dev->dev); } } diff --git a/arch/arm/kernel/ecard.c b/arch/arm/kernel/ecard.c index 8bfd299bfe7..2de425f62c2 100644 --- a/arch/arm/kernel/ecard.c +++ b/arch/arm/kernel/ecard.c @@ -853,8 +853,7 @@ static struct expansion_card *__init ecard_alloc_card(int type, int slot) for (i = 0; i < ECARD_NUM_RESOURCES; i++) { if (ec->resource[i].flags && request_resource(&iomem_resource, &ec->resource[i])) { - printk(KERN_ERR "%s: resource(s) not available\n", - ec->dev.bus_id); + dev_err(&ec->dev, "resource(s) not available\n"); ec->resource[i].end -= ec->resource[i].start; ec->resource[i].start = 0; ec->resource[i].flags = 0; diff --git a/arch/arm/mach-integrator/impd1.c b/arch/arm/mach-integrator/impd1.c index 62e653a3ea1..619d05e6cf7 100644 --- a/arch/arm/mach-integrator/impd1.c +++ b/arch/arm/mach-integrator/impd1.c @@ -407,8 +407,7 @@ static int impd1_probe(struct lm_device *dev) ret = amba_device_register(d, &dev->resource); if (ret) { - printk("unable to register device %s: %d\n", - d->dev.bus_id, ret); + dev_err(&d->dev, "unable to register device: %d\n"); kfree(d); } } diff --git a/arch/powerpc/platforms/chrp/pci.c b/arch/powerpc/platforms/chrp/pci.c index 609c46db4a1..768c262b936 100644 --- a/arch/powerpc/platforms/chrp/pci.c +++ b/arch/powerpc/platforms/chrp/pci.c @@ -367,7 +367,7 @@ static void chrp_pci_fixup_vt8231_ata(struct pci_dev *viaide) viaisa = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8231, NULL); if (!viaisa) return; - printk("Fixing VIA IDE, force legacy mode on '%s'\n", viaide->dev.bus_id); + dev_info(&viaide->dev, "Fixing VIA IDE, force legacy mode on\n"); pci_read_config_byte(viaide, PCI_CLASS_PROG, &progif); pci_write_config_byte(viaide, PCI_CLASS_PROG, progif & ~0x5); diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index a4213c00dff..cbecb05551b 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -314,8 +314,7 @@ int dma_supported(struct device *dev, u64 mask) { #ifdef CONFIG_PCI if (mask > 0xffffffff && forbid_dac > 0) { - printk(KERN_INFO "PCI: Disallowing DAC for device %s\n", - dev->bus_id); + dev_info(dev, "PCI: Disallowing DAC for device\n"); return 0; } #endif @@ -342,8 +341,7 @@ int dma_supported(struct device *dev, u64 mask) type. Normally this doesn't make any difference, but gives more gentle handling of IOMMU overflow. */ if (iommu_sac_force && (mask >= DMA_40BIT_MASK)) { - printk(KERN_INFO "%s: Force SAC with mask %Lx\n", - dev->bus_id, mask); + dev_info(dev, "Force SAC with mask %Lx\n", mask); return 0; } diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index be60961f869..df5f142657d 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -198,9 +198,7 @@ static void iommu_full(struct device *dev, size_t size, int dir) * out. Hopefully no network devices use single mappings that big. */ - printk(KERN_ERR - "PCI-DMA: Out of IOMMU space for %lu bytes at device %s\n", - size, dev->bus_id); + dev_err(dev, "PCI-DMA: Out of IOMMU space for %lu bytes\n", size); if (size > PAGE_SIZE*EMERGENCY_PAGES) { if (dir == PCI_DMA_FROMDEVICE || dir == PCI_DMA_BIDIRECTIONAL) diff --git a/drivers/acpi/fan.c b/drivers/acpi/fan.c index 55c17afbe66..2655bc1b4ee 100644 --- a/drivers/acpi/fan.c +++ b/drivers/acpi/fan.c @@ -263,22 +263,22 @@ static int acpi_fan_add(struct acpi_device *device) goto end; } - printk(KERN_INFO PREFIX - "%s is registered as cooling_device%d\n", - device->dev.bus_id, cdev->id); + dev_info(&device->dev, "registered as cooling_device%d\n", cdev->id); acpi_driver_data(device) = cdev; result = sysfs_create_link(&device->dev.kobj, &cdev->device.kobj, "thermal_cooling"); if (result) - printk(KERN_ERR PREFIX "Create sysfs link\n"); + dev_err(&device->dev, "Failed to create sysfs link " + "'thermal_cooling'\n"); result = sysfs_create_link(&cdev->device.kobj, &device->dev.kobj, "device"); if (result) - printk(KERN_ERR PREFIX "Create sysfs link\n"); + dev_err(&device->dev, "Failed to create sysfs link " + "'device'\n"); result = acpi_fan_add_fs(device); if (result) diff --git a/drivers/acpi/glue.c b/drivers/acpi/glue.c index 2f173e83f8a..084109507c9 100644 --- a/drivers/acpi/glue.c +++ b/drivers/acpi/glue.c @@ -146,8 +146,7 @@ static int acpi_bind_one(struct device *dev, acpi_handle handle) acpi_status status; if (dev->archdata.acpi_handle) { - printk(KERN_WARNING PREFIX - "Drivers changed 'acpi_handle' for %s\n", dev->bus_id); + dev_warn(dev, "Drivers changed 'acpi_handle'\n"); return -EINVAL; } get_device(dev); @@ -195,8 +194,7 @@ static int acpi_unbind_one(struct device *dev) /* acpi_bind_one increase refcnt by one */ put_device(dev); } else { - printk(KERN_ERR PREFIX - "Oops, 'acpi_handle' corrupt for %s\n", dev->bus_id); + dev_err(dev, "Oops, 'acpi_handle' corrupt\n"); } return 0; } diff --git a/drivers/acpi/processor_core.c b/drivers/acpi/processor_core.c index ec0f2d581ec..e36422a7122 100644 --- a/drivers/acpi/processor_core.c +++ b/drivers/acpi/processor_core.c @@ -714,9 +714,8 @@ static int __cpuinit acpi_processor_start(struct acpi_device *device) goto end; } - printk(KERN_INFO PREFIX - "%s is registered as cooling_device%d\n", - device->dev.bus_id, pr->cdev->id); + dev_info(&device->dev, "registered as cooling_device%d\n", + pr->cdev->id); result = sysfs_create_link(&device->dev.kobj, &pr->cdev->device.kobj, diff --git a/drivers/acpi/scan.c b/drivers/acpi/scan.c index f3132aa47a6..f6f52c1a2ab 100644 --- a/drivers/acpi/scan.c +++ b/drivers/acpi/scan.c @@ -471,7 +471,7 @@ static int acpi_device_register(struct acpi_device *device, device->dev.release = &acpi_device_release; result = device_add(&device->dev); if(result) { - printk(KERN_ERR PREFIX "Error adding device %s", device->dev.bus_id); + dev_err(&device->dev, "Error adding device\n"); goto end; } diff --git a/drivers/acpi/thermal.c b/drivers/acpi/thermal.c index 84c795fb9b1..30a34133793 100644 --- a/drivers/acpi/thermal.c +++ b/drivers/acpi/thermal.c @@ -1179,8 +1179,8 @@ static int acpi_thermal_register_thermal_zone(struct acpi_thermal *tz) tz->tz_enabled = 1; - printk(KERN_INFO PREFIX "%s is registered as thermal_zone%d\n", - tz->device->dev.bus_id, tz->thermal_zone->id); + dev_info(&tz->device->dev, "registered as thermal_zone%d\n", + tz->thermal_zone->id); return 0; } diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index 64c889331f3..37b9e16710d 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -762,9 +762,8 @@ static void acpi_video_device_find_cap(struct acpi_video_device *device) if (IS_ERR(device->cdev)) return; - printk(KERN_INFO PREFIX - "%s is registered as cooling_device%d\n", - device->dev->dev.bus_id, device->cdev->id); + dev_info(&device->dev->dev, "registered as cooling_device%d\n", + device->cdev->id); result = sysfs_create_link(&device->dev->dev.kobj, &device->cdev->device.kobj, "thermal_cooling"); diff --git a/drivers/base/power/trace.c b/drivers/base/power/trace.c index 9b1b20b59e0..2aa6e8fc4de 100644 --- a/drivers/base/power/trace.c +++ b/drivers/base/power/trace.c @@ -194,7 +194,7 @@ static int show_dev_hash(unsigned int value) struct device * dev = to_device(entry); unsigned int hash = hash_string(DEVSEED, dev->bus_id, DEVHASH); if (hash == value) { - printk(" hash matches device %s\n", dev->bus_id); + dev_info(dev, "hash matches\n"); match++; } entry = entry->prev; -- cgit v1.2.3-70-g09d2 From 838ea8e8dbfc8d907c7a55bbad507a44c1b81d57 Mon Sep 17 00:00:00 2001 From: Ben Dooks Date: Thu, 12 Jun 2008 19:00:34 +0100 Subject: sysdev: fix debugging statements in registration code. The systdev_class_register() and sysdev_register() functions have pr_debug() statements which are enabled when the user selects the driver core debug. Both of these routines do not produce the correct output, as they make assumptions about data which has not been initialised. In sysdev_class_register() the code uses the kobject_name(&cls->kset.kobj) at the start of the function, but this is not setup until later in the same call. Change this to use cls->name which is passed in from the caller. The sysdev_register() function tries to get the name of the sysdev by kobject_name(&sysdev->kobj), but that isn't setup until later in the same function so change this message to use the name of the sysdev's class and add another message once the name is initialised. Signed-off-by: Ben Dooks Signed-off-by: Greg Kroah-Hartman --- drivers/base/sys.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/sys.c b/drivers/base/sys.c index 358bb0be3c0..50690d9df24 100644 --- a/drivers/base/sys.c +++ b/drivers/base/sys.c @@ -130,8 +130,8 @@ static struct kset *system_kset; int sysdev_class_register(struct sysdev_class * cls) { - pr_debug("Registering sysdev class '%s'\n", - kobject_name(&cls->kset.kobj)); + pr_debug("Registering sysdev class '%s'\n", cls->name); + INIT_LIST_HEAD(&cls->drivers); memset(&cls->kset.kobj, 0x00, sizeof(struct kobject)); cls->kset.kobj.parent = &system_kset->kobj; @@ -241,7 +241,8 @@ int sysdev_register(struct sys_device * sysdev) if (!cls) return -EINVAL; - pr_debug("Registering sys device '%s'\n", kobject_name(&sysdev->kobj)); + pr_debug("Registering sys device of class '%s'\n", + kobject_name(&cls->kset.kobj)); /* initialize the kobject to 0, in case it had previously been used */ memset(&sysdev->kobj, 0x00, sizeof(struct kobject)); @@ -257,6 +258,9 @@ int sysdev_register(struct sys_device * sysdev) if (!error) { struct sysdev_driver * drv; + pr_debug("Registering sys device '%s'\n", + kobject_name(&sysdev->kobj)); + mutex_lock(&sysdev_drivers_lock); /* Generic notification is implicit, because it's that * code that should have called us. @@ -269,6 +273,7 @@ int sysdev_register(struct sys_device * sysdev) } mutex_unlock(&sysdev_drivers_lock); } + kobject_uevent(&sysdev->kobj, KOBJ_ADD); return error; } -- cgit v1.2.3-70-g09d2 From d9b19199e4894089456aaad295023263b5225c1a Mon Sep 17 00:00:00 2001 From: Adrian Bunk Date: Tue, 10 Jun 2008 19:04:08 +0300 Subject: always enable FW_LOADER unless EMBEDDED=y James Bottomley recently discovered that we have {request,release}_firmware() dummies for the case of the actual functions not being available and has a fix for the bug that was actually causing build errors for built-in users with CONFIG_FW_LOADER=m. But now missing selects on FW_LOADER are no longer visible at compile-time at all and can become runtime problems. FW_LOADER is infrastructure with relatively small codesize we can safely enable for everyone, and only for people who really need small kernels (and can be expected to know what they are doing) it matters being able to disable it. This patch therefore always sets FW_LOADER=y and allows users only to disable it with EMBEDDED=y. As a bonus, we can then get rid of all "select FW_LOADER" plus the due to it required "depends on HOTPLUG" which removes some complexity from our Kconfig files. Signed-off-by: Adrian Bunk Signed-off-by: Greg Kroah-Hartman --- drivers/base/Kconfig | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/base') diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig index d47482fa1d2..6318f6b5736 100644 --- a/drivers/base/Kconfig +++ b/drivers/base/Kconfig @@ -27,8 +27,9 @@ config PREVENT_FIRMWARE_BUILD If unsure say Y here. config FW_LOADER - tristate "Userspace firmware loading support" + tristate "Userspace firmware loading support" if EMBEDDED depends on HOTPLUG + default y ---help--- This option is provided for the case where no in-kernel-tree modules require userspace firmware loading support, but a module built outside -- cgit v1.2.3-70-g09d2 From 36ce6dad6e3cb3f050ed41e0beac0070d2062b25 Mon Sep 17 00:00:00 2001 From: Cornelia Huck Date: Tue, 10 Jun 2008 11:09:08 +0200 Subject: driver core: Suppress sysfs warnings for device_rename(). driver core: Suppress sysfs warnings for device_rename(). Renaming network devices to an already existing name is not something we want sysfs to print a scary warning for, since the callers can deal with this correctly. So let's introduce sysfs_create_link_nowarn() which gets rid of the common warning. Signed-off-by: Cornelia Huck Signed-off-by: Greg Kroah-Hartman --- drivers/base/core.c | 9 +++++---- fs/sysfs/dir.c | 37 +++++++++++++++++++++++++++++++++++-- fs/sysfs/symlink.c | 41 +++++++++++++++++++++++++++++++++-------- fs/sysfs/sysfs.h | 1 + include/linux/sysfs.h | 10 ++++++++++ 5 files changed, 84 insertions(+), 14 deletions(-) (limited to 'drivers/base') diff --git a/drivers/base/core.c b/drivers/base/core.c index c05b1159023..7d5c63c81a5 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -1345,8 +1345,9 @@ int device_rename(struct device *dev, char *new_name) if (old_class_name) { new_class_name = make_class_name(dev->class->name, &dev->kobj); if (new_class_name) { - error = sysfs_create_link(&dev->parent->kobj, - &dev->kobj, new_class_name); + error = sysfs_create_link_nowarn(&dev->parent->kobj, + &dev->kobj, + new_class_name); if (error) goto out; sysfs_remove_link(&dev->parent->kobj, old_class_name); @@ -1354,8 +1355,8 @@ int device_rename(struct device *dev, char *new_name) } #else if (dev->class) { - error = sysfs_create_link(&dev->class->p->class_subsys.kobj, - &dev->kobj, dev->bus_id); + error = sysfs_create_link_nowarn(&dev->class->p->class_subsys.kobj, + &dev->kobj, dev->bus_id); if (error) goto out; sysfs_remove_link(&dev->class->p->class_subsys.kobj, diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 8c0e4b92574..c1a7efb310b 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -398,7 +398,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt, } /** - * sysfs_add_one - add sysfs_dirent to parent + * __sysfs_add_one - add sysfs_dirent to parent without warning * @acxt: addrm context to use * @sd: sysfs_dirent to be added * @@ -417,7 +417,7 @@ void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt, * 0 on success, -EEXIST if entry with the given name already * exists. */ -int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) +int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) { if (sysfs_find_dirent(acxt->parent_sd, sd->s_name)) return -EEXIST; @@ -434,6 +434,39 @@ int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) return 0; } +/** + * sysfs_add_one - add sysfs_dirent to parent + * @acxt: addrm context to use + * @sd: sysfs_dirent to be added + * + * Get @acxt->parent_sd and set sd->s_parent to it and increment + * nlink of parent inode if @sd is a directory and link into the + * children list of the parent. + * + * This function should be called between calls to + * sysfs_addrm_start() and sysfs_addrm_finish() and should be + * passed the same @acxt as passed to sysfs_addrm_start(). + * + * LOCKING: + * Determined by sysfs_addrm_start(). + * + * RETURNS: + * 0 on success, -EEXIST if entry with the given name already + * exists. + */ +int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd) +{ + int ret; + + ret = __sysfs_add_one(acxt, sd); + if (ret == -EEXIST) { + printk(KERN_WARNING "sysfs: duplicate filename '%s' " + "can not be created\n", sd->s_name); + WARN_ON(1); + } + return ret; +} + /** * sysfs_remove_one - remove sysfs_dirent from parent * @acxt: addrm context to use diff --git a/fs/sysfs/symlink.c b/fs/sysfs/symlink.c index 817f5966edc..a3ba217fbe7 100644 --- a/fs/sysfs/symlink.c +++ b/fs/sysfs/symlink.c @@ -19,13 +19,8 @@ #include "sysfs.h" -/** - * sysfs_create_link - create symlink between two objects. - * @kobj: object whose directory we're creating the link in. - * @target: object we're pointing to. - * @name: name of the symlink. - */ -int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char * name) +static int sysfs_do_create_link(struct kobject *kobj, struct kobject *target, + const char *name, int warn) { struct sysfs_dirent *parent_sd = NULL; struct sysfs_dirent *target_sd = NULL; @@ -65,7 +60,10 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char target_sd = NULL; /* reference is now owned by the symlink */ sysfs_addrm_start(&acxt, parent_sd); - error = sysfs_add_one(&acxt, sd); + if (warn) + error = sysfs_add_one(&acxt, sd); + else + error = __sysfs_add_one(&acxt, sd); sysfs_addrm_finish(&acxt); if (error) @@ -79,6 +77,33 @@ int sysfs_create_link(struct kobject * kobj, struct kobject * target, const char return error; } +/** + * sysfs_create_link - create symlink between two objects. + * @kobj: object whose directory we're creating the link in. + * @target: object we're pointing to. + * @name: name of the symlink. + */ +int sysfs_create_link(struct kobject *kobj, struct kobject *target, + const char *name) +{ + return sysfs_do_create_link(kobj, target, name, 1); +} + +/** + * sysfs_create_link_nowarn - create symlink between two objects. + * @kobj: object whose directory we're creating the link in. + * @target: object we're pointing to. + * @name: name of the symlink. + * + * This function does the same as sysf_create_link(), but it + * doesn't warn if the link already exists. + */ +int sysfs_create_link_nowarn(struct kobject *kobj, struct kobject *target, + const char *name) +{ + return sysfs_do_create_link(kobj, target, name, 0); +} + /** * sysfs_remove_link - remove symlink in object's directory. * @kobj: object we're acting for. diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index ce4e15f8aae..a5db496f71c 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -107,6 +107,7 @@ struct sysfs_dirent *sysfs_get_active_two(struct sysfs_dirent *sd); void sysfs_put_active_two(struct sysfs_dirent *sd); void sysfs_addrm_start(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *parent_sd); +int __sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd); int sysfs_add_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd); void sysfs_remove_one(struct sysfs_addrm_cxt *acxt, struct sysfs_dirent *sd); void sysfs_addrm_finish(struct sysfs_addrm_cxt *acxt); diff --git a/include/linux/sysfs.h b/include/linux/sysfs.h index 7858eac40aa..37fa24152bd 100644 --- a/include/linux/sysfs.h +++ b/include/linux/sysfs.h @@ -101,6 +101,9 @@ void sysfs_remove_bin_file(struct kobject *kobj, struct bin_attribute *attr); int __must_check sysfs_create_link(struct kobject *kobj, struct kobject *target, const char *name); +int __must_check sysfs_create_link_nowarn(struct kobject *kobj, + struct kobject *target, + const char *name); void sysfs_remove_link(struct kobject *kobj, const char *name); int __must_check sysfs_create_group(struct kobject *kobj, @@ -180,6 +183,13 @@ static inline int sysfs_create_link(struct kobject *kobj, return 0; } +static inline int sysfs_create_link_nowarn(struct kobject *kobj, + struct kobject *target, + const char *name) +{ + return 0; +} + static inline void sysfs_remove_link(struct kobject *kobj, const char *name) { } -- cgit v1.2.3-70-g09d2 From 4a0b2b4dbe1335b8b9886ba3dc85a145d5d938ed Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 1 Jul 2008 18:48:41 +0200 Subject: sysdev: Pass the attribute to the low level sysdev show/store function This allow to dynamically generate attributes and share show/store functions between attributes. Right now most attributes are generated by special macros and lots of duplicated code. With the attribute passed it's instead possible to attach some data to the attribute and then use that in shared low level functions to do different things. I need this for the dynamically generated bank attributes in the x86 machine check code, but it'll allow some further cleanups. I converted all users in tree to the new show/store prototype. It's a single huge patch to avoid unbisectable sections. Runtime tested: x86-32, x86-64 Compiled only: ia64, powerpc Not compile tested/only grep converted: sh, arm, avr32 Signed-off-by: Andi Kleen Signed-off-by: Greg Kroah-Hartman --- arch/arm/kernel/time.c | 4 ++- arch/avr32/kernel/cpu.c | 38 +++++++++++++++++--------- arch/ia64/kernel/err_inject.c | 22 ++++++++++----- arch/powerpc/kernel/sysfs.c | 15 ++++++++--- arch/powerpc/platforms/cell/cbe_thermal.c | 45 ++++++++++++++++++++----------- arch/powerpc/platforms/cell/spu_base.c | 3 ++- arch/s390/kernel/smp.c | 36 ++++++++++++++++--------- arch/s390/kernel/time.c | 35 ++++++++++++++++-------- arch/sh/drivers/dma/dma-sysfs.c | 15 ++++++++--- arch/sparc64/kernel/sysfs.c | 16 +++++++---- arch/x86/kernel/cpu/mcheck/mce_64.c | 14 +++++++--- arch/x86/kernel/cpu/mcheck/therm_throt.c | 1 + arch/x86/kernel/microcode.c | 10 ++++--- drivers/base/cpu.c | 10 ++++--- drivers/base/memory.c | 12 ++++++--- drivers/base/node.c | 15 +++++++---- drivers/base/sys.c | 4 +-- drivers/base/topology.c | 17 ++++++++---- drivers/cpuidle/sysfs.c | 10 ++++--- drivers/xen/balloon.c | 1 + include/linux/sysdev.h | 5 ++-- kernel/rtmutex-tester.c | 7 ++--- kernel/sched.c | 8 ++++-- kernel/time/clocksource.c | 8 ++++-- 24 files changed, 239 insertions(+), 112 deletions(-) (limited to 'drivers/base') diff --git a/arch/arm/kernel/time.c b/arch/arm/kernel/time.c index cc5145b28e7..368d171754c 100644 --- a/arch/arm/kernel/time.c +++ b/arch/arm/kernel/time.c @@ -130,7 +130,9 @@ static const struct leds_evt_name evt_names[] = { { "red", led_red_on, led_red_off }, }; -static ssize_t leds_store(struct sys_device *dev, const char *buf, size_t size) +static ssize_t leds_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t size) { int ret = -EINVAL, len = strcspn(buf, " "); diff --git a/arch/avr32/kernel/cpu.c b/arch/avr32/kernel/cpu.c index b8409caeb23..e84faffbbec 100644 --- a/arch/avr32/kernel/cpu.c +++ b/arch/avr32/kernel/cpu.c @@ -26,14 +26,16 @@ static DEFINE_PER_CPU(struct cpu, cpu_devices); * XXX: If/when a SMP-capable implementation of AVR32 will ever be * made, we must make sure that the code executes on the correct CPU. */ -static ssize_t show_pc0event(struct sys_device *dev, char *buf) +static ssize_t show_pc0event(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pccr; pccr = sysreg_read(PCCR); return sprintf(buf, "0x%lx\n", (pccr >> 12) & 0x3f); } -static ssize_t store_pc0event(struct sys_device *dev, const char *buf, +static ssize_t store_pc0event(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { unsigned long val; @@ -46,15 +48,17 @@ static ssize_t store_pc0event(struct sys_device *dev, const char *buf, sysreg_write(PCCR, val); return count; } -static ssize_t show_pc0count(struct sys_device *dev, char *buf) +static ssize_t show_pc0count(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pcnt0; pcnt0 = sysreg_read(PCNT0); return sprintf(buf, "%lu\n", pcnt0); } -static ssize_t store_pc0count(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t store_pc0count(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { unsigned long val; char *endp; @@ -67,14 +71,16 @@ static ssize_t store_pc0count(struct sys_device *dev, const char *buf, return count; } -static ssize_t show_pc1event(struct sys_device *dev, char *buf) +static ssize_t show_pc1event(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pccr; pccr = sysreg_read(PCCR); return sprintf(buf, "0x%lx\n", (pccr >> 18) & 0x3f); } -static ssize_t store_pc1event(struct sys_device *dev, const char *buf, +static ssize_t store_pc1event(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { unsigned long val; @@ -87,14 +93,16 @@ static ssize_t store_pc1event(struct sys_device *dev, const char *buf, sysreg_write(PCCR, val); return count; } -static ssize_t show_pc1count(struct sys_device *dev, char *buf) +static ssize_t show_pc1count(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pcnt1; pcnt1 = sysreg_read(PCNT1); return sprintf(buf, "%lu\n", pcnt1); } -static ssize_t store_pc1count(struct sys_device *dev, const char *buf, +static ssize_t store_pc1count(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { unsigned long val; @@ -108,14 +116,16 @@ static ssize_t store_pc1count(struct sys_device *dev, const char *buf, return count; } -static ssize_t show_pccycles(struct sys_device *dev, char *buf) +static ssize_t show_pccycles(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pccnt; pccnt = sysreg_read(PCCNT); return sprintf(buf, "%lu\n", pccnt); } -static ssize_t store_pccycles(struct sys_device *dev, const char *buf, +static ssize_t store_pccycles(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { unsigned long val; @@ -129,14 +139,16 @@ static ssize_t store_pccycles(struct sys_device *dev, const char *buf, return count; } -static ssize_t show_pcenable(struct sys_device *dev, char *buf) +static ssize_t show_pcenable(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned long pccr; pccr = sysreg_read(PCCR); return sprintf(buf, "%c\n", (pccr & 1)?'1':'0'); } -static ssize_t store_pcenable(struct sys_device *dev, const char *buf, +static ssize_t store_pcenable(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { unsigned long pccr, val; diff --git a/arch/ia64/kernel/err_inject.c b/arch/ia64/kernel/err_inject.c index b642648cc2a..c539c689493 100644 --- a/arch/ia64/kernel/err_inject.c +++ b/arch/ia64/kernel/err_inject.c @@ -55,7 +55,8 @@ static u64 resources[NR_CPUS]; #define show(name) \ static ssize_t \ -show_##name(struct sys_device *dev, char *buf) \ +show_##name(struct sys_device *dev, struct sysdev_attribute *attr, \ + char *buf) \ { \ u32 cpu=dev->id; \ return sprintf(buf, "%lx\n", name[cpu]); \ @@ -63,7 +64,8 @@ show_##name(struct sys_device *dev, char *buf) \ #define store(name) \ static ssize_t \ -store_##name(struct sys_device *dev, const char *buf, size_t size) \ +store_##name(struct sys_device *dev, struct sysdev_attribute *attr, \ + const char *buf, size_t size) \ { \ unsigned int cpu=dev->id; \ name[cpu] = simple_strtoull(buf, NULL, 16); \ @@ -76,7 +78,8 @@ show(call_start) * processor. The cpu number in driver is only used for storing data. */ static ssize_t -store_call_start(struct sys_device *dev, const char *buf, size_t size) +store_call_start(struct sys_device *dev, struct sysdev_attribute *attr, + const char *buf, size_t size) { unsigned int cpu=dev->id; unsigned long call_start = simple_strtoull(buf, NULL, 16); @@ -124,14 +127,16 @@ show(err_type_info) store(err_type_info) static ssize_t -show_virtual_to_phys(struct sys_device *dev, char *buf) +show_virtual_to_phys(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { unsigned int cpu=dev->id; return sprintf(buf, "%lx\n", phys_addr[cpu]); } static ssize_t -store_virtual_to_phys(struct sys_device *dev, const char *buf, size_t size) +store_virtual_to_phys(struct sys_device *dev, struct sysdev_attribute *attr, + const char *buf, size_t size) { unsigned int cpu=dev->id; u64 virt_addr=simple_strtoull(buf, NULL, 16); @@ -154,7 +159,8 @@ show(err_struct_info) store(err_struct_info) static ssize_t -show_err_data_buffer(struct sys_device *dev, char *buf) +show_err_data_buffer(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned int cpu=dev->id; @@ -165,7 +171,9 @@ show_err_data_buffer(struct sys_device *dev, char *buf) } static ssize_t -store_err_data_buffer(struct sys_device *dev, const char *buf, size_t size) +store_err_data_buffer(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t size) { unsigned int cpu=dev->id; int ret; diff --git a/arch/powerpc/kernel/sysfs.c b/arch/powerpc/kernel/sysfs.c index c8127f832df..aba0ba95f06 100644 --- a/arch/powerpc/kernel/sysfs.c +++ b/arch/powerpc/kernel/sysfs.c @@ -28,7 +28,9 @@ static DEFINE_PER_CPU(struct cpu, cpu_devices); /* Time in microseconds we delay before sleeping in the idle loop */ DEFINE_PER_CPU(unsigned long, smt_snooze_delay) = { 100 }; -static ssize_t store_smt_snooze_delay(struct sys_device *dev, const char *buf, +static ssize_t store_smt_snooze_delay(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { struct cpu *cpu = container_of(dev, struct cpu, sysdev); @@ -44,7 +46,9 @@ static ssize_t store_smt_snooze_delay(struct sys_device *dev, const char *buf, return count; } -static ssize_t show_smt_snooze_delay(struct sys_device *dev, char *buf) +static ssize_t show_smt_snooze_delay(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { struct cpu *cpu = container_of(dev, struct cpu, sysdev); @@ -152,14 +156,17 @@ static unsigned long write_##NAME(unsigned long val) \ mtspr(ADDRESS, val); \ return 0; \ } \ -static ssize_t show_##NAME(struct sys_device *dev, char *buf) \ +static ssize_t show_##NAME(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ { \ struct cpu *cpu = container_of(dev, struct cpu, sysdev); \ unsigned long val = run_on_cpu(cpu->sysdev.id, read_##NAME, 0); \ return sprintf(buf, "%lx\n", val); \ } \ static ssize_t __used \ - store_##NAME(struct sys_device *dev, const char *buf, size_t count) \ + store_##NAME(struct sys_device *dev, struct sysdev_attribute *attr, \ + const char *buf, size_t count) \ { \ struct cpu *cpu = container_of(dev, struct cpu, sysdev); \ unsigned long val; \ diff --git a/arch/powerpc/platforms/cell/cbe_thermal.c b/arch/powerpc/platforms/cell/cbe_thermal.c index 4852bf312d8..4d4c8c16912 100644 --- a/arch/powerpc/platforms/cell/cbe_thermal.c +++ b/arch/powerpc/platforms/cell/cbe_thermal.c @@ -97,7 +97,8 @@ static u8 spu_read_register_value(struct sys_device *sysdev, union spe_reg __iom return value.spe[spu->spe_id]; } -static ssize_t spu_show_temp(struct sys_device *sysdev, char *buf) +static ssize_t spu_show_temp(struct sys_device *sysdev, struct sysdev_attribute *attr, + char *buf) { u8 value; struct cbe_pmd_regs __iomem *pmd_regs; @@ -146,32 +147,38 @@ static ssize_t store_throttle(struct cbe_pmd_regs __iomem *pmd_regs, const char return size; } -static ssize_t spu_show_throttle_end(struct sys_device *sysdev, char *buf) +static ssize_t spu_show_throttle_end(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(get_pmd_regs(sysdev), buf, 0); } -static ssize_t spu_show_throttle_begin(struct sys_device *sysdev, char *buf) +static ssize_t spu_show_throttle_begin(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(get_pmd_regs(sysdev), buf, 8); } -static ssize_t spu_show_throttle_full_stop(struct sys_device *sysdev, char *buf) +static ssize_t spu_show_throttle_full_stop(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(get_pmd_regs(sysdev), buf, 16); } -static ssize_t spu_store_throttle_end(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t spu_store_throttle_end(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(get_pmd_regs(sysdev), buf, size, 0); } -static ssize_t spu_store_throttle_begin(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t spu_store_throttle_begin(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(get_pmd_regs(sysdev), buf, size, 8); } -static ssize_t spu_store_throttle_full_stop(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t spu_store_throttle_full_stop(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(get_pmd_regs(sysdev), buf, size, 16); } @@ -192,43 +199,51 @@ static ssize_t ppe_show_temp(struct sys_device *sysdev, char *buf, int pos) /* shows the temperature of the DTS on the PPE, * located near the linear thermal sensor */ -static ssize_t ppe_show_temp0(struct sys_device *sysdev, char *buf) +static ssize_t ppe_show_temp0(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return ppe_show_temp(sysdev, buf, 32); } /* shows the temperature of the second DTS on the PPE */ -static ssize_t ppe_show_temp1(struct sys_device *sysdev, char *buf) +static ssize_t ppe_show_temp1(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return ppe_show_temp(sysdev, buf, 0); } -static ssize_t ppe_show_throttle_end(struct sys_device *sysdev, char *buf) +static ssize_t ppe_show_throttle_end(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 32); } -static ssize_t ppe_show_throttle_begin(struct sys_device *sysdev, char *buf) +static ssize_t ppe_show_throttle_begin(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 40); } -static ssize_t ppe_show_throttle_full_stop(struct sys_device *sysdev, char *buf) +static ssize_t ppe_show_throttle_full_stop(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { return show_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, 48); } -static ssize_t ppe_store_throttle_end(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t ppe_store_throttle_end(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 32); } -static ssize_t ppe_store_throttle_begin(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t ppe_store_throttle_begin(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 40); } -static ssize_t ppe_store_throttle_full_stop(struct sys_device *sysdev, const char *buf, size_t size) +static ssize_t ppe_store_throttle_full_stop(struct sys_device *sysdev, + struct sysdev_attribute *attr, const char *buf, size_t size) { return store_throttle(cbe_get_cpu_pmd_regs(sysdev->id), buf, size, 48); } diff --git a/arch/powerpc/platforms/cell/spu_base.c b/arch/powerpc/platforms/cell/spu_base.c index 78f905bc6a4..a5bdb89a17c 100644 --- a/arch/powerpc/platforms/cell/spu_base.c +++ b/arch/powerpc/platforms/cell/spu_base.c @@ -703,7 +703,8 @@ static unsigned long long spu_acct_time(struct spu *spu, } -static ssize_t spu_stat_show(struct sys_device *sysdev, char *buf) +static ssize_t spu_stat_show(struct sys_device *sysdev, + struct sysdev_attribute *attr, char *buf) { struct spu *spu = container_of(sysdev, struct spu, sysdev); diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c index b6781030cfb..b795b3e24af 100644 --- a/arch/s390/kernel/smp.c +++ b/arch/s390/kernel/smp.c @@ -864,7 +864,8 @@ int setup_profiling_timer(unsigned int multiplier) } #ifdef CONFIG_HOTPLUG_CPU -static ssize_t cpu_configure_show(struct sys_device *dev, char *buf) +static ssize_t cpu_configure_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t count; @@ -874,8 +875,9 @@ static ssize_t cpu_configure_show(struct sys_device *dev, char *buf) return count; } -static ssize_t cpu_configure_store(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t cpu_configure_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int cpu = dev->id; int val, rc; @@ -922,7 +924,8 @@ out: static SYSDEV_ATTR(configure, 0644, cpu_configure_show, cpu_configure_store); #endif /* CONFIG_HOTPLUG_CPU */ -static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf) +static ssize_t cpu_polarization_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { int cpu = dev->id; ssize_t count; @@ -950,7 +953,8 @@ static ssize_t cpu_polarization_show(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(polarization, 0444, cpu_polarization_show, NULL); -static ssize_t show_cpu_address(struct sys_device *dev, char *buf) +static ssize_t show_cpu_address(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { return sprintf(buf, "%d\n", __cpu_logical_map[dev->id]); } @@ -970,7 +974,8 @@ static struct attribute_group cpu_common_attr_group = { .attrs = cpu_common_attrs, }; -static ssize_t show_capability(struct sys_device *dev, char *buf) +static ssize_t show_capability(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { unsigned int capability; int rc; @@ -982,7 +987,8 @@ static ssize_t show_capability(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(capability, 0444, show_capability, NULL); -static ssize_t show_idle_count(struct sys_device *dev, char *buf) +static ssize_t show_idle_count(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct s390_idle_data *idle; unsigned long long idle_count; @@ -995,7 +1001,8 @@ static ssize_t show_idle_count(struct sys_device *dev, char *buf) } static SYSDEV_ATTR(idle_count, 0444, show_idle_count, NULL); -static ssize_t show_idle_time(struct sys_device *dev, char *buf) +static ssize_t show_idle_time(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct s390_idle_data *idle; unsigned long long new_time; @@ -1112,7 +1119,9 @@ out: return rc; } -static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf, +static ssize_t __ref rescan_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int rc; @@ -1123,7 +1132,9 @@ static ssize_t __ref rescan_store(struct sys_device *dev, const char *buf, static SYSDEV_ATTR(rescan, 0200, NULL, rescan_store); #endif /* CONFIG_HOTPLUG_CPU */ -static ssize_t dispatching_show(struct sys_device *dev, char *buf) +static ssize_t dispatching_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { ssize_t count; @@ -1133,8 +1144,9 @@ static ssize_t dispatching_show(struct sys_device *dev, char *buf) return count; } -static ssize_t dispatching_store(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t dispatching_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t count) { int val, rc; char delim; diff --git a/arch/s390/kernel/time.c b/arch/s390/kernel/time.c index f2cede3947b..ab70d9bd926 100644 --- a/arch/s390/kernel/time.c +++ b/arch/s390/kernel/time.c @@ -1100,7 +1100,9 @@ static inline struct etr_aib *etr_aib_from_dev(struct sys_device *dev) return etr_port1_online ? &etr_port1 : NULL; } -static ssize_t etr_online_show(struct sys_device *dev, char *buf) +static ssize_t etr_online_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { unsigned int online; @@ -1109,7 +1111,8 @@ static ssize_t etr_online_show(struct sys_device *dev, char *buf) } static ssize_t etr_online_store(struct sys_device *dev, - const char *buf, size_t count) + struct sysdev_attribute *attr, + const char *buf, size_t count) { unsigned int value; @@ -1136,7 +1139,9 @@ static ssize_t etr_online_store(struct sys_device *dev, static SYSDEV_ATTR(online, 0600, etr_online_show, etr_online_store); -static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf) +static ssize_t etr_stepping_control_show(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { return sprintf(buf, "%i\n", (dev == &etr_port0_dev) ? etr_eacr.e0 : etr_eacr.e1); @@ -1144,7 +1149,8 @@ static ssize_t etr_stepping_control_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(stepping_control, 0400, etr_stepping_control_show, NULL); -static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf) +static ssize_t etr_mode_code_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { if (!etr_port0_online && !etr_port1_online) /* Status word is not uptodate if both ports are offline. */ @@ -1155,7 +1161,8 @@ static ssize_t etr_mode_code_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(state_code, 0400, etr_mode_code_show, NULL); -static ssize_t etr_untuned_show(struct sys_device *dev, char *buf) +static ssize_t etr_untuned_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1166,7 +1173,8 @@ static ssize_t etr_untuned_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(untuned, 0400, etr_untuned_show, NULL); -static ssize_t etr_network_id_show(struct sys_device *dev, char *buf) +static ssize_t etr_network_id_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1177,7 +1185,8 @@ static ssize_t etr_network_id_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(network, 0400, etr_network_id_show, NULL); -static ssize_t etr_id_show(struct sys_device *dev, char *buf) +static ssize_t etr_id_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1188,7 +1197,8 @@ static ssize_t etr_id_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(id, 0400, etr_id_show, NULL); -static ssize_t etr_port_number_show(struct sys_device *dev, char *buf) +static ssize_t etr_port_number_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1199,7 +1209,8 @@ static ssize_t etr_port_number_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(port, 0400, etr_port_number_show, NULL); -static ssize_t etr_coupled_show(struct sys_device *dev, char *buf) +static ssize_t etr_coupled_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1210,7 +1221,8 @@ static ssize_t etr_coupled_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(coupled, 0400, etr_coupled_show, NULL); -static ssize_t etr_local_time_show(struct sys_device *dev, char *buf) +static ssize_t etr_local_time_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); @@ -1221,7 +1233,8 @@ static ssize_t etr_local_time_show(struct sys_device *dev, char *buf) static SYSDEV_ATTR(local_time, 0400, etr_local_time_show, NULL); -static ssize_t etr_utc_offset_show(struct sys_device *dev, char *buf) +static ssize_t etr_utc_offset_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct etr_aib *aib = etr_aib_from_dev(dev); diff --git a/arch/sh/drivers/dma/dma-sysfs.c b/arch/sh/drivers/dma/dma-sysfs.c index 51b57c0d1a3..347ee11351e 100644 --- a/arch/sh/drivers/dma/dma-sysfs.c +++ b/arch/sh/drivers/dma/dma-sysfs.c @@ -23,7 +23,8 @@ static struct sysdev_class dma_sysclass = { }; EXPORT_SYMBOL(dma_sysclass); -static ssize_t dma_show_devices(struct sys_device *dev, char *buf) +static ssize_t dma_show_devices(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t len = 0; int i; @@ -57,13 +58,15 @@ static int __init dma_sysclass_init(void) } postcore_initcall(dma_sysclass_init); -static ssize_t dma_show_dev_id(struct sys_device *dev, char *buf) +static ssize_t dma_show_dev_id(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct dma_channel *channel = to_dma_channel(dev); return sprintf(buf, "%s\n", channel->dev_id); } static ssize_t dma_store_dev_id(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { struct dma_channel *channel = to_dma_channel(dev); @@ -74,6 +77,7 @@ static ssize_t dma_store_dev_id(struct sys_device *dev, static SYSDEV_ATTR(dev_id, S_IRUGO | S_IWUSR, dma_show_dev_id, dma_store_dev_id); static ssize_t dma_store_config(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { struct dma_channel *channel = to_dma_channel(dev); @@ -87,13 +91,15 @@ static ssize_t dma_store_config(struct sys_device *dev, static SYSDEV_ATTR(config, S_IWUSR, NULL, dma_store_config); -static ssize_t dma_show_mode(struct sys_device *dev, char *buf) +static ssize_t dma_show_mode(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct dma_channel *channel = to_dma_channel(dev); return sprintf(buf, "0x%08x\n", channel->mode); } static ssize_t dma_store_mode(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { struct dma_channel *channel = to_dma_channel(dev); @@ -104,7 +110,8 @@ static ssize_t dma_store_mode(struct sys_device *dev, static SYSDEV_ATTR(mode, S_IRUGO | S_IWUSR, dma_show_mode, dma_store_mode); #define dma_ro_attr(field, fmt) \ -static ssize_t dma_show_##field(struct sys_device *dev, char *buf) \ +static ssize_t dma_show_##field(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf)\ { \ struct dma_channel *channel = to_dma_channel(dev); \ return sprintf(buf, fmt, channel->field); \ diff --git a/arch/sparc64/kernel/sysfs.c b/arch/sparc64/kernel/sysfs.c index e885034a6b7..84e5ce14671 100644 --- a/arch/sparc64/kernel/sysfs.c +++ b/arch/sparc64/kernel/sysfs.c @@ -14,7 +14,8 @@ static DEFINE_PER_CPU(struct hv_mmu_statistics, mmu_stats) __attribute__((aligned(64))); #define SHOW_MMUSTAT_ULONG(NAME) \ -static ssize_t show_##NAME(struct sys_device *dev, char *buf) \ +static ssize_t show_##NAME(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ struct hv_mmu_statistics *p = &per_cpu(mmu_stats, dev->id); \ return sprintf(buf, "%lu\n", p->NAME); \ @@ -135,13 +136,16 @@ static unsigned long write_mmustat_enable(unsigned long val) return sun4v_mmustat_conf(ra, &orig_ra); } -static ssize_t show_mmustat_enable(struct sys_device *s, char *buf) +static ssize_t show_mmustat_enable(struct sys_device *s, + struct sysdev_attribute *attr, char *buf) { unsigned long val = run_on_cpu(s->id, read_mmustat_enable, 0); return sprintf(buf, "%lx\n", val); } -static ssize_t store_mmustat_enable(struct sys_device *s, const char *buf, size_t count) +static ssize_t store_mmustat_enable(struct sys_device *s, + struct sysdev_attribute *attr, const char *buf, + size_t count) { unsigned long val, err; int ret = sscanf(buf, "%ld", &val); @@ -179,14 +183,16 @@ static void unregister_mmu_stats(struct sys_device *s) #endif #define SHOW_CPUDATA_ULONG_NAME(NAME, MEMBER) \ -static ssize_t show_##NAME(struct sys_device *dev, char *buf) \ +static ssize_t show_##NAME(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ cpuinfo_sparc *c = &cpu_data(dev->id); \ return sprintf(buf, "%lu\n", c->MEMBER); \ } #define SHOW_CPUDATA_UINT_NAME(NAME, MEMBER) \ -static ssize_t show_##NAME(struct sys_device *dev, char *buf) \ +static ssize_t show_##NAME(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ cpuinfo_sparc *c = &cpu_data(dev->id); \ return sprintf(buf, "%u\n", c->MEMBER); \ diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index c4a7ec31394..e6a4d5f6764 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -762,10 +762,14 @@ DEFINE_PER_CPU(struct sys_device, device_mce); /* Why are there no generic functions for this? */ #define ACCESSOR(name, var, start) \ - static ssize_t show_ ## name(struct sys_device *s, char *buf) { \ + static ssize_t show_ ## name(struct sys_device *s, \ + struct sysdev_attribute *attr, \ + char *buf) { \ return sprintf(buf, "%lx\n", (unsigned long)var); \ } \ - static ssize_t set_ ## name(struct sys_device *s,const char *buf,size_t siz) { \ + static ssize_t set_ ## name(struct sys_device *s, \ + struct sysdev_attribute *attr, \ + const char *buf, size_t siz) { \ char *end; \ unsigned long new = simple_strtoul(buf, &end, 0); \ if (end == buf) return -EINVAL; \ @@ -786,14 +790,16 @@ ACCESSOR(bank3ctl,bank[3],mce_restart()) ACCESSOR(bank4ctl,bank[4],mce_restart()) ACCESSOR(bank5ctl,bank[5],mce_restart()) -static ssize_t show_trigger(struct sys_device *s, char *buf) +static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, + char *buf) { strcpy(buf, trigger); strcat(buf, "\n"); return strlen(trigger) + 1; } -static ssize_t set_trigger(struct sys_device *s,const char *buf,size_t siz) +static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, + const char *buf,size_t siz) { char *p; int len; diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index 1f4cc48c14c..d5ae2243f0b 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -35,6 +35,7 @@ atomic_t therm_throt_en = ATOMIC_INIT(0); #define define_therm_throt_sysdev_show_func(name) \ static ssize_t therm_throt_sysdev_show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ char *buf) \ { \ unsigned int cpu = dev->id; \ diff --git a/arch/x86/kernel/microcode.c b/arch/x86/kernel/microcode.c index 56b933119a0..fc4790638b6 100644 --- a/arch/x86/kernel/microcode.c +++ b/arch/x86/kernel/microcode.c @@ -644,7 +644,9 @@ static void microcode_fini_cpu(int cpu) mutex_unlock(µcode_mutex); } -static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz) +static ssize_t reload_store(struct sys_device *dev, + struct sysdev_attribute *attr, + const char *buf, size_t sz) { struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; char *end; @@ -674,14 +676,16 @@ static ssize_t reload_store(struct sys_device *dev, const char *buf, size_t sz) return sz; } -static ssize_t version_show(struct sys_device *dev, char *buf) +static ssize_t version_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; return sprintf(buf, "0x%x\n", uci->rev); } -static ssize_t pf_show(struct sys_device *dev, char *buf) +static ssize_t pf_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct ucode_cpu_info *uci = ucode_cpu_info + dev->id; diff --git a/drivers/base/cpu.c b/drivers/base/cpu.c index e38dfed41d8..20537d50790 100644 --- a/drivers/base/cpu.c +++ b/drivers/base/cpu.c @@ -21,15 +21,16 @@ EXPORT_SYMBOL(cpu_sysdev_class); static DEFINE_PER_CPU(struct sys_device *, cpu_sys_devices); #ifdef CONFIG_HOTPLUG_CPU -static ssize_t show_online(struct sys_device *dev, char *buf) +static ssize_t show_online(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { struct cpu *cpu = container_of(dev, struct cpu, sysdev); return sprintf(buf, "%u\n", !!cpu_online(cpu->sysdev.id)); } -static ssize_t __ref store_online(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t __ref store_online(struct sys_device *dev, struct sysdev_attribute *attr, + const char *buf, size_t count) { struct cpu *cpu = container_of(dev, struct cpu, sysdev); ssize_t ret; @@ -80,7 +81,8 @@ static inline void register_cpu_control(struct cpu *cpu) #ifdef CONFIG_KEXEC #include -static ssize_t show_crash_notes(struct sys_device *dev, char *buf) +static ssize_t show_crash_notes(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { struct cpu *cpu = container_of(dev, struct cpu, sysdev); ssize_t rc; diff --git a/drivers/base/memory.c b/drivers/base/memory.c index 937e8258981..4d4e0e7b6e9 100644 --- a/drivers/base/memory.c +++ b/drivers/base/memory.c @@ -92,7 +92,8 @@ unregister_memory(struct memory_block *memory, struct mem_section *section) * uses. */ -static ssize_t show_mem_phys_index(struct sys_device *dev, char *buf) +static ssize_t show_mem_phys_index(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct memory_block *mem = container_of(dev, struct memory_block, sysdev); @@ -102,7 +103,8 @@ static ssize_t show_mem_phys_index(struct sys_device *dev, char *buf) /* * online, offline, going offline, etc. */ -static ssize_t show_mem_state(struct sys_device *dev, char *buf) +static ssize_t show_mem_state(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct memory_block *mem = container_of(dev, struct memory_block, sysdev); @@ -217,7 +219,8 @@ out: } static ssize_t -store_mem_state(struct sys_device *dev, const char *buf, size_t count) +store_mem_state(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { struct memory_block *mem; unsigned int phys_section_nr; @@ -248,7 +251,8 @@ out: * s.t. if I offline all of these sections I can then * remove the physical device? */ -static ssize_t show_phys_device(struct sys_device *dev, char *buf) +static ssize_t show_phys_device(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { struct memory_block *mem = container_of(dev, struct memory_block, sysdev); diff --git a/drivers/base/node.c b/drivers/base/node.c index 0f867a08333..5116b78c632 100644 --- a/drivers/base/node.c +++ b/drivers/base/node.c @@ -36,11 +36,13 @@ static ssize_t node_read_cpumap(struct sys_device *dev, int type, char *buf) return len; } -static inline ssize_t node_read_cpumask(struct sys_device *dev, char *buf) +static inline ssize_t node_read_cpumask(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { return node_read_cpumap(dev, 0, buf); } -static inline ssize_t node_read_cpulist(struct sys_device *dev, char *buf) +static inline ssize_t node_read_cpulist(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { return node_read_cpumap(dev, 1, buf); } @@ -49,7 +51,8 @@ static SYSDEV_ATTR(cpumap, S_IRUGO, node_read_cpumask, NULL); static SYSDEV_ATTR(cpulist, S_IRUGO, node_read_cpulist, NULL); #define K(x) ((x) << (PAGE_SHIFT - 10)) -static ssize_t node_read_meminfo(struct sys_device * dev, char * buf) +static ssize_t node_read_meminfo(struct sys_device * dev, + struct sysdev_attribute *attr, char * buf) { int n; int nid = dev->id; @@ -112,7 +115,8 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf) #undef K static SYSDEV_ATTR(meminfo, S_IRUGO, node_read_meminfo, NULL); -static ssize_t node_read_numastat(struct sys_device * dev, char * buf) +static ssize_t node_read_numastat(struct sys_device * dev, + struct sysdev_attribute *attr, char * buf) { return sprintf(buf, "numa_hit %lu\n" @@ -130,7 +134,8 @@ static ssize_t node_read_numastat(struct sys_device * dev, char * buf) } static SYSDEV_ATTR(numastat, S_IRUGO, node_read_numastat, NULL); -static ssize_t node_read_distance(struct sys_device * dev, char * buf) +static ssize_t node_read_distance(struct sys_device * dev, + struct sysdev_attribute *attr, char * buf) { int nid = dev->id; int len = 0; diff --git a/drivers/base/sys.c b/drivers/base/sys.c index 50690d9df24..dc7dace14e1 100644 --- a/drivers/base/sys.c +++ b/drivers/base/sys.c @@ -36,7 +36,7 @@ sysdev_show(struct kobject * kobj, struct attribute * attr, char * buffer) struct sysdev_attribute * sysdev_attr = to_sysdev_attr(attr); if (sysdev_attr->show) - return sysdev_attr->show(sysdev, buffer); + return sysdev_attr->show(sysdev, sysdev_attr, buffer); return -EIO; } @@ -49,7 +49,7 @@ sysdev_store(struct kobject * kobj, struct attribute * attr, struct sysdev_attribute * sysdev_attr = to_sysdev_attr(attr); if (sysdev_attr->store) - return sysdev_attr->store(sysdev, buffer, count); + return sysdev_attr->store(sysdev, sysdev_attr, buffer, count); return -EIO; } diff --git a/drivers/base/topology.c b/drivers/base/topology.c index 3f6d9b0a6ab..199cd97e32e 100644 --- a/drivers/base/topology.c +++ b/drivers/base/topology.c @@ -34,7 +34,8 @@ static SYSDEV_ATTR(_name, 0444, show_##_name, NULL) #define define_id_show_func(name) \ -static ssize_t show_##name(struct sys_device *dev, char *buf) \ +static ssize_t show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ unsigned int cpu = dev->id; \ return sprintf(buf, "%d\n", topology_##name(cpu)); \ @@ -59,14 +60,17 @@ static ssize_t show_cpumap(int type, cpumask_t *mask, char *buf) #ifdef arch_provides_topology_pointers #define define_siblings_show_map(name) \ -static ssize_t show_##name(struct sys_device *dev, char *buf) \ +static ssize_t show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ unsigned int cpu = dev->id; \ return show_cpumap(0, &(topology_##name(cpu)), buf); \ } #define define_siblings_show_list(name) \ -static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ +static ssize_t show_##name##_list(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ { \ unsigned int cpu = dev->id; \ return show_cpumap(1, &(topology_##name(cpu)), buf); \ @@ -74,7 +78,8 @@ static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ #else #define define_siblings_show_map(name) \ -static ssize_t show_##name(struct sys_device *dev, char *buf) \ +static ssize_t show_##name(struct sys_device *dev, \ + struct sysdev_attribute *attr, char *buf) \ { \ unsigned int cpu = dev->id; \ cpumask_t mask = topology_##name(cpu); \ @@ -82,7 +87,9 @@ static ssize_t show_##name(struct sys_device *dev, char *buf) \ } #define define_siblings_show_list(name) \ -static ssize_t show_##name##_list(struct sys_device *dev, char *buf) \ +static ssize_t show_##name##_list(struct sys_device *dev, \ + struct sysdev_attribute *attr, \ + char *buf) \ { \ unsigned int cpu = dev->id; \ cpumask_t mask = topology_##name(cpu); \ diff --git a/drivers/cpuidle/sysfs.c b/drivers/cpuidle/sysfs.c index e949618b9be..31a0e0b455b 100644 --- a/drivers/cpuidle/sysfs.c +++ b/drivers/cpuidle/sysfs.c @@ -21,7 +21,8 @@ static int __init cpuidle_sysfs_setup(char *unused) } __setup("cpuidle_sysfs_switch", cpuidle_sysfs_setup); -static ssize_t show_available_governors(struct sys_device *dev, char *buf) +static ssize_t show_available_governors(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t i = 0; struct cpuidle_governor *tmp; @@ -39,7 +40,8 @@ out: return i; } -static ssize_t show_current_driver(struct sys_device *dev, char *buf) +static ssize_t show_current_driver(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t ret; @@ -53,7 +55,8 @@ static ssize_t show_current_driver(struct sys_device *dev, char *buf) return ret; } -static ssize_t show_current_governor(struct sys_device *dev, char *buf) +static ssize_t show_current_governor(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t ret; @@ -68,6 +71,7 @@ static ssize_t show_current_governor(struct sys_device *dev, char *buf) } static ssize_t store_current_governor(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { char gov_name[CPUIDLE_NAME_LEN]; diff --git a/drivers/xen/balloon.c b/drivers/xen/balloon.c index 591bc29b55f..d4427cb8697 100644 --- a/drivers/xen/balloon.c +++ b/drivers/xen/balloon.c @@ -610,6 +610,7 @@ static ssize_t show_target_kb(struct sys_device *dev, char *buf) } static ssize_t store_target_kb(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { diff --git a/include/linux/sysdev.h b/include/linux/sysdev.h index f2767bc6b73..8dcf3162b21 100644 --- a/include/linux/sysdev.h +++ b/include/linux/sysdev.h @@ -99,8 +99,9 @@ extern void sysdev_unregister(struct sys_device *); struct sysdev_attribute { struct attribute attr; - ssize_t (*show)(struct sys_device *, char *); - ssize_t (*store)(struct sys_device *, const char *, size_t); + ssize_t (*show)(struct sys_device *, struct sysdev_attribute *, char *); + ssize_t (*store)(struct sys_device *, struct sysdev_attribute *, + const char *, size_t); }; diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index 092e4c620af..a56f629b057 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -297,8 +297,8 @@ static int test_func(void *data) * * opcode:data */ -static ssize_t sysfs_test_command(struct sys_device *dev, const char *buf, - size_t count) +static ssize_t sysfs_test_command(struct sys_device *dev, struct sysdev_attribute *attr, + const char *buf, size_t count) { struct sched_param schedpar; struct test_thread_data *td; @@ -360,7 +360,8 @@ static ssize_t sysfs_test_command(struct sys_device *dev, const char *buf, * @dev: thread to query * @buf: char buffer to be filled with thread status info */ -static ssize_t sysfs_test_status(struct sys_device *dev, char *buf) +static ssize_t sysfs_test_status(struct sys_device *dev, struct sysdev_attribute *attr, + char *buf) { struct test_thread_data *td; struct task_struct *tsk; diff --git a/kernel/sched.c b/kernel/sched.c index 99e6d850eca..b1104ea5d25 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -7737,11 +7737,13 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) } #ifdef CONFIG_SCHED_MC -static ssize_t sched_mc_power_savings_show(struct sys_device *dev, char *page) +static ssize_t sched_mc_power_savings_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *page) { return sprintf(page, "%u\n", sched_mc_power_savings); } static ssize_t sched_mc_power_savings_store(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { return sched_power_savings_store(buf, count, 0); @@ -7751,11 +7753,13 @@ static SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show, #endif #ifdef CONFIG_SCHED_SMT -static ssize_t sched_smt_power_savings_show(struct sys_device *dev, char *page) +static ssize_t sched_smt_power_savings_show(struct sys_device *dev, + struct sysdev_attribute *attr, char *page) { return sprintf(page, "%u\n", sched_smt_power_savings); } static ssize_t sched_smt_power_savings_store(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { return sched_power_savings_store(buf, count, 1); diff --git a/kernel/time/clocksource.c b/kernel/time/clocksource.c index dadde5361f3..b1c2da81b05 100644 --- a/kernel/time/clocksource.c +++ b/kernel/time/clocksource.c @@ -376,7 +376,8 @@ void clocksource_unregister(struct clocksource *cs) * Provides sysfs interface for listing current clocksource. */ static ssize_t -sysfs_show_current_clocksources(struct sys_device *dev, char *buf) +sysfs_show_current_clocksources(struct sys_device *dev, + struct sysdev_attribute *attr, char *buf) { ssize_t count = 0; @@ -397,6 +398,7 @@ sysfs_show_current_clocksources(struct sys_device *dev, char *buf) * clocksource selction. */ static ssize_t sysfs_override_clocksource(struct sys_device *dev, + struct sysdev_attribute *attr, const char *buf, size_t count) { struct clocksource *ovr = NULL; @@ -449,7 +451,9 @@ static ssize_t sysfs_override_clocksource(struct sys_device *dev, * Provides sysfs interface for listing registered clocksources */ static ssize_t -sysfs_show_available_clocksources(struct sys_device *dev, char *buf) +sysfs_show_available_clocksources(struct sys_device *dev, + struct sysdev_attribute *attr, + char *buf) { struct clocksource *src; ssize_t count = 0; -- cgit v1.2.3-70-g09d2 From 9800794ac11d4646384b3a310dfd1fe2eed577bf Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 1 Jul 2008 18:48:42 +0200 Subject: sysdev: Add utility functions for simple int/ulong variable sysdev attributes This adds a new sysdev_ext_attribute that stores a pointer to the variable it manages and some utility functions/macro to easily use them. Previously all users wrote custom macros to generate show/store functions for each variable, with this it is possible to avoid that in many cases. Signed-off-by: Andi Kleen Signed-off-by: Greg Kroah-Hartman --- drivers/base/sys.c | 49 +++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/sysdev.h | 34 ++++++++++++++++++++++++++++++++++ 2 files changed, 83 insertions(+) (limited to 'drivers/base') diff --git a/drivers/base/sys.c b/drivers/base/sys.c index dc7dace14e1..40fc14f0354 100644 --- a/drivers/base/sys.c +++ b/drivers/base/sys.c @@ -479,3 +479,52 @@ int __init system_bus_init(void) EXPORT_SYMBOL_GPL(sysdev_register); EXPORT_SYMBOL_GPL(sysdev_unregister); + +#define to_ext_attr(x) container_of(x, struct sysdev_ext_attribute, attr) + +ssize_t sysdev_store_ulong(struct sys_device *sysdev, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + struct sysdev_ext_attribute *ea = to_ext_attr(attr); + char *end; + unsigned long new = simple_strtoul(buf, &end, 0); + if (end == buf) + return -EINVAL; + *(unsigned long *)(ea->var) = new; + return end - buf; +} +EXPORT_SYMBOL_GPL(sysdev_store_ulong); + +ssize_t sysdev_show_ulong(struct sys_device *sysdev, + struct sysdev_attribute *attr, + char *buf) +{ + struct sysdev_ext_attribute *ea = to_ext_attr(attr); + return snprintf(buf, PAGE_SIZE, "%lx\n", *(unsigned long *)(ea->var)); +} +EXPORT_SYMBOL_GPL(sysdev_show_ulong); + +ssize_t sysdev_store_int(struct sys_device *sysdev, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + struct sysdev_ext_attribute *ea = to_ext_attr(attr); + char *end; + long new = simple_strtol(buf, &end, 0); + if (end == buf || new > INT_MAX || new < INT_MIN) + return -EINVAL; + *(int *)(ea->var) = new; + return end - buf; +} +EXPORT_SYMBOL_GPL(sysdev_store_int); + +ssize_t sysdev_show_int(struct sys_device *sysdev, + struct sysdev_attribute *attr, + char *buf) +{ + struct sysdev_ext_attribute *ea = to_ext_attr(attr); + return snprintf(buf, PAGE_SIZE, "%d\n", *(int *)(ea->var)); +} +EXPORT_SYMBOL_GPL(sysdev_show_int); + diff --git a/include/linux/sysdev.h b/include/linux/sysdev.h index 8dcf3162b21..f395bb3fa2f 100644 --- a/include/linux/sysdev.h +++ b/include/linux/sysdev.h @@ -119,4 +119,38 @@ struct sysdev_attribute { extern int sysdev_create_file(struct sys_device *, struct sysdev_attribute *); extern void sysdev_remove_file(struct sys_device *, struct sysdev_attribute *); +struct sysdev_ext_attribute { + struct sysdev_attribute attr; + void *var; +}; + +/* + * Support for simple variable sysdev attributes. + * The pointer to the variable is stored in a sysdev_ext_attribute + */ + +/* Add more types as needed */ + +extern ssize_t sysdev_show_ulong(struct sys_device *, struct sysdev_attribute *, + char *); +extern ssize_t sysdev_store_ulong(struct sys_device *, + struct sysdev_attribute *, const char *, size_t); +extern ssize_t sysdev_show_int(struct sys_device *, struct sysdev_attribute *, + char *); +extern ssize_t sysdev_store_int(struct sys_device *, + struct sysdev_attribute *, const char *, size_t); + +#define _SYSDEV_ULONG_ATTR(_name, _mode, _var) \ + { _SYSDEV_ATTR(_name, _mode, sysdev_show_ulong, sysdev_store_ulong), \ + &(_var) } +#define SYSDEV_ULONG_ATTR(_name, _mode, _var) \ + struct sysdev_ext_attribute attr_##_name = \ + _SYSDEV_ULONG_ATTR(_name, _mode, _var); +#define _SYSDEV_INT_ATTR(_name, _mode, _var) \ + { _SYSDEV_ATTR(_name, _mode, sysdev_show_int, sysdev_store_int), \ + &(_var) } +#define SYSDEV_INT_ATTR(_name, _mode, _var) \ + struct sysdev_ext_attribute attr_##_name = \ + _SYSDEV_INT_ATTR(_name, _mode, _var); + #endif /* _SYSDEV_H_ */ -- cgit v1.2.3-70-g09d2