diff options
Diffstat (limited to 'kernel')
49 files changed, 1885 insertions, 845 deletions
diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz index 248e1c396f8b..4af15802ccd4 100644 --- a/kernel/Kconfig.hz +++ b/kernel/Kconfig.hz @@ -7,7 +7,7 @@ choice default HZ_250 help Allows the configuration of the timer frequency. It is customary - to have the timer interrupt run at 1000 HZ but 100 HZ may be more + to have the timer interrupt run at 1000 Hz but 100 Hz may be more beneficial for servers and NUMA systems that do not need to have a fast response for user interaction and that may experience bus contention and cacheline bounces as a result of timer interrupts. @@ -19,21 +19,30 @@ choice config HZ_100 bool "100 HZ" help - 100 HZ is a typical choice for servers, SMP and NUMA systems + 100 Hz is a typical choice for servers, SMP and NUMA systems with lots of processors that may show reduced performance if too many timer interrupts are occurring. config HZ_250 bool "250 HZ" help - 250 HZ is a good compromise choice allowing server performance + 250 Hz is a good compromise choice allowing server performance while also showing good interactive responsiveness even - on SMP and NUMA systems. + on SMP and NUMA systems. If you are going to be using NTSC video + or multimedia, selected 300Hz instead. + + config HZ_300 + bool "300 HZ" + help + 300 Hz is a good compromise choice allowing server performance + while also showing good interactive responsiveness even + on SMP and NUMA systems and exactly dividing by both PAL and + NTSC frame rates for video and multimedia work. config HZ_1000 bool "1000 HZ" help - 1000 HZ is the preferred choice for desktop systems and other + 1000 Hz is the preferred choice for desktop systems and other systems requiring fast interactive responses to events. endchoice @@ -42,5 +51,6 @@ config HZ int default 100 if HZ_100 default 250 if HZ_250 + default 300 if HZ_300 default 1000 if HZ_1000 diff --git a/kernel/acct.c b/kernel/acct.c index 0aad5ca36a81..dc12db8600e7 100644 --- a/kernel/acct.c +++ b/kernel/acct.c @@ -89,7 +89,8 @@ struct acct_glbs { struct timer_list timer; }; -static struct acct_glbs acct_globals __cacheline_aligned = {SPIN_LOCK_UNLOCKED}; +static struct acct_glbs acct_globals __cacheline_aligned = + {__SPIN_LOCK_UNLOCKED(acct_globals.lock)}; /* * Called whenever the timer says to check the free space. diff --git a/kernel/audit.c b/kernel/audit.c index 98106f6078b0..d9b690ac684b 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -57,6 +57,7 @@ #include <linux/netlink.h> #include <linux/selinux.h> #include <linux/inotify.h> +#include <linux/freezer.h> #include "audit.h" diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 4f40d923af8e..2e896f8ae29e 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -636,10 +636,9 @@ static struct audit_rule *audit_krule_to_rule(struct audit_krule *krule) struct audit_rule *rule; int i; - rule = kmalloc(sizeof(*rule), GFP_KERNEL); + rule = kzalloc(sizeof(*rule), GFP_KERNEL); if (unlikely(!rule)) return NULL; - memset(rule, 0, sizeof(*rule)); rule->flags = krule->flags | krule->listnr; rule->action = krule->action; diff --git a/kernel/auditsc.c b/kernel/auditsc.c index ab97e5101232..40722e26de98 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -731,7 +731,7 @@ static inline void audit_free_context(struct audit_context *context) printk(KERN_ERR "audit: freed %d contexts\n", count); } -static void audit_log_task_context(struct audit_buffer *ab) +void audit_log_task_context(struct audit_buffer *ab) { char *ctx = NULL; ssize_t len = 0; @@ -760,6 +760,8 @@ error_path: return; } +EXPORT_SYMBOL(audit_log_task_context); + static void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk) { char name[sizeof(tsk->comm)]; @@ -1488,6 +1490,8 @@ uid_t audit_get_loginuid(struct audit_context *ctx) return ctx ? ctx->loginuid : -1; } +EXPORT_SYMBOL(audit_get_loginuid); + /** * __audit_mq_open - record audit data for a POSIX MQ open * @oflag: open flag diff --git a/kernel/configs.c b/kernel/configs.c index f9e31974f4ad..8fa1fb28f8a7 100644 --- a/kernel/configs.c +++ b/kernel/configs.c @@ -75,7 +75,7 @@ ikconfig_read_current(struct file *file, char __user *buf, return count; } -static struct file_operations ikconfig_file_ops = { +static const struct file_operations ikconfig_file_ops = { .owner = THIS_MODULE, .read = ikconfig_read_current, }; diff --git a/kernel/cpu.c b/kernel/cpu.c index 272254f20d97..9124669f4586 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -270,11 +270,7 @@ int disable_nonboot_cpus(void) 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 */ diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 6313c38c930e..0a6b4d89f9a0 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -729,9 +729,11 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) } /* Remaining checks don't apply to root cpuset */ - if ((par = cur->parent) == NULL) + if (cur == &top_cpuset) return 0; + par = cur->parent; + /* We must be a subset of our parent cpuset */ if (!is_cpuset_subset(trial, par)) return -EACCES; @@ -1060,10 +1062,7 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf) cpu_exclusive_changed = (is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs)); mutex_lock(&callback_mutex); - if (turning_on) - set_bit(bit, &cs->flags); - else - clear_bit(bit, &cs->flags); + cs->flags = trialcs.flags; mutex_unlock(&callback_mutex); if (cpu_exclusive_changed) @@ -1281,7 +1280,8 @@ typedef enum { FILE_TASKLIST, } cpuset_filetype_t; -static ssize_t cpuset_common_file_write(struct file *file, const char __user *userbuf, +static ssize_t cpuset_common_file_write(struct file *file, + const char __user *userbuf, size_t nbytes, loff_t *unused_ppos) { struct cpuset *cs = __d_cs(file->f_dentry->d_parent); @@ -1292,7 +1292,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us int retval = 0; /* Crude upper limit on largest legitimate cpulist user might write. */ - if (nbytes > 100 + 6 * NR_CPUS) + if (nbytes > 100 + 6 * max(NR_CPUS, MAX_NUMNODES)) return -E2BIG; /* +1 for nul-terminator */ @@ -1532,7 +1532,7 @@ static int cpuset_rename(struct inode *old_dir, struct dentry *old_dentry, return simple_rename(old_dir, old_dentry, new_dir, new_dentry); } -static struct file_operations cpuset_file_operations = { +static const struct file_operations cpuset_file_operations = { .read = cpuset_file_read, .write = cpuset_file_write, .llseek = generic_file_llseek, @@ -2045,7 +2045,6 @@ out: return err; } -#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_MEMORY_HOTPLUG) /* * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, @@ -2109,9 +2108,7 @@ static void common_cpu_mem_hotplug_unplug(void) mutex_unlock(&callback_mutex); mutex_unlock(&manage_mutex); } -#endif -#ifdef CONFIG_HOTPLUG_CPU /* * The top_cpuset tracks what CPUs and Memory Nodes are online, * period. This is necessary in order to make cpusets transparent @@ -2128,7 +2125,6 @@ static int cpuset_handle_cpuhp(struct notifier_block *nb, common_cpu_mem_hotplug_unplug(); return 0; } -#endif #ifdef CONFIG_MEMORY_HOTPLUG /* @@ -2610,7 +2606,7 @@ static int cpuset_open(struct inode *inode, struct file *file) return single_open(file, proc_cpuset_show, pid); } -struct file_operations proc_cpuset_operations = { +const struct file_operations proc_cpuset_operations = { .open = cpuset_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/delayacct.c b/kernel/delayacct.c index 66a0ea48751d..766d5912b26a 100644 --- a/kernel/delayacct.c +++ b/kernel/delayacct.c @@ -20,7 +20,7 @@ #include <linux/delayacct.h> int delayacct_on __read_mostly = 1; /* Delay accounting turned on/off */ -kmem_cache_t *delayacct_cache; +struct kmem_cache *delayacct_cache; static int __init delayacct_setup_disable(char *str) { @@ -41,7 +41,7 @@ void delayacct_init(void) void __delayacct_tsk_init(struct task_struct *tsk) { - tsk->delays = kmem_cache_zalloc(delayacct_cache, SLAB_KERNEL); + tsk->delays = kmem_cache_zalloc(delayacct_cache, GFP_KERNEL); if (tsk->delays) spin_lock_init(&tsk->delays->lock); } diff --git a/kernel/dma.c b/kernel/dma.c index 2020644c938a..937b13ca33ba 100644 --- a/kernel/dma.c +++ b/kernel/dma.c @@ -140,7 +140,7 @@ static int proc_dma_open(struct inode *inode, struct file *file) return single_open(file, proc_dma_show, NULL); } -static struct file_operations proc_dma_operations = { +static const struct file_operations proc_dma_operations = { .open = proc_dma_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/exit.c b/kernel/exit.c index 06de6c4e8ca3..4e3f919edc48 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -850,9 +850,7 @@ static void exit_notify(struct task_struct *tsk) fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; - struct taskstats *tidstats; int group_dead; - unsigned int mycpu; profile_task_exit(tsk); @@ -890,8 +888,6 @@ fastcall NORET_TYPE void do_exit(long code) current->comm, current->pid, preempt_count()); - taskstats_exit_alloc(&tidstats, &mycpu); - acct_update_integrals(tsk); if (tsk->mm) { update_hiwater_rss(tsk->mm); @@ -911,8 +907,8 @@ fastcall NORET_TYPE void do_exit(long code) #endif if (unlikely(tsk->audit_context)) audit_free(tsk); - taskstats_exit_send(tsk, tidstats, group_dead, mycpu); - taskstats_exit_free(tidstats); + + taskstats_exit(tsk, group_dead); exit_mm(tsk); diff --git a/kernel/fork.c b/kernel/fork.c index 8cdd3e72ba55..7f2e31ba33af 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -82,26 +82,26 @@ int nr_processes(void) #ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR # define alloc_task_struct() kmem_cache_alloc(task_struct_cachep, GFP_KERNEL) # define free_task_struct(tsk) kmem_cache_free(task_struct_cachep, (tsk)) -static kmem_cache_t *task_struct_cachep; +static struct kmem_cache *task_struct_cachep; #endif /* SLAB cache for signal_struct structures (tsk->signal) */ -static kmem_cache_t *signal_cachep; +static struct kmem_cache *signal_cachep; /* SLAB cache for sighand_struct structures (tsk->sighand) */ -kmem_cache_t *sighand_cachep; +struct kmem_cache *sighand_cachep; /* SLAB cache for files_struct structures (tsk->files) */ -kmem_cache_t *files_cachep; +struct kmem_cache *files_cachep; /* SLAB cache for fs_struct structures (tsk->fs) */ -kmem_cache_t *fs_cachep; +struct kmem_cache *fs_cachep; /* SLAB cache for vm_area_struct structures */ -kmem_cache_t *vm_area_cachep; +struct kmem_cache *vm_area_cachep; /* SLAB cache for mm_struct structures (tsk->mm) */ -static kmem_cache_t *mm_cachep; +static struct kmem_cache *mm_cachep; void free_task(struct task_struct *tsk) { @@ -237,7 +237,7 @@ static inline int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm) goto fail_nomem; charge = len; } - tmp = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL); + tmp = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL); if (!tmp) goto fail_nomem; *tmp = *mpnt; @@ -319,7 +319,7 @@ static inline void mm_free_pgd(struct mm_struct * mm) __cacheline_aligned_in_smp DEFINE_SPINLOCK(mmlist_lock); -#define allocate_mm() (kmem_cache_alloc(mm_cachep, SLAB_KERNEL)) +#define allocate_mm() (kmem_cache_alloc(mm_cachep, GFP_KERNEL)) #define free_mm(mm) (kmem_cache_free(mm_cachep, (mm))) #include <linux/init_task.h> @@ -448,7 +448,16 @@ void mm_release(struct task_struct *tsk, struct mm_struct *mm) tsk->vfork_done = NULL; complete(vfork_done); } - if (tsk->clear_child_tid && atomic_read(&mm->mm_users) > 1) { + + /* + * If we're exiting normally, clear a user-space tid field if + * requested. We leave this alone when dying by signal, to leave + * the value intact in a core dump, and to save the unnecessary + * trouble otherwise. Userland only wants this done for a sys_exit. + */ + if (tsk->clear_child_tid + && !(tsk->flags & PF_SIGNALED) + && atomic_read(&mm->mm_users) > 1) { u32 __user * tidptr = tsk->clear_child_tid; tsk->clear_child_tid = NULL; @@ -479,6 +488,10 @@ static struct mm_struct *dup_mm(struct task_struct *tsk) memcpy(mm, oldmm, sizeof(*mm)); + /* Initializing for Swap token stuff */ + mm->token_priority = 0; + mm->last_interval = 0; + if (!mm_init(mm)) goto fail_nomem; @@ -542,6 +555,10 @@ static int copy_mm(unsigned long clone_flags, struct task_struct * tsk) goto fail_nomem; good_mm: + /* Initializing for Swap token stuff */ + mm->token_priority = 0; + mm->last_interval = 0; + tsk->mm = mm; tsk->active_mm = mm; return 0; @@ -613,7 +630,7 @@ static struct files_struct *alloc_files(void) struct files_struct *newf; struct fdtable *fdt; - newf = kmem_cache_alloc(files_cachep, SLAB_KERNEL); + newf = kmem_cache_alloc(files_cachep, GFP_KERNEL); if (!newf) goto out; @@ -830,7 +847,6 @@ static inline int copy_signal(unsigned long clone_flags, struct task_struct * ts if (clone_flags & CLONE_THREAD) { atomic_inc(¤t->signal->count); atomic_inc(¤t->signal->live); - taskstats_tgid_alloc(current); return 0; } sig = kmem_cache_alloc(signal_cachep, GFP_KERNEL); @@ -1303,7 +1319,7 @@ fork_out: return ERR_PTR(retval); } -struct pt_regs * __devinit __attribute__((weak)) idle_regs(struct pt_regs *regs) +noinline struct pt_regs * __devinit __attribute__((weak)) idle_regs(struct pt_regs *regs) { memset(regs, 0, sizeof(struct pt_regs)); return regs; @@ -1413,7 +1429,7 @@ long do_fork(unsigned long clone_flags, #define ARCH_MIN_MMSTRUCT_ALIGN 0 #endif -static void sighand_ctor(void *data, kmem_cache_t *cachep, unsigned long flags) +static void sighand_ctor(void *data, struct kmem_cache *cachep, unsigned long flags) { struct sighand_struct *sighand = data; diff --git a/kernel/futex.c b/kernel/futex.c index 93ef30ba209f..95989a3b4168 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -282,9 +282,9 @@ static inline int get_futex_value_locked(u32 *dest, u32 __user *from) { int ret; - inc_preempt_count(); + pagefault_disable(); ret = __copy_from_user_inatomic(dest, from, sizeof(u32)); - dec_preempt_count(); + pagefault_enable(); return ret ? -EFAULT : 0; } @@ -324,12 +324,11 @@ static int refill_pi_state_cache(void) if (likely(current->pi_state_cache)) return 0; - pi_state = kmalloc(sizeof(*pi_state), GFP_KERNEL); + pi_state = kzalloc(sizeof(*pi_state), GFP_KERNEL); if (!pi_state) return -ENOMEM; - memset(pi_state, 0, sizeof(*pi_state)); INIT_LIST_HEAD(&pi_state->list); /* pi_mutex gets initialized later */ pi_state->owner = NULL; @@ -553,7 +552,7 @@ static void wake_futex(struct futex_q *q) * at the end of wake_up_all() does not prevent this store from * moving. */ - wmb(); + smp_wmb(); q->lock_ptr = NULL; } @@ -585,9 +584,9 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this) if (!(uval & FUTEX_OWNER_DIED)) { newval = FUTEX_WAITERS | new_owner->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (curval == -EFAULT) return -EFAULT; if (curval != uval) @@ -618,9 +617,9 @@ static int unlock_futex_pi(u32 __user *uaddr, u32 uval) * There is no waiter, so we unlock the futex. The owner died * bit has not to be preserved here. We are the owner: */ - inc_preempt_count(); + pagefault_disable(); oldval = futex_atomic_cmpxchg_inatomic(uaddr, uval, 0); - dec_preempt_count(); + pagefault_enable(); if (oldval == -EFAULT) return oldval; @@ -1158,9 +1157,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, */ newval = current->pid; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, 0, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1183,9 +1182,9 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, uval = curval; newval = uval | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1215,10 +1214,10 @@ static int futex_lock_pi(u32 __user *uaddr, int detect, unsigned long sec, newval = current->pid | FUTEX_OWNER_DIED | FUTEX_WAITERS; - inc_preempt_count(); + pagefault_disable(); curval = futex_atomic_cmpxchg_inatomic(uaddr, uval, newval); - dec_preempt_count(); + pagefault_enable(); if (unlikely(curval == -EFAULT)) goto uaddr_faulted; @@ -1390,9 +1389,9 @@ retry_locked: * anyone else up: */ if (!(uval & FUTEX_OWNER_DIED)) { - inc_preempt_count(); + pagefault_disable(); uval = futex_atomic_cmpxchg_inatomic(uaddr, current->pid, 0); - dec_preempt_count(); + pagefault_enable(); } if (unlikely(uval == -EFAULT)) @@ -1493,7 +1492,7 @@ static unsigned int futex_poll(struct file *filp, return ret; } -static struct file_operations futex_fops = { +static const struct file_operations futex_fops = { .release = futex_close, .poll = futex_poll, }; @@ -1858,10 +1857,16 @@ static struct file_system_type futex_fs_type = { static int __init init(void) { - unsigned int i; + int i = register_filesystem(&futex_fs_type); + + if (i) + return i; - register_filesystem(&futex_fs_type); futex_mnt = kern_mount(&futex_fs_type); + if (IS_ERR(futex_mnt)) { + unregister_filesystem(&futex_fs_type); + return PTR_ERR(futex_mnt); + } for (i = 0; i < ARRAY_SIZE(futex_queues); i++) { INIT_LIST_HEAD(&futex_queues[i].chain); diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c index a681912bc89a..aff1f0fabb0d 100644 --- a/kernel/irq/handle.c +++ b/kernel/irq/handle.c @@ -54,7 +54,7 @@ struct irq_desc irq_desc[NR_IRQS] __cacheline_aligned = { .chip = &no_irq_chip, .handle_irq = handle_bad_irq, .depth = 1, - .lock = SPIN_LOCK_UNLOCKED, + .lock = __SPIN_LOCK_UNLOCKED(irq_desc->lock), #ifdef CONFIG_SMP .affinity = CPU_MASK_ALL #endif diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index eeac3e313b2b..ab63cfc42992 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -20,6 +20,7 @@ #include <linux/proc_fs.h> #include <linux/sched.h> /* for cond_resched */ #include <linux/mm.h> +#include <linux/ctype.h> #include <asm/sections.h> @@ -301,13 +302,6 @@ struct kallsym_iter char name[KSYM_NAME_LEN+1]; }; -/* Only label it "global" if it is exported. */ -static void upcase_if_global(struct kallsym_iter *iter) -{ - if (is_exported(iter->name, iter->owner)) - iter->type += 'A' - 'a'; -} - static int get_ksymbol_mod(struct kallsym_iter *iter) { iter->owner = module_get_kallsym(iter->pos - kallsyms_num_syms, @@ -316,7 +310,10 @@ static int get_ksymbol_mod(struct kallsym_iter *iter) if (iter->owner == NULL) return 0; - upcase_if_global(iter); + /* Label it "global" if it is exported, "local" if not exported. */ + iter->type = is_exported(iter->name, iter->owner) + ? toupper(iter->type) : tolower(iter->type); + return 1; } @@ -401,7 +398,7 @@ static int s_show(struct seq_file *m, void *p) return 0; } -static struct seq_operations kallsyms_op = { +static const struct seq_operations kallsyms_op = { .start = s_start, .next = s_next, .stop = s_stop, @@ -436,7 +433,7 @@ static int kallsyms_release(struct inode *inode, struct file *file) return seq_release(inode, file); } -static struct file_operations kallsyms_operations = { +static const struct file_operations kallsyms_operations = { .open = kallsyms_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/kexec.c b/kernel/kexec.c index fcdd5d2bc3f4..afbbbe981be2 100644 --- a/kernel/kexec.c +++ b/kernel/kexec.c @@ -20,6 +20,8 @@ #include <linux/syscalls.h> #include <linux/ioport.h> #include <linux/hardirq.h> +#include <linux/elf.h> +#include <linux/elfcore.h> #include <asm/page.h> #include <asm/uaccess.h> @@ -108,11 +110,10 @@ static int do_kimage_alloc(struct kimage **rimage, unsigned long entry, /* Allocate a controlling structure */ result = -ENOMEM; - image = kmalloc(sizeof(*image), GFP_KERNEL); + image = kzalloc(sizeof(*image), GFP_KERNEL); if (!image) goto out; - memset(image, 0, sizeof(*image)); image->head = 0; image->entry = &image->head; image->last_entry = &image->head; @@ -1067,6 +1068,60 @@ void crash_kexec(struct pt_regs *regs) } } +static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data, + size_t data_len) +{ + struct elf_note note; + + note.n_namesz = strlen(name) + 1; + note.n_descsz = data_len; + note.n_type = type; + memcpy(buf, ¬e, sizeof(note)); + buf += (sizeof(note) + 3)/4; + memcpy(buf, name, note.n_namesz); + buf += (note.n_namesz + 3)/4; + memcpy(buf, data, note.n_descsz); + buf += (note.n_descsz + 3)/4; + + return buf; +} + +static void final_note(u32 *buf) +{ + struct elf_note note; + + note.n_namesz = 0; + note.n_descsz = 0; + note.n_type = 0; + memcpy(buf, ¬e, sizeof(note)); +} + +void crash_save_cpu(struct pt_regs *regs, int cpu) +{ + struct elf_prstatus prstatus; + u32 *buf; + + if ((cpu < 0) || (cpu >= NR_CPUS)) + return; + + /* Using ELF notes here is opportunistic. + * I need a well defined structure format + * for the data I pass, and I need tags + * on the data to indicate what information I have + * squirrelled away. ELF notes happen to provide + * all of that, so there is no need to invent something new. + */ + buf = (u32*)per_cpu_ptr(crash_notes, cpu); + if (!buf) + return; + memset(&prstatus, 0, sizeof(prstatus)); + prstatus.pr_pid = current->pid; + elf_core_copy_regs(&prstatus.pr_reg, regs); + buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus, + sizeof(prstatus)); + final_note(buf); +} + static int __init crash_notes_memory_init(void) { /* Allocate memory for saving cpu registers. */ diff --git a/kernel/kprobes.c b/kernel/kprobes.c index 610c837ad9e0..17ec4afb0994 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -38,6 +38,7 @@ #include <linux/module.h> #include <linux/moduleloader.h> #include <linux/kallsyms.h> +#include <linux/freezer.h> #include <asm-generic/sections.h> #include <asm/cacheflush.h> #include <asm/errno.h> @@ -83,9 +84,36 @@ struct kprobe_insn_page { kprobe_opcode_t *insns; /* Page of instruction slots */ char slot_used[INSNS_PER_PAGE]; int nused; + int ngarbage; }; static struct hlist_head kprobe_insn_pages; +static int kprobe_garbage_slots; +static int collect_garbage_slots(void); + +static int __kprobes check_safety(void) +{ + int ret = 0; +#if defined(CONFIG_PREEMPT) && defined(CONFIG_PM) + ret = freeze_processes(); + if (ret == 0) { + struct task_struct *p, *q; + do_each_thread(p, q) { + if (p != current && p->state == TASK_RUNNING && + p->pid != 0) { + printk("Check failed: %s is running\n",p->comm); + ret = -1; + goto loop_end; + } + } while_each_thread(p, q); + } +loop_end: + thaw_processes(); +#else + synchronize_sched(); +#endif + return ret; +} /** * get_insn_slot() - Find a slot on an executable page for an instruction. @@ -96,6 +124,7 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) struct kprobe_insn_page *kip; struct hlist_node *pos; + retry: hlist_for_each(pos, &kprobe_insn_pages) { kip = hlist_entry(pos, struct kprobe_insn_page, hlist); if (kip->nused < INSNS_PER_PAGE) { @@ -112,7 +141,11 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) } } - /* All out of space. Need to allocate a new page. Use slot 0.*/ + /* If there are any garbage slots, collect it and try again. */ + if (kprobe_garbage_slots && collect_garbage_slots() == 0) { + goto retry; + } + /* All out of space. Need to allocate a new page. Use slot 0. */ kip = kmalloc(sizeof(struct kprobe_insn_page), GFP_KERNEL); if (!kip) { return NULL; @@ -133,10 +166,62 @@ kprobe_opcode_t __kprobes *get_insn_slot(void) memset(kip->slot_used, 0, INSNS_PER_PAGE); kip->slot_used[0] = 1; kip->nused = 1; + kip->ngarbage = 0; return kip->insns; } -void __kprobes free_insn_slot(kprobe_opcode_t *slot) +/* Return 1 if all garbages are collected, otherwise 0. */ +static int __kprobes collect_one_slot(struct kprobe_insn_page *kip, int idx) +{ + kip->slot_used[idx] = 0; + kip->nused--; + if (kip->nused == 0) { + /* + * Page is no longer in use. Free it unless + * it's the last one. We keep the last one + * so as not to have to set it up again the + * next time somebody inserts a probe. + */ + hlist_del(&kip->hlist); + if (hlist_empty(&kprobe_insn_pages)) { + INIT_HLIST_NODE(&kip->hlist); + hlist_add_head(&kip->hlist, + &kprobe_insn_pages); + } else { + module_free(NULL, kip->insns); + kfree(kip); + } + return 1; + } + return 0; +} + +static int __kprobes collect_garbage_slots(void) +{ + struct kprobe_insn_page *kip; + struct hlist_node *pos, *next; + + /* Ensure no-one is preepmted on the garbages */ + if (check_safety() != 0) + return -EAGAIN; + + hlist_for_each_safe(pos, next, &kprobe_insn_pages) { + int i; + kip = hlist_entry(pos, struct kprobe_insn_page, hlist); + if (kip->ngarbage == 0) + continue; + kip->ngarbage = 0; /* we will collect all garbages */ + for (i = 0; i < INSNS_PER_PAGE; i++) { + if (kip->slot_used[i] == -1 && + collect_one_slot(kip, i)) + break; + } + } + kprobe_garbage_slots = 0; + return 0; +} + +void __kprobes free_insn_slot(kprobe_opcode_t * slot, int dirty) { struct kprobe_insn_page *kip; struct hlist_node *pos; @@ -146,28 +231,18 @@ void __kprobes free_insn_slot(kprobe_opcode_t *slot) if (kip->insns <= slot && slot < kip->insns + (INSNS_PER_PAGE * MAX_INSN_SIZE)) { int i = (slot - kip->insns) / MAX_INSN_SIZE; - kip->slot_used[i] = 0; - kip->nused--; - if (kip->nused == 0) { - /* - * Page is no longer in use. Free it unless - * it's the last one. We keep the last one - * so as not to have to set it up again the - * next time somebody inserts a probe. - */ - hlist_del(&kip->hlist); - if (hlist_empty(&kprobe_insn_pages)) { - INIT_HLIST_NODE(&kip->hlist); - hlist_add_head(&kip->hlist, - &kprobe_insn_pages); - } else { - module_free(NULL, kip->insns); - kfree(kip); - } + if (dirty) { + kip->slot_used[i] = -1; + kip->ngarbage++; + } else { + collect_one_slot(kip, i); } - return; + break; } } + if (dirty && (++kprobe_garbage_slots > INSNS_PER_PAGE)) { + collect_garbage_slots(); + } } #endif diff --git a/kernel/lockdep.c b/kernel/lockdep.c index c9fefdb1a7db..b02032476dc2 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -140,13 +140,6 @@ void lockdep_on(void) EXPORT_SYMBOL(lockdep_on); -int lockdep_internal(void) -{ - return current->lockdep_recursion != 0; -} - -EXPORT_SYMBOL(lockdep_internal); - /* * Debugging switches: */ @@ -228,17 +221,15 @@ 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; nr_stack_trace_entries += trace->nr_entries; - if (DEBUG_LOCKS_WARN_ON(nr_stack_trace_entries > MAX_STACK_TRACE_ENTRIES)) + if (DEBUG_LOCKS_WARN_ON(nr_stack_trace_entries > MAX_STACK_TRACE_ENTRIES)) { + __raw_spin_unlock(&hash_lock); return 0; + } if (nr_stack_trace_entries == MAX_STACK_TRACE_ENTRIES) { __raw_spin_unlock(&hash_lock); @@ -357,7 +348,7 @@ get_usage_chars(struct lock_class *class, char *c1, char *c2, char *c3, char *c4 static void print_lock_name(struct lock_class *class) { - char str[128], c1, c2, c3, c4; + char str[KSYM_NAME_LEN + 1], c1, c2, c3, c4; const char *name; get_usage_chars(class, &c1, &c2, &c3, &c4); @@ -379,7 +370,7 @@ static void print_lock_name(struct lock_class *class) static void print_lockdep_cache(struct lockdep_map *lock) { const char *name; - char str[128]; + char str[KSYM_NAME_LEN + 1]; name = lock->name; if (!name) @@ -449,7 +440,9 @@ static void print_lock_dependencies(struct lock_class *class, int depth) print_lock_class_header(class, depth); list_for_each_entry(entry, &class->locks_after, entry) { - DEBUG_LOCKS_WARN_ON(!entry->class); + if (DEBUG_LOCKS_WARN_ON(!entry->class)) + return; + print_lock_dependencies(entry->class, depth + 1); printk("%*s ... acquired at:\n",depth,""); @@ -474,7 +467,8 @@ static int add_lock_to_list(struct lock_class *class, struct lock_class *this, return 0; entry->class = this; - save_trace(&entry->trace); + if (!save_trace(&entry->trace)) + return 0; /* * Since we never remove from the dependency list, the list can @@ -562,8 +556,12 @@ static noinline int print_circular_bug_tail(void) if (debug_locks_silent) return 0; + /* hash_lock unlocked by the header */ + __raw_spin_lock(&hash_lock); this.class = check_source->class; - save_trace(&this.trace); + if (!save_trace(&this.trace)) + return 0; + __raw_spin_unlock(&hash_lock); print_circular_bug_entry(&this, 0); printk("\nother info that might help us debug this:\n\n"); @@ -966,14 +964,11 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, &prev->class->locks_after, next->acquire_ip); if (!ret) return 0; - /* - * Return value of 2 signals 'dependency already added', - * in that case we dont have to add the backlink either. - */ - if (ret == 2) - return 2; + ret = add_lock_to_list(next->class, prev->class, &next->class->locks_before, next->acquire_ip); + if (!ret) + return 0; /* * Debugging printouts: @@ -1025,7 +1020,8 @@ check_prevs_add(struct task_struct *curr, struct held_lock *next) * added: */ if (hlock->read != 2) { - check_prev_add(curr, hlock, next); + if (!check_prev_add(curr, hlock, next)) + return 0; /* * Stop after the first non-trylock entry, * as non-trylock entries have added their @@ -1182,6 +1178,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) struct lockdep_subclass_key *key; struct list_head *hash_head; struct lock_class *class; + unsigned long flags; class = look_up_lock_class(lock, subclass); if (likely(class)) @@ -1203,6 +1200,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) key = lock->key->subkeys + subclass; hash_head = classhashentry(key); + raw_local_irq_save(flags); __raw_spin_lock(&hash_lock); /* * We have to do the hash-walk again, to avoid races @@ -1217,6 +1215,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) */ if (nr_lock_classes >= MAX_LOCKDEP_KEYS) { __raw_spin_unlock(&hash_lock); + raw_local_irq_restore(flags); debug_locks_off(); printk("BUG: MAX_LOCKDEP_KEYS too low!\n"); printk("turning off the locking correctness validator.\n"); @@ -1239,15 +1238,18 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force) if (verbose(class)) { __raw_spin_unlock(&hash_lock); + raw_local_irq_restore(flags); printk("\nnew class %p: %s", class->key, class->name); if (class->name_version > 1) printk("#%d", class->name_version); printk("\n"); dump_stack(); + raw_local_irq_save(flags); __raw_spin_lock(&hash_lock); } out_unlock_set: __raw_spin_unlock(&hash_lock); + raw_local_irq_restore(flags); if (!subclass || force) lock->class_cache = class; @@ -1728,6 +1730,7 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, debug_atomic_dec(&nr_unused_locks); break; default: + __raw_spin_unlock(&hash_lock); debug_locks_off(); WARN_ON(1); return 0; @@ -2645,6 +2648,7 @@ void debug_check_no_locks_freed(const void *mem_from, unsigned long mem_len) } local_irq_restore(flags); } +EXPORT_SYMBOL_GPL(debug_check_no_locks_freed); static void print_held_locks_bug(struct task_struct *curr) { diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index eab043c83bb2..8ce09bc4613d 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -20,7 +20,7 @@ #define MAX_LOCKDEP_KEYS_BITS 11 #define MAX_LOCKDEP_KEYS (1UL << MAX_LOCKDEP_KEYS_BITS) -#define MAX_LOCKDEP_CHAINS_BITS 13 +#define MAX_LOCKDEP_CHAINS_BITS 14 #define MAX_LOCKDEP_CHAINS (1UL << MAX_LOCKDEP_CHAINS_BITS) /* diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index f6e72eaab3fa..b554b40a4aa6 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -113,7 +113,7 @@ static int l_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations lockdep_ops = { +static const struct seq_operations lockdep_ops = { .start = l_start, .next = l_next, .stop = l_stop, @@ -135,7 +135,7 @@ static int lockdep_open(struct inode *inode, struct file *file) return res; } -static struct file_operations proc_lockdep_operations = { +static const struct file_operations proc_lockdep_operations = { .open = lockdep_open, .read = seq_read, .llseek = seq_lseek, @@ -319,7 +319,7 @@ static int lockdep_stats_open(struct inode *inode, struct file *file) return single_open(file, lockdep_stats_show, NULL); } -static struct file_operations proc_lockdep_stats_operations = { +static const struct file_operations proc_lockdep_stats_operations = { .open = lockdep_stats_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/module.c b/kernel/module.c index e2d09d604ca0..d9eae45d0145 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -2209,7 +2209,7 @@ static int m_show(struct seq_file *m, void *p) Where refcount is a number or -, and deps is a comma-separated list of depends or -. */ -struct seq_operations modules_op = { +const struct seq_operations modules_op = { .start = m_start, .next = m_next, .stop = m_stop, diff --git a/kernel/mutex-debug.c b/kernel/mutex-debug.c index 18651641a7b5..841539d72c55 100644 --- a/kernel/mutex-debug.c +++ b/kernel/mutex-debug.c @@ -77,6 +77,9 @@ void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter, void debug_mutex_unlock(struct mutex *lock) { + if (unlikely(!debug_locks)) + return; + DEBUG_LOCKS_WARN_ON(lock->owner != current_thread_info()); DEBUG_LOCKS_WARN_ON(lock->magic != lock); DEBUG_LOCKS_WARN_ON(!lock->wait_list.prev && !lock->wait_list.next); diff --git a/kernel/pid.c b/kernel/pid.c index b914392085f9..a48879b0b921 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -31,7 +31,7 @@ #define pid_hashfn(nr) hash_long((unsigned long)nr, pidhash_shift) static struct hlist_head *pid_hash; static int pidhash_shift; -static kmem_cache_t *pid_cachep; +static struct kmem_cache *pid_cachep; int pid_max = PID_MAX_DEFAULT; diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 9cbb5d1be06f..5fe87de10ff0 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -70,7 +70,7 @@ /* * Lets keep our timers in a slab cache :-) */ -static kmem_cache_t *posix_timers_cache; +static struct kmem_cache *posix_timers_cache; static struct idr posix_timers_id; static DEFINE_SPINLOCK(idr_lock); diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 825068ca3479..710ed084e7c5 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -78,7 +78,7 @@ config PM_SYSFS_DEPRECATED config SOFTWARE_SUSPEND bool "Software Suspend" - depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP) && !X86_PAE) || ((FRV || PPC32) && !SMP)) + depends on PM && SWAP && ((X86 && (!SMP || SUSPEND_SMP)) || ((FRV || PPC32) && !SMP)) ---help--- Enable the possibility of suspending the machine. It doesn't need ACPI or APM. diff --git a/kernel/power/disk.c b/kernel/power/disk.c index b1fb7866b0b3..0b00f56c2ad0 100644 --- a/kernel/power/disk.c +++ b/kernel/power/disk.c @@ -20,6 +20,7 @@ #include <linux/pm.h> #include <linux/console.h> #include <linux/cpu.h> +#include <linux/freezer.h> #include "power.h" @@ -27,6 +28,23 @@ static int noresume = 0; char resume_file[256] = CONFIG_PM_STD_PARTITION; dev_t swsusp_resume_device; +sector_t swsusp_resume_block; + +/** + * platform_prepare - prepare the machine for hibernation using the + * platform driver if so configured and return an error code if it fails + */ + +static inline int platform_prepare(void) +{ + int error = 0; + + if (pm_disk_mode == PM_DISK_PLATFORM) { + if (pm_ops && pm_ops->prepare) + error = pm_ops->prepare(PM_SUSPEND_DISK); + } + return error; +} /** * power_down - Shut machine down for hibernate. @@ -40,12 +58,10 @@ dev_t swsusp_resume_device; static void power_down(suspend_disk_method_t mode) { - int error = 0; - switch(mode) { case PM_DISK_PLATFORM: kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); - error = pm_ops->enter(PM_SUSPEND_DISK); + pm_ops->enter(PM_SUSPEND_DISK); break; case PM_DISK_SHUTDOWN: kernel_power_off(); @@ -90,12 +106,18 @@ static int prepare_processes(void) goto thaw; } + error = platform_prepare(); + if (error) + goto thaw; + /* Free memory before shutting down devices. */ if (!(error = swsusp_shrink_memory())) return 0; -thaw: + + platform_finish(); + thaw: thaw_processes(); -enable_cpus: + enable_cpus: enable_nonboot_cpus(); pm_restore_console(); return error; @@ -127,7 +149,7 @@ int pm_suspend_disk(void) return error; if (pm_disk_mode == PM_DISK_TESTPROC) - goto Thaw; + return 0; suspend_console(); error = device_suspend(PMSG_FREEZE); @@ -189,10 +211,10 @@ static int software_resume(void) { int error; - down(&pm_sem); + mutex_lock(&pm_mutex); if (!swsusp_resume_device) { if (!strlen(resume_file)) { - up(&pm_sem); + mutex_unlock(&pm_mutex); return -ENOENT; } swsusp_resume_device = name_to_dev_t(resume_file); @@ -207,7 +229,7 @@ static int software_resume(void) * FIXME: If noresume is specified, we need to find the partition * and reset it back to normal swap space. */ - up(&pm_sem); + mutex_unlock(&pm_mutex); return 0; } @@ -251,7 +273,7 @@ static int software_resume(void) unprepare_processes(); Done: /* For success case, the suspend path will release the lock */ - up(&pm_sem); + mutex_unlock(&pm_mutex); pr_debug("PM: Resume from disk failed.\n"); return 0; } @@ -312,7 +334,7 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n) p = memchr(buf, '\n', n); len = p ? p - buf : n; - down(&pm_sem); + mutex_lock(&pm_mutex); for (i = PM_DISK_FIRMWARE; i < PM_DISK_MAX; i++) { if (!strncmp(buf, pm_disk_modes[i], len)) { mode = i; @@ -336,7 +358,7 @@ static ssize_t disk_store(struct subsystem * s, const char * buf, size_t n) pr_debug("PM: suspend-to-disk mode set to '%s'\n", pm_disk_modes[mode]); - up(&pm_sem); + mutex_unlock(&pm_mutex); return error ? error : n; } @@ -361,14 +383,14 @@ static ssize_t resume_store(struct subsystem *subsys, const char *buf, size_t n) if (maj != MAJOR(res) || min != MINOR(res)) goto out; - down(&pm_sem); + mutex_lock(&pm_mutex); swsusp_resume_device = res; - up(&pm_sem); + mutex_unlock(&pm_mutex); printk("Attempting manual resume\n"); noresume = 0; software_resume(); ret = n; -out: + out: return ret; } @@ -423,6 +445,19 @@ static int __init resume_setup(char *str) return 1; } +static int __init resume_offset_setup(char *str) +{ + unsigned long long offset; + + if (noresume) + return 1; + + if (sscanf(str, "%llu", &offset) == 1) + swsusp_resume_block = offset; + + return 1; +} + static int __init noresume_setup(char *str) { noresume = 1; @@ -430,4 +465,5 @@ static int __init noresume_setup(char *str) } __setup("noresume", noresume_setup); +__setup("resume_offset=", resume_offset_setup); __setup("resume=", resume_setup); diff --git a/kernel/power/main.c b/kernel/power/main.c index 873228c71dab..500eb87f643d 100644 --- a/kernel/power/main.c +++ b/kernel/power/main.c @@ -8,6 +8,7 @@ * */ +#include <linux/module.h> #include <linux/suspend.h> #include <linux/kobject.h> #include <linux/string.h> @@ -18,13 +19,14 @@ #include <linux/console.h> #include <linux/cpu.h> #include <linux/resume-trace.h> +#include <linux/freezer.h> #include "power.h" /*This is just an arbitrary number */ #define FREE_PAGE_NUMBER (100) -DECLARE_MUTEX(pm_sem); +DEFINE_MUTEX(pm_mutex); struct pm_ops *pm_ops; suspend_disk_method_t pm_disk_mode = PM_DISK_SHUTDOWN; @@ -36,9 +38,9 @@ suspend_disk_method_t pm_disk_mode = PM_DISK_SHUTDOWN; void pm_set_ops(struct pm_ops * ops) { - down(&pm_sem); + mutex_lock(&pm_mutex); pm_ops = ops; - up(&pm_sem); + mutex_unlock(&pm_mutex); } @@ -182,7 +184,7 @@ static int enter_state(suspend_state_t state) if (!valid_state(state)) return -ENODEV; - if (down_trylock(&pm_sem)) + if (!mutex_trylock(&pm_mutex)) return -EBUSY; if (state == PM_SUSPEND_DISK) { @@ -200,7 +202,7 @@ static int enter_state(suspend_state_t state) pr_debug("PM: Finishing wakeup.\n"); suspend_finish(state); Unlock: - up(&pm_sem); + mutex_unlock(&pm_mutex); return error; } @@ -229,7 +231,7 @@ int pm_suspend(suspend_state_t state) return -EINVAL; } - +EXPORT_SYMBOL(pm_suspend); decl_subsys(power,NULL,NULL); diff --git a/kernel/power/power.h b/kernel/power/power.h index bfe999f7b272..eb461b816bf4 100644 --- a/kernel/power/power.h +++ b/kernel/power/power.h @@ -22,7 +22,9 @@ static inline int pm_suspend_disk(void) return -EPERM; } #endif -extern struct semaphore pm_sem; + +extern struct mutex pm_mutex; + #define power_attr(_name) \ static struct subsys_attribute _name##_attr = { \ .attr = { \ @@ -42,6 +44,7 @@ extern const void __nosave_begin, __nosave_end; extern unsigned long image_size; extern int in_suspend; extern dev_t swsusp_resume_device; +extern sector_t swsusp_resume_block; extern asmlinkage int swsusp_arch_suspend(void); extern asmlinkage int swsusp_arch_resume(void); @@ -102,8 +105,18 @@ struct snapshot_handle { 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 void snapshot_write_finalize(struct snapshot_handle *handle); extern int snapshot_image_loaded(struct snapshot_handle *handle); -extern void snapshot_free_unused_memory(struct snapshot_handle *handle); + +/* + * This structure is used to pass the values needed for the identification + * of the resume swap area from a user space to the kernel via the + * SNAPSHOT_SET_SWAP_AREA ioctl + */ +struct resume_swap_area { + loff_t offset; + u_int32_t dev; +} __attribute__((packed)); #define SNAPSHOT_IOC_MAGIC '3' #define SNAPSHOT_FREEZE _IO(SNAPSHOT_IOC_MAGIC, 1) @@ -117,7 +130,14 @@ extern void snapshot_free_unused_memory(struct snapshot_handle *handle); #define SNAPSHOT_FREE_SWAP_PAGES _IO(SNAPSHOT_IOC_MAGIC, 9) #define SNAPSHOT_SET_SWAP_FILE _IOW(SNAPSHOT_IOC_MAGIC, 10, unsigned int) #define SNAPSHOT_S2RAM _IO(SNAPSHOT_IOC_MAGIC, 11) -#define SNAPSHOT_IOC_MAXNR 11 +#define SNAPSHOT_PMOPS _IOW(SNAPSHOT_IOC_MAGIC, 12, unsigned int) +#define SNAPSHOT_SET_SWAP_AREA _IOW(SNAPSHOT_IOC_MAGIC, 13, \ + struct resume_swap_area) +#define SNAPSHOT_IOC_MAXNR 13 + +#define PMOPS_PREPARE 1 +#define PMOPS_ENTER 2 +#define PMOPS_FINISH 3 /** * The bitmap is used for tracing allocated swap pages @@ -141,7 +161,7 @@ struct bitmap_page { extern void free_bitmap(struct bitmap_page *bitmap); extern struct bitmap_page *alloc_bitmap(unsigned int nr_bits); -extern unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap); +extern sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap); extern void free_all_swap_pages(int swap, struct bitmap_page *bitmap); extern int swsusp_check(void); @@ -153,3 +173,7 @@ extern int swsusp_read(void); extern int swsusp_write(void); extern void swsusp_close(void); extern int suspend_enter(suspend_state_t state); + +struct timeval; +extern void swsusp_show_speed(struct timeval *, struct timeval *, + unsigned int, char *); diff --git a/kernel/power/process.c b/kernel/power/process.c index 72e72d2c61e6..99eeb119b06d 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -13,12 +13,15 @@ #include <linux/suspend.h> #include <linux/module.h> #include <linux/syscalls.h> +#include <linux/freezer.h> /* * Timeout for stopping processes */ #define TIMEOUT (20 * HZ) +#define FREEZER_KERNEL_THREADS 0 +#define FREEZER_USER_SPACE 1 static inline int freezeable(struct task_struct * p) { @@ -39,7 +42,6 @@ void refrigerator(void) long save; save = current->state; pr_debug("%s entered refrigerator\n", current->comm); - printk("="); frozen_process(current); spin_lock_irq(¤t->sighand->siglock); @@ -79,96 +81,136 @@ static void cancel_freezing(struct task_struct *p) } } -/* 0 = success, else # of processes that we failed to stop */ -int freeze_processes(void) +static inline int is_user_space(struct task_struct *p) +{ + return p->mm && !(p->flags & PF_BORROWED_MM); +} + +static unsigned int try_to_freeze_tasks(int freeze_user_space) { - int todo, nr_user, user_frozen; - unsigned long start_time; struct task_struct *g, *p; + unsigned long end_time; + unsigned int todo; - printk( "Stopping tasks: " ); - start_time = jiffies; - user_frozen = 0; + end_time = jiffies + TIMEOUT; do { - nr_user = todo = 0; + todo = 0; read_lock(&tasklist_lock); do_each_thread(g, p) { if (!freezeable(p)) continue; + if (frozen(p)) continue; - if (p->state == TASK_TRACED && frozen(p->parent)) { + + if (p->state == TASK_TRACED && + (frozen(p->parent) || + p->parent->state == TASK_STOPPED)) { cancel_freezing(p); continue; } - if (p->mm && !(p->flags & PF_BORROWED_MM)) { - /* The task is a user-space one. - * Freeze it unless there's a vfork completion - * pending + if (is_user_space(p)) { + if (!freeze_user_space) + continue; + + /* Freeze the task unless there is a vfork + * completion pending */ if (!p->vfork_done) freeze_process(p); - nr_user++; } else { - /* Freeze only if the user space is frozen */ - if (user_frozen) - freeze_process(p); - todo++; + if (freeze_user_space) + continue; + + freeze_process(p); } + todo++; } while_each_thread(g, p); read_unlock(&tasklist_lock); - todo += nr_user; - if (!user_frozen && !nr_user) { - sys_sync(); - start_time = jiffies; - } - user_frozen = !nr_user; yield(); /* Yield is okay here */ - if (todo && time_after(jiffies, start_time + TIMEOUT)) + if (todo && time_after(jiffies, end_time)) break; - } while(todo); + } while (todo); - /* This does not unfreeze processes that are already frozen - * (we have slightly ugly calling convention in that respect, - * and caller must call thaw_processes() if something fails), - * but it cleans up leftover PF_FREEZE requests. - */ if (todo) { - printk( "\n" ); - printk(KERN_ERR " stopping tasks timed out " - "after %d seconds (%d tasks remaining):\n", - TIMEOUT / HZ, todo); + /* This does not unfreeze processes that are already frozen + * (we have slightly ugly calling convention in that respect, + * and caller must call thaw_processes() if something fails), + * but it cleans up leftover PF_FREEZE requests. + */ + printk("\n"); + printk(KERN_ERR "Stopping %s timed out after %d seconds " + "(%d tasks refusing to freeze):\n", + freeze_user_space ? "user space processes" : + "kernel threads", + TIMEOUT / HZ, todo); read_lock(&tasklist_lock); do_each_thread(g, p) { + if (is_user_space(p) == !freeze_user_space) + continue; + if (freezeable(p) && !frozen(p)) - printk(KERN_ERR " %s\n", p->comm); + printk(KERN_ERR " %s\n", p->comm); + cancel_freezing(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); - return todo; } - printk( "|\n" ); + return todo; +} + +/** + * freeze_processes - tell processes to enter the refrigerator + * + * Returns 0 on success, or the number of processes that didn't freeze, + * although they were told to. + */ +int freeze_processes(void) +{ + unsigned int nr_unfrozen; + + printk("Stopping tasks ... "); + nr_unfrozen = try_to_freeze_tasks(FREEZER_USER_SPACE); + if (nr_unfrozen) + return nr_unfrozen; + + sys_sync(); + nr_unfrozen = try_to_freeze_tasks(FREEZER_KERNEL_THREADS); + if (nr_unfrozen) + return nr_unfrozen; + + printk("done.\n"); BUG_ON(in_atomic()); return 0; } -void thaw_processes(void) +static void thaw_tasks(int thaw_user_space) { struct task_struct *g, *p; - printk( "Restarting tasks..." ); read_lock(&tasklist_lock); do_each_thread(g, p) { if (!freezeable(p)) continue; + + if (is_user_space(p) == !thaw_user_space) + continue; + if (!thaw_process(p)) - printk(KERN_INFO " Strange, %s not stopped\n", p->comm ); + printk(KERN_WARNING " Strange, %s not stopped\n", + p->comm ); } while_each_thread(g, p); - read_unlock(&tasklist_lock); +} + +void thaw_processes(void) +{ + printk("Restarting tasks ... "); + thaw_tasks(FREEZER_KERNEL_THREADS); + thaw_tasks(FREEZER_USER_SPACE); schedule(); - printk( " done\n" ); + printk("done.\n"); } EXPORT_SYMBOL(refrigerator); diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c index 99f9b7d177d6..c024606221c4 100644 --- a/kernel/power/snapshot.c +++ b/kernel/power/snapshot.c @@ -1,15 +1,15 @@ /* * linux/kernel/power/snapshot.c * - * This file provide system snapshot/restore functionality. + * This file provides system snapshot/restore functionality for swsusp. * * Copyright (C) 1998-2005 Pavel Machek <pavel@suse.cz> + * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl> * - * This file is released under the GPLv2, and is based on swsusp.c. + * This file is released under the GPLv2. * */ - #include <linux/version.h> #include <linux/module.h> #include <linux/mm.h> @@ -34,137 +34,24 @@ #include "power.h" -/* List of PBEs used for creating and restoring the suspend image */ +/* List of PBEs needed for restoring the pages that were allocated before + * the suspend and included in the suspend image, but have also been + * allocated by the "resume" kernel, so their contents cannot be written + * directly to their "original" page frames. + */ struct pbe *restore_pblist; -static unsigned int nr_copy_pages; -static unsigned int nr_meta_pages; +/* Pointer to an auxiliary buffer (1 page) */ static void *buffer; -#ifdef CONFIG_HIGHMEM -unsigned int count_highmem_pages(void) -{ - struct zone *zone; - unsigned long zone_pfn; - unsigned int n = 0; - - for_each_zone (zone) - if (is_highmem(zone)) { - mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; zone_pfn++) { - struct page *page; - unsigned long pfn = zone_pfn + zone->zone_start_pfn; - if (!pfn_valid(pfn)) - continue; - page = pfn_to_page(pfn); - if (PageReserved(page)) - continue; - if (PageNosaveFree(page)) - continue; - n++; - } - } - return n; -} - -struct highmem_page { - char *data; - struct page *page; - struct highmem_page *next; -}; - -static struct highmem_page *highmem_copy; - -static int save_highmem_zone(struct zone *zone) -{ - unsigned long zone_pfn; - mark_free_pages(zone); - for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn) { - struct page *page; - struct highmem_page *save; - void *kaddr; - unsigned long pfn = zone_pfn + zone->zone_start_pfn; - - if (!(pfn%10000)) - printk("."); - if (!pfn_valid(pfn)) - continue; - page = pfn_to_page(pfn); - /* - * This condition results from rvmalloc() sans vmalloc_32() - * and architectural memory reservations. This should be - * corrected eventually when the cases giving rise to this - * are better understood. - */ - if (PageReserved(page)) - continue; - BUG_ON(PageNosave(page)); - if (PageNosaveFree(page)) - continue; - save = kmalloc(sizeof(struct highmem_page), GFP_ATOMIC); - if (!save) - return -ENOMEM; - save->next = highmem_copy; - save->page = page; - save->data = (void *) get_zeroed_page(GFP_ATOMIC); - if (!save->data) { - kfree(save); - return -ENOMEM; - } - kaddr = kmap_atomic(page, KM_USER0); - memcpy(save->data, kaddr, PAGE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - highmem_copy = save; - } - return 0; -} - -int save_highmem(void) -{ - struct zone *zone; - int res = 0; - - pr_debug("swsusp: Saving Highmem"); - drain_local_pages(); - for_each_zone (zone) { - if (is_highmem(zone)) - res = save_highmem_zone(zone); - if (res) - return res; - } - printk("\n"); - return 0; -} - -int restore_highmem(void) -{ - printk("swsusp: Restoring Highmem\n"); - while (highmem_copy) { - struct highmem_page *save = highmem_copy; - void *kaddr; - highmem_copy = save->next; - - kaddr = kmap_atomic(save->page, KM_USER0); - memcpy(kaddr, save->data, PAGE_SIZE); - kunmap_atomic(kaddr, KM_USER0); - free_page((long) save->data); - kfree(save); - } - return 0; -} -#else -static inline unsigned int count_highmem_pages(void) {return 0;} -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. + * used before suspend. The unsafe pages have PageNosaveFree set + * and we count them using unsafe_pages. * - * The unsafe pages are marked with the PG_nosave_free flag - * and we count them using unsafe_pages + * Each allocated image page is marked as PageNosave and PageNosaveFree + * so that swsusp_free() can release it. */ #define PG_ANY 0 @@ -174,7 +61,7 @@ static inline int restore_highmem(void) {return 0;} static unsigned int allocated_unsafe_pages; -static void *alloc_image_page(gfp_t gfp_mask, int safe_needed) +static void *get_image_page(gfp_t gfp_mask, int safe_needed) { void *res; @@ -195,20 +82,39 @@ 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, PG_SAFE); + return (unsigned long)get_image_page(gfp_mask, PG_SAFE); +} + +static struct page *alloc_image_page(gfp_t gfp_mask) +{ + struct page *page; + + page = alloc_page(gfp_mask); + if (page) { + SetPageNosave(page); + SetPageNosaveFree(page); + } + return page; } /** * free_image_page - free page represented by @addr, allocated with - * alloc_image_page (page flags set by it must be cleared) + * get_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)); + struct page *page; + + BUG_ON(!virt_addr_valid(addr)); + + page = virt_to_page(addr); + + ClearPageNosave(page); if (clear_nosave_free) - ClearPageNosaveFree(virt_to_page(addr)); - free_page((unsigned long)addr); + ClearPageNosaveFree(page); + + __free_page(page); } /* struct linked_page is used to build chains of pages */ @@ -269,7 +175,7 @@ static void *chain_alloc(struct chain_allocator *ca, unsigned int size) if (LINKED_PAGE_DATA_SIZE - ca->used_space < size) { struct linked_page *lp; - lp = alloc_image_page(ca->gfp_mask, ca->safe_needed); + lp = get_image_page(ca->gfp_mask, ca->safe_needed); if (!lp) return NULL; @@ -446,8 +352,8 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) /* Compute the number of zones */ nr = 0; - for_each_zone (zone) - if (populated_zone(zone) && !is_highmem(zone)) + for_each_zone(zone) + if (populated_zone(zone)) nr++; /* Allocate the list of zones bitmap objects */ @@ -459,10 +365,10 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) } /* Initialize the zone bitmap objects */ - for_each_zone (zone) { + for_each_zone(zone) { unsigned long pfn; - if (!populated_zone(zone) || is_highmem(zone)) + if (!populated_zone(zone)) continue; zone_bm->start_pfn = zone->zone_start_pfn; @@ -481,7 +387,7 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) while (bb) { unsigned long *ptr; - ptr = alloc_image_page(gfp_mask, safe_needed); + ptr = get_image_page(gfp_mask, safe_needed); bb->data = ptr; if (!ptr) goto Free; @@ -505,7 +411,7 @@ memory_bm_create(struct memory_bitmap *bm, gfp_t gfp_mask, int safe_needed) memory_bm_position_reset(bm); return 0; -Free: + Free: bm->p_list = ca.chain; memory_bm_free(bm, PG_UNSAFE_CLEAR); return -ENOMEM; @@ -651,7 +557,7 @@ static unsigned long memory_bm_next_pfn(struct memory_bitmap *bm) memory_bm_position_reset(bm); return BM_END_OF_MAP; -Return_pfn: + Return_pfn: bm->cur.chunk = chunk; bm->cur.bit = bit; return bb->start_pfn + chunk * BM_BITS_PER_CHUNK + bit; @@ -669,10 +575,82 @@ unsigned int snapshot_additional_pages(struct zone *zone) res = DIV_ROUND_UP(zone->spanned_pages, BM_BITS_PER_BLOCK); res += DIV_ROUND_UP(res * sizeof(struct bm_block), PAGE_SIZE); - return res; + return 2 * res; +} + +#ifdef CONFIG_HIGHMEM +/** + * count_free_highmem_pages - compute the total number of free highmem + * pages, system-wide. + */ + +static unsigned int count_free_highmem_pages(void) +{ + struct zone *zone; + unsigned int cnt = 0; + + for_each_zone(zone) + if (populated_zone(zone) && is_highmem(zone)) + cnt += zone->free_pages; + + return cnt; +} + +/** + * saveable_highmem_page - Determine whether a highmem page should be + * included in the suspend image. + * + * We should save the page if it isn't Nosave or NosaveFree, or Reserved, + * and it isn't a part of a free chunk of pages. + */ + +static struct page *saveable_highmem_page(unsigned long pfn) +{ + struct page *page; + + if (!pfn_valid(pfn)) + return NULL; + + page = pfn_to_page(pfn); + + BUG_ON(!PageHighMem(page)); + + if (PageNosave(page) || PageReserved(page) || PageNosaveFree(page)) + return NULL; + + return page; } /** + * count_highmem_pages - compute the total number of saveable highmem + * pages. + */ + +unsigned int count_highmem_pages(void) +{ + struct zone *zone; + unsigned int n = 0; + + for_each_zone(zone) { + unsigned long pfn, 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++) + if (saveable_highmem_page(pfn)) + n++; + } + return n; +} +#else +static inline void *saveable_highmem_page(unsigned long pfn) { return NULL; } +static inline unsigned int count_highmem_pages(void) { return 0; } +#endif /* CONFIG_HIGHMEM */ + +/** * pfn_is_nosave - check if given pfn is in the 'nosave' section */ @@ -684,12 +662,12 @@ static inline int pfn_is_nosave(unsigned long pfn) } /** - * saveable - Determine whether a page should be cloned or not. - * @pfn: The page + * saveable - Determine whether a non-highmem page should be included in + * the suspend image. * - * 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. + * We should save the 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 struct page *saveable_page(unsigned long pfn) @@ -701,76 +679,130 @@ static struct page *saveable_page(unsigned long pfn) page = pfn_to_page(pfn); - if (PageNosave(page)) + BUG_ON(PageHighMem(page)); + + if (PageNosave(page) || PageNosaveFree(page)) return NULL; + if (PageReserved(page) && pfn_is_nosave(pfn)) return NULL; - if (PageNosaveFree(page)) - return NULL; return page; } +/** + * count_data_pages - compute the total number of saveable non-highmem + * pages. + */ + unsigned int count_data_pages(void) { struct zone *zone; unsigned long pfn, max_zone_pfn; unsigned int n = 0; - for_each_zone (zone) { + for_each_zone(zone) { 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++) - n += !!saveable_page(pfn); + if(saveable_page(pfn)) + n++; } return n; } -static inline void copy_data_page(long *dst, long *src) +/* This is needed, because copy_page and memcpy are not usable for copying + * task structs. + */ +static inline void do_copy_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++; } +#ifdef CONFIG_HIGHMEM +static inline struct page * +page_is_saveable(struct zone *zone, unsigned long pfn) +{ + return is_highmem(zone) ? + saveable_highmem_page(pfn) : saveable_page(pfn); +} + +static inline void +copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +{ + struct page *s_page, *d_page; + void *src, *dst; + + s_page = pfn_to_page(src_pfn); + d_page = pfn_to_page(dst_pfn); + if (PageHighMem(s_page)) { + src = kmap_atomic(s_page, KM_USER0); + dst = kmap_atomic(d_page, KM_USER1); + do_copy_page(dst, src); + kunmap_atomic(src, KM_USER0); + kunmap_atomic(dst, KM_USER1); + } else { + src = page_address(s_page); + if (PageHighMem(d_page)) { + /* Page pointed to by src may contain some kernel + * data modified by kmap_atomic() + */ + do_copy_page(buffer, src); + dst = kmap_atomic(pfn_to_page(dst_pfn), KM_USER0); + memcpy(dst, buffer, PAGE_SIZE); + kunmap_atomic(dst, KM_USER0); + } else { + dst = page_address(d_page); + do_copy_page(dst, src); + } + } +} +#else +#define page_is_saveable(zone, pfn) saveable_page(pfn) + +static inline void +copy_data_page(unsigned long dst_pfn, unsigned long src_pfn) +{ + do_copy_page(page_address(pfn_to_page(dst_pfn)), + page_address(pfn_to_page(src_pfn))); +} +#endif /* CONFIG_HIGHMEM */ + static void copy_data_pages(struct memory_bitmap *copy_bm, struct memory_bitmap *orig_bm) { struct zone *zone; unsigned long pfn; - for_each_zone (zone) { + 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++) - if (saveable_page(pfn)) + if (page_is_saveable(zone, pfn)) memory_bm_set_bit(orig_bm, pfn); } 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); - } + if (likely(pfn != BM_END_OF_MAP)) + copy_data_page(memory_bm_next_pfn(copy_bm), pfn); } while (pfn != BM_END_OF_MAP); } +/* Total number of image pages */ +static unsigned int nr_copy_pages; +/* Number of pages needed for saving the original pfns of the image pages */ +static unsigned int nr_meta_pages; + /** * swsusp_free - free pages allocated for the suspend. * @@ -792,7 +824,7 @@ void swsusp_free(void) if (PageNosave(page) && PageNosaveFree(page)) { ClearPageNosave(page); ClearPageNosaveFree(page); - free_page((long) page_address(page)); + __free_page(page); } } } @@ -802,34 +834,108 @@ void swsusp_free(void) buffer = NULL; } +#ifdef CONFIG_HIGHMEM +/** + * count_pages_for_highmem - compute the number of non-highmem pages + * that will be necessary for creating copies of highmem pages. + */ + +static unsigned int count_pages_for_highmem(unsigned int nr_highmem) +{ + unsigned int free_highmem = count_free_highmem_pages(); + + if (free_highmem >= nr_highmem) + nr_highmem = 0; + else + nr_highmem -= free_highmem; + + return nr_highmem; +} +#else +static unsigned int +count_pages_for_highmem(unsigned int nr_highmem) { return 0; } +#endif /* CONFIG_HIGHMEM */ /** - * enough_free_mem - Make sure we enough free memory to snapshot. - * - * Returns TRUE or FALSE after checking the number of available - * free pages. + * enough_free_mem - Make sure we have enough free memory for the + * snapshot image. */ -static int enough_free_mem(unsigned int nr_pages) +static int enough_free_mem(unsigned int nr_pages, unsigned int nr_highmem) { struct zone *zone; unsigned int free = 0, meta = 0; - for_each_zone (zone) - if (!is_highmem(zone)) { + for_each_zone(zone) { + meta += snapshot_additional_pages(zone); + 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 += count_pages_for_highmem(nr_highmem); + pr_debug("swsusp: Normal pages needed: %u + %u + %u, available pages: %u\n", nr_pages, PAGES_FOR_IO, meta, free); return free > nr_pages + PAGES_FOR_IO + meta; } +#ifdef CONFIG_HIGHMEM +/** + * get_highmem_buffer - if there are some highmem pages in the suspend + * image, we may need the buffer to copy them and/or load their data. + */ + +static inline int get_highmem_buffer(int safe_needed) +{ + buffer = get_image_page(GFP_ATOMIC | __GFP_COLD, safe_needed); + return buffer ? 0 : -ENOMEM; +} + +/** + * alloc_highmem_image_pages - allocate some highmem pages for the image. + * Try to allocate as many pages as needed, but if the number of free + * highmem pages is lesser than that, allocate them all. + */ + +static inline unsigned int +alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int nr_highmem) +{ + unsigned int to_alloc = count_free_highmem_pages(); + + if (to_alloc > nr_highmem) + to_alloc = nr_highmem; + + nr_highmem -= to_alloc; + while (to_alloc-- > 0) { + struct page *page; + + page = alloc_image_page(__GFP_HIGHMEM); + memory_bm_set_bit(bm, page_to_pfn(page)); + } + return nr_highmem; +} +#else +static inline int get_highmem_buffer(int safe_needed) { return 0; } + +static inline unsigned int +alloc_highmem_image_pages(struct memory_bitmap *bm, unsigned int n) { return 0; } +#endif /* CONFIG_HIGHMEM */ + +/** + * swsusp_alloc - allocate memory for the suspend image + * + * We first try to allocate as many highmem pages as there are + * saveable highmem pages in the system. If that fails, we allocate + * non-highmem pages for the copies of the remaining highmem ones. + * + * In this approach it is likely that the copies of highmem pages will + * also be located in the high memory, because of the way in which + * copy_data_pages() works. + */ + static int swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, - unsigned int nr_pages) + unsigned int nr_pages, unsigned int nr_highmem) { int error; @@ -841,46 +947,61 @@ swsusp_alloc(struct memory_bitmap *orig_bm, struct memory_bitmap *copy_bm, if (error) goto Free; + if (nr_highmem > 0) { + error = get_highmem_buffer(PG_ANY); + if (error) + goto Free; + + nr_pages += alloc_highmem_image_pages(copy_bm, nr_highmem); + } while (nr_pages-- > 0) { - struct page *page = alloc_page(GFP_ATOMIC | __GFP_COLD); + struct page *page = alloc_image_page(GFP_ATOMIC | __GFP_COLD); + if (!page) goto Free; - SetPageNosave(page); - SetPageNosaveFree(page); memory_bm_set_bit(copy_bm, page_to_pfn(page)); } return 0; -Free: + Free: swsusp_free(); return -ENOMEM; } -/* Memory bitmap used for marking saveable pages */ +/* Memory bitmap used for marking saveable pages (during suspend) or the + * suspend image pages (during resume) + */ static struct memory_bitmap orig_bm; -/* Memory bitmap used for marking allocated pages that will contain the copies - * of saveable pages +/* Memory bitmap used on suspend for marking allocated pages that will contain + * the copies of saveable pages. During resume it is initially used for + * marking the suspend image pages, but then its set bits are duplicated in + * @orig_bm and it is released. Next, on systems with high memory, it may be + * used for marking "safe" highmem pages, but it has to be reinitialized for + * this purpose. */ static struct memory_bitmap copy_bm; asmlinkage int swsusp_save(void) { - unsigned int nr_pages; + unsigned int nr_pages, nr_highmem; - pr_debug("swsusp: critical section: \n"); + printk("swsusp: critical section: \n"); drain_local_pages(); nr_pages = count_data_pages(); - printk("swsusp: Need to copy %u pages\n", nr_pages); + nr_highmem = count_highmem_pages(); + printk("swsusp: Need to copy %u pages\n", nr_pages + nr_highmem); - if (!enough_free_mem(nr_pages)) { + if (!enough_free_mem(nr_pages, nr_highmem)) { printk(KERN_ERR "swsusp: Not enough free memory\n"); return -ENOMEM; } - if (swsusp_alloc(&orig_bm, ©_bm, nr_pages)) + if (swsusp_alloc(&orig_bm, ©_bm, nr_pages, nr_highmem)) { + printk(KERN_ERR "swsusp: Memory allocation failed\n"); return -ENOMEM; + } /* During allocating of suspend pagedir, new cold pages may appear. * Kill them. @@ -894,10 +1015,12 @@ asmlinkage int swsusp_save(void) * touch swap space! Except we must write out our image of course. */ + nr_pages += nr_highmem; nr_copy_pages = nr_pages; - nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT; + nr_meta_pages = DIV_ROUND_UP(nr_pages * sizeof(long), PAGE_SIZE); printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages); + return 0; } @@ -960,7 +1083,7 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) if (!buffer) { /* This makes the buffer be freed by swsusp_free() */ - buffer = alloc_image_page(GFP_ATOMIC, PG_ANY); + buffer = get_image_page(GFP_ATOMIC, PG_ANY); if (!buffer) return -ENOMEM; } @@ -975,9 +1098,23 @@ int snapshot_read_next(struct snapshot_handle *handle, size_t count) memset(buffer, 0, PAGE_SIZE); pack_pfns(buffer, &orig_bm); } else { - unsigned long pfn = memory_bm_next_pfn(©_bm); + struct page *page; - handle->buffer = page_address(pfn_to_page(pfn)); + page = pfn_to_page(memory_bm_next_pfn(©_bm)); + if (PageHighMem(page)) { + /* Highmem pages are copied to the buffer, + * because we can't return with a kmapped + * highmem page (we may not be called again). + */ + void *kaddr; + + kaddr = kmap_atomic(page, KM_USER0); + memcpy(buffer, kaddr, PAGE_SIZE); + kunmap_atomic(kaddr, KM_USER0); + handle->buffer = buffer; + } else { + handle->buffer = page_address(page); + } } handle->prev = handle->cur; } @@ -1005,7 +1142,7 @@ static int mark_unsafe_pages(struct memory_bitmap *bm) unsigned long pfn, max_zone_pfn; /* Clear page flags */ - for_each_zone (zone) { + for_each_zone(zone) { 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)) @@ -1101,6 +1238,218 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) } } +/* List of "safe" pages that may be used to store data loaded from the suspend + * image + */ +static struct linked_page *safe_pages_list; + +#ifdef CONFIG_HIGHMEM +/* struct highmem_pbe is used for creating the list of highmem pages that + * should be restored atomically during the resume from disk, because the page + * frames they have occupied before the suspend are in use. + */ +struct highmem_pbe { + struct page *copy_page; /* data is here now */ + struct page *orig_page; /* data was here before the suspend */ + struct highmem_pbe *next; +}; + +/* List of highmem PBEs needed for restoring the highmem pages that were + * allocated before the suspend and included in the suspend image, but have + * also been allocated by the "resume" kernel, so their contents cannot be + * written directly to their "original" page frames. + */ +static struct highmem_pbe *highmem_pblist; + +/** + * count_highmem_image_pages - compute the number of highmem pages in the + * suspend image. The bits in the memory bitmap @bm that correspond to the + * image pages are assumed to be set. + */ + +static unsigned int count_highmem_image_pages(struct memory_bitmap *bm) +{ + unsigned long pfn; + unsigned int cnt = 0; + + memory_bm_position_reset(bm); + pfn = memory_bm_next_pfn(bm); + while (pfn != BM_END_OF_MAP) { + if (PageHighMem(pfn_to_page(pfn))) + cnt++; + + pfn = memory_bm_next_pfn(bm); + } + return cnt; +} + +/** + * prepare_highmem_image - try to allocate as many highmem pages as + * there are highmem image pages (@nr_highmem_p points to the variable + * containing the number of highmem image pages). The pages that are + * "safe" (ie. will not be overwritten when the suspend image is + * restored) have the corresponding bits set in @bm (it must be + * unitialized). + * + * NOTE: This function should not be called if there are no highmem + * image pages. + */ + +static unsigned int safe_highmem_pages; + +static struct memory_bitmap *safe_highmem_bm; + +static int +prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p) +{ + unsigned int to_alloc; + + if (memory_bm_create(bm, GFP_ATOMIC, PG_SAFE)) + return -ENOMEM; + + if (get_highmem_buffer(PG_SAFE)) + return -ENOMEM; + + to_alloc = count_free_highmem_pages(); + if (to_alloc > *nr_highmem_p) + to_alloc = *nr_highmem_p; + else + *nr_highmem_p = to_alloc; + + safe_highmem_pages = 0; + while (to_alloc-- > 0) { + struct page *page; + + page = alloc_page(__GFP_HIGHMEM); + if (!PageNosaveFree(page)) { + /* The page is "safe", set its bit the bitmap */ + memory_bm_set_bit(bm, page_to_pfn(page)); + safe_highmem_pages++; + } + /* Mark the page as allocated */ + SetPageNosave(page); + SetPageNosaveFree(page); + } + memory_bm_position_reset(bm); + safe_highmem_bm = bm; + return 0; +} + +/** + * get_highmem_page_buffer - for given highmem image page find the buffer + * that suspend_write_next() should set for its caller to write to. + * + * If the page is to be saved to its "original" page frame or a copy of + * the page is to be made in the highmem, @buffer is returned. Otherwise, + * the copy of the page is to be made in normal memory, so the address of + * the copy is returned. + * + * If @buffer is returned, the caller of suspend_write_next() will write + * the page's contents to @buffer, so they will have to be copied to the + * right location on the next call to suspend_write_next() and it is done + * with the help of copy_last_highmem_page(). For this purpose, if + * @buffer is returned, @last_highmem page is set to the page to which + * the data will have to be copied from @buffer. + */ + +static struct page *last_highmem_page; + +static void * +get_highmem_page_buffer(struct page *page, struct chain_allocator *ca) +{ + struct highmem_pbe *pbe; + void *kaddr; + + if (PageNosave(page) && PageNosaveFree(page)) { + /* We have allocated the "original" page frame and we can + * use it directly to store the loaded page. + */ + last_highmem_page = page; + return buffer; + } + /* The "original" page frame has not been allocated and we have to + * use a "safe" page frame to store the loaded page. + */ + pbe = chain_alloc(ca, sizeof(struct highmem_pbe)); + if (!pbe) { + swsusp_free(); + return NULL; + } + pbe->orig_page = page; + if (safe_highmem_pages > 0) { + struct page *tmp; + + /* Copy of the page will be stored in high memory */ + kaddr = buffer; + tmp = pfn_to_page(memory_bm_next_pfn(safe_highmem_bm)); + safe_highmem_pages--; + last_highmem_page = tmp; + pbe->copy_page = tmp; + } else { + /* Copy of the page will be stored in normal memory */ + kaddr = safe_pages_list; + safe_pages_list = safe_pages_list->next; + pbe->copy_page = virt_to_page(kaddr); + } + pbe->next = highmem_pblist; + highmem_pblist = pbe; + return kaddr; +} + +/** + * copy_last_highmem_page - copy the contents of a highmem image from + * @buffer, where the caller of snapshot_write_next() has place them, + * to the right location represented by @last_highmem_page . + */ + +static void copy_last_highmem_page(void) +{ + if (last_highmem_page) { + void *dst; + + dst = kmap_atomic(last_highmem_page, KM_USER0); + memcpy(dst, buffer, PAGE_SIZE); + kunmap_atomic(dst, KM_USER0); + last_highmem_page = NULL; + } +} + +static inline int last_highmem_page_copied(void) +{ + return !last_highmem_page; +} + +static inline void free_highmem_data(void) +{ + if (safe_highmem_bm) + memory_bm_free(safe_highmem_bm, PG_UNSAFE_CLEAR); + + if (buffer) + free_image_page(buffer, PG_UNSAFE_CLEAR); +} +#else +static inline int get_safe_write_buffer(void) { return 0; } + +static unsigned int +count_highmem_image_pages(struct memory_bitmap *bm) { return 0; } + +static inline int +prepare_highmem_image(struct memory_bitmap *bm, unsigned int *nr_highmem_p) +{ + return 0; +} + +static inline void * +get_highmem_page_buffer(struct page *page, struct chain_allocator *ca) +{ + return NULL; +} + +static inline void copy_last_highmem_page(void) {} +static inline int last_highmem_page_copied(void) { return 1; } +static inline void free_highmem_data(void) {} +#endif /* CONFIG_HIGHMEM */ + /** * prepare_image - use the memory bitmap @bm to mark the pages that will * be overwritten in the process of restoring the system memory state @@ -1110,20 +1459,25 @@ unpack_orig_pfns(unsigned long *buf, struct memory_bitmap *bm) * 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. + * allocated and create a lists of "safe" pages that will be used + * later. On systems with high memory a list of "safe" highmem pages is + * also created. */ #define PBES_PER_LINKED_PAGE (LINKED_PAGE_DATA_SIZE / sizeof(struct pbe)) -static struct linked_page *safe_pages_list; - static int prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) { - unsigned int nr_pages; + unsigned int nr_pages, nr_highmem; struct linked_page *sp_list, *lp; int error; + /* If there is no highmem, the buffer will not be necessary */ + free_image_page(buffer, PG_UNSAFE_CLEAR); + buffer = NULL; + + nr_highmem = count_highmem_image_pages(bm); error = mark_unsafe_pages(bm); if (error) goto Free; @@ -1134,6 +1488,11 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) duplicate_memory_bitmap(new_bm, bm); memory_bm_free(bm, PG_UNSAFE_KEEP); + if (nr_highmem > 0) { + error = prepare_highmem_image(bm, &nr_highmem); + if (error) + goto Free; + } /* Reserve some safe pages for potential later use. * * NOTE: This way we make sure there will be enough safe pages for the @@ -1142,10 +1501,10 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) */ sp_list = NULL; /* nr_copy_pages cannot be lesser than allocated_unsafe_pages */ - nr_pages = nr_copy_pages - allocated_unsafe_pages; + nr_pages = nr_copy_pages - nr_highmem - 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); + lp = get_image_page(GFP_ATOMIC, PG_SAFE); if (!lp) { error = -ENOMEM; goto Free; @@ -1156,7 +1515,7 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) } /* Preallocate memory for the image */ safe_pages_list = NULL; - nr_pages = nr_copy_pages - allocated_unsafe_pages; + nr_pages = nr_copy_pages - nr_highmem - allocated_unsafe_pages; while (nr_pages > 0) { lp = (struct linked_page *)get_zeroed_page(GFP_ATOMIC); if (!lp) { @@ -1181,7 +1540,7 @@ prepare_image(struct memory_bitmap *new_bm, struct memory_bitmap *bm) } return 0; -Free: + Free: swsusp_free(); return error; } @@ -1196,6 +1555,9 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca) struct pbe *pbe; struct page *page = pfn_to_page(memory_bm_next_pfn(bm)); + if (PageHighMem(page)) + return get_highmem_page_buffer(page, ca); + if (PageNosave(page) && PageNosaveFree(page)) /* We have allocated the "original" page frame and we can * use it directly to store the loaded page. @@ -1210,12 +1572,12 @@ static void *get_buffer(struct memory_bitmap *bm, struct chain_allocator *ca) swsusp_free(); return NULL; } - pbe->orig_address = (unsigned long)page_address(page); - pbe->address = (unsigned long)safe_pages_list; + pbe->orig_address = page_address(page); + pbe->address = safe_pages_list; safe_pages_list = safe_pages_list->next; pbe->next = restore_pblist; restore_pblist = pbe; - return (void *)pbe->address; + return pbe->address; } /** @@ -1249,14 +1611,16 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) 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); + if (handle->offset == 0) { + if (!buffer) + /* This makes the buffer be freed by swsusp_free() */ + buffer = get_image_page(GFP_ATOMIC, PG_ANY); + if (!buffer) return -ENOMEM; - } - if (!handle->offset) + handle->buffer = buffer; + } handle->sync_read = 1; if (handle->prev < handle->cur) { if (handle->prev == 0) { @@ -1284,8 +1648,10 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) return -ENOMEM; } } else { + copy_last_highmem_page(); handle->buffer = get_buffer(&orig_bm, &ca); - handle->sync_read = 0; + if (handle->buffer != buffer) + handle->sync_read = 0; } handle->prev = handle->cur; } @@ -1301,15 +1667,73 @@ int snapshot_write_next(struct snapshot_handle *handle, size_t count) return count; } +/** + * snapshot_write_finalize - must be called after the last call to + * snapshot_write_next() in case the last page in the image happens + * to be a highmem page and its contents should be stored in the + * highmem. Additionally, it releases the memory that will not be + * used any more. + */ + +void snapshot_write_finalize(struct snapshot_handle *handle) +{ + copy_last_highmem_page(); + /* 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); + free_highmem_data(); + } +} + int snapshot_image_loaded(struct snapshot_handle *handle) { - return !(!nr_copy_pages || + return !(!nr_copy_pages || !last_highmem_page_copied() || handle->cur <= nr_meta_pages + nr_copy_pages); } -void snapshot_free_unused_memory(struct snapshot_handle *handle) +#ifdef CONFIG_HIGHMEM +/* Assumes that @buf is ready and points to a "safe" page */ +static inline void +swap_two_pages_data(struct page *p1, struct page *p2, void *buf) { - /* 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); + void *kaddr1, *kaddr2; + + kaddr1 = kmap_atomic(p1, KM_USER0); + kaddr2 = kmap_atomic(p2, KM_USER1); + memcpy(buf, kaddr1, PAGE_SIZE); + memcpy(kaddr1, kaddr2, PAGE_SIZE); + memcpy(kaddr2, buf, PAGE_SIZE); + kunmap_atomic(kaddr1, KM_USER0); + kunmap_atomic(kaddr2, KM_USER1); +} + +/** + * restore_highmem - for each highmem page that was allocated before + * the suspend and included in the suspend image, and also has been + * allocated by the "resume" kernel swap its current (ie. "before + * resume") contents with the previous (ie. "before suspend") one. + * + * If the resume eventually fails, we can call this function once + * again and restore the "before resume" highmem state. + */ + +int restore_highmem(void) +{ + struct highmem_pbe *pbe = highmem_pblist; + void *buf; + + if (!pbe) + return 0; + + buf = get_image_page(GFP_ATOMIC, PG_SAFE); + if (!buf) + return -ENOMEM; + + while (pbe) { + swap_two_pages_data(pbe->copy_page, pbe->orig_page, buf); + pbe = pbe->next; + } + free_image_page(buf, PG_UNSAFE_CLEAR); + return 0; } +#endif /* CONFIG_HIGHMEM */ diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 1a3b0dd2c3fc..f133d4a6d817 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -34,34 +34,123 @@ extern char resume_file[]; #define SWSUSP_SIG "S1SUSPEND" static struct swsusp_header { - char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)]; - swp_entry_t image; + char reserved[PAGE_SIZE - 20 - sizeof(sector_t)]; + sector_t image; char orig_sig[10]; char sig[10]; } __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header; /* - * Saving part... + * General things */ static unsigned short root_swap = 0xffff; +static struct block_device *resume_bdev; + +/** + * submit - submit BIO request. + * @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 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, struct page *page, + struct bio **bio_chain) +{ + struct bio *bio; + + bio = bio_alloc(__GFP_WAIT | __GFP_HIGH, 1); + if (!bio) + return -ENOMEM; + bio->bi_sector = page_off * (PAGE_SIZE >> 9); + bio->bi_bdev = resume_bdev; + bio->bi_end_io = end_swap_bio_read; + + 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; + } + + 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 { + if (rw == READ) + get_page(page); /* These pages are freed later */ + 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 *addr, struct bio **bio_chain) +{ + return submit(READ, page_off, virt_to_page(addr), bio_chain); +} + +static int bio_write_page(pgoff_t page_off, void *addr, struct bio **bio_chain) +{ + return submit(WRITE, page_off, virt_to_page(addr), bio_chain); +} + +static int wait_on_bio_chain(struct bio **bio_chain) +{ + struct bio *bio; + struct bio *next_bio; + int ret = 0; + + if (bio_chain == NULL) + return 0; + + bio = *bio_chain; + if (bio == NULL) + return 0; + 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; +} + +/* + * Saving part + */ -static int mark_swapfiles(swp_entry_t start) +static int mark_swapfiles(sector_t start) { int error; - rw_swap_page_sync(READ, swp_entry(root_swap, 0), - virt_to_page((unsigned long)&swsusp_header), NULL); + bio_read_page(swsusp_resume_block, &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), - NULL); + error = bio_write_page(swsusp_resume_block, + &swsusp_header, NULL); } else { - pr_debug("swsusp: Partition is not swap space.\n"); + printk(KERN_ERR "swsusp: Swap header not found!\n"); error = -ENODEV; } return error; @@ -74,12 +163,21 @@ static int mark_swapfiles(swp_entry_t start) static int swsusp_swap_check(void) /* This is called before saving image */ { - int res = swap_type_of(swsusp_resume_device); + int res; + + res = swap_type_of(swsusp_resume_device, swsusp_resume_block); + if (res < 0) + return res; + + root_swap = res; + resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_WRITE); + if (IS_ERR(resume_bdev)) + return PTR_ERR(resume_bdev); + + res = set_blocksize(resume_bdev, PAGE_SIZE); + if (res < 0) + blkdev_put(resume_bdev); - if (res >= 0) { - root_swap = res; - return 0; - } return res; } @@ -90,36 +188,26 @@ static int swsusp_swap_check(void) /* This is called before saving image */ * @bio_chain: Link the next write BIO here */ -static int write_page(void *buf, unsigned long offset, struct bio **bio_chain) +static int write_page(void *buf, sector_t 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; - } + void *src; + + if (!offset) + return -ENOSPC; + + if (bio_chain) { + src = (void *)__get_free_page(__GFP_WAIT | __GFP_HIGH); + if (src) { + memcpy(src, buf, PAGE_SIZE); + } else { + WARN_ON_ONCE(1); + bio_chain = NULL; /* Go synchronous */ + src = buf; } - entry = swp_entry(root_swap, offset); - error = rw_swap_page_sync(WRITE, entry, page, bio_chain); + } else { + src = buf; } - return error; + return bio_write_page(offset, src, bio_chain); } /* @@ -137,11 +225,11 @@ static int write_page(void *buf, unsigned long offset, struct bio **bio_chain) * at a time. */ -#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(long) - 1) +#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(sector_t) - 1) struct swap_map_page { - unsigned long entries[MAP_PAGE_ENTRIES]; - unsigned long next_swap; + sector_t entries[MAP_PAGE_ENTRIES]; + sector_t next_swap; }; /** @@ -151,7 +239,7 @@ struct swap_map_page { struct swap_map_handle { struct swap_map_page *cur; - unsigned long cur_swap; + sector_t cur_swap; struct bitmap_page *bitmap; unsigned int k; }; @@ -166,26 +254,6 @@ 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); @@ -196,7 +264,7 @@ static int get_swap_writer(struct swap_map_handle *handle) release_swap_writer(handle); return -ENOMEM; } - handle->cur_swap = alloc_swap_page(root_swap, handle->bitmap); + handle->cur_swap = alloc_swapdev_block(root_swap, handle->bitmap); if (!handle->cur_swap) { release_swap_writer(handle); return -ENOSPC; @@ -205,43 +273,15 @@ static int get_swap_writer(struct swap_map_handle *handle) return 0; } -static int wait_on_bio_chain(struct bio **bio_chain) -{ - struct bio *bio; - struct bio *next_bio; - int ret = 0; - - if (bio_chain == NULL) - return 0; - - bio = *bio_chain; - if (bio == NULL) - return 0; - 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; + sector_t offset; if (!handle->cur) return -EINVAL; - offset = alloc_swap_page(root_swap, handle->bitmap); + offset = alloc_swapdev_block(root_swap, handle->bitmap); error = write_page(buf, offset, bio_chain); if (error) return error; @@ -250,7 +290,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, error = wait_on_bio_chain(bio_chain); if (error) goto out; - offset = alloc_swap_page(root_swap, handle->bitmap); + offset = alloc_swapdev_block(root_swap, handle->bitmap); if (!offset) return -ENOSPC; handle->cur->next_swap = offset; @@ -261,7 +301,7 @@ static int swap_write_page(struct swap_map_handle *handle, void *buf, handle->cur_swap = offset; handle->k = 0; } -out: + out: return error; } @@ -315,7 +355,7 @@ static int save_image(struct swap_map_handle *handle, error = err2; if (!error) printk("\b\b\b\bdone\n"); - show_speed(&start, &stop, nr_to_write, "Wrote"); + swsusp_show_speed(&start, &stop, nr_to_write, "Wrote"); return error; } @@ -350,100 +390,50 @@ int swsusp_write(void) struct swsusp_info *header; int error; - if ((error = swsusp_swap_check())) { + error = swsusp_swap_check(); + if (error) { printk(KERN_ERR "swsusp: Cannot find swap device, try " "swapon -a.\n"); return error; } memset(&snapshot, 0, sizeof(struct snapshot_handle)); error = snapshot_read_next(&snapshot, PAGE_SIZE); - if (error < PAGE_SIZE) - return error < 0 ? error : -EFAULT; + if (error < PAGE_SIZE) { + if (error >= 0) + error = -EFAULT; + + goto out; + } header = (struct swsusp_info *)data_of(snapshot); if (!enough_swap(header->pages)) { printk(KERN_ERR "swsusp: Not enough free swap\n"); - return -ENOSPC; + error = -ENOSPC; + goto out; } error = get_swap_writer(&handle); if (!error) { - unsigned long start = handle.cur_swap; + sector_t start = handle.cur_swap; + error = swap_write_page(&handle, header, NULL); if (!error) error = save_image(&handle, &snapshot, header->pages - 1); + if (!error) { flush_swap_writer(&handle); printk("S"); - error = mark_swapfiles(swp_entry(root_swap, start)); + error = mark_swapfiles(start); printk("|\n"); } } if (error) free_all_swap_pages(root_swap, handle.bitmap); release_swap_writer(&handle); + out: + swsusp_close(); return error; } -static struct block_device *resume_bdev; - -/** - * submit - submit BIO request. - * @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 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, struct page *page, - struct bio **bio_chain) -{ - struct bio *bio; - - bio = bio_alloc(GFP_ATOMIC, 1); - if (!bio) - return -ENOMEM; - bio->bi_sector = page_off * (PAGE_SIZE >> 9); - bio->bi_bdev = resume_bdev; - bio->bi_end_io = end_swap_bio_read; - - 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; - } - - 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 { - if (rw == READ) - get_page(page); /* These pages are freed later */ - 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 *addr, struct bio **bio_chain) -{ - return submit(READ, page_off, virt_to_page(addr), bio_chain); -} - -static int bio_write_page(pgoff_t page_off, void *addr) -{ - return submit(WRITE, page_off, virt_to_page(addr), NULL); -} - /** * The following functions allow us to read data using a swap map * in a file-alike way @@ -456,17 +446,18 @@ static void release_swap_reader(struct swap_map_handle *handle) handle->cur = NULL; } -static int get_swap_reader(struct swap_map_handle *handle, - swp_entry_t start) +static int get_swap_reader(struct swap_map_handle *handle, sector_t start) { int error; - if (!swp_offset(start)) + if (!start) return -EINVAL; - handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC); + + handle->cur = (struct swap_map_page *)get_zeroed_page(__GFP_WAIT | __GFP_HIGH); if (!handle->cur) return -ENOMEM; - error = bio_read_page(swp_offset(start), handle->cur, NULL); + + error = bio_read_page(start, handle->cur, NULL); if (error) { release_swap_reader(handle); return error; @@ -478,7 +469,7 @@ static int get_swap_reader(struct swap_map_handle *handle, static int swap_read_page(struct swap_map_handle *handle, void *buf, struct bio **bio_chain) { - unsigned long offset; + sector_t offset; int error; if (!handle->cur) @@ -547,11 +538,11 @@ static int load_image(struct swap_map_handle *handle, error = err2; if (!error) { printk("\b\b\b\bdone\n"); - snapshot_free_unused_memory(snapshot); + snapshot_write_finalize(snapshot); if (!snapshot_image_loaded(snapshot)) error = -ENODATA; } - show_speed(&start, &stop, nr_to_read, "Read"); + swsusp_show_speed(&start, &stop, nr_to_read, "Read"); return error; } @@ -600,12 +591,16 @@ 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, NULL))) + error = bio_read_page(swsusp_resume_block, + &swsusp_header, NULL); + if (error) return error; + if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) { memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10); /* Reset swap signature now */ - error = bio_write_page(0, &swsusp_header); + error = bio_write_page(swsusp_resume_block, + &swsusp_header, NULL); } else { return -EINVAL; } diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c index 0b66659dc516..31aa0390c777 100644 --- a/kernel/power/swsusp.c +++ b/kernel/power/swsusp.c @@ -49,6 +49,7 @@ #include <linux/bootmem.h> #include <linux/syscalls.h> #include <linux/highmem.h> +#include <linux/time.h> #include "power.h" @@ -64,10 +65,8 @@ int in_suspend __nosavedata = 0; #ifdef CONFIG_HIGHMEM unsigned int count_highmem_pages(void); -int save_highmem(void); int restore_highmem(void); #else -static inline int save_highmem(void) { return 0; } static inline int restore_highmem(void) { return 0; } static inline unsigned int count_highmem_pages(void) { return 0; } #endif @@ -134,18 +133,18 @@ static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit) return 0; } -unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap) +sector_t alloc_swapdev_block(int swap, struct bitmap_page *bitmap) { unsigned long offset; offset = swp_offset(get_swap_page_of_type(swap)); if (offset) { - if (bitmap_set(bitmap, offset)) { + if (bitmap_set(bitmap, offset)) swap_free(swp_entry(swap, offset)); - offset = 0; - } + else + return swapdev_block(swap, offset); } - return offset; + return 0; } void free_all_swap_pages(int swap, struct bitmap_page *bitmap) @@ -166,6 +165,34 @@ void free_all_swap_pages(int swap, struct bitmap_page *bitmap) } /** + * swsusp_show_speed - print the time elapsed between two events represented by + * @start and @stop + * + * @nr_pages - number of pages processed between @start and @stop + * @msg - introductory message to print + */ + +void swsusp_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); +} + +/** * swsusp_shrink_memory - Try to free as much memory as needed * * ... but do not OOM-kill anyone @@ -184,23 +211,37 @@ static inline unsigned long __shrink_memory(long tmp) int swsusp_shrink_memory(void) { - long size, tmp; + long tmp; struct zone *zone; unsigned long pages = 0; unsigned int i = 0; char *p = "-\\|/"; + struct timeval start, stop; printk("Shrinking memory... "); + do_gettimeofday(&start); do { - size = 2 * count_highmem_pages(); - size += size / 50 + count_data_pages() + PAGES_FOR_IO; + long size, highmem_size; + + highmem_size = count_highmem_pages(); + size = count_data_pages() + PAGES_FOR_IO; tmp = size; + size += highmem_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 (populated_zone(zone)) { + if (is_highmem(zone)) { + highmem_size -= zone->free_pages; + } else { + tmp -= zone->free_pages; + tmp += zone->lowmem_reserve[ZONE_NORMAL]; + tmp += snapshot_additional_pages(zone); + } } + + if (highmem_size < 0) + highmem_size = 0; + + tmp += highmem_size; if (tmp > 0) { tmp = __shrink_memory(tmp); if (!tmp) @@ -212,7 +253,9 @@ int swsusp_shrink_memory(void) } printk("\b%c", p[i++%4]); } while (tmp > 0); + do_gettimeofday(&stop); printk("\bdone (%lu pages freed)\n", pages); + swsusp_show_speed(&start, &stop, pages, "Freed"); return 0; } @@ -223,6 +266,7 @@ int swsusp_suspend(void) if ((error = arch_prepare_suspend())) return error; + local_irq_disable(); /* At this point, device_suspend() has been called, but *not* * device_power_down(). We *must* device_power_down() now. @@ -235,23 +279,16 @@ int swsusp_suspend(void) goto Enable_irqs; } - if ((error = save_highmem())) { - printk(KERN_ERR "swsusp: Not enough free pages for highmem\n"); - goto Restore_highmem; - } - save_processor_state(); if ((error = swsusp_arch_suspend())) printk(KERN_ERR "Error %d suspending\n", error); /* Restore control flow magically appears here */ 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: + Enable_irqs: local_irq_enable(); return error; } @@ -268,18 +305,23 @@ int swsusp_resume(void) 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(); - error = swsusp_arch_resume(); - /* Code below is only ever reached in case of failure. Otherwise - * execution continues at place where swsusp_arch_suspend was called - */ - BUG_ON(!error); + error = restore_highmem(); + if (!error) { + error = swsusp_arch_resume(); + /* The code below is only ever reached in case of a failure. + * Otherwise execution continues at place where + * swsusp_arch_suspend() was called + */ + BUG_ON(!error); + /* This call to restore_highmem() undos the previous one */ + restore_highmem(); + } /* The only reason why swsusp_arch_resume() can fail is memory being * very tight, so we have to free it as soon as we can to avoid * subsequent failures */ swsusp_free(); restore_processor_state(); - restore_highmem(); touch_softlockup_watchdog(); device_power_up(); local_irq_enable(); diff --git a/kernel/power/user.c b/kernel/power/user.c index d991d3b0e5a4..89443b85163b 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -11,6 +11,7 @@ #include <linux/suspend.h> #include <linux/syscalls.h> +#include <linux/reboot.h> #include <linux/string.h> #include <linux/device.h> #include <linux/miscdevice.h> @@ -21,6 +22,7 @@ #include <linux/fs.h> #include <linux/console.h> #include <linux/cpu.h> +#include <linux/freezer.h> #include <asm/uaccess.h> @@ -54,7 +56,8 @@ static int snapshot_open(struct inode *inode, struct file *filp) filp->private_data = data; memset(&data->handle, 0, sizeof(struct snapshot_handle)); if ((filp->f_flags & O_ACCMODE) == O_RDONLY) { - data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device) : -1; + data->swap = swsusp_resume_device ? + swap_type_of(swsusp_resume_device, 0) : -1; data->mode = O_RDONLY; } else { data->swap = -1; @@ -76,10 +79,10 @@ static int snapshot_release(struct inode *inode, struct file *filp) free_all_swap_pages(data->swap, data->bitmap); free_bitmap(data->bitmap); if (data->frozen) { - down(&pm_sem); + mutex_lock(&pm_mutex); thaw_processes(); enable_nonboot_cpus(); - up(&pm_sem); + mutex_unlock(&pm_mutex); } atomic_inc(&device_available); return 0; @@ -124,7 +127,8 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, { int error = 0; struct snapshot_data *data; - loff_t offset, avail; + loff_t avail; + sector_t offset; if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC) return -ENOTTY; @@ -140,7 +144,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, case SNAPSHOT_FREEZE: if (data->frozen) break; - down(&pm_sem); + mutex_lock(&pm_mutex); error = disable_nonboot_cpus(); if (!error) { error = freeze_processes(); @@ -150,7 +154,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -EBUSY; } } - up(&pm_sem); + mutex_unlock(&pm_mutex); if (!error) data->frozen = 1; break; @@ -158,10 +162,10 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, case SNAPSHOT_UNFREEZE: if (!data->frozen) break; - down(&pm_sem); + mutex_lock(&pm_mutex); thaw_processes(); enable_nonboot_cpus(); - up(&pm_sem); + mutex_unlock(&pm_mutex); data->frozen = 0; break; @@ -170,7 +174,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, error = -EPERM; break; } - down(&pm_sem); + mutex_lock(&pm_mutex); /* Free memory before shutting down devices. */ error = swsusp_shrink_memory(); if (!error) { @@ -183,7 +187,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, } resume_console(); } - up(&pm_sem); + mutex_unlock(&pm_mutex); if (!error) error = put_user(in_suspend, (unsigned int __user *)arg); if (!error) @@ -191,13 +195,13 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; case SNAPSHOT_ATOMIC_RESTORE: + snapshot_write_finalize(&data->handle); if (data->mode != O_WRONLY || !data->frozen || !snapshot_image_loaded(&data->handle)) { error = -EPERM; break; } - snapshot_free_unused_memory(&data->handle); - down(&pm_sem); + mutex_lock(&pm_mutex); pm_prepare_console(); suspend_console(); error = device_suspend(PMSG_PRETHAW); @@ -207,7 +211,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, } resume_console(); pm_restore_console(); - up(&pm_sem); + mutex_unlock(&pm_mutex); break; case SNAPSHOT_FREE: @@ -238,10 +242,10 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; } } - offset = alloc_swap_page(data->swap, data->bitmap); + offset = alloc_swapdev_block(data->swap, data->bitmap); if (offset) { offset <<= PAGE_SHIFT; - error = put_user(offset, (loff_t __user *)arg); + error = put_user(offset, (sector_t __user *)arg); } else { error = -ENOSPC; } @@ -264,7 +268,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, * so we need to recode them */ if (old_decode_dev(arg)) { - data->swap = swap_type_of(old_decode_dev(arg)); + data->swap = swap_type_of(old_decode_dev(arg), 0); if (data->swap < 0) error = -ENODEV; } else { @@ -282,7 +286,7 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, break; } - if (down_trylock(&pm_sem)) { + if (!mutex_trylock(&pm_mutex)) { error = -EBUSY; break; } @@ -309,8 +313,66 @@ static int snapshot_ioctl(struct inode *inode, struct file *filp, if (pm_ops->finish) pm_ops->finish(PM_SUSPEND_MEM); -OutS3: - up(&pm_sem); + OutS3: + mutex_unlock(&pm_mutex); + break; + + case SNAPSHOT_PMOPS: + switch (arg) { + + case PMOPS_PREPARE: + if (pm_ops->prepare) { + error = pm_ops->prepare(PM_SUSPEND_DISK); + } + break; + + case PMOPS_ENTER: + kernel_shutdown_prepare(SYSTEM_SUSPEND_DISK); + error = pm_ops->enter(PM_SUSPEND_DISK); + break; + + case PMOPS_FINISH: + if (pm_ops && pm_ops->finish) { + pm_ops->finish(PM_SUSPEND_DISK); + } + break; + + default: + printk(KERN_ERR "SNAPSHOT_PMOPS: invalid argument %ld\n", arg); + error = -EINVAL; + + } + break; + + case SNAPSHOT_SET_SWAP_AREA: + if (data->bitmap) { + error = -EPERM; + } else { + struct resume_swap_area swap_area; + dev_t swdev; + + error = copy_from_user(&swap_area, (void __user *)arg, + sizeof(struct resume_swap_area)); + if (error) { + error = -EFAULT; + break; + } + + /* + * User space encodes device types as two-byte values, + * so we need to recode them + */ + swdev = old_decode_dev(swap_area.dev); + if (swdev) { + offset = swap_area.offset; + data->swap = swap_type_of(swdev, offset); + if (data->swap < 0) + error = -ENODEV; + } else { + data->swap = -1; + error = -EINVAL; + } + } break; default: @@ -321,7 +383,7 @@ OutS3: return error; } -static struct file_operations snapshot_fops = { +static const struct file_operations snapshot_fops = { .open = snapshot_open, .release = snapshot_release, .read = snapshot_read, diff --git a/kernel/printk.c b/kernel/printk.c index 66426552fbfe..185bb45eacf7 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -53,8 +53,6 @@ int console_printk[4] = { DEFAULT_CONSOLE_LOGLEVEL, /* default_console_loglevel */ }; -EXPORT_UNUSED_SYMBOL(console_printk); /* June 2006 */ - /* * Low lever drivers may need that to know if they can schedule in * their unblank() callback or not. So let's export it. @@ -335,13 +333,25 @@ static void __call_console_drivers(unsigned long start, unsigned long end) } } +static int __read_mostly ignore_loglevel; + +int __init ignore_loglevel_setup(char *str) +{ + ignore_loglevel = 1; + printk(KERN_INFO "debug: ignoring loglevel setting.\n"); + + return 1; +} + +__setup("ignore_loglevel", ignore_loglevel_setup); + /* * Write out chars from start to end - 1 inclusive */ static void _call_console_drivers(unsigned long start, unsigned long end, int msg_log_level) { - if (msg_log_level < console_loglevel && + if ((msg_log_level < console_loglevel || ignore_loglevel) && console_drivers && start != end) { if ((start & LOG_BUF_MASK) > (end & LOG_BUF_MASK)) { /* wrapped write */ @@ -631,12 +641,7 @@ EXPORT_SYMBOL(vprintk); asmlinkage long sys_syslog(int type, char __user *buf, int len) { - return 0; -} - -int do_syslog(int type, char __user *buf, int len) -{ - return 0; + return -ENOSYS; } static void call_console_drivers(unsigned long start, unsigned long end) @@ -777,7 +782,6 @@ int is_console_locked(void) { return console_locked; } -EXPORT_UNUSED_SYMBOL(is_console_locked); /* June 2006 */ /** * release_console_sem - unlock the console system diff --git a/kernel/profile.c b/kernel/profile.c index f940b462eec9..fb5e03d57e9d 100644 --- a/kernel/profile.c +++ b/kernel/profile.c @@ -40,7 +40,7 @@ int (*timer_hook)(struct pt_regs *) __read_mostly; static atomic_t *prof_buffer; static unsigned long prof_len, prof_shift; -static int prof_on __read_mostly; +int prof_on __read_mostly; static cpumask_t prof_cpu_mask = CPU_MASK_ALL; #ifdef CONFIG_SMP static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits); @@ -51,9 +51,19 @@ static DEFINE_MUTEX(profile_flip_mutex); static int __init profile_setup(char * str) { static char __initdata schedstr[] = "schedule"; + static char __initdata sleepstr[] = "sleep"; int par; - if (!strncmp(str, schedstr, strlen(schedstr))) { + if (!strncmp(str, sleepstr, strlen(sleepstr))) { + prof_on = SLEEP_PROFILING; + if (str[strlen(sleepstr)] == ',') + str += strlen(sleepstr) + 1; + if (get_option(&str, &par)) + prof_shift = par; + printk(KERN_INFO + "kernel sleep profiling enabled (shift: %ld)\n", + prof_shift); + } else if (!strncmp(str, sleepstr, strlen(sleepstr))) { prof_on = SCHED_PROFILING; if (str[strlen(schedstr)] == ',') str += strlen(schedstr) + 1; @@ -204,7 +214,8 @@ EXPORT_SYMBOL_GPL(profile_event_unregister); * positions to which hits are accounted during short intervals (e.g. * several seconds) is usually very small. Exclusion from buffer * flipping is provided by interrupt disablement (note that for - * SCHED_PROFILING profile_hit() may be called from process context). + * SCHED_PROFILING or SLEEP_PROFILING profile_hit() may be called from + * process context). * The hash function is meant to be lightweight as opposed to strong, * and was vaguely inspired by ppc64 firmware-supported inverted * pagetable hash functions, but uses a full hashtable full of finite @@ -257,7 +268,7 @@ static void profile_discard_flip_buffers(void) mutex_unlock(&profile_flip_mutex); } -void profile_hit(int type, void *__pc) +void profile_hits(int type, void *__pc, unsigned int nr_hits) { unsigned long primary, secondary, flags, pc = (unsigned long)__pc; int i, j, cpu; @@ -274,21 +285,31 @@ void profile_hit(int type, void *__pc) put_cpu(); return; } + /* + * We buffer the global profiler buffer into a per-CPU + * queue and thus reduce the number of global (and possibly + * NUMA-alien) accesses. The write-queue is self-coalescing: + */ local_irq_save(flags); do { for (j = 0; j < PROFILE_GRPSZ; ++j) { if (hits[i + j].pc == pc) { - hits[i + j].hits++; + hits[i + j].hits += nr_hits; goto out; } else if (!hits[i + j].hits) { hits[i + j].pc = pc; - hits[i + j].hits = 1; + hits[i + j].hits = nr_hits; goto out; } } i = (i + secondary) & (NR_PROFILE_HIT - 1); } while (i != primary); - atomic_inc(&prof_buffer[pc]); + + /* + * Add the current hit(s) and flush the write-queue out + * to the global buffer: + */ + atomic_add(nr_hits, &prof_buffer[pc]); for (i = 0; i < NR_PROFILE_HIT; ++i) { atomic_add(hits[i].hits, &prof_buffer[hits[i].pc]); hits[i].pc = hits[i].hits = 0; @@ -298,7 +319,6 @@ out: put_cpu(); } -#ifdef CONFIG_HOTPLUG_CPU static int __devinit profile_cpu_callback(struct notifier_block *info, unsigned long action, void *__cpu) { @@ -351,19 +371,19 @@ static int __devinit profile_cpu_callback(struct notifier_block *info, } return NOTIFY_OK; } -#endif /* CONFIG_HOTPLUG_CPU */ #else /* !CONFIG_SMP */ #define profile_flip_buffers() do { } while (0) #define profile_discard_flip_buffers() do { } while (0) +#define profile_cpu_callback NULL -void profile_hit(int type, void *__pc) +void profile_hits(int type, void *__pc, unsigned int nr_hits) { unsigned long pc; if (prof_on != type || !prof_buffer) return; pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift; - atomic_inc(&prof_buffer[min(pc, prof_len - 1)]); + atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]); } #endif /* !CONFIG_SMP */ @@ -442,7 +462,8 @@ read_profile(struct file *file, char __user *buf, size_t count, loff_t *ppos) read = 0; while (p < sizeof(unsigned int) && count > 0) { - put_user(*((char *)(&sample_step)+p),buf); + if (put_user(*((char *)(&sample_step)+p),buf)) + return -EFAULT; buf++; p++; count--; read++; } pnt = (char *)prof_buffer + p - sizeof(atomic_t); @@ -480,7 +501,7 @@ static ssize_t write_profile(struct file *file, const char __user *buf, return count; } -static struct file_operations proc_profile_operations = { +static const struct file_operations proc_profile_operations = { .read = read_profile, .write = write_profile, }; diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c index 26bb5ffe1ef1..3554b76da84c 100644 --- a/kernel/rcupdate.c +++ b/kernel/rcupdate.c @@ -235,12 +235,14 @@ static void rcu_do_batch(struct rcu_data *rdp) list = rdp->donelist; while (list) { - next = rdp->donelist = list->next; + next = list->next; + prefetch(next); list->func(list); list = next; if (++count >= rdp->blimit) break; } + rdp->donelist = list; local_irq_disable(); rdp->qlen -= count; diff --git a/kernel/rcutorture.c b/kernel/rcutorture.c index e2bda18f6f42..c52f981ea008 100644 --- a/kernel/rcutorture.c +++ b/kernel/rcutorture.c @@ -401,7 +401,7 @@ static void srcu_torture_cleanup(void) cleanup_srcu_struct(&srcu_ctl); } -static int srcu_torture_read_lock(void) +static int srcu_torture_read_lock(void) __acquires(&srcu_ctl) { return srcu_read_lock(&srcu_ctl); } @@ -419,7 +419,7 @@ static void srcu_read_delay(struct rcu_random_state *rrsp) schedule_timeout_interruptible(longdelay); } -static void srcu_torture_read_unlock(int idx) +static void srcu_torture_read_unlock(int idx) __releases(&srcu_ctl) { srcu_read_unlock(&srcu_ctl, idx); } diff --git a/kernel/relay.c b/kernel/relay.c index 2b92e8ece85b..75a3a9a7efc2 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -1013,7 +1013,7 @@ static ssize_t relay_file_sendfile(struct file *filp, actor, &desc); } -struct file_operations relay_file_operations = { +const struct file_operations relay_file_operations = { .open = relay_file_open, .poll = relay_file_poll, .mmap = relay_file_mmap, diff --git a/kernel/resource.c b/kernel/resource.c index 6de60c12143e..7b9a497419d9 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -88,7 +88,7 @@ static int r_show(struct seq_file *m, void *v) return 0; } -static struct seq_operations resource_op = { +static const struct seq_operations resource_op = { .start = r_start, .next = r_next, .stop = r_stop, @@ -115,14 +115,14 @@ static int iomem_open(struct inode *inode, struct file *file) return res; } -static struct file_operations proc_ioports_operations = { +static const struct file_operations proc_ioports_operations = { .open = ioports_open, .read = seq_read, .llseek = seq_lseek, .release = seq_release, }; -static struct file_operations proc_iomem_operations = { +static const struct file_operations proc_iomem_operations = { .open = iomem_open, .read = seq_read, .llseek = seq_lseek, diff --git a/kernel/rtmutex-tester.c b/kernel/rtmutex-tester.c index 6dcea9dd8c94..015fc633c96c 100644 --- a/kernel/rtmutex-tester.c +++ b/kernel/rtmutex-tester.c @@ -13,6 +13,7 @@ #include <linux/spinlock.h> #include <linux/sysdev.h> #include <linux/timer.h> +#include <linux/freezer.h> #include "rtmutex.h" diff --git a/kernel/sched.c b/kernel/sched.c index 3399701c680e..f385eff4682d 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -34,7 +34,7 @@ #include <linux/security.h> #include <linux/notifier.h> #include <linux/profile.h> -#include <linux/suspend.h> +#include <linux/freezer.h> #include <linux/vmalloc.h> #include <linux/blkdev.h> #include <linux/delay.h> @@ -505,7 +505,7 @@ static int schedstat_open(struct inode *inode, struct file *file) return res; } -struct file_operations proc_schedstat_operations = { +const struct file_operations proc_schedstat_operations = { .open = schedstat_open, .read = seq_read, .llseek = seq_lseek, @@ -948,6 +948,17 @@ static void activate_task(struct task_struct *p, struct rq *rq, int local) } #endif + /* + * Sleep time is in units of nanosecs, so shift by 20 to get a + * milliseconds-range estimation of the amount of time that the task + * spent sleeping: + */ + if (unlikely(prof_on == SLEEP_PROFILING)) { + if (p->state == TASK_UNINTERRUPTIBLE) + profile_hits(SLEEP_PROFILING, (void *)get_wchan(p), + (now - p->timestamp) >> 20); + } + if (!rt_task(p)) p->prio = recalc_task_prio(p, now); @@ -3333,6 +3344,7 @@ asmlinkage void __sched schedule(void) printk(KERN_ERR "BUG: scheduling while atomic: " "%s/0x%08x/%d\n", current->comm, preempt_count(), current->pid); + debug_show_held_locks(current); dump_stack(); } profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -4804,18 +4816,18 @@ static void show_task(struct task_struct *p) show_stack(p, NULL); } -void show_state(void) +void show_state_filter(unsigned long state_filter) { struct task_struct *g, *p; #if (BITS_PER_LONG == 32) printk("\n" - " sibling\n"); - printk(" task PC pid father child younger older\n"); + " free sibling\n"); + printk(" task PC stack pid father child younger older\n"); #else printk("\n" - " sibling\n"); - printk(" task PC pid father child younger older\n"); + " free sibling\n"); + printk(" task PC stack pid father child younger older\n"); #endif read_lock(&tasklist_lock); do_each_thread(g, p) { @@ -4824,11 +4836,16 @@ void show_state(void) * console might take alot of time: */ touch_nmi_watchdog(); - show_task(p); + if (p->state & state_filter) + show_task(p); } while_each_thread(g, p); read_unlock(&tasklist_lock); - debug_show_all_locks(); + /* + * Only show locks if all tasks are dumped: + */ + if (state_filter == -1) + debug_show_all_locks(); } /** @@ -6723,8 +6740,6 @@ SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, sched_smt_power_savings_store); #endif - -#ifdef CONFIG_HOTPLUG_CPU /* * Force a reinitialization of the sched domains hierarchy. The domains * and groups cannot be updated in place without racing with the balancing @@ -6757,7 +6772,6 @@ static int update_sched_domains(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void __init sched_init_smp(void) { @@ -6867,6 +6881,7 @@ void __might_sleep(char *file, int line) " context at %s:%d\n", file, line); printk("in_atomic():%d, irqs_disabled():%d\n", in_atomic(), irqs_disabled()); + debug_show_held_locks(current); dump_stack(); } #endif diff --git a/kernel/signal.c b/kernel/signal.c index df18c167a2a7..ec81defde339 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -23,6 +23,7 @@ #include <linux/ptrace.h> #include <linux/signal.h> #include <linux/capability.h> +#include <linux/freezer.h> #include <asm/param.h> #include <asm/uaccess.h> #include <asm/unistd.h> @@ -33,7 +34,7 @@ * SLAB caches for signal bits. */ -static kmem_cache_t *sigqueue_cachep; +static struct kmem_cache *sigqueue_cachep; /* * In POSIX a signal is sent either to a specific thread (Linux task) @@ -1133,8 +1134,7 @@ int kill_pid_info(int sig, struct siginfo *info, struct pid *pid) return error; } -int -kill_proc_info(int sig, struct siginfo *info, pid_t pid) +static int kill_proc_info(int sig, struct siginfo *info, pid_t pid) { int error; rcu_read_lock(); diff --git a/kernel/softirq.c b/kernel/softirq.c index bf25015dce16..918e52df090e 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -574,8 +574,6 @@ static int __cpuinit cpu_callback(struct notifier_block *nfb, switch (action) { case CPU_UP_PREPARE: - BUG_ON(per_cpu(tasklet_vec, hotcpu).list); - BUG_ON(per_cpu(tasklet_hi_vec, hotcpu).list); p = kthread_create(ksoftirqd, hcpu, "ksoftirqd/%d", hotcpu); if (IS_ERR(p)) { printk("ksoftirqd for %i failed\n", hotcpu); diff --git a/kernel/sys.c b/kernel/sys.c index c87b461de38d..a0c1a29a507f 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1102,14 +1102,14 @@ asmlinkage long sys_setreuid(uid_t ruid, uid_t euid) asmlinkage long sys_setuid(uid_t uid) { int old_euid = current->euid; - int old_ruid, old_suid, new_ruid, new_suid; + int old_ruid, old_suid, new_suid; int retval; retval = security_task_setuid(uid, (uid_t)-1, (uid_t)-1, LSM_SETID_ID); if (retval) return retval; - old_ruid = new_ruid = current->uid; + old_ruid = current->uid; old_suid = current->suid; new_suid = old_suid; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 09e569f4792b..8e9f00fd6d18 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -54,6 +54,7 @@ extern int proc_nr_files(ctl_table *table, int write, struct file *filp, #ifdef CONFIG_X86 #include <asm/nmi.h> +#include <asm/stacktrace.h> #endif #if defined(CONFIG_SYSCTL) @@ -170,7 +171,7 @@ 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 *); static int proc_opensys(struct inode *, struct file *); -struct file_operations proc_sys_file_operations = { +const struct file_operations proc_sys_file_operations = { .open = proc_opensys, .read = proc_readsys, .write = proc_writesys, @@ -707,6 +708,14 @@ static ctl_table kern_table[] = { .mode = 0444, .proc_handler = &proc_dointvec, }, + { + .ctl_name = CTL_UNNUMBERED, + .procname = "kstack_depth_to_print", + .data = &kstack_depth_to_print, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, #endif #if defined(CONFIG_MMU) { @@ -977,17 +986,6 @@ static ctl_table vm_table[] = { .extra1 = &zero, }, #endif -#ifdef CONFIG_SWAP - { - .ctl_name = VM_SWAP_TOKEN_TIMEOUT, - .procname = "swap_token_timeout", - .data = &swap_token_default_timeout, - .maxlen = sizeof(swap_token_default_timeout), - .mode = 0644, - .proc_handler = &proc_dointvec_jiffies, - .strategy = &sysctl_jiffies, - }, -#endif #ifdef CONFIG_NUMA { .ctl_name = VM_ZONE_RECLAIM_MODE, @@ -1886,7 +1884,7 @@ static int __do_proc_dointvec(void *tbl_data, ctl_table *table, p = buf; if (*p == '-' && left > 1) { neg = 1; - left--, p++; + p++; } if (*p < '0' || *p > '9') break; @@ -2137,7 +2135,7 @@ static int __do_proc_doulongvec_minmax(void *data, ctl_table *table, int write, p = buf; if (*p == '-' && left > 1) { neg = 1; - left--, p++; + p++; } if (*p < '0' || *p > '9') break; diff --git a/kernel/taskstats.c b/kernel/taskstats.c index d3d28919d4b4..4c3476fa058d 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -34,7 +34,7 @@ static DEFINE_PER_CPU(__u32, taskstats_seqnum) = { 0 }; static int family_registered; -kmem_cache_t *taskstats_cache; +struct kmem_cache *taskstats_cache; static struct genl_family family = { .id = GENL_ID_GENERATE, @@ -69,7 +69,7 @@ enum actions { }; static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, - void **replyp, size_t size) + size_t size) { struct sk_buff *skb; void *reply; @@ -94,7 +94,6 @@ static int prepare_reply(struct genl_info *info, u8 cmd, struct sk_buff **skbp, } *skbp = skb; - *replyp = reply; return 0; } @@ -119,10 +118,10 @@ static int send_reply(struct sk_buff *skb, pid_t pid) /* * Send taskstats data in @skb to listeners registered for @cpu's exit data */ -static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) +static void send_cpu_listeners(struct sk_buff *skb, + struct listener_list *listeners) { struct genlmsghdr *genlhdr = nlmsg_data((struct nlmsghdr *)skb->data); - struct listener_list *listeners; struct listener *s, *tmp; struct sk_buff *skb_next, *skb_cur = skb; void *reply = genlmsg_data(genlhdr); @@ -135,7 +134,6 @@ static void send_cpu_listeners(struct sk_buff *skb, unsigned int cpu) } rc = 0; - listeners = &per_cpu(listener_array, cpu); down_read(&listeners->sem); list_for_each_entry(s, &listeners->list, list) { skb_next = NULL; @@ -186,6 +184,7 @@ static int fill_pid(pid_t pid, struct task_struct *tsk, } else get_task_struct(tsk); + memset(stats, 0, sizeof(*stats)); /* * Each accounting subsystem adds calls to its functions to * fill in relevant parts of struct taskstsats as follows @@ -228,6 +227,8 @@ static int fill_tgid(pid_t tgid, struct task_struct *first, if (first->signal->stats) memcpy(stats, first->signal->stats, sizeof(*stats)); + else + memset(stats, 0, sizeof(*stats)); tsk = first; do { @@ -344,14 +345,36 @@ static int parse(struct nlattr *na, cpumask_t *mask) return ret; } +static struct taskstats *mk_reply(struct sk_buff *skb, int type, u32 pid) +{ + struct nlattr *na, *ret; + int aggr; + + aggr = (type == TASKSTATS_TYPE_PID) + ? TASKSTATS_TYPE_AGGR_PID + : TASKSTATS_TYPE_AGGR_TGID; + + na = nla_nest_start(skb, aggr); + if (!na) + goto err; + if (nla_put(skb, type, sizeof(pid), &pid) < 0) + goto err; + ret = nla_reserve(skb, TASKSTATS_TYPE_STATS, sizeof(struct taskstats)); + if (!ret) + goto err; + nla_nest_end(skb, na); + + return nla_data(ret); +err: + return NULL; +} + static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) { int rc = 0; struct sk_buff *rep_skb; - struct taskstats stats; - void *reply; + struct taskstats *stats; size_t size; - struct nlattr *na; cpumask_t mask; rc = parse(info->attrs[TASKSTATS_CMD_ATTR_REGISTER_CPUMASK], &mask); @@ -372,83 +395,71 @@ static int taskstats_user_cmd(struct sk_buff *skb, struct genl_info *info) size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); - memset(&stats, 0, sizeof(stats)); - rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); + rc = prepare_reply(info, TASKSTATS_CMD_NEW, &rep_skb, size); if (rc < 0) return rc; + rc = -EINVAL; if (info->attrs[TASKSTATS_CMD_ATTR_PID]) { u32 pid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_PID]); - rc = fill_pid(pid, NULL, &stats); - if (rc < 0) + stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, pid); + if (!stats) goto err; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, pid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - stats); + rc = fill_pid(pid, NULL, stats); + if (rc < 0) + goto err; } else if (info->attrs[TASKSTATS_CMD_ATTR_TGID]) { u32 tgid = nla_get_u32(info->attrs[TASKSTATS_CMD_ATTR_TGID]); - rc = fill_tgid(tgid, NULL, &stats); - if (rc < 0) + stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tgid); + if (!stats) goto err; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, tgid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - stats); - } else { - rc = -EINVAL; + rc = fill_tgid(tgid, NULL, stats); + if (rc < 0) + goto err; + } else goto err; - } - - nla_nest_end(rep_skb, na); return send_reply(rep_skb, info->snd_pid); - -nla_put_failure: - rc = genlmsg_cancel(rep_skb, reply); err: nlmsg_free(rep_skb); return rc; } -void taskstats_exit_alloc(struct taskstats **ptidstats, unsigned int *mycpu) +static struct taskstats *taskstats_tgid_alloc(struct task_struct *tsk) { - struct listener_list *listeners; - struct taskstats *tmp; - /* - * This is the cpu on which the task is exiting currently and will - * be the one for which the exit event is sent, even if the cpu - * on which this function is running changes later. - */ - *mycpu = raw_smp_processor_id(); + struct signal_struct *sig = tsk->signal; + struct taskstats *stats; - *ptidstats = NULL; - tmp = kmem_cache_zalloc(taskstats_cache, SLAB_KERNEL); - if (!tmp) - return; + if (sig->stats || thread_group_empty(tsk)) + goto ret; - listeners = &per_cpu(listener_array, *mycpu); - down_read(&listeners->sem); - if (!list_empty(&listeners->list)) { - *ptidstats = tmp; - tmp = NULL; + /* No problem if kmem_cache_zalloc() fails */ + stats = kmem_cache_zalloc(taskstats_cache, GFP_KERNEL); + + spin_lock_irq(&tsk->sighand->siglock); + if (!sig->stats) { + sig->stats = stats; + stats = NULL; } - up_read(&listeners->sem); - kfree(tmp); + spin_unlock_irq(&tsk->sighand->siglock); + + if (stats) + kmem_cache_free(taskstats_cache, stats); +ret: + return sig->stats; } /* Send pid data out on exit */ -void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, - int group_dead, unsigned int mycpu) +void taskstats_exit(struct task_struct *tsk, int group_dead) { int rc; + struct listener_list *listeners; + struct taskstats *stats; struct sk_buff *rep_skb; - void *reply; size_t size; int is_thread_group; - struct nlattr *na; if (!family_registered) return; @@ -459,7 +470,7 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, size = nla_total_size(sizeof(u32)) + nla_total_size(sizeof(struct taskstats)) + nla_total_size(0); - is_thread_group = (tsk->signal->stats != NULL); + is_thread_group = !!taskstats_tgid_alloc(tsk); if (is_thread_group) { /* PID + STATS + TGID + STATS */ size = 2 * size; @@ -467,49 +478,39 @@ void taskstats_exit_send(struct task_struct *tsk, struct taskstats *tidstats, fill_tgid_exit(tsk); } - if (!tidstats) + listeners = &__raw_get_cpu_var(listener_array); + if (list_empty(&listeners->list)) return; - rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, &reply, size); - if (rc < 0) - goto ret; - - rc = fill_pid(tsk->pid, tsk, tidstats); + rc = prepare_reply(NULL, TASKSTATS_CMD_NEW, &rep_skb, size); if (rc < 0) - goto err_skb; + return; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_PID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_PID, (u32)tsk->pid); - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - *tidstats); - nla_nest_end(rep_skb, na); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_PID, tsk->pid); + if (!stats) + goto err; - if (!is_thread_group) - goto send; + rc = fill_pid(tsk->pid, tsk, stats); + if (rc < 0) + goto err; /* * Doesn't matter if tsk is the leader or the last group member leaving */ - if (!group_dead) + if (!is_thread_group || !group_dead) goto send; - na = nla_nest_start(rep_skb, TASKSTATS_TYPE_AGGR_TGID); - NLA_PUT_U32(rep_skb, TASKSTATS_TYPE_TGID, (u32)tsk->tgid); - /* No locking needed for tsk->signal->stats since group is dead */ - NLA_PUT_TYPE(rep_skb, struct taskstats, TASKSTATS_TYPE_STATS, - *tsk->signal->stats); - nla_nest_end(rep_skb, na); + stats = mk_reply(rep_skb, TASKSTATS_TYPE_TGID, tsk->tgid); + if (!stats) + goto err; + + memcpy(stats, tsk->signal->stats, sizeof(*stats)); send: - send_cpu_listeners(rep_skb, mycpu); + send_cpu_listeners(rep_skb, listeners); return; - -nla_put_failure: - genlmsg_cancel(rep_skb, reply); -err_skb: +err: nlmsg_free(rep_skb); -ret: - return; } static struct genl_ops taskstats_ops = { diff --git a/kernel/unwind.c b/kernel/unwind.c index ed0a21d4a902..09c261329249 100644 --- a/kernel/unwind.c +++ b/kernel/unwind.c @@ -14,11 +14,12 @@ #include <linux/bootmem.h> #include <linux/sort.h> #include <linux/stop_machine.h> +#include <linux/uaccess.h> #include <asm/sections.h> #include <asm/uaccess.h> #include <asm/unaligned.h> -extern char __start_unwind[], __end_unwind[]; +extern const char __start_unwind[], __end_unwind[]; extern const u8 __start_unwind_hdr[], __end_unwind_hdr[]; #define MAX_STACK_DEPTH 8 @@ -94,6 +95,7 @@ static const struct { typedef unsigned long uleb128_t; typedef signed long sleb128_t; +#define sleb128abs __builtin_labs static struct unwind_table { struct { @@ -135,6 +137,17 @@ struct unwind_state { static const struct cfa badCFA = { ARRAY_SIZE(reg_info), 1 }; +static unsigned unwind_debug; +static int __init unwind_debug_setup(char *s) +{ + unwind_debug = simple_strtoul(s, NULL, 0); + return 1; +} +__setup("unwind_debug=", unwind_debug_setup); +#define dprintk(lvl, fmt, args...) \ + ((void)(lvl > unwind_debug \ + || printk(KERN_DEBUG "unwind: " fmt "\n", ##args))) + static struct unwind_table *find_table(unsigned long pc) { struct unwind_table *table; @@ -151,7 +164,9 @@ static struct unwind_table *find_table(unsigned long pc) static unsigned long read_pointer(const u8 **pLoc, const void *end, - signed ptrType); + signed ptrType, + unsigned long text_base, + unsigned long data_base); static void init_unwind_table(struct unwind_table *table, const char *name, @@ -176,10 +191,13 @@ static void init_unwind_table(struct unwind_table *table, /* See if the linker provided table looks valid. */ if (header_size <= 4 || header_start[0] != 1 - || (void *)read_pointer(&ptr, end, header_start[1]) != table_start - || header_start[2] == DW_EH_PE_omit - || read_pointer(&ptr, end, header_start[2]) <= 0 - || header_start[3] == DW_EH_PE_omit) + || (void *)read_pointer(&ptr, end, header_start[1], 0, 0) + != table_start + || !read_pointer(&ptr, end, header_start[2], 0, 0) + || !read_pointer(&ptr, end, header_start[3], 0, + (unsigned long)header_start) + || !read_pointer(&ptr, end, header_start[3], 0, + (unsigned long)header_start)) header_start = NULL; table->hdrsz = header_size; smp_wmb(); @@ -269,7 +287,7 @@ static void __init setup_unwind_table(struct unwind_table *table, ptr = (const u8 *)(fde + 2); if (!read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - ptrType)) + ptrType, 0, 0)) return; ++n; } @@ -279,6 +297,7 @@ static void __init setup_unwind_table(struct unwind_table *table, hdrSize = 4 + sizeof(unsigned long) + sizeof(unsigned int) + 2 * n * sizeof(unsigned long); + dprintk(2, "Binary lookup table size for %s: %lu bytes", table->name, hdrSize); header = alloc(hdrSize); if (!header) return; @@ -303,7 +322,7 @@ static void __init setup_unwind_table(struct unwind_table *table, ptr = (const u8 *)(fde + 2); header->table[n].start = read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - fde_pointer_type(cie)); + fde_pointer_type(cie), 0, 0); header->table[n].fde = (unsigned long)fde; ++n; } @@ -486,7 +505,9 @@ static const u32 *cie_for_fde(const u32 *fde, const struct unwind_table *table) static unsigned long read_pointer(const u8 **pLoc, const void *end, - signed ptrType) + signed ptrType, + unsigned long text_base, + unsigned long data_base) { unsigned long value = 0; union { @@ -498,13 +519,17 @@ static unsigned long read_pointer(const u8 **pLoc, const unsigned long *pul; } ptr; - if (ptrType < 0 || ptrType == DW_EH_PE_omit) + if (ptrType < 0 || ptrType == DW_EH_PE_omit) { + dprintk(1, "Invalid pointer encoding %02X (%p,%p).", ptrType, *pLoc, end); return 0; + } ptr.p8 = *pLoc; switch(ptrType & DW_EH_PE_FORM) { case DW_EH_PE_data2: - if (end < (const void *)(ptr.p16u + 1)) + if (end < (const void *)(ptr.p16u + 1)) { + dprintk(1, "Data16 overrun (%p,%p).", ptr.p8, end); return 0; + } if(ptrType & DW_EH_PE_signed) value = get_unaligned(ptr.p16s++); else @@ -512,8 +537,10 @@ static unsigned long read_pointer(const u8 **pLoc, break; case DW_EH_PE_data4: #ifdef CONFIG_64BIT - if (end < (const void *)(ptr.p32u + 1)) + if (end < (const void *)(ptr.p32u + 1)) { + dprintk(1, "Data32 overrun (%p,%p).", ptr.p8, end); return 0; + } if(ptrType & DW_EH_PE_signed) value = get_unaligned(ptr.p32s++); else @@ -525,8 +552,10 @@ static unsigned long read_pointer(const u8 **pLoc, BUILD_BUG_ON(sizeof(u32) != sizeof(value)); #endif case DW_EH_PE_native: - if (end < (const void *)(ptr.pul + 1)) + if (end < (const void *)(ptr.pul + 1)) { + dprintk(1, "DataUL overrun (%p,%p).", ptr.p8, end); return 0; + } value = get_unaligned(ptr.pul++); break; case DW_EH_PE_leb128: @@ -534,10 +563,14 @@ static unsigned long read_pointer(const u8 **pLoc, value = ptrType & DW_EH_PE_signed ? get_sleb128(&ptr.p8, end) : get_uleb128(&ptr.p8, end); - if ((const void *)ptr.p8 > end) + if ((const void *)ptr.p8 > end) { + dprintk(1, "DataLEB overrun (%p,%p).", ptr.p8, end); return 0; + } break; default: + dprintk(2, "Cannot decode pointer type %02X (%p,%p).", + ptrType, ptr.p8, end); return 0; } switch(ptrType & DW_EH_PE_ADJUST) { @@ -546,12 +579,33 @@ static unsigned long read_pointer(const u8 **pLoc, case DW_EH_PE_pcrel: value += (unsigned long)*pLoc; break; + case DW_EH_PE_textrel: + if (likely(text_base)) { + value += text_base; + break; + } + dprintk(2, "Text-relative encoding %02X (%p,%p), but zero text base.", + ptrType, *pLoc, end); + return 0; + case DW_EH_PE_datarel: + if (likely(data_base)) { + value += data_base; + break; + } + dprintk(2, "Data-relative encoding %02X (%p,%p), but zero data base.", + ptrType, *pLoc, end); + return 0; default: + dprintk(2, "Cannot adjust pointer type %02X (%p,%p).", + ptrType, *pLoc, end); return 0; } if ((ptrType & DW_EH_PE_indirect) - && __get_user(value, (unsigned long *)value)) + && probe_kernel_address((unsigned long *)value, value)) { + dprintk(1, "Cannot read indirect value %lx (%p,%p).", + value, *pLoc, end); return 0; + } *pLoc = ptr.p8; return value; @@ -594,7 +648,8 @@ static signed fde_pointer_type(const u32 *cie) case 'P': { signed ptrType = *ptr++; - if (!read_pointer(&ptr, end, ptrType) || ptr > end) + if (!read_pointer(&ptr, end, ptrType, 0, 0) + || ptr > end) return -1; } break; @@ -654,7 +709,8 @@ static int processCFI(const u8 *start, case DW_CFA_nop: break; case DW_CFA_set_loc: - if ((state->loc = read_pointer(&ptr.p8, end, ptrType)) == 0) + state->loc = read_pointer(&ptr.p8, end, ptrType, 0, 0); + if (state->loc == 0) result = 0; break; case DW_CFA_advance_loc1: @@ -700,8 +756,10 @@ static int processCFI(const u8 *start, state->label = NULL; return 1; } - if (state->stackDepth >= MAX_STACK_DEPTH) + if (state->stackDepth >= MAX_STACK_DEPTH) { + dprintk(1, "State stack overflow (%p,%p).", ptr.p8, end); return 0; + } state->stack[state->stackDepth++] = ptr.p8; break; case DW_CFA_restore_state: @@ -716,8 +774,10 @@ static int processCFI(const u8 *start, result = processCFI(start, end, 0, ptrType, state); state->loc = loc; state->label = label; - } else + } else { + dprintk(1, "State stack underflow (%p,%p).", ptr.p8, end); return 0; + } break; case DW_CFA_def_cfa: state->cfa.reg = get_uleb128(&ptr.p8, end); @@ -749,6 +809,7 @@ static int processCFI(const u8 *start, break; case DW_CFA_GNU_window_save: default: + dprintk(1, "Unrecognized CFI op %02X (%p,%p).", ptr.p8[-1], ptr.p8 - 1, end); result = 0; break; } @@ -764,12 +825,17 @@ static int processCFI(const u8 *start, set_rule(*ptr.p8++ & 0x3f, Nowhere, 0, state); break; } - if (ptr.p8 > end) + if (ptr.p8 > end) { + dprintk(1, "Data overrun (%p,%p).", ptr.p8, end); result = 0; + } if (result && targetLoc != 0 && targetLoc < state->loc) return 1; } + if (result && ptr.p8 < end) + dprintk(1, "Data underrun (%p,%p).", ptr.p8, end); + return result && ptr.p8 == end && (targetLoc == 0 @@ -786,7 +852,7 @@ 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 pc = UNW_PC(frame) - frame->call_frame, sp; unsigned long startLoc = 0, endLoc = 0, cfa; unsigned i; signed ptrType = -1; @@ -813,9 +879,9 @@ int unwind(struct unwind_frame_info *frame) ptr = hdr + 4; end = hdr + table->hdrsz; if (tableSize - && read_pointer(&ptr, end, hdr[1]) + && read_pointer(&ptr, end, hdr[1], 0, 0) == (unsigned long)table->address - && (i = read_pointer(&ptr, end, hdr[2])) > 0 + && (i = read_pointer(&ptr, end, hdr[2], 0, 0)) > 0 && i == (end - ptr) / (2 * tableSize) && !((end - ptr) % (2 * tableSize))) { do { @@ -823,7 +889,8 @@ int unwind(struct unwind_frame_info *frame) startLoc = read_pointer(&cur, cur + tableSize, - hdr[3]); + hdr[3], 0, + (unsigned long)hdr); if (pc < startLoc) i /= 2; else { @@ -834,13 +901,17 @@ int unwind(struct unwind_frame_info *frame) if (i == 1 && (startLoc = read_pointer(&ptr, ptr + tableSize, - hdr[3])) != 0 + hdr[3], 0, + (unsigned long)hdr)) != 0 && pc >= startLoc) fde = (void *)read_pointer(&ptr, ptr + tableSize, - hdr[3]); + hdr[3], 0, + (unsigned long)hdr); } } + if(hdr && !fde) + dprintk(3, "Binary lookup for %lx failed.", pc); if (fde != NULL) { cie = cie_for_fde(fde, table); @@ -851,17 +922,19 @@ int unwind(struct unwind_frame_info *frame) && (ptrType = fde_pointer_type(cie)) >= 0 && read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - ptrType) == startLoc) { + ptrType, 0, 0) == startLoc) { if (!(ptrType & DW_EH_PE_indirect)) ptrType &= DW_EH_PE_FORM|DW_EH_PE_signed; endLoc = startLoc + read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - ptrType); + ptrType, 0, 0); if(pc >= endLoc) fde = NULL; } else fde = NULL; + if(!fde) + dprintk(1, "Binary lookup result for %lx discarded.", pc); } if (fde == NULL) { for (fde = table->address, tableSize = table->size; @@ -881,7 +954,7 @@ int unwind(struct unwind_frame_info *frame) ptr = (const u8 *)(fde + 2); startLoc = read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - ptrType); + ptrType, 0, 0); if (!startLoc) continue; if (!(ptrType & DW_EH_PE_indirect)) @@ -889,10 +962,12 @@ int unwind(struct unwind_frame_info *frame) endLoc = startLoc + read_pointer(&ptr, (const u8 *)(fde + 1) + *fde, - ptrType); + ptrType, 0, 0); if (pc >= startLoc && pc < endLoc) break; } + if(!fde) + dprintk(3, "Linear lookup for %lx failed.", pc); } } if (cie != NULL) { @@ -926,6 +1001,8 @@ int unwind(struct unwind_frame_info *frame) if (ptr >= end || *ptr) cie = NULL; } + if(!cie) + dprintk(1, "CIE unusable (%p,%p).", ptr, end); ++ptr; } if (cie != NULL) { @@ -935,7 +1012,12 @@ int unwind(struct unwind_frame_info *frame) state.dataAlign = get_sleb128(&ptr, end); if (state.codeAlign == 0 || state.dataAlign == 0 || ptr >= end) cie = NULL; - else { + else if (UNW_PC(frame) % state.codeAlign + || UNW_SP(frame) % sleb128abs(state.dataAlign)) { + dprintk(1, "Input pointer(s) misaligned (%lx,%lx).", + UNW_PC(frame), UNW_SP(frame)); + return -EPERM; + } else { retAddrReg = state.version <= 1 ? *ptr++ : get_uleb128(&ptr, end); /* skip augmentation */ if (((const char *)(cie + 2))[1] == 'z') { @@ -949,6 +1031,8 @@ int unwind(struct unwind_frame_info *frame) || reg_info[retAddrReg].width != sizeof(unsigned long)) cie = NULL; } + if(!cie) + dprintk(1, "CIE validation failed (%p,%p).", ptr, end); } if (cie != NULL) { state.cieStart = ptr; @@ -962,11 +1046,15 @@ int unwind(struct unwind_frame_info *frame) if ((ptr += augSize) > end) fde = NULL; } + if(!fde) + dprintk(1, "FDE validation failed (%p,%p).", ptr, end); } if (cie == NULL || fde == NULL) { #ifdef CONFIG_FRAME_POINTER unsigned long top, bottom; + if ((UNW_SP(frame) | UNW_FP(frame)) % sizeof(unsigned long)) + return -EPERM; top = STACK_TOP(frame->task); bottom = STACK_BOTTOM(frame->task); # if FRAME_RETADDR_OFFSET < 0 @@ -982,18 +1070,19 @@ int unwind(struct unwind_frame_info *frame) & (sizeof(unsigned long) - 1))) { unsigned long link; - if (!__get_user(link, + if (!probe_kernel_address( (unsigned long *)(UNW_FP(frame) - + FRAME_LINK_OFFSET)) + + FRAME_LINK_OFFSET), + link) # if FRAME_RETADDR_OFFSET < 0 && link > bottom && link < UNW_FP(frame) # else && link > UNW_FP(frame) && link < bottom # endif && !(link & (sizeof(link) - 1)) - && !__get_user(UNW_PC(frame), + && !probe_kernel_address( (unsigned long *)(UNW_FP(frame) - + FRAME_RETADDR_OFFSET))) { + + FRAME_RETADDR_OFFSET), UNW_PC(frame))) { UNW_SP(frame) = UNW_FP(frame) + FRAME_RETADDR_OFFSET # if FRAME_RETADDR_OFFSET < 0 - @@ -1016,8 +1105,11 @@ int unwind(struct unwind_frame_info *frame) || state.regs[retAddrReg].where == Nowhere || state.cfa.reg >= ARRAY_SIZE(reg_info) || reg_info[state.cfa.reg].width != sizeof(unsigned long) - || state.cfa.offs % sizeof(unsigned long)) + || FRAME_REG(state.cfa.reg, unsigned long) % sizeof(unsigned long) + || state.cfa.offs % sizeof(unsigned long)) { + dprintk(1, "Unusable unwind info (%p,%p).", ptr, end); return -EIO; + } /* update frame */ #ifndef CONFIG_AS_CFI_SIGNAL_FRAME if(frame->call_frame @@ -1036,10 +1128,14 @@ int unwind(struct unwind_frame_info *frame) #else # define CASES CASE(8); CASE(16); CASE(32); CASE(64) #endif + pc = UNW_PC(frame); + sp = UNW_SP(frame); for (i = 0; i < ARRAY_SIZE(state.regs); ++i) { if (REG_INVALID(i)) { if (state.regs[i].where == Nowhere) continue; + dprintk(1, "Cannot restore register %u (%d).", + i, state.regs[i].where); return -EIO; } switch(state.regs[i].where) { @@ -1048,8 +1144,11 @@ int unwind(struct unwind_frame_info *frame) case Register: if (state.regs[i].value >= ARRAY_SIZE(reg_info) || REG_INVALID(state.regs[i].value) - || reg_info[i].width > reg_info[state.regs[i].value].width) + || reg_info[i].width > reg_info[state.regs[i].value].width) { + dprintk(1, "Cannot restore register %u from register %lu.", + i, state.regs[i].value); return -EIO; + } switch(reg_info[state.regs[i].value].width) { #define CASE(n) \ case sizeof(u##n): \ @@ -1059,6 +1158,9 @@ int unwind(struct unwind_frame_info *frame) CASES; #undef CASE default: + dprintk(1, "Unsupported register size %u (%lu).", + reg_info[state.regs[i].value].width, + state.regs[i].value); return -EIO; } break; @@ -1083,12 +1185,17 @@ int unwind(struct unwind_frame_info *frame) CASES; #undef CASE default: + dprintk(1, "Unsupported register size %u (%u).", + reg_info[i].width, i); return -EIO; } break; case Value: - if (reg_info[i].width != sizeof(unsigned long)) + if (reg_info[i].width != sizeof(unsigned long)) { + dprintk(1, "Unsupported value size %u (%u).", + reg_info[i].width, i); return -EIO; + } FRAME_REG(i, unsigned long) = cfa + state.regs[i].value * state.dataAlign; break; @@ -1100,15 +1207,20 @@ int unwind(struct unwind_frame_info *frame) % sizeof(unsigned long) || addr < startLoc || addr + sizeof(unsigned long) < addr - || addr + sizeof(unsigned long) > endLoc) + || addr + sizeof(unsigned long) > endLoc) { + dprintk(1, "Bad memory location %lx (%lx).", + addr, state.regs[i].value); return -EIO; + } switch(reg_info[i].width) { #define CASE(n) case sizeof(u##n): \ - __get_user(FRAME_REG(i, u##n), (u##n *)addr); \ + probe_kernel_address((u##n *)addr, FRAME_REG(i, u##n)); \ break CASES; #undef CASE default: + dprintk(1, "Unsupported memory size %u (%u).", + reg_info[i].width, i); return -EIO; } } @@ -1116,6 +1228,17 @@ int unwind(struct unwind_frame_info *frame) } } + if (UNW_PC(frame) % state.codeAlign + || UNW_SP(frame) % sleb128abs(state.dataAlign)) { + dprintk(1, "Output pointer(s) misaligned (%lx,%lx).", + UNW_PC(frame), UNW_SP(frame)); + return -EIO; + } + if (pc == UNW_PC(frame) && sp == UNW_SP(frame)) { + dprintk(1, "No progress (%lx,%lx).", pc, sp); + return -EIO; + } + return 0; #undef CASES #undef FRAME_REG diff --git a/kernel/user.c b/kernel/user.c index 220e586127a0..4869563080e9 100644 --- a/kernel/user.c +++ b/kernel/user.c @@ -26,7 +26,7 @@ #define __uidhashfn(uid) (((uid >> UIDHASH_BITS) + uid) & UIDHASH_MASK) #define uidhashentry(uid) (uidhash_table + __uidhashfn((uid))) -static kmem_cache_t *uid_cachep; +static struct kmem_cache *uid_cachep; static struct list_head uidhash_table[UIDHASH_SZ]; /* @@ -132,7 +132,7 @@ struct user_struct * alloc_uid(uid_t uid) if (!up) { struct user_struct *new; - new = kmem_cache_alloc(uid_cachep, SLAB_KERNEL); + new = kmem_cache_alloc(uid_cachep, GFP_KERNEL); if (!new) return NULL; new->uid = uid; diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 8d1e7cb8a51a..6b186750e9be 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -29,6 +29,9 @@ #include <linux/kthread.h> #include <linux/hardirq.h> #include <linux/mempolicy.h> +#include <linux/freezer.h> +#include <linux/kallsyms.h> +#include <linux/debug_locks.h> /* * The per-CPU workqueue (if single thread, we always use the first @@ -55,6 +58,8 @@ struct cpu_workqueue_struct { struct task_struct *thread; int run_depth; /* Detect run_workqueue() recursion depth */ + + int freezeable; /* Freeze the thread during suspend */ } ____cacheline_aligned; /* @@ -103,6 +108,79 @@ static inline void *get_wq_data(struct work_struct *work) return (void *) (work->management & WORK_STRUCT_WQ_DATA_MASK); } +static int __run_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) +{ + int ret = 0; + unsigned long flags; + + spin_lock_irqsave(&cwq->lock, flags); + /* + * We need to re-validate the work info after we've gotten + * the cpu_workqueue lock. We can run the work now iff: + * + * - the wq_data still matches the cpu_workqueue_struct + * - AND the work is still marked pending + * - AND the work is still on a list (which will be this + * workqueue_struct list) + * + * All these conditions are important, because we + * need to protect against the work being run right + * now on another CPU (all but the last one might be + * true if it's currently running and has not been + * released yet, for example). + */ + if (get_wq_data(work) == cwq + && work_pending(work) + && !list_empty(&work->entry)) { + work_func_t f = work->func; + list_del_init(&work->entry); + spin_unlock_irqrestore(&cwq->lock, flags); + + if (!test_bit(WORK_STRUCT_NOAUTOREL, &work->management)) + work_release(work); + f(work); + + spin_lock_irqsave(&cwq->lock, flags); + cwq->remove_sequence++; + wake_up(&cwq->work_done); + ret = 1; + } + spin_unlock_irqrestore(&cwq->lock, flags); + return ret; +} + +/** + * run_scheduled_work - run scheduled work synchronously + * @work: work to run + * + * This checks if the work was pending, and runs it + * synchronously if so. It returns a boolean to indicate + * whether it had any scheduled work to run or not. + * + * NOTE! This _only_ works for normal work_structs. You + * CANNOT use this for delayed work, because the wq data + * for delayed work will not point properly to the per- + * CPU workqueue struct, but will change! + */ +int fastcall run_scheduled_work(struct work_struct *work) +{ + for (;;) { + struct cpu_workqueue_struct *cwq; + + if (!work_pending(work)) + return 0; + if (list_empty(&work->entry)) + return 0; + /* NOTE! This depends intimately on __queue_work! */ + cwq = get_wq_data(work); + if (!cwq) + return 0; + if (__run_work(cwq, work)) + return 1; + } +} +EXPORT_SYMBOL(run_scheduled_work); + /* Preempt must be disabled. */ static void __queue_work(struct cpu_workqueue_struct *cwq, struct work_struct *work) @@ -250,6 +328,17 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) work_release(work); f(work); + if (unlikely(in_atomic() || lockdep_depth(current) > 0)) { + printk(KERN_ERR "BUG: workqueue leaked lock or atomic: " + "%s/0x%08x/%d\n", + current->comm, preempt_count(), + current->pid); + printk(KERN_ERR " last function: "); + print_symbol("%s\n", (unsigned long)f); + debug_show_held_locks(current); + dump_stack(); + } + spin_lock_irqsave(&cwq->lock, flags); cwq->remove_sequence++; wake_up(&cwq->work_done); @@ -265,7 +354,8 @@ static int worker_thread(void *__cwq) struct k_sigaction sa; sigset_t blocked; - current->flags |= PF_NOFREEZE; + if (!cwq->freezeable) + current->flags |= PF_NOFREEZE; set_user_nice(current, -5); @@ -288,6 +378,9 @@ static int worker_thread(void *__cwq) set_current_state(TASK_INTERRUPTIBLE); while (!kthread_should_stop()) { + if (cwq->freezeable) + try_to_freeze(); + add_wait_queue(&cwq->more_work, &wait); if (list_empty(&cwq->worklist)) schedule(); @@ -364,7 +457,7 @@ void fastcall flush_workqueue(struct workqueue_struct *wq) EXPORT_SYMBOL_GPL(flush_workqueue); static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, - int cpu) + int cpu, int freezeable) { struct cpu_workqueue_struct *cwq = per_cpu_ptr(wq->cpu_wq, cpu); struct task_struct *p; @@ -374,6 +467,7 @@ static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, cwq->thread = NULL; cwq->insert_sequence = 0; cwq->remove_sequence = 0; + cwq->freezeable = freezeable; INIT_LIST_HEAD(&cwq->worklist); init_waitqueue_head(&cwq->more_work); init_waitqueue_head(&cwq->work_done); @@ -389,7 +483,7 @@ static struct task_struct *create_workqueue_thread(struct workqueue_struct *wq, } struct workqueue_struct *__create_workqueue(const char *name, - int singlethread) + int singlethread, int freezeable) { int cpu, destroy = 0; struct workqueue_struct *wq; @@ -409,7 +503,7 @@ struct workqueue_struct *__create_workqueue(const char *name, mutex_lock(&workqueue_mutex); if (singlethread) { INIT_LIST_HEAD(&wq->list); - p = create_workqueue_thread(wq, singlethread_cpu); + p = create_workqueue_thread(wq, singlethread_cpu, freezeable); if (!p) destroy = 1; else @@ -417,7 +511,7 @@ struct workqueue_struct *__create_workqueue(const char *name, } else { list_add(&wq->list, &workqueues); for_each_online_cpu(cpu) { - p = create_workqueue_thread(wq, cpu); + p = create_workqueue_thread(wq, cpu, freezeable); if (p) { kthread_bind(p, cpu); wake_up_process(p); @@ -634,7 +728,6 @@ int current_is_keventd(void) } -#ifdef CONFIG_HOTPLUG_CPU /* Take the work from this (downed) CPU. */ static void take_over_work(struct workqueue_struct *wq, unsigned int cpu) { @@ -667,7 +760,7 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, mutex_lock(&workqueue_mutex); /* Create a new workqueue thread for it. */ list_for_each_entry(wq, &workqueues, list) { - if (!create_workqueue_thread(wq, hotcpu)) { + if (!create_workqueue_thread(wq, hotcpu, 0)) { printk("workqueue for %i failed\n", hotcpu); return NOTIFY_BAD; } @@ -717,7 +810,6 @@ static int __devinit workqueue_cpu_callback(struct notifier_block *nfb, return NOTIFY_OK; } -#endif void init_workqueues(void) { |
