From 1cc5f7142eca352109895fe20b1fc6405dd17727 Mon Sep 17 00:00:00 2001 From: Ed Swierk Date: Mon, 25 Sep 2006 16:25:36 -0700 Subject: [PATCH] load_module: no BUG if module_subsys uninitialized Invoking load_module() before param_sysfs_init() is called crashes in mod_sysfs_setup(), since the kset in module_subsys is not initialized yet. In my case, net-pf-1 is getting modprobed as a result of hotplug trying to create a UNIX socket. Calls to hotplug begin after the topology_init initcall. Another patch for the same symptom (module_subsys-initialize-earlier.patch) moves param_sysfs_init() to the subsys initcalls, but this is still not early enough in the boot process in some cases. In particular, topology_init() causes /sbin/hotplug to run, which requests net-pf-1 (the UNIX socket protocol) which can be compiled as a module. Moving param_sysfs_init() to the postcore initcalls fixes this particular race, but there might well be other cases where a usermodehelper causes a module to load earlier still. The patch makes load_module() return an error rather than crashing the kernel if invoked before module_subsys is initialized. Cc: Mark Huang Cc: Greg KH Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/module.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'kernel') diff --git a/kernel/module.c b/kernel/module.c index 2a19cd47c046..b7fe6e840963 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1054,6 +1054,12 @@ static int mod_sysfs_setup(struct module *mod, { int err; + if (!module_subsys.kset.subsys) { + printk(KERN_ERR "%s: module_subsys not initialized\n", + mod->name); + err = -EINVAL; + goto out; + } memset(&mod->mkobj.kobj, 0, sizeof(mod->mkobj.kobj)); err = kobject_set_name(&mod->mkobj.kobj, "%s", mod->name); if (err) -- cgit v1.2.3 From 7c8265f51073bc8632a99de78d5fd19117ed78b7 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 24 Jun 2006 14:50:29 -0700 Subject: Suspend infrastructure cleanup and extension Allow devices to participate in the suspend process more intimately, in particular, allow the final phase (with interrupts disabled) to also be open to normal devices, not just system devices. Also, allow classes to participate in device suspend. Signed-off-by: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/resume.c | 28 ++++++++-- drivers/base/power/suspend.c | 122 ++++++++++++++++++++++++++++++++----------- include/linux/device.h | 11 +++- include/linux/pm.h | 1 + kernel/power/main.c | 4 ++ 5 files changed, 130 insertions(+), 36 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/resume.c b/drivers/base/power/resume.c index 826093ef4c7e..48e3d49d7b65 100644 --- a/drivers/base/power/resume.c +++ b/drivers/base/power/resume.c @@ -38,13 +38,35 @@ int resume_device(struct device * dev) dev_dbg(dev,"resuming\n"); error = dev->bus->resume(dev); } + if (dev->class && dev->class->resume) { + dev_dbg(dev,"class resume\n"); + error = dev->class->resume(dev); + } up(&dev->sem); TRACE_RESUME(error); return error; } +static int resume_device_early(struct device * dev) +{ + int error = 0; + TRACE_DEVICE(dev); + TRACE_RESUME(0); + if (dev->bus && dev->bus->resume_early) { + dev_dbg(dev,"EARLY resume\n"); + error = dev->bus->resume_early(dev); + } + TRACE_RESUME(error); + return error; +} + +/* + * 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 + */ void dpm_resume(void) { down(&dpm_list_sem); @@ -99,10 +121,8 @@ void dpm_power_up(void) struct list_head * entry = dpm_off_irq.next; struct device * dev = to_device(entry); - get_device(dev); - list_move_tail(entry, &dpm_active); - resume_device(dev); - put_device(dev); + list_move_tail(entry, &dpm_off); + resume_device_early(dev); } } diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c index 69509e02f703..10e8032aee1a 100644 --- a/drivers/base/power/suspend.c +++ b/drivers/base/power/suspend.c @@ -65,7 +65,19 @@ int suspend_device(struct device * dev, pm_message_t state) dev->power.prev_state = dev->power.power_state; - if (dev->bus && dev->bus->suspend && !dev->power.power_state.event) { + if (dev->class && dev->class->suspend && !dev->power.power_state.event) { + dev_dbg(dev, "class %s%s\n", + suspend_verb(state.event), + ((state.event == PM_EVENT_SUSPEND) + && device_may_wakeup(dev)) + ? ", may wakeup" + : "" + ); + error = dev->class->suspend(dev, state); + suspend_report_result(dev->class->suspend, error); + } + + if (!error && dev->bus && dev->bus->suspend && !dev->power.power_state.event) { dev_dbg(dev, "%s%s\n", suspend_verb(state.event), ((state.event == PM_EVENT_SUSPEND) @@ -81,15 +93,74 @@ int suspend_device(struct device * dev, pm_message_t state) } +/* + * This is called with interrupts off, only a single CPU + * running. We can't do down() on a semaphore (and we don't + * need the protection) + */ +static int suspend_device_late(struct device *dev, pm_message_t state) +{ + int error = 0; + + if (dev->power.power_state.event) { + dev_dbg(dev, "PM: suspend_late %d-->%d\n", + dev->power.power_state.event, state.event); + } + + if (dev->bus && dev->bus->suspend_late && !dev->power.power_state.event) { + dev_dbg(dev, "LATE %s%s\n", + suspend_verb(state.event), + ((state.event == PM_EVENT_SUSPEND) + && device_may_wakeup(dev)) + ? ", may wakeup" + : "" + ); + error = dev->bus->suspend_late(dev, state); + suspend_report_result(dev->bus->suspend_late, error); + } + return error; +} + +/** + * device_prepare_suspend - save state and prepare to suspend + * + * NOTE! Devices cannot detach at this point - not only do we + * hold the device list semaphores over the whole prepare, but + * the whole point is to do non-invasive preparatory work, not + * the actual suspend. + */ +int device_prepare_suspend(pm_message_t state) +{ + int error = 0; + struct device * dev; + + down(&dpm_sem); + down(&dpm_list_sem); + list_for_each_entry_reverse(dev, &dpm_active, power.entry) { + if (!dev->bus || !dev->bus->suspend_prepare) + continue; + error = dev->bus->suspend_prepare(dev, state); + if (error) + break; + } + up(&dpm_list_sem); + up(&dpm_sem); + return error; +} + /** * device_suspend - Save state and stop all devices in system. * @state: Power state to put each device in. * * Walk the dpm_active list, call ->suspend() for each device, and move - * it to dpm_off. - * Check the return value for each. If it returns 0, then we move the - * the device to the dpm_off list. If it returns -EAGAIN, we move it to - * the dpm_off_irq list. If we get a different error, try and back out. + * 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). + * + * If we get a different error, try and back out. * * If we hit a failure with any of the devices, call device_resume() * above to bring the suspended devices back to life. @@ -115,39 +186,27 @@ int device_suspend(pm_message_t state) /* Check if the device got removed */ if (!list_empty(&dev->power.entry)) { - /* Move it to the dpm_off or dpm_off_irq list */ + /* Move it to the dpm_off list */ if (!error) list_move(&dev->power.entry, &dpm_off); - else if (error == -EAGAIN) { - list_move(&dev->power.entry, &dpm_off_irq); - error = 0; - } } if (error) printk(KERN_ERR "Could not suspend device %s: " - "error %d\n", kobject_name(&dev->kobj), error); + "error %d%s\n", + kobject_name(&dev->kobj), error, + error == -EAGAIN ? " (please convert to suspend_late)" : ""); put_device(dev); } up(&dpm_list_sem); - if (error) { - /* we failed... before resuming, bring back devices from - * dpm_off_irq list back to main dpm_off list, we do want - * to call resume() on them, in case they partially suspended - * despite returning -EAGAIN - */ - while (!list_empty(&dpm_off_irq)) { - struct list_head * entry = dpm_off_irq.next; - list_move(entry, &dpm_off); - } + if (error) dpm_resume(); - } + up(&dpm_sem); return error; } EXPORT_SYMBOL_GPL(device_suspend); - /** * device_power_down - Shut down special devices. * @state: Power state to enter. @@ -162,14 +221,17 @@ int device_power_down(pm_message_t state) int error = 0; struct device * dev; - list_for_each_entry_reverse(dev, &dpm_off_irq, power.entry) { - if ((error = suspend_device(dev, state))) - break; + while (!list_empty(&dpm_off)) { + struct list_head * entry = dpm_off.prev; + + dev = to_device(entry); + error = suspend_device_late(dev, state); + if (error) + goto Error; + list_move(&dev->power.entry, &dpm_off_irq); } - if (error) - goto Error; - if ((error = sysdev_suspend(state))) - goto Error; + + error = sysdev_suspend(state); Done: return error; Error: diff --git a/include/linux/device.h b/include/linux/device.h index 8a648cd94fa9..b40be6fca6fa 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -51,8 +51,12 @@ struct bus_type { int (*probe)(struct device * dev); int (*remove)(struct device * dev); void (*shutdown)(struct device * dev); - int (*suspend)(struct device * dev, pm_message_t state); - int (*resume)(struct device * dev); + + int (*suspend_prepare)(struct device * dev, pm_message_t state); + int (*suspend)(struct device * dev, pm_message_t state); + int (*suspend_late)(struct device * dev, pm_message_t state); + int (*resume_early)(struct device * dev); + int (*resume)(struct device * dev); }; extern int bus_register(struct bus_type * bus); @@ -154,6 +158,9 @@ struct class { void (*release)(struct class_device *dev); void (*class_release)(struct class *class); + + int (*suspend)(struct device *, pm_message_t state); + int (*resume)(struct device *); }; extern int class_register(struct class *); diff --git a/include/linux/pm.h b/include/linux/pm.h index 658c1b93d5bb..096fb6f754cf 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -190,6 +190,7 @@ extern void device_resume(void); extern suspend_disk_method_t pm_disk_mode; extern int device_suspend(pm_message_t state); +extern int device_prepare_suspend(pm_message_t state); #define device_set_wakeup_enable(dev,val) \ ((dev)->power.should_wakeup = !!(val)) diff --git a/kernel/power/main.c b/kernel/power/main.c index 6d295c776794..0c3ed6ac938e 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -57,6 +57,10 @@ static int suspend_prepare(suspend_state_t state) if (!pm_ops || !pm_ops->enter) return -EPERM; + error = device_prepare_suspend(PMSG_SUSPEND); + if (error) + return error; + pm_prepare_console(); disable_nonboot_cpus(); -- cgit v1.2.3 From f1cc0a894c963923b766eb2d455747495e6e982d Mon Sep 17 00:00:00 2001 From: David Brownell Date: Mon, 14 Aug 2006 23:11:08 -0700 Subject: PM: issue PM_EVENT_PRETHAW This patch is the first of this series that should actually change any behavior ... by issuing the new event, now tha the rest of the kernel is prepared to receive it. This converts the PM core to issue the new PRETHAW message, which the rest of the kernel is now ready to receive. Signed-off-by: David Brownell Cc: "Rafael J. Wysocki" Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/suspend.c | 1 + kernel/power/disk.c | 4 ++-- kernel/power/swsusp.c | 9 ++++++++- kernel/power/user.c | 2 +- 4 files changed, 12 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c index 0bda4a7f2042..e86db83746ac 100644 --- a/drivers/base/power/suspend.c +++ b/drivers/base/power/suspend.c @@ -34,6 +34,7 @@ static inline char *suspend_verb(u32 event) 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)"; } } diff --git a/kernel/power/disk.c b/kernel/power/disk.c index e13e74067845..a3c34fb14321 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -98,7 +98,7 @@ static void unprepare_processes(void) } /** - * pm_suspend_disk - The granpappy of power management. + * pm_suspend_disk - The granpappy of hibernation power management. * * If we're going through the firmware, then get it over with quickly. * @@ -207,7 +207,7 @@ static int software_resume(void) pr_debug("PM: Preparing devices for restore.\n"); - if ((error = device_suspend(PMSG_FREEZE))) { + if ((error = device_suspend(PMSG_PRETHAW))) { printk("Some devices failed to suspend\n"); swsusp_free(); goto Thaw; diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 17f669c83012..62752899b1a1 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -248,6 +248,9 @@ int swsusp_suspend(void) restore_processor_state(); Restore_highmem: restore_highmem(); + /* NOTE: device_power_up() is just a resume() for devices + * that suspended with irqs off ... no overall powerup. + */ device_power_up(); Enable_irqs: local_irq_enable(); @@ -257,8 +260,12 @@ Enable_irqs: int swsusp_resume(void) { int error; + local_irq_disable(); - if (device_power_down(PMSG_FREEZE)) + /* NOTE: device_power_down() is just a suspend() with irqs off; + * it has no special "power things down" semantics + */ + if (device_power_down(PMSG_PRETHAW)) printk(KERN_ERR "Some devices failed to power down, very bad\n"); /* We'll ignore saved state, but this gets preempt count (etc) right */ save_processor_state(); diff --git a/kernel/power/user.c b/kernel/power/user.c index 3f1539fbe48a..5a8d060d7909 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -191,7 +191,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, } down(&pm_sem); pm_prepare_console(); - error = device_suspend(PMSG_FREEZE); + error = device_suspend(PMSG_PRETHAW); if (!error) { error = swsusp_resume(); device_resume(); -- cgit v1.2.3 From 2bca293e56b6a8cd16bb6e70a09b2adac9c723b5 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 30 Aug 2006 13:54:36 -0700 Subject: PM: add kconfig option for deprecated .../power/state files Add a new PM_SYSFS_DEPRECATED config option to control whether or not the /sys/devices/.../power/state files are provided. This will make it easier to get rid of that mechanism when the time comes, and to verify that userspace tools work right without it. Signed-off-by: David Brownell Acked-by: Pavel Machek Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/sysfs.c | 6 ++++++ kernel/power/Kconfig | 11 +++++++++++ 2 files changed, 17 insertions(+) (limited to 'kernel') diff --git a/drivers/base/power/sysfs.c b/drivers/base/power/sysfs.c index e55b3c2779e9..2d47517dbe32 100644 --- a/drivers/base/power/sysfs.c +++ b/drivers/base/power/sysfs.c @@ -7,6 +7,8 @@ #include "power.h" +#ifdef CONFIG_PM_SYSFS_DEPRECATED + /** * state - Control current power state of device * @@ -66,6 +68,8 @@ static ssize_t state_store(struct device * dev, struct device_attribute *attr, c static DEVICE_ATTR(state, 0644, state_show, state_store); +#endif /* CONFIG_PM_SYSFS_DEPRECATED */ + /* * wakeup - Report/change current wakeup option for device * @@ -139,7 +143,9 @@ static DEVICE_ATTR(wakeup, 0644, wake_show, wake_store); static struct attribute * power_attrs[] = { +#ifdef CONFIG_PM_SYSFS_DEPRECATED &dev_attr_state.attr, +#endif &dev_attr_wakeup.attr, NULL, }; diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 619ecabf7c58..1ed972070d19 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -53,6 +53,17 @@ config PM_TRACE CAUTION: this option will cause your machine's real-time clock to be set to an invalid time after a resume. +config PM_SYSFS_DEPRECATED + bool "Driver model /sys/devices/.../power/state files (DEPRECATED)" + depends on PM && SYSFS + default n + help + The driver model started out with a sysfs file intended to provide + a userspace hook for device power management. This feature has never + worked very well, except for limited testing purposes, and so it will + be removed. It's not clear that a generic mechanism could really + handle the wide variability of device power states; any replacements + are likely to be bus or driver specific. config SOFTWARE_SUSPEND bool "Software Suspend" -- cgit v1.2.3 From 1d3a82af45428c5e8deaa119cdeb79611ae46371 Mon Sep 17 00:00:00 2001 From: David Brownell Date: Wed, 30 Aug 2006 14:09:47 -0700 Subject: PM: no suspend_prepare() phase Remove the new suspend_prepare() phase. It doesn't seem very usable, has never been tested, doesn't address fault cleanup, and would need a sibling resume_complete(); plus there are no real use cases. It could be restored later if those issues get resolved. Signed-off-by: David Brownell Cc: Linus Torvalds Signed-off-by: Greg Kroah-Hartman --- drivers/base/power/suspend.c | 27 --------------------------- drivers/pci/pci-driver.c | 14 -------------- include/linux/device.h | 1 - include/linux/pci.h | 1 - kernel/power/main.c | 4 ---- 5 files changed, 47 deletions(-) (limited to 'kernel') diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c index e86db83746ac..6453c25103d2 100644 --- a/drivers/base/power/suspend.c +++ b/drivers/base/power/suspend.c @@ -117,33 +117,6 @@ static int suspend_device_late(struct device *dev, pm_message_t state) return error; } -/** - * device_prepare_suspend - save state and prepare to suspend - * - * NOTE! Devices cannot detach at this point - not only do we - * hold the device list semaphores over the whole prepare, but - * the whole point is to do non-invasive preparatory work, not - * the actual suspend. - */ -int device_prepare_suspend(pm_message_t state) -{ - int error = 0; - struct device * dev; - - down(&dpm_sem); - down(&dpm_list_sem); - list_for_each_entry_reverse(dev, &dpm_active, power.entry) { - if (!dev->bus || !dev->bus->suspend_prepare) - continue; - error = dev->bus->suspend_prepare(dev, state); - if (error) - break; - } - up(&dpm_list_sem); - up(&dpm_sem); - return error; -} - /** * device_suspend - Save state and stop all devices in system. * @state: Power state to put each device in. diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 9e7d6ceb3805..8948ac9ab681 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -264,19 +264,6 @@ static int pci_device_remove(struct device * dev) return 0; } -static int pci_device_suspend_prepare(struct device * dev, pm_message_t state) -{ - struct pci_dev * pci_dev = to_pci_dev(dev); - struct pci_driver * drv = pci_dev->driver; - int i = 0; - - if (drv && drv->suspend_prepare) { - i = drv->suspend_prepare(pci_dev, state); - suspend_report_result(drv->suspend_prepare, i); - } - return i; -} - static int pci_device_suspend(struct device * dev, pm_message_t state) { struct pci_dev * pci_dev = to_pci_dev(dev); @@ -544,7 +531,6 @@ struct bus_type pci_bus_type = { .uevent = pci_uevent, .probe = pci_device_probe, .remove = pci_device_remove, - .suspend_prepare= pci_device_suspend_prepare, .suspend = pci_device_suspend, .suspend_late = pci_device_suspend_late, .resume_early = pci_device_resume_early, diff --git a/include/linux/device.h b/include/linux/device.h index b40be6fca6fa..b3646462d6dc 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -52,7 +52,6 @@ struct bus_type { int (*remove)(struct device * dev); void (*shutdown)(struct device * dev); - int (*suspend_prepare)(struct device * dev, pm_message_t state); int (*suspend)(struct device * dev, pm_message_t state); int (*suspend_late)(struct device * dev, pm_message_t state); int (*resume_early)(struct device * dev); diff --git a/include/linux/pci.h b/include/linux/pci.h index 4b2e629467c7..9514bbfe96e2 100644 --- a/include/linux/pci.h +++ b/include/linux/pci.h @@ -345,7 +345,6 @@ struct pci_driver { const struct pci_device_id *id_table; /* must be non-NULL for probe to be called */ int (*probe) (struct pci_dev *dev, const struct pci_device_id *id); /* New device inserted */ void (*remove) (struct pci_dev *dev); /* Device removed (NULL if not a hot-plug capable driver) */ - int (*suspend_prepare) (struct pci_dev *dev, pm_message_t state); int (*suspend) (struct pci_dev *dev, pm_message_t state); /* Device suspended */ int (*suspend_late) (struct pci_dev *dev, pm_message_t state); int (*resume_early) (struct pci_dev *dev); diff --git a/kernel/power/main.c b/kernel/power/main.c index 0c3ed6ac938e..6d295c776794 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -57,10 +57,6 @@ static int suspend_prepare(suspend_state_t state) if (!pm_ops || !pm_ops->enter) return -EPERM; - error = device_prepare_suspend(PMSG_SUSPEND); - if (error) - return error; - pm_prepare_console(); disable_nonboot_cpus(); -- cgit v1.2.3 From 2fbe7b25c8edaf2d10e6c1a4cc9f8afe714c4764 Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Tue, 26 Sep 2006 10:52:27 +0200 Subject: [PATCH] i386/x86-64: Remove un/set_nmi_callback and reserve/release_lapic_nmi functions Removes the un/set_nmi_callback and reserve/release_lapic_nmi functions as they are no longer needed. The various subsystems are modified to register with the die_notifier instead. Also includes compile fixes by Andrew Morton. Signed-off-by: Don Zickus Signed-off-by: Andi Kleen --- arch/i386/kernel/crash.c | 20 +++++++- arch/i386/kernel/nmi.c | 85 ++++--------------------------- arch/i386/kernel/traps.c | 23 +-------- arch/i386/oprofile/nmi_int.c | 47 ++++++++++------- arch/i386/oprofile/nmi_timer_int.c | 33 +++++++++--- arch/x86_64/kernel/crash.c | 20 +++++++- arch/x86_64/kernel/nmi.c | 102 +++---------------------------------- include/asm-i386/nmi.h | 21 ++------ include/asm-x86_64/nmi.h | 21 -------- kernel/sysctl.c | 4 +- 10 files changed, 116 insertions(+), 260 deletions(-) (limited to 'kernel') diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c index 5b96f038367f..736c76d6b31d 100644 --- a/arch/i386/kernel/crash.c +++ b/arch/i386/kernel/crash.c @@ -22,6 +22,8 @@ #include #include #include +#include + #include @@ -93,9 +95,18 @@ static void crash_save_self(struct pt_regs *regs) #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) static atomic_t waiting_for_crash_ipi; -static int crash_nmi_callback(struct pt_regs *regs, int cpu) +static int crash_nmi_callback(struct notifier_block *self, + unsigned long val, void *data) { + struct pt_regs *regs; struct pt_regs fixed_regs; + int cpu; + + if (val != DIE_NMI) + return NOTIFY_OK; + + regs = ((struct die_args *)data)->regs; + cpu = raw_smp_processor_id(); /* Don't do anything if this handler is invoked on crashing cpu. * Otherwise, system will completely hang. Crashing cpu can get @@ -125,13 +136,18 @@ static void smp_send_nmi_allbutself(void) send_IPI_allbutself(NMI_VECTOR); } +static struct notifier_block crash_nmi_nb = { + .notifier_call = crash_nmi_callback, +}; + static void nmi_shootdown_cpus(void) { unsigned long msecs; atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); /* Would it be better to replace the trap vector here? */ - set_nmi_callback(crash_nmi_callback); + if (register_die_notifier(&crash_nmi_nb)) + return; /* return what? */ /* Ensure the new callback function is set before sending * out the NMI */ diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index bd96ea4f2942..acd3fdea2a21 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -42,20 +42,6 @@ static DEFINE_PER_CPU(unsigned long, evntsel_nmi_owner[3]); */ #define NMI_MAX_COUNTER_BITS 66 -/* - * lapic_nmi_owner tracks the ownership of the lapic NMI hardware: - * - it may be reserved by some other driver, or not - * - when not reserved by some other driver, it may be used for - * the NMI watchdog, or not - * - * This is maintained separately from nmi_active because the NMI - * watchdog may also be driven from the I/O APIC timer. - */ -static DEFINE_SPINLOCK(lapic_nmi_owner_lock); -static unsigned int lapic_nmi_owner; -#define LAPIC_NMI_WATCHDOG (1<<0) -#define LAPIC_NMI_RESERVED (1<<1) - /* nmi_active: * >0: the lapic NMI watchdog is active, but can be disabled * <0: the lapic NMI watchdog has not been set up, and cannot @@ -325,33 +311,6 @@ static void enable_lapic_nmi_watchdog(void) touch_nmi_watchdog(); } -int reserve_lapic_nmi(void) -{ - unsigned int old_owner; - - spin_lock(&lapic_nmi_owner_lock); - old_owner = lapic_nmi_owner; - lapic_nmi_owner |= LAPIC_NMI_RESERVED; - spin_unlock(&lapic_nmi_owner_lock); - if (old_owner & LAPIC_NMI_RESERVED) - return -EBUSY; - if (old_owner & LAPIC_NMI_WATCHDOG) - disable_lapic_nmi_watchdog(); - return 0; -} - -void release_lapic_nmi(void) -{ - unsigned int new_owner; - - spin_lock(&lapic_nmi_owner_lock); - new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED; - lapic_nmi_owner = new_owner; - spin_unlock(&lapic_nmi_owner_lock); - if (new_owner & LAPIC_NMI_WATCHDOG) - enable_lapic_nmi_watchdog(); -} - void disable_timer_nmi_watchdog(void) { BUG_ON(nmi_watchdog != NMI_IO_APIC); @@ -866,6 +825,15 @@ done: return rc; } +int do_nmi_callback(struct pt_regs * regs, int cpu) +{ +#ifdef CONFIG_SYSCTL + if (unknown_nmi_panic) + return unknown_nmi_panic_callback(regs, cpu); +#endif + return 0; +} + #ifdef CONFIG_SYSCTL static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) @@ -873,37 +841,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) unsigned char reason = get_nmi_reason(); char buf[64]; - if (!(reason & 0xc0)) { - sprintf(buf, "NMI received for unknown reason %02x\n", reason); - die_nmi(regs, buf); - } - return 0; -} - -/* - * proc handler for /proc/sys/kernel/unknown_nmi_panic - */ -int proc_unknown_nmi_panic(ctl_table *table, int write, struct file *file, - void __user *buffer, size_t *length, loff_t *ppos) -{ - int old_state; - - old_state = unknown_nmi_panic; - proc_dointvec(table, write, file, buffer, length, ppos); - if (!!old_state == !!unknown_nmi_panic) - return 0; - - if (unknown_nmi_panic) { - if (reserve_lapic_nmi() < 0) { - unknown_nmi_panic = 0; - return -EBUSY; - } else { - set_nmi_callback(unknown_nmi_panic_callback); - } - } else { - release_lapic_nmi(); - unset_nmi_callback(); - } + sprintf(buf, "NMI received for unknown reason %02x\n", reason); + die_nmi(regs, buf); return 0; } @@ -917,7 +856,5 @@ EXPORT_SYMBOL(reserve_perfctr_nmi); EXPORT_SYMBOL(release_perfctr_nmi); EXPORT_SYMBOL(reserve_evntsel_nmi); EXPORT_SYMBOL(release_evntsel_nmi); -EXPORT_SYMBOL(reserve_lapic_nmi); -EXPORT_SYMBOL(release_lapic_nmi); EXPORT_SYMBOL(disable_timer_nmi_watchdog); EXPORT_SYMBOL(enable_timer_nmi_watchdog); diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 282f0bd40dfd..7db664d0b25c 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -706,13 +706,6 @@ void die_nmi (struct pt_regs *regs, const char *msg) do_exit(SIGSEGV); } -static int dummy_nmi_callback(struct pt_regs * regs, int cpu) -{ - return 0; -} - -static nmi_callback_t nmi_callback = dummy_nmi_callback; - static void default_do_nmi(struct pt_regs * regs) { unsigned char reason = 0; @@ -732,9 +725,10 @@ static void default_do_nmi(struct pt_regs * regs) */ if (nmi_watchdog_tick(regs, reason)) return; + if (!do_nmi_callback(regs, smp_processor_id())) #endif - if (!rcu_dereference(nmi_callback)(regs, smp_processor_id())) unknown_nmi_error(reason, regs); + return; } if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP) @@ -765,19 +759,6 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code) nmi_exit(); } -void set_nmi_callback(nmi_callback_t callback) -{ - vmalloc_sync_all(); - rcu_assign_pointer(nmi_callback, callback); -} -EXPORT_SYMBOL_GPL(set_nmi_callback); - -void unset_nmi_callback(void) -{ - nmi_callback = dummy_nmi_callback; -} -EXPORT_SYMBOL_GPL(unset_nmi_callback); - #ifdef CONFIG_KPROBES fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code) { diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c index 8710ca081b1e..b3610188bcf0 100644 --- a/arch/i386/oprofile/nmi_int.c +++ b/arch/i386/oprofile/nmi_int.c @@ -17,14 +17,15 @@ #include #include #include +#include #include "op_counter.h" #include "op_x86_model.h" - + static struct op_x86_model_spec const * model; static struct op_msrs cpu_msrs[NR_CPUS]; static unsigned long saved_lvtpc[NR_CPUS]; - + static int nmi_start(void); static void nmi_stop(void); @@ -82,13 +83,24 @@ static void exit_driverfs(void) #define exit_driverfs() do { } while (0) #endif /* CONFIG_PM */ - -static int nmi_callback(struct pt_regs * regs, int cpu) +int profile_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data) { - return model->check_ctrs(regs, &cpu_msrs[cpu]); + struct die_args *args = (struct die_args *)data; + int ret = NOTIFY_DONE; + int cpu = smp_processor_id(); + + switch(val) { + case DIE_NMI: + if (model->check_ctrs(args->regs, &cpu_msrs[cpu])) + ret = NOTIFY_STOP; + break; + default: + break; + } + return ret; } - - + static void nmi_cpu_save_registers(struct op_msrs * msrs) { unsigned int const nr_ctrs = model->num_counters; @@ -174,27 +186,29 @@ static void nmi_cpu_setup(void * dummy) apic_write(APIC_LVTPC, APIC_DM_NMI); } +static struct notifier_block profile_exceptions_nb = { + .notifier_call = profile_exceptions_notify, + .next = NULL, + .priority = 0 +}; static int nmi_setup(void) { + int err=0; + if (!allocate_msrs()) return -ENOMEM; - /* We walk a thin line between law and rape here. - * We need to be careful to install our NMI handler - * without actually triggering any NMIs as this will - * break the core code horrifically. - */ - if (reserve_lapic_nmi() < 0) { + if ((err = register_die_notifier(&profile_exceptions_nb))){ free_msrs(); - return -EBUSY; + return err; } + /* We need to serialize save and setup for HT because the subset * of msrs are distinct for save and setup operations */ on_each_cpu(nmi_save_registers, NULL, 0, 1); on_each_cpu(nmi_cpu_setup, NULL, 0, 1); - set_nmi_callback(nmi_callback); nmi_enabled = 1; return 0; } @@ -250,8 +264,7 @@ static void nmi_shutdown(void) { nmi_enabled = 0; on_each_cpu(nmi_cpu_shutdown, NULL, 0, 1); - unset_nmi_callback(); - release_lapic_nmi(); + unregister_die_notifier(&profile_exceptions_nb); free_msrs(); } diff --git a/arch/i386/oprofile/nmi_timer_int.c b/arch/i386/oprofile/nmi_timer_int.c index a33a73bb502d..93ca48f804ac 100644 --- a/arch/i386/oprofile/nmi_timer_int.c +++ b/arch/i386/oprofile/nmi_timer_int.c @@ -17,32 +17,49 @@ #include #include #include +#include -static int nmi_timer_callback(struct pt_regs * regs, int cpu) +int profile_timer_exceptions_notify(struct notifier_block *self, + unsigned long val, void *data) { - oprofile_add_sample(regs, 0); - return 1; + struct die_args *args = (struct die_args *)data; + int ret = NOTIFY_DONE; + + switch(val) { + case DIE_NMI: + oprofile_add_sample(args->regs, 0); + ret = NOTIFY_STOP; + break; + default: + break; + } + return ret; } +static struct notifier_block profile_timer_exceptions_nb = { + .notifier_call = profile_timer_exceptions_notify, + .next = NULL, + .priority = 0 +}; + static int timer_start(void) { - disable_timer_nmi_watchdog(); - set_nmi_callback(nmi_timer_callback); + if (register_die_notifier(&profile_timer_exceptions_nb)) + return 1; return 0; } static void timer_stop(void) { - enable_timer_nmi_watchdog(); - unset_nmi_callback(); + unregister_die_notifier(&profile_timer_exceptions_nb); synchronize_sched(); /* Allow already-started NMIs to complete. */ } int __init op_nmi_timer_init(struct oprofile_operations * ops) { - if (atomic_read(&nmi_active) <= 0) + if ((nmi_watchdog != NMI_IO_APIC) || (atomic_read(&nmi_active) <= 0)) return -ENODEV; ops->start = timer_start; diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c index d8d5750d6106..44c8af65325e 100644 --- a/arch/x86_64/kernel/crash.c +++ b/arch/x86_64/kernel/crash.c @@ -23,6 +23,7 @@ #include #include #include +#include /* This keeps a track of which one is crashing cpu. */ static int crashing_cpu; @@ -95,8 +96,18 @@ static void crash_save_self(struct pt_regs *regs) #ifdef CONFIG_SMP static atomic_t waiting_for_crash_ipi; -static int crash_nmi_callback(struct pt_regs *regs, int cpu) +static int crash_nmi_callback(struct notifier_block *self, + unsigned long val, void *data) { + struct pt_regs *regs; + int cpu; + + if (val != DIE_NMI) + return NOTIFY_OK; + + regs = ((struct die_args *)data)->regs; + cpu = raw_smp_processor_id(); + /* * Don't do anything if this handler is invoked on crashing cpu. * Otherwise, system will completely hang. Crashing cpu can get @@ -127,12 +138,17 @@ static void smp_send_nmi_allbutself(void) * cpu hotplug shouldn't matter. */ +static struct notifier_block crash_nmi_nb = { + .notifier_call = crash_nmi_callback, +}; + static void nmi_shootdown_cpus(void) { unsigned long msecs; atomic_set(&waiting_for_crash_ipi, num_online_cpus() - 1); - set_nmi_callback(crash_nmi_callback); + if (register_die_notifier(&crash_nmi_nb)) + return; /* return what? */ /* * Ensure the new callback function is set before sending diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index f6b881b23a70..9d175dcf3a2d 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -41,20 +41,6 @@ static DEFINE_PER_CPU(unsigned, evntsel_nmi_owner[2]); */ #define NMI_MAX_COUNTER_BITS 66 -/* - * lapic_nmi_owner tracks the ownership of the lapic NMI hardware: - * - it may be reserved by some other driver, or not - * - when not reserved by some other driver, it may be used for - * the NMI watchdog, or not - * - * This is maintained separately from nmi_active because the NMI - * watchdog may also be driven from the I/O APIC timer. - */ -static DEFINE_SPINLOCK(lapic_nmi_owner_lock); -static unsigned int lapic_nmi_owner; -#define LAPIC_NMI_WATCHDOG (1<<0) -#define LAPIC_NMI_RESERVED (1<<1) - /* nmi_active: * >0: the lapic NMI watchdog is active, but can be disabled * <0: the lapic NMI watchdog has not been set up, and cannot @@ -321,33 +307,6 @@ static void enable_lapic_nmi_watchdog(void) touch_nmi_watchdog(); } -int reserve_lapic_nmi(void) -{ - unsigned int old_owner; - - spin_lock(&lapic_nmi_owner_lock); - old_owner = lapic_nmi_owner; - lapic_nmi_owner |= LAPIC_NMI_RESERVED; - spin_unlock(&lapic_nmi_owner_lock); - if (old_owner & LAPIC_NMI_RESERVED) - return -EBUSY; - if (old_owner & LAPIC_NMI_WATCHDOG) - disable_lapic_nmi_watchdog(); - return 0; -} - -void release_lapic_nmi(void) -{ - unsigned int new_owner; - - spin_lock(&lapic_nmi_owner_lock); - new_owner = lapic_nmi_owner & ~LAPIC_NMI_RESERVED; - lapic_nmi_owner = new_owner; - spin_unlock(&lapic_nmi_owner_lock); - if (new_owner & LAPIC_NMI_WATCHDOG) - enable_lapic_nmi_watchdog(); -} - void disable_timer_nmi_watchdog(void) { BUG_ON(nmi_watchdog != NMI_IO_APIC); @@ -762,13 +721,6 @@ done: return rc; } -static __kprobes int dummy_nmi_callback(struct pt_regs * regs, int cpu) -{ - return 0; -} - -static nmi_callback_t nmi_callback = dummy_nmi_callback; - asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code) { nmi_enter(); @@ -779,21 +731,12 @@ asmlinkage __kprobes void do_nmi(struct pt_regs * regs, long error_code) int do_nmi_callback(struct pt_regs * regs, int cpu) { - return rcu_dereference(nmi_callback)(regs, cpu); -} - -void set_nmi_callback(nmi_callback_t callback) -{ - vmalloc_sync_all(); - rcu_assign_pointer(nmi_callback, callback); -} -EXPORT_SYMBOL_GPL(set_nmi_callback); - -void unset_nmi_callback(void) -{ - nmi_callback = dummy_nmi_callback; +#ifdef CONFIG_SYSCTL + if (unknown_nmi_panic) + return unknown_nmi_panic_callback(regs, cpu); +#endif + return 0; } -EXPORT_SYMBOL_GPL(unset_nmi_callback); #ifdef CONFIG_SYSCTL @@ -802,37 +745,8 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) unsigned char reason = get_nmi_reason(); char buf[64]; - if (!(reason & 0xc0)) { - sprintf(buf, "NMI received for unknown reason %02x\n", reason); - die_nmi(buf,regs); - } - return 0; -} - -/* - * proc handler for /proc/sys/kernel/unknown_nmi_panic - */ -int proc_unknown_nmi_panic(struct ctl_table *table, int write, struct file *file, - void __user *buffer, size_t *length, loff_t *ppos) -{ - int old_state; - - old_state = unknown_nmi_panic; - proc_dointvec(table, write, file, buffer, length, ppos); - if (!!old_state == !!unknown_nmi_panic) - return 0; - - if (unknown_nmi_panic) { - if (reserve_lapic_nmi() < 0) { - unknown_nmi_panic = 0; - return -EBUSY; - } else { - set_nmi_callback(unknown_nmi_panic_callback); - } - } else { - release_lapic_nmi(); - unset_nmi_callback(); - } + sprintf(buf, "NMI received for unknown reason %02x\n", reason); + die_nmi(buf,regs); return 0; } @@ -846,8 +760,6 @@ EXPORT_SYMBOL(reserve_perfctr_nmi); EXPORT_SYMBOL(release_perfctr_nmi); EXPORT_SYMBOL(reserve_evntsel_nmi); EXPORT_SYMBOL(release_evntsel_nmi); -EXPORT_SYMBOL(reserve_lapic_nmi); -EXPORT_SYMBOL(release_lapic_nmi); EXPORT_SYMBOL(disable_timer_nmi_watchdog); EXPORT_SYMBOL(enable_timer_nmi_watchdog); EXPORT_SYMBOL(touch_nmi_watchdog); diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h index da0e0b4e9139..34d6bf063b6e 100644 --- a/include/asm-i386/nmi.h +++ b/include/asm-i386/nmi.h @@ -6,24 +6,13 @@ #include -struct pt_regs; - -typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu); - -/** - * set_nmi_callback - * - * Set a handler for an NMI. Only one handler may be - * set. Return 1 if the NMI was handled. - */ -void set_nmi_callback(nmi_callback_t callback); - /** - * unset_nmi_callback + * do_nmi_callback * - * Remove the handler previously set. + * Check to see if a callback exists and execute it. Return 1 + * if the handler exists and was handled successfully. */ -void unset_nmi_callback(void); +int do_nmi_callback(struct pt_regs *regs, int cpu); extern int avail_to_resrv_perfctr_nmi_bit(unsigned int); extern int avail_to_resrv_perfctr_nmi(unsigned int); @@ -33,8 +22,6 @@ extern int reserve_evntsel_nmi(unsigned int); extern void release_evntsel_nmi(unsigned int); extern void setup_apic_nmi_watchdog (void *); -extern int reserve_lapic_nmi(void); -extern void release_lapic_nmi(void); extern void disable_timer_nmi_watchdog(void); extern void enable_timer_nmi_watchdog(void); extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason); diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h index 8f02a2a416e6..8818c39d34e0 100644 --- a/include/asm-x86_64/nmi.h +++ b/include/asm-x86_64/nmi.h @@ -7,25 +7,6 @@ #include #include -struct pt_regs; - -typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu); - -/** - * set_nmi_callback - * - * Set a handler for an NMI. Only one handler may be - * set. Return 1 if the NMI was handled. - */ -void set_nmi_callback(nmi_callback_t callback); - -/** - * unset_nmi_callback - * - * Remove the handler previously set. - */ -void unset_nmi_callback(void); - /** * do_nmi_callback * @@ -72,8 +53,6 @@ extern int reserve_evntsel_nmi(unsigned int); extern void release_evntsel_nmi(unsigned int); extern void setup_apic_nmi_watchdog (void *); -extern int reserve_lapic_nmi(void); -extern void release_lapic_nmi(void); extern void disable_timer_nmi_watchdog(void); extern void enable_timer_nmi_watchdog(void); extern int nmi_watchdog_tick (struct pt_regs * regs, unsigned reason); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 362a0cc37138..83f168361624 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -76,8 +76,6 @@ extern int compat_log; #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) int unknown_nmi_panic; -extern int proc_unknown_nmi_panic(ctl_table *, int, struct file *, - void __user *, size_t *, loff_t *); #endif /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ @@ -628,7 +626,7 @@ static ctl_table kern_table[] = { .data = &unknown_nmi_panic, .maxlen = sizeof (int), .mode = 0644, - .proc_handler = &proc_unknown_nmi_panic, + .proc_handler = &proc_dointvec, }, #endif #if defined(CONFIG_X86) -- cgit v1.2.3 From 407984f1af259b31957c7c05075a454a751bb801 Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Tue, 26 Sep 2006 10:52:27 +0200 Subject: [PATCH] x86: Add abilty to enable/disable nmi watchdog with sysctl Adds a new /proc/sys/kernel/nmi call that will enable/disable the nmi watchdog. Signed-off-by: Don Zickus Signed-off-by: Andi Kleen --- arch/i386/kernel/nmi.c | 52 ++++++++++++++++++++++++++++++++++++++++++++++++ arch/x86_64/kernel/nmi.c | 48 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-i386/nmi.h | 1 + include/asm-x86_64/nmi.h | 1 + include/linux/sysctl.h | 1 + kernel/sysctl.c | 11 ++++++++++ 6 files changed, 114 insertions(+) (limited to 'kernel') diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c index acd3fdea2a21..28065d0b71a9 100644 --- a/arch/i386/kernel/nmi.c +++ b/arch/i386/kernel/nmi.c @@ -846,6 +846,58 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) return 0; } +/* + * proc handler for /proc/sys/kernel/nmi_watchdog + */ +int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, + void __user *buffer, size_t *length, loff_t *ppos) +{ + int old_state; + + nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0; + old_state = nmi_watchdog_enabled; + proc_dointvec(table, write, file, buffer, length, ppos); + if (!!old_state == !!nmi_watchdog_enabled) + return 0; + + if (atomic_read(&nmi_active) < 0) { + printk(KERN_WARNING "NMI watchdog is permanently disabled\n"); + return -EINVAL; + } + + if (nmi_watchdog == NMI_DEFAULT) { + if (nmi_known_cpu() > 0) + nmi_watchdog = NMI_LOCAL_APIC; + else + nmi_watchdog = NMI_IO_APIC; + } + + if (nmi_watchdog == NMI_LOCAL_APIC) + { + if (nmi_watchdog_enabled) + enable_lapic_nmi_watchdog(); + else + disable_lapic_nmi_watchdog(); + } else if (nmi_watchdog == NMI_IO_APIC) { + /* FIXME + * for some reason these functions don't work + */ + printk("Can not enable/disable NMI on IO APIC\n"); + return -EINVAL; +#if 0 + if (nmi_watchdog_enabled) + enable_timer_nmi_watchdog(); + else + disable_timer_nmi_watchdog(); +#endif + } else { + printk( KERN_WARNING + "NMI watchdog doesn't know what hardware to touch\n"); + return -EIO; + } + return 0; +} + #endif EXPORT_SYMBOL(nmi_active); diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c index 9d175dcf3a2d..3a17411a9a19 100644 --- a/arch/x86_64/kernel/nmi.c +++ b/arch/x86_64/kernel/nmi.c @@ -750,6 +750,54 @@ static int unknown_nmi_panic_callback(struct pt_regs *regs, int cpu) return 0; } +/* + * proc handler for /proc/sys/kernel/nmi + */ +int proc_nmi_enabled(struct ctl_table *table, int write, struct file *file, + void __user *buffer, size_t *length, loff_t *ppos) +{ + int old_state; + + nmi_watchdog_enabled = (atomic_read(&nmi_active) > 0) ? 1 : 0; + old_state = nmi_watchdog_enabled; + proc_dointvec(table, write, file, buffer, length, ppos); + if (!!old_state == !!nmi_watchdog_enabled) + return 0; + + if (atomic_read(&nmi_active) < 0) { + printk( KERN_WARNING "NMI watchdog is permanently disabled\n"); + return -EINVAL; + } + + /* if nmi_watchdog is not set yet, then set it */ + nmi_watchdog_default(); + + if (nmi_watchdog == NMI_LOCAL_APIC) + { + if (nmi_watchdog_enabled) + enable_lapic_nmi_watchdog(); + else + disable_lapic_nmi_watchdog(); + } else if (nmi_watchdog == NMI_IO_APIC) { + /* FIXME + * for some reason these functions don't work + */ + printk("Can not enable/disable NMI on IO APIC\n"); + return -EIO; +#if 0 + if (nmi_watchdog_enabled) + enable_timer_nmi_watchdog(); + else + disable_timer_nmi_watchdog(); +#endif + } else { + printk(KERN_WARNING + "NMI watchdog doesn't know what hardware to touch\n"); + return -EIO; + } + return 0; +} + #endif EXPORT_SYMBOL(nmi_active); diff --git a/include/asm-i386/nmi.h b/include/asm-i386/nmi.h index 34d6bf063b6e..13b5d8311bf7 100644 --- a/include/asm-i386/nmi.h +++ b/include/asm-i386/nmi.h @@ -14,6 +14,7 @@ */ int do_nmi_callback(struct pt_regs *regs, int cpu); +extern int nmi_watchdog_enabled; extern int avail_to_resrv_perfctr_nmi_bit(unsigned int); extern int avail_to_resrv_perfctr_nmi(unsigned int); extern int reserve_perfctr_nmi(unsigned int); diff --git a/include/asm-x86_64/nmi.h b/include/asm-x86_64/nmi.h index 8818c39d34e0..2c23b0df87d2 100644 --- a/include/asm-x86_64/nmi.h +++ b/include/asm-x86_64/nmi.h @@ -43,6 +43,7 @@ extern void die_nmi(char *str, struct pt_regs *regs); extern int panic_on_timeout; extern int unknown_nmi_panic; +extern int nmi_watchdog_enabled; extern int check_nmi_watchdog(void); extern int avail_to_resrv_perfctr_nmi_bit(unsigned int); diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 736ed917a4f8..ecb79ba52ae1 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -150,6 +150,7 @@ enum KERN_IA64_UNALIGNED=72, /* int: ia64 unaligned userland trap enable */ KERN_COMPAT_LOG=73, /* int: print compat layer messages */ KERN_MAX_LOCK_DEPTH=74, + KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */ }; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 83f168361624..040de6bd74dd 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -76,6 +76,9 @@ extern int compat_log; #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_X86) int unknown_nmi_panic; +int nmi_watchdog_enabled; +extern int proc_nmi_enabled(struct ctl_table *, int , struct file *, + void __user *, size_t *, loff_t *); #endif /* this is needed for the proc_dointvec_minmax for [fs_]overflow UID and GID */ @@ -628,6 +631,14 @@ static ctl_table kern_table[] = { .mode = 0644, .proc_handler = &proc_dointvec, }, + { + .ctl_name = KERN_NMI_WATCHDOG, + .procname = "nmi_watchdog", + .data = &nmi_watchdog_enabled, + .maxlen = sizeof (int), + .mode = 0644, + .proc_handler = &proc_nmi_enabled, + }, #endif #if defined(CONFIG_X86) { -- cgit v1.2.3 From 8da5adda91df3d2fcc5300e68da491694c9af019 Mon Sep 17 00:00:00 2001 From: Don Zickus Date: Tue, 26 Sep 2006 10:52:27 +0200 Subject: [PATCH] x86: Allow users to force a panic on NMI To quote Alan Cox: The default Linux behaviour on an NMI of either memory or unknown is to continue operation. For many environments such as scientific computing it is preferable that the box is taken out and the error dealt with than an uncorrected parity/ECC error get propogated. A small number of systems do generate NMI's for bizarre random reasons such as power management so the default is unchanged. In other respects the new proc/sys entry works like the existing panic controls already in that directory. This is separate to the edac support - EDAC allows supported chipsets to handle ECC errors well, this change allows unsupported cases to at least panic rather than cause problems further down the line. Signed-off-by: Don Zickus Signed-off-by: Andi Kleen --- arch/i386/kernel/traps.c | 6 ++++++ arch/x86_64/kernel/traps.c | 6 ++++++ include/linux/kernel.h | 1 + include/linux/sysctl.h | 1 + kernel/panic.c | 1 + kernel/sysctl.c | 8 ++++++++ 6 files changed, 23 insertions(+) (limited to 'kernel') diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 7db664d0b25c..2f6cb8276480 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -635,6 +635,8 @@ static void mem_parity_error(unsigned char reason, struct pt_regs * regs) "to continue\n"); printk(KERN_EMERG "You probably have a hardware problem with your RAM " "chips\n"); + if (panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); /* Clear and disable the memory parity error line. */ clear_mem_error(reason); @@ -670,6 +672,10 @@ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs) reason, smp_processor_id()); printk("Dazed and confused, but trying to continue\n"); printk("Do you have a strange power saving mode enabled?\n"); + + if (panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); + } static DEFINE_SPINLOCK(nmi_print_lock); diff --git a/arch/x86_64/kernel/traps.c b/arch/x86_64/kernel/traps.c index 42bc070fdf11..b18829db2a6a 100644 --- a/arch/x86_64/kernel/traps.c +++ b/arch/x86_64/kernel/traps.c @@ -732,6 +732,8 @@ mem_parity_error(unsigned char reason, struct pt_regs * regs) { printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n"); printk("You probably have a hardware problem with your RAM chips\n"); + if (panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); /* Clear and disable the memory parity error line. */ reason = (reason & 0xf) | 4; @@ -757,6 +759,10 @@ unknown_nmi_error(unsigned char reason, struct pt_regs * regs) { printk("Uhhuh. NMI received for unknown reason %02x.\n", reason); printk("Dazed and confused, but trying to continue\n"); printk("Do you have a strange power saving mode enabled?\n"); + + if (panic_on_unrecovered_nmi) + panic("NMI: Not continuing"); + } /* Runs on IST stack. This code must keep interrupts off all the time. diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 2b2ae4fdce8b..1ff9609300b4 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -186,6 +186,7 @@ extern void bust_spinlocks(int yes); extern int oops_in_progress; /* If set, an oops, panic(), BUG() or die() is in progress */ extern int panic_timeout; extern int panic_on_oops; +extern int panic_on_unrecovered_nmi; extern int tainted; extern const char *print_tainted(void); extern void add_taint(unsigned); diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index ecb79ba52ae1..432778446ad2 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -151,6 +151,7 @@ enum KERN_COMPAT_LOG=73, /* int: print compat layer messages */ KERN_MAX_LOCK_DEPTH=74, KERN_NMI_WATCHDOG=75, /* int: enable/disable nmi watchdog */ + KERN_PANIC_ON_NMI=76, /* int: whether we will panic on an unrecovered */ }; diff --git a/kernel/panic.c b/kernel/panic.c index 8010b9b17aca..d2db3e2209e0 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -21,6 +21,7 @@ #include int panic_on_oops; +int panic_on_unrecovered_nmi; int tainted; static int pause_on_oops; static int pause_on_oops_flag; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 040de6bd74dd..220e20564124 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -641,6 +641,14 @@ static ctl_table kern_table[] = { }, #endif #if defined(CONFIG_X86) + { + .ctl_name = KERN_PANIC_ON_NMI, + .procname = "panic_on_unrecovered_nmi", + .data = &panic_on_unrecovered_nmi, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, { .ctl_name = KERN_BOOTLOADER_TYPE, .procname = "bootloader_type", -- cgit v1.2.3 From 3cfc348bf90ffaa777c188652aa297f04eb94de8 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 26 Sep 2006 10:52:28 +0200 Subject: [PATCH] x86: Add portable getcpu call For NUMA optimization and some other algorithms it is useful to have a fast to get the current CPU and node numbers in user space. x86-64 added a fast way to do this in a vsyscall. This adds a generic syscall for other architectures to make it a generic portable facility. I expect some of them will also implement it as a faster vsyscall. The cache is an optimization for the x86-64 vsyscall optimization. Since what the syscall returns is an approximation anyways and user space often wants very fast results it can be cached for some time. The norma methods to get this information in user space are relatively slow The vsyscall is in a better position to manage the cache because it has direct access to a fast time stamp (jiffies). For the generic syscall optimization it doesn't help much, but enforce a valid argument to keep programs portable I only added an i386 syscall entry for now. Other architectures can follow as needed. AK: Also added some cleanups from Andrew Morton Signed-off-by: Andi Kleen --- arch/i386/kernel/syscall_table.S | 1 + arch/x86_64/ia32/ia32entry.S | 1 + include/asm-i386/unistd.h | 3 ++- include/linux/syscalls.h | 2 ++ kernel/sys.c | 31 +++++++++++++++++++++++++++++++ 5 files changed, 37 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S index dd63d4775398..7e639f78b0b9 100644 --- a/arch/i386/kernel/syscall_table.S +++ b/arch/i386/kernel/syscall_table.S @@ -317,3 +317,4 @@ ENTRY(sys_call_table) .long sys_tee /* 315 */ .long sys_vmsplice .long sys_move_pages + .long sys_getcpu diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index 30ed0f6f4a2b..32fd32bea07c 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -713,4 +713,5 @@ ia32_sys_call_table: .quad sys_tee .quad compat_sys_vmsplice .quad compat_sys_move_pages + .quad sys_getcpu ia32_syscall_end: diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h index fc1c8ddae149..565d0897b205 100644 --- a/include/asm-i386/unistd.h +++ b/include/asm-i386/unistd.h @@ -323,10 +323,11 @@ #define __NR_tee 315 #define __NR_vmsplice 316 #define __NR_move_pages 317 +#define __NR_getcpu 318 #ifdef __KERNEL__ -#define NR_syscalls 318 +#define NR_syscalls 319 /* * user-visible error numbers are in the range -1 - -128: see diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 008f04c56737..3f0f716225ec 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -53,6 +53,7 @@ struct mq_attr; struct compat_stat; struct compat_timeval; struct robust_list_head; +struct getcpu_cache; #include #include @@ -596,5 +597,6 @@ asmlinkage long sys_get_robust_list(int pid, size_t __user *len_ptr); asmlinkage long sys_set_robust_list(struct robust_list_head __user *head, size_t len); +asmlinkage long sys_getcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *cache); #endif diff --git a/kernel/sys.c b/kernel/sys.c index e236f98f7ec5..3f894775488d 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -28,6 +28,7 @@ #include #include #include +#include #include #include @@ -2062,3 +2063,33 @@ asmlinkage long sys_prctl(int option, unsigned long arg2, unsigned long arg3, } return error; } + +asmlinkage long sys_getcpu(unsigned __user *cpup, unsigned __user *nodep, + struct getcpu_cache __user *cache) +{ + int err = 0; + int cpu = raw_smp_processor_id(); + if (cpup) + err |= put_user(cpu, cpup); + if (nodep) + err |= put_user(cpu_to_node(cpu), nodep); + if (cache) { + /* + * The cache is not needed for this implementation, + * but make sure user programs pass something + * valid. vsyscall implementations can instead make + * good use of the cache. Only use t0 and t1 because + * these are available in both 32bit and 64bit ABI (no + * need for a compat_getcpu). 32bit has enough + * padding + */ + unsigned long t0, t1; + get_user(t0, &cache->t0); + get_user(t1, &cache->t1); + t0++; + t1++; + put_user(t0, &cache->t0); + put_user(t1, &cache->t1); + } + return err ? -EFAULT : 0; +} -- cgit v1.2.3 From 0cb91a2293648507886563ccb91979cfc94d6a4b Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 26 Sep 2006 10:52:28 +0200 Subject: [PATCH] i386: Account spinlocks to the caller during profiling for !FP kernels This ports the algorithm from x86-64 (with improvements) to i386. Previously this only worked for frame pointer enabled kernels. But spinlocks have a very simple stack frame that can be manually analyzed. Do this. Signed-off-by: Andi Kleen --- arch/i386/kernel/time.c | 23 +++++++++++++++++++---- include/asm-i386/ptrace.h | 4 ---- kernel/spinlock.c | 5 +++++ 3 files changed, 24 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c index edd00f6cee37..5af802ef00b2 100644 --- a/arch/i386/kernel/time.c +++ b/arch/i386/kernel/time.c @@ -130,18 +130,33 @@ static int set_rtc_mmss(unsigned long nowtime) int timer_ack; -#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER) unsigned long profile_pc(struct pt_regs *regs) { unsigned long pc = instruction_pointer(regs); - if (!user_mode_vm(regs) && in_lock_functions(pc)) +#ifdef CONFIG_SMP + if (!user_mode_vm(regs) && in_lock_functions(pc)) { +#ifdef CONFIG_FRAME_POINTER return *(unsigned long *)(regs->ebp + 4); - +#else + unsigned long *sp; + if ((regs->xcs & 3) == 0) + sp = (unsigned long *)®s->esp; + else + sp = (unsigned long *)regs->esp; + /* Return address is either directly at stack pointer + or above a saved eflags. Eflags has bits 22-31 zero, + kernel addresses don't. */ + if (sp[0] >> 22) + return sp[0]; + if (sp[1] >> 22) + return sp[1]; +#endif + } +#endif return pc; } EXPORT_SYMBOL(profile_pc); -#endif /* * This is the same as the above, except we _also_ save the current diff --git a/include/asm-i386/ptrace.h b/include/asm-i386/ptrace.h index f324c53b6f9a..30a442ec2059 100644 --- a/include/asm-i386/ptrace.h +++ b/include/asm-i386/ptrace.h @@ -80,11 +80,7 @@ static inline int user_mode_vm(struct pt_regs *regs) return ((regs->xcs & 3) | (regs->eflags & VM_MASK)) != 0; } #define instruction_pointer(regs) ((regs)->eip) -#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER) extern unsigned long profile_pc(struct pt_regs *regs); -#else -#define profile_pc(regs) instruction_pointer(regs) -#endif #endif /* __KERNEL__ */ #endif diff --git a/kernel/spinlock.c b/kernel/spinlock.c index fb524b009eef..9644a41e0bef 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -7,6 +7,11 @@ * * This file contains the spinlock/rwlock implementations for the * SMP and the DEBUG_SPINLOCK cases. (UP-nondebug inlines them) + * + * Note that some architectures have special knowledge about the + * stack frames of these functions in their profile_pc. If you + * change anything significant here that could change the stack + * frame contact the architecture maintainers. */ #include -- cgit v1.2.3 From 5a1b3999d6cb7ab87f1f3b1700bc91839fd6fa29 Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 26 Sep 2006 10:52:34 +0200 Subject: [PATCH] x86: Some preparationary cleanup for stack trace - Remove unused all_contexts parameter No caller used it - Move skip argument into the structure (needed for followon patches) Cc: mingo@elte.hu Signed-off-by: Andi Kleen --- arch/i386/kernel/stacktrace.c | 11 +++-------- arch/s390/kernel/stacktrace.c | 17 ++++++++--------- arch/x86_64/kernel/stacktrace.c | 14 +++++--------- include/linux/stacktrace.h | 7 ++++--- kernel/lockdep.c | 5 ++++- 5 files changed, 24 insertions(+), 30 deletions(-) (limited to 'kernel') diff --git a/arch/i386/kernel/stacktrace.c b/arch/i386/kernel/stacktrace.c index e62a037ab399..ae3c32a87add 100644 --- a/arch/i386/kernel/stacktrace.c +++ b/arch/i386/kernel/stacktrace.c @@ -61,12 +61,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip, /* * Save stack-backtrace addresses into a stack_trace buffer. - * If all_contexts is set, all contexts (hardirq, softirq and process) - * are saved. If not set then only the current context is saved. */ -void save_stack_trace(struct stack_trace *trace, - struct task_struct *task, int all_contexts, - unsigned int skip) +void save_stack_trace(struct stack_trace *trace, struct task_struct *task) { unsigned long ebp; unsigned long *stack = &ebp; @@ -85,10 +81,9 @@ void save_stack_trace(struct stack_trace *trace, struct thread_info *context = (struct thread_info *) ((unsigned long)stack & (~(THREAD_SIZE - 1))); - ebp = save_context_stack(trace, skip, context, stack, ebp); + ebp = save_context_stack(trace, trace->skip, context, stack, ebp); stack = (unsigned long *)context->previous_esp; - if (!all_contexts || !stack || - trace->nr_entries >= trace->max_entries) + if (!stack || trace->nr_entries >= trace->max_entries) break; trace->entries[trace->nr_entries++] = ULONG_MAX; if (trace->nr_entries >= trace->max_entries) diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c index de83f38288d0..d9428a0fc8fb 100644 --- a/arch/s390/kernel/stacktrace.c +++ b/arch/s390/kernel/stacktrace.c @@ -59,9 +59,7 @@ static inline unsigned long save_context_stack(struct stack_trace *trace, } } -void save_stack_trace(struct stack_trace *trace, - struct task_struct *task, int all_contexts, - unsigned int skip) +void save_stack_trace(struct stack_trace *trace, struct task_struct *task) { register unsigned long sp asm ("15"); unsigned long orig_sp; @@ -69,22 +67,23 @@ void save_stack_trace(struct stack_trace *trace, sp &= PSW_ADDR_INSN; orig_sp = sp; - sp = save_context_stack(trace, &skip, sp, + sp = save_context_stack(trace, &trace->skip, sp, S390_lowcore.panic_stack - PAGE_SIZE, S390_lowcore.panic_stack); - if ((sp != orig_sp) && !all_contexts) + if ((sp != orig_sp) && !trace->all_contexts) return; - sp = save_context_stack(trace, &skip, sp, + sp = save_context_stack(trace, &trace->skip, sp, S390_lowcore.async_stack - ASYNC_SIZE, S390_lowcore.async_stack); - if ((sp != orig_sp) && !all_contexts) + if ((sp != orig_sp) && !trace->all_contexts) return; if (task) - save_context_stack(trace, &skip, sp, + save_context_stack(trace, &trace->skip, sp, (unsigned long) task_stack_page(task), (unsigned long) task_stack_page(task) + THREAD_SIZE); else - save_context_stack(trace, &skip, sp, S390_lowcore.thread_info, + save_context_stack(trace, &trace->skip, sp, + S390_lowcore.thread_info, S390_lowcore.thread_info + THREAD_SIZE); return; } diff --git a/arch/x86_64/kernel/stacktrace.c b/arch/x86_64/kernel/stacktrace.c index 32cf55eb9af8..1c022af8fe1e 100644 --- a/arch/x86_64/kernel/stacktrace.c +++ b/arch/x86_64/kernel/stacktrace.c @@ -109,9 +109,10 @@ out_restore: * Save stack-backtrace addresses into a stack_trace buffer: */ static inline unsigned long -save_context_stack(struct stack_trace *trace, unsigned int skip, +save_context_stack(struct stack_trace *trace, unsigned long stack, unsigned long stack_end) { + int skip = trace->skip; unsigned long addr; #ifdef CONFIG_FRAME_POINTER @@ -159,12 +160,8 @@ save_context_stack(struct stack_trace *trace, unsigned int skip, /* * Save stack-backtrace addresses into a stack_trace buffer. - * If all_contexts is set, all contexts (hardirq, softirq and process) - * are saved. If not set then only the current context is saved. */ -void save_stack_trace(struct stack_trace *trace, - struct task_struct *task, int all_contexts, - unsigned int skip) +void save_stack_trace(struct stack_trace *trace, struct task_struct *task) { unsigned long stack = (unsigned long)&stack; int i, nr_stacks = 0, stacks_done[MAX_STACKS]; @@ -207,9 +204,8 @@ void save_stack_trace(struct stack_trace *trace, return; stacks_done[nr_stacks] = stack_end; - stack = save_context_stack(trace, skip, stack, stack_end); - if (!all_contexts || !stack || - trace->nr_entries >= trace->max_entries) + stack = save_context_stack(trace, stack, stack_end); + if (!stack || trace->nr_entries >= trace->max_entries) return; trace->entries[trace->nr_entries++] = ULONG_MAX; if (trace->nr_entries >= trace->max_entries) diff --git a/include/linux/stacktrace.h b/include/linux/stacktrace.h index 9cc81e572224..50e2b01e517c 100644 --- a/include/linux/stacktrace.h +++ b/include/linux/stacktrace.h @@ -5,15 +5,16 @@ struct stack_trace { unsigned int nr_entries, max_entries; unsigned long *entries; + int skip; /* input argument: How many entries to skip */ + int all_contexts; /* input argument: if true do than one stack */ }; extern void save_stack_trace(struct stack_trace *trace, - struct task_struct *task, int all_contexts, - unsigned int skip); + struct task_struct *task); extern void print_stack_trace(struct stack_trace *trace, int spaces); #else -# define save_stack_trace(trace, task, all, skip) do { } while (0) +# define save_stack_trace(trace, task) do { } while (0) # define print_stack_trace(trace) do { } while (0) #endif diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 9bad17884513..900b4cb1a024 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -224,7 +224,10 @@ static int save_trace(struct stack_trace *trace) trace->max_entries = MAX_STACK_TRACE_ENTRIES - nr_stack_trace_entries; trace->entries = stack_trace + nr_stack_trace_entries; - save_stack_trace(trace, NULL, 0, 3); + trace->skip = 3; + trace->all_contexts = 0; + + save_stack_trace(trace, NULL); trace->max_entries = trace->nr_entries; -- cgit v1.2.3 From 3fa7c794fe4dc127f7fac3fad4d13628e68f89ce Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Tue, 26 Sep 2006 10:52:34 +0200 Subject: [PATCH] Avoid recursion in lockdep when stack tracer takes locks The new dwarf2 unwinder needs to take locks to do backtraces inside modules. This patch makes sure lockdep which calls stacktrace is not reentered. Thanks to Ingo for suggesting this simpler approach. Cc: mingo@elte.hu Signed-off-by: Andi Kleen --- kernel/lockdep.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'kernel') diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 900b4cb1a024..c088e5542e84 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -227,7 +227,11 @@ static int save_trace(struct stack_trace *trace) trace->skip = 3; trace->all_contexts = 0; + /* Make sure to not recurse in case the the unwinder needs to tak +e locks. */ + lockdep_off(); save_stack_trace(trace, NULL); + lockdep_on(); trace->max_entries = trace->nr_entries; -- cgit v1.2.3 From 0a4254058037eb172758961d0a5b94f4320a1425 Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 26 Sep 2006 10:52:38 +0200 Subject: [PATCH] Add the canary field to the PDA area and the task struct This patch adds the per thread cookie field to the task struct and the PDA. Also it makes sure that the PDA value gets the new cookie value at context switch, and that a new task gets a new cookie at task creation time. Signed-off-by: Arjan van Ven Signed-off-by: Ingo Molnar Signed-off-by: Andi Kleen CC: Andi Kleen --- arch/x86_64/kernel/process.c | 8 ++++++++ include/asm-x86_64/pda.h | 7 ++++++- include/linux/sched.h | 5 +++++ kernel/fork.c | 5 +++++ 4 files changed, 24 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/arch/x86_64/kernel/process.c b/arch/x86_64/kernel/process.c index 9e9a70e50c72..fba8dfeda67c 100644 --- a/arch/x86_64/kernel/process.c +++ b/arch/x86_64/kernel/process.c @@ -625,6 +625,14 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p) unlazy_fpu(prev_p); write_pda(kernelstack, task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET); +#ifdef CONFIG_CC_STACKPROTECTOR + write_pda(stack_canary, next_p->stack_canary); + /* + * Build time only check to make sure the stack_canary is at + * offset 40 in the pda; this is a gcc ABI requirement + */ + BUILD_BUG_ON(offsetof(struct x8664_pda, stack_canary) != 40); +#endif /* * Now maybe reload the debug registers and handle I/O bitmaps diff --git a/include/asm-x86_64/pda.h b/include/asm-x86_64/pda.h index 6794ffaae433..e7773e0af865 100644 --- a/include/asm-x86_64/pda.h +++ b/include/asm-x86_64/pda.h @@ -16,7 +16,12 @@ struct x8664_pda { unsigned long oldrsp; /* 24 user rsp for system call */ int irqcount; /* 32 Irq nesting counter. Starts with -1 */ int cpunumber; /* 36 Logical CPU number */ - char *irqstackptr; /* 40 top of irqstack */ +#ifdef CONFIG_CC_STACKPROTECTOR + unsigned long stack_canary; /* 40 stack canary value */ + /* gcc-ABI: this canary MUST be at + offset 40!!! */ +#endif + char *irqstackptr; int nodenumber; /* number of current node */ unsigned int __softirq_pending; unsigned int __nmi_count; /* number of NMI on this CPUs */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 807556c5bcd2..9d4aa7f95bc8 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -819,6 +819,11 @@ struct task_struct { unsigned did_exec:1; pid_t pid; pid_t tgid; + +#ifdef CONFIG_CC_STACKPROTECTOR + /* Canary value for the -fstack-protector gcc feature */ + unsigned long stack_canary; +#endif /* * pointers to (original) parent process, youngest child, younger sibling, * older sibling, respectively. (p->father can be replaced with diff --git a/kernel/fork.c b/kernel/fork.c index f9b014e3e700..a0dad84567c9 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -45,6 +45,7 @@ #include #include #include +#include #include #include @@ -175,6 +176,10 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) tsk->thread_info = ti; setup_thread_stack(tsk, orig); +#ifdef CONFIG_CC_STACKPROTECTOR + tsk->stack_canary = get_random_int(); +#endif + /* One for us, one for whoever does the "release_task()" (usually parent) */ atomic_set(&tsk->usage,2); atomic_set(&tsk->fs_excl, 0); -- cgit v1.2.3 From 3162f751d04086a9d006342de63ac8f44fe0f72a Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 26 Sep 2006 10:52:39 +0200 Subject: [PATCH] Add the __stack_chk_fail() function GCC emits a call to a __stack_chk_fail() function when the stack canary is not matching the expected value. Since this is a bad security issue; lets panic the kernel rather than limping along; the kernel really can't be trusted anymore when this happens. Signed-off-by: Arjan van de Ven Signed-off-by: Ingo Molnar Signed-off-by: Andi Kleen CC: Andi Kleen --- kernel/panic.c | 12 ++++++++++++ 1 file changed, 12 insertions(+) (limited to 'kernel') diff --git a/kernel/panic.c b/kernel/panic.c index d2db3e2209e0..6ceb664fb52a 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -271,3 +271,15 @@ void oops_exit(void) { do_oops_enter_exit(); } + +#ifdef CONFIG_CC_STACKPROTECTOR +/* + * Called when gcc's -fstack-protector feature is used, and + * gcc detects corruption of the on-stack canary value + */ +void __stack_chk_fail(void) +{ + panic("stack-protector: Kernel stack is corrupted"); +} +EXPORT_SYMBOL(__stack_chk_fail); +#endif -- cgit v1.2.3 From adf1423698f00d00b267f7dca8231340ce7d65ef Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Tue, 26 Sep 2006 10:52:41 +0200 Subject: [PATCH] i386/x86-64: Work around gcc bug with noreturn functions in unwinder Current gcc generates calls not jumps to noreturn functions. When that happens the return address can point to the next function, which confuses the unwinder. This patch works around it by marking asynchronous exception frames in contrast normal call frames in the unwind information. Then teach the unwinder to decode this. For normal call frames the unwinder now subtracts one from the address which avoids this problem. The standard libgcc unwinder uses the same trick. It doesn't include adjustment of the printed address (i.e. for the original example, it'd still be kernel_math_error+0 that gets displayed, but the unwinder wouldn't get confused anymore. This only works with binutils 2.6.17+ and some versions of H.J.Lu's 2.6.16 unfortunately because earlier binutils don't support .cfi_signal_frame [AK: added automatic detection of the new binutils and wrote description] Signed-off-by: Jan Beulich Signed-off-by: Andi Kleen --- arch/i386/Makefile | 4 ++++ arch/i386/kernel/entry.S | 4 ++++ arch/x86_64/Makefile | 4 ++++ arch/x86_64/ia32/ia32entry.S | 4 ++++ arch/x86_64/kernel/entry.S | 4 ++++ include/asm-i386/dwarf2.h | 7 +++++++ include/asm-i386/unwind.h | 5 +++++ include/asm-x86_64/dwarf2.h | 6 ++++++ include/asm-x86_64/unwind.h | 5 +++++ kernel/unwind.c | 35 ++++++++++++++++++++++++++++------- scripts/Kbuild.include | 4 ++-- 11 files changed, 73 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/arch/i386/Makefile b/arch/i386/Makefile index 508cdbeb3a09..7cc0b189b82b 100644 --- a/arch/i386/Makefile +++ b/arch/i386/Makefile @@ -50,6 +50,10 @@ CFLAGS += $(shell if [ $(call cc-version) -lt 0400 ] ; then echo $(call cc-op cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) +# is .cfi_signal_frame supported too? +cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) +AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) + CFLAGS += $(cflags-y) # Default subarch .c files diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S index 284f2e908ad0..5a63d6fdb70e 100644 --- a/arch/i386/kernel/entry.S +++ b/arch/i386/kernel/entry.S @@ -183,18 +183,21 @@ VM_MASK = 0x00020000 #define RING0_INT_FRAME \ CFI_STARTPROC simple;\ + CFI_SIGNAL_FRAME;\ CFI_DEF_CFA esp, 3*4;\ /*CFI_OFFSET cs, -2*4;*/\ CFI_OFFSET eip, -3*4 #define RING0_EC_FRAME \ CFI_STARTPROC simple;\ + CFI_SIGNAL_FRAME;\ CFI_DEF_CFA esp, 4*4;\ /*CFI_OFFSET cs, -2*4;*/\ CFI_OFFSET eip, -3*4 #define RING0_PTREGS_FRAME \ CFI_STARTPROC simple;\ + CFI_SIGNAL_FRAME;\ CFI_DEF_CFA esp, OLDESP-EBX;\ /*CFI_OFFSET cs, CS-OLDESP;*/\ CFI_OFFSET eip, EIP-OLDESP;\ @@ -275,6 +278,7 @@ need_resched: # sysenter call handler stub ENTRY(sysenter_entry) CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA esp, 0 CFI_REGISTER esp, ebp movl TSS_sysenter_esp0(%esp),%esp diff --git a/arch/x86_64/Makefile b/arch/x86_64/Makefile index 2b8d07c70106..1c0f18d4f887 100644 --- a/arch/x86_64/Makefile +++ b/arch/x86_64/Makefile @@ -58,6 +58,10 @@ cflags-y += $(call cc-option,-mno-sse -mno-mmx -mno-sse2 -mno-3dnow,) cflags-y += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_endproc,-DCONFIG_AS_CFI=1,) +# is .cfi_signal_frame supported too? +cflags-y += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) +AFLAGS += $(call as-instr,.cfi_startproc\n.cfi_signal_frame\n.cfi_endproc,-DCONFIG_AS_CFI_SIGNAL_FRAME=1,) + cflags-$(CONFIG_CC_STACKPROTECTOR) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector ) cflags-$(CONFIG_CC_STACKPROTECTOR_ALL) += $(shell $(CONFIG_SHELL) $(srctree)/scripts/gcc-x86_64-has-stack-protector.sh $(CC) -fstack-protector-all ) diff --git a/arch/x86_64/ia32/ia32entry.S b/arch/x86_64/ia32/ia32entry.S index 32fd32bea07c..b4aa875e175b 100644 --- a/arch/x86_64/ia32/ia32entry.S +++ b/arch/x86_64/ia32/ia32entry.S @@ -71,6 +71,7 @@ */ ENTRY(ia32_sysenter_target) CFI_STARTPROC32 simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,0 CFI_REGISTER rsp,rbp swapgs @@ -186,6 +187,7 @@ ENDPROC(ia32_sysenter_target) */ ENTRY(ia32_cstar_target) CFI_STARTPROC32 simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,PDA_STACKOFFSET CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ @@ -293,6 +295,7 @@ ia32_badarg: ENTRY(ia32_syscall) CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,SS+8-RIP /*CFI_REL_OFFSET ss,SS-RIP*/ CFI_REL_OFFSET rsp,RSP-RIP @@ -370,6 +373,7 @@ ENTRY(ia32_ptregs_common) popq %r11 CFI_ENDPROC CFI_STARTPROC32 simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,SS+8-ARGOFFSET CFI_REL_OFFSET rax,RAX-ARGOFFSET CFI_REL_OFFSET rcx,RCX-ARGOFFSET diff --git a/arch/x86_64/kernel/entry.S b/arch/x86_64/kernel/entry.S index ea32688386fd..4cbc65290ae7 100644 --- a/arch/x86_64/kernel/entry.S +++ b/arch/x86_64/kernel/entry.S @@ -123,6 +123,7 @@ .macro CFI_DEFAULT_STACK start=1 .if \start CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,SS+8 .else CFI_DEF_CFA_OFFSET SS+8 @@ -207,6 +208,7 @@ END(ret_from_fork) ENTRY(system_call) CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,PDA_STACKOFFSET CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ @@ -324,6 +326,7 @@ END(system_call) */ ENTRY(int_ret_from_sys_call) CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,SS+8-ARGOFFSET /*CFI_REL_OFFSET ss,SS-ARGOFFSET*/ CFI_REL_OFFSET rsp,RSP-ARGOFFSET @@ -484,6 +487,7 @@ END(stub_rt_sigreturn) */ .macro _frame ref CFI_STARTPROC simple + CFI_SIGNAL_FRAME CFI_DEF_CFA rsp,SS+8-\ref /*CFI_REL_OFFSET ss,SS-\ref*/ CFI_REL_OFFSET rsp,RSP-\ref diff --git a/include/asm-i386/dwarf2.h b/include/asm-i386/dwarf2.h index 5d1a8db5a9b0..6d66398a307d 100644 --- a/include/asm-i386/dwarf2.h +++ b/include/asm-i386/dwarf2.h @@ -28,6 +28,12 @@ #define CFI_RESTORE_STATE .cfi_restore_state #define CFI_UNDEFINED .cfi_undefined +#ifdef CONFIG_AS_CFI_SIGNAL_FRAME +#define CFI_SIGNAL_FRAME .cfi_signal_frame +#else +#define CFI_SIGNAL_FRAME +#endif + #else /* Due to the structure of pre-exisiting code, don't use assembler line @@ -48,6 +54,7 @@ #define CFI_REMEMBER_STATE ignore #define CFI_RESTORE_STATE ignore #define CFI_UNDEFINED ignore +#define CFI_SIGNAL_FRAME ignore #endif diff --git a/include/asm-i386/unwind.h b/include/asm-i386/unwind.h index f0ac399bae3c..5031d693b89d 100644 --- a/include/asm-i386/unwind.h +++ b/include/asm-i386/unwind.h @@ -18,6 +18,7 @@ struct unwind_frame_info { struct pt_regs regs; struct task_struct *task; + unsigned call_frame:1; }; #define UNW_PC(frame) (frame)->regs.eip @@ -44,6 +45,10 @@ struct unwind_frame_info PTREGS_INFO(edi), \ PTREGS_INFO(eip) +#define UNW_DEFAULT_RA(raItem, dataAlign) \ + ((raItem).where == Memory && \ + !((raItem).value * (dataAlign) + 4)) + static inline void arch_unw_init_frame_info(struct unwind_frame_info *info, /*const*/ struct pt_regs *regs) { diff --git a/include/asm-x86_64/dwarf2.h b/include/asm-x86_64/dwarf2.h index 2b9368365fad..eedc08526b0b 100644 --- a/include/asm-x86_64/dwarf2.h +++ b/include/asm-x86_64/dwarf2.h @@ -28,6 +28,11 @@ #define CFI_REMEMBER_STATE .cfi_remember_state #define CFI_RESTORE_STATE .cfi_restore_state #define CFI_UNDEFINED .cfi_undefined +#ifdef CONFIG_AS_CFI_SIGNAL_FRAME +#define CFI_SIGNAL_FRAME .cfi_signal_frame +#else +#define CFI_SIGNAL_FRAME +#endif #else @@ -45,6 +50,7 @@ #define CFI_REMEMBER_STATE # #define CFI_RESTORE_STATE # #define CFI_UNDEFINED # +#define CFI_SIGNAL_FRAME # #endif diff --git a/include/asm-x86_64/unwind.h b/include/asm-x86_64/unwind.h index 1f6e9bfb569e..b8fa5cb7ff88 100644 --- a/include/asm-x86_64/unwind.h +++ b/include/asm-x86_64/unwind.h @@ -18,6 +18,7 @@ struct unwind_frame_info { struct pt_regs regs; struct task_struct *task; + unsigned call_frame:1; }; #define UNW_PC(frame) (frame)->regs.rip @@ -57,6 +58,10 @@ struct unwind_frame_info PTREGS_INFO(r15), \ PTREGS_INFO(rip) +#define UNW_DEFAULT_RA(raItem, dataAlign) \ + ((raItem).where == Memory && \ + !((raItem).value * (dataAlign) + 8)) + static inline void arch_unw_init_frame_info(struct unwind_frame_info *info, /*const*/ struct pt_regs *regs) { diff --git a/kernel/unwind.c b/kernel/unwind.c index f69c804c8e62..3430475fcd88 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -603,6 +603,7 @@ int unwind(struct unwind_frame_info *frame) #define FRAME_REG(r, t) (((t *)frame)[reg_info[r].offs]) const u32 *fde = NULL, *cie = NULL; const u8 *ptr = NULL, *end = NULL; + unsigned long pc = UNW_PC(frame) - frame->call_frame; unsigned long startLoc = 0, endLoc = 0, cfa; unsigned i; signed ptrType = -1; @@ -612,7 +613,7 @@ int unwind(struct unwind_frame_info *frame) if (UNW_PC(frame) == 0) return -EINVAL; - if ((table = find_table(UNW_PC(frame))) != NULL + if ((table = find_table(pc)) != NULL && !(table->size & (sizeof(*fde) - 1))) { unsigned long tableSize = table->size; @@ -647,7 +648,7 @@ int unwind(struct unwind_frame_info *frame) ptrType & DW_EH_PE_indirect ? ptrType : ptrType & (DW_EH_PE_FORM|DW_EH_PE_signed)); - if (UNW_PC(frame) >= startLoc && UNW_PC(frame) < endLoc) + if (pc >= startLoc && pc < endLoc) break; cie = NULL; } @@ -657,16 +658,28 @@ int unwind(struct unwind_frame_info *frame) state.cieEnd = ptr; /* keep here temporarily */ ptr = (const u8 *)(cie + 2); end = (const u8 *)(cie + 1) + *cie; + frame->call_frame = 1; if ((state.version = *ptr) != 1) cie = NULL; /* unsupported version */ else if (*++ptr) { /* check if augmentation size is first (and thus present) */ if (*ptr == 'z') { - /* check for ignorable (or already handled) - * nul-terminated augmentation string */ - while (++ptr < end && *ptr) - if (strchr("LPR", *ptr) == NULL) + while (++ptr < end && *ptr) { + switch(*ptr) { + /* check for ignorable (or already handled) + * nul-terminated augmentation string */ + case 'L': + case 'P': + case 'R': + continue; + case 'S': + frame->call_frame = 0; + continue; + default: break; + } + break; + } } if (ptr >= end || *ptr) cie = NULL; @@ -755,7 +768,7 @@ int unwind(struct unwind_frame_info *frame) state.org = startLoc; memcpy(&state.cfa, &badCFA, sizeof(state.cfa)); /* process instructions */ - if (!processCFI(ptr, end, UNW_PC(frame), ptrType, &state) + if (!processCFI(ptr, end, pc, ptrType, &state) || state.loc > endLoc || state.regs[retAddrReg].where == Nowhere || state.cfa.reg >= ARRAY_SIZE(reg_info) @@ -763,6 +776,11 @@ int unwind(struct unwind_frame_info *frame) || state.cfa.offs % sizeof(unsigned long)) return -EIO; /* update frame */ +#ifndef CONFIG_AS_CFI_SIGNAL_FRAME + if(frame->call_frame + && !UNW_DEFAULT_RA(state.regs[retAddrReg], state.dataAlign)) + frame->call_frame = 0; +#endif cfa = FRAME_REG(state.cfa.reg, unsigned long) + state.cfa.offs; startLoc = min((unsigned long)UNW_SP(frame), cfa); endLoc = max((unsigned long)UNW_SP(frame), cfa); @@ -866,6 +884,7 @@ int unwind_init_frame_info(struct unwind_frame_info *info, /*const*/ struct pt_regs *regs) { info->task = tsk; + info->call_frame = 0; arch_unw_init_frame_info(info, regs); return 0; @@ -879,6 +898,7 @@ int unwind_init_blocked(struct unwind_frame_info *info, struct task_struct *tsk) { info->task = tsk; + info->call_frame = 0; arch_unw_init_blocked(info); return 0; @@ -894,6 +914,7 @@ int unwind_init_running(struct unwind_frame_info *info, void *arg) { info->task = current; + info->call_frame = 0; return arch_unwind_init_running(info, callback, arg); } diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include index 7adef12a0c26..4f5ff19b992b 100644 --- a/scripts/Kbuild.include +++ b/scripts/Kbuild.include @@ -66,8 +66,8 @@ as-option = $(shell if $(CC) $(CFLAGS) $(1) -Wa,-Z -c -o /dev/null \ # as-instr # Usage: cflags-y += $(call as-instr, instr, option1, option2) -as-instr = $(shell if echo -e "$(1)" | $(AS) -Z -o astest$$$$.out \ - 2>&1 >/dev/null ; then echo "$(2)"; else echo "$(3)"; fi; \ +as-instr = $(shell if echo -e "$(1)" | $(AS) >/dev/null 2>&1 -W -Z -o astest$$$$.out ; \ + then echo "$(2)"; else echo "$(3)"; fi; \ rm -f astest$$$$.out) # cc-option -- cgit v1.2.3 From 0a2966b48fb784e437520e400ddc94874ddbd4e8 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 25 Sep 2006 23:30:51 -0700 Subject: [PATCH] Fix longstanding load balancing bug in the scheduler The scheduler will stop load balancing if the most busy processor contains processes pinned via processor affinity. The scheduler currently only does one search for busiest cpu. If it cannot pull any tasks away from the busiest cpu because they were pinned then the scheduler goes into a corner and sulks leaving the idle processors idle. F.e. If you have processor 0 busy running four tasks pinned via taskset, there are none on processor 1 and one just started two processes on processor 2 then the scheduler will not move one of the two processes away from processor 2. This patch fixes that issue by forcing the scheduler to come out of its corner and retrying the load balancing by considering other processors for load balancing. This patch was originally developed by John Hawkes and discussed at http://marc.theaimsgroup.com/?l=linux-kernel&m=113901368523205&w=2. I have removed extraneous material and gone back to equipping struct rq with the cpu the queue is associated with since this makes the patch much easier and it is likely that others in the future will have the same difficulty of figuring out which processor owns which runqueue. The overhead added through these patches is a single word on the stack if the kernel is configured to support 32 cpus or less (32 bit). For 32 bit environments the maximum number of cpus that can be configued is 255 which would result in the use of 32 bytes additional on the stack. On IA64 up to 1k cpus can be configured which will result in the use of 128 additional bytes on the stack. The maximum additional cache footprint is one cacheline. Typically memory use will be much less than a cacheline and the additional cpumask will be placed on the stack in a cacheline that already contains other local variable. Signed-off-by: Christoph Lameter Cc: John Hawkes Cc: "Siddha, Suresh B" Cc: Ingo Molnar Cc: Nick Piggin Cc: Peter Williams Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/sched.c | 54 ++++++++++++++++++++++++++++++++++++++++++++++-------- 1 file changed, 46 insertions(+), 8 deletions(-) (limited to 'kernel') diff --git a/kernel/sched.c b/kernel/sched.c index a234fbee1238..5c848fd4e461 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -238,6 +238,7 @@ struct rq { /* For active balancing */ int active_balance; int push_cpu; + int cpu; /* cpu of this runqueue */ struct task_struct *migration_thread; struct list_head migration_queue; @@ -267,6 +268,15 @@ struct rq { static DEFINE_PER_CPU(struct rq, runqueues); +static inline int cpu_of(struct rq *rq) +{ +#ifdef CONFIG_SMP + return rq->cpu; +#else + return 0; +#endif +} + /* * The domain tree (rq->sd) is protected by RCU's quiescent state transition. * See detach_destroy_domains: synchronize_sched for details. @@ -2211,7 +2221,8 @@ out: */ static struct sched_group * find_busiest_group(struct sched_domain *sd, int this_cpu, - unsigned long *imbalance, enum idle_type idle, int *sd_idle) + unsigned long *imbalance, enum idle_type idle, int *sd_idle, + cpumask_t *cpus) { struct sched_group *busiest = NULL, *this = NULL, *group = sd->groups; unsigned long max_load, avg_load, total_load, this_load, total_pwr; @@ -2248,7 +2259,12 @@ find_busiest_group(struct sched_domain *sd, int this_cpu, sum_weighted_load = sum_nr_running = avg_load = 0; for_each_cpu_mask(i, group->cpumask) { - struct rq *rq = cpu_rq(i); + struct rq *rq; + + if (!cpu_isset(i, *cpus)) + continue; + + rq = cpu_rq(i); if (*sd_idle && !idle_cpu(i)) *sd_idle = 0; @@ -2466,13 +2482,17 @@ ret: */ static struct rq * find_busiest_queue(struct sched_group *group, enum idle_type idle, - unsigned long imbalance) + unsigned long imbalance, cpumask_t *cpus) { struct rq *busiest = NULL, *rq; unsigned long max_load = 0; int i; for_each_cpu_mask(i, group->cpumask) { + + if (!cpu_isset(i, *cpus)) + continue; + rq = cpu_rq(i); if (rq->nr_running == 1 && rq->raw_weighted_load > imbalance) @@ -2511,6 +2531,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct sched_group *group; unsigned long imbalance; struct rq *busiest; + cpumask_t cpus = CPU_MASK_ALL; if (idle != NOT_IDLE && sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) @@ -2518,13 +2539,15 @@ static int load_balance(int this_cpu, struct rq *this_rq, schedstat_inc(sd, lb_cnt[idle]); - group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle); +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, idle, &sd_idle, + &cpus); if (!group) { schedstat_inc(sd, lb_nobusyg[idle]); goto out_balanced; } - busiest = find_busiest_queue(group, idle, imbalance); + busiest = find_busiest_queue(group, idle, imbalance, &cpus); if (!busiest) { schedstat_inc(sd, lb_nobusyq[idle]); goto out_balanced; @@ -2549,8 +2572,12 @@ static int load_balance(int this_cpu, struct rq *this_rq, double_rq_unlock(this_rq, busiest); /* All tasks on this runqueue were pinned by CPU affinity */ - if (unlikely(all_pinned)) + if (unlikely(all_pinned)) { + cpu_clear(cpu_of(busiest), cpus); + if (!cpus_empty(cpus)) + goto redo; goto out_balanced; + } } if (!nr_moved) { @@ -2639,18 +2666,22 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) unsigned long imbalance; int nr_moved = 0; int sd_idle = 0; + cpumask_t cpus = CPU_MASK_ALL; if (sd->flags & SD_SHARE_CPUPOWER && !sched_smt_power_savings) sd_idle = 1; schedstat_inc(sd, lb_cnt[NEWLY_IDLE]); - group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, &sd_idle); +redo: + group = find_busiest_group(sd, this_cpu, &imbalance, NEWLY_IDLE, + &sd_idle, &cpus); if (!group) { schedstat_inc(sd, lb_nobusyg[NEWLY_IDLE]); goto out_balanced; } - busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance); + busiest = find_busiest_queue(group, NEWLY_IDLE, imbalance, + &cpus); if (!busiest) { schedstat_inc(sd, lb_nobusyq[NEWLY_IDLE]); goto out_balanced; @@ -2668,6 +2699,12 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd) minus_1_or_zero(busiest->nr_running), imbalance, sd, NEWLY_IDLE, NULL); spin_unlock(&busiest->lock); + + if (!nr_moved) { + cpu_clear(cpu_of(busiest), cpus); + if (!cpus_empty(cpus)) + goto redo; + } } if (!nr_moved) { @@ -6747,6 +6784,7 @@ void __init sched_init(void) rq->cpu_load[j] = 0; rq->active_balance = 0; rq->push_cpu = 0; + rq->cpu = i; rq->migration_thread = NULL; INIT_LIST_HEAD(&rq->migration_queue); #endif -- cgit v1.2.3 From 9b819d204cf602eab1a53a9ec4b8d2ca51e02a1d Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 25 Sep 2006 23:31:40 -0700 Subject: [PATCH] Add __GFP_THISNODE to avoid fallback to other nodes and ignore cpuset/memory policy restrictions Add a new gfp flag __GFP_THISNODE to avoid fallback to other nodes. This flag is essential if a kernel component requires memory to be located on a certain node. It will be needed for alloc_pages_node() to force allocation on the indicated node and for alloc_pages() to force allocation on the current node. Signed-off-by: Christoph Lameter Cc: Andy Whitcroft Cc: Mel Gorman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/gfp.h | 3 ++- kernel/cpuset.c | 2 +- mm/mempolicy.c | 2 +- mm/page_alloc.c | 3 +++ 4 files changed, 7 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/include/linux/gfp.h b/include/linux/gfp.h index 63ab88a36f39..0eda5b6dedbd 100644 --- a/include/linux/gfp.h +++ b/include/linux/gfp.h @@ -45,6 +45,7 @@ struct vm_area_struct; #define __GFP_ZERO ((__force gfp_t)0x8000u)/* Return zeroed page on success */ #define __GFP_NOMEMALLOC ((__force gfp_t)0x10000u) /* Don't use emergency reserves */ #define __GFP_HARDWALL ((__force gfp_t)0x20000u) /* Enforce hardwall cpuset memory allocs */ +#define __GFP_THISNODE ((__force gfp_t)0x40000u)/* No fallback, no policies */ #define __GFP_BITS_SHIFT 20 /* Room for 20 __GFP_FOO bits */ #define __GFP_BITS_MASK ((__force gfp_t)((1 << __GFP_BITS_SHIFT) - 1)) @@ -53,7 +54,7 @@ struct vm_area_struct; #define GFP_LEVEL_MASK (__GFP_WAIT|__GFP_HIGH|__GFP_IO|__GFP_FS| \ __GFP_COLD|__GFP_NOWARN|__GFP_REPEAT| \ __GFP_NOFAIL|__GFP_NORETRY|__GFP_NO_GROW|__GFP_COMP| \ - __GFP_NOMEMALLOC|__GFP_HARDWALL) + __GFP_NOMEMALLOC|__GFP_HARDWALL|__GFP_THISNODE) /* This equals 0, but use constants in case they ever change */ #define GFP_NOWAIT (GFP_ATOMIC & ~__GFP_HIGH) diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 4ea6f0dc2fc5..76940361273e 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2316,7 +2316,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) const struct cpuset *cs; /* current cpuset ancestors */ int allowed; /* is allocation in zone z allowed? */ - if (in_interrupt()) + if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; node = z->zone_pgdat->node_id; might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); diff --git a/mm/mempolicy.c b/mm/mempolicy.c index c3429a710ab1..8002e1faccda 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -1290,7 +1290,7 @@ struct page *alloc_pages_current(gfp_t gfp, unsigned order) if ((gfp & __GFP_WAIT) && !in_interrupt()) cpuset_update_task_memory_state(); - if (!pol || in_interrupt()) + if (!pol || in_interrupt() || (gfp & __GFP_THISNODE)) pol = &default_policy; if (pol->policy == MPOL_INTERLEAVE) return alloc_page_interleave(gfp, order, interleave_nodes(pol)); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 208a6b03aa78..ea498788af53 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -893,6 +893,9 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, * See also cpuset_zone_allowed() comment in kernel/cpuset.c. */ do { + if (unlikely((gfp_mask & __GFP_THISNODE) && + (*z)->zone_pgdat != zonelist->zones[0]->zone_pgdat)) + break; if ((alloc_flags & ALLOC_CPUSET) && !cpuset_zone_allowed(*z, gfp_mask)) continue; -- cgit v1.2.3 From fbd98167e653535c5816be154f2149c0efa7757d Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 25 Sep 2006 23:31:45 -0700 Subject: [PATCH] Profiling: require buffer allocation on the correct node Profiling really suffers with off node buffers. Fail if no memory is available on the nodes. The profiling code can deal with these failures should they occur. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/profile.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) (limited to 'kernel') diff --git a/kernel/profile.c b/kernel/profile.c index d5bd75e7501c..fb660c7d35ba 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -309,13 +309,17 @@ static int __devinit profile_cpu_callback(struct notifier_block *info, node = cpu_to_node(cpu); per_cpu(cpu_profile_flip, cpu) = 0; if (!per_cpu(cpu_profile_hits, cpu)[1]) { - page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0); + page = alloc_pages_node(node, + GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, + 0); if (!page) return NOTIFY_BAD; per_cpu(cpu_profile_hits, cpu)[1] = page_address(page); } if (!per_cpu(cpu_profile_hits, cpu)[0]) { - page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0); + page = alloc_pages_node(node, + GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, + 0); if (!page) goto out_free; per_cpu(cpu_profile_hits, cpu)[0] = page_address(page); @@ -491,12 +495,16 @@ static int __init create_hash_tables(void) int node = cpu_to_node(cpu); struct page *page; - page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0); + page = alloc_pages_node(node, + GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, + 0); if (!page) goto out_cleanup; per_cpu(cpu_profile_hits, cpu)[1] = (struct profile_hit *)page_address(page); - page = alloc_pages_node(node, GFP_KERNEL | __GFP_ZERO, 0); + page = alloc_pages_node(node, + GFP_KERNEL | __GFP_ZERO | GFP_THISNODE, + 0); if (!page) goto out_cleanup; per_cpu(cpu_profile_hits, cpu)[0] -- cgit v1.2.3 From 0ff38490c836dc379ff7ec45b10a15a662f4e5f6 Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 25 Sep 2006 23:31:52 -0700 Subject: [PATCH] zone_reclaim: dynamic slab reclaim Currently one can enable slab reclaim by setting an explicit option in /proc/sys/vm/zone_reclaim_mode. Slab reclaim is then used as a final option if the freeing of unmapped file backed pages is not enough to free enough pages to allow a local allocation. However, that means that the slab can grow excessively and that most memory of a node may be used by slabs. We have had a case where a machine with 46GB of memory was using 40-42GB for slab. Zone reclaim was effective in dealing with pagecache pages. However, slab reclaim was only done during global reclaim (which is a bit rare on NUMA systems). This patch implements slab reclaim during zone reclaim. Zone reclaim occurs if there is a danger of an off node allocation. At that point we 1. Shrink the per node page cache if the number of pagecache pages is more than min_unmapped_ratio percent of pages in a zone. 2. Shrink the slab cache if the number of the nodes reclaimable slab pages (patch depends on earlier one that implements that counter) are more than min_slab_ratio (a new /proc/sys/vm tunable). The shrinking of the slab cache is a bit problematic since it is not node specific. So we simply calculate what point in the slab we want to reach (current per node slab use minus the number of pages that neeed to be allocated) and then repeately run the global reclaim until that is unsuccessful or we have reached the limit. I hope we will have zone based slab reclaim at some point which will make that easier. The default for the min_slab_ratio is 5% Also remove the slab option from /proc/sys/vm/zone_reclaim_mode. [akpm@osdl.org: cleanups] Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/sysctl/vm.txt | 27 +++++++++++++++------ include/linux/mmzone.h | 3 +++ include/linux/swap.h | 1 + include/linux/sysctl.h | 1 + kernel/sysctl.c | 11 +++++++++ mm/page_alloc.c | 17 +++++++++++++ mm/vmscan.c | 58 +++++++++++++++++++++++++++++---------------- 7 files changed, 90 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt index 7cee90223d3a..20d0d797f539 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -29,6 +29,7 @@ Currently, these files are in /proc/sys/vm: - drop-caches - zone_reclaim_mode - min_unmapped_ratio +- min_slab_ratio - panic_on_oom ============================================================== @@ -138,7 +139,6 @@ This is value ORed together of 1 = Zone reclaim on 2 = Zone reclaim writes dirty pages out 4 = Zone reclaim swaps pages -8 = Also do a global slab reclaim pass zone_reclaim_mode is set during bootup to 1 if it is determined that pages from remote zones will cause a measurable performance reduction. The @@ -162,18 +162,13 @@ Allowing regular swap effectively restricts allocations to the local node unless explicitly overridden by memory policies or cpuset configurations. -It may be advisable to allow slab reclaim if the system makes heavy -use of files and builds up large slab caches. However, the slab -shrink operation is global, may take a long time and free slabs -in all nodes of the system. - ============================================================= min_unmapped_ratio: This is available only on NUMA kernels. -A percentage of the file backed pages in each zone. Zone reclaim will only +A percentage of the total pages in each zone. Zone reclaim will only occur if more than this percentage of pages are file backed and unmapped. This is to insure that a minimal amount of local pages is still available for file I/O even if the node is overallocated. @@ -182,6 +177,24 @@ The default is 1 percent. ============================================================= +min_slab_ratio: + +This is available only on NUMA kernels. + +A percentage of the total pages in each zone. On Zone reclaim +(fallback from the local zone occurs) slabs will be reclaimed if more +than this percentage of pages in a zone are reclaimable slab pages. +This insures that the slab growth stays under control even in NUMA +systems that rarely perform global reclaim. + +The default is 5 percent. + +Note that slab reclaim is triggered in a per zone / node fashion. +The process of reclaiming slab memory is currently not node specific +and may not be fast. + +============================================================= + panic_on_oom This enables or disables panic on out-of-memory feature. If this is set to 1, diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h index 08c41b9f92e0..3693f1a52788 100644 --- a/include/linux/mmzone.h +++ b/include/linux/mmzone.h @@ -171,6 +171,7 @@ struct zone { * zone reclaim becomes active if more unmapped pages exist. */ unsigned long min_unmapped_pages; + unsigned long min_slab_pages; struct per_cpu_pageset *pageset[NR_CPUS]; #else struct per_cpu_pageset pageset[NR_CPUS]; @@ -448,6 +449,8 @@ int percpu_pagelist_fraction_sysctl_handler(struct ctl_table *, int, struct file void __user *, size_t *, loff_t *); int sysctl_min_unmapped_ratio_sysctl_handler(struct ctl_table *, int, struct file *, void __user *, size_t *, loff_t *); +int sysctl_min_slab_ratio_sysctl_handler(struct ctl_table *, int, + struct file *, void __user *, size_t *, loff_t *); #include /* Returns the number of the current Node. */ diff --git a/include/linux/swap.h b/include/linux/swap.h index 32db06c8ffe0..a2f5ad7c2d2e 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -193,6 +193,7 @@ extern long vm_total_pages; #ifdef CONFIG_NUMA extern int zone_reclaim_mode; extern int sysctl_min_unmapped_ratio; +extern int sysctl_min_slab_ratio; extern int zone_reclaim(struct zone *, gfp_t, unsigned int); #else #define zone_reclaim_mode 0 diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h index 736ed917a4f8..eca555781d05 100644 --- a/include/linux/sysctl.h +++ b/include/linux/sysctl.h @@ -191,6 +191,7 @@ enum VM_MIN_UNMAPPED=32, /* Set min percent of unmapped pages */ VM_PANIC_ON_OOM=33, /* panic at out-of-memory */ VM_VDSO_ENABLED=34, /* map VDSO into new processes? */ + VM_MIN_SLAB=35, /* Percent pages ignored by zone reclaim */ }; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 362a0cc37138..fd43c3e6786b 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -943,6 +943,17 @@ static ctl_table vm_table[] = { .extra1 = &zero, .extra2 = &one_hundred, }, + { + .ctl_name = VM_MIN_SLAB, + .procname = "min_slab_ratio", + .data = &sysctl_min_slab_ratio, + .maxlen = sizeof(sysctl_min_slab_ratio), + .mode = 0644, + .proc_handler = &sysctl_min_slab_ratio_sysctl_handler, + .strategy = &sysctl_intvec, + .extra1 = &zero, + .extra2 = &one_hundred, + }, #endif #ifdef CONFIG_X86_32 { diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 47e98423b30d..cf913bdd433e 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2005,6 +2005,7 @@ static void __meminit free_area_init_core(struct pglist_data *pgdat, #ifdef CONFIG_NUMA zone->min_unmapped_pages = (realsize*sysctl_min_unmapped_ratio) / 100; + zone->min_slab_pages = (realsize * sysctl_min_slab_ratio) / 100; #endif zone->name = zone_names[j]; spin_lock_init(&zone->lock); @@ -2318,6 +2319,22 @@ int sysctl_min_unmapped_ratio_sysctl_handler(ctl_table *table, int write, sysctl_min_unmapped_ratio) / 100; return 0; } + +int sysctl_min_slab_ratio_sysctl_handler(ctl_table *table, int write, + struct file *file, void __user *buffer, size_t *length, loff_t *ppos) +{ + struct zone *zone; + int rc; + + rc = proc_dointvec_minmax(table, write, file, buffer, length, ppos); + if (rc) + return rc; + + for_each_zone(zone) + zone->min_slab_pages = (zone->present_pages * + sysctl_min_slab_ratio) / 100; + return 0; +} #endif /* diff --git a/mm/vmscan.c b/mm/vmscan.c index 349797ba4bac..089e943c4d38 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1527,7 +1527,6 @@ int zone_reclaim_mode __read_mostly; #define RECLAIM_ZONE (1<<0) /* Run shrink_cache on the zone */ #define RECLAIM_WRITE (1<<1) /* Writeout pages during reclaim */ #define RECLAIM_SWAP (1<<2) /* Swap pages out during reclaim */ -#define RECLAIM_SLAB (1<<3) /* Do a global slab shrink if the zone is out of memory */ /* * Priority for ZONE_RECLAIM. This determines the fraction of pages @@ -1542,6 +1541,12 @@ int zone_reclaim_mode __read_mostly; */ int sysctl_min_unmapped_ratio = 1; +/* + * If the number of slab pages in a zone grows beyond this percentage then + * slab reclaim needs to occur. + */ +int sysctl_min_slab_ratio = 5; + /* * Try to free up some pages from this zone through reclaim. */ @@ -1573,29 +1578,37 @@ static int __zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) reclaim_state.reclaimed_slab = 0; p->reclaim_state = &reclaim_state; - /* - * Free memory by calling shrink zone with increasing priorities - * until we have enough memory freed. - */ - priority = ZONE_RECLAIM_PRIORITY; - do { - nr_reclaimed += shrink_zone(priority, zone, &sc); - priority--; - } while (priority >= 0 && nr_reclaimed < nr_pages); + if (zone_page_state(zone, NR_FILE_PAGES) - + zone_page_state(zone, NR_FILE_MAPPED) > + zone->min_unmapped_pages) { + /* + * Free memory by calling shrink zone with increasing + * priorities until we have enough memory freed. + */ + priority = ZONE_RECLAIM_PRIORITY; + do { + nr_reclaimed += shrink_zone(priority, zone, &sc); + priority--; + } while (priority >= 0 && nr_reclaimed < nr_pages); + } - if (nr_reclaimed < nr_pages && (zone_reclaim_mode & RECLAIM_SLAB)) { + if (zone_page_state(zone, NR_SLAB_RECLAIMABLE) > zone->min_slab_pages) { /* * shrink_slab() does not currently allow us to determine how - * many pages were freed in this zone. So we just shake the slab - * a bit and then go off node for this particular allocation - * despite possibly having freed enough memory to allocate in - * this zone. If we freed local memory then the next - * allocations will be local again. + * many pages were freed in this zone. So we take the current + * number of slab pages and shake the slab until it is reduced + * by the same nr_pages that we used for reclaiming unmapped + * pages. * - * shrink_slab will free memory on all zones and may take - * a long time. + * Note that shrink_slab will free memory on all zones and may + * take a long time. */ - shrink_slab(sc.nr_scanned, gfp_mask, order); + unsigned long limit = zone_page_state(zone, + NR_SLAB_RECLAIMABLE) - nr_pages; + + while (shrink_slab(sc.nr_scanned, gfp_mask, order) && + zone_page_state(zone, NR_SLAB_RECLAIMABLE) > limit) + ; } p->reclaim_state = NULL; @@ -1609,7 +1622,8 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) int node_id; /* - * Zone reclaim reclaims unmapped file backed pages. + * Zone reclaim reclaims unmapped file backed pages and + * slab pages if we are over the defined limits. * * A small portion of unmapped file backed pages is needed for * file I/O otherwise pages read by file I/O will be immediately @@ -1618,7 +1632,9 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) * unmapped file backed pages. */ if (zone_page_state(zone, NR_FILE_PAGES) - - zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages) + zone_page_state(zone, NR_FILE_MAPPED) <= zone->min_unmapped_pages + && zone_page_state(zone, NR_SLAB_RECLAIMABLE) + <= zone->min_slab_pages) return 0; /* -- cgit v1.2.3 From 89fa30242facca249aead2aac03c4c69764f911c Mon Sep 17 00:00:00 2001 From: Christoph Lameter Date: Mon, 25 Sep 2006 23:31:55 -0700 Subject: [PATCH] NUMA: Add zone_to_nid function There are many places where we need to determine the node of a zone. Currently we use a difficult to read sequence of pointer dereferencing. Put that into an inline function and use throughout VM. Maybe we can find a way to optimize the lookup in the future. Signed-off-by: Christoph Lameter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/mm/discontig.c | 2 +- arch/parisc/mm/init.c | 2 +- include/linux/mm.h | 7 ++++++- kernel/cpuset.c | 4 ++-- mm/hugetlb.c | 2 +- mm/mempolicy.c | 6 +++--- mm/oom_kill.c | 3 +-- mm/page_alloc.c | 2 +- mm/vmscan.c | 2 +- 9 files changed, 17 insertions(+), 13 deletions(-) (limited to 'kernel') diff --git a/arch/i386/mm/discontig.c b/arch/i386/mm/discontig.c index 07c300f93764..fb5d8b747de4 100644 --- a/arch/i386/mm/discontig.c +++ b/arch/i386/mm/discontig.c @@ -422,7 +422,7 @@ void __init set_highmem_pages_init(int bad_ppro) zone_end_pfn = zone_start_pfn + zone->spanned_pages; printk("Initializing %s for node %d (%08lx:%08lx)\n", - zone->name, zone->zone_pgdat->node_id, + zone->name, zone_to_nid(zone), zone_start_pfn, zone_end_pfn); for (node_pfn = zone_start_pfn; node_pfn < zone_end_pfn; node_pfn++) { diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c index c7329615ef94..25ad28d63e88 100644 --- a/arch/parisc/mm/init.c +++ b/arch/parisc/mm/init.c @@ -551,7 +551,7 @@ void show_mem(void) printk("Zone list for zone %d on node %d: ", j, i); for (k = 0; zl->zones[k] != NULL; k++) - printk("[%d/%s] ", zl->zones[k]->zone_pgdat->node_id, zl->zones[k]->name); + printk("[%d/%s] ", zone_to_nid(zl->zones[k]), zl->zones[k]->name); printk("\n"); } } diff --git a/include/linux/mm.h b/include/linux/mm.h index f2018775b995..856f0ee7e84a 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -499,12 +499,17 @@ static inline struct zone *page_zone(struct page *page) return zone_table[page_zone_id(page)]; } +static inline unsigned long zone_to_nid(struct zone *zone) +{ + return zone->zone_pgdat->node_id; +} + static inline unsigned long page_to_nid(struct page *page) { if (FLAGS_HAS_NODE) return (page->flags >> NODES_PGSHIFT) & NODES_MASK; else - return page_zone(page)->zone_pgdat->node_id; + return zone_to_nid(page_zone(page)); } static inline unsigned long page_to_section(struct page *page) { diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 76940361273e..cff41511269f 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -2245,7 +2245,7 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl) int i; for (i = 0; zl->zones[i]; i++) { - int nid = zl->zones[i]->zone_pgdat->node_id; + int nid = zone_to_nid(zl->zones[i]); if (node_isset(nid, current->mems_allowed)) return 1; @@ -2318,7 +2318,7 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask) if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = z->zone_pgdat->node_id; + node = zone_to_nid(z); might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); if (node_isset(node, current->mems_allowed)) return 1; diff --git a/mm/hugetlb.c b/mm/hugetlb.c index 3aceadce1a76..7c7d03dbf73d 100644 --- a/mm/hugetlb.c +++ b/mm/hugetlb.c @@ -72,7 +72,7 @@ static struct page *dequeue_huge_page(struct vm_area_struct *vma, struct zone **z; for (z = zonelist->zones; *z; z++) { - nid = (*z)->zone_pgdat->node_id; + nid = zone_to_nid(*z); if (cpuset_zone_allowed(*z, GFP_HIGHUSER) && !list_empty(&hugepage_freelists[nid])) break; diff --git a/mm/mempolicy.c b/mm/mempolicy.c index 8002e1faccda..38f89650bc84 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -487,7 +487,7 @@ static void get_zonemask(struct mempolicy *p, nodemask_t *nodes) switch (p->policy) { case MPOL_BIND: for (i = 0; p->v.zonelist->zones[i]; i++) - node_set(p->v.zonelist->zones[i]->zone_pgdat->node_id, + node_set(zone_to_nid(p->v.zonelist->zones[i]), *nodes); break; case MPOL_DEFAULT: @@ -1145,7 +1145,7 @@ unsigned slab_node(struct mempolicy *policy) * Follow bind policy behavior and start allocation at the * first node. */ - return policy->v.zonelist->zones[0]->zone_pgdat->node_id; + return zone_to_nid(policy->v.zonelist->zones[0]); case MPOL_PREFERRED: if (policy->v.preferred_node >= 0) @@ -1649,7 +1649,7 @@ void mpol_rebind_policy(struct mempolicy *pol, const nodemask_t *newmask) nodes_clear(nodes); for (z = pol->v.zonelist->zones; *z; z++) - node_set((*z)->zone_pgdat->node_id, nodes); + node_set(zone_to_nid(*z), nodes); nodes_remap(tmp, nodes, *mpolmask, *newmask); nodes = tmp; diff --git a/mm/oom_kill.c b/mm/oom_kill.c index f1c0ef1fd21f..bada3d03119f 100644 --- a/mm/oom_kill.c +++ b/mm/oom_kill.c @@ -177,8 +177,7 @@ static inline int constrained_alloc(struct zonelist *zonelist, gfp_t gfp_mask) for (z = zonelist->zones; *z; z++) if (cpuset_zone_allowed(*z, gfp_mask)) - node_clear((*z)->zone_pgdat->node_id, - nodes); + node_clear(zone_to_nid(*z), nodes); else return CONSTRAINT_CPUSET; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index cf913bdd433e..51070b6d593f 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -1217,7 +1217,7 @@ unsigned int nr_free_pagecache_pages(void) #ifdef CONFIG_NUMA static void show_node(struct zone *zone) { - printk("Node %d ", zone->zone_pgdat->node_id); + printk("Node %ld ", zone_to_nid(zone)); } #else #define show_node(zone) do { } while (0) diff --git a/mm/vmscan.c b/mm/vmscan.c index b950f193816e..87779dda4ec6 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -1661,7 +1661,7 @@ int zone_reclaim(struct zone *zone, gfp_t gfp_mask, unsigned int order) * over remote processors and spread off node memory allocations * as wide as possible. */ - node_id = zone->zone_pgdat->node_id; + node_id = zone_to_nid(zone); mask = node_to_cpumask(node_id); if (!cpus_empty(mask) && node_id != numa_node_id()) return 0; -- cgit v1.2.3 From 62bac0185ad3dfef11d9602980445c54d45199c6 Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Mon, 25 Sep 2006 23:31:56 -0700 Subject: [PATCH] selinux: eliminate selinux_task_ctxid Eliminate selinux_task_ctxid since it duplicates selinux_task_get_sid. Signed-off-by: Stephen Smalley Acked-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/selinux.h | 15 --------------- kernel/auditsc.c | 2 +- security/selinux/exports.c | 9 --------- 3 files changed, 1 insertion(+), 25 deletions(-) (limited to 'kernel') diff --git a/include/linux/selinux.h b/include/linux/selinux.h index aad4e390d6a5..79e4707ca772 100644 --- a/include/linux/selinux.h +++ b/include/linux/selinux.h @@ -69,16 +69,6 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op, */ void selinux_audit_set_callback(int (*callback)(void)); -/** - * selinux_task_ctxid - determine a context ID for a process. - * @tsk: the task object - * @ctxid: ID value returned via this - * - * On return, ctxid will contain an ID for the context. This value - * should only be used opaquely. - */ -void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid); - /** * selinux_ctxid_to_string - map a security context ID to a string * @ctxid: security context ID to be converted. @@ -166,11 +156,6 @@ static inline void selinux_audit_set_callback(int (*callback)(void)) return; } -static inline void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid) -{ - *ctxid = 0; -} - static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen) { *ctx = NULL; diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 1bd8827a0102..331e17010393 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -385,7 +385,7 @@ static int audit_filter_rules(struct task_struct *tsk, logged upon error */ if (f->se_rule) { if (need_sid) { - selinux_task_ctxid(tsk, &sid); + selinux_get_task_sid(tsk, &sid); need_sid = 0; } result = selinux_audit_rule_match(sid, f->type, diff --git a/security/selinux/exports.c b/security/selinux/exports.c index 9d7737db5e51..ee0fb47f81ae 100644 --- a/security/selinux/exports.c +++ b/security/selinux/exports.c @@ -21,15 +21,6 @@ #include "security.h" #include "objsec.h" -void selinux_task_ctxid(struct task_struct *tsk, u32 *ctxid) -{ - struct task_security_struct *tsec = tsk->security; - if (selinux_enabled) - *ctxid = tsec->sid; - else - *ctxid = 0; -} - int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen) { if (selinux_enabled) -- cgit v1.2.3 From 1a70cd40cb291c25b67ec0da715a49d76719329d Mon Sep 17 00:00:00 2001 From: Stephen Smalley Date: Mon, 25 Sep 2006 23:31:57 -0700 Subject: [PATCH] selinux: rename selinux_ctxid_to_string Rename selinux_ctxid_to_string to selinux_sid_to_string to be consistent with other interfaces. Signed-off-by: Stephen Smalley Acked-by: James Morris Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/selinux.h | 8 ++++---- kernel/audit.c | 14 +++++++------- kernel/auditfilter.c | 2 +- kernel/auditsc.c | 4 ++-- security/selinux/exports.c | 4 ++-- 5 files changed, 16 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/include/linux/selinux.h b/include/linux/selinux.h index 79e4707ca772..df9098de4c99 100644 --- a/include/linux/selinux.h +++ b/include/linux/selinux.h @@ -70,8 +70,8 @@ int selinux_audit_rule_match(u32 ctxid, u32 field, u32 op, void selinux_audit_set_callback(int (*callback)(void)); /** - * selinux_ctxid_to_string - map a security context ID to a string - * @ctxid: security context ID to be converted. + * selinux_sid_to_string - map a security context ID to a string + * @sid: security context ID to be converted. * @ctx: address of context string to be returned * @ctxlen: length of returned context string. * @@ -79,7 +79,7 @@ void selinux_audit_set_callback(int (*callback)(void)); * string will be allocated internally, and the caller must call * kfree() on it after use. */ -int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen); +int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen); /** * selinux_get_inode_sid - get the inode's security context ID @@ -156,7 +156,7 @@ static inline void selinux_audit_set_callback(int (*callback)(void)) return; } -static inline int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen) +static inline int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen) { *ctx = NULL; *ctxlen = 0; diff --git a/kernel/audit.c b/kernel/audit.c index 963fd15c9621..f9889ee77825 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -244,7 +244,7 @@ static int audit_set_rate_limit(int limit, uid_t loginuid, u32 sid) char *ctx = NULL; u32 len; int rc; - if ((rc = selinux_ctxid_to_string(sid, &ctx, &len))) + if ((rc = selinux_sid_to_string(sid, &ctx, &len))) return rc; else audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE, @@ -267,7 +267,7 @@ static int audit_set_backlog_limit(int limit, uid_t loginuid, u32 sid) char *ctx = NULL; u32 len; int rc; - if ((rc = selinux_ctxid_to_string(sid, &ctx, &len))) + if ((rc = selinux_sid_to_string(sid, &ctx, &len))) return rc; else audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE, @@ -293,7 +293,7 @@ static int audit_set_enabled(int state, uid_t loginuid, u32 sid) char *ctx = NULL; u32 len; int rc; - if ((rc = selinux_ctxid_to_string(sid, &ctx, &len))) + if ((rc = selinux_sid_to_string(sid, &ctx, &len))) return rc; else audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE, @@ -321,7 +321,7 @@ static int audit_set_failure(int state, uid_t loginuid, u32 sid) char *ctx = NULL; u32 len; int rc; - if ((rc = selinux_ctxid_to_string(sid, &ctx, &len))) + if ((rc = selinux_sid_to_string(sid, &ctx, &len))) return rc; else audit_log(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE, @@ -538,7 +538,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (status_get->mask & AUDIT_STATUS_PID) { int old = audit_pid; if (sid) { - if ((err = selinux_ctxid_to_string( + if ((err = selinux_sid_to_string( sid, &ctx, &len))) return err; else @@ -576,7 +576,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) "user pid=%d uid=%u auid=%u", pid, uid, loginuid); if (sid) { - if (selinux_ctxid_to_string( + if (selinux_sid_to_string( sid, &ctx, &len)) { audit_log_format(ab, " ssid=%u", sid); @@ -614,7 +614,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) loginuid, sid); break; case AUDIT_SIGNAL_INFO: - err = selinux_ctxid_to_string(audit_sig_sid, &ctx, &len); + err = selinux_sid_to_string(audit_sig_sid, &ctx, &len); if (err) return err; sig_data = kmalloc(sizeof(*sig_data) + len, GFP_KERNEL); diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index a44879b0c72f..1a58a81fb09d 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -1398,7 +1398,7 @@ static void audit_log_rule_change(uid_t loginuid, u32 sid, char *action, if (sid) { char *ctx = NULL; u32 len; - if (selinux_ctxid_to_string(sid, &ctx, &len)) + if (selinux_sid_to_string(sid, &ctx, &len)) audit_log_format(ab, " ssid=%u", sid); else audit_log_format(ab, " subj=%s", ctx); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 331e17010393..fb83c5cb8c32 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -898,7 +898,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts if (axi->osid != 0) { char *ctx = NULL; u32 len; - if (selinux_ctxid_to_string( + if (selinux_sid_to_string( axi->osid, &ctx, &len)) { audit_log_format(ab, " osid=%u", axi->osid); @@ -1005,7 +1005,7 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts if (n->osid != 0) { char *ctx = NULL; u32 len; - if (selinux_ctxid_to_string( + if (selinux_sid_to_string( n->osid, &ctx, &len)) { audit_log_format(ab, " osid=%u", n->osid); call_panic = 2; diff --git a/security/selinux/exports.c b/security/selinux/exports.c index ee0fb47f81ae..b6f96943be1f 100644 --- a/security/selinux/exports.c +++ b/security/selinux/exports.c @@ -21,10 +21,10 @@ #include "security.h" #include "objsec.h" -int selinux_ctxid_to_string(u32 ctxid, char **ctx, u32 *ctxlen) +int selinux_sid_to_string(u32 sid, char **ctx, u32 *ctxlen) { if (selinux_enabled) - return security_sid_to_context(ctxid, ctx, ctxlen); + return security_sid_to_context(sid, ctx, ctxlen); else { *ctx = NULL; *ctxlen = 0; -- cgit v1.2.3 From af8c65b57aaa4ae321af34dbfc5ca7f5625263fe Mon Sep 17 00:00:00 2001 From: David Howells Date: Mon, 25 Sep 2006 23:32:07 -0700 Subject: [PATCH] FRV: permit __do_IRQ() to be dispensed with Permit __do_IRQ() to be dispensed with based on a configuration option. Signed-off-by: David Howells Cc: Benjamin Herrenschmidt Cc: Thomas Gleixner Cc: Ingo Molnar Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/frv/Kconfig | 4 ++++ include/linux/irq.h | 6 ++++++ kernel/irq/handle.c | 2 ++ 3 files changed, 12 insertions(+) (limited to 'kernel') diff --git a/arch/frv/Kconfig b/arch/frv/Kconfig index 5833808fb823..f7b171b92ea2 100644 --- a/arch/frv/Kconfig +++ b/arch/frv/Kconfig @@ -29,6 +29,10 @@ config GENERIC_HARDIRQS bool default y +config GENERIC_HARDIRQS_NO__DO_IRQ + bool + default y + config GENERIC_TIME bool default y diff --git a/include/linux/irq.h b/include/linux/irq.h index fbf6d901e9c2..48d3cb3b6a47 100644 --- a/include/linux/irq.h +++ b/include/linux/irq.h @@ -320,7 +320,9 @@ handle_irq_name(void fastcall (*handle)(unsigned int, struct irq_desc *, * Monolithic do_IRQ implementation. * (is an explicit fastcall, because i386 4KSTACKS calls it from assembly) */ +#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs); +#endif /* * Architectures call this to let the generic IRQ layer @@ -332,10 +334,14 @@ static inline void generic_handle_irq(unsigned int irq, struct pt_regs *regs) { struct irq_desc *desc = irq_desc + irq; +#ifdef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ + desc->handle_irq(irq, desc, regs); +#else if (likely(desc->handle_irq)) desc->handle_irq(irq, desc, regs); else __do_IRQ(irq, regs); +#endif } /* Handling of unhandled and spurious interrupts: */ diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index 48a53f68af96..4c6cdbaed661 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -154,6 +154,7 @@ irqreturn_t handle_IRQ_event(unsigned int irq, struct pt_regs *regs, return retval; } +#ifndef CONFIG_GENERIC_HARDIRQS_NO__DO_IRQ /** * __do_IRQ - original all in one highlevel IRQ handler * @irq: the interrupt number @@ -253,6 +254,7 @@ out: return 1; } +#endif #ifdef CONFIG_TRACE_IRQFLAGS -- cgit v1.2.3 From 3a4f7577c9ef393ca80c783f02ffbc125de771c7 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 25 Sep 2006 23:32:41 -0700 Subject: [PATCH] swsusp: add write-speed instrumentation Add some instrumentation to the swsusp writeout code to show what bandwidth we're achieving. Cc: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/swap.c | 32 +++++++++++++++++++++++++++++--- 1 file changed, 29 insertions(+), 3 deletions(-) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index f1dd146bd64d..79b66e734bdb 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -146,6 +146,26 @@ static void release_swap_writer(struct swap_map_handle *handle) handle->bitmap = NULL; } +static void show_speed(struct timeval *start, struct timeval *stop, + unsigned nr_pages, char *msg) +{ + s64 elapsed_centisecs64; + int centisecs; + int k; + int kps; + + elapsed_centisecs64 = timeval_to_ns(stop) - timeval_to_ns(start); + do_div(elapsed_centisecs64, NSEC_PER_SEC / 100); + centisecs = elapsed_centisecs64; + if (centisecs == 0) + centisecs = 1; /* avoid div-by-zero */ + k = nr_pages * (PAGE_SIZE / 1024); + kps = (k * 100) / centisecs; + printk("%s %d kbytes in %d.%02d seconds (%d.%02d MB/s)\n", msg, k, + centisecs / 100, centisecs % 100, + kps / 1000, (kps % 1000) / 10); +} + static int get_swap_writer(struct swap_map_handle *handle) { handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL); @@ -206,17 +226,21 @@ static int flush_swap_writer(struct swap_map_handle *handle) static int save_image(struct swap_map_handle *handle, struct snapshot_handle *snapshot, - unsigned int nr_pages) + unsigned int nr_to_write) { unsigned int m; int ret; int error = 0; + int nr_pages; + struct timeval start; + struct timeval stop; - printk("Saving image data pages (%u pages) ... ", nr_pages); - m = nr_pages / 100; + printk("Saving image data pages (%u pages) ... ", nr_to_write); + m = nr_to_write / 100; if (!m) m = 1; nr_pages = 0; + do_gettimeofday(&start); do { ret = snapshot_read_next(snapshot, PAGE_SIZE); if (ret > 0) { @@ -228,8 +252,10 @@ static int save_image(struct swap_map_handle *handle, nr_pages++; } } while (ret > 0); + do_gettimeofday(&stop); if (!error) printk("\b\b\b\bdone\n"); + show_speed(&start, &stop, nr_to_write, "Wrote"); return error; } -- cgit v1.2.3 From ab954160350c91c77ae03740ef90458c3ad5412c Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 25 Sep 2006 23:32:42 -0700 Subject: [PATCH] swsusp: write speedup Switch the swsusp writeout code from 4k-at-a-time to 4MB-at-a-time. Crufty old PIII testbox: 12.9 MB/s -> 20.9 MB/s Sony Vaio: 14.7 MB/s -> 26.5 MB/s The implementation is crude. A better one would use larger BIOs, but wouldn't gain any performance. The memcpys will be mostly pipelined with the IO and basically come for free. The ENOMEM path has not been tested. It should be. Cc: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 5 ++- kernel/power/swap.c | 93 ++++++++++++++++++++++++++++++++++++++++++---------- mm/page_io.c | 25 ++++++++++---- 3 files changed, 98 insertions(+), 25 deletions(-) (limited to 'kernel') diff --git a/include/linux/swap.h b/include/linux/swap.h index a2f5ad7c2d2e..3d434cbffe26 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -12,6 +12,8 @@ struct notifier_block; +struct bio; + #define SWAP_FLAG_PREFER 0x8000 /* set if swap priority specified */ #define SWAP_FLAG_PRIO_MASK 0x7fff #define SWAP_FLAG_PRIO_SHIFT 0 @@ -216,7 +218,8 @@ extern void swap_unplug_io_fn(struct backing_dev_info *, struct page *); /* linux/mm/page_io.c */ extern int swap_readpage(struct file *, struct page *); extern int swap_writepage(struct page *page, struct writeback_control *wbc); -extern int rw_swap_page_sync(int, swp_entry_t, struct page *); +extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page, + struct bio **bio_chain); /* linux/mm/swap_state.c */ extern struct address_space swapper_space; diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 79b66e734bdb..2dc883d361d5 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -49,18 +49,16 @@ static int mark_swapfiles(swp_entry_t start) { int error; - rw_swap_page_sync(READ, - swp_entry(root_swap, 0), - virt_to_page((unsigned long)&swsusp_header)); + rw_swap_page_sync(READ, swp_entry(root_swap, 0), + virt_to_page((unsigned long)&swsusp_header), NULL); if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) || !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) { memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10); memcpy(swsusp_header.sig,SWSUSP_SIG, 10); swsusp_header.image = start; - error = rw_swap_page_sync(WRITE, - swp_entry(root_swap, 0), - virt_to_page((unsigned long) - &swsusp_header)); + error = rw_swap_page_sync(WRITE, swp_entry(root_swap, 0), + virt_to_page((unsigned long)&swsusp_header), + NULL); } else { pr_debug("swsusp: Partition is not swap space.\n"); error = -ENODEV; @@ -88,16 +86,37 @@ static int swsusp_swap_check(void) /* This is called before saving image */ * write_page - Write one page to given swap location. * @buf: Address we're writing. * @offset: Offset of the swap page we're writing to. + * @bio_chain: Link the next write BIO here */ -static int write_page(void *buf, unsigned long offset) +static int write_page(void *buf, unsigned long offset, struct bio **bio_chain) { swp_entry_t entry; int error = -ENOSPC; if (offset) { + struct page *page = virt_to_page(buf); + + if (bio_chain) { + /* + * Whether or not we successfully allocated a copy page, + * we take a ref on the page here. It gets undone in + * wait_on_bio_chain(). + */ + struct page *page_copy; + page_copy = alloc_page(GFP_ATOMIC); + if (page_copy == NULL) { + WARN_ON_ONCE(1); + bio_chain = NULL; /* Go synchronous */ + get_page(page); + } else { + memcpy(page_address(page_copy), + page_address(page), PAGE_SIZE); + page = page_copy; + } + } entry = swp_entry(root_swap, offset); - error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf)); + error = rw_swap_page_sync(WRITE, entry, page, bio_chain); } return error; } @@ -185,37 +204,68 @@ static int get_swap_writer(struct swap_map_handle *handle) return 0; } -static int swap_write_page(struct swap_map_handle *handle, void *buf) +static int wait_on_bio_chain(struct bio **bio_chain) { - int error; + struct bio *bio; + struct bio *next_bio; + int ret = 0; + + if (bio_chain == NULL) + return 0; + + bio = *bio_chain; + while (bio) { + struct page *page; + + next_bio = bio->bi_private; + page = bio->bi_io_vec[0].bv_page; + wait_on_page_locked(page); + if (!PageUptodate(page) || PageError(page)) + ret = -EIO; + put_page(page); + bio_put(bio); + bio = next_bio; + } + *bio_chain = NULL; + return ret; +} + +static int swap_write_page(struct swap_map_handle *handle, void *buf, + struct bio **bio_chain) +{ + int error = 0; unsigned long offset; if (!handle->cur) return -EINVAL; offset = alloc_swap_page(root_swap, handle->bitmap); - error = write_page(buf, offset); + error = write_page(buf, offset, bio_chain); if (error) return error; handle->cur->entries[handle->k++] = offset; if (handle->k >= MAP_PAGE_ENTRIES) { + error = wait_on_bio_chain(bio_chain); + if (error) + goto out; offset = alloc_swap_page(root_swap, handle->bitmap); if (!offset) return -ENOSPC; handle->cur->next_swap = offset; - error = write_page(handle->cur, handle->cur_swap); + error = write_page(handle->cur, handle->cur_swap, NULL); if (error) - return error; + goto out; memset(handle->cur, 0, PAGE_SIZE); handle->cur_swap = offset; handle->k = 0; } - return 0; +out: + return error; } static int flush_swap_writer(struct swap_map_handle *handle) { if (handle->cur && handle->cur_swap) - return write_page(handle->cur, handle->cur_swap); + return write_page(handle->cur, handle->cur_swap, NULL); else return -EINVAL; } @@ -232,6 +282,8 @@ static int save_image(struct swap_map_handle *handle, int ret; int error = 0; int nr_pages; + int err2; + struct bio *bio; struct timeval start; struct timeval stop; @@ -240,11 +292,13 @@ static int save_image(struct swap_map_handle *handle, if (!m) m = 1; nr_pages = 0; + bio = NULL; do_gettimeofday(&start); do { ret = snapshot_read_next(snapshot, PAGE_SIZE); if (ret > 0) { - error = swap_write_page(handle, data_of(*snapshot)); + error = swap_write_page(handle, data_of(*snapshot), + &bio); if (error) break; if (!(nr_pages % m)) @@ -252,7 +306,10 @@ static int save_image(struct swap_map_handle *handle, nr_pages++; } } while (ret > 0); + err2 = wait_on_bio_chain(&bio); do_gettimeofday(&stop); + if (!error) + error = err2; if (!error) printk("\b\b\b\bdone\n"); show_speed(&start, &stop, nr_to_write, "Wrote"); @@ -307,7 +364,7 @@ int swsusp_write(void) error = get_swap_writer(&handle); if (!error) { unsigned long start = handle.cur_swap; - error = swap_write_page(&handle, header); + error = swap_write_page(&handle, header, NULL); if (!error) error = save_image(&handle, &snapshot, header->pages - 1); diff --git a/mm/page_io.c b/mm/page_io.c index d2f0a5783370..f46a9862b7ef 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -156,10 +156,12 @@ out: * We use end_swap_bio_read() even for writes, because it happens to do what * we want. */ -int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page) +int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page, + struct bio **bio_chain) { struct bio *bio; int ret = 0; + int bio_rw; lock_page(page); @@ -170,11 +172,22 @@ int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page) goto out; } - submit_bio(rw | (1 << BIO_RW_SYNC), bio); - wait_on_page_locked(page); - - if (!PageUptodate(page) || PageError(page)) - ret = -EIO; + bio_rw = rw; + if (!bio_chain) + bio_rw |= (1 << BIO_RW_SYNC); + if (bio_chain) + bio_get(bio); + submit_bio(bio_rw, bio); + if (bio_chain == NULL) { + wait_on_page_locked(page); + + if (!PageUptodate(page) || PageError(page)) + ret = -EIO; + } + if (bio_chain) { + bio->bi_private = *bio_chain; + *bio_chain = bio; + } out: return ret; } -- cgit v1.2.3 From 8c002494b55119a3fd1dddee83b4fb75cfda47e5 Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 25 Sep 2006 23:32:43 -0700 Subject: [PATCH] swsusp: add read-speed instrumentation Add some instrumentation to the swsusp readin code to show what bandwidth we're achieving. Cc: Pavel Machek Cc: "Rafael J. Wysocki" Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/swap.c | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 2dc883d361d5..9ab989572164 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -522,12 +522,15 @@ static int load_image(struct swap_map_handle *handle, unsigned int m; int ret; int error = 0; + struct timeval start; + struct timeval stop; printk("Loading image data pages (%u pages) ... ", nr_pages); m = nr_pages / 100; if (!m) m = 1; nr_pages = 0; + do_gettimeofday(&start); do { ret = snapshot_write_next(snapshot, PAGE_SIZE); if (ret > 0) { @@ -539,11 +542,13 @@ static int load_image(struct swap_map_handle *handle, nr_pages++; } } while (ret > 0); + do_gettimeofday(&stop); if (!error) { printk("\b\b\b\bdone\n"); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; } + show_speed(&start, &stop, nr_pages, "Read"); return error; } -- cgit v1.2.3 From 546e0d271941dd1ff6961e2a1f7eac75f1fc277e Mon Sep 17 00:00:00 2001 From: Andrew Morton Date: Mon, 25 Sep 2006 23:32:44 -0700 Subject: [PATCH] swsusp: read speedup Implement async reads for swsusp resuming. Crufty old PIII testbox: 15.7 MB/s -> 20.3 MB/s Sony Vaio: 14.6 MB/s -> 33.3 MB/s I didn't implement the post-resume bio_set_pages_dirty(). I don't really understand why resume needs to run set_page_dirty() against these pages. It might be a worry that this code modifies PG_Uptodate, PG_Error and PG_Locked against the image pages. Can this possibly affect the resumed-into kernel? Hopefully not, if we're atomically restoring its mem_map? Cc: Pavel Machek Cc: "Rafael J. Wysocki" Cc: Jens Axboe Cc: Laurent Riffard Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/swap.h | 1 + kernel/power/power.h | 1 + kernel/power/snapshot.c | 11 ++-- kernel/power/swap.c | 138 ++++++++++++++++++++++++------------------------ mm/page_io.c | 2 +- 5 files changed, 81 insertions(+), 72 deletions(-) (limited to 'kernel') diff --git a/include/linux/swap.h b/include/linux/swap.h index 3d434cbffe26..e7c36ba2a2db 100644 --- a/include/linux/swap.h +++ b/include/linux/swap.h @@ -220,6 +220,7 @@ extern int swap_readpage(struct file *, struct page *); extern int swap_writepage(struct page *page, struct writeback_control *wbc); extern int rw_swap_page_sync(int rw, swp_entry_t entry, struct page *page, struct bio **bio_chain); +extern int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err); /* linux/mm/swap_state.c */ extern struct address_space swapper_space; diff --git a/kernel/power/power.h b/kernel/power/power.h index 57a792982fb9..59ce712f082d 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -58,6 +58,7 @@ struct snapshot_handle { struct pbe *pbe, *last_pbe; void *buffer; unsigned int buf_offset; + int sync_read; }; #define data_of(handle) ((handle).buffer + (handle).buf_offset) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 75d4886e648e..591301ae8b7d 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -314,7 +314,7 @@ static unsigned int unsafe_pages; * and we count them using unsafe_pages */ -static inline void *alloc_image_page(gfp_t gfp_mask, int safe_needed) +static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) { void *res; @@ -828,13 +828,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) } if (!handle->offset) handle->buffer = buffer; + handle->sync_read = 1; if (handle->prev < handle->page) { if (!handle->prev) { - error = load_header(handle, (struct swsusp_info *)buffer); + error = load_header(handle, + (struct swsusp_info *)buffer); if (error) return error; } else if (handle->prev <= nr_meta_pages) { - handle->pbe = unpack_orig_addresses(buffer, handle->pbe); + handle->pbe = unpack_orig_addresses(buffer, + handle->pbe); if (!handle->pbe) { error = prepare_image(handle); if (error) @@ -842,10 +845,12 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) handle->pbe = pagedir_nosave; handle->last_pbe = NULL; handle->buffer = get_buffer(handle); + handle->sync_read = 0; } } else { handle->pbe = handle->pbe->next; handle->buffer = get_buffer(handle); + handle->sync_read = 0; } handle->prev = handle->page; } diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 9ab989572164..8309d20b2563 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -22,6 +22,7 @@ #include #include #include +#include #include #include #include @@ -214,6 +215,8 @@ static int wait_on_bio_chain(struct bio **bio_chain) return 0; bio = *bio_chain; + if (bio == NULL) + return 0; while (bio) { struct page *page; @@ -349,7 +352,8 @@ int swsusp_write(void) int error; if ((error = swsusp_swap_check())) { - printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n"); + printk(KERN_ERR "swsusp: Cannot find swap device, try " + "swapon -a.\n"); return error; } memset(&snapshot, 0, sizeof(struct snapshot_handle)); @@ -381,27 +385,6 @@ int swsusp_write(void) return error; } -/* - * Using bio to read from swap. - * This code requires a bit more work than just using buffer heads - * but, it is the recommended way for 2.5/2.6. - * The following are to signal the beginning and end of I/O. Bios - * finish asynchronously, while we want them to happen synchronously. - * A simple atomic_t, and a wait loop take care of this problem. - */ - -static atomic_t io_done = ATOMIC_INIT(0); - -static int end_io(struct bio *bio, unsigned int num, int err) -{ - if (!test_bit(BIO_UPTODATE, &bio->bi_flags)) { - printk(KERN_ERR "I/O error reading swsusp image.\n"); - return -EIO; - } - atomic_set(&io_done, 0); - return 0; -} - static struct block_device *resume_bdev; /** @@ -409,15 +392,15 @@ static struct block_device *resume_bdev; * @rw: READ or WRITE. * @off physical offset of page. * @page: page we're reading or writing. + * @bio_chain: list of pending biod (for async reading) * * Straight from the textbook - allocate and initialize the bio. - * If we're writing, make sure the page is marked as dirty. - * Then submit it and wait. + * If we're reading, make sure the page is marked as dirty. + * Then submit it and, if @bio_chain == NULL, wait. */ - -static int submit(int rw, pgoff_t page_off, void *page) +static int submit(int rw, pgoff_t page_off, struct page *page, + struct bio **bio_chain) { - int error = 0; struct bio *bio; bio = bio_alloc(GFP_ATOMIC, 1); @@ -425,33 +408,40 @@ static int submit(int rw, pgoff_t page_off, void *page) return -ENOMEM; bio->bi_sector = page_off * (PAGE_SIZE >> 9); bio->bi_bdev = resume_bdev; - bio->bi_end_io = end_io; + bio->bi_end_io = end_swap_bio_read; - if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) { - printk("swsusp: ERROR: adding page to bio at %ld\n",page_off); - error = -EFAULT; - goto Done; + if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) { + printk("swsusp: ERROR: adding page to bio at %ld\n", page_off); + bio_put(bio); + return -EFAULT; } - atomic_set(&io_done, 1); - submit_bio(rw | (1 << BIO_RW_SYNC), bio); - while (atomic_read(&io_done)) - yield(); - if (rw == READ) - bio_set_pages_dirty(bio); - Done: - bio_put(bio); - return error; + lock_page(page); + bio_get(bio); + + if (bio_chain == NULL) { + submit_bio(rw | (1 << BIO_RW_SYNC), bio); + wait_on_page_locked(page); + if (rw == READ) + bio_set_pages_dirty(bio); + bio_put(bio); + } else { + get_page(page); + bio->bi_private = *bio_chain; + *bio_chain = bio; + submit_bio(rw | (1 << BIO_RW_SYNC), bio); + } + return 0; } -static int bio_read_page(pgoff_t page_off, void *page) +static int bio_read_page(pgoff_t page_off, void *addr, struct bio **bio_chain) { - return submit(READ, page_off, page); + return submit(READ, page_off, virt_to_page(addr), bio_chain); } -static int bio_write_page(pgoff_t page_off, void *page) +static int bio_write_page(pgoff_t page_off, void *addr) { - return submit(WRITE, page_off, page); + return submit(WRITE, page_off, virt_to_page(addr), NULL); } /** @@ -476,7 +466,7 @@ static int get_swap_reader(struct swap_map_handle *handle, handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC); if (!handle->cur) return -ENOMEM; - error = bio_read_page(swp_offset(start), handle->cur); + error = bio_read_page(swp_offset(start), handle->cur, NULL); if (error) { release_swap_reader(handle); return error; @@ -485,7 +475,8 @@ static int get_swap_reader(struct swap_map_handle *handle, return 0; } -static int swap_read_page(struct swap_map_handle *handle, void *buf) +static int swap_read_page(struct swap_map_handle *handle, void *buf, + struct bio **bio_chain) { unsigned long offset; int error; @@ -495,16 +486,17 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf) offset = handle->cur->entries[handle->k]; if (!offset) return -EFAULT; - error = bio_read_page(offset, buf); + error = bio_read_page(offset, buf, bio_chain); if (error) return error; if (++handle->k >= MAP_PAGE_ENTRIES) { + error = wait_on_bio_chain(bio_chain); handle->k = 0; offset = handle->cur->next_swap; if (!offset) release_swap_reader(handle); - else - error = bio_read_page(offset, handle->cur); + else if (!error) + error = bio_read_page(offset, handle->cur, NULL); } return error; } @@ -517,38 +509,48 @@ static int swap_read_page(struct swap_map_handle *handle, void *buf) static int load_image(struct swap_map_handle *handle, struct snapshot_handle *snapshot, - unsigned int nr_pages) + unsigned int nr_to_read) { unsigned int m; - int ret; int error = 0; struct timeval start; struct timeval stop; + struct bio *bio; + int err2; + unsigned nr_pages; - printk("Loading image data pages (%u pages) ... ", nr_pages); - m = nr_pages / 100; + printk("Loading image data pages (%u pages) ... ", nr_to_read); + m = nr_to_read / 100; if (!m) m = 1; nr_pages = 0; + bio = NULL; do_gettimeofday(&start); - do { - ret = snapshot_write_next(snapshot, PAGE_SIZE); - if (ret > 0) { - error = swap_read_page(handle, data_of(*snapshot)); - if (error) - break; - if (!(nr_pages % m)) - printk("\b\b\b\b%3d%%", nr_pages / m); - nr_pages++; - } - } while (ret > 0); + for ( ; ; ) { + error = snapshot_write_next(snapshot, PAGE_SIZE); + if (error <= 0) + break; + error = swap_read_page(handle, data_of(*snapshot), &bio); + if (error) + break; + if (snapshot->sync_read) + error = wait_on_bio_chain(&bio); + if (error) + break; + if (!(nr_pages % m)) + printk("\b\b\b\b%3d%%", nr_pages / m); + nr_pages++; + } + err2 = wait_on_bio_chain(&bio); do_gettimeofday(&stop); + if (!error) + error = err2; if (!error) { printk("\b\b\b\bdone\n"); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; } - show_speed(&start, &stop, nr_pages, "Read"); + show_speed(&start, &stop, nr_to_read, "Read"); return error; } @@ -571,7 +573,7 @@ int swsusp_read(void) header = (struct swsusp_info *)data_of(snapshot); error = get_swap_reader(&handle, swsusp_header.image); if (!error) - error = swap_read_page(&handle, header); + error = swap_read_page(&handle, header, NULL); if (!error) error = load_image(&handle, &snapshot, header->pages - 1); release_swap_reader(&handle); @@ -597,7 +599,7 @@ int swsusp_check(void) if (!IS_ERR(resume_bdev)) { set_blocksize(resume_bdev, PAGE_SIZE); memset(&swsusp_header, 0, sizeof(swsusp_header)); - if ((error = bio_read_page(0, &swsusp_header))) + if ((error = bio_read_page(0, &swsusp_header, NULL))) return error; if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) { memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10); diff --git a/mm/page_io.c b/mm/page_io.c index f46a9862b7ef..d4840ecbf8f9 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -74,7 +74,7 @@ static int end_swap_bio_write(struct bio *bio, unsigned int bytes_done, int err) return 0; } -static int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err) +int end_swap_bio_read(struct bio *bio, unsigned int bytes_done, int err) { const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); struct page *page = bio->bi_io_vec[0].bv_page; -- cgit v1.2.3 From ae83c5eef59ffe6eb61110b8c8fd1ea3e0881712 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:45 -0700 Subject: [PATCH] swsusp: clean up browsing of pfns Clean up some loops over pfns for each zone in snapshot.c: reduce the number of additions to perform, rework detection of saveable pages and make the code a bit less difficult to understand, hopefully. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 62 +++++++++++++++++++++++++------------------------ 1 file changed, 32 insertions(+), 30 deletions(-) (limited to 'kernel') diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 591301ae8b7d..979096c27773 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -156,7 +156,7 @@ static inline int save_highmem(void) {return 0;} static inline int restore_highmem(void) {return 0;} #endif -static int pfn_is_nosave(unsigned long pfn) +static inline int pfn_is_nosave(unsigned long pfn) { unsigned long nosave_begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT; unsigned long nosave_end_pfn = PAGE_ALIGN(__pa(&__nosave_end)) >> PAGE_SHIFT; @@ -167,43 +167,43 @@ static int pfn_is_nosave(unsigned long pfn) * saveable - Determine whether a page should be cloned or not. * @pfn: The page * - * We save a page if it's Reserved, and not in the range of pages - * statically defined as 'unsaveable', or if it isn't reserved, and - * isn't part of a free chunk of pages. + * We save a page if it isn't Nosave, and is not in the range of pages + * statically defined as 'unsaveable', and it + * isn't a part of a free chunk of pages. */ -static int saveable(struct zone *zone, unsigned long *zone_pfn) +static struct page *saveable_page(unsigned long pfn) { - unsigned long pfn = *zone_pfn + zone->zone_start_pfn; struct page *page; if (!pfn_valid(pfn)) - return 0; + return NULL; page = pfn_to_page(pfn); - BUG_ON(PageReserved(page) && PageNosave(page)); + if (PageNosave(page)) - return 0; + return NULL; if (PageReserved(page) && pfn_is_nosave(pfn)) - return 0; + return NULL; if (PageNosaveFree(page)) - return 0; + return NULL; - return 1; + return page; } unsigned int count_data_pages(void) { struct zone *zone; - unsigned long zone_pfn; + unsigned long pfn, max_zone_pfn; unsigned int n = 0; for_each_zone (zone) { if (is_highmem(zone)) continue; mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - n += saveable(zone, &zone_pfn); + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + n += !!saveable_page(pfn); } return n; } @@ -211,7 +211,7 @@ unsigned int count_data_pages(void) static void copy_data_pages(struct pbe *pblist) { struct zone *zone; - unsigned long zone_pfn; + unsigned long pfn, max_zone_pfn; struct pbe *pbe, *p; pbe = pblist; @@ -224,13 +224,14 @@ static void copy_data_pages(struct pbe *pblist) SetPageNosaveFree(virt_to_page(p)); for_each_pbe (p, pblist) SetPageNosaveFree(virt_to_page(p->address)); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) { - if (saveable(zone, &zone_pfn)) { - struct page *page; + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) { + struct page *page = saveable_page(pfn); + + if (page) { long *src, *dst; int n; - page = pfn_to_page(zone_pfn + zone->zone_start_pfn); BUG_ON(!pbe); pbe->orig_address = (unsigned long)page_address(page); /* copy_page and memcpy are not usable for copying task structs. */ @@ -383,13 +384,14 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, void swsusp_free(void) { struct zone *zone; - unsigned long zone_pfn; + unsigned long pfn, max_zone_pfn; for_each_zone(zone) { - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - if (pfn_valid(zone_pfn + zone->zone_start_pfn)) { - struct page *page; - page = pfn_to_page(zone_pfn + zone->zone_start_pfn); + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + if (pfn_valid(pfn)) { + struct page *page = pfn_to_page(pfn); + if (PageNosave(page) && PageNosaveFree(page)) { ClearPageNosave(page); ClearPageNosaveFree(page); @@ -598,7 +600,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) static int mark_unsafe_pages(struct pbe *pblist) { struct zone *zone; - unsigned long zone_pfn; + unsigned long pfn, max_zone_pfn; struct pbe *p; if (!pblist) /* a sanity check */ @@ -606,10 +608,10 @@ static int mark_unsafe_pages(struct pbe *pblist) /* Clear page flags */ for_each_zone (zone) { - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - if (pfn_valid(zone_pfn + zone->zone_start_pfn)) - ClearPageNosaveFree(pfn_to_page(zone_pfn + - zone->zone_start_pfn)); + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + if (pfn_valid(pfn)) + ClearPageNosaveFree(pfn_to_page(pfn)); } /* Mark orig addresses */ -- cgit v1.2.3 From fb13a28b0f5ada60861868c4fa48a12bd0cb8dea Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:46 -0700 Subject: [PATCH] swsusp: struct snapshot_handle cleanup Add comments describing struct snapshot_handle and its members, change the confusing name of its member 'page' to 'cur'. Signed-off-by: Rafael J. Wysocki Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/power.h | 64 ++++++++++++++++++++++++++++++++++++++++++------- kernel/power/snapshot.c | 40 +++++++++++++++---------------- 2 files changed, 76 insertions(+), 28 deletions(-) (limited to 'kernel') diff --git a/kernel/power/power.h b/kernel/power/power.h index 59ce712f082d..1cefcf87a694 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -50,17 +50,65 @@ extern asmlinkage int swsusp_arch_resume(void); extern unsigned int count_data_pages(void); +/** + * Auxiliary structure used for reading the snapshot image data and + * metadata from and writing them to the list of page backup entries + * (PBEs) which is the main data structure of swsusp. + * + * Using struct snapshot_handle we can transfer the image, including its + * metadata, as a continuous sequence of bytes with the help of + * snapshot_read_next() and snapshot_write_next(). + * + * The code that writes the image to a storage or transfers it to + * the user land is required to use snapshot_read_next() for this + * purpose and it should not make any assumptions regarding the internal + * structure of the image. Similarly, the code that reads the image from + * a storage or transfers it from the user land is required to use + * snapshot_write_next(). + * + * This may allow us to change the internal structure of the image + * in the future with considerably less effort. + */ + struct snapshot_handle { - loff_t offset; - unsigned int page; - unsigned int page_offset; - unsigned int prev; - struct pbe *pbe, *last_pbe; - void *buffer; - unsigned int buf_offset; - int sync_read; + loff_t offset; /* number of the last byte ready for reading + * or writing in the sequence + */ + unsigned int cur; /* number of the block of PAGE_SIZE bytes the + * next operation will refer to (ie. current) + */ + unsigned int cur_offset; /* offset with respect to the current + * block (for the next operation) + */ + unsigned int prev; /* number of the block of PAGE_SIZE bytes that + * was the current one previously + */ + struct pbe *pbe; /* PBE that corresponds to 'buffer' */ + struct pbe *last_pbe; /* When the image is restored (eg. read + * from disk) we can store some image + * data directly in the page frames + * in which they were before suspend. + * In such a case the PBEs that + * correspond to them will be unused. + * This is the last PBE, so far, that + * does not correspond to such data. + */ + void *buffer; /* address of the block to read from + * or write to + */ + unsigned int buf_offset; /* location to read from or write to, + * given as a displacement from 'buffer' + */ + int sync_read; /* Set to one to notify the caller of + * snapshot_write_next() that it may + * need to call wait_on_bio_chain() + */ }; +/* This macro returns the address from/to which the caller of + * snapshot_read_next()/snapshot_write_next() is allowed to + * read/write data after the function returns + */ #define data_of(handle) ((handle).buffer + (handle).buf_offset) extern int snapshot_read_next(struct snapshot_handle *handle, size_t count); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 979096c27773..81fe8de9e604 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -555,7 +555,7 @@ static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pb int snapshot_read_next(struct snapshot_handle *handle, size_t count) { - if (handle->page > nr_meta_pages + nr_copy_pages) + if (handle->cur > nr_meta_pages + nr_copy_pages) return 0; if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ @@ -568,8 +568,8 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) handle->buffer = buffer; handle->pbe = pagedir_nosave; } - if (handle->prev < handle->page) { - if (handle->page <= nr_meta_pages) { + if (handle->prev < handle->cur) { + if (handle->cur <= nr_meta_pages) { handle->pbe = pack_orig_addresses(buffer, handle->pbe); if (!handle->pbe) handle->pbe = pagedir_nosave; @@ -577,15 +577,15 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) handle->buffer = (void *)handle->pbe->address; handle->pbe = handle->pbe->next; } - handle->prev = handle->page; + handle->prev = handle->cur; } - handle->buf_offset = handle->page_offset; - if (handle->page_offset + count >= PAGE_SIZE) { - count = PAGE_SIZE - handle->page_offset; - handle->page_offset = 0; - handle->page++; + handle->buf_offset = handle->cur_offset; + if (handle->cur_offset + count >= PAGE_SIZE) { + count = PAGE_SIZE - handle->cur_offset; + handle->cur_offset = 0; + handle->cur++; } else { - handle->page_offset += count; + handle->cur_offset += count; } handle->offset += count; return count; @@ -820,7 +820,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) { int error = 0; - if (handle->prev && handle->page > nr_meta_pages + nr_copy_pages) + if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) return 0; if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ @@ -831,7 +831,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) if (!handle->offset) handle->buffer = buffer; handle->sync_read = 1; - if (handle->prev < handle->page) { + if (handle->prev < handle->cur) { if (!handle->prev) { error = load_header(handle, (struct swsusp_info *)buffer); @@ -854,15 +854,15 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) handle->buffer = get_buffer(handle); handle->sync_read = 0; } - handle->prev = handle->page; + handle->prev = handle->cur; } - handle->buf_offset = handle->page_offset; - if (handle->page_offset + count >= PAGE_SIZE) { - count = PAGE_SIZE - handle->page_offset; - handle->page_offset = 0; - handle->page++; + handle->buf_offset = handle->cur_offset; + if (handle->cur_offset + count >= PAGE_SIZE) { + count = PAGE_SIZE - handle->cur_offset; + handle->cur_offset = 0; + handle->cur++; } else { - handle->page_offset += count; + handle->cur_offset += count; } handle->offset += count; return count; @@ -871,5 +871,5 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) int snapshot_image_loaded(struct snapshot_handle *handle) { return !(!handle->pbe || handle->pbe->next || !nr_copy_pages || - handle->page <= nr_meta_pages + nr_copy_pages); + handle->cur <= nr_meta_pages + nr_copy_pages); } -- cgit v1.2.3 From e3920fb42c8ddfe63befb54d95c0e13eabacea9b Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:48 -0700 Subject: [PATCH] Disable CPU hotplug during suspend The current suspend code has to be run on one CPU, so we use the CPU hotplug to take the non-boot CPUs offline on SMP machines. However, we should also make sure that these CPUs will not be enabled by someone else after we have disabled them. The functions disable_nonboot_cpus() and enable_nonboot_cpus() are moved to kernel/cpu.c, because they now refer to some stuff in there that should better be static. Also it's better if disable_nonboot_cpus() returns an error instead of panicking if something goes wrong, and enable_nonboot_cpus() has no reason to panic(), because the CPUs may have been enabled by the userland before it tries to take them online. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/cpu.h | 8 +++ include/linux/suspend.h | 8 --- kernel/cpu.c | 138 +++++++++++++++++++++++++++++++++++++++++------- kernel/power/Makefile | 2 - kernel/power/disk.c | 7 ++- kernel/power/main.c | 10 ++-- kernel/power/smp.c | 62 ---------------------- kernel/power/user.c | 14 +++-- 8 files changed, 145 insertions(+), 104 deletions(-) delete mode 100644 kernel/power/smp.c (limited to 'kernel') diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 8fb344a9abd8..3fef7d67aedc 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -89,4 +89,12 @@ int cpu_down(unsigned int cpu); static inline int cpu_is_offline(int cpu) { return 0; } #endif +#ifdef CONFIG_SUSPEND_SMP +extern int disable_nonboot_cpus(void); +extern void enable_nonboot_cpus(void); +#else +static inline int disable_nonboot_cpus(void) { return 0; } +static inline void enable_nonboot_cpus(void) {} +#endif + #endif /* _LINUX_CPU_H_ */ diff --git a/include/linux/suspend.h b/include/linux/suspend.h index 96e31aa64cc7..6e8a06c950f4 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -57,14 +57,6 @@ static inline int software_suspend(void) } #endif /* CONFIG_PM */ -#ifdef CONFIG_SUSPEND_SMP -extern void disable_nonboot_cpus(void); -extern void enable_nonboot_cpus(void); -#else -static inline void disable_nonboot_cpus(void) {} -static inline void enable_nonboot_cpus(void) {} -#endif - void save_processor_state(void); void restore_processor_state(void); struct saved_context; diff --git a/kernel/cpu.c b/kernel/cpu.c index f230f9ae01c2..32c96628463e 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -21,6 +21,11 @@ static DEFINE_MUTEX(cpu_bitmask_lock); static __cpuinitdata BLOCKING_NOTIFIER_HEAD(cpu_chain); +/* If set, cpu_up and cpu_down will return -EBUSY and do nothing. + * Should always be manipulated under cpu_add_remove_lock + */ +static int cpu_hotplug_disabled; + #ifdef CONFIG_HOTPLUG_CPU /* Crappy recursive lock-takers in cpufreq! Complain loudly about idiots */ @@ -108,30 +113,25 @@ static int take_cpu_down(void *unused) return 0; } -int cpu_down(unsigned int cpu) +/* Requires cpu_add_remove_lock to be held */ +static int _cpu_down(unsigned int cpu) { int err; struct task_struct *p; cpumask_t old_allowed, tmp; - mutex_lock(&cpu_add_remove_lock); - if (num_online_cpus() == 1) { - err = -EBUSY; - goto out; - } + if (num_online_cpus() == 1) + return -EBUSY; - if (!cpu_online(cpu)) { - err = -EINVAL; - goto out; - } + if (!cpu_online(cpu)) + return -EINVAL; err = blocking_notifier_call_chain(&cpu_chain, CPU_DOWN_PREPARE, (void *)(long)cpu); if (err == NOTIFY_BAD) { printk("%s: attempt to take down CPU %u failed\n", __FUNCTION__, cpu); - err = -EINVAL; - goto out; + return -EINVAL; } /* Ensure that we are not runnable on dying cpu */ @@ -179,22 +179,32 @@ out_thread: err = kthread_stop(p); out_allowed: set_cpus_allowed(current, old_allowed); -out: + return err; +} + +int cpu_down(unsigned int cpu) +{ + int err = 0; + + mutex_lock(&cpu_add_remove_lock); + if (cpu_hotplug_disabled) + err = -EBUSY; + else + err = _cpu_down(cpu); + mutex_unlock(&cpu_add_remove_lock); return err; } #endif /*CONFIG_HOTPLUG_CPU*/ -int __devinit cpu_up(unsigned int cpu) +/* Requires cpu_add_remove_lock to be held */ +static int __devinit _cpu_up(unsigned int cpu) { int ret; void *hcpu = (void *)(long)cpu; - mutex_lock(&cpu_add_remove_lock); - if (cpu_online(cpu) || !cpu_present(cpu)) { - ret = -EINVAL; - goto out; - } + if (cpu_online(cpu) || !cpu_present(cpu)) + return -EINVAL; ret = blocking_notifier_call_chain(&cpu_chain, CPU_UP_PREPARE, hcpu); if (ret == NOTIFY_BAD) { @@ -219,7 +229,95 @@ out_notify: if (ret != 0) blocking_notifier_call_chain(&cpu_chain, CPU_UP_CANCELED, hcpu); + + return ret; +} + +int __devinit cpu_up(unsigned int cpu) +{ + int err = 0; + + mutex_lock(&cpu_add_remove_lock); + if (cpu_hotplug_disabled) + err = -EBUSY; + else + err = _cpu_up(cpu); + + mutex_unlock(&cpu_add_remove_lock); + return err; +} + +#ifdef CONFIG_SUSPEND_SMP +static cpumask_t frozen_cpus; + +int disable_nonboot_cpus(void) +{ + int cpu, first_cpu, error; + + mutex_lock(&cpu_add_remove_lock); + first_cpu = first_cpu(cpu_present_map); + if (!cpu_online(first_cpu)) { + error = _cpu_up(first_cpu); + if (error) { + printk(KERN_ERR "Could not bring CPU%d up.\n", + first_cpu); + goto out; + } + } + error = set_cpus_allowed(current, cpumask_of_cpu(first_cpu)); + if (error) { + printk(KERN_ERR "Could not run on CPU%d\n", first_cpu); + goto out; + } + /* We take down all of the non-boot CPUs in one shot to avoid races + * with the userspace trying to use the CPU hotplug at the same time + */ + cpus_clear(frozen_cpus); + printk("Disabling non-boot CPUs ...\n"); + for_each_online_cpu(cpu) { + if (cpu == first_cpu) + continue; + error = _cpu_down(cpu); + if (!error) { + cpu_set(cpu, frozen_cpus); + printk("CPU%d is down\n", cpu); + } else { + printk(KERN_ERR "Error taking CPU%d down: %d\n", + cpu, error); + break; + } + } + if (!error) { + BUG_ON(num_online_cpus() > 1); + /* Make sure the CPUs won't be enabled by someone else */ + cpu_hotplug_disabled = 1; + } else { + printk(KERN_ERR "Non-boot CPUs are not disabled"); + } out: mutex_unlock(&cpu_add_remove_lock); - return ret; + return error; +} + +void enable_nonboot_cpus(void) +{ + int cpu, error; + + /* Allow everyone to use the CPU hotplug again */ + mutex_lock(&cpu_add_remove_lock); + cpu_hotplug_disabled = 0; + mutex_unlock(&cpu_add_remove_lock); + + printk("Enabling non-boot CPUs ...\n"); + for_each_cpu_mask(cpu, frozen_cpus) { + error = cpu_up(cpu); + if (!error) { + printk("CPU%d is up\n", cpu); + continue; + } + printk(KERN_WARNING "Error taking CPU%d up: %d\n", + cpu, error); + } + cpus_clear(frozen_cpus); } +#endif diff --git a/kernel/power/Makefile b/kernel/power/Makefile index 8d0af3d37a4b..38725f526afc 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -7,6 +7,4 @@ obj-y := main.o process.o console.o obj-$(CONFIG_PM_LEGACY) += pm.o obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o snapshot.o swap.o user.o -obj-$(CONFIG_SUSPEND_SMP) += smp.o - obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o diff --git a/kernel/power/disk.c b/kernel/power/disk.c index e13e74067845..7c7b9b65e365 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -18,6 +18,7 @@ #include #include #include +#include #include "power.h" @@ -72,7 +73,10 @@ static int prepare_processes(void) int error; pm_prepare_console(); - disable_nonboot_cpus(); + + error = disable_nonboot_cpus(); + if (error) + goto enable_cpus; if (freeze_processes()) { error = -EBUSY; @@ -84,6 +88,7 @@ static int prepare_processes(void) return 0; thaw: thaw_processes(); +enable_cpus: enable_nonboot_cpus(); pm_restore_console(); return error; diff --git a/kernel/power/main.c b/kernel/power/main.c index 6d295c776794..4d403323a7bb 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "power.h" @@ -51,7 +52,7 @@ void pm_set_ops(struct pm_ops * ops) static int suspend_prepare(suspend_state_t state) { - int error = 0; + int error; unsigned int free_pages; if (!pm_ops || !pm_ops->enter) @@ -59,12 +60,9 @@ static int suspend_prepare(suspend_state_t state) pm_prepare_console(); - disable_nonboot_cpus(); - - if (num_online_cpus() != 1) { - error = -EPERM; + error = disable_nonboot_cpus(); + if (error) goto Enable_cpu; - } if (freeze_processes()) { error = -EAGAIN; diff --git a/kernel/power/smp.c b/kernel/power/smp.c deleted file mode 100644 index 5957312b2d68..000000000000 --- a/kernel/power/smp.c +++ /dev/null @@ -1,62 +0,0 @@ -/* - * drivers/power/smp.c - Functions for stopping other CPUs. - * - * Copyright 2004 Pavel Machek - * Copyright (C) 2002-2003 Nigel Cunningham - * - * This file is released under the GPLv2. - */ - -#undef DEBUG - -#include -#include -#include -#include -#include -#include -#include - -/* This is protected by pm_sem semaphore */ -static cpumask_t frozen_cpus; - -void disable_nonboot_cpus(void) -{ - int cpu, error; - - error = 0; - cpus_clear(frozen_cpus); - printk("Freezing cpus ...\n"); - for_each_online_cpu(cpu) { - if (cpu == 0) - continue; - error = cpu_down(cpu); - if (!error) { - cpu_set(cpu, frozen_cpus); - printk("CPU%d is down\n", cpu); - continue; - } - printk("Error taking cpu %d down: %d\n", cpu, error); - } - BUG_ON(raw_smp_processor_id() != 0); - if (error) - panic("cpus not sleeping"); -} - -void enable_nonboot_cpus(void) -{ - int cpu, error; - - printk("Thawing cpus ...\n"); - for_each_cpu_mask(cpu, frozen_cpus) { - error = cpu_up(cpu); - if (!error) { - printk("CPU%d is up\n", cpu); - continue; - } - printk("Error taking cpu %d up: %d\n", cpu, error); - panic("Not enough cpus"); - } - cpus_clear(frozen_cpus); -} - diff --git a/kernel/power/user.c b/kernel/power/user.c index 3f1539fbe48a..0ef5e4ba39e5 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -19,6 +19,7 @@ #include #include #include +#include #include @@ -139,12 +140,15 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, if (data->frozen) break; down(&pm_sem); - disable_nonboot_cpus(); - if (freeze_processes()) { - thaw_processes(); - enable_nonboot_cpus(); - error = -EBUSY; + error = disable_nonboot_cpus(); + if (!error) { + error = freeze_processes(); + if (error) { + thaw_processes(); + error = -EBUSY; + } } + enable_nonboot_cpus(); up(&pm_sem); if (!error) data->frozen = 1; -- cgit v1.2.3 From f623f0db8e6aa86a37be86167e4ff478821a9f4f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:49 -0700 Subject: [PATCH] swsusp: Fix mark_free_pages Clean up mm/page_alloc.c#mark_free_pages() and make it avoid clearing PageNosaveFree for PageNosave pages. This allows us to get rid of an ugly hack in kernel/power/snapshot.c#copy_data_pages(). Additionally, the page-copying loop in copy_data_pages() is moved to an inline function. Signed-off-by: Rafael J. Wysocki Cc: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 27 +++++++++++++-------------- mm/page_alloc.c | 24 ++++++++++++++++-------- 2 files changed, 29 insertions(+), 22 deletions(-) (limited to 'kernel') diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 81fe8de9e604..4afb7900002b 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -208,37 +208,36 @@ unsigned int count_data_pages(void) return n; } +static inline void copy_data_page(long *dst, long *src) +{ + int n; + + /* copy_page and memcpy are not usable for copying task structs. */ + for (n = PAGE_SIZE / sizeof(long); n; n--) + *dst++ = *src++; +} + static void copy_data_pages(struct pbe *pblist) { struct zone *zone; unsigned long pfn, max_zone_pfn; - struct pbe *pbe, *p; + struct pbe *pbe; pbe = pblist; for_each_zone (zone) { if (is_highmem(zone)) continue; mark_free_pages(zone); - /* This is necessary for swsusp_free() */ - for_each_pb_page (p, pblist) - SetPageNosaveFree(virt_to_page(p)); - for_each_pbe (p, pblist) - SetPageNosaveFree(virt_to_page(p->address)); max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) { struct page *page = saveable_page(pfn); if (page) { - long *src, *dst; - int n; + void *ptr = page_address(page); BUG_ON(!pbe); - pbe->orig_address = (unsigned long)page_address(page); - /* copy_page and memcpy are not usable for copying task structs. */ - dst = (long *)pbe->address; - src = (long *)pbe->orig_address; - for (n = PAGE_SIZE / sizeof(long); n; n--) - *dst++ = *src++; + copy_data_page((void *)pbe->address, ptr); + pbe->orig_address = (unsigned long)ptr; pbe = pbe->next; } } diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 51070b6d593f..9810f0a60db7 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -694,7 +694,8 @@ static void __drain_pages(unsigned int cpu) void mark_free_pages(struct zone *zone) { - unsigned long zone_pfn, flags; + unsigned long pfn, max_zone_pfn; + unsigned long flags; int order; struct list_head *curr; @@ -702,18 +703,25 @@ void mark_free_pages(struct zone *zone) return; spin_lock_irqsave(&zone->lock, flags); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) - ClearPageNosaveFree(pfn_to_page(zone_pfn + zone->zone_start_pfn)); + + max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + if (pfn_valid(pfn)) { + struct page *page = pfn_to_page(pfn); + + if (!PageNosave(page)) + ClearPageNosaveFree(page); + } for (order = MAX_ORDER - 1; order >= 0; --order) list_for_each(curr, &zone->free_area[order].free_list) { - unsigned long start_pfn, i; + unsigned long i; - start_pfn = page_to_pfn(list_entry(curr, struct page, lru)); + pfn = page_to_pfn(list_entry(curr, struct page, lru)); + for (i = 0; i < (1UL << order); i++) + SetPageNosaveFree(pfn_to_page(pfn + i)); + } - for (i=0; i < (1<lock, flags); } -- cgit v1.2.3 From f6143aa60ed71e58578bc92cc64d98158a694d99 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:50 -0700 Subject: [PATCH] swsusp: Reorder memory-allocating functions Move some functions in kernel/power/snapshot.c to a better place (in the same file) and introduce free_image_page() (will be necessary in the future). Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 93 ++++++++++++++++++++++++++++--------------------- 1 file changed, 53 insertions(+), 40 deletions(-) (limited to 'kernel') diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 4afb7900002b..7ad0c0465524 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -156,6 +156,58 @@ static inline int save_highmem(void) {return 0;} static inline int restore_highmem(void) {return 0;} #endif +/** + * @safe_needed - on resume, for storing the PBE list and the image, + * we can only use memory pages that do not conflict with the pages + * used before suspend. + * + * The unsafe pages are marked with the PG_nosave_free flag + * and we count them using unsafe_pages + */ + +static unsigned int unsafe_pages; + +static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) +{ + void *res; + + res = (void *)get_zeroed_page(gfp_mask); + if (safe_needed) + while (res && PageNosaveFree(virt_to_page(res))) { + /* The page is unsafe, mark it for swsusp_free() */ + SetPageNosave(virt_to_page(res)); + unsafe_pages++; + res = (void *)get_zeroed_page(gfp_mask); + } + if (res) { + SetPageNosave(virt_to_page(res)); + SetPageNosaveFree(virt_to_page(res)); + } + return res; +} + +unsigned long get_safe_page(gfp_t gfp_mask) +{ + return (unsigned long)alloc_image_page(gfp_mask, 1); +} + +/** + * free_image_page - free page represented by @addr, allocated with + * alloc_image_page (page flags set by it must be cleared) + */ + +static inline void free_image_page(void *addr, int clear_nosave_free) +{ + ClearPageNosave(virt_to_page(addr)); + if (clear_nosave_free) + ClearPageNosaveFree(virt_to_page(addr)); + free_page((unsigned long)addr); +} + +/** + * pfn_is_nosave - check if given pfn is in the 'nosave' section + */ + static inline int pfn_is_nosave(unsigned long pfn) { unsigned long nosave_begin_pfn = __pa(&__nosave_begin) >> PAGE_SHIFT; @@ -245,7 +297,6 @@ static void copy_data_pages(struct pbe *pblist) BUG_ON(pbe); } - /** * free_pagedir - free pages allocated with alloc_pagedir() */ @@ -256,10 +307,7 @@ static void free_pagedir(struct pbe *pblist, int clear_nosave_free) while (pblist) { pbe = (pblist + PB_PAGE_SKIP)->next; - ClearPageNosave(virt_to_page(pblist)); - if (clear_nosave_free) - ClearPageNosaveFree(virt_to_page(pblist)); - free_page((unsigned long)pblist); + free_image_page(pblist, clear_nosave_free); pblist = pbe; } } @@ -303,41 +351,6 @@ static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages) } } -static unsigned int unsafe_pages; - -/** - * @safe_needed - on resume, for storing the PBE list and the image, - * we can only use memory pages that do not conflict with the pages - * used before suspend. - * - * The unsafe pages are marked with the PG_nosave_free flag - * and we count them using unsafe_pages - */ - -static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) -{ - void *res; - - res = (void *)get_zeroed_page(gfp_mask); - if (safe_needed) - while (res && PageNosaveFree(virt_to_page(res))) { - /* The page is unsafe, mark it for swsusp_free() */ - SetPageNosave(virt_to_page(res)); - unsafe_pages++; - res = (void *)get_zeroed_page(gfp_mask); - } - if (res) { - SetPageNosave(virt_to_page(res)); - SetPageNosaveFree(virt_to_page(res)); - } - return res; -} - -unsigned long get_safe_page(gfp_t gfp_mask) -{ - return (unsigned long)alloc_image_page(gfp_mask, 1); -} - /** * alloc_pagedir - Allocate the page directory. * -- cgit v1.2.3 From cd560bb2f9e2cd451bb3942af43da19632ba4a8e Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:50 -0700 Subject: [PATCH] swsusp: Fix alloc_pagedir Get rid of the FIXME in kernel/power/snapshot.c#alloc_pagedir() and simplify the functions called by it. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 32 +++++++++++++++++--------------- 1 file changed, 17 insertions(+), 15 deletions(-) (limited to 'kernel') diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 7ad0c0465524..4ca372f2bc14 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -316,12 +316,12 @@ static void free_pagedir(struct pbe *pblist, int clear_nosave_free) * fill_pb_page - Create a list of PBEs on a given memory page */ -static inline void fill_pb_page(struct pbe *pbpage) +static inline void fill_pb_page(struct pbe *pbpage, unsigned int n) { struct pbe *p; p = pbpage; - pbpage += PB_PAGE_SKIP; + pbpage += n - 1; do p->next = p + 1; while (++p < pbpage); @@ -330,24 +330,26 @@ static inline void fill_pb_page(struct pbe *pbpage) /** * create_pbe_list - Create a list of PBEs on top of a given chain * of memory pages allocated with alloc_pagedir() + * + * This function assumes that pages allocated by alloc_image_page() will + * always be zeroed. */ static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages) { - struct pbe *pbpage, *p; + struct pbe *pbpage; unsigned int num = PBES_PER_PAGE; for_each_pb_page (pbpage, pblist) { if (num >= nr_pages) break; - fill_pb_page(pbpage); + fill_pb_page(pbpage, PBES_PER_PAGE); num += PBES_PER_PAGE; } if (pbpage) { - for (num -= PBES_PER_PAGE - 1, p = pbpage; num < nr_pages; p++, num++) - p->next = p + 1; - p->next = NULL; + num -= PBES_PER_PAGE; + fill_pb_page(pbpage, nr_pages - num); } } @@ -374,17 +376,17 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, return NULL; pblist = alloc_image_page(gfp_mask, safe_needed); - /* FIXME: rewrite this ugly loop */ - for (pbe = pblist, num = PBES_PER_PAGE; pbe && num < nr_pages; - pbe = pbe->next, num += PBES_PER_PAGE) { + pbe = pblist; + for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) { + if (!pbe) { + free_pagedir(pblist, 1); + return NULL; + } pbe += PB_PAGE_SKIP; pbe->next = alloc_image_page(gfp_mask, safe_needed); + pbe = pbe->next; } - if (!pbe) { /* get_zeroed_page() failed */ - free_pagedir(pblist, 1); - pblist = NULL; - } else - create_pbe_list(pblist, nr_pages); + create_pbe_list(pblist, nr_pages); return pblist; } -- cgit v1.2.3 From 75534b50cc658e951bcb213c2763c81e9f7b0b48 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:52 -0700 Subject: [PATCH] Change the name of pagedir_nosave The name of the pagedir_nosave variable does not make sense any more, so it seems reasonable to change it to something more meaningful. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/i386/power/swsusp.S | 2 +- arch/powerpc/kernel/swsusp_32.S | 4 ++-- arch/x86_64/kernel/suspend_asm.S | 2 +- kernel/power/power.h | 2 -- kernel/power/snapshot.c | 26 ++++++++++++++------------ 5 files changed, 18 insertions(+), 18 deletions(-) (limited to 'kernel') diff --git a/arch/i386/power/swsusp.S b/arch/i386/power/swsusp.S index c893b897217f..8a2b50a0aaad 100644 --- a/arch/i386/power/swsusp.S +++ b/arch/i386/power/swsusp.S @@ -32,7 +32,7 @@ ENTRY(swsusp_arch_resume) movl $swsusp_pg_dir-__PAGE_OFFSET, %ecx movl %ecx, %cr3 - movl pagedir_nosave, %edx + movl restore_pblist, %edx .p2align 4,,7 copy_loop: diff --git a/arch/powerpc/kernel/swsusp_32.S b/arch/powerpc/kernel/swsusp_32.S index 7369f9a6ad25..69e8f86aa4f8 100644 --- a/arch/powerpc/kernel/swsusp_32.S +++ b/arch/powerpc/kernel/swsusp_32.S @@ -159,8 +159,8 @@ END_FTR_SECTION_IFSET(CPU_FTR_ALTIVEC) isync /* Load ptr the list of pages to copy in r3 */ - lis r11,(pagedir_nosave - KERNELBASE)@h - ori r11,r11,pagedir_nosave@l + lis r11,(restore_pblist - KERNELBASE)@h + ori r11,r11,restore_pblist@l lwz r10,0(r11) /* Copy the pages. This is a very basic implementation, to diff --git a/arch/x86_64/kernel/suspend_asm.S b/arch/x86_64/kernel/suspend_asm.S index 320b6fb00cca..bfbe00763c68 100644 --- a/arch/x86_64/kernel/suspend_asm.S +++ b/arch/x86_64/kernel/suspend_asm.S @@ -54,7 +54,7 @@ ENTRY(restore_image) movq %rcx, %cr3; movq %rax, %cr4; # turn PGE back on - movq pagedir_nosave(%rip), %rdx + movq restore_pblist(%rip), %rdx loop: testq %rdx, %rdx jz done diff --git a/kernel/power/power.h b/kernel/power/power.h index 1cefcf87a694..e18ba207e784 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -38,8 +38,6 @@ extern struct subsystem power_subsys; /* References to section boundaries */ extern const void __nosave_begin, __nosave_end; -extern struct pbe *pagedir_nosave; - /* Preferred image size in bytes (default 500 MB) */ extern unsigned long image_size; extern int in_suspend; diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 4ca372f2bc14..1d276b3ae152 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -34,7 +34,9 @@ #include "power.h" -struct pbe *pagedir_nosave; +/* List of PBEs used for creating and restoring the suspend image */ +struct pbe *restore_pblist; + static unsigned int nr_copy_pages; static unsigned int nr_meta_pages; static unsigned long *buffer; @@ -415,7 +417,7 @@ void swsusp_free(void) } nr_copy_pages = 0; nr_meta_pages = 0; - pagedir_nosave = NULL; + restore_pblist = NULL; buffer = NULL; } @@ -490,15 +492,15 @@ asmlinkage int swsusp_save(void) return -ENOMEM; } - pagedir_nosave = swsusp_alloc(nr_pages); - if (!pagedir_nosave) + restore_pblist = swsusp_alloc(nr_pages); + if (!restore_pblist) return -ENOMEM; /* During allocating of suspend pagedir, new cold pages may appear. * Kill them. */ drain_local_pages(); - copy_data_pages(pagedir_nosave); + copy_data_pages(restore_pblist); /* * End of critical section. From now on, we can write to memory, @@ -580,13 +582,13 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) if (!handle->offset) { init_header((struct swsusp_info *)buffer); handle->buffer = buffer; - handle->pbe = pagedir_nosave; + handle->pbe = restore_pblist; } if (handle->prev < handle->cur) { if (handle->cur <= nr_meta_pages) { handle->pbe = pack_orig_addresses(buffer, handle->pbe); if (!handle->pbe) - handle->pbe = pagedir_nosave; + handle->pbe = restore_pblist; } else { handle->buffer = (void *)handle->pbe->address; handle->pbe = handle->pbe->next; @@ -689,7 +691,7 @@ static int load_header(struct snapshot_handle *handle, pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0); if (!pblist) return -ENOMEM; - pagedir_nosave = pblist; + restore_pblist = pblist; handle->pbe = pblist; nr_copy_pages = info->image_pages; nr_meta_pages = info->pages - info->image_pages - 1; @@ -716,7 +718,7 @@ static inline struct pbe *unpack_orig_addresses(unsigned long *buf, /** * prepare_image - use metadata contained in the PBE list - * pointed to by pagedir_nosave to mark the pages that will + * pointed to by restore_pblist to mark the pages that will * be overwritten in the process of restoring the system * memory state from the image ("unsafe" pages) and allocate * memory for the image @@ -741,7 +743,7 @@ static int prepare_image(struct snapshot_handle *handle) unsigned int nr_pages = nr_copy_pages; struct pbe *p, *pblist = NULL; - p = pagedir_nosave; + p = restore_pblist; error = mark_unsafe_pages(p); if (!error) { pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1); @@ -773,7 +775,7 @@ static int prepare_image(struct snapshot_handle *handle) } } if (!error) { - pagedir_nosave = pblist; + restore_pblist = pblist; } else { handle->pbe = NULL; swsusp_free(); @@ -858,7 +860,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) error = prepare_image(handle); if (error) return error; - handle->pbe = pagedir_nosave; + handle->pbe = restore_pblist; handle->last_pbe = NULL; handle->buffer = get_buffer(handle); handle->sync_read = 0; -- cgit v1.2.3 From 0bcd888d64684f896ffa70c1d16a42b00753c184 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:52 -0700 Subject: [PATCH] swsusp: Introduce some helpful constants Introduce some constants that hopefully will help improve the readability of code in kernel/power/snapshot.c. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/snapshot.c | 24 +++++++++++++++--------- 1 file changed, 15 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 1d276b3ae152..d0d691f976d8 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -167,6 +167,11 @@ static inline int restore_highmem(void) {return 0;} * and we count them using unsafe_pages */ +#define PG_ANY 0 +#define PG_SAFE 1 +#define PG_UNSAFE_CLEAR 1 +#define PG_UNSAFE_KEEP 0 + static unsigned int unsafe_pages; static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) @@ -190,7 +195,7 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) unsigned long get_safe_page(gfp_t gfp_mask) { - return (unsigned long)alloc_image_page(gfp_mask, 1); + return (unsigned long)alloc_image_page(gfp_mask, PG_SAFE); } /** @@ -381,7 +386,7 @@ static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, pbe = pblist; for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) { if (!pbe) { - free_pagedir(pblist, 1); + free_pagedir(pblist, PG_UNSAFE_CLEAR); return NULL; } pbe += PB_PAGE_SKIP; @@ -458,12 +463,13 @@ static struct pbe *swsusp_alloc(unsigned int nr_pages) { struct pbe *pblist; - if (!(pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, 0))) { + pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, PG_ANY); + if (!pblist) { printk(KERN_ERR "suspend: Allocating pagedir failed.\n"); return NULL; } - if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, 0)) { + if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, PG_ANY)) { printk(KERN_ERR "suspend: Allocating image pages failed.\n"); swsusp_free(); return NULL; @@ -575,7 +581,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) return 0; if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ - buffer = alloc_image_page(GFP_ATOMIC, 0); + buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); if (!buffer) return -ENOMEM; } @@ -688,7 +694,7 @@ static int load_header(struct snapshot_handle *handle, error = check_header(info); if (!error) { - pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0); + pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY); if (!pblist) return -ENOMEM; restore_pblist = pblist; @@ -746,10 +752,10 @@ static int prepare_image(struct snapshot_handle *handle) p = restore_pblist; error = mark_unsafe_pages(p); if (!error) { - pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1); + pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE); if (pblist) copy_page_backup_list(pblist, p); - free_pagedir(p, 0); + free_pagedir(p, PG_UNSAFE_KEEP); if (!pblist) error = -ENOMEM; } @@ -840,7 +846,7 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) return 0; if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ - buffer = alloc_image_page(GFP_ATOMIC, 0); + buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); if (!buffer) return -ENOMEM; } -- cgit v1.2.3 From b788db79896ef2a5817b9395ad63573b254a6d93 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:54 -0700 Subject: [PATCH] swsusp: Introduce memory bitmaps Introduce the memory bitmap data structure and make swsusp use in the suspend phase. The current swsusp's internal data structure is not very efficient from the memory usage point of view, so it seems reasonable to replace it with a data structure that will require less memory, such as a pair of bitmaps. The idea is to use bitmaps that may be allocated as sets of individual pages, so that we can avoid making allocations of order greater than 0. For this reason the memory bitmap structure consists of several linked lists of objects that contain pointers to memory pages with the actual bitmap data. Still, for a typical system all of these lists fit in a single page, so it's reasonable to introduce an additional mechanism allowing us to allocate all of them efficiently without sacrificing the generality of the design. This is done with the help of the chain_allocator structure and associated functions. We need to use two memory bitmaps during the suspend phase of the suspend-resume cycle. One of them is necessary for marking the saveable pages, and the second is used to mark the pages in which to store the copies of them (aka image pages). First, the bitmaps are created and we allocate as many image pages as needed (the corresponding bits in the second bitmap are set as soon as the pages are allocated). Second, the bits corresponding to the saveable pages are set in the first bitmap and the saveable pages are copied to the image pages. Finally, the first bitmap is used to save the kernel virtual addresses of the saveable pages and the second one is used to save the contents of the image pages. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/power/power.h | 3 +- kernel/power/snapshot.c | 602 ++++++++++++++++++++++++++++++++++++++++++------ kernel/power/swsusp.c | 5 +- 3 files changed, 540 insertions(+), 70 deletions(-) (limited to 'kernel') diff --git a/kernel/power/power.h b/kernel/power/power.h index e18ba207e784..6e9e2acc34f8 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -109,9 +109,10 @@ struct snapshot_handle { */ #define data_of(handle) ((handle).buffer + (handle).buf_offset) +extern unsigned int snapshot_additional_pages(struct zone *zone); extern int snapshot_read_next(struct snapshot_handle *handle, size_t count); extern int snapshot_write_next(struct snapshot_handle *handle, size_t count); -int snapshot_image_loaded(struct snapshot_handle *handle); +extern int snapshot_image_loaded(struct snapshot_handle *handle); #define SNAPSHOT_IOC_MAGIC '3' #define SNAPSHOT_FREEZE _IO(SNAPSHOT_IOC_MAGIC, 1) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index d0d691f976d8..852e0df41719 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -211,6 +211,467 @@ static inline void free_image_page(void *addr, int clear_nosave_free) free_page((unsigned long)addr); } +/* struct linked_page is used to build chains of pages */ + +#define LINKED_PAGE_DATA_SIZE (PAGE_SIZE - sizeof(void *)) + +struct linked_page { + struct linked_page *next; + char data[LINKED_PAGE_DATA_SIZE]; +} __attribute__((packed)); + +static inline void +free_list_of_pages(struct linked_page *list, int clear_page_nosave) +{ + while (list) { + struct linked_page *lp = list->next; + + free_image_page(list, clear_page_nosave); + list = lp; + } +} + +/** + * struct chain_allocator is used for allocating small objects out of + * a linked list of pages called 'the chain'. + * + * The chain grows each time when there is no room for a new object in + * the current page. The allocated objects cannot be freed individually. + * It is only possible to free them all at once, by freeing the entire + * chain. + * + * NOTE: The chain allocator may be inefficient if the allocated objects + * are not much smaller than PAGE_SIZE. + */ + +struct chain_allocator { + struct linked_page *chain; /* the chain */ + unsigned int used_space; /* total size of objects allocated out + * of the current page + */ + gfp_t gfp_mask; /* mask for allocating pages */ + int safe_needed; /* if set, only "safe" pages are allocated */ +}; + +static void +chain_init(struct chain_allocator *ca, gfp_t gfp_mask, int safe_needed) +{ + ca->chain = NULL; + ca->used_space = LINKED_PAGE_DATA_SIZE; + ca->gfp_mask = gfp_mask; + ca->safe_needed = safe_needed; +} + +static void *chain_alloc(struct chain_allocator *ca, unsigned int size) +{ + void *ret; + + if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) { + struct linked_page *lp; + + lp = alloc_image_page(ca->gfp_mask, ca->safe_needed); + if (!lp) + return NULL; + + lp->next = ca->chain; + ca->chain = lp; + ca->used_space = 0; + } + ret = ca->chain->data + ca->used_space; + ca->used_space += size; + return ret; +} + +static void chain_free(struct chain_allocator *ca, int clear_page_nosave) +{ + free_list_of_pages(ca->chain, clear_page_nosave); + memset(ca, 0, sizeof(struct chain_allocator)); +} + +/** + * Data types related to memory bitmaps. + * + * Memory bitmap is a structure consiting of many linked lists of + * objects. The main list's elements are of type struct zone_bitmap + * and each of them corresonds to one zone. For each zone bitmap + * object there is a list of objects of type struct bm_block that + * represent each blocks of bit chunks in which information is + * stored. + * + * struct memory_bitmap contains a pointer to the main list of zone + * bitmap objects, a struct bm_position used for browsing the bitmap, + * and a pointer to the list of pages used for allocating all of the + * zone bitmap objects and bitmap block objects. + * + * NOTE: It has to be possible to lay out the bitmap in memory + * using only allocations of order 0. Additionally, the bitmap is + * designed to work with arbitrary number of zones (this is over the + * top for now, but let's avoid making unnecessary assumptions ;-). + * + * struct zone_bitmap contains a pointer to a list of bitmap block + * objects and a pointer to the bitmap block object that has been + * most recently used for setting bits. Additionally, it contains the + * pfns that correspond to the start and end of the represented zone. + * + * struct bm_block contains a pointer to the memory page in which + * information is stored (in the form of a block of bit chunks + * of type unsigned long each). It also contains the pfns that + * correspond to the start and end of the represented memory area and + * the number of bit chunks in the block. + * + * NOTE: Memory bitmaps are used for two types of operations only: + * "set a bit" and "find the next bit set". Moreover, the searching + * is always carried out after all of the "set a bit" operations + * on given bitmap. + */ + +#define BM_END_OF_MAP (~0UL) + +#define BM_CHUNKS_PER_BLOCK (PAGE_SIZE / sizeof(long)) +#define BM_BITS_PER_CHUNK (sizeof(long) << 3) +#define BM_BITS_PER_BLOCK (PAGE_SIZE << 3) + +struct bm_block { + struct bm_block *next; /* next element of the list */ + unsigned long start_pfn; /* pfn represented by the first bit */ + unsigned long end_pfn; /* pfn represented by the last bit plus 1 */ + unsigned int size; /* number of bit chunks */ + unsigned long *data; /* chunks of bits representing pages */ +}; + +struct zone_bitmap { + struct zone_bitmap *next; /* next element of the list */ + unsigned long start_pfn; /* minimal pfn in this zone */ + unsigned long end_pfn; /* maximal pfn in this zone plus 1 */ + struct bm_block *bm_blocks; /* list of bitmap blocks */ + struct bm_block *cur_block; /* recently used bitmap block */ +}; + +/* strcut bm_position is used for browsing memory bitmaps */ + +struct bm_position { + struct zone_bitmap *zone_bm; + struct bm_block *block; + int chunk; + int bit; +}; + +struct memory_bitmap { + struct zone_bitmap *zone_bm_list; /* list of zone bitmaps */ + struct linked_page *p_list; /* list of pages used to store zone + * bitmap objects and bitmap block + * objects + */ + struct bm_position cur; /* most recently used bit position */ +}; + +/* Functions that operate on memory bitmaps */ + +static inline void memory_bm_reset_chunk(struct memory_bitmap *bm) +{ + bm->cur.chunk = 0; + bm->cur.bit = -1; +} + +static void memory_bm_position_reset(struct memory_bitmap *bm) +{ + struct zone_bitmap *zone_bm; + + zone_bm = bm->zone_bm_list; + bm->cur.zone_bm = zone_bm; + bm->cur.block = zone_bm->bm_blocks; + memory_bm_reset_chunk(bm); +} + +static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free); + +/** + * create_bm_block_list - create a list of block bitmap objects + */ + +static inline struct bm_block * +create_bm_block_list(unsigned int nr_blocks, struct chain_allocator *ca) +{ + struct bm_block *bblist = NULL; + + while (nr_blocks-- > 0) { + struct bm_block *bb; + + bb = chain_alloc(ca, sizeof(struct bm_block)); + if (!bb) + return NULL; + + bb->next = bblist; + bblist = bb; + } + return bblist; +} + +/** + * create_zone_bm_list - create a list of zone bitmap objects + */ + +static inline struct zone_bitmap * +create_zone_bm_list(unsigned int nr_zones, struct chain_allocator *ca) +{ + struct zone_bitmap *zbmlist = NULL; + + while (nr_zones-- > 0) { + struct zone_bitmap *zbm; + + zbm = chain_alloc(ca, sizeof(struct zone_bitmap)); + if (!zbm) + return NULL; + + zbm->next = zbmlist; + zbmlist = zbm; + } + return zbmlist; +} + +/** + * memory_bm_create - allocate memory for a memory bitmap + */ + +static int +memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) +{ + struct chain_allocator ca; + struct zone *zone; + struct zone_bitmap *zone_bm; + struct bm_block *bb; + unsigned int nr; + + chain_init(&ca, gfp_mask, safe_needed); + + /* Compute the number of zones */ + nr = 0; + for_each_zone (zone) + if (populated_zone(zone) && !is_highmem(zone)) + nr++; + + /* Allocate the list of zones bitmap objects */ + zone_bm = create_zone_bm_list(nr, &ca); + bm->zone_bm_list = zone_bm; + if (!zone_bm) { + chain_free(&ca, PG_UNSAFE_CLEAR); + return -ENOMEM; + } + + /* Initialize the zone bitmap objects */ + for_each_zone (zone) { + unsigned long pfn; + + if (!populated_zone(zone) || is_highmem(zone)) + continue; + + zone_bm->start_pfn = zone->zone_start_pfn; + zone_bm->end_pfn = zone->zone_start_pfn + zone->spanned_pages; + /* Allocate the list of bitmap block objects */ + nr = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK); + bb = create_bm_block_list(nr, &ca); + zone_bm->bm_blocks = bb; + zone_bm->cur_block = bb; + if (!bb) + goto Free; + + nr = zone->spanned_pages; + pfn = zone->zone_start_pfn; + /* Initialize the bitmap block objects */ + while (bb) { + unsigned long *ptr; + + ptr = alloc_image_page(gfp_mask, safe_needed); + bb->data = ptr; + if (!ptr) + goto Free; + + bb->start_pfn = pfn; + if (nr >= BM_BITS_PER_BLOCK) { + pfn += BM_BITS_PER_BLOCK; + bb->size = BM_CHUNKS_PER_BLOCK; + nr -= BM_BITS_PER_BLOCK; + } else { + /* This is executed only once in the loop */ + pfn += nr; + bb->size = DIV_ROUND_UP(nr, BM_BITS_PER_CHUNK); + } + bb->end_pfn = pfn; + bb = bb->next; + } + zone_bm = zone_bm->next; + } + bm->p_list = ca.chain; + memory_bm_position_reset(bm); + return 0; + +Free: + bm->p_list = ca.chain; + memory_bm_free(bm, PG_UNSAFE_CLEAR); + return -ENOMEM; +} + +/** + * memory_bm_free - free memory occupied by the memory bitmap @bm + */ + +static void memory_bm_free(struct memory_bitmap *bm, int clear_nosave_free) +{ + struct zone_bitmap *zone_bm; + + /* Free the list of bit blocks for each zone_bitmap object */ + zone_bm = bm->zone_bm_list; + while (zone_bm) { + struct bm_block *bb; + + bb = zone_bm->bm_blocks; + while (bb) { + if (bb->data) + free_image_page(bb->data, clear_nosave_free); + bb = bb->next; + } + zone_bm = zone_bm->next; + } + free_list_of_pages(bm->p_list, clear_nosave_free); + bm->zone_bm_list = NULL; +} + +/** + * memory_bm_set_bit - set the bit in the bitmap @bm that corresponds + * to given pfn. The cur_zone_bm member of @bm and the cur_block member + * of @bm->cur_zone_bm are updated. + * + * If the bit cannot be set, the function returns -EINVAL . + */ + +static int +memory_bm_set_bit(struct memory_bitmap *bm, unsigned long pfn) +{ + struct zone_bitmap *zone_bm; + struct bm_block *bb; + + /* Check if the pfn is from the current zone */ + zone_bm = bm->cur.zone_bm; + if (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) { + zone_bm = bm->zone_bm_list; + /* We don't assume that the zones are sorted by pfns */ + while (pfn < zone_bm->start_pfn || pfn >= zone_bm->end_pfn) { + zone_bm = zone_bm->next; + if (unlikely(!zone_bm)) + return -EINVAL; + } + bm->cur.zone_bm = zone_bm; + } + /* Check if the pfn corresponds to the current bitmap block */ + bb = zone_bm->cur_block; + if (pfn < bb->start_pfn) + bb = zone_bm->bm_blocks; + + while (pfn >= bb->end_pfn) { + bb = bb->next; + if (unlikely(!bb)) + return -EINVAL; + } + zone_bm->cur_block = bb; + pfn -= bb->start_pfn; + set_bit(pfn % BM_BITS_PER_CHUNK, bb->data + pfn / BM_BITS_PER_CHUNK); + return 0; +} + +/* Two auxiliary functions for memory_bm_next_pfn */ + +/* Find the first set bit in the given chunk, if there is one */ + +static inline int next_bit_in_chunk(int bit, unsigned long *chunk_p) +{ + bit++; + while (bit < BM_BITS_PER_CHUNK) { + if (test_bit(bit, chunk_p)) + return bit; + + bit++; + } + return -1; +} + +/* Find a chunk containing some bits set in given block of bits */ + +static inline int next_chunk_in_block(int n, struct bm_block *bb) +{ + n++; + while (n < bb->size) { + if (bb->data[n]) + return n; + + n++; + } + return -1; +} + +/** + * memory_bm_next_pfn - find the pfn that corresponds to the next set bit + * in the bitmap @bm. If the pfn cannot be found, BM_END_OF_MAP is + * returned. + * + * It is required to run memory_bm_position_reset() before the first call to + * this function. + */ + +static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm) +{ + struct zone_bitmap *zone_bm; + struct bm_block *bb; + int chunk; + int bit; + + do { + bb = bm->cur.block; + do { + chunk = bm->cur.chunk; + bit = bm->cur.bit; + do { + bit = next_bit_in_chunk(bit, bb->data + chunk); + if (bit >= 0) + goto Return_pfn; + + chunk = next_chunk_in_block(chunk, bb); + bit = -1; + } while (chunk >= 0); + bb = bb->next; + bm->cur.block = bb; + memory_bm_reset_chunk(bm); + } while (bb); + zone_bm = bm->cur.zone_bm->next; + if (zone_bm) { + bm->cur.zone_bm = zone_bm; + bm->cur.block = zone_bm->bm_blocks; + memory_bm_reset_chunk(bm); + } + } while (zone_bm); + memory_bm_position_reset(bm); + return BM_END_OF_MAP; + +Return_pfn: + bm->cur.chunk = chunk; + bm->cur.bit = bit; + return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit; +} + +/** + * snapshot_additional_pages - estimate the number of additional pages + * be needed for setting up the suspend image data structures for given + * zone (usually the returned value is greater than the exact number) + */ + +unsigned int snapshot_additional_pages(struct zone *zone) +{ + unsigned int res; + + res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK); + res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE); + return res; +} + /** * pfn_is_nosave - check if given pfn is in the 'nosave' section */ @@ -276,32 +737,38 @@ static inline void copy_data_page(long *dst, long *src) *dst++ = *src++; } -static void copy_data_pages(struct pbe *pblist) +static void +copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) { struct zone *zone; - unsigned long pfn, max_zone_pfn; - struct pbe *pbe; + unsigned long pfn; - pbe = pblist; for_each_zone (zone) { + unsigned long max_zone_pfn; + if (is_highmem(zone)) continue; + mark_free_pages(zone); max_zone_pfn = zone->zone_start_pfn + zone->spanned_pages; - for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) { - struct page *page = saveable_page(pfn); - - if (page) { - void *ptr = page_address(page); - - BUG_ON(!pbe); - copy_data_page((void *)pbe->address, ptr); - pbe->orig_address = (unsigned long)ptr; - pbe = pbe->next; - } - } + for (pfn = zone->zone_start_pfn; pfn < max_zone_pfn; pfn++) + if (saveable_page(pfn)) + memory_bm_set_bit(orig_bm, pfn); } - BUG_ON(pbe); + memory_bm_position_reset(orig_bm); + memory_bm_position_reset(copy_bm); + do { + pfn = memory_bm_next_pfn(orig_bm); + if (likely(pfn != BM_END_OF_MAP)) { + struct page *page; + void *src; + + page = pfn_to_page(pfn); + src = page_address(page); + page = pfn_to_page(memory_bm_next_pfn(copy_bm)); + copy_data_page(page_address(page), src); + } + } while (pfn != BM_END_OF_MAP); } /** @@ -447,37 +914,43 @@ static int enough_free_mem(unsigned int nr_pages) (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE); } -static int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed) +static int +swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, + unsigned int nr_pages) { - struct pbe *p; + int error; - for_each_pbe (p, pblist) { - p->address = (unsigned long)alloc_image_page(gfp_mask, safe_needed); - if (!p->address) - return -ENOMEM; - } - return 0; -} + error = memory_bm_create(orig_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY); + if (error) + goto Free; -static struct pbe *swsusp_alloc(unsigned int nr_pages) -{ - struct pbe *pblist; + error = memory_bm_create(copy_bm, GFP_ATOMIC | __GFP_COLD, PG_ANY); + if (error) + goto Free; - pblist = alloc_pagedir(nr_pages, GFP_ATOMIC | __GFP_COLD, PG_ANY); - if (!pblist) { - printk(KERN_ERR "suspend: Allocating pagedir failed.\n"); - return NULL; - } + while (nr_pages-- > 0) { + struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD); + if (!page) + goto Free; - if (alloc_data_pages(pblist, GFP_ATOMIC | __GFP_COLD, PG_ANY)) { - printk(KERN_ERR "suspend: Allocating image pages failed.\n"); - swsusp_free(); - return NULL; + SetPageNosave(page); + SetPageNosaveFree(page); + memory_bm_set_bit(copy_bm, page_to_pfn(page)); } + return 0; - return pblist; +Free: + swsusp_free(); + return -ENOMEM; } +/* Memory bitmap used for marking saveable pages */ +static struct memory_bitmap orig_bm; +/* Memory bitmap used for marking allocated pages that will contain the copies + * of saveable pages + */ +static struct memory_bitmap copy_bm; + asmlinkage int swsusp_save(void) { unsigned int nr_pages; @@ -498,15 +971,14 @@ asmlinkage int swsusp_save(void) return -ENOMEM; } - restore_pblist = swsusp_alloc(nr_pages); - if (!restore_pblist) + if (swsusp_alloc(&orig_bm, ©_bm, nr_pages)) return -ENOMEM; /* During allocating of suspend pagedir, new cold pages may appear. * Kill them. */ drain_local_pages(); - copy_data_pages(restore_pblist); + copy_data_pages(©_bm, &orig_bm); /* * End of critical section. From now on, we can write to memory, @@ -535,22 +1007,23 @@ static void init_header(struct swsusp_info *info) } /** - * pack_orig_addresses - the .orig_address fields of the PBEs from the - * list starting at @pbe are stored in the array @buf[] (1 page) + * pack_addresses - the addresses corresponding to pfns found in the + * bitmap @bm are stored in the array @buf[] (1 page) */ -static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pbe) +static inline void +pack_addresses(unsigned long *buf, struct memory_bitmap *bm) { int j; - for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) { - buf[j] = pbe->orig_address; - pbe = pbe->next; + for (j = 0; j < PAGE_SIZE / sizeof(long); j++) { + unsigned long pfn = memory_bm_next_pfn(bm); + + if (unlikely(pfn == BM_END_OF_MAP)) + break; + + buf[j] = (unsigned long)page_address(pfn_to_page(pfn)); } - if (!pbe) - for (; j < PAGE_SIZE / sizeof(long); j++) - buf[j] = 0; - return pbe; } /** @@ -579,6 +1052,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) { if (handle->cur > nr_meta_pages + nr_copy_pages) return 0; + if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); @@ -588,16 +1062,17 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) if (!handle->offset) { init_header((struct swsusp_info *)buffer); handle->buffer = buffer; - handle->pbe = restore_pblist; + memory_bm_position_reset(&orig_bm); + memory_bm_position_reset(©_bm); } if (handle->prev < handle->cur) { if (handle->cur <= nr_meta_pages) { - handle->pbe = pack_orig_addresses(buffer, handle->pbe); - if (!handle->pbe) - handle->pbe = restore_pblist; + memset(buffer, 0, PAGE_SIZE); + pack_addresses(buffer, &orig_bm); } else { - handle->buffer = (void *)handle->pbe->address; - handle->pbe = handle->pbe->next; + unsigned long pfn = memory_bm_next_pfn(©_bm); + + handle->buffer = page_address(pfn_to_page(pfn)); } handle->prev = handle->cur; } @@ -736,12 +1211,7 @@ static inline struct pbe *unpack_orig_addresses(unsigned long *buf, * of "safe" which will be used later */ -struct safe_page { - struct safe_page *next; - char padding[PAGE_SIZE - sizeof(void *)]; -}; - -static struct safe_page *safe_pages; +static struct linked_page *safe_pages; static int prepare_image(struct snapshot_handle *handle) { @@ -763,9 +1233,9 @@ static int prepare_image(struct snapshot_handle *handle) if (!error && nr_pages > unsafe_pages) { nr_pages -= unsafe_pages; while (nr_pages--) { - struct safe_page *ptr; + struct linked_page *ptr; - ptr = (struct safe_page *)get_zeroed_page(GFP_ATOMIC); + ptr = (void *)get_zeroed_page(GFP_ATOMIC); if (!ptr) { error = -ENOMEM; break; diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 17f669c83012..8ef677ea0cea 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -193,14 +193,13 @@ int swsusp_shrink_memory(void) printk("Shrinking memory... "); do { size = 2 * count_highmem_pages(); - size += size / 50 + count_data_pages(); - size += (size + PBES_PER_PAGE - 1) / PBES_PER_PAGE + - PAGES_FOR_IO; + size += size / 50 + count_data_pages() + PAGES_FOR_IO; tmp = size; for_each_zone (zone) if (!is_highmem(zone) && populated_zone(zone)) { tmp -= zone->free_pages; tmp += zone->lowmem_reserve[ZONE_NORMAL]; + tmp += snapshot_additional_pages(zone); } if (tmp > 0) { tmp = __shrink_memory(tmp); -- cgit v1.2.3 From 940864ddabdb180e02041c4dcd46ba6f9eee732f Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:55 -0700 Subject: [PATCH] swsusp: Use memory bitmaps during resume Make swsusp use memory bitmaps to store its internal information during the resume phase of the suspend-resume cycle. If the pfns of saveable pages are saved during the suspend phase instead of the kernel virtual addresses of these pages, we can use them during the resume phase directly to set the corresponding bits in a memory bitmap. Then, this bitmap is used to mark the page frames corresponding to the pages that were saveable before the suspend (aka "unsafe" page frames). Next, we allocate as many page frames as needed to store the entire suspend image and make sure that there will be some extra free "safe" page frames for the list of PBEs constructed later. Subsequently, the image is loaded and, if possible, the data loaded from it are written into their "original" page frames (ie. the ones they had occupied before the suspend). The image data that cannot be written into their "original" page frames are loaded into "safe" page frames and their "original" kernel virtual addresses, as well as the addresses of the "safe" pages containing their copies, are stored in a list of PBEs. Finally, the list of PBEs is used to copy the remaining image data into their "original" page frames (this is done atomically, by the architecture-dependent parts of swsusp). Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/suspend.h | 9 -- kernel/power/power.h | 11 +- kernel/power/snapshot.c | 416 +++++++++++++++++++++--------------------------- kernel/power/swap.c | 4 +- kernel/power/user.c | 1 + 5 files changed, 186 insertions(+), 255 deletions(-) (limited to 'kernel') diff --git a/include/linux/suspend.h b/include/linux/suspend.h index c11cacf1a13b..b1237f16ecde 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -16,15 +16,6 @@ struct pbe { struct pbe *next; }; -#define for_each_pbe(pbe, pblist) \ - for (pbe = pblist ; pbe ; pbe = pbe->next) - -#define PBES_PER_PAGE (PAGE_SIZE/sizeof(struct pbe)) -#define PB_PAGE_SKIP (PBES_PER_PAGE-1) - -#define for_each_pb_page(pbe, pblist) \ - for (pbe = pblist ; pbe ; pbe = (pbe+PB_PAGE_SKIP)->next) - /* mm/page_alloc.c */ extern void drain_local_pages(void); extern void mark_free_pages(struct zone *zone); diff --git a/kernel/power/power.h b/kernel/power/power.h index 6e9e2acc34f8..bfe999f7b272 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -81,16 +81,6 @@ struct snapshot_handle { unsigned int prev; /* number of the block of PAGE_SIZE bytes that * was the current one previously */ - struct pbe *pbe; /* PBE that corresponds to 'buffer' */ - struct pbe *last_pbe; /* When the image is restored (eg. read - * from disk) we can store some image - * data directly in the page frames - * in which they were before suspend. - * In such a case the PBEs that - * correspond to them will be unused. - * This is the last PBE, so far, that - * does not correspond to such data. - */ void *buffer; /* address of the block to read from * or write to */ @@ -113,6 +103,7 @@ extern unsigned int snapshot_additional_pages(struct zone *zone); extern int snapshot_read_next(struct snapshot_handle *handle, size_t count); extern int snapshot_write_next(struct snapshot_handle *handle, size_t count); extern int snapshot_image_loaded(struct snapshot_handle *handle); +extern void snapshot_free_unused_memory(struct snapshot_handle *handle); #define SNAPSHOT_IOC_MAGIC '3' #define SNAPSHOT_FREEZE _IO(SNAPSHOT_IOC_MAGIC, 1) diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 852e0df41719..1b84313cbab5 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -39,7 +39,7 @@ struct pbe *restore_pblist; static unsigned int nr_copy_pages; static unsigned int nr_meta_pages; -static unsigned long *buffer; +static void *buffer; #ifdef CONFIG_HIGHMEM unsigned int count_highmem_pages(void) @@ -172,7 +172,7 @@ static inline int restore_highmem(void) {return 0;} #define PG_UNSAFE_CLEAR 1 #define PG_UNSAFE_KEEP 0 -static unsigned int unsafe_pages; +static unsigned int allocated_unsafe_pages; static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) { @@ -183,7 +183,7 @@ static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) while (res && PageNosaveFree(virt_to_page(res))) { /* The page is unsafe, mark it for swsusp_free() */ SetPageNosave(virt_to_page(res)); - unsafe_pages++; + allocated_unsafe_pages++; res = (void *)get_zeroed_page(gfp_mask); } if (res) { @@ -772,101 +772,10 @@ copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) } /** - * free_pagedir - free pages allocated with alloc_pagedir() - */ - -static void free_pagedir(struct pbe *pblist, int clear_nosave_free) -{ - struct pbe *pbe; - - while (pblist) { - pbe = (pblist + PB_PAGE_SKIP)->next; - free_image_page(pblist, clear_nosave_free); - pblist = pbe; - } -} - -/** - * fill_pb_page - Create a list of PBEs on a given memory page - */ - -static inline void fill_pb_page(struct pbe *pbpage, unsigned int n) -{ - struct pbe *p; - - p = pbpage; - pbpage += n - 1; - do - p->next = p + 1; - while (++p < pbpage); -} - -/** - * create_pbe_list - Create a list of PBEs on top of a given chain - * of memory pages allocated with alloc_pagedir() + * swsusp_free - free pages allocated for the suspend. * - * This function assumes that pages allocated by alloc_image_page() will - * always be zeroed. - */ - -static inline void create_pbe_list(struct pbe *pblist, unsigned int nr_pages) -{ - struct pbe *pbpage; - unsigned int num = PBES_PER_PAGE; - - for_each_pb_page (pbpage, pblist) { - if (num >= nr_pages) - break; - - fill_pb_page(pbpage, PBES_PER_PAGE); - num += PBES_PER_PAGE; - } - if (pbpage) { - num -= PBES_PER_PAGE; - fill_pb_page(pbpage, nr_pages - num); - } -} - -/** - * alloc_pagedir - Allocate the page directory. - * - * First, determine exactly how many pages we need and - * allocate them. - * - * We arrange the pages in a chain: each page is an array of PBES_PER_PAGE - * struct pbe elements (pbes) and the last element in the page points - * to the next page. - * - * On each page we set up a list of struct_pbe elements. - */ - -static struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, - int safe_needed) -{ - unsigned int num; - struct pbe *pblist, *pbe; - - if (!nr_pages) - return NULL; - - pblist = alloc_image_page(gfp_mask, safe_needed); - pbe = pblist; - for (num = PBES_PER_PAGE; num < nr_pages; num += PBES_PER_PAGE) { - if (!pbe) { - free_pagedir(pblist, PG_UNSAFE_CLEAR); - return NULL; - } - pbe += PB_PAGE_SKIP; - pbe->next = alloc_image_page(gfp_mask, safe_needed); - pbe = pbe->next; - } - create_pbe_list(pblist, nr_pages); - return pblist; -} - -/** - * Free pages we allocated for suspend. Suspend pages are alocated - * before atomic copy, so we need to free them after resume. + * Suspend pages are alocated before the atomic copy is made, so we + * need to release them after the resume. */ void swsusp_free(void) @@ -904,14 +813,18 @@ void swsusp_free(void) static int enough_free_mem(unsigned int nr_pages) { struct zone *zone; - unsigned int n = 0; + unsigned int free = 0, meta = 0; for_each_zone (zone) - if (!is_highmem(zone)) - n += zone->free_pages; - pr_debug("swsusp: available memory: %u pages\n", n); - return n > (nr_pages + PAGES_FOR_IO + - (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE); + if (!is_highmem(zone)) { + free += zone->free_pages; + meta += snapshot_additional_pages(zone); + } + + pr_debug("swsusp: pages needed: %u + %u + %u, available pages: %u\n", + nr_pages, PAGES_FOR_IO, meta, free); + + return free > nr_pages + PAGES_FOR_IO + meta; } static int @@ -961,11 +874,6 @@ asmlinkage int swsusp_save(void) nr_pages = count_data_pages(); printk("swsusp: Need to copy %u pages\n", nr_pages); - pr_debug("swsusp: pages needed: %u + %lu + %u, free: %u\n", - nr_pages, - (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE, - PAGES_FOR_IO, nr_free_pages()); - if (!enough_free_mem(nr_pages)) { printk(KERN_ERR "swsusp: Not enough free memory\n"); return -ENOMEM; @@ -1007,22 +915,19 @@ static void init_header(struct swsusp_info *info) } /** - * pack_addresses - the addresses corresponding to pfns found in the - * bitmap @bm are stored in the array @buf[] (1 page) + * pack_pfns - pfns corresponding to the set bits found in the bitmap @bm + * are stored in the array @buf[] (1 page at a time) */ static inline void -pack_addresses(unsigned long *buf, struct memory_bitmap *bm) +pack_pfns(unsigned long *buf, struct memory_bitmap *bm) { int j; for (j = 0; j < PAGE_SIZE / sizeof(long); j++) { - unsigned long pfn = memory_bm_next_pfn(bm); - - if (unlikely(pfn == BM_END_OF_MAP)) + buf[j] = memory_bm_next_pfn(bm); + if (unlikely(buf[j] == BM_END_OF_MAP)) break; - - buf[j] = (unsigned long)page_address(pfn_to_page(pfn)); } } @@ -1068,7 +973,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) if (handle->prev < handle->cur) { if (handle->cur <= nr_meta_pages) { memset(buffer, 0, PAGE_SIZE); - pack_addresses(buffer, &orig_bm); + pack_pfns(buffer, &orig_bm); } else { unsigned long pfn = memory_bm_next_pfn(©_bm); @@ -1094,14 +999,10 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) * had been used before suspend */ -static int mark_unsafe_pages(struct pbe *pblist) +static int mark_unsafe_pages(struct memory_bitmap *bm) { struct zone *zone; unsigned long pfn, max_zone_pfn; - struct pbe *p; - - if (!pblist) /* a sanity check */ - return -EINVAL; /* Clear page flags */ for_each_zone (zone) { @@ -1111,30 +1012,37 @@ static int mark_unsafe_pages(struct pbe *pblist) ClearPageNosaveFree(pfn_to_page(pfn)); } - /* Mark orig addresses */ - for_each_pbe (p, pblist) { - if (virt_addr_valid(p->orig_address)) - SetPageNosaveFree(virt_to_page(p->orig_address)); - else - return -EFAULT; - } + /* Mark pages that correspond to the "original" pfns as "unsafe" */ + memory_bm_position_reset(bm); + do { + pfn = memory_bm_next_pfn(bm); + if (likely(pfn != BM_END_OF_MAP)) { + if (likely(pfn_valid(pfn))) + SetPageNosaveFree(pfn_to_page(pfn)); + else + return -EFAULT; + } + } while (pfn != BM_END_OF_MAP); - unsafe_pages = 0; + allocated_unsafe_pages = 0; return 0; } -static void copy_page_backup_list(struct pbe *dst, struct pbe *src) +static void +duplicate_memory_bitmap(struct memory_bitmap *dst, struct memory_bitmap *src) { - /* We assume both lists contain the same number of elements */ - while (src) { - dst->orig_address = src->orig_address; - dst = dst->next; - src = src->next; + unsigned long pfn; + + memory_bm_position_reset(src); + pfn = memory_bm_next_pfn(src); + while (pfn != BM_END_OF_MAP) { + memory_bm_set_bit(dst, pfn); + pfn = memory_bm_next_pfn(src); } } -static int check_header(struct swsusp_info *info) +static inline int check_header(struct swsusp_info *info) { char *reason = NULL; @@ -1161,19 +1069,14 @@ static int check_header(struct swsusp_info *info) * load header - check the image header and copy data from it */ -static int load_header(struct snapshot_handle *handle, - struct swsusp_info *info) +static int +load_header(struct swsusp_info *info) { int error; - struct pbe *pblist; + restore_pblist = NULL; error = check_header(info); if (!error) { - pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, PG_ANY); - if (!pblist) - return -ENOMEM; - restore_pblist = pblist; - handle->pbe = pblist; nr_copy_pages = info->image_pages; nr_meta_pages = info->pages - info->image_pages - 1; } @@ -1181,108 +1084,137 @@ static int load_header(struct snapshot_handle *handle, } /** - * unpack_orig_addresses - copy the elements of @buf[] (1 page) to - * the PBEs in the list starting at @pbe + * unpack_orig_pfns - for each element of @buf[] (1 page at a time) set + * the corresponding bit in the memory bitmap @bm */ -static inline struct pbe *unpack_orig_addresses(unsigned long *buf, - struct pbe *pbe) +static inline void +unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) { int j; - for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) { - pbe->orig_address = buf[j]; - pbe = pbe->next; + for (j = 0; j < PAGE_SIZE / sizeof(long); j++) { + if (unlikely(buf[j] == BM_END_OF_MAP)) + break; + + memory_bm_set_bit(bm, buf[j]); } - return pbe; } /** - * prepare_image - use metadata contained in the PBE list - * pointed to by restore_pblist to mark the pages that will - * be overwritten in the process of restoring the system - * memory state from the image ("unsafe" pages) and allocate - * memory for the image + * prepare_image - use the memory bitmap @bm to mark the pages that will + * be overwritten in the process of restoring the system memory state + * from the suspend image ("unsafe" pages) and allocate memory for the + * image. * - * The idea is to allocate the PBE list first and then - * allocate as many pages as it's needed for the image data, - * but not to assign these pages to the PBEs initially. - * Instead, we just mark them as allocated and create a list - * of "safe" which will be used later + * The idea is to allocate a new memory bitmap first and then allocate + * as many pages as needed for the image data, but not to assign these + * pages to specific tasks initially. Instead, we just mark them as + * allocated and create a list of "safe" pages that will be used later. */ -static struct linked_page *safe_pages; +#define PBES_PER_LINKED_PAGE (LINKED_PAGE_DATA_SIZE / sizeof(struct pbe)) + +static struct linked_page *safe_pages_list; -static int prepare_image(struct snapshot_handle *handle) +static int +prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) { - int error = 0; - unsigned int nr_pages = nr_copy_pages; - struct pbe *p, *pblist = NULL; + unsigned int nr_pages; + struct linked_page *sp_list, *lp; + int error; - p = restore_pblist; - error = mark_unsafe_pages(p); - if (!error) { - pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, PG_SAFE); - if (pblist) - copy_page_backup_list(pblist, p); - free_pagedir(p, PG_UNSAFE_KEEP); - if (!pblist) + error = mark_unsafe_pages(bm); + if (error) + goto Free; + + error = memory_bm_create(new_bm, GFP_ATOMIC, PG_SAFE); + if (error) + goto Free; + + duplicate_memory_bitmap(new_bm, bm); + memory_bm_free(bm, PG_UNSAFE_KEEP); + /* Reserve some safe pages for potential later use. + * + * NOTE: This way we make sure there will be enough safe pages for the + * chain_alloc() in get_buffer(). It is a bit wasteful, but + * nr_copy_pages cannot be greater than 50% of the memory anyway. + */ + sp_list = NULL; + /* nr_copy_pages cannot be lesser than allocated_unsafe_pages */ + nr_pages = nr_copy_pages - allocated_unsafe_pages; + nr_pages = DIV_ROUND_UP(nr_pages, PBES_PER_LINKED_PAGE); + while (nr_pages > 0) { + lp = alloc_image_page(GFP_ATOMIC, PG_SAFE); + if (!lp) { error = -ENOMEM; + goto Free; + } + lp->next = sp_list; + sp_list = lp; + nr_pages--; } - safe_pages = NULL; - if (!error && nr_pages > unsafe_pages) { - nr_pages -= unsafe_pages; - while (nr_pages--) { - struct linked_page *ptr; - - ptr = (void *)get_zeroed_page(GFP_ATOMIC); - if (!ptr) { - error = -ENOMEM; - break; - } - if (!PageNosaveFree(virt_to_page(ptr))) { - /* The page is "safe", add it to the list */ - ptr->next = safe_pages; - safe_pages = ptr; - } - /* Mark the page as allocated */ - SetPageNosave(virt_to_page(ptr)); - SetPageNosaveFree(virt_to_page(ptr)); + /* Preallocate memory for the image */ + safe_pages_list = NULL; + nr_pages = nr_copy_pages - allocated_unsafe_pages; + while (nr_pages > 0) { + lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC); + if (!lp) { + error = -ENOMEM; + goto Free; + } + if (!PageNosaveFree(virt_to_page(lp))) { + /* The page is "safe", add it to the list */ + lp->next = safe_pages_list; + safe_pages_list = lp; } + /* Mark the page as allocated */ + SetPageNosave(virt_to_page(lp)); + SetPageNosaveFree(virt_to_page(lp)); + nr_pages--; } - if (!error) { - restore_pblist = pblist; - } else { - handle->pbe = NULL; - swsusp_free(); + /* Free the reserved safe pages so that chain_alloc() can use them */ + while (sp_list) { + lp = sp_list->next; + free_image_page(sp_list, PG_UNSAFE_CLEAR); + sp_list = lp; } + return 0; + +Free: + swsusp_free(); return error; } -static void *get_buffer(struct snapshot_handle *handle) +/** + * get_buffer - compute the address that snapshot_write_next() should + * set for its caller to write to. + */ + +static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca) { - struct pbe *pbe = handle->pbe, *last = handle->last_pbe; - struct page *page = virt_to_page(pbe->orig_address); + struct pbe *pbe; + struct page *page = pfn_to_page(memory_bm_next_pfn(bm)); - if (PageNosave(page) && PageNosaveFree(page)) { - /* - * We have allocated the "original" page frame and we can - * use it directly to store the read page + if (PageNosave(page) && PageNosaveFree(page)) + /* We have allocated the "original" page frame and we can + * use it directly to store the loaded page. */ - pbe->address = 0; - if (last && last->next) - last->next = NULL; - return (void *)pbe->orig_address; - } - /* - * The "original" page frame has not been allocated and we have to - * use a "safe" page frame to store the read page + return page_address(page); + + /* The "original" page frame has not been allocated and we have to + * use a "safe" page frame to store the loaded page. */ - pbe->address = (unsigned long)safe_pages; - safe_pages = safe_pages->next; - if (last) - last->next = pbe; - handle->last_pbe = pbe; + pbe = chain_alloc(ca, sizeof(struct pbe)); + if (!pbe) { + swsusp_free(); + return NULL; + } + pbe->orig_address = (unsigned long)page_address(page); + pbe->address = (unsigned long)safe_pages_list; + safe_pages_list = safe_pages_list->next; + pbe->next = restore_pblist; + restore_pblist = pbe; return (void *)pbe->address; } @@ -1310,10 +1242,13 @@ static void *get_buffer(struct snapshot_handle *handle) int snapshot_write_next(struct snapshot_handle *handle, size_t count) { + static struct chain_allocator ca; int error = 0; + /* Check if we have already loaded the entire image */ if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) return 0; + if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); @@ -1324,26 +1259,32 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) handle->buffer = buffer; handle->sync_read = 1; if (handle->prev < handle->cur) { - if (!handle->prev) { - error = load_header(handle, - (struct swsusp_info *)buffer); + if (handle->prev == 0) { + error = load_header(buffer); + if (error) + return error; + + error = memory_bm_create(©_bm, GFP_ATOMIC, PG_ANY); if (error) return error; + } else if (handle->prev <= nr_meta_pages) { - handle->pbe = unpack_orig_addresses(buffer, - handle->pbe); - if (!handle->pbe) { - error = prepare_image(handle); + unpack_orig_pfns(buffer, ©_bm); + if (handle->prev == nr_meta_pages) { + error = prepare_image(&orig_bm, ©_bm); if (error) return error; - handle->pbe = restore_pblist; - handle->last_pbe = NULL; - handle->buffer = get_buffer(handle); + + chain_init(&ca, GFP_ATOMIC, PG_SAFE); + memory_bm_position_reset(&orig_bm); + restore_pblist = NULL; + handle->buffer = get_buffer(&orig_bm, &ca); handle->sync_read = 0; + if (!handle->buffer) + return -ENOMEM; } } else { - handle->pbe = handle->pbe->next; - handle->buffer = get_buffer(handle); + handle->buffer = get_buffer(&orig_bm, &ca); handle->sync_read = 0; } handle->prev = handle->cur; @@ -1362,6 +1303,13 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) int snapshot_image_loaded(struct snapshot_handle *handle) { - return !(!handle->pbe || handle->pbe->next || !nr_copy_pages || - handle->cur <= nr_meta_pages + nr_copy_pages); + return !(!nr_copy_pages || + handle->cur <= nr_meta_pages + nr_copy_pages); +} + +void snapshot_free_unused_memory(struct snapshot_handle *handle) +{ + /* Free only if we have loaded the image entirely */ + if (handle->prev && handle->cur > nr_meta_pages + nr_copy_pages) + memory_bm_free(&orig_bm, PG_UNSAFE_CLEAR); } diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 8309d20b2563..9b2ee5344dee 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -331,8 +331,7 @@ static int enough_swap(unsigned int nr_pages) unsigned int free_swap = count_swap_pages(root_swap, 1); pr_debug("swsusp: free swap pages: %u\n", free_swap); - return free_swap > (nr_pages + PAGES_FOR_IO + - (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE); + return free_swap > nr_pages + PAGES_FOR_IO; } /** @@ -547,6 +546,7 @@ static int load_image(struct swap_map_handle *handle, error = err2; if (!error) { printk("\b\b\b\bdone\n"); + snapshot_free_unused_memory(snapshot); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; } diff --git a/kernel/power/user.c b/kernel/power/user.c index 0ef5e4ba39e5..2e4499f3e4d9 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -193,6 +193,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -EPERM; break; } + snapshot_free_unused_memory(&data->handle); down(&pm_sem); pm_prepare_console(); error = device_suspend(PMSG_FREEZE); -- cgit v1.2.3 From c8eb8b4025175f967af0ba8e933f23aa9954dc35 Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:56 -0700 Subject: [PATCH] PM: make it possible to disable console suspending Change suspend_console() so that it waits for all consoles to flush the remaining messages and make it possible to switch the console suspending off with the help of a Kconfig option. Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Cc: Stefan Seyfried Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/console.h | 5 +++++ kernel/power/Kconfig | 11 +++++++++++ kernel/printk.c | 3 +++ 3 files changed, 19 insertions(+) (limited to 'kernel') diff --git a/include/linux/console.h b/include/linux/console.h index 3bdf2155e565..76a1807726eb 100644 --- a/include/linux/console.h +++ b/include/linux/console.h @@ -120,9 +120,14 @@ extern void console_stop(struct console *); extern void console_start(struct console *); extern int is_console_locked(void); +#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND /* Suspend and resume console messages over PM events */ extern void suspend_console(void); extern void resume_console(void); +#else +static inline void suspend_console(void) {} +static inline void resume_console(void) {} +#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */ /* Some debug stub to catch some of the obvious races in the VT code */ #if 1 diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 619ecabf7c58..4b6e2f18e056 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -36,6 +36,17 @@ config PM_DEBUG code. This is helpful when debugging and reporting various PM bugs, like suspend support. +config DISABLE_CONSOLE_SUSPEND + bool "Keep console(s) enabled during suspend/resume (DANGEROUS)" + depends on PM && PM_DEBUG + default n + ---help--- + This option turns off the console suspend mechanism that prevents + debug messages from reaching the console during the suspend/resume + operations. This may be helpful when debugging device drivers' + suspend/resume routines, but may itself lead to problems, for example + if netconsole is used. + config PM_TRACE bool "Suspend/resume event tracing" depends on PM && PM_DEBUG && X86_32 && EXPERIMENTAL diff --git a/kernel/printk.c b/kernel/printk.c index 1149365e989e..771f5e861bcd 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -721,6 +721,7 @@ int __init add_preferred_console(char *name, int idx, char *options) return 0; } +#ifndef CONFIG_DISABLE_CONSOLE_SUSPEND /** * suspend_console - suspend the console subsystem * @@ -728,6 +729,7 @@ int __init add_preferred_console(char *name, int idx, char *options) */ void suspend_console(void) { + printk("Suspending console(s)\n"); acquire_console_sem(); console_suspended = 1; } @@ -737,6 +739,7 @@ void resume_console(void) console_suspended = 0; release_console_sem(); } +#endif /* CONFIG_DISABLE_CONSOLE_SUSPEND */ /** * acquire_console_sem - lock the console system for exclusive use. -- cgit v1.2.3 From c5c6ba4e08ab9c9e390a0f3a7d9a5c332f5cc6ef Mon Sep 17 00:00:00 2001 From: "Rafael J. Wysocki" Date: Mon, 25 Sep 2006 23:32:58 -0700 Subject: [PATCH] PM: Add pm_trace switch Add the pm_trace attribute in /sys/power which has to be explicitly set to one to really enable the "PM tracing" code compiled in when CONFIG_PM_TRACE is set (which modifies the machine's CMOS clock in unpredictable ways). Signed-off-by: Rafael J. Wysocki Acked-by: Pavel Machek Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- Documentation/power/interface.txt | 15 +++++++++++++++ include/linux/resume-trace.h | 24 ++++++++++++++---------- kernel/power/main.c | 30 ++++++++++++++++++++++++++++++ 3 files changed, 59 insertions(+), 10 deletions(-) (limited to 'kernel') diff --git a/Documentation/power/interface.txt b/Documentation/power/interface.txt index 4117802af0f8..a66bec222b16 100644 --- a/Documentation/power/interface.txt +++ b/Documentation/power/interface.txt @@ -52,3 +52,18 @@ suspend image will be as small as possible. Reading from this file will display the current image size limit, which is set to 500 MB by default. + +/sys/power/pm_trace controls the code which saves the last PM event point in +the RTC across reboots, so that you can debug a machine that just hangs +during suspend (or more commonly, during resume). Namely, the RTC is only +used to save the last PM event point if this file contains '1'. Initially it +contains '0' which may be changed to '1' by writing a string representing a +nonzero integer into it. + +To use this debugging feature you should attempt to suspend the machine, then +reboot it and run + + dmesg -s 1000000 | grep 'hash matches' + +CAUTION: Using it will cause your machine's real-time (CMOS) clock to be +set to a random invalid time after a resume. diff --git a/include/linux/resume-trace.h b/include/linux/resume-trace.h index a376bd4ade39..81e9299ca148 100644 --- a/include/linux/resume-trace.h +++ b/include/linux/resume-trace.h @@ -3,21 +3,25 @@ #ifdef CONFIG_PM_TRACE +extern int pm_trace_enabled; + struct device; extern void set_trace_device(struct device *); extern void generate_resume_trace(void *tracedata, unsigned int user); #define TRACE_DEVICE(dev) set_trace_device(dev) -#define TRACE_RESUME(user) do { \ - void *tracedata; \ - asm volatile("movl $1f,%0\n" \ - ".section .tracedata,\"a\"\n" \ - "1:\t.word %c1\n" \ - "\t.long %c2\n" \ - ".previous" \ - :"=r" (tracedata) \ - : "i" (__LINE__), "i" (__FILE__)); \ - generate_resume_trace(tracedata, user); \ +#define TRACE_RESUME(user) do { \ + if (pm_trace_enabled) { \ + void *tracedata; \ + asm volatile("movl $1f,%0\n" \ + ".section .tracedata,\"a\"\n" \ + "1:\t.word %c1\n" \ + "\t.long %c2\n" \ + ".previous" \ + :"=r" (tracedata) \ + : "i" (__LINE__), "i" (__FILE__)); \ + generate_resume_trace(tracedata, user); \ + } \ } while (0) #else diff --git a/kernel/power/main.c b/kernel/power/main.c index 4d403323a7bb..873228c71dab 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -17,6 +17,7 @@ #include #include #include +#include #include "power.h" @@ -281,10 +282,39 @@ static ssize_t state_store(struct subsystem * subsys, const char * buf, size_t n power_attr(state); +#ifdef CONFIG_PM_TRACE +int pm_trace_enabled; + +static ssize_t pm_trace_show(struct subsystem * subsys, char * buf) +{ + return sprintf(buf, "%d\n", pm_trace_enabled); +} + +static ssize_t +pm_trace_store(struct subsystem * subsys, const char * buf, size_t n) +{ + int val; + + if (sscanf(buf, "%d", &val) == 1) { + pm_trace_enabled = !!val; + return n; + } + return -EINVAL; +} + +power_attr(pm_trace); + +static struct attribute * g[] = { + &state_attr.attr, + &pm_trace_attr.attr, + NULL, +}; +#else static struct attribute * g[] = { &state_attr.attr, NULL, }; +#endif /* CONFIG_PM_TRACE */ static struct attribute_group attr_group = { .attrs = g, -- cgit v1.2.3 From d33b6fba2c4350651f3f61ff2ab858a2f116e9a4 Mon Sep 17 00:00:00 2001 From: Matthew Wilcox Date: Fri, 30 Jun 2006 02:31:24 -0700 Subject: Resources: insert identical resources above existing resources If you have two resources which aree exactly the same size, insert_resource() currently inserts the new one below the existing one. This is wrong because there's no way to insert a resource of the same size above an existing one. I took this opportunity to rewrite the initial loop to be a for-loop instead of a goto-loop and fix the documentation. Signed-off-by: Matthew Wilcox Cc: Ivan Kokshaysky Cc: Dominik Brodowski Signed-off-by: Andrew Morton Signed-off-by: Greg Kroah-Hartman --- kernel/resource.c | 32 ++++++++++++++++---------------- 1 file changed, 16 insertions(+), 16 deletions(-) (limited to 'kernel') diff --git a/kernel/resource.c b/kernel/resource.c index 46286434af80..9db38a1a7520 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -344,12 +344,11 @@ EXPORT_SYMBOL(allocate_resource); * * Returns 0 on success, -EBUSY if the resource can't be inserted. * - * This function is equivalent of request_resource when no conflict + * This function is equivalent to request_resource when no conflict * happens. If a conflict happens, and the conflicting resources * entirely fit within the range of the new resource, then the new - * resource is inserted and the conflicting resources become childs of - * the new resource. Otherwise the new resource becomes the child of - * the conflicting resource + * resource is inserted and the conflicting resources become children of + * the new resource. */ int insert_resource(struct resource *parent, struct resource *new) { @@ -357,20 +356,21 @@ int insert_resource(struct resource *parent, struct resource *new) struct resource *first, *next; write_lock(&resource_lock); - begin: - result = 0; - first = __request_resource(parent, new); - if (!first) - goto out; - result = -EBUSY; - if (first == parent) - goto out; + for (;; parent = first) { + result = 0; + first = __request_resource(parent, new); + if (!first) + goto out; - /* Resource fully contained by the clashing resource? Recurse into it */ - if (first->start <= new->start && first->end >= new->end) { - parent = first; - goto begin; + result = -EBUSY; + if (first == parent) + goto out; + + if ((first->start > new->start) || (first->end < new->end)) + break; + if ((first->start == new->start) && (first->end == new->end)) + break; } for (next = first; ; next = next->sibling) { -- cgit v1.2.3 From 0ec76a110f432e98277e464b82ace8dd66571689 Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Sep 2006 01:50:15 -0700 Subject: [PATCH] NOMMU: Check that access_process_vm() has a valid target Check that access_process_vm() is accessing a valid mapping in the target process. This limits ptrace() accesses and accesses through /proc//maps to only those regions actually mapped by a program. Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/ptrace.c | 54 ------------------------------------------------------ mm/memory.c | 53 +++++++++++++++++++++++++++++++++++++++++++++++++++++ mm/nommu.c | 47 +++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 100 insertions(+), 54 deletions(-) (limited to 'kernel') diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 9a111f70145c..8aad0331d82e 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -241,60 +241,6 @@ int ptrace_detach(struct task_struct *child, unsigned int data) return 0; } -/* - * Access another process' address space. - * Source/target buffer must be kernel space, - * Do not walk the page table directly, use get_user_pages - */ - -int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write) -{ - struct mm_struct *mm; - struct vm_area_struct *vma; - struct page *page; - void *old_buf = buf; - - mm = get_task_mm(tsk); - if (!mm) - return 0; - - down_read(&mm->mmap_sem); - /* ignore errors, just check how much was sucessfully transfered */ - while (len) { - int bytes, ret, offset; - void *maddr; - - ret = get_user_pages(tsk, mm, addr, 1, - write, 1, &page, &vma); - if (ret <= 0) - break; - - bytes = len; - offset = addr & (PAGE_SIZE-1); - if (bytes > PAGE_SIZE-offset) - bytes = PAGE_SIZE-offset; - - maddr = kmap(page); - if (write) { - copy_to_user_page(vma, page, addr, - maddr + offset, buf, bytes); - set_page_dirty_lock(page); - } else { - copy_from_user_page(vma, page, addr, - buf, maddr + offset, bytes); - } - kunmap(page); - page_cache_release(page); - len -= bytes; - buf += bytes; - addr += bytes; - } - up_read(&mm->mmap_sem); - mmput(mm); - - return buf - old_buf; -} - int ptrace_readdata(struct task_struct *tsk, unsigned long src, char __user *dst, int len) { int copied = 0; diff --git a/mm/memory.c b/mm/memory.c index f2ef1dcfff77..601159a46ab6 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2604,3 +2604,56 @@ int in_gate_area_no_task(unsigned long addr) } #endif /* __HAVE_ARCH_GATE_AREA */ + +/* + * Access another process' address space. + * Source/target buffer must be kernel space, + * Do not walk the page table directly, use get_user_pages + */ +int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write) +{ + struct mm_struct *mm; + struct vm_area_struct *vma; + struct page *page; + void *old_buf = buf; + + mm = get_task_mm(tsk); + if (!mm) + return 0; + + down_read(&mm->mmap_sem); + /* ignore errors, just check how much was sucessfully transfered */ + while (len) { + int bytes, ret, offset; + void *maddr; + + ret = get_user_pages(tsk, mm, addr, 1, + write, 1, &page, &vma); + if (ret <= 0) + break; + + bytes = len; + offset = addr & (PAGE_SIZE-1); + if (bytes > PAGE_SIZE-offset) + bytes = PAGE_SIZE-offset; + + maddr = kmap(page); + if (write) { + copy_to_user_page(vma, page, addr, + maddr + offset, buf, bytes); + set_page_dirty_lock(page); + } else { + copy_from_user_page(vma, page, addr, + buf, maddr + offset, bytes); + } + kunmap(page); + page_cache_release(page); + len -= bytes; + buf += bytes; + addr += bytes; + } + up_read(&mm->mmap_sem); + mmput(mm); + + return buf - old_buf; +} diff --git a/mm/nommu.c b/mm/nommu.c index d99dea31e443..d08acdae0036 100644 --- a/mm/nommu.c +++ b/mm/nommu.c @@ -1206,3 +1206,50 @@ struct page *filemap_nopage(struct vm_area_struct *area, BUG(); return NULL; } + +/* + * Access another process' address space. + * - source/target buffer must be kernel space + */ +int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write) +{ + struct vm_list_struct *vml; + struct vm_area_struct *vma; + struct mm_struct *mm; + + if (addr + len < addr) + return 0; + + mm = get_task_mm(tsk); + if (!mm) + return 0; + + down_read(&mm->mmap_sem); + + /* the access must start within one of the target process's mappings */ + for (vml = mm->context.vmlist; vml; vml = vml->next) + if (addr >= vml->vma->vm_start && addr < vml->vma->vm_end) + break; + + if (vml) { + vma = vml->vma; + + /* don't overrun this mapping */ + if (addr + len >= vma->vm_end) + len = vma->vm_end - addr; + + /* only read or write mappings where it is permitted */ + if (write && vma->vm_flags & VM_WRITE) + len -= copy_to_user((void *) addr, buf, len); + else if (!write && vma->vm_flags & VM_READ) + len -= copy_from_user(buf, (void *) addr, len); + else + len = 0; + } else { + len = 0; + } + + up_read(&mm->mmap_sem); + mmput(mm); + return len; +} -- cgit v1.2.3 From f269fdd1829acc5e53bf57b145003e5733133f2b Mon Sep 17 00:00:00 2001 From: David Howells Date: Wed, 27 Sep 2006 01:50:23 -0700 Subject: [PATCH] NOMMU: move the fallback arch_vma_name() to a sensible place Move the fallback arch_vma_name() to a sensible place (kernel/signal.c). Currently it's in fs/proc/task_mmu.c, a file that is dependent on both CONFIG_PROC_FS and CONFIG_MMU being enabled, but it's used from kernel/signal.c from where it is called unconditionally. [akpm@osdl.org: build fix] Signed-off-by: David Howells Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/proc/task_mmu.c | 5 ----- include/linux/mm.h | 2 +- kernel/signal.c | 5 +++++ 3 files changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 0a163a4f7764..6b769afac55a 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -122,11 +122,6 @@ struct mem_size_stats unsigned long private_dirty; }; -__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma) -{ - return NULL; -} - static int show_map_internal(struct seq_file *m, void *v, struct mem_size_stats *mss) { struct proc_maps_private *priv = m->private; diff --git a/include/linux/mm.h b/include/linux/mm.h index 22165cb18906..7b703b6d4358 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1131,7 +1131,7 @@ void drop_slab(void); extern int randomize_va_space; #endif -const char *arch_vma_name(struct vm_area_struct *vma); +__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma); #endif /* __KERNEL__ */ #endif /* _LINUX_MM_H */ diff --git a/kernel/signal.c b/kernel/signal.c index bfdb5686fa3e..05853a7337e3 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2577,6 +2577,11 @@ asmlinkage long sys_rt_sigsuspend(sigset_t __user *unewset, size_t sigsetsize) } #endif /* __ARCH_WANT_SYS_RT_SIGSUSPEND */ +__attribute__((weak)) const char *arch_vma_name(struct vm_area_struct *vma) +{ + return NULL; +} + void __init signals_init(void) { sigqueue_cachep = -- cgit v1.2.3 From 8e18e2941c53416aa219708e7dcad21fb4bd6794 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 27 Sep 2006 01:50:46 -0700 Subject: [PATCH] inode_diet: Replace inode.u.generic_ip with inode.i_private The following patches reduce the size of the VFS inode structure by 28 bytes on a UP x86. (It would be more on an x86_64 system). This is a 10% reduction in the inode size on a UP kernel that is configured in a production mode (i.e., with no spinlock or other debugging functions enabled; if you want to save memory taken up by in-core inodes, the first thing you should do is disable the debugging options; they are responsible for a huge amount of bloat in the VFS inode structure). This patch: The filesystem or device-specific pointer in the inode is inside a union, which is pretty pointless given that all 30+ users of this field have been using the void pointer. Get rid of the union and rename it to i_private, with a comment to explain who is allowed to use the void pointer. This is just a cleanup, but it allows us to reuse the union 'u' for something something where the union will actually be used. [judith@osdl.org: powerpc build fix] Signed-off-by: "Theodore Ts'o" Signed-off-by: Judith Lebzelter Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/cell/spufs/inode.c | 2 +- arch/powerpc/platforms/pseries/hvCall_inst.c | 2 +- arch/s390/hypfs/inode.c | 6 ++--- arch/s390/kernel/debug.c | 2 +- block/blktrace.c | 2 +- drivers/i2c/chips/tps65010.c | 2 +- drivers/infiniband/hw/ipath/ipath_fs.c | 12 ++++----- drivers/infiniband/ulp/ipoib/ipoib_fs.c | 4 +-- drivers/misc/ibmasm/ibmasmfs.c | 16 ++++++------ drivers/net/irda/vlsi_ir.h | 2 +- drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c | 2 +- drivers/oprofile/oprofilefs.c | 10 ++++---- drivers/pci/hotplug/cpqphp_sysfs.c | 2 +- drivers/usb/core/devio.c | 2 +- drivers/usb/core/inode.c | 6 ++--- drivers/usb/gadget/inode.c | 6 ++--- drivers/usb/host/isp116x-hcd.c | 2 +- drivers/usb/host/uhci-debug.c | 2 +- drivers/usb/mon/mon_stat.c | 2 +- drivers/usb/mon/mon_text.c | 4 +-- fs/autofs/inode.c | 2 +- fs/autofs/symlink.c | 2 +- fs/binfmt_misc.c | 8 +++--- fs/debugfs/file.c | 4 +-- fs/debugfs/inode.c | 4 +-- fs/devpts/inode.c | 4 +-- fs/freevxfs/vxfs.h | 2 +- fs/freevxfs/vxfs_inode.c | 4 +-- fs/fuse/control.c | 6 ++--- fs/inode.c | 2 +- fs/jffs/inode-v23.c | 34 +++++++++++++------------- fs/libfs.c | 2 +- fs/ocfs2/dlmglue.c | 2 +- include/linux/fs.h | 4 +-- kernel/relay.c | 2 +- security/inode.c | 8 +++--- 36 files changed, 88 insertions(+), 90 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index 7b4572805db9..b837f12a84ae 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -120,7 +120,7 @@ spufs_new_file(struct super_block *sb, struct dentry *dentry, ret = 0; inode->i_op = &spufs_file_iops; inode->i_fop = fops; - inode->u.generic_ip = SPUFS_I(inode)->i_ctx = get_spu_context(ctx); + inode->i_private = SPUFS_I(inode)->i_ctx = get_spu_context(ctx); d_add(dentry, inode); out: return ret; diff --git a/arch/powerpc/platforms/pseries/hvCall_inst.c b/arch/powerpc/platforms/pseries/hvCall_inst.c index 641e6511cf06..446e17d162a5 100644 --- a/arch/powerpc/platforms/pseries/hvCall_inst.c +++ b/arch/powerpc/platforms/pseries/hvCall_inst.c @@ -85,7 +85,7 @@ static int hcall_inst_seq_open(struct inode *inode, struct file *file) rc = seq_open(file, &hcall_inst_seq_ops); seq = file->private_data; - seq->private = file->f_dentry->d_inode->u.generic_ip; + seq->private = file->f_dentry->d_inode->i_private; return rc; } diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index bdade5f2e325..8735024d235b 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -104,13 +104,13 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode) static void hypfs_drop_inode(struct inode *inode) { - kfree(inode->u.generic_ip); + kfree(inode->i_private); generic_delete_inode(inode); } static int hypfs_open(struct inode *inode, struct file *filp) { - char *data = filp->f_dentry->d_inode->u.generic_ip; + char *data = filp->f_dentry->d_inode->i_private; struct hypfs_sb_info *fs_info; if (filp->f_mode & FMODE_WRITE) { @@ -352,7 +352,7 @@ static struct dentry *hypfs_create_file(struct super_block *sb, parent->d_inode->i_nlink++; } else BUG(); - inode->u.generic_ip = data; + inode->i_private = data; d_instantiate(dentry, inode); dget(dentry); return dentry; diff --git a/arch/s390/kernel/debug.c b/arch/s390/kernel/debug.c index 7ba20922a535..43f3d0c7e132 100644 --- a/arch/s390/kernel/debug.c +++ b/arch/s390/kernel/debug.c @@ -603,7 +603,7 @@ debug_open(struct inode *inode, struct file *file) debug_info_t *debug_info, *debug_info_snapshot; down(&debug_lock); - debug_info = (struct debug_info*)file->f_dentry->d_inode->u.generic_ip; + debug_info = file->f_dentry->d_inode->i_private; /* find debug view */ for (i = 0; i < DEBUG_MAX_VIEWS; i++) { if (!debug_info->views[i]) diff --git a/block/blktrace.c b/block/blktrace.c index 265f7a830619..2b4ef2b89b8d 100644 --- a/block/blktrace.c +++ b/block/blktrace.c @@ -217,7 +217,7 @@ static int blk_trace_remove(request_queue_t *q) static int blk_dropped_open(struct inode *inode, struct file *filp) { - filp->private_data = inode->u.generic_ip; + filp->private_data = inode->i_private; return 0; } diff --git a/drivers/i2c/chips/tps65010.c b/drivers/i2c/chips/tps65010.c index 0be6fd6a267d..6a7578217177 100644 --- a/drivers/i2c/chips/tps65010.c +++ b/drivers/i2c/chips/tps65010.c @@ -305,7 +305,7 @@ static int dbg_show(struct seq_file *s, void *_) static int dbg_tps_open(struct inode *inode, struct file *file) { - return single_open(file, dbg_show, inode->u.generic_ip); + return single_open(file, dbg_show, inode->i_private); } static struct file_operations debug_fops = { diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c index a5eb30a06a5c..055cdd089b28 100644 --- a/drivers/infiniband/hw/ipath/ipath_fs.c +++ b/drivers/infiniband/hw/ipath/ipath_fs.c @@ -64,7 +64,7 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry, inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; - inode->u.generic_ip = data; + inode->i_private = data; if ((mode & S_IFMT) == S_IFDIR) { inode->i_op = &simple_dir_inode_operations; inode->i_nlink++; @@ -119,7 +119,7 @@ static ssize_t atomic_counters_read(struct file *file, char __user *buf, u16 i; struct ipath_devdata *dd; - dd = file->f_dentry->d_inode->u.generic_ip; + dd = file->f_dentry->d_inode->i_private; for (i = 0; i < NUM_COUNTERS; i++) counters[i] = ipath_snap_cntr(dd, i); @@ -139,7 +139,7 @@ static ssize_t atomic_node_info_read(struct file *file, char __user *buf, struct ipath_devdata *dd; u64 guid; - dd = file->f_dentry->d_inode->u.generic_ip; + dd = file->f_dentry->d_inode->i_private; guid = be64_to_cpu(dd->ipath_guid); @@ -178,7 +178,7 @@ static ssize_t atomic_port_info_read(struct file *file, char __user *buf, u32 tmp, tmp2; struct ipath_devdata *dd; - dd = file->f_dentry->d_inode->u.generic_ip; + dd = file->f_dentry->d_inode->i_private; /* so we only initialize non-zero fields. */ memset(portinfo, 0, sizeof portinfo); @@ -325,7 +325,7 @@ static ssize_t flash_read(struct file *file, char __user *buf, goto bail; } - dd = file->f_dentry->d_inode->u.generic_ip; + dd = file->f_dentry->d_inode->i_private; if (ipath_eeprom_read(dd, pos, tmp, count)) { ipath_dev_err(dd, "failed to read from flash\n"); ret = -ENXIO; @@ -381,7 +381,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf, goto bail_tmp; } - dd = file->f_dentry->d_inode->u.generic_ip; + dd = file->f_dentry->d_inode->i_private; if (ipath_eeprom_write(dd, pos, tmp, count)) { ret = -ENXIO; ipath_dev_err(dd, "failed to write to flash\n"); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_fs.c b/drivers/infiniband/ulp/ipoib/ipoib_fs.c index 5dde380e8dbe..f1cb83688b31 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_fs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_fs.c @@ -141,7 +141,7 @@ static int ipoib_mcg_open(struct inode *inode, struct file *file) return ret; seq = file->private_data; - seq->private = inode->u.generic_ip; + seq->private = inode->i_private; return 0; } @@ -247,7 +247,7 @@ static int ipoib_path_open(struct inode *inode, struct file *file) return ret; seq = file->private_data; - seq->private = inode->u.generic_ip; + seq->private = inode->i_private; return 0; } diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c index 4a35caff5d02..0e909b617226 100644 --- a/drivers/misc/ibmasm/ibmasmfs.c +++ b/drivers/misc/ibmasm/ibmasmfs.c @@ -175,7 +175,7 @@ static struct dentry *ibmasmfs_create_file (struct super_block *sb, } inode->i_fop = fops; - inode->u.generic_ip = data; + inode->i_private = data; d_add(dentry, inode); return dentry; @@ -244,7 +244,7 @@ static int command_file_open(struct inode *inode, struct file *file) { struct ibmasmfs_command_data *command_data; - if (!inode->u.generic_ip) + if (!inode->i_private) return -ENODEV; command_data = kmalloc(sizeof(struct ibmasmfs_command_data), GFP_KERNEL); @@ -252,7 +252,7 @@ static int command_file_open(struct inode *inode, struct file *file) return -ENOMEM; command_data->command = NULL; - command_data->sp = inode->u.generic_ip; + command_data->sp = inode->i_private; file->private_data = command_data; return 0; } @@ -351,10 +351,10 @@ static int event_file_open(struct inode *inode, struct file *file) struct ibmasmfs_event_data *event_data; struct service_processor *sp; - if (!inode->u.generic_ip) + if (!inode->i_private) return -ENODEV; - sp = inode->u.generic_ip; + sp = inode->i_private; event_data = kmalloc(sizeof(struct ibmasmfs_event_data), GFP_KERNEL); if (!event_data) @@ -439,14 +439,14 @@ static int r_heartbeat_file_open(struct inode *inode, struct file *file) { struct ibmasmfs_heartbeat_data *rhbeat; - if (!inode->u.generic_ip) + if (!inode->i_private) return -ENODEV; rhbeat = kmalloc(sizeof(struct ibmasmfs_heartbeat_data), GFP_KERNEL); if (!rhbeat) return -ENOMEM; - rhbeat->sp = (struct service_processor *)inode->u.generic_ip; + rhbeat->sp = inode->i_private; rhbeat->active = 0; ibmasm_init_reverse_heartbeat(rhbeat->sp, &rhbeat->heartbeat); file->private_data = rhbeat; @@ -508,7 +508,7 @@ static ssize_t r_heartbeat_file_write(struct file *file, const char __user *buf, static int remote_settings_file_open(struct inode *inode, struct file *file) { - file->private_data = inode->u.generic_ip; + file->private_data = inode->i_private; return 0; } diff --git a/drivers/net/irda/vlsi_ir.h b/drivers/net/irda/vlsi_ir.h index a82a4ba8de4f..c37f0bc4c7f9 100644 --- a/drivers/net/irda/vlsi_ir.h +++ b/drivers/net/irda/vlsi_ir.h @@ -58,7 +58,7 @@ typedef void irqreturn_t; /* PDE() introduced in 2.5.4 */ #ifdef CONFIG_PROC_FS -#define PDE(inode) ((inode)->u.generic_ip) +#define PDE(inode) ((inode)->i_private) #endif /* irda crc16 calculation exported in 2.5.42 */ diff --git a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c index 923275ea0789..b9df06a06ea9 100644 --- a/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c +++ b/drivers/net/wireless/bcm43xx/bcm43xx_debugfs.c @@ -54,7 +54,7 @@ static ssize_t write_file_dummy(struct file *file, const char __user *buf, static int open_file_generic(struct inode *inode, struct file *file) { - file->private_data = inode->u.generic_ip; + file->private_data = inode->i_private; return 0; } diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c index 71c2da277d6e..deb37354785b 100644 --- a/drivers/oprofile/oprofilefs.c +++ b/drivers/oprofile/oprofilefs.c @@ -110,8 +110,8 @@ static ssize_t ulong_write_file(struct file * file, char const __user * buf, siz static int default_open(struct inode * inode, struct file * filp) { - if (inode->u.generic_ip) - filp->private_data = inode->u.generic_ip; + if (inode->i_private) + filp->private_data = inode->i_private; return 0; } @@ -158,7 +158,7 @@ int oprofilefs_create_ulong(struct super_block * sb, struct dentry * root, if (!d) return -EFAULT; - d->d_inode->u.generic_ip = val; + d->d_inode->i_private = val; return 0; } @@ -171,7 +171,7 @@ int oprofilefs_create_ro_ulong(struct super_block * sb, struct dentry * root, if (!d) return -EFAULT; - d->d_inode->u.generic_ip = val; + d->d_inode->i_private = val; return 0; } @@ -197,7 +197,7 @@ int oprofilefs_create_ro_atomic(struct super_block * sb, struct dentry * root, if (!d) return -EFAULT; - d->d_inode->u.generic_ip = val; + d->d_inode->i_private = val; return 0; } diff --git a/drivers/pci/hotplug/cpqphp_sysfs.c b/drivers/pci/hotplug/cpqphp_sysfs.c index 8b3da007e859..5bab666cd67e 100644 --- a/drivers/pci/hotplug/cpqphp_sysfs.c +++ b/drivers/pci/hotplug/cpqphp_sysfs.c @@ -140,7 +140,7 @@ struct ctrl_dbg { static int open(struct inode *inode, struct file *file) { - struct controller *ctrl = inode->u.generic_ip; + struct controller *ctrl = inode->i_private; struct ctrl_dbg *dbg; int retval = -ENOMEM; diff --git a/drivers/usb/core/devio.c b/drivers/usb/core/devio.c index 218621b9958e..32e03000420c 100644 --- a/drivers/usb/core/devio.c +++ b/drivers/usb/core/devio.c @@ -555,7 +555,7 @@ static int usbdev_open(struct inode *inode, struct file *file) if (imajor(inode) == USB_DEVICE_MAJOR) dev = usbdev_lookup_minor(iminor(inode)); if (!dev) - dev = inode->u.generic_ip; + dev = inode->i_private; if (!dev) { kfree(ps); goto out; diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c index 3182c2224ba2..482f253085e5 100644 --- a/drivers/usb/core/inode.c +++ b/drivers/usb/core/inode.c @@ -402,8 +402,8 @@ static loff_t default_file_lseek (struct file *file, loff_t offset, int orig) static int default_open (struct inode *inode, struct file *file) { - if (inode->u.generic_ip) - file->private_data = inode->u.generic_ip; + if (inode->i_private) + file->private_data = inode->i_private; return 0; } @@ -509,7 +509,7 @@ static struct dentry *fs_create_file (const char *name, mode_t mode, } else { if (dentry->d_inode) { if (data) - dentry->d_inode->u.generic_ip = data; + dentry->d_inode->i_private = data; if (fops) dentry->d_inode->i_fop = fops; dentry->d_inode->i_uid = uid; diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index 3bdc5e3ba234..ffaa8c1afad8 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -844,7 +844,7 @@ fail1: static int ep_open (struct inode *inode, struct file *fd) { - struct ep_data *data = inode->u.generic_ip; + struct ep_data *data = inode->i_private; int value = -EBUSY; if (down_interruptible (&data->lock) != 0) @@ -1909,7 +1909,7 @@ fail: static int dev_open (struct inode *inode, struct file *fd) { - struct dev_data *dev = inode->u.generic_ip; + struct dev_data *dev = inode->i_private; int value = -EBUSY; if (dev->state == STATE_DEV_DISABLED) { @@ -1970,7 +1970,7 @@ gadgetfs_make_inode (struct super_block *sb, inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; - inode->u.generic_ip = data; + inode->i_private = data; inode->i_fop = fops; } return inode; diff --git a/drivers/usb/host/isp116x-hcd.c b/drivers/usb/host/isp116x-hcd.c index 5147ed4a6662..8c6b38a0b5bb 100644 --- a/drivers/usb/host/isp116x-hcd.c +++ b/drivers/usb/host/isp116x-hcd.c @@ -1204,7 +1204,7 @@ static int isp116x_show_dbg(struct seq_file *s, void *unused) static int isp116x_open_seq(struct inode *inode, struct file *file) { - return single_open(file, isp116x_show_dbg, inode->u.generic_ip); + return single_open(file, isp116x_show_dbg, inode->i_private); } static struct file_operations isp116x_debug_fops = { diff --git a/drivers/usb/host/uhci-debug.c b/drivers/usb/host/uhci-debug.c index dc286a48cafd..d1372cb27f33 100644 --- a/drivers/usb/host/uhci-debug.c +++ b/drivers/usb/host/uhci-debug.c @@ -428,7 +428,7 @@ struct uhci_debug { static int uhci_debug_open(struct inode *inode, struct file *file) { - struct uhci_hcd *uhci = inode->u.generic_ip; + struct uhci_hcd *uhci = inode->i_private; struct uhci_debug *up; int ret = -ENOMEM; unsigned long flags; diff --git a/drivers/usb/mon/mon_stat.c b/drivers/usb/mon/mon_stat.c index 1fe01d994a79..86ad2b381c4b 100644 --- a/drivers/usb/mon/mon_stat.c +++ b/drivers/usb/mon/mon_stat.c @@ -28,7 +28,7 @@ static int mon_stat_open(struct inode *inode, struct file *file) if ((sp = kmalloc(sizeof(struct snap), GFP_KERNEL)) == NULL) return -ENOMEM; - mbus = inode->u.generic_ip; + mbus = inode->i_private; sp->slen = snprintf(sp->str, STAT_BUF_SIZE, "nreaders %d events %u text_lost %u\n", diff --git a/drivers/usb/mon/mon_text.c b/drivers/usb/mon/mon_text.c index f961a770cee2..2fd39b4fa166 100644 --- a/drivers/usb/mon/mon_text.c +++ b/drivers/usb/mon/mon_text.c @@ -238,7 +238,7 @@ static int mon_text_open(struct inode *inode, struct file *file) int rc; mutex_lock(&mon_lock); - mbus = inode->u.generic_ip; + mbus = inode->i_private; ubus = mbus->u_bus; rp = kzalloc(sizeof(struct mon_reader_text), GFP_KERNEL); @@ -401,7 +401,7 @@ static int mon_text_release(struct inode *inode, struct file *file) struct mon_event_text *ep; mutex_lock(&mon_lock); - mbus = inode->u.generic_ip; + mbus = inode->i_private; if (mbus->nreaders <= 0) { printk(KERN_ERR TAG ": consistency error on close\n"); diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index c81d6b8c2828..d39d2acd9b38 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -241,7 +241,7 @@ static void autofs_read_inode(struct inode *inode) inode->i_op = &autofs_symlink_inode_operations; sl = &sbi->symlink[n]; - inode->u.generic_ip = sl; + inode->i_private = sl; inode->i_mode = S_IFLNK | S_IRWXUGO; inode->i_mtime.tv_sec = inode->i_ctime.tv_sec = sl->mtime; inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0; diff --git a/fs/autofs/symlink.c b/fs/autofs/symlink.c index 52e8772b066e..c74f2eb65775 100644 --- a/fs/autofs/symlink.c +++ b/fs/autofs/symlink.c @@ -15,7 +15,7 @@ /* Nothing to release.. */ static void *autofs_follow_link(struct dentry *dentry, struct nameidata *nd) { - char *s=((struct autofs_symlink *)dentry->d_inode->u.generic_ip)->data; + char *s=((struct autofs_symlink *)dentry->d_inode->i_private)->data; nd_set_link(nd, s); return NULL; } diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 34ebbc191e46..6759b9839ce8 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -517,7 +517,7 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode) static void bm_clear_inode(struct inode *inode) { - kfree(inode->u.generic_ip); + kfree(inode->i_private); } static void kill_node(Node *e) @@ -545,7 +545,7 @@ static void kill_node(Node *e) static ssize_t bm_entry_read(struct file * file, char __user * buf, size_t nbytes, loff_t *ppos) { - Node *e = file->f_dentry->d_inode->u.generic_ip; + Node *e = file->f_dentry->d_inode->i_private; loff_t pos = *ppos; ssize_t res; char *page; @@ -579,7 +579,7 @@ static ssize_t bm_entry_write(struct file *file, const char __user *buffer, size_t count, loff_t *ppos) { struct dentry *root; - Node *e = file->f_dentry->d_inode->u.generic_ip; + Node *e = file->f_dentry->d_inode->i_private; int res = parse_command(buffer, count); switch (res) { @@ -646,7 +646,7 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, } e->dentry = dget(dentry); - inode->u.generic_ip = e; + inode->i_private = e; inode->i_fop = &bm_entry_operations; d_instantiate(dentry, inode); diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index e4b430552c88..bf3901ab1744 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -32,8 +32,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf, static int default_open(struct inode *inode, struct file *file) { - if (inode->u.generic_ip) - file->private_data = inode->u.generic_ip; + if (inode->i_private) + file->private_data = inode->i_private; return 0; } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 3ca268d2e5a2..717f4821ed02 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -168,7 +168,7 @@ static int debugfs_create_by_name(const char *name, mode_t mode, * directory dentry if set. If this paramater is NULL, then the * file will be created in the root of the debugfs filesystem. * @data: a pointer to something that the caller will want to get to later - * on. The inode.u.generic_ip pointer will point to this value on + * on. The inode.i_private pointer will point to this value on * the open() call. * @fops: a pointer to a struct file_operations that should be used for * this file. @@ -209,7 +209,7 @@ struct dentry *debugfs_create_file(const char *name, mode_t mode, if (dentry->d_inode) { if (data) - dentry->d_inode->u.generic_ip = data; + dentry->d_inode->i_private = data; if (fops) dentry->d_inode->i_fop = fops; } diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index f7aef5bb584a..5bf06a10dddf 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -177,7 +177,7 @@ int devpts_pty_new(struct tty_struct *tty) inode->i_gid = config.setgid ? config.gid : current->fsgid; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; init_special_inode(inode, S_IFCHR|config.mode, device); - inode->u.generic_ip = tty; + inode->i_private = tty; dentry = get_node(number); if (!IS_ERR(dentry) && !dentry->d_inode) @@ -196,7 +196,7 @@ struct tty_struct *devpts_get_tty(int number) tty = NULL; if (!IS_ERR(dentry)) { if (dentry->d_inode) - tty = dentry->d_inode->u.generic_ip; + tty = dentry->d_inode->i_private; dput(dentry); } diff --git a/fs/freevxfs/vxfs.h b/fs/freevxfs/vxfs.h index d35979a58743..c8a92652612a 100644 --- a/fs/freevxfs/vxfs.h +++ b/fs/freevxfs/vxfs.h @@ -252,7 +252,7 @@ enum { * Get filesystem private data from VFS inode. */ #define VXFS_INO(ip) \ - ((struct vxfs_inode_info *)(ip)->u.generic_ip) + ((struct vxfs_inode_info *)(ip)->i_private) /* * Get filesystem private data from VFS superblock. diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index ca6a39714771..32a82ed108e4 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -243,7 +243,7 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip) ip->i_blocks = vip->vii_blocks; ip->i_generation = vip->vii_gen; - ip->u.generic_ip = (void *)vip; + ip->i_private = vip; } @@ -338,5 +338,5 @@ vxfs_read_inode(struct inode *ip) void vxfs_clear_inode(struct inode *ip) { - kmem_cache_free(vxfs_inode_cachep, ip->u.generic_ip); + kmem_cache_free(vxfs_inode_cachep, ip->i_private); } diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 46fe60b2da23..79ec1f23d4d2 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -23,7 +23,7 @@ static struct fuse_conn *fuse_ctl_file_conn_get(struct file *file) { struct fuse_conn *fc; mutex_lock(&fuse_mutex); - fc = file->f_dentry->d_inode->u.generic_ip; + fc = file->f_dentry->d_inode->i_private; if (fc) fc = fuse_conn_get(fc); mutex_unlock(&fuse_mutex); @@ -98,7 +98,7 @@ static struct dentry *fuse_ctl_add_dentry(struct dentry *parent, inode->i_op = iop; inode->i_fop = fop; inode->i_nlink = nlink; - inode->u.generic_ip = fc; + inode->i_private = fc; d_add(dentry, inode); return dentry; } @@ -150,7 +150,7 @@ void fuse_ctl_remove_conn(struct fuse_conn *fc) for (i = fc->ctl_ndents - 1; i >= 0; i--) { struct dentry *dentry = fc->ctl_dentry[i]; - dentry->d_inode->u.generic_ip = NULL; + dentry->d_inode->i_private = NULL; d_drop(dentry); dput(dentry); } diff --git a/fs/inode.c b/fs/inode.c index 0bf9f0444a96..77e254792025 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -163,7 +163,7 @@ static struct inode *alloc_inode(struct super_block *sb) bdi = sb->s_bdev->bd_inode->i_mapping->backing_dev_info; mapping->backing_dev_info = bdi; } - memset(&inode->u, 0, sizeof(inode->u)); + inode->i_private = 0; inode->i_mapping = mapping; } return inode; diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index b59553d28d13..7358ef87f16b 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -369,7 +369,7 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode, f = jffs_find_file(c, raw_inode->ino); - inode->u.generic_ip = (void *)f; + inode->i_private = (void *)f; insert_inode_hash(inode); return inode; @@ -442,7 +442,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry, }); result = -ENOTDIR; - if (!(old_dir_f = (struct jffs_file *)old_dir->u.generic_ip)) { + if (!(old_dir_f = old_dir->i_private)) { D(printk("jffs_rename(): Old dir invalid.\n")); goto jffs_rename_end; } @@ -456,7 +456,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry, /* Find the new directory. */ result = -ENOTDIR; - if (!(new_dir_f = (struct jffs_file *)new_dir->u.generic_ip)) { + if (!(new_dir_f = new_dir->i_private)) { D(printk("jffs_rename(): New dir invalid.\n")); goto jffs_rename_end; } @@ -593,7 +593,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir) } else { ddino = ((struct jffs_file *) - inode->u.generic_ip)->pino; + inode->i_private)->pino; } D3(printk("jffs_readdir(): \"..\" %u\n", ddino)); if (filldir(dirent, "..", 2, filp->f_pos, ddino, DT_DIR) < 0) { @@ -604,7 +604,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir) } filp->f_pos++; } - f = ((struct jffs_file *)inode->u.generic_ip)->children; + f = ((struct jffs_file *)inode->i_private)->children; j = 2; while(f && (f->deleted || j++ < filp->f_pos )) { @@ -668,7 +668,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd) } r = -EACCES; - if (!(d = (struct jffs_file *)dir->u.generic_ip)) { + if (!(d = (struct jffs_file *)dir->i_private)) { D(printk("jffs_lookup(): No such inode! (%lu)\n", dir->i_ino)); goto jffs_lookup_end; @@ -739,7 +739,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page) unsigned long read_len; int result; struct inode *inode = (struct inode*)page->mapping->host; - struct jffs_file *f = (struct jffs_file *)inode->u.generic_ip; + struct jffs_file *f = (struct jffs_file *)inode->i_private; struct jffs_control *c = (struct jffs_control *)inode->i_sb->s_fs_info; int r; loff_t offset; @@ -828,7 +828,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode) }); lock_kernel(); - dir_f = (struct jffs_file *)dir->u.generic_ip; + dir_f = dir->i_private; ASSERT(if (!dir_f) { printk(KERN_ERR "jffs_mkdir(): No reference to a " @@ -972,7 +972,7 @@ jffs_remove(struct inode *dir, struct dentry *dentry, int type) kfree(_name); }); - dir_f = (struct jffs_file *) dir->u.generic_ip; + dir_f = dir->i_private; c = dir_f->c; result = -ENOENT; @@ -1082,7 +1082,7 @@ jffs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev) if (!old_valid_dev(rdev)) return -EINVAL; lock_kernel(); - dir_f = (struct jffs_file *)dir->u.generic_ip; + dir_f = dir->i_private; c = dir_f->c; D3(printk (KERN_NOTICE "mknod(): down biglock\n")); @@ -1186,7 +1186,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) kfree(_symname); }); - dir_f = (struct jffs_file *)dir->u.generic_ip; + dir_f = dir->i_private; ASSERT(if (!dir_f) { printk(KERN_ERR "jffs_symlink(): No reference to a " "jffs_file struct in inode.\n"); @@ -1289,7 +1289,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode, kfree(s); }); - dir_f = (struct jffs_file *)dir->u.generic_ip; + dir_f = dir->i_private; ASSERT(if (!dir_f) { printk(KERN_ERR "jffs_create(): No reference to a " "jffs_file struct in inode.\n"); @@ -1403,9 +1403,9 @@ jffs_file_write(struct file *filp, const char *buf, size_t count, goto out_isem; } - if (!(f = (struct jffs_file *)inode->u.generic_ip)) { - D(printk("jffs_file_write(): inode->u.generic_ip = 0x%p\n", - inode->u.generic_ip)); + if (!(f = inode->i_private)) { + D(printk("jffs_file_write(): inode->i_private = 0x%p\n", + inode->i_private)); goto out_isem; } @@ -1693,7 +1693,7 @@ jffs_read_inode(struct inode *inode) mutex_unlock(&c->fmc->biglock); return; } - inode->u.generic_ip = (void *)f; + inode->i_private = f; inode->i_mode = f->mode; inode->i_nlink = f->nlink; inode->i_uid = f->uid; @@ -1748,7 +1748,7 @@ jffs_delete_inode(struct inode *inode) lock_kernel(); inode->i_size = 0; inode->i_blocks = 0; - inode->u.generic_ip = NULL; + inode->i_private = NULL; clear_inode(inode); if (inode->i_nlink == 0) { c = (struct jffs_control *) inode->i_sb->s_fs_info; diff --git a/fs/libfs.c b/fs/libfs.c index ac02ea602c3d..2751793beeaa 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -547,7 +547,7 @@ int simple_attr_open(struct inode *inode, struct file *file, attr->get = get; attr->set = set; - attr->data = inode->u.generic_ip; + attr->data = inode->i_private; attr->fmt = fmt; mutex_init(&attr->mutex); diff --git a/fs/ocfs2/dlmglue.c b/fs/ocfs2/dlmglue.c index de887063dcfc..8801e41afe80 100644 --- a/fs/ocfs2/dlmglue.c +++ b/fs/ocfs2/dlmglue.c @@ -2052,7 +2052,7 @@ static int ocfs2_dlm_debug_open(struct inode *inode, struct file *file) mlog_errno(ret); goto out; } - osb = (struct ocfs2_super *) inode->u.generic_ip; + osb = inode->i_private; ocfs2_get_dlm_debug(osb->osb_dlm_debug); priv->p_dlm_debug = osb->osb_dlm_debug; INIT_LIST_HEAD(&priv->p_iter_res.l_debug_list); diff --git a/include/linux/fs.h b/include/linux/fs.h index 1d3e601ece73..4f77ec9c3353 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -554,9 +554,7 @@ struct inode { atomic_t i_writecount; void *i_security; - union { - void *generic_ip; - } u; + void *i_private; /* fs or device private pointer */ #ifdef __NEED_I_SIZE_ORDERED seqcount_t i_size_seqcount; #endif diff --git a/kernel/relay.c b/kernel/relay.c index 33345e73485c..85786ff2a4f9 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -669,7 +669,7 @@ EXPORT_SYMBOL_GPL(relay_flush); */ static int relay_file_open(struct inode *inode, struct file *filp) { - struct rchan_buf *buf = inode->u.generic_ip; + struct rchan_buf *buf = inode->i_private; kref_get(&buf->kref); filp->private_data = buf; diff --git a/security/inode.c b/security/inode.c index 47eb63480dac..176aacea8ca4 100644 --- a/security/inode.c +++ b/security/inode.c @@ -44,8 +44,8 @@ static ssize_t default_write_file(struct file *file, const char __user *buf, static int default_open(struct inode *inode, struct file *file) { - if (inode->u.generic_ip) - file->private_data = inode->u.generic_ip; + if (inode->i_private) + file->private_data = inode->i_private; return 0; } @@ -194,7 +194,7 @@ static int create_by_name(const char *name, mode_t mode, * directory dentry if set. If this paramater is NULL, then the * file will be created in the root of the securityfs filesystem. * @data: a pointer to something that the caller will want to get to later - * on. The inode.u.generic_ip pointer will point to this value on + * on. The inode.i_private pointer will point to this value on * the open() call. * @fops: a pointer to a struct file_operations that should be used for * this file. @@ -240,7 +240,7 @@ struct dentry *securityfs_create_file(const char *name, mode_t mode, if (fops) dentry->d_inode->i_fop = fops; if (data) - dentry->d_inode->u.generic_ip = data; + dentry->d_inode->i_private = data; } exit: return dentry; -- cgit v1.2.3 From ba52de123d454b57369f291348266d86f4b35070 Mon Sep 17 00:00:00 2001 From: Theodore Ts'o Date: Wed, 27 Sep 2006 01:50:49 -0700 Subject: [PATCH] inode-diet: Eliminate i_blksize from the inode structure This eliminates the i_blksize field from struct inode. Filesystems that want to provide a per-inode st_blksize can do so by providing their own getattr routine instead of using the generic_fillattr() function. Note that some filesystems were providing pretty much random (and incorrect) values for i_blksize. [bunk@stusta.de: cleanup] [akpm@osdl.org: generic_fillattr() fix] Signed-off-by: "Theodore Ts'o" Signed-off-by: Adrian Bunk Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/powerpc/platforms/cell/spufs/inode.c | 1 - arch/s390/hypfs/inode.c | 1 - drivers/block/loop.c | 7 +++++-- drivers/infiniband/hw/ipath/ipath_fs.c | 1 - drivers/isdn/capi/capifs.c | 2 -- drivers/misc/ibmasm/ibmasmfs.c | 1 - drivers/oprofile/oprofilefs.c | 1 - drivers/usb/core/inode.c | 1 - drivers/usb/gadget/inode.c | 1 - fs/9p/vfs_inode.c | 4 +--- fs/adfs/inode.c | 1 - fs/afs/inode.c | 1 - fs/autofs/inode.c | 1 - fs/autofs4/inode.c | 1 - fs/befs/linuxvfs.c | 1 - fs/bfs/dir.c | 2 +- fs/bfs/inode.c | 1 - fs/binfmt_misc.c | 1 - fs/cifs/cifsfs.c | 1 - fs/cifs/readdir.c | 5 ++--- fs/coda/coda_linux.c | 2 -- fs/configfs/inode.c | 1 - fs/cramfs/inode.c | 1 - fs/debugfs/inode.c | 1 - fs/devpts/inode.c | 2 -- fs/eventpoll.c | 1 - fs/ext2/ialloc.c | 1 - fs/ext2/inode.c | 1 - fs/ext3/ialloc.c | 1 - fs/ext3/inode.c | 3 --- fs/fat/inode.c | 3 --- fs/freevxfs/vxfs_inode.c | 1 - fs/fuse/inode.c | 1 - fs/hfs/inode.c | 2 -- fs/hfsplus/inode.c | 2 -- fs/hostfs/hostfs_kern.c | 1 - fs/hpfs/inode.c | 1 - fs/hppfs/hppfs_kern.c | 1 - fs/hugetlbfs/inode.c | 1 - fs/isofs/inode.c | 3 +-- fs/jffs/inode-v23.c | 2 -- fs/jffs2/fs.c | 2 -- fs/jfs/jfs_extent.c | 2 +- fs/jfs/jfs_imap.c | 1 - fs/jfs/jfs_inode.c | 1 - fs/jfs/jfs_metapage.c | 2 +- fs/libfs.c | 2 -- fs/minix/bitmap.c | 2 +- fs/minix/inode.c | 4 ++-- fs/ncpfs/inode.c | 1 - fs/nfs/inode.c | 4 ---- fs/ntfs/inode.c | 4 ---- fs/ntfs/mft.c | 5 ----- fs/ocfs2/dlm/dlmfs.c | 2 -- fs/ocfs2/inode.c | 4 ---- fs/pipe.c | 1 - fs/qnx4/inode.c | 1 - fs/ramfs/inode.c | 1 - fs/reiserfs/inode.c | 4 ---- fs/smbfs/inode.c | 2 -- fs/smbfs/proc.c | 1 - fs/stat.c | 3 ++- fs/sysfs/inode.c | 1 - fs/sysv/ialloc.c | 2 +- fs/sysv/inode.c | 2 +- fs/udf/ialloc.c | 1 - fs/udf/inode.c | 2 -- fs/ufs/ialloc.c | 1 - fs/ufs/inode.c | 1 - fs/xfs/linux-2.6/xfs_super.c | 1 - fs/xfs/linux-2.6/xfs_vnode.c | 1 - include/linux/fs.h | 1 - include/linux/nfsd/nfsfh.h | 10 ++-------- include/linux/smb.h | 1 - ipc/mqueue.c | 1 - kernel/cpuset.c | 1 - mm/shmem.c | 1 - net/sunrpc/rpc_pipe.c | 1 - security/inode.c | 1 - security/selinux/selinuxfs.c | 1 - 80 files changed, 21 insertions(+), 125 deletions(-) (limited to 'kernel') diff --git a/arch/powerpc/platforms/cell/spufs/inode.c b/arch/powerpc/platforms/cell/spufs/inode.c index b837f12a84ae..3950ddccb2c8 100644 --- a/arch/powerpc/platforms/cell/spufs/inode.c +++ b/arch/powerpc/platforms/cell/spufs/inode.c @@ -82,7 +82,6 @@ spufs_new_inode(struct super_block *sb, int mode) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; out: diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 8735024d235b..813fc21358f9 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -91,7 +91,6 @@ static struct inode *hypfs_make_inode(struct super_block *sb, int mode) ret->i_mode = mode; ret->i_uid = hypfs_info->uid; ret->i_gid = hypfs_info->gid; - ret->i_blksize = PAGE_CACHE_SIZE; ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; if (mode & S_IFDIR) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 7b3b94ddddcc..c774121684d7 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -662,7 +662,8 @@ static void do_loop_switch(struct loop_device *lo, struct switch_request *p) mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask); lo->lo_backing_file = file; - lo->lo_blocksize = mapping->host->i_blksize; + lo->lo_blocksize = S_ISBLK(mapping->host->i_mode) ? + mapping->host->i_bdev->bd_block_size : PAGE_SIZE; lo->old_gfp_mask = mapping_gfp_mask(mapping); mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS)); complete(&p->wait); @@ -794,7 +795,9 @@ static int loop_set_fd(struct loop_device *lo, struct file *lo_file, if (!(lo_flags & LO_FLAGS_USE_AOPS) && !file->f_op->write) lo_flags |= LO_FLAGS_READ_ONLY; - lo_blocksize = inode->i_blksize; + lo_blocksize = S_ISBLK(inode->i_mode) ? + inode->i_bdev->bd_block_size : PAGE_SIZE; + error = 0; } else { goto out_putf; diff --git a/drivers/infiniband/hw/ipath/ipath_fs.c b/drivers/infiniband/hw/ipath/ipath_fs.c index 055cdd089b28..c8a8af0fe471 100644 --- a/drivers/infiniband/hw/ipath/ipath_fs.c +++ b/drivers/infiniband/hw/ipath/ipath_fs.c @@ -61,7 +61,6 @@ static int ipathfs_mknod(struct inode *dir, struct dentry *dentry, inode->i_mode = mode; inode->i_uid = 0; inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_private = data; diff --git a/drivers/isdn/capi/capifs.c b/drivers/isdn/capi/capifs.c index 9ea6bd0ddc35..2dd1b57b0ba4 100644 --- a/drivers/isdn/capi/capifs.c +++ b/drivers/isdn/capi/capifs.c @@ -104,7 +104,6 @@ capifs_fill_super(struct super_block *s, void *data, int silent) inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_blocks = 0; - inode->i_blksize = 1024; inode->i_uid = inode->i_gid = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; @@ -149,7 +148,6 @@ void capifs_new_ncci(unsigned int number, dev_t device) if (!inode) return; inode->i_ino = number+2; - inode->i_blksize = 1024; inode->i_uid = config.setuid ? config.uid : current->fsuid; inode->i_gid = config.setgid ? config.gid : current->fsgid; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; diff --git a/drivers/misc/ibmasm/ibmasmfs.c b/drivers/misc/ibmasm/ibmasmfs.c index 0e909b617226..b99dc507de2e 100644 --- a/drivers/misc/ibmasm/ibmasmfs.c +++ b/drivers/misc/ibmasm/ibmasmfs.c @@ -147,7 +147,6 @@ static struct inode *ibmasmfs_make_inode(struct super_block *sb, int mode) if (ret) { ret->i_mode = mode; ret->i_uid = ret->i_gid = 0; - ret->i_blksize = PAGE_CACHE_SIZE; ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; } diff --git a/drivers/oprofile/oprofilefs.c b/drivers/oprofile/oprofilefs.c index deb37354785b..5756401fb15b 100644 --- a/drivers/oprofile/oprofilefs.c +++ b/drivers/oprofile/oprofilefs.c @@ -31,7 +31,6 @@ static struct inode * oprofilefs_get_inode(struct super_block * sb, int mode) inode->i_mode = mode; inode->i_uid = 0; inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; } diff --git a/drivers/usb/core/inode.c b/drivers/usb/core/inode.c index 482f253085e5..58b4b1012120 100644 --- a/drivers/usb/core/inode.c +++ b/drivers/usb/core/inode.c @@ -249,7 +249,6 @@ static struct inode *usbfs_get_inode (struct super_block *sb, int mode, dev_t de inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { diff --git a/drivers/usb/gadget/inode.c b/drivers/usb/gadget/inode.c index ffaa8c1afad8..2a7162d89799 100644 --- a/drivers/usb/gadget/inode.c +++ b/drivers/usb/gadget/inode.c @@ -1966,7 +1966,6 @@ gadgetfs_make_inode (struct super_block *sb, inode->i_mode = mode; inode->i_uid = default_uid; inode->i_gid = default_gid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index eae50c9d6dc4..7a7ec2d1d2f4 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -204,7 +204,6 @@ struct inode *v9fs_get_inode(struct super_block *sb, int mode) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = sb->s_blocksize; inode->i_blocks = 0; inode->i_rdev = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; @@ -950,9 +949,8 @@ v9fs_stat2inode(struct v9fs_stat *stat, struct inode *inode, inode->i_size = stat->length; - inode->i_blksize = sb->s_blocksize; inode->i_blocks = - (inode->i_size + inode->i_blksize - 1) >> sb->s_blocksize_bits; + (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits; } /** diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c index 534f3eecc985..7e7a04be1278 100644 --- a/fs/adfs/inode.c +++ b/fs/adfs/inode.c @@ -269,7 +269,6 @@ adfs_iget(struct super_block *sb, struct object_info *obj) inode->i_ino = obj->file_id; inode->i_size = obj->size; inode->i_nlink = 2; - inode->i_blksize = PAGE_SIZE; inode->i_blocks = (inode->i_size + sb->s_blocksize - 1) >> sb->s_blocksize_bits; diff --git a/fs/afs/inode.c b/fs/afs/inode.c index 4ebb30a50ed5..6f37754906c2 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -72,7 +72,6 @@ static int afs_inode_map_status(struct afs_vnode *vnode) inode->i_ctime.tv_sec = vnode->status.mtime_server; inode->i_ctime.tv_nsec = 0; inode->i_atime = inode->i_mtime = inode->i_ctime; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_version = vnode->fid.unique; inode->i_mapping->a_ops = &afs_fs_aops; diff --git a/fs/autofs/inode.c b/fs/autofs/inode.c index d39d2acd9b38..2c9759baad61 100644 --- a/fs/autofs/inode.c +++ b/fs/autofs/inode.c @@ -216,7 +216,6 @@ static void autofs_read_inode(struct inode *inode) inode->i_nlink = 2; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_blocks = 0; - inode->i_blksize = 1024; if ( ino == AUTOFS_ROOT_INO ) { inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index 11a6a9ae51b7..800ce876caec 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -447,7 +447,6 @@ struct inode *autofs4_get_inode(struct super_block *sb, inode->i_uid = 0; inode->i_gid = 0; } - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index f6676fbe9484..57020c7a7e65 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -365,7 +365,6 @@ befs_read_inode(struct inode *inode) inode->i_mtime.tv_nsec = 0; /* lower 16 bits are not a time */ inode->i_ctime = inode->i_mtime; inode->i_atime = inode->i_mtime; - inode->i_blksize = befs_sb->block_size; befs_ino->i_inode_num = fsrun_to_cpu(sb, raw_inode->inode_num); befs_ino->i_parent = fsrun_to_cpu(sb, raw_inode->parent); diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index 26fad9621738..dcf04cb13283 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -102,7 +102,7 @@ static int bfs_create(struct inode * dir, struct dentry * dentry, int mode, inode->i_uid = current->fsuid; inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; inode->i_op = &bfs_file_inops; inode->i_fop = &bfs_file_operations; inode->i_mapping->a_ops = &bfs_aops; diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 8fc2e8e49dbe..ed27ffb3459e 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -76,7 +76,6 @@ static void bfs_read_inode(struct inode * inode) inode->i_size = BFS_FILESIZE(di); inode->i_blocks = BFS_FILEBLOCKS(di); if (inode->i_size || inode->i_blocks) dprintf("Registered inode with %lld size, %ld blocks\n", inode->i_size, inode->i_blocks); - inode->i_blksize = PAGE_SIZE; inode->i_atime.tv_sec = le32_to_cpu(di->i_atime); inode->i_mtime.tv_sec = le32_to_cpu(di->i_mtime); inode->i_ctime.tv_sec = le32_to_cpu(di->i_ctime); diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 6759b9839ce8..66ba137f8661 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -507,7 +507,6 @@ static struct inode *bm_get_inode(struct super_block *sb, int mode) inode->i_mode = mode; inode->i_uid = 0; inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = current_fs_time(inode->i_sb); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 4197a5043f13..22bcf4d7e7ae 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -253,7 +253,6 @@ cifs_alloc_inode(struct super_block *sb) file data or metadata */ cifs_inode->clientCanCacheRead = FALSE; cifs_inode->clientCanCacheAll = FALSE; - cifs_inode->vfs_inode.i_blksize = CIFS_MAX_MSGSIZE; cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME; INIT_LIST_HEAD(&cifs_inode->openFileList); diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c index 9aeb58a7d369..b27b34537bf2 100644 --- a/fs/cifs/readdir.c +++ b/fs/cifs/readdir.c @@ -216,10 +216,9 @@ static void fill_in_inode(struct inode *tmp_inode, int new_buf_type, if (allocation_size < end_of_file) cFYI(1, ("May be sparse file, allocation less than file size")); - cFYI(1, ("File Size %ld and blocks %llu and blocksize %ld", + cFYI(1, ("File Size %ld and blocks %llu", (unsigned long)tmp_inode->i_size, - (unsigned long long)tmp_inode->i_blocks, - tmp_inode->i_blksize)); + (unsigned long long)tmp_inode->i_blocks)); if (S_ISREG(tmp_inode->i_mode)) { cFYI(1, ("File inode")); tmp_inode->i_op = &cifs_file_inode_ops; diff --git a/fs/coda/coda_linux.c b/fs/coda/coda_linux.c index 5597080cb811..95a54253c047 100644 --- a/fs/coda/coda_linux.c +++ b/fs/coda/coda_linux.c @@ -110,8 +110,6 @@ void coda_vattr_to_iattr(struct inode *inode, struct coda_vattr *attr) inode->i_nlink = attr->va_nlink; if (attr->va_size != -1) inode->i_size = attr->va_size; - if (attr->va_blocksize != -1) - inode->i_blksize = attr->va_blocksize; if (attr->va_size != -1) inode->i_blocks = (attr->va_size + 511) >> 9; if (attr->va_atime.tv_sec != -1) diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 5047e6a7581e..fb18917954a9 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -135,7 +135,6 @@ struct inode * configfs_new_inode(mode_t mode, struct configfs_dirent * sd) { struct inode * inode = new_inode(configfs_sb); if (inode) { - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->a_ops = &configfs_aops; inode->i_mapping->backing_dev_info = &configfs_backing_dev_info; diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index d09b6777c41a..ad96b6990715 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -73,7 +73,6 @@ static int cramfs_iget5_set(struct inode *inode, void *opaque) inode->i_uid = cramfs_inode->uid; inode->i_size = cramfs_inode->size; inode->i_blocks = (cramfs_inode->size - 1) / 512 + 1; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_gid = cramfs_inode->gid; /* Struct copy intentional */ inode->i_mtime = inode->i_atime = inode->i_ctime = zerotime; diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 717f4821ed02..269e649e6dc6 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -40,7 +40,6 @@ static struct inode *debugfs_get_inode(struct super_block *sb, int mode, dev_t d inode->i_mode = mode; inode->i_uid = 0; inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index 5bf06a10dddf..5f7b5a6025bf 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -113,7 +113,6 @@ devpts_fill_super(struct super_block *s, void *data, int silent) inode->i_ino = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; inode->i_blocks = 0; - inode->i_blksize = 1024; inode->i_uid = inode->i_gid = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO | S_IWUSR; inode->i_op = &simple_dir_inode_operations; @@ -172,7 +171,6 @@ int devpts_pty_new(struct tty_struct *tty) return -ENOMEM; inode->i_ino = number+2; - inode->i_blksize = 1024; inode->i_uid = config.setuid ? config.uid : current->fsuid; inode->i_gid = config.setgid ? config.gid : current->fsgid; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 3a3567433b92..8d544334bcd2 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1590,7 +1590,6 @@ static struct inode *ep_eventpoll_inode(void) inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; - inode->i_blksize = PAGE_SIZE; return inode; eexit_1: diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index 695f69ccf908..2cb545bf0f3c 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -574,7 +574,6 @@ got: inode->i_mode = mode; inode->i_ino = ino; - inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size (for stat), not the fs block size */ inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; memset(ei->i_data, 0, sizeof(ei->i_data)); diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index fb4d3220eb8d..dd4e14c221e0 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -1094,7 +1094,6 @@ void ext2_read_inode (struct inode * inode) brelse (bh); goto bad_inode; } - inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size (for stat), not the fs block size */ inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); ei->i_flags = le32_to_cpu(raw_inode->i_flags); ei->i_faddr = le32_to_cpu(raw_inode->i_faddr); diff --git a/fs/ext3/ialloc.c b/fs/ext3/ialloc.c index 1e4ded8aa3cd..e45dbd651736 100644 --- a/fs/ext3/ialloc.c +++ b/fs/ext3/ialloc.c @@ -559,7 +559,6 @@ got: inode->i_ino = ino; /* This is the optimal IO size (for stat), not the fs block size */ - inode->i_blksize = PAGE_SIZE; inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index c9fc15f8b609..dcf4f1dd108b 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -2632,9 +2632,6 @@ void ext3_read_inode(struct inode * inode) * recovery code: that's fine, we're about to complete * the process of deleting those. */ } - inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size - * (for stat), not the fs block - * size */ inode->i_blocks = le32_to_cpu(raw_inode->i_blocks); ei->i_flags = le32_to_cpu(raw_inode->i_flags); #ifdef EXT3_FRAGMENTS diff --git a/fs/fat/inode.c b/fs/fat/inode.c index e1035a590664..ab96ae823753 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -370,8 +370,6 @@ static int fat_fill_inode(struct inode *inode, struct msdos_dir_entry *de) inode->i_flags |= S_IMMUTABLE; } MSDOS_I(inode)->i_attrs = de->attr & ATTR_UNUSED; - /* this is as close to the truth as we can get ... */ - inode->i_blksize = sbi->cluster_size; inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1)) & ~((loff_t)sbi->cluster_size - 1)) >> 9; inode->i_mtime.tv_sec = @@ -1131,7 +1129,6 @@ static int fat_read_root(struct inode *inode) MSDOS_I(inode)->i_start = 0; inode->i_size = sbi->dir_entries * sizeof(struct msdos_dir_entry); } - inode->i_blksize = sbi->cluster_size; inode->i_blocks = ((inode->i_size + (sbi->cluster_size - 1)) & ~((loff_t)sbi->cluster_size - 1)) >> 9; MSDOS_I(inode)->i_logstart = 0; diff --git a/fs/freevxfs/vxfs_inode.c b/fs/freevxfs/vxfs_inode.c index 32a82ed108e4..4786d51ad3bd 100644 --- a/fs/freevxfs/vxfs_inode.c +++ b/fs/freevxfs/vxfs_inode.c @@ -239,7 +239,6 @@ vxfs_iinit(struct inode *ip, struct vxfs_inode_info *vip) ip->i_ctime.tv_nsec = 0; ip->i_mtime.tv_nsec = 0; - ip->i_blksize = PAGE_SIZE; ip->i_blocks = vip->vii_blocks; ip->i_generation = vip->vii_gen; diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 7d25092262ae..cb7cadb0b790 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -118,7 +118,6 @@ void fuse_change_attributes(struct inode *inode, struct fuse_attr *attr) inode->i_uid = attr->uid; inode->i_gid = attr->gid; i_size_write(inode, attr->size); - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = attr->blocks; inode->i_atime.tv_sec = attr->atime; inode->i_atime.tv_nsec = attr->atimensec; diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 315cf44a90b2..d05641c35fc9 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -154,7 +154,6 @@ struct inode *hfs_new_inode(struct inode *dir, struct qstr *name, int mode) inode->i_gid = current->fsgid; inode->i_nlink = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - inode->i_blksize = HFS_SB(sb)->alloc_blksz; HFS_I(inode)->flags = 0; HFS_I(inode)->rsrc_inode = NULL; HFS_I(inode)->fs_blocks = 0; @@ -284,7 +283,6 @@ static int hfs_read_inode(struct inode *inode, void *data) inode->i_uid = hsb->s_uid; inode->i_gid = hsb->s_gid; inode->i_nlink = 1; - inode->i_blksize = HFS_SB(inode->i_sb)->alloc_blksz; if (idata->key) HFS_I(inode)->cat_key = *idata->key; diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 924ecdef8091..0eb1a6092668 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -304,7 +304,6 @@ struct inode *hfsplus_new_inode(struct super_block *sb, int mode) inode->i_gid = current->fsgid; inode->i_nlink = 1; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - inode->i_blksize = HFSPLUS_SB(sb).alloc_blksz; INIT_LIST_HEAD(&HFSPLUS_I(inode).open_dir_list); init_MUTEX(&HFSPLUS_I(inode).extents_lock); atomic_set(&HFSPLUS_I(inode).opencnt, 0); @@ -407,7 +406,6 @@ int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd) type = hfs_bnode_read_u16(fd->bnode, fd->entryoffset); HFSPLUS_I(inode).dev = 0; - inode->i_blksize = HFSPLUS_SB(inode->i_sb).alloc_blksz; if (type == HFSPLUS_FOLDER) { struct hfsplus_cat_folder *folder = &entry.folder; diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index b82e3d9c8790..322e876c35ed 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -156,7 +156,6 @@ static int read_name(struct inode *ino, char *name) ino->i_mode = i_mode; ino->i_nlink = i_nlink; ino->i_size = i_size; - ino->i_blksize = i_blksize; ino->i_blocks = i_blocks; return(0); } diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c index 56f2c338c4d9..bcf6ee36e065 100644 --- a/fs/hpfs/inode.c +++ b/fs/hpfs/inode.c @@ -17,7 +17,6 @@ void hpfs_init_inode(struct inode *i) i->i_gid = hpfs_sb(sb)->sb_gid; i->i_mode = hpfs_sb(sb)->sb_mode; hpfs_inode->i_conv = hpfs_sb(sb)->sb_conv; - i->i_blksize = 512; i->i_size = -1; i->i_blocks = -1; diff --git a/fs/hppfs/hppfs_kern.c b/fs/hppfs/hppfs_kern.c index 3a9bdf58166f..dcb6d2e988b8 100644 --- a/fs/hppfs/hppfs_kern.c +++ b/fs/hppfs/hppfs_kern.c @@ -152,7 +152,6 @@ static void hppfs_read_inode(struct inode *ino) ino->i_mode = proc_ino->i_mode; ino->i_nlink = proc_ino->i_nlink; ino->i_size = proc_ino->i_size; - ino->i_blksize = proc_ino->i_blksize; ino->i_blocks = proc_ino->i_blocks; } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index c3920c96dadf..e025a31b4c64 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -357,7 +357,6 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, inode->i_mode = mode; inode->i_uid = uid; inode->i_gid = gid; - inode->i_blksize = HPAGE_SIZE; inode->i_blocks = 0; inode->i_mapping->a_ops = &hugetlbfs_aops; inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info; diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index 10e47897bac7..4527692f432b 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -1235,7 +1235,7 @@ static void isofs_read_inode(struct inode *inode) } inode->i_uid = sbi->s_uid; inode->i_gid = sbi->s_gid; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; ei->i_format_parm[0] = 0; ei->i_format_parm[1] = 0; @@ -1291,7 +1291,6 @@ static void isofs_read_inode(struct inode *inode) isonum_711 (de->ext_attr_length)); /* Set the number of blocks for stat() - should be done before RR */ - inode->i_blksize = PAGE_CACHE_SIZE; /* For stat() only */ inode->i_blocks = (inode->i_size + 511) >> 9; /* diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c index 7358ef87f16b..f5cf9c93e243 100644 --- a/fs/jffs/inode-v23.c +++ b/fs/jffs/inode-v23.c @@ -364,7 +364,6 @@ jffs_new_inode(const struct inode * dir, struct jffs_raw_inode *raw_inode, inode->i_ctime.tv_nsec = 0; inode->i_mtime.tv_nsec = 0; inode->i_atime.tv_nsec = 0; - inode->i_blksize = PAGE_SIZE; inode->i_blocks = (inode->i_size + 511) >> 9; f = jffs_find_file(c, raw_inode->ino); @@ -1706,7 +1705,6 @@ jffs_read_inode(struct inode *inode) inode->i_mtime.tv_nsec = inode->i_ctime.tv_nsec = 0; - inode->i_blksize = PAGE_SIZE; inode->i_blocks = (inode->i_size + 511) >> 9; if (S_ISREG(inode->i_mode)) { inode->i_op = &jffs_file_inode_operations; diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 4780f82825d6..72d9909d95ff 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -263,7 +263,6 @@ void jffs2_read_inode (struct inode *inode) inode->i_nlink = f->inocache->nlink; - inode->i_blksize = PAGE_SIZE; inode->i_blocks = (inode->i_size + 511) >> 9; switch (inode->i_mode & S_IFMT) { @@ -449,7 +448,6 @@ struct inode *jffs2_new_inode (struct inode *dir_i, int mode, struct jffs2_raw_i inode->i_atime = inode->i_ctime = inode->i_mtime = CURRENT_TIME_SEC; ri->atime = ri->mtime = ri->ctime = cpu_to_je32(I_SEC(inode->i_mtime)); - inode->i_blksize = PAGE_SIZE; inode->i_blocks = 0; inode->i_size = 0; diff --git a/fs/jfs/jfs_extent.c b/fs/jfs/jfs_extent.c index 4d52593a5fc6..4c74f0944f7e 100644 --- a/fs/jfs/jfs_extent.c +++ b/fs/jfs/jfs_extent.c @@ -468,7 +468,7 @@ int extRecord(struct inode *ip, xad_t * xp) int extFill(struct inode *ip, xad_t * xp) { int rc, nbperpage = JFS_SBI(ip->i_sb)->nbperpage; - s64 blkno = offsetXAD(xp) >> ip->i_blksize; + s64 blkno = offsetXAD(xp) >> ip->i_blkbits; // assert(ISSPARSE(ip)); diff --git a/fs/jfs/jfs_imap.c b/fs/jfs/jfs_imap.c index ccbe60aff83d..369d7f39c040 100644 --- a/fs/jfs/jfs_imap.c +++ b/fs/jfs/jfs_imap.c @@ -3115,7 +3115,6 @@ static int copy_from_dinode(struct dinode * dip, struct inode *ip) ip->i_mtime.tv_nsec = le32_to_cpu(dip->di_mtime.tv_nsec); ip->i_ctime.tv_sec = le32_to_cpu(dip->di_ctime.tv_sec); ip->i_ctime.tv_nsec = le32_to_cpu(dip->di_ctime.tv_nsec); - ip->i_blksize = ip->i_sb->s_blocksize; ip->i_blocks = LBLK2PBLK(ip->i_sb, le64_to_cpu(dip->di_nblocks)); ip->i_generation = le32_to_cpu(dip->di_gen); diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c index 495df402916d..bffaca9ae3a2 100644 --- a/fs/jfs/jfs_inode.c +++ b/fs/jfs/jfs_inode.c @@ -115,7 +115,6 @@ struct inode *ialloc(struct inode *parent, umode_t mode) } jfs_inode->mode2 |= mode; - inode->i_blksize = sb->s_blocksize; inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME; jfs_inode->otime = inode->i_ctime.tv_sec; diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index e1e0a6e6ebdf..f5afc129d6b1 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -257,7 +257,7 @@ static sector_t metapage_get_blocks(struct inode *inode, sector_t lblock, int rc = 0; int xflag; s64 xaddr; - sector_t file_blocks = (inode->i_size + inode->i_blksize - 1) >> + sector_t file_blocks = (inode->i_size + inode->i_sb->s_blocksize - 1) >> inode->i_blkbits; if (lblock >= file_blocks) diff --git a/fs/libfs.c b/fs/libfs.c index 2751793beeaa..8db5afb7b0a7 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -383,7 +383,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files return -ENOMEM; inode->i_mode = S_IFDIR | 0755; inode->i_uid = inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_op = &simple_dir_inode_operations; @@ -405,7 +404,6 @@ int simple_fill_super(struct super_block *s, int magic, struct tree_descr *files goto out; inode->i_mode = S_IFREG | files->mode; inode->i_uid = inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_fop = files->ops; diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c index 4a6abc49418e..df6b1075b549 100644 --- a/fs/minix/bitmap.c +++ b/fs/minix/bitmap.c @@ -254,7 +254,7 @@ struct inode * minix_new_inode(const struct inode * dir, int * error) inode->i_gid = (dir->i_mode & S_ISGID) ? dir->i_gid : current->fsgid; inode->i_ino = j; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; memset(&minix_i(inode)->u, 0, sizeof(minix_i(inode)->u)); insert_inode_hash(inode); mark_inode_dirty(inode); diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 826b9d830650..c11a4b9fb863 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -396,7 +396,7 @@ static void V1_minix_read_inode(struct inode * inode) inode->i_mtime.tv_nsec = 0; inode->i_atime.tv_nsec = 0; inode->i_ctime.tv_nsec = 0; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; for (i = 0; i < 9; i++) minix_inode->u.i1_data[i] = raw_inode->i_zone[i]; minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0])); @@ -429,7 +429,7 @@ static void V2_minix_read_inode(struct inode * inode) inode->i_mtime.tv_nsec = 0; inode->i_atime.tv_nsec = 0; inode->i_ctime.tv_nsec = 0; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; for (i = 0; i < 10; i++) minix_inode->u.i2_data[i] = raw_inode->i_zone[i]; minix_set_inode(inode, old_decode_dev(raw_inode->i_zone[0])); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 8244710e97dd..42e3bef270c9 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -223,7 +223,6 @@ static void ncp_set_attr(struct inode *inode, struct ncp_entry_info *nwinfo) inode->i_nlink = 1; inode->i_uid = server->m.uid; inode->i_gid = server->m.gid; - inode->i_blksize = NCP_BLOCK_SIZE; ncp_update_dates(inode, &nwinfo->i); ncp_update_inode(inode, nwinfo); diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 931f52a19579..bc9376ca86cd 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -277,10 +277,8 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr) * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - inode->i_blksize = inode->i_sb->s_blocksize; } else { inode->i_blocks = fattr->du.nfs2.blocks; - inode->i_blksize = fattr->du.nfs2.blocksize; } nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = jiffies; @@ -969,10 +967,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) * report the blocks in 512byte units */ inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); - inode->i_blksize = inode->i_sb->s_blocksize; } else { inode->i_blocks = fattr->du.nfs2.blocks; - inode->i_blksize = fattr->du.nfs2.blocksize; } if ((fattr->valid & NFS_ATTR_FATTR_V4) != 0 && diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index 31852121b3f5..933dbd89c2a4 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -556,8 +556,6 @@ static int ntfs_read_locked_inode(struct inode *vi) /* Setup the generic vfs inode parts now. */ - /* This is the optimal IO size (for stat), not the fs block size. */ - vi->i_blksize = PAGE_CACHE_SIZE; /* * This is for checking whether an inode has changed w.r.t. a file so * that the file can be updated if necessary (compare with f_version). @@ -1234,7 +1232,6 @@ static int ntfs_read_locked_attr_inode(struct inode *base_vi, struct inode *vi) base_ni = NTFS_I(base_vi); /* Just mirror the values from the base inode. */ - vi->i_blksize = base_vi->i_blksize; vi->i_version = base_vi->i_version; vi->i_uid = base_vi->i_uid; vi->i_gid = base_vi->i_gid; @@ -1504,7 +1501,6 @@ static int ntfs_read_locked_index_inode(struct inode *base_vi, struct inode *vi) ni = NTFS_I(vi); base_ni = NTFS_I(base_vi); /* Just mirror the values from the base inode. */ - vi->i_blksize = base_vi->i_blksize; vi->i_version = base_vi->i_version; vi->i_uid = base_vi->i_uid; vi->i_gid = base_vi->i_gid; diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c index 578fb3d5e803..584260fd6848 100644 --- a/fs/ntfs/mft.c +++ b/fs/ntfs/mft.c @@ -2637,11 +2637,6 @@ mft_rec_already_initialized: goto undo_mftbmp_alloc; } vi->i_ino = bit; - /* - * This is the optimal IO size (for stat), not the fs block - * size. - */ - vi->i_blksize = PAGE_CACHE_SIZE; /* * This is for checking whether an inode has changed w.r.t. a * file so that the file can be updated if necessary (compare diff --git a/fs/ocfs2/dlm/dlmfs.c b/fs/ocfs2/dlm/dlmfs.c index 0ff0898a0b9c..0368c6402182 100644 --- a/fs/ocfs2/dlm/dlmfs.c +++ b/fs/ocfs2/dlm/dlmfs.c @@ -335,7 +335,6 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; @@ -362,7 +361,6 @@ static struct inode *dlmfs_get_inode(struct inode *parent, inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->backing_dev_info = &dlmfs_backing_dev_info; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; diff --git a/fs/ocfs2/inode.c b/fs/ocfs2/inode.c index 69d3db569166..16e8e74dc966 100644 --- a/fs/ocfs2/inode.c +++ b/fs/ocfs2/inode.c @@ -269,7 +269,6 @@ int ocfs2_populate_inode(struct inode *inode, struct ocfs2_dinode *fe, inode->i_mode = le16_to_cpu(fe->i_mode); inode->i_uid = le32_to_cpu(fe->i_uid); inode->i_gid = le32_to_cpu(fe->i_gid); - inode->i_blksize = (u32)osb->s_clustersize; /* Fast symlinks will have i_size but no allocated clusters. */ if (S_ISLNK(inode->i_mode) && !fe->i_clusters) @@ -1258,8 +1257,6 @@ leave: void ocfs2_refresh_inode(struct inode *inode, struct ocfs2_dinode *fe) { - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); - spin_lock(&OCFS2_I(inode)->ip_lock); OCFS2_I(inode)->ip_clusters = le32_to_cpu(fe->i_clusters); @@ -1270,7 +1267,6 @@ void ocfs2_refresh_inode(struct inode *inode, inode->i_uid = le32_to_cpu(fe->i_uid); inode->i_gid = le32_to_cpu(fe->i_gid); inode->i_mode = le16_to_cpu(fe->i_mode); - inode->i_blksize = (u32) osb->s_clustersize; if (S_ISLNK(inode->i_mode) && le32_to_cpu(fe->i_clusters) == 0) inode->i_blocks = 0; else diff --git a/fs/pipe.c b/fs/pipe.c index 20352573e025..f3b6f71e9d0b 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -879,7 +879,6 @@ static struct inode * get_pipe_inode(void) inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; - inode->i_blksize = PAGE_SIZE; return inode; diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index fddbd61c68d0..5a41db2a218d 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -496,7 +496,6 @@ static void qnx4_read_inode(struct inode *inode) inode->i_ctime.tv_sec = le32_to_cpu(raw_inode->di_ctime); inode->i_ctime.tv_nsec = 0; inode->i_blocks = le32_to_cpu(raw_inode->di_first_xtnt.xtnt_size); - inode->i_blksize = QNX4_DIR_ENTRY_SIZE; memcpy(qnx4_inode, raw_inode, QNX4_DIR_ENTRY_SIZE); if (S_ISREG(inode->i_mode)) { diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index b9677335cc8d..bc0e51662424 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -58,7 +58,6 @@ struct inode *ramfs_get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->a_ops = &ramfs_aops; inode->i_mapping->backing_dev_info = &ramfs_backing_dev_info; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 52f1e2136546..8810fda0da46 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -17,8 +17,6 @@ #include #include -extern int reiserfs_default_io_size; /* default io size devuned in super.c */ - static int reiserfs_commit_write(struct file *f, struct page *page, unsigned from, unsigned to); static int reiserfs_prepare_write(struct file *f, struct page *page, @@ -1122,7 +1120,6 @@ static void init_inode(struct inode *inode, struct path *path) ih = PATH_PITEM_HEAD(path); copy_key(INODE_PKEY(inode), &(ih->ih_key)); - inode->i_blksize = reiserfs_default_io_size; INIT_LIST_HEAD(&(REISERFS_I(inode)->i_prealloc_list)); REISERFS_I(inode)->i_flags = 0; @@ -1877,7 +1874,6 @@ int reiserfs_new_inode(struct reiserfs_transaction_handle *th, } // these do not go to on-disk stat data inode->i_ino = le32_to_cpu(ih.ih_key.k_objectid); - inode->i_blksize = reiserfs_default_io_size; // store in in-core inode the key of stat data and version all // object items will have (directory items will have old offset diff --git a/fs/smbfs/inode.c b/fs/smbfs/inode.c index 92cf60aa6121..2c122ee83adb 100644 --- a/fs/smbfs/inode.c +++ b/fs/smbfs/inode.c @@ -166,7 +166,6 @@ smb_get_inode_attr(struct inode *inode, struct smb_fattr *fattr) fattr->f_mtime = inode->i_mtime; fattr->f_ctime = inode->i_ctime; fattr->f_atime = inode->i_atime; - fattr->f_blksize= inode->i_blksize; fattr->f_blocks = inode->i_blocks; fattr->attr = SMB_I(inode)->attr; @@ -200,7 +199,6 @@ smb_set_inode_attr(struct inode *inode, struct smb_fattr *fattr) inode->i_uid = fattr->f_uid; inode->i_gid = fattr->f_gid; inode->i_ctime = fattr->f_ctime; - inode->i_blksize= fattr->f_blksize; inode->i_blocks = fattr->f_blocks; inode->i_size = fattr->f_size; inode->i_mtime = fattr->f_mtime; diff --git a/fs/smbfs/proc.c b/fs/smbfs/proc.c index c3495059889d..40e174db9872 100644 --- a/fs/smbfs/proc.c +++ b/fs/smbfs/proc.c @@ -1826,7 +1826,6 @@ smb_init_dirent(struct smb_sb_info *server, struct smb_fattr *fattr) fattr->f_nlink = 1; fattr->f_uid = server->mnt->uid; fattr->f_gid = server->mnt->gid; - fattr->f_blksize = SMB_ST_BLKSIZE; fattr->f_unix = 0; } diff --git a/fs/stat.c b/fs/stat.c index 3a44dcf97da2..60a31d5e5966 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include @@ -32,7 +33,7 @@ void generic_fillattr(struct inode *inode, struct kstat *stat) stat->ctime = inode->i_ctime; stat->size = i_size_read(inode); stat->blocks = inode->i_blocks; - stat->blksize = inode->i_blksize; + stat->blksize = (1 << inode->i_blkbits); } EXPORT_SYMBOL(generic_fillattr); diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index fd7cd5f843d2..e79e38d52c00 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -125,7 +125,6 @@ struct inode * sysfs_new_inode(mode_t mode, struct sysfs_dirent * sd) { struct inode * inode = new_inode(sysfs_sb); if (inode) { - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->a_ops = &sysfs_aops; inode->i_mapping->backing_dev_info = &sysfs_backing_dev_info; diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c index 9b585d1081c0..115ab0d6f4bc 100644 --- a/fs/sysv/ialloc.c +++ b/fs/sysv/ialloc.c @@ -170,7 +170,7 @@ struct inode * sysv_new_inode(const struct inode * dir, mode_t mode) inode->i_uid = current->fsuid; inode->i_ino = fs16_to_cpu(sbi, ino); inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; memset(SYSV_I(inode)->i_data, 0, sizeof(SYSV_I(inode)->i_data)); SYSV_I(inode)->i_dir_start_lookup = 0; insert_inode_hash(inode); diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index 58b2d22142ba..d63c5e48b050 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -201,7 +201,7 @@ static void sysv_read_inode(struct inode *inode) inode->i_ctime.tv_nsec = 0; inode->i_atime.tv_nsec = 0; inode->i_mtime.tv_nsec = 0; - inode->i_blocks = inode->i_blksize = 0; + inode->i_blocks = 0; si = SYSV_I(inode); for (block = 0; block < 10+1+1+1; block++) diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index d1d38238ce65..8206983f2ebf 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -121,7 +121,6 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err) UDF_I_LOCATION(inode).logicalBlockNum = block; UDF_I_LOCATION(inode).partitionReferenceNum = UDF_I_LOCATION(dir).partitionReferenceNum; inode->i_ino = udf_get_lb_pblock(sb, UDF_I_LOCATION(inode), 0); - inode->i_blksize = PAGE_SIZE; inode->i_blocks = 0; UDF_I_LENEATTR(inode) = 0; UDF_I_LENALLOC(inode) = 0; diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 605f5111b6d8..b223b32db991 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -916,8 +916,6 @@ __udf_read_inode(struct inode *inode) * i_nlink = 1 * i_op = NULL; */ - inode->i_blksize = PAGE_SIZE; - bh = udf_read_ptagged(inode->i_sb, UDF_I_LOCATION(inode), 0, &ident); if (!bh) diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c index 9501dcd3b213..2ad1259c6eca 100644 --- a/fs/ufs/ialloc.c +++ b/fs/ufs/ialloc.c @@ -255,7 +255,6 @@ cg_found: inode->i_gid = current->fsgid; inode->i_ino = cg * uspi->s_ipg + bit; - inode->i_blksize = PAGE_SIZE; /* This is the optimal IO size (for stat), not the fs block size */ inode->i_blocks = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME_SEC; ufsi->i_flags = UFS_I(dir)->i_flags; diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index 30c6e8a9446c..ee1eaa6f4ec2 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -741,7 +741,6 @@ void ufs_read_inode(struct inode * inode) ufs1_read_inode(inode, ufs_inode + ufs_inotofsbo(inode->i_ino)); } - inode->i_blksize = PAGE_SIZE;/*This is the optimal IO size (for stat)*/ inode->i_version++; ufsi->i_lastfrag = (inode->i_size + uspi->s_fsize - 1) >> uspi->s_fshift; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 4754f342a5d3..9df9ed37d219 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -171,7 +171,6 @@ xfs_revalidate_inode( break; } - inode->i_blksize = xfs_preferred_iosize(mp); inode->i_generation = ip->i_d.di_gen; i_size_write(inode, ip->i_d.di_size); inode->i_blocks = diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c index 6628d96b6fd6..553fa731ade5 100644 --- a/fs/xfs/linux-2.6/xfs_vnode.c +++ b/fs/xfs/linux-2.6/xfs_vnode.c @@ -122,7 +122,6 @@ vn_revalidate_core( inode->i_blocks = vap->va_nblocks; inode->i_mtime = vap->va_mtime; inode->i_ctime = vap->va_ctime; - inode->i_blksize = vap->va_blocksize; if (vap->va_xflags & XFS_XFLAG_IMMUTABLE) inode->i_flags |= S_IMMUTABLE; else diff --git a/include/linux/fs.h b/include/linux/fs.h index 192e69bb55b5..8f74dfbb2edd 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -512,7 +512,6 @@ struct inode { struct timespec i_mtime; struct timespec i_ctime; unsigned int i_blkbits; - unsigned long i_blksize; unsigned long i_version; blkcnt_t i_blocks; unsigned short i_bytes; diff --git a/include/linux/nfsd/nfsfh.h b/include/linux/nfsd/nfsfh.h index f9edcd2ff3c8..31a3cb617ce0 100644 --- a/include/linux/nfsd/nfsfh.h +++ b/include/linux/nfsd/nfsfh.h @@ -269,14 +269,8 @@ fill_post_wcc(struct svc_fh *fhp) fhp->fh_post_uid = inode->i_uid; fhp->fh_post_gid = inode->i_gid; fhp->fh_post_size = inode->i_size; - if (inode->i_blksize) { - fhp->fh_post_blksize = inode->i_blksize; - fhp->fh_post_blocks = inode->i_blocks; - } else { - fhp->fh_post_blksize = BLOCK_SIZE; - /* how much do we care for accuracy with MinixFS? */ - fhp->fh_post_blocks = (inode->i_size+511) >> 9; - } + fhp->fh_post_blksize = BLOCK_SIZE; + fhp->fh_post_blocks = inode->i_blocks; fhp->fh_post_rdev[0] = htonl((u32)imajor(inode)); fhp->fh_post_rdev[1] = htonl((u32)iminor(inode)); fhp->fh_post_atime = inode->i_atime; diff --git a/include/linux/smb.h b/include/linux/smb.h index 6df3b1501559..f098dff93f6b 100644 --- a/include/linux/smb.h +++ b/include/linux/smb.h @@ -89,7 +89,6 @@ struct smb_fattr { struct timespec f_atime; struct timespec f_mtime; struct timespec f_ctime; - unsigned long f_blksize; unsigned long f_blocks; int f_unix; }; diff --git a/ipc/mqueue.c b/ipc/mqueue.c index f8381f0660b2..840f8a6fb85f 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -115,7 +115,6 @@ static struct inode *mqueue_get_inode(struct super_block *sb, int mode, inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mtime = inode->i_ctime = inode->i_atime = CURRENT_TIME; diff --git a/kernel/cpuset.c b/kernel/cpuset.c index cff41511269f..1b32c2c04c15 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -289,7 +289,6 @@ static struct inode *cpuset_new_inode(mode_t mode) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; inode->i_mapping->backing_dev_info = &cpuset_backing_dev_info; diff --git a/mm/shmem.c b/mm/shmem.c index 0a8e29cf87e0..eda907c3a86a 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -1351,7 +1351,6 @@ shmem_get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mode = mode; inode->i_uid = current->fsuid; inode->i_gid = current->fsgid; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_mapping->a_ops = &shmem_aops; inode->i_mapping->backing_dev_info = &shmem_backing_dev_info; diff --git a/net/sunrpc/rpc_pipe.c b/net/sunrpc/rpc_pipe.c index f65bea74734d..700c6e061a04 100644 --- a/net/sunrpc/rpc_pipe.c +++ b/net/sunrpc/rpc_pipe.c @@ -488,7 +488,6 @@ rpc_get_inode(struct super_block *sb, int mode) return NULL; inode->i_mode = mode; inode->i_uid = inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch(mode & S_IFMT) { diff --git a/security/inode.c b/security/inode.c index 176aacea8ca4..49ee51529396 100644 --- a/security/inode.c +++ b/security/inode.c @@ -64,7 +64,6 @@ static struct inode *get_inode(struct super_block *sb, int mode, dev_t dev) inode->i_mode = mode; inode->i_uid = 0; inode->i_gid = 0; - inode->i_blksize = PAGE_CACHE_SIZE; inode->i_blocks = 0; inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; switch (mode & S_IFMT) { diff --git a/security/selinux/selinuxfs.c b/security/selinux/selinuxfs.c index 00534c302ba2..bab7b386cb8d 100644 --- a/security/selinux/selinuxfs.c +++ b/security/selinux/selinuxfs.c @@ -771,7 +771,6 @@ static struct inode *sel_make_inode(struct super_block *sb, int mode) if (ret) { ret->i_mode = mode; ret->i_uid = ret->i_gid = 0; - ret->i_blksize = PAGE_CACHE_SIZE; ret->i_blocks = 0; ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME; } -- cgit v1.2.3 From b89a81712f486e4f7a606987413e387605fdeaf4 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 27 Sep 2006 01:51:04 -0700 Subject: [PATCH] sysctl: Allow /proc/sys without sys_sysctl Since sys_sysctl is deprecated start allow it to be compiled out. This should catch any remaining user space code that cares, and paves the way for further sysctl cleanups. [akpm@osdl.org: If sys_sysctl() is not compiled-in, emit a warning] Signed-off-by: Eric W. Biederman Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/ia32/sys_ia32.c | 2 +- arch/mips/kernel/linux32.c | 4 +- arch/powerpc/kernel/sys_ppc32.c | 2 +- arch/s390/kernel/compat_linux.c | 2 +- arch/sparc64/kernel/sys_sparc32.c | 2 +- arch/x86_64/ia32/sys_ia32.c | 2 +- fs/Kconfig | 19 +++++++ init/Kconfig | 31 ++++++----- kernel/sysctl.c | 113 +++++++++++++------------------------- 9 files changed, 81 insertions(+), 96 deletions(-) (limited to 'kernel') diff --git a/arch/ia64/ia32/sys_ia32.c b/arch/ia64/ia32/sys_ia32.c index 6aa3c51619ca..bddbd22706ed 100644 --- a/arch/ia64/ia32/sys_ia32.c +++ b/arch/ia64/ia32/sys_ia32.c @@ -1942,7 +1942,7 @@ struct sysctl32 { unsigned int __unused[4]; }; -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL asmlinkage long sys32_sysctl (struct sysctl32 __user *args) { diff --git a/arch/mips/kernel/linux32.c b/arch/mips/kernel/linux32.c index dc500e20cf14..43b1162d714f 100644 --- a/arch/mips/kernel/linux32.c +++ b/arch/mips/kernel/linux32.c @@ -991,7 +991,7 @@ struct sysctl_args32 unsigned int __unused[4]; }; -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL asmlinkage long sys32_sysctl(struct sysctl_args32 __user *args) { @@ -1032,7 +1032,7 @@ asmlinkage long sys32_sysctl(struct sysctl_args32 __user *args) return error; } -#endif /* CONFIG_SYSCTL */ +#endif /* CONFIG_SYSCTL_SYSCALL */ asmlinkage long sys32_newuname(struct new_utsname __user * name) { diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index 2e292863e982..5e391fc25340 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -740,7 +740,7 @@ asmlinkage long compat_sys_umask(u32 mask) return sys_umask((int)mask); } -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL struct __sysctl_args32 { u32 name; int nlen; diff --git a/arch/s390/kernel/compat_linux.c b/arch/s390/kernel/compat_linux.c index 785c9f70ac98..91b2884fa5c4 100644 --- a/arch/s390/kernel/compat_linux.c +++ b/arch/s390/kernel/compat_linux.c @@ -708,7 +708,7 @@ asmlinkage long sys32_sendfile64(int out_fd, int in_fd, return ret; } -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL struct __sysctl_args32 { u32 name; int nlen; diff --git a/arch/sparc64/kernel/sys_sparc32.c b/arch/sparc64/kernel/sys_sparc32.c index c88ae23ce812..69444f266e2d 100644 --- a/arch/sparc64/kernel/sys_sparc32.c +++ b/arch/sparc64/kernel/sys_sparc32.c @@ -1016,7 +1016,7 @@ struct __sysctl_args32 { asmlinkage long sys32_sysctl(struct __sysctl_args32 __user *args) { -#ifndef CONFIG_SYSCTL +#ifndef CONFIG_SYSCTL_SYSCALL return -ENOSYS; #else struct __sysctl_args32 tmp; diff --git a/arch/x86_64/ia32/sys_ia32.c b/arch/x86_64/ia32/sys_ia32.c index b0e82c7947dc..f280d3665f4b 100644 --- a/arch/x86_64/ia32/sys_ia32.c +++ b/arch/x86_64/ia32/sys_ia32.c @@ -648,7 +648,7 @@ sys32_pause(void) } -#ifdef CONFIG_SYSCTL +#ifdef CONFIG_SYSCTL_SYSCALL struct sysctl_ia32 { unsigned int name; int nlen; diff --git a/fs/Kconfig b/fs/Kconfig index a27002668bd3..d311198bba43 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -826,6 +826,25 @@ config PROC_VMCORE help Exports the dump image of crashed kernel in ELF format. +config PROC_SYSCTL + bool "Sysctl support (/proc/sys)" if EMBEDDED + depends on PROC_FS + select SYSCTL + default y + ---help--- + The sysctl interface provides a means of dynamically changing + certain kernel parameters and variables on the fly without requiring + a recompile of the kernel or reboot of the system. The primary + interface is through /proc/sys. If you say Y here a tree of + modifiable sysctl entries will be generated beneath the + /proc/sys directory. They are explained in the files + in . Note that enabling this + option will enlarge the kernel by at least 8 KB. + + As it is generally a good thing, you should say Y here unless + building a kernel for install/rescue disks or your system is very + limited in memory. + config SYSFS bool "sysfs file system support" if EMBEDDED default y diff --git a/init/Kconfig b/init/Kconfig index 9a7656f0b5ec..4381006dd666 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -273,21 +273,24 @@ config UID16 This enables the legacy 16-bit UID syscall wrappers. config SYSCTL - bool "Sysctl support" if EMBEDDED - default y + bool + +config SYSCTL_SYSCALL + bool "Sysctl syscall support" + default n + select SYSCTL ---help--- - The sysctl interface provides a means of dynamically changing - certain kernel parameters and variables on the fly without requiring - a recompile of the kernel or reboot of the system. The primary - interface consists of a system call, but if you say Y to "/proc - file system support", a tree of modifiable sysctl entries will be - generated beneath the /proc/sys directory. They are explained in the - files in . Note that enabling this - option will enlarge the kernel by at least 8 KB. - - As it is generally a good thing, you should say Y here unless - building a kernel for install/rescue disks or your system is very - limited in memory. + Enable the deprecated sysctl system call. sys_sysctl uses + binary paths that have been found to be a major pain to maintain + and use. The interface in /proc/sys is now the primary and what + everyone uses. + + Nothing has been using the binary sysctl interface for some time + time now so nothing should break if you disable sysctl syscall + support, and you kernel will get marginally smaller. + + Unless you have an application that uses the sys_syscall interface + you should probably say N here. config KALLSYMS bool "Load all symbols for debugging/kksymoops" if EMBEDDED diff --git a/kernel/sysctl.c b/kernel/sysctl.c index bcb3a181dbb2..8bfa7d117c54 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -137,8 +137,11 @@ extern int no_unaligned_warning; extern int max_lock_depth; #endif -static int parse_table(int __user *, int, void __user *, size_t __user *, void __user *, size_t, - ctl_table *, void **); +#ifdef CONFIG_SYSCTL_SYSCALL +static int parse_table(int __user *, int, void __user *, size_t __user *, + void __user *, size_t, ctl_table *, void **); +#endif + static int proc_doutsstring(ctl_table *table, int write, struct file *filp, void __user *buffer, size_t *lenp, loff_t *ppos); @@ -165,7 +168,7 @@ int sysctl_legacy_va_layout; /* /proc declarations: */ -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL static ssize_t proc_readsys(struct file *, char __user *, size_t, loff_t *); static ssize_t proc_writesys(struct file *, const char __user *, size_t, loff_t *); @@ -1166,12 +1169,13 @@ static void start_unregistering(struct ctl_table_header *p) void __init sysctl_init(void) { -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL register_proc_table(root_table, proc_sys_root, &root_table_header); init_irq_proc(); #endif } +#ifdef CONFIG_SYSCTL_SYSCALL int do_sysctl(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen) { @@ -1225,6 +1229,7 @@ asmlinkage long sys_sysctl(struct __sysctl_args __user *args) unlock_kernel(); return error; } +#endif /* CONFIG_SYSCTL_SYSCALL */ /* * ctl_perm does NOT grant the superuser all rights automatically, because @@ -1251,6 +1256,7 @@ static inline int ctl_perm(ctl_table *table, int op) return test_perm(table->mode, op); } +#ifdef CONFIG_SYSCTL_SYSCALL static int parse_table(int __user *name, int nlen, void __user *oldval, size_t __user *oldlenp, void __user *newval, size_t newlen, @@ -1340,6 +1346,7 @@ int do_sysctl_strategy (ctl_table *table, } return 0; } +#endif /* CONFIG_SYSCTL_SYSCALL */ /** * register_sysctl_table - register a sysctl hierarchy @@ -1427,7 +1434,7 @@ struct ctl_table_header *register_sysctl_table(ctl_table * table, else list_add_tail(&tmp->ctl_entry, &root_table_header.ctl_entry); spin_unlock(&sysctl_lock); -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL register_proc_table(table, proc_sys_root, tmp); #endif return tmp; @@ -1445,18 +1452,31 @@ void unregister_sysctl_table(struct ctl_table_header * header) might_sleep(); spin_lock(&sysctl_lock); start_unregistering(header); -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL unregister_proc_table(header->ctl_table, proc_sys_root); #endif spin_unlock(&sysctl_lock); kfree(header); } +#else /* !CONFIG_SYSCTL */ +struct ctl_table_header * register_sysctl_table(ctl_table * table, + int insert_at_head) +{ + return NULL; +} + +void unregister_sysctl_table(struct ctl_table_header * table) +{ +} + +#endif /* CONFIG_SYSCTL */ + /* * /proc/sys support */ -#ifdef CONFIG_PROC_FS +#ifdef CONFIG_PROC_SYSCTL /* Scan the sysctl entries in table and add them all into /proc */ static void register_proc_table(ctl_table * table, struct proc_dir_entry *root, void *set) @@ -2318,6 +2338,7 @@ int proc_doulongvec_ms_jiffies_minmax(ctl_table *table, int write, #endif /* CONFIG_PROC_FS */ +#ifdef CONFIG_SYSCTL_SYSCALL /* * General sysctl support routines */ @@ -2460,11 +2481,19 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, return 1; } -#else /* CONFIG_SYSCTL */ +#else /* CONFIG_SYSCTL_SYSCALL */ asmlinkage long sys_sysctl(struct __sysctl_args __user *args) { + static int msg_count; + + if (msg_count < 5) { + msg_count++; + printk(KERN_INFO + "warning: process `%s' used the removed sysctl " + "system call\n", current->comm); + } return -ENOSYS; } @@ -2496,73 +2525,7 @@ int sysctl_ms_jiffies(ctl_table *table, int __user *name, int nlen, return -ENOSYS; } -int proc_dostring(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec_bset(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec_minmax(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec_jiffies(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec_userhz_jiffies(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_dointvec_ms_jiffies(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_doulongvec_minmax(ctl_table *table, int write, struct file *filp, - void __user *buffer, size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -int proc_doulongvec_ms_jiffies_minmax(ctl_table *table, int write, - struct file *filp, - void __user *buffer, - size_t *lenp, loff_t *ppos) -{ - return -ENOSYS; -} - -struct ctl_table_header * register_sysctl_table(ctl_table * table, - int insert_at_head) -{ - return NULL; -} - -void unregister_sysctl_table(struct ctl_table_header * table) -{ -} - -#endif /* CONFIG_SYSCTL */ +#endif /* CONFIG_SYSCTL_SYSCALL */ /* * No sense putting this after each symbol definition, twice, -- cgit v1.2.3 From c18258c6f0848f97e85287f6271c511a092bb784 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 27 Sep 2006 01:51:06 -0700 Subject: [PATCH] pid: Implement transfer_pid and use it to simplify de_thread In de_thread we move pids from one process to another, a rather ugly case. The function transfer_pid makes it clear what we are doing, and makes the action atomic. This is useful we ever want to atomically traverse the process group and session lists, in a rcu safe manner. Even if the atomic properties this change should be a win as transfer_pid should be less code to execute than executing both attach_pid and detach_pid, and this should make de_thread slightly smaller as only a single function call needs to be emitted. The only downside is that the code might be slower to execute as the odds are against transfer_pid being in cache. Signed-off-by: Eric W. Biederman Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- fs/exec.c | 11 ++++------- include/linux/pid.h | 2 ++ kernel/pid.c | 9 +++++++++ 3 files changed, 15 insertions(+), 7 deletions(-) (limited to 'kernel') diff --git a/fs/exec.c b/fs/exec.c index 54135df2a966..b7aa3d6422d6 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -696,23 +696,20 @@ static int de_thread(struct task_struct *tsk) */ /* Become a process group leader with the old leader's pid. - * Note: The old leader also uses thispid until release_task + * The old leader becomes a thread of the this thread group. + * Note: The old leader also uses this pid until release_task * is called. Odd but simple and correct. */ detach_pid(current, PIDTYPE_PID); current->pid = leader->pid; attach_pid(current, PIDTYPE_PID, current->pid); - attach_pid(current, PIDTYPE_PGID, current->signal->pgrp); - attach_pid(current, PIDTYPE_SID, current->signal->session); + transfer_pid(leader, current, PIDTYPE_PGID); + transfer_pid(leader, current, PIDTYPE_SID); list_replace_rcu(&leader->tasks, ¤t->tasks); current->group_leader = current; leader->group_leader = current; - /* Reduce leader to a thread */ - detach_pid(leader, PIDTYPE_PGID); - detach_pid(leader, PIDTYPE_SID); - current->exit_signal = SIGCHLD; BUG_ON(leader->exit_state != EXIT_ZOMBIE); diff --git a/include/linux/pid.h b/include/linux/pid.h index 29960b03bef7..93da7e2d9f30 100644 --- a/include/linux/pid.h +++ b/include/linux/pid.h @@ -76,6 +76,8 @@ extern int FASTCALL(attach_pid(struct task_struct *task, enum pid_type type, int nr)); extern void FASTCALL(detach_pid(struct task_struct *task, enum pid_type)); +extern void FASTCALL(transfer_pid(struct task_struct *old, + struct task_struct *new, enum pid_type)); /* * look up a PID in the hash table. Must be called with the tasklist_lock diff --git a/kernel/pid.c b/kernel/pid.c index 93e212f20671..6db82b68e2f8 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -252,6 +252,15 @@ void fastcall detach_pid(struct task_struct *task, enum pid_type type) free_pid(pid); } +/* transfer_pid is an optimization of attach_pid(new), detach_pid(old) */ +void fastcall transfer_pid(struct task_struct *old, struct task_struct *new, + enum pid_type type) +{ + new->pids[type].pid = old->pids[type].pid; + hlist_replace_rcu(&old->pids[type].node, &new->pids[type].node); + old->pids[type].pid = NULL; +} + struct task_struct * fastcall pid_task(struct pid *pid, enum pid_type type) { struct task_struct *result = NULL; -- cgit v1.2.3 From 65800ac77e080cf159d6c1207b6886e18f22bc08 Mon Sep 17 00:00:00 2001 From: "Eric W. Biederman" Date: Wed, 27 Sep 2006 01:51:11 -0700 Subject: [PATCH] pid: remove temporary debug code in attach_pid With the patches flying between Oleg and myself somehow this temporary debug code got left in pid.c. It was never intended to make it to the stable kernel. Signed-off-by: Eric W. Biederman Cc: Oleg Nesterov Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- kernel/pid.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'kernel') diff --git a/kernel/pid.c b/kernel/pid.c index 6db82b68e2f8..8387e8c68193 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -223,9 +223,6 @@ int fastcall attach_pid(struct task_struct *task, enum pid_type type, int nr) struct pid_link *link; struct pid *pid; - WARN_ON(!task->pid); /* to be removed soon */ - WARN_ON(!nr); /* to be removed soon */ - link = &task->pids[type]; link->pid = pid = find_pid(nr); hlist_add_head_rcu(&link->node, &pid->tasks[type]); -- cgit v1.2.3