diff options
Diffstat (limited to 'kernel')
40 files changed, 1060 insertions, 629 deletions
diff --git a/kernel/Kconfig.hz b/kernel/Kconfig.hz index 382dd5a8b2d..94fabd534b0 100644 --- a/kernel/Kconfig.hz +++ b/kernel/Kconfig.hz @@ -55,4 +55,4 @@ config HZ default 1000 if HZ_1000 config SCHED_HRTICK - def_bool HIGH_RES_TIMERS && USE_GENERIC_SMP_HELPERS + def_bool HIGH_RES_TIMERS && (!SMP || USE_GENERIC_SMP_HELPERS) diff --git a/kernel/Makefile b/kernel/Makefile index 54f69837d35..4e1d7df7c3e 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -84,6 +84,7 @@ obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o obj-$(CONFIG_TASKSTATS) += taskstats.o tsacct.o obj-$(CONFIG_MARKERS) += marker.o obj-$(CONFIG_LATENCYTOP) += latencytop.o +obj-$(CONFIG_HAVE_GENERIC_DMA_COHERENT) += dma-coherent.o obj-$(CONFIG_FTRACE) += trace/ obj-$(CONFIG_TRACING) += trace/ obj-$(CONFIG_SMP) += sched_cpupri.o diff --git a/kernel/audit.c b/kernel/audit.c index e092f1c0ce3..4414e93d875 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -707,12 +707,14 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) if (status_get->mask & AUDIT_STATUS_ENABLED) { err = audit_set_enabled(status_get->enabled, loginuid, sessionid, sid); - if (err < 0) return err; + if (err < 0) + return err; } if (status_get->mask & AUDIT_STATUS_FAILURE) { err = audit_set_failure(status_get->failure, loginuid, sessionid, sid); - if (err < 0) return err; + if (err < 0) + return err; } if (status_get->mask & AUDIT_STATUS_PID) { int new_pid = status_get->pid; @@ -725,9 +727,12 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh) audit_pid = new_pid; audit_nlk_pid = NETLINK_CB(skb).pid; } - if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) + if (status_get->mask & AUDIT_STATUS_RATE_LIMIT) { err = audit_set_rate_limit(status_get->rate_limit, loginuid, sessionid, sid); + if (err < 0) + return err; + } if (status_get->mask & AUDIT_STATUS_BACKLOG_LIMIT) err = audit_set_backlog_limit(status_get->backlog_limit, loginuid, sessionid, sid); @@ -1366,7 +1371,7 @@ int audit_string_contains_control(const char *string, size_t len) { const unsigned char *p; for (p = string; p < (const unsigned char *)string + len && *p; p++) { - if (*p == '"' || *p < 0x21 || *p > 0x7f) + if (*p == '"' || *p < 0x21 || *p > 0x7e) return 1; } return 0; diff --git a/kernel/auditfilter.c b/kernel/auditfilter.c index 98c50cc671b..b7d354e2b0e 100644 --- a/kernel/auditfilter.c +++ b/kernel/auditfilter.c @@ -1022,8 +1022,11 @@ static void audit_update_watch(struct audit_parent *parent, struct audit_buffer *ab; ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE); + audit_log_format(ab, "auid=%u ses=%u", + audit_get_loginuid(current), + audit_get_sessionid(current)); audit_log_format(ab, - "op=updated rules specifying path="); + " op=updated rules specifying path="); audit_log_untrustedstring(ab, owatch->path); audit_log_format(ab, " with dev=%u ino=%lu\n", dev, ino); @@ -1058,7 +1061,10 @@ static void audit_remove_parent_watches(struct audit_parent *parent) struct audit_buffer *ab; ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_CONFIG_CHANGE); - audit_log_format(ab, "op=remove rule path="); + audit_log_format(ab, "auid=%u ses=%u", + audit_get_loginuid(current), + audit_get_sessionid(current)); + audit_log_format(ab, " op=remove rule path="); audit_log_untrustedstring(ab, w->path); if (r->filterkey) { audit_log_format(ab, " key="); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index 4699950e65b..972f8e61d36 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -243,6 +243,9 @@ static inline int open_arg(int flags, int mask) static int audit_match_perm(struct audit_context *ctx, int mask) { + if (unlikely(!ctx)) + return 0; + unsigned n = ctx->major; switch (audit_classify_syscall(ctx->arch, n)) { case 0: /* native */ @@ -284,6 +287,10 @@ static int audit_match_filetype(struct audit_context *ctx, int which) { unsigned index = which & ~S_IFMT; mode_t mode = which & S_IFMT; + + if (unlikely(!ctx)) + return 0; + if (index >= ctx->name_count) return 0; if (ctx->names[index].ino == -1) @@ -610,7 +617,7 @@ static int audit_filter_rules(struct task_struct *tsk, if (!result) return 0; } - if (rule->filterkey) + if (rule->filterkey && ctx) ctx->filterkey = kstrdup(rule->filterkey, GFP_ATOMIC); switch (rule->action) { case AUDIT_NEVER: *state = AUDIT_DISABLED; break; @@ -2375,7 +2382,7 @@ int __audit_signal_info(int sig, struct task_struct *t) struct audit_context *ctx = tsk->audit_context; if (audit_pid && t->tgid == audit_pid) { - if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1) { + if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) { audit_sig_pid = tsk->pid; if (tsk->loginuid != -1) audit_sig_uid = tsk->loginuid; diff --git a/kernel/cgroup.c b/kernel/cgroup.c index 657f8f8d93a..13932abde15 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -355,6 +355,17 @@ static struct css_set *find_existing_css_set( return NULL; } +static void free_cg_links(struct list_head *tmp) +{ + struct cg_cgroup_link *link; + struct cg_cgroup_link *saved_link; + + list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) { + list_del(&link->cgrp_link_list); + kfree(link); + } +} + /* * allocate_cg_links() allocates "count" cg_cgroup_link structures * and chains them on tmp through their cgrp_link_list fields. Returns 0 on @@ -363,17 +374,12 @@ static struct css_set *find_existing_css_set( static int allocate_cg_links(int count, struct list_head *tmp) { struct cg_cgroup_link *link; - struct cg_cgroup_link *saved_link; int i; INIT_LIST_HEAD(tmp); for (i = 0; i < count; i++) { link = kmalloc(sizeof(*link), GFP_KERNEL); if (!link) { - list_for_each_entry_safe(link, saved_link, tmp, - cgrp_link_list) { - list_del(&link->cgrp_link_list); - kfree(link); - } + free_cg_links(tmp); return -ENOMEM; } list_add(&link->cgrp_link_list, tmp); @@ -381,17 +387,6 @@ static int allocate_cg_links(int count, struct list_head *tmp) return 0; } -static void free_cg_links(struct list_head *tmp) -{ - struct cg_cgroup_link *link; - struct cg_cgroup_link *saved_link; - - list_for_each_entry_safe(link, saved_link, tmp, cgrp_link_list) { - list_del(&link->cgrp_link_list); - kfree(link); - } -} - /* * find_css_set() takes an existing cgroup group and a * cgroup object, and returns a css_set object that's @@ -956,7 +951,6 @@ static int cgroup_get_sb(struct file_system_type *fs_type, struct super_block *sb; struct cgroupfs_root *root; struct list_head tmp_cg_links; - INIT_LIST_HEAD(&tmp_cg_links); /* First find the desired set of subsystems */ ret = parse_cgroupfs_options(data, &opts); @@ -1424,14 +1418,17 @@ static ssize_t cgroup_write_string(struct cgroup *cgrp, struct cftype *cft, if (buffer == NULL) return -ENOMEM; } - if (nbytes && copy_from_user(buffer, userbuf, nbytes)) - return -EFAULT; + if (nbytes && copy_from_user(buffer, userbuf, nbytes)) { + retval = -EFAULT; + goto out; + } buffer[nbytes] = 0; /* nul-terminate */ strstrip(buffer); retval = cft->write_string(cgrp, cft, buffer); if (!retval) retval = nbytes; +out: if (buffer != local_buffer) kfree(buffer); return retval; @@ -2371,7 +2368,7 @@ static int cgroup_mkdir(struct inode *dir, struct dentry *dentry, int mode) return cgroup_create(c_parent, dentry, mode | S_IFDIR); } -static inline int cgroup_has_css_refs(struct cgroup *cgrp) +static int cgroup_has_css_refs(struct cgroup *cgrp) { /* Check the reference count on each subsystem. Since we * already established that there are no tasks in the diff --git a/kernel/cpu.c b/kernel/cpu.c index 10ba5f1004a..f17e9854c24 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -216,7 +216,6 @@ static int __ref take_cpu_down(void *_param) static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) { int err, nr_calls = 0; - struct task_struct *p; cpumask_t old_allowed, tmp; void *hcpu = (void *)(long)cpu; unsigned long mod = tasks_frozen ? CPU_TASKS_FROZEN : 0; @@ -249,21 +248,18 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) cpus_setall(tmp); cpu_clear(cpu, tmp); set_cpus_allowed_ptr(current, &tmp); + tmp = cpumask_of_cpu(cpu); - p = __stop_machine_run(take_cpu_down, &tcd_param, cpu); - - if (IS_ERR(p) || cpu_online(cpu)) { + err = __stop_machine(take_cpu_down, &tcd_param, &tmp); + if (err) { /* CPU didn't die: tell everyone. Can't complain. */ if (raw_notifier_call_chain(&cpu_chain, CPU_DOWN_FAILED | mod, hcpu) == NOTIFY_BAD) BUG(); - if (IS_ERR(p)) { - err = PTR_ERR(p); - goto out_allowed; - } - goto out_thread; + goto out_allowed; } + BUG_ON(cpu_online(cpu)); /* Wait for it to sleep (leaving idle task). */ while (!idle_cpu(cpu)) @@ -279,8 +275,6 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen) check_for_tasks(cpu); -out_thread: - err = kthread_stop(p); out_allowed: set_cpus_allowed_ptr(current, &old_allowed); out_release: @@ -355,6 +349,8 @@ static int __cpuinit _cpu_up(unsigned int cpu, int tasks_frozen) goto out_notify; BUG_ON(!cpu_online(cpu)); + cpu_set(cpu, cpu_active_map); + /* Now call notifier in preparation. */ raw_notifier_call_chain(&cpu_chain, CPU_ONLINE | mod, hcpu); @@ -373,7 +369,7 @@ int __cpuinit cpu_up(unsigned int cpu) if (!cpu_isset(cpu, cpu_possible_map)) { printk(KERN_ERR "can't online cpu %d because it is not " "configured as may-hotadd at boot time\n", cpu); -#if defined(CONFIG_IA64) || defined(CONFIG_X86_64) || defined(CONFIG_S390) +#if defined(CONFIG_IA64) || defined(CONFIG_X86_64) printk(KERN_ERR "please check additional_cpus= boot " "parameter\n"); #endif @@ -389,9 +385,6 @@ int __cpuinit cpu_up(unsigned int cpu) err = _cpu_up(cpu, 0); - if (cpu_online(cpu)) - cpu_set(cpu, cpu_active_map); - out: cpu_maps_update_done(); return err; @@ -461,3 +454,28 @@ out: #endif /* CONFIG_PM_SLEEP_SMP */ #endif /* CONFIG_SMP */ + +/* + * cpu_bit_bitmap[] is a special, "compressed" data structure that + * represents all NR_CPUS bits binary values of 1<<nr. + * + * It is used by cpumask_of_cpu() to get a constant address to a CPU + * mask value that has a single bit set only. + */ + +/* cpu_bit_bitmap[0] is empty - so we can back into it */ +#define MASK_DECLARE_1(x) [x+1][0] = 1UL << (x) +#define MASK_DECLARE_2(x) MASK_DECLARE_1(x), MASK_DECLARE_1(x+1) +#define MASK_DECLARE_4(x) MASK_DECLARE_2(x), MASK_DECLARE_2(x+2) +#define MASK_DECLARE_8(x) MASK_DECLARE_4(x), MASK_DECLARE_4(x+4) + +const unsigned long cpu_bit_bitmap[BITS_PER_LONG+1][BITS_TO_LONGS(NR_CPUS)] = { + + MASK_DECLARE_8(0), MASK_DECLARE_8(8), + MASK_DECLARE_8(16), MASK_DECLARE_8(24), +#if BITS_PER_LONG > 32 + MASK_DECLARE_8(32), MASK_DECLARE_8(40), + MASK_DECLARE_8(48), MASK_DECLARE_8(56), +#endif +}; +EXPORT_SYMBOL_GPL(cpu_bit_bitmap); diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 91cf85b36dd..d5ab79cf516 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -54,7 +54,6 @@ #include <asm/uaccess.h> #include <asm/atomic.h> #include <linux/mutex.h> -#include <linux/kfifo.h> #include <linux/workqueue.h> #include <linux/cgroup.h> @@ -486,13 +485,38 @@ static int cpusets_overlap(struct cpuset *a, struct cpuset *b) static void update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) { - if (!dattr) - return; if (dattr->relax_domain_level < c->relax_domain_level) dattr->relax_domain_level = c->relax_domain_level; return; } +static void +update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) +{ + LIST_HEAD(q); + + list_add(&c->stack_list, &q); + while (!list_empty(&q)) { + struct cpuset *cp; + struct cgroup *cont; + struct cpuset *child; + + cp = list_first_entry(&q, struct cpuset, stack_list); + list_del(q.next); + + if (cpus_empty(cp->cpus_allowed)) + continue; + + if (is_sched_load_balance(cp)) + update_domain_attr(dattr, cp); + + list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { + child = cgroup_cs(cont); + list_add_tail(&child->stack_list, &q); + } + } +} + /* * rebuild_sched_domains() * @@ -532,7 +556,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) * So the reverse nesting would risk an ABBA deadlock. * * The three key local variables below are: - * q - a kfifo queue of cpuset pointers, used to implement a + * q - a linked-list queue of cpuset pointers, used to implement a * top-down scan of all cpusets. This scan loads a pointer * to each cpuset marked is_sched_load_balance into the * array 'csa'. For our purposes, rebuilding the schedulers @@ -567,7 +591,7 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) void rebuild_sched_domains(void) { - struct kfifo *q; /* queue of cpusets to be scanned */ + LIST_HEAD(q); /* queue of cpusets to be scanned*/ struct cpuset *cp; /* scans q */ struct cpuset **csa; /* array of all cpuset ptrs */ int csn; /* how many cpuset ptrs in csa so far */ @@ -577,7 +601,6 @@ void rebuild_sched_domains(void) int ndoms; /* number of sched domains in result */ int nslot; /* next empty doms[] cpumask_t slot */ - q = NULL; csa = NULL; doms = NULL; dattr = NULL; @@ -591,35 +614,42 @@ void rebuild_sched_domains(void) dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL); if (dattr) { *dattr = SD_ATTR_INIT; - update_domain_attr(dattr, &top_cpuset); + update_domain_attr_tree(dattr, &top_cpuset); } *doms = top_cpuset.cpus_allowed; goto rebuild; } - q = kfifo_alloc(number_of_cpusets * sizeof(cp), GFP_KERNEL, NULL); - if (IS_ERR(q)) - goto done; csa = kmalloc(number_of_cpusets * sizeof(cp), GFP_KERNEL); if (!csa) goto done; csn = 0; - cp = &top_cpuset; - __kfifo_put(q, (void *)&cp, sizeof(cp)); - while (__kfifo_get(q, (void *)&cp, sizeof(cp))) { + list_add(&top_cpuset.stack_list, &q); + while (!list_empty(&q)) { struct cgroup *cont; struct cpuset *child; /* scans child cpusets of cp */ + cp = list_first_entry(&q, struct cpuset, stack_list); + list_del(q.next); + if (cpus_empty(cp->cpus_allowed)) continue; - if (is_sched_load_balance(cp)) + /* + * All child cpusets contain a subset of the parent's cpus, so + * just skip them, and then we call update_domain_attr_tree() + * to calc relax_domain_level of the corresponding sched + * domain. + */ + if (is_sched_load_balance(cp)) { csa[csn++] = cp; + continue; + } list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { child = cgroup_cs(cont); - __kfifo_put(q, (void *)&child, sizeof(cp)); + list_add_tail(&child->stack_list, &q); } } @@ -686,7 +716,7 @@ restart: cpus_or(*dp, *dp, b->cpus_allowed); b->pn = -1; if (dattr) - update_domain_attr(dattr + update_domain_attr_tree(dattr + nslot, b); } } @@ -702,8 +732,6 @@ rebuild: put_online_cpus(); done: - if (q && !IS_ERR(q)) - kfifo_free(q); kfree(csa); /* Don't kfree(doms) -- partition_sched_domains() does that. */ /* Don't kfree(dattr) -- partition_sched_domains() does that. */ @@ -1833,24 +1861,21 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs) */ static void scan_for_empty_cpusets(const struct cpuset *root) { + LIST_HEAD(queue); struct cpuset *cp; /* scans cpusets being updated */ struct cpuset *child; /* scans child cpusets of cp */ - struct list_head queue; struct cgroup *cont; nodemask_t oldmems; - INIT_LIST_HEAD(&queue); - list_add_tail((struct list_head *)&root->stack_list, &queue); while (!list_empty(&queue)) { - cp = container_of(queue.next, struct cpuset, stack_list); + cp = list_first_entry(&queue, struct cpuset, stack_list); list_del(queue.next); list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { child = cgroup_cs(cont); list_add_tail(&child->stack_list, &queue); } - cont = cp->css.cgroup; /* Continue past cpusets with all cpus, mems online */ if (cpus_subset(cp->cpus_allowed, cpu_online_map) && diff --git a/kernel/dma-coherent.c b/kernel/dma-coherent.c new file mode 100644 index 00000000000..c1d4d5b4c61 --- /dev/null +++ b/kernel/dma-coherent.c @@ -0,0 +1,153 @@ +/* + * Coherent per-device memory handling. + * Borrowed from i386 + */ +#include <linux/kernel.h> +#include <linux/dma-mapping.h> + +struct dma_coherent_mem { + void *virt_base; + u32 device_base; + int size; + int flags; + unsigned long *bitmap; +}; + +int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr, + dma_addr_t device_addr, size_t size, int flags) +{ + void __iomem *mem_base = NULL; + int pages = size >> PAGE_SHIFT; + int bitmap_size = BITS_TO_LONGS(pages) * sizeof(long); + + if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0) + goto out; + if (!size) + goto out; + if (dev->dma_mem) + goto out; + + /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */ + + mem_base = ioremap(bus_addr, size); + if (!mem_base) + goto out; + + dev->dma_mem = kzalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL); + if (!dev->dma_mem) + goto out; + dev->dma_mem->bitmap = kzalloc(bitmap_size, GFP_KERNEL); + if (!dev->dma_mem->bitmap) + goto free1_out; + + dev->dma_mem->virt_base = mem_base; + dev->dma_mem->device_base = device_addr; + dev->dma_mem->size = pages; + dev->dma_mem->flags = flags; + + if (flags & DMA_MEMORY_MAP) + return DMA_MEMORY_MAP; + + return DMA_MEMORY_IO; + + free1_out: + kfree(dev->dma_mem); + out: + if (mem_base) + iounmap(mem_base); + return 0; +} +EXPORT_SYMBOL(dma_declare_coherent_memory); + +void dma_release_declared_memory(struct device *dev) +{ + struct dma_coherent_mem *mem = dev->dma_mem; + + if (!mem) + return; + dev->dma_mem = NULL; + iounmap(mem->virt_base); + kfree(mem->bitmap); + kfree(mem); +} +EXPORT_SYMBOL(dma_release_declared_memory); + +void *dma_mark_declared_memory_occupied(struct device *dev, + dma_addr_t device_addr, size_t size) +{ + struct dma_coherent_mem *mem = dev->dma_mem; + int pos, err; + + size += device_addr & ~PAGE_MASK; + + if (!mem) + return ERR_PTR(-EINVAL); + + pos = (device_addr - mem->device_base) >> PAGE_SHIFT; + err = bitmap_allocate_region(mem->bitmap, pos, get_order(size)); + if (err != 0) + return ERR_PTR(err); + return mem->virt_base + (pos << PAGE_SHIFT); +} +EXPORT_SYMBOL(dma_mark_declared_memory_occupied); + +/** + * dma_alloc_from_coherent() - try to allocate memory from the per-device coherent area + * + * @dev: device from which we allocate memory + * @size: size of requested memory area + * @dma_handle: This will be filled with the correct dma handle + * @ret: This pointer will be filled with the virtual address + * to allocated area. + * + * This function should be only called from per-arch dma_alloc_coherent() + * to support allocation from per-device coherent memory pools. + * + * Returns 0 if dma_alloc_coherent should continue with allocating from + * generic memory areas, or !0 if dma_alloc_coherent should return @ret. + */ +int dma_alloc_from_coherent(struct device *dev, ssize_t size, + dma_addr_t *dma_handle, void **ret) +{ + struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL; + int order = get_order(size); + + if (mem) { + int page = bitmap_find_free_region(mem->bitmap, mem->size, + order); + if (page >= 0) { + *dma_handle = mem->device_base + (page << PAGE_SHIFT); + *ret = mem->virt_base + (page << PAGE_SHIFT); + memset(*ret, 0, size); + } else if (mem->flags & DMA_MEMORY_EXCLUSIVE) + *ret = NULL; + } + return (mem != NULL); +} + +/** + * dma_release_from_coherent() - try to free the memory allocated from per-device coherent memory pool + * @dev: device from which the memory was allocated + * @order: the order of pages allocated + * @vaddr: virtual address of allocated pages + * + * This checks whether the memory was allocated from the per-device + * coherent memory pool and if so, releases that memory. + * + * Returns 1 if we correctly released the memory, or 0 if + * dma_release_coherent() should proceed with releasing memory from + * generic pools. + */ +int dma_release_from_coherent(struct device *dev, int order, void *vaddr) +{ + struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL; + + if (mem && vaddr >= mem->virt_base && vaddr < + (mem->virt_base + (mem->size << PAGE_SHIFT))) { + int page = (vaddr - mem->virt_base) >> PAGE_SHIFT; + + bitmap_release_region(mem->bitmap, page, order); + return 1; + } + return 0; +} diff --git a/kernel/exit.c b/kernel/exit.c index eb4d6470d1d..38ec4063014 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -911,10 +911,10 @@ static void exit_notify(struct task_struct *tsk, int group_dead) tsk->exit_signal = SIGCHLD; signal = tracehook_notify_death(tsk, &cookie, group_dead); - if (signal > 0) + if (signal >= 0) signal = do_notify_parent(tsk, signal); - tsk->exit_state = signal < 0 ? EXIT_DEAD : EXIT_ZOMBIE; + tsk->exit_state = signal == DEATH_REAP ? EXIT_DEAD : EXIT_ZOMBIE; /* mt-exec, de_thread() is waiting for us */ if (thread_group_leader(tsk) && @@ -927,7 +927,7 @@ static void exit_notify(struct task_struct *tsk, int group_dead) tracehook_report_death(tsk, signal, cookie, group_dead); /* If the process is dead, release it - nobody will wait for it */ - if (signal < 0) + if (signal == DEATH_REAP) release_task(tsk); } diff --git a/kernel/fork.c b/kernel/fork.c index 8214ba7c8bb..7ce2ebe8479 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -27,6 +27,7 @@ #include <linux/key.h> #include <linux/binfmts.h> #include <linux/mman.h> +#include <linux/mmu_notifier.h> #include <linux/fs.h> #include <linux/nsproxy.h> #include <linux/capability.h> @@ -414,6 +415,7 @@ static struct mm_struct * mm_init(struct mm_struct * mm, struct task_struct *p) if (likely(!mm_alloc_pgd(mm))) { mm->def_flags = 0; + mmu_notifier_mm_init(mm); return mm; } @@ -446,6 +448,7 @@ void __mmdrop(struct mm_struct *mm) BUG_ON(mm == &init_mm); mm_free_pgd(mm); destroy_context(mm); + mmu_notifier_mm_destroy(mm); free_mm(mm); } EXPORT_SYMBOL_GPL(__mmdrop); diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c index 152abfd3589..0314074fa23 100644 --- a/kernel/irq/manage.c +++ b/kernel/irq/manage.c @@ -323,7 +323,8 @@ static int __irq_set_trigger(struct irq_chip *chip, unsigned int irq, ret = chip->set_type(irq, flags & IRQF_TRIGGER_MASK); if (ret) - pr_err("setting flow type for irq %u failed (%pF)\n", + pr_err("setting trigger mode %d for irq %u failed (%pF)\n", + (int)(flags & IRQF_TRIGGER_MASK), irq, chip->set_type); return ret; diff --git a/kernel/irq/proc.c b/kernel/irq/proc.c index 6c6d35d68ee..a09dd29c2fd 100644 --- a/kernel/irq/proc.c +++ b/kernel/irq/proc.c @@ -8,6 +8,7 @@ #include <linux/irq.h> #include <linux/proc_fs.h> +#include <linux/seq_file.h> #include <linux/interrupt.h> #include "internals.h" @@ -16,23 +17,18 @@ static struct proc_dir_entry *root_irq_dir; #ifdef CONFIG_SMP -static int irq_affinity_read_proc(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int irq_affinity_proc_show(struct seq_file *m, void *v) { - struct irq_desc *desc = irq_desc + (long)data; + struct irq_desc *desc = irq_desc + (long)m->private; cpumask_t *mask = &desc->affinity; - int len; #ifdef CONFIG_GENERIC_PENDING_IRQ if (desc->status & IRQ_MOVE_PENDING) mask = &desc->pending_mask; #endif - len = cpumask_scnprintf(page, count, *mask); - - if (count - len < 2) - return -EINVAL; - len += sprintf(page + len, "\n"); - return len; + seq_cpumask(m, mask); + seq_putc(m, '\n'); + return 0; } #ifndef is_affinity_mask_valid @@ -40,11 +36,12 @@ static int irq_affinity_read_proc(char *page, char **start, off_t off, #endif int no_irq_affinity; -static int irq_affinity_write_proc(struct file *file, const char __user *buffer, - unsigned long count, void *data) +static ssize_t irq_affinity_proc_write(struct file *file, + const char __user *buffer, size_t count, loff_t *pos) { - unsigned int irq = (int)(long)data, full_count = count, err; + unsigned int irq = (int)(long)PDE(file->f_path.dentry->d_inode)->data; cpumask_t new_value; + int err; if (!irq_desc[irq].chip->set_affinity || no_irq_affinity || irq_balancing_disabled(irq)) @@ -65,28 +62,38 @@ static int irq_affinity_write_proc(struct file *file, const char __user *buffer, if (!cpus_intersects(new_value, cpu_online_map)) /* Special case for empty set - allow the architecture code to set default SMP affinity. */ - return irq_select_affinity(irq) ? -EINVAL : full_count; + return irq_select_affinity(irq) ? -EINVAL : count; irq_set_affinity(irq, new_value); - return full_count; + return count; } -static int default_affinity_read(char *page, char **start, off_t off, - int count, int *eof, void *data) +static int irq_affinity_proc_open(struct inode *inode, struct file *file) { - int len = cpumask_scnprintf(page, count, irq_default_affinity); - if (count - len < 2) - return -EINVAL; - len += sprintf(page + len, "\n"); - return len; + return single_open(file, irq_affinity_proc_show, PDE(inode)->data); } -static int default_affinity_write(struct file *file, const char __user *buffer, - unsigned long count, void *data) +static const struct file_operations irq_affinity_proc_fops = { + .open = irq_affinity_proc_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = irq_affinity_proc_write, +}; + +static int default_affinity_show(struct seq_file *m, void *v) +{ + seq_cpumask(m, &irq_default_affinity); + seq_putc(m, '\n'); + return 0; +} + +static ssize_t default_affinity_write(struct file *file, + const char __user *buffer, size_t count, loff_t *ppos) { - unsigned int full_count = count, err; cpumask_t new_value; + int err; err = cpumask_parse_user(buffer, count, new_value); if (err) @@ -105,8 +112,21 @@ static int default_affinity_write(struct file *file, const char __user *buffer, irq_default_affinity = new_value; - return full_count; + return count; } + +static int default_affinity_open(struct inode *inode, struct file *file) +{ + return single_open(file, default_affinity_show, NULL); +} + +static const struct file_operations default_affinity_proc_fops = { + .open = default_affinity_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = default_affinity_write, +}; #endif static int irq_spurious_read(char *page, char **start, off_t off, @@ -178,16 +198,9 @@ void register_irq_proc(unsigned int irq) irq_desc[irq].dir = proc_mkdir(name, root_irq_dir); #ifdef CONFIG_SMP - { - /* create /proc/irq/<irq>/smp_affinity */ - entry = create_proc_entry("smp_affinity", 0600, irq_desc[irq].dir); - - if (entry) { - entry->data = (void *)(long)irq; - entry->read_proc = irq_affinity_read_proc; - entry->write_proc = irq_affinity_write_proc; - } - } + /* create /proc/irq/<irq>/smp_affinity */ + proc_create_data("smp_affinity", 0600, irq_desc[irq].dir, + &irq_affinity_proc_fops, (void *)(long)irq); #endif entry = create_proc_entry("spurious", 0444, irq_desc[irq].dir); @@ -208,15 +221,8 @@ void unregister_handler_proc(unsigned int irq, struct irqaction *action) void register_default_affinity_proc(void) { #ifdef CONFIG_SMP - struct proc_dir_entry *entry; - - /* create /proc/irq/default_smp_affinity */ - entry = create_proc_entry("default_smp_affinity", 0600, root_irq_dir); - if (entry) { - entry->data = NULL; - entry->read_proc = default_affinity_read; - entry->write_proc = default_affinity_write; - } + proc_create("irq/default_smp_affinity", 0600, NULL, + &default_affinity_proc_fops); #endif } diff --git a/kernel/kgdb.c b/kernel/kgdb.c index 3ec23c3ec97..eaa21fc9ad1 100644 --- a/kernel/kgdb.c +++ b/kernel/kgdb.c @@ -56,12 +56,14 @@ static int kgdb_break_asap; +#define KGDB_MAX_THREAD_QUERY 17 struct kgdb_state { int ex_vector; int signo; int err_code; int cpu; int pass_exception; + unsigned long thr_query; unsigned long threadid; long kgdb_usethreadid; struct pt_regs *linux_regs; @@ -166,13 +168,6 @@ early_param("nokgdbroundup", opt_nokgdbroundup); * Weak aliases for breakpoint management, * can be overriden by architectures when needed: */ -int __weak kgdb_validate_break_address(unsigned long addr) -{ - char tmp_variable[BREAK_INSTR_SIZE]; - - return probe_kernel_read(tmp_variable, (char *)addr, BREAK_INSTR_SIZE); -} - int __weak kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr) { int err; @@ -191,6 +186,25 @@ int __weak kgdb_arch_remove_breakpoint(unsigned long addr, char *bundle) (char *)bundle, BREAK_INSTR_SIZE); } +int __weak kgdb_validate_break_address(unsigned long addr) +{ + char tmp_variable[BREAK_INSTR_SIZE]; + int err; + /* Validate setting the breakpoint and then removing it. In the + * remove fails, the kernel needs to emit a bad message because we + * are deep trouble not being able to put things back the way we + * found them. + */ + err = kgdb_arch_set_breakpoint(addr, tmp_variable); + if (err) + return err; + err = kgdb_arch_remove_breakpoint(addr, tmp_variable); + if (err) + printk(KERN_ERR "KGDB: Critical breakpoint error, kernel " + "memory destroyed at: %lx", addr); + return err; +} + unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs) { return instruction_pointer(regs); @@ -433,9 +447,14 @@ int kgdb_hex2long(char **ptr, unsigned long *long_val) { int hex_val; int num = 0; + int negate = 0; *long_val = 0; + if (**ptr == '-') { + negate = 1; + (*ptr)++; + } while (**ptr) { hex_val = hex(**ptr); if (hex_val < 0) @@ -446,6 +465,9 @@ int kgdb_hex2long(char **ptr, unsigned long *long_val) (*ptr)++; } + if (negate) + *long_val = -*long_val; + return num; } @@ -515,10 +537,16 @@ static void int_to_threadref(unsigned char *id, int value) static struct task_struct *getthread(struct pt_regs *regs, int tid) { /* - * Non-positive TIDs are remapped idle tasks: + * Non-positive TIDs are remapped to the cpu shadow information */ - if (tid <= 0) - return idle_task(-tid); + if (tid == 0 || tid == -1) + tid = -atomic_read(&kgdb_active) - 2; + if (tid < 0) { + if (kgdb_info[-tid - 2].task) + return kgdb_info[-tid - 2].task; + else + return idle_task(-tid - 2); + } /* * find_task_by_pid_ns() does not take the tasklist lock anymore @@ -725,14 +753,15 @@ setundefined: } /* - * Remap normal tasks to their real PID, idle tasks to -1 ... -NR_CPUs: + * Remap normal tasks to their real PID, + * CPU shadow threads are mapped to -CPU - 2 */ static inline int shadow_pid(int realpid) { if (realpid) return realpid; - return -1-raw_smp_processor_id(); + return -raw_smp_processor_id() - 2; } static char gdbmsgbuf[BUFMAX + 1]; @@ -826,7 +855,7 @@ static void gdb_cmd_getregs(struct kgdb_state *ks) local_debuggerinfo = kgdb_info[ks->cpu].debuggerinfo; } else { local_debuggerinfo = NULL; - for (i = 0; i < NR_CPUS; i++) { + for_each_online_cpu(i) { /* * Try to find the task on some other * or possibly this node if we do not @@ -960,10 +989,13 @@ static int gdb_cmd_reboot(struct kgdb_state *ks) /* Handle the 'q' query packets */ static void gdb_cmd_query(struct kgdb_state *ks) { - struct task_struct *thread; + struct task_struct *g; + struct task_struct *p; unsigned char thref[8]; char *ptr; int i; + int cpu; + int finished = 0; switch (remcom_in_buffer[1]) { case 's': @@ -973,22 +1005,34 @@ static void gdb_cmd_query(struct kgdb_state *ks) break; } - if (remcom_in_buffer[1] == 'f') - ks->threadid = 1; - + i = 0; remcom_out_buffer[0] = 'm'; ptr = remcom_out_buffer + 1; - - for (i = 0; i < 17; ks->threadid++) { - thread = getthread(ks->linux_regs, ks->threadid); - if (thread) { - int_to_threadref(thref, ks->threadid); + if (remcom_in_buffer[1] == 'f') { + /* Each cpu is a shadow thread */ + for_each_online_cpu(cpu) { + ks->thr_query = 0; + int_to_threadref(thref, -cpu - 2); pack_threadid(ptr, thref); ptr += BUF_THREAD_ID_SIZE; *(ptr++) = ','; i++; } } + + do_each_thread(g, p) { + if (i >= ks->thr_query && !finished) { + int_to_threadref(thref, p->pid); + pack_threadid(ptr, thref); + ptr += BUF_THREAD_ID_SIZE; + *(ptr++) = ','; + ks->thr_query++; + if (ks->thr_query % KGDB_MAX_THREAD_QUERY == 0) + finished = 1; + } + i++; + } while_each_thread(g, p); + *(--ptr) = '\0'; break; @@ -1011,15 +1055,15 @@ static void gdb_cmd_query(struct kgdb_state *ks) error_packet(remcom_out_buffer, -EINVAL); break; } - if (ks->threadid > 0) { + if ((int)ks->threadid > 0) { kgdb_mem2hex(getthread(ks->linux_regs, ks->threadid)->comm, remcom_out_buffer, 16); } else { static char tmpstr[23 + BUF_THREAD_ID_SIZE]; - sprintf(tmpstr, "Shadow task %d for pid 0", - (int)(-ks->threadid-1)); + sprintf(tmpstr, "shadowCPU%d", + (int)(-ks->threadid - 2)); kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr)); } break; diff --git a/kernel/lockdep.c b/kernel/lockdep.c index d38a6436297..1aa91fd6b06 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -124,6 +124,15 @@ static struct lock_list list_entries[MAX_LOCKDEP_ENTRIES]; unsigned long nr_lock_classes; static struct lock_class lock_classes[MAX_LOCKDEP_KEYS]; +static inline struct lock_class *hlock_class(struct held_lock *hlock) +{ + if (!hlock->class_idx) { + DEBUG_LOCKS_WARN_ON(1); + return NULL; + } + return lock_classes + hlock->class_idx - 1; +} + #ifdef CONFIG_LOCK_STAT static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], lock_stats); @@ -222,7 +231,7 @@ static void lock_release_holdtime(struct held_lock *hlock) holdtime = sched_clock() - hlock->holdtime_stamp; - stats = get_lock_stats(hlock->class); + stats = get_lock_stats(hlock_class(hlock)); if (hlock->read) lock_time_inc(&stats->read_holdtime, holdtime); else @@ -372,6 +381,19 @@ unsigned int nr_process_chains; unsigned int max_lockdep_depth; unsigned int max_recursion_depth; +static unsigned int lockdep_dependency_gen_id; + +static bool lockdep_dependency_visit(struct lock_class *source, + unsigned int depth) +{ + if (!depth) + lockdep_dependency_gen_id++; + if (source->dep_gen_id == lockdep_dependency_gen_id) + return true; + source->dep_gen_id = lockdep_dependency_gen_id; + return false; +} + #ifdef CONFIG_DEBUG_LOCKDEP /* * We cannot printk in early bootup code. Not even early_printk() @@ -505,7 +527,7 @@ static void print_lockdep_cache(struct lockdep_map *lock) static void print_lock(struct held_lock *hlock) { - print_lock_name(hlock->class); + print_lock_name(hlock_class(hlock)); printk(", at: "); print_ip_sym(hlock->acquire_ip); } @@ -558,6 +580,9 @@ static void print_lock_dependencies(struct lock_class *class, int depth) { struct lock_list *entry; + if (lockdep_dependency_visit(class, depth)) + return; + if (DEBUG_LOCKS_WARN_ON(depth >= 20)) return; @@ -932,7 +957,7 @@ static noinline int print_circular_bug_tail(void) if (debug_locks_silent) return 0; - this.class = check_source->class; + this.class = hlock_class(check_source); if (!save_trace(&this.trace)) return 0; @@ -959,6 +984,67 @@ static int noinline print_infinite_recursion_bug(void) return 0; } +unsigned long __lockdep_count_forward_deps(struct lock_class *class, + unsigned int depth) +{ + struct lock_list *entry; + unsigned long ret = 1; + + if (lockdep_dependency_visit(class, depth)) + return 0; + + /* + * Recurse this class's dependency list: + */ + list_for_each_entry(entry, &class->locks_after, entry) + ret += __lockdep_count_forward_deps(entry->class, depth + 1); + + return ret; +} + +unsigned long lockdep_count_forward_deps(struct lock_class *class) +{ + unsigned long ret, flags; + + local_irq_save(flags); + __raw_spin_lock(&lockdep_lock); + ret = __lockdep_count_forward_deps(class, 0); + __raw_spin_unlock(&lockdep_lock); + local_irq_restore(flags); + + return ret; +} + +unsigned long __lockdep_count_backward_deps(struct lock_class *class, + unsigned int depth) +{ + struct lock_list *entry; + unsigned long ret = 1; + + if (lockdep_dependency_visit(class, depth)) + return 0; + /* + * Recurse this class's dependency list: + */ + list_for_each_entry(entry, &class->locks_before, entry) + ret += __lockdep_count_backward_deps(entry->class, depth + 1); + + return ret; +} + +unsigned long lockdep_count_backward_deps(struct lock_class *class) +{ + unsigned long ret, flags; + + local_irq_save(flags); + __raw_spin_lock(&lockdep_lock); + ret = __lockdep_count_backward_deps(class, 0); + __raw_spin_unlock(&lockdep_lock); + local_irq_restore(flags); + + return ret; +} + /* * Prove that the dependency graph starting at <entry> can not * lead to <target>. Print an error and return 0 if it does. @@ -968,6 +1054,9 @@ check_noncircular(struct lock_class *source, unsigned int depth) { struct lock_list *entry; + if (lockdep_dependency_visit(source, depth)) + return 1; + debug_atomic_inc(&nr_cyclic_check_recursions); if (depth > max_recursion_depth) max_recursion_depth = depth; @@ -977,7 +1066,7 @@ check_noncircular(struct lock_class *source, unsigned int depth) * Check this lock's dependency list: */ list_for_each_entry(entry, &source->locks_after, entry) { - if (entry->class == check_target->class) + if (entry->class == hlock_class(check_target)) return print_circular_bug_header(entry, depth+1); debug_atomic_inc(&nr_cyclic_checks); if (!check_noncircular(entry->class, depth+1)) @@ -1011,6 +1100,9 @@ find_usage_forwards(struct lock_class *source, unsigned int depth) struct lock_list *entry; int ret; + if (lockdep_dependency_visit(source, depth)) + return 1; + if (depth > max_recursion_depth) max_recursion_depth = depth; if (depth >= RECURSION_LIMIT) @@ -1050,6 +1142,9 @@ find_usage_backwards(struct lock_class *source, unsigned int depth) struct lock_list *entry; int ret; + if (lockdep_dependency_visit(source, depth)) + return 1; + if (!__raw_spin_is_locked(&lockdep_lock)) return DEBUG_LOCKS_WARN_ON(1); @@ -1064,6 +1159,11 @@ find_usage_backwards(struct lock_class *source, unsigned int depth) return 2; } + if (!source && debug_locks_off_graph_unlock()) { + WARN_ON(1); + return 0; + } + /* * Check this lock's dependency list: */ @@ -1103,9 +1203,9 @@ print_bad_irq_dependency(struct task_struct *curr, printk("\nand this task is already holding:\n"); print_lock(prev); printk("which would create a new lock dependency:\n"); - print_lock_name(prev->class); + print_lock_name(hlock_class(prev)); printk(" ->"); - print_lock_name(next->class); + print_lock_name(hlock_class(next)); printk("\n"); printk("\nbut this new dependency connects a %s-irq-safe lock:\n", @@ -1146,12 +1246,12 @@ check_usage(struct task_struct *curr, struct held_lock *prev, find_usage_bit = bit_backwards; /* fills in <backwards_match> */ - ret = find_usage_backwards(prev->class, 0); + ret = find_usage_backwards(hlock_class(prev), 0); if (!ret || ret == 1) return ret; find_usage_bit = bit_forwards; - ret = find_usage_forwards(next->class, 0); + ret = find_usage_forwards(hlock_class(next), 0); if (!ret || ret == 1) return ret; /* ret == 2 */ @@ -1272,18 +1372,32 @@ check_deadlock(struct task_struct *curr, struct held_lock *next, struct lockdep_map *next_instance, int read) { struct held_lock *prev; + struct held_lock *nest = NULL; int i; for (i = 0; i < curr->lockdep_depth; i++) { prev = curr->held_locks + i; - if (prev->class != next->class) + + if (prev->instance == next->nest_lock) + nest = prev; + + if (hlock_class(prev) != hlock_class(next)) continue; + /* * Allow read-after-read recursion of the same * lock class (i.e. read_lock(lock)+read_lock(lock)): */ if ((read == 2) && prev->read) return 2; + + /* + * We're holding the nest_lock, which serializes this lock's + * nesting behaviour. + */ + if (nest) + return 2; + return print_deadlock_bug(curr, prev, next); } return 1; @@ -1329,7 +1443,7 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, */ check_source = next; check_target = prev; - if (!(check_noncircular(next->class, 0))) + if (!(check_noncircular(hlock_class(next), 0))) return print_circular_bug_tail(); if (!check_prev_add_irq(curr, prev, next)) @@ -1353,8 +1467,8 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, * chains - the second one will be new, but L1 already has * L2 added to its dependency list, due to the first chain.) */ - list_for_each_entry(entry, &prev->class->locks_after, entry) { - if (entry->class == next->class) { + list_for_each_entry(entry, &hlock_class(prev)->locks_after, entry) { + if (entry->class == hlock_class(next)) { if (distance == 1) entry->distance = 1; return 2; @@ -1365,26 +1479,28 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev, * Ok, all validations passed, add the new lock * to the previous lock's dependency list: */ - ret = add_lock_to_list(prev->class, next->class, - &prev->class->locks_after, next->acquire_ip, distance); + ret = add_lock_to_list(hlock_class(prev), hlock_class(next), + &hlock_class(prev)->locks_after, + next->acquire_ip, distance); if (!ret) return 0; - ret = add_lock_to_list(next->class, prev->class, - &next->class->locks_before, next->acquire_ip, distance); + ret = add_lock_to_list(hlock_class(next), hlock_class(prev), + &hlock_class(next)->locks_before, + next->acquire_ip, distance); if (!ret) return 0; /* * Debugging printouts: */ - if (verbose(prev->class) || verbose(next->class)) { + if (verbose(hlock_class(prev)) || verbose(hlock_class(next))) { graph_unlock(); printk("\n new dependency: "); - print_lock_name(prev->class); + print_lock_name(hlock_class(prev)); printk(" => "); - print_lock_name(next->class); + print_lock_name(hlock_class(next)); printk("\n"); dump_stack(); return graph_lock(); @@ -1481,7 +1597,7 @@ static inline int lookup_chain_cache(struct task_struct *curr, struct held_lock *hlock, u64 chain_key) { - struct lock_class *class = hlock->class; + struct lock_class *class = hlock_class(hlock); struct list_head *hash_head = chainhashentry(chain_key); struct lock_chain *chain; struct held_lock *hlock_curr, *hlock_next; @@ -1554,7 +1670,7 @@ cache_hit: if (likely(cn + chain->depth <= MAX_LOCKDEP_CHAIN_HLOCKS)) { chain->base = cn; for (j = 0; j < chain->depth - 1; j++, i++) { - int lock_id = curr->held_locks[i].class - lock_classes; + int lock_id = curr->held_locks[i].class_idx - 1; chain_hlocks[chain->base + j] = lock_id; } chain_hlocks[chain->base + j] = class - lock_classes; @@ -1650,7 +1766,7 @@ static void check_chain_key(struct task_struct *curr) WARN_ON(1); return; } - id = hlock->class - lock_classes; + id = hlock->class_idx - 1; if (DEBUG_LOCKS_WARN_ON(id >= MAX_LOCKDEP_KEYS)) return; @@ -1695,7 +1811,7 @@ print_usage_bug(struct task_struct *curr, struct held_lock *this, print_lock(this); printk("{%s} state was registered at:\n", usage_str[prev_bit]); - print_stack_trace(this->class->usage_traces + prev_bit, 1); + print_stack_trace(hlock_class(this)->usage_traces + prev_bit, 1); print_irqtrace_events(curr); printk("\nother info that might help us debug this:\n"); @@ -1714,7 +1830,7 @@ static inline int valid_state(struct task_struct *curr, struct held_lock *this, enum lock_usage_bit new_bit, enum lock_usage_bit bad_bit) { - if (unlikely(this->class->usage_mask & (1 << bad_bit))) + if (unlikely(hlock_class(this)->usage_mask & (1 << bad_bit))) return print_usage_bug(curr, this, bad_bit, new_bit); return 1; } @@ -1753,7 +1869,7 @@ print_irq_inversion_bug(struct task_struct *curr, struct lock_class *other, lockdep_print_held_locks(curr); printk("\nthe first lock's dependencies:\n"); - print_lock_dependencies(this->class, 0); + print_lock_dependencies(hlock_class(this), 0); printk("\nthe second lock's dependencies:\n"); print_lock_dependencies(other, 0); @@ -1776,7 +1892,7 @@ check_usage_forwards(struct task_struct *curr, struct held_lock *this, find_usage_bit = bit; /* fills in <forwards_match> */ - ret = find_usage_forwards(this->class, 0); + ret = find_usage_forwards(hlock_class(this), 0); if (!ret || ret == 1) return ret; @@ -1795,7 +1911,7 @@ check_usage_backwards(struct task_struct *curr, struct held_lock *this, find_usage_bit = bit; /* fills in <backwards_match> */ - ret = find_usage_backwards(this->class, 0); + ret = find_usage_backwards(hlock_class(this), 0); if (!ret || ret == 1) return ret; @@ -1861,7 +1977,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_ENABLED_HARDIRQS_READ, "hard-read")) return 0; #endif - if (hardirq_verbose(this->class)) + if (hardirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_USED_IN_SOFTIRQ: @@ -1886,7 +2002,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_ENABLED_SOFTIRQS_READ, "soft-read")) return 0; #endif - if (softirq_verbose(this->class)) + if (softirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_USED_IN_HARDIRQ_READ: @@ -1899,7 +2015,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, if (!check_usage_forwards(curr, this, LOCK_ENABLED_HARDIRQS, "hard")) return 0; - if (hardirq_verbose(this->class)) + if (hardirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_USED_IN_SOFTIRQ_READ: @@ -1912,7 +2028,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, if (!check_usage_forwards(curr, this, LOCK_ENABLED_SOFTIRQS, "soft")) return 0; - if (softirq_verbose(this->class)) + if (softirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_ENABLED_HARDIRQS: @@ -1938,7 +2054,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_USED_IN_HARDIRQ_READ, "hard-read")) return 0; #endif - if (hardirq_verbose(this->class)) + if (hardirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_ENABLED_SOFTIRQS: @@ -1964,7 +2080,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_USED_IN_SOFTIRQ_READ, "soft-read")) return 0; #endif - if (softirq_verbose(this->class)) + if (softirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_ENABLED_HARDIRQS_READ: @@ -1979,7 +2095,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_USED_IN_HARDIRQ, "hard")) return 0; #endif - if (hardirq_verbose(this->class)) + if (hardirq_verbose(hlock_class(this))) ret = 2; break; case LOCK_ENABLED_SOFTIRQS_READ: @@ -1994,7 +2110,7 @@ static int mark_lock_irq(struct task_struct *curr, struct held_lock *this, LOCK_USED_IN_SOFTIRQ, "soft")) return 0; #endif - if (softirq_verbose(this->class)) + if (softirq_verbose(hlock_class(this))) ret = 2; break; default: @@ -2310,7 +2426,7 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, * If already set then do not dirty the cacheline, * nor do any checks: */ - if (likely(this->class->usage_mask & new_mask)) + if (likely(hlock_class(this)->usage_mask & new_mask)) return 1; if (!graph_lock()) @@ -2318,14 +2434,14 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, /* * Make sure we didnt race: */ - if (unlikely(this->class->usage_mask & new_mask)) { + if (unlikely(hlock_class(this)->usage_mask & new_mask)) { graph_unlock(); return 1; } - this->class->usage_mask |= new_mask; + hlock_class(this)->usage_mask |= new_mask; - if (!save_trace(this->class->usage_traces + new_bit)) + if (!save_trace(hlock_class(this)->usage_traces + new_bit)) return 0; switch (new_bit) { @@ -2405,7 +2521,7 @@ EXPORT_SYMBOL_GPL(lockdep_init_map); */ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, int trylock, int read, int check, int hardirqs_off, - unsigned long ip) + struct lockdep_map *nest_lock, unsigned long ip) { struct task_struct *curr = current; struct lock_class *class = NULL; @@ -2459,10 +2575,12 @@ static int __lock_acquire(struct lockdep_map *lock, unsigned int subclass, return 0; hlock = curr->held_locks + depth; - - hlock->class = class; + if (DEBUG_LOCKS_WARN_ON(!class)) + return 0; + hlock->class_idx = class - lock_classes + 1; hlock->acquire_ip = ip; hlock->instance = lock; + hlock->nest_lock = nest_lock; hlock->trylock = trylock; hlock->read = read; hlock->check = check; @@ -2574,6 +2692,55 @@ static int check_unlock(struct task_struct *curr, struct lockdep_map *lock, return 1; } +static int +__lock_set_subclass(struct lockdep_map *lock, + unsigned int subclass, unsigned long ip) +{ + struct task_struct *curr = current; + struct held_lock *hlock, *prev_hlock; + struct lock_class *class; + unsigned int depth; + int i; + + depth = curr->lockdep_depth; + if (DEBUG_LOCKS_WARN_ON(!depth)) + return 0; + + prev_hlock = NULL; + for (i = depth-1; i >= 0; i--) { + hlock = curr->held_locks + i; + /* + * We must not cross into another context: + */ + if (prev_hlock && prev_hlock->irq_context != hlock->irq_context) + break; + if (hlock->instance == lock) + goto found_it; + prev_hlock = hlock; + } + return print_unlock_inbalance_bug(curr, lock, ip); + +found_it: + class = register_lock_class(lock, subclass, 0); + hlock->class_idx = class - lock_classes + 1; + + curr->lockdep_depth = i; + curr->curr_chain_key = hlock->prev_chain_key; + + for (; i < depth; i++) { + hlock = curr->held_locks + i; + if (!__lock_acquire(hlock->instance, + hlock_class(hlock)->subclass, hlock->trylock, + hlock->read, hlock->check, hlock->hardirqs_off, + hlock->nest_lock, hlock->acquire_ip)) + return 0; + } + + if (DEBUG_LOCKS_WARN_ON(curr->lockdep_depth != depth)) + return 0; + return 1; +} + /* * Remove the lock to the list of currently held locks in a * potentially non-nested (out of order) manner. This is a @@ -2624,9 +2791,9 @@ found_it: for (i++; i < depth; i++) { hlock = curr->held_locks + i; if (!__lock_acquire(hlock->instance, - hlock->class->subclass, hlock->trylock, + hlock_class(hlock)->subclass, hlock->trylock, hlock->read, hlock->check, hlock->hardirqs_off, - hlock->acquire_ip)) + hlock->nest_lock, hlock->acquire_ip)) return 0; } @@ -2669,7 +2836,7 @@ static int lock_release_nested(struct task_struct *curr, #ifdef CONFIG_DEBUG_LOCKDEP hlock->prev_chain_key = 0; - hlock->class = NULL; + hlock->class_idx = 0; hlock->acquire_ip = 0; hlock->irq_context = 0; #endif @@ -2738,18 +2905,36 @@ static void check_flags(unsigned long flags) #endif } +void +lock_set_subclass(struct lockdep_map *lock, + unsigned int subclass, unsigned long ip) +{ + unsigned long flags; + + if (unlikely(current->lockdep_recursion)) + return; + + raw_local_irq_save(flags); + current->lockdep_recursion = 1; + check_flags(flags); + if (__lock_set_subclass(lock, subclass, ip)) + check_chain_key(current); + current->lockdep_recursion = 0; + raw_local_irq_restore(flags); +} + +EXPORT_SYMBOL_GPL(lock_set_subclass); + /* * We are not always called with irqs disabled - do that here, * and also avoid lockdep recursion: */ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, - int trylock, int read, int check, unsigned long ip) + int trylock, int read, int check, + struct lockdep_map *nest_lock, unsigned long ip) { unsigned long flags; - if (unlikely(!lock_stat && !prove_locking)) - return; - if (unlikely(current->lockdep_recursion)) return; @@ -2758,7 +2943,7 @@ void lock_acquire(struct lockdep_map *lock, unsigned int subclass, current->lockdep_recursion = 1; __lock_acquire(lock, subclass, trylock, read, check, - irqs_disabled_flags(flags), ip); + irqs_disabled_flags(flags), nest_lock, ip); current->lockdep_recursion = 0; raw_local_irq_restore(flags); } @@ -2770,9 +2955,6 @@ void lock_release(struct lockdep_map *lock, int nested, { unsigned long flags; - if (unlikely(!lock_stat && !prove_locking)) - return; - if (unlikely(current->lockdep_recursion)) return; @@ -2845,9 +3027,9 @@ __lock_contended(struct lockdep_map *lock, unsigned long ip) found_it: hlock->waittime_stamp = sched_clock(); - point = lock_contention_point(hlock->class, ip); + point = lock_contention_point(hlock_class(hlock), ip); - stats = get_lock_stats(hlock->class); + stats = get_lock_stats(hlock_class(hlock)); if (point < ARRAY_SIZE(stats->contention_point)) stats->contention_point[i]++; if (lock->cpu != smp_processor_id()) @@ -2893,7 +3075,7 @@ found_it: hlock->holdtime_stamp = now; } - stats = get_lock_stats(hlock->class); + stats = get_lock_stats(hlock_class(hlock)); if (waittime) { if (hlock->read) lock_time_inc(&stats->read_waittime, waittime); @@ -2988,6 +3170,7 @@ static void zap_class(struct lock_class *class) list_del_rcu(&class->hash_entry); list_del_rcu(&class->lock_entry); + class->key = NULL; } static inline int within(const void *addr, void *start, unsigned long size) diff --git a/kernel/lockdep_internals.h b/kernel/lockdep_internals.h index c3600a091a2..55db193d366 100644 --- a/kernel/lockdep_internals.h +++ b/kernel/lockdep_internals.h @@ -17,9 +17,6 @@ */ #define MAX_LOCKDEP_ENTRIES 8192UL -#define MAX_LOCKDEP_KEYS_BITS 11 -#define MAX_LOCKDEP_KEYS (1UL << MAX_LOCKDEP_KEYS_BITS) - #define MAX_LOCKDEP_CHAINS_BITS 14 #define MAX_LOCKDEP_CHAINS (1UL << MAX_LOCKDEP_CHAINS_BITS) @@ -53,6 +50,9 @@ extern unsigned int nr_process_chains; extern unsigned int max_lockdep_depth; extern unsigned int max_recursion_depth; +extern unsigned long lockdep_count_forward_deps(struct lock_class *); +extern unsigned long lockdep_count_backward_deps(struct lock_class *); + #ifdef CONFIG_DEBUG_LOCKDEP /* * Various lockdep statistics: diff --git a/kernel/lockdep_proc.c b/kernel/lockdep_proc.c index 9b0e940e254..fa19aee604c 100644 --- a/kernel/lockdep_proc.c +++ b/kernel/lockdep_proc.c @@ -63,34 +63,6 @@ static void l_stop(struct seq_file *m, void *v) { } -static unsigned long count_forward_deps(struct lock_class *class) -{ - struct lock_list *entry; - unsigned long ret = 1; - - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_after, entry) - ret += count_forward_deps(entry->class); - - return ret; -} - -static unsigned long count_backward_deps(struct lock_class *class) -{ - struct lock_list *entry; - unsigned long ret = 1; - - /* - * Recurse this class's dependency list: - */ - list_for_each_entry(entry, &class->locks_before, entry) - ret += count_backward_deps(entry->class); - - return ret; -} - static void print_name(struct seq_file *m, struct lock_class *class) { char str[128]; @@ -124,10 +96,10 @@ static int l_show(struct seq_file *m, void *v) #ifdef CONFIG_DEBUG_LOCKDEP seq_printf(m, " OPS:%8ld", class->ops); #endif - nr_forward_deps = count_forward_deps(class); + nr_forward_deps = lockdep_count_forward_deps(class); seq_printf(m, " FD:%5ld", nr_forward_deps); - nr_backward_deps = count_backward_deps(class); + nr_backward_deps = lockdep_count_backward_deps(class); seq_printf(m, " BD:%5ld", nr_backward_deps); get_usage_chars(class, &c1, &c2, &c3, &c4); @@ -229,6 +201,9 @@ static int lc_show(struct seq_file *m, void *v) for (i = 0; i < chain->depth; i++) { class = lock_chain_get_class(chain, i); + if (!class->key) + continue; + seq_printf(m, "[%p] ", class->key); print_name(m, class); seq_puts(m, "\n"); @@ -350,7 +325,7 @@ static int lockdep_stats_show(struct seq_file *m, void *v) if (class->usage_mask & LOCKF_ENABLED_HARDIRQS_READ) nr_hardirq_read_unsafe++; - sum_forward_deps += count_forward_deps(class); + sum_forward_deps += lockdep_count_forward_deps(class); } #ifdef CONFIG_DEBUG_LOCKDEP DEBUG_LOCKS_WARN_ON(debug_atomic_read(&nr_unused_locks) != nr_unused); diff --git a/kernel/marker.c b/kernel/marker.c index 971da531790..7d1faecd7a5 100644 --- a/kernel/marker.c +++ b/kernel/marker.c @@ -126,6 +126,11 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...) struct marker_probe_closure *multi; int i; /* + * Read mdata->ptype before mdata->multi. + */ + smp_rmb(); + multi = mdata->multi; + /* * multi points to an array, therefore accessing the array * depends on reading multi. However, even in this case, * we must insure that the pointer is read _before_ the array @@ -133,7 +138,6 @@ void marker_probe_cb(const struct marker *mdata, void *call_private, ...) * in the fast path, so put the explicit barrier here. */ smp_read_barrier_depends(); - multi = mdata->multi; for (i = 0; multi[i].func; i++) { va_start(args, call_private); multi[i].func(multi[i].probe_private, call_private, @@ -175,6 +179,11 @@ void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...) struct marker_probe_closure *multi; int i; /* + * Read mdata->ptype before mdata->multi. + */ + smp_rmb(); + multi = mdata->multi; + /* * multi points to an array, therefore accessing the array * depends on reading multi. However, even in this case, * we must insure that the pointer is read _before_ the array @@ -182,7 +191,6 @@ void marker_probe_cb_noarg(const struct marker *mdata, void *call_private, ...) * in the fast path, so put the explicit barrier here. */ smp_read_barrier_depends(); - multi = mdata->multi; for (i = 0; multi[i].func; i++) multi[i].func(multi[i].probe_private, call_private, mdata->format, &args); diff --git a/kernel/module.c b/kernel/module.c index d8b5605132a..08864d257eb 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -325,18 +325,6 @@ static unsigned long find_symbol(const char *name, return -ENOENT; } -/* lookup symbol in given range of kernel_symbols */ -static const struct kernel_symbol *lookup_symbol(const char *name, - const struct kernel_symbol *start, - const struct kernel_symbol *stop) -{ - const struct kernel_symbol *ks = start; - for (; ks < stop; ks++) - if (strcmp(ks->name, name) == 0) - return ks; - return NULL; -} - /* Search for module by name: must hold module_mutex. */ static struct module *find_module(const char *name) { @@ -690,7 +678,7 @@ static int try_stop_module(struct module *mod, int flags, int *forced) if (flags & O_NONBLOCK) { struct stopref sref = { mod, flags, forced }; - return stop_machine_run(__try_stop_module, &sref, NR_CPUS); + return stop_machine(__try_stop_module, &sref, NULL); } else { /* We don't need to stop the machine for this. */ mod->state = MODULE_STATE_GOING; @@ -1428,7 +1416,7 @@ static int __unlink_module(void *_mod) static void free_module(struct module *mod) { /* Delete from various lists */ - stop_machine_run(__unlink_module, mod, NR_CPUS); + stop_machine(__unlink_module, mod, NULL); remove_notes_attrs(mod); remove_sect_attrs(mod); mod_kobject_remove(mod); @@ -1703,6 +1691,19 @@ static void setup_modinfo(struct module *mod, Elf_Shdr *sechdrs, } #ifdef CONFIG_KALLSYMS + +/* lookup symbol in given range of kernel_symbols */ +static const struct kernel_symbol *lookup_symbol(const char *name, + const struct kernel_symbol *start, + const struct kernel_symbol *stop) +{ + const struct kernel_symbol *ks = start; + for (; ks < stop; ks++) + if (strcmp(ks->name, name) == 0) + return ks; + return NULL; +} + static int is_exported(const char *name, const struct module *mod) { if (!mod && lookup_symbol(name, __start___ksymtab, __stop___ksymtab)) @@ -2196,7 +2197,7 @@ static struct module *load_module(void __user *umod, /* Now sew it into the lists so we can get lockdep and oops * info during argument parsing. Noone should access us, since * strong_try_module_get() will fail. */ - stop_machine_run(__link_module, mod, NR_CPUS); + stop_machine(__link_module, mod, NULL); /* Size of section 0 is 0, so this works well if no params */ err = parse_args(mod->name, mod->args, @@ -2230,7 +2231,7 @@ static struct module *load_module(void __user *umod, return mod; unlink: - stop_machine_run(__unlink_module, mod, NR_CPUS); + stop_machine(__unlink_module, mod, NULL); module_arch_cleanup(mod); cleanup: kobject_del(&mod->mkobj.kobj); @@ -2287,7 +2288,7 @@ sys_init_module(void __user *umod, /* Start the module */ if (mod->init != NULL) - ret = mod->init(); + ret = do_one_initcall(mod->init); if (ret < 0) { /* Init routine failed: abort. Try to protect us from buggy refcounters. */ diff --git a/kernel/mutex.c b/kernel/mutex.c index bcdc9ac8ef6..12c779dc65d 100644 --- a/kernel/mutex.c +++ b/kernel/mutex.c @@ -34,6 +34,7 @@ /*** * mutex_init - initialize the mutex * @lock: the mutex to be initialized + * @key: the lock_class_key for the class; used by mutex lock debugging * * Initialize the mutex to unlocked state. * diff --git a/kernel/pm_qos_params.c b/kernel/pm_qos_params.c index 8cb75702638..da9c2dda6a4 100644 --- a/kernel/pm_qos_params.c +++ b/kernel/pm_qos_params.c @@ -24,7 +24,7 @@ * requirement that the application has is cleaned up when closes the file * pointer or exits the pm_qos_object will get an opportunity to clean up. * - * mark gross mgross@linux.intel.com + * Mark Gross <mgross@linux.intel.com> */ #include <linux/pm_qos_params.h> @@ -211,8 +211,8 @@ EXPORT_SYMBOL_GPL(pm_qos_requirement); * @value: defines the qos request * * This function inserts a new entry in the pm_qos_class list of requested qos - * performance charactoistics. It recomputes the agregate QoS expectations for - * the pm_qos_class of parrameters. + * performance characteristics. It recomputes the aggregate QoS expectations + * for the pm_qos_class of parameters. */ int pm_qos_add_requirement(int pm_qos_class, char *name, s32 value) { @@ -250,10 +250,10 @@ EXPORT_SYMBOL_GPL(pm_qos_add_requirement); * @name: identifies the request * @value: defines the qos request * - * Updates an existing qos requierement for the pm_qos_class of parameters along + * Updates an existing qos requirement for the pm_qos_class of parameters along * with updating the target pm_qos_class value. * - * If the named request isn't in the lest then no change is made. + * If the named request isn't in the list then no change is made. */ int pm_qos_update_requirement(int pm_qos_class, char *name, s32 new_value) { @@ -287,7 +287,7 @@ EXPORT_SYMBOL_GPL(pm_qos_update_requirement); * @pm_qos_class: identifies which list of qos request to us * @name: identifies the request * - * Will remove named qos request from pm_qos_class list of parrameters and + * Will remove named qos request from pm_qos_class list of parameters and * recompute the current target value for the pm_qos_class. */ void pm_qos_remove_requirement(int pm_qos_class, char *name) @@ -319,7 +319,7 @@ EXPORT_SYMBOL_GPL(pm_qos_remove_requirement); * @notifier: notifier block managed by caller. * * will register the notifier into a notification chain that gets called - * uppon changes to the pm_qos_class target value. + * upon changes to the pm_qos_class target value. */ int pm_qos_add_notifier(int pm_qos_class, struct notifier_block *notifier) { @@ -338,7 +338,7 @@ EXPORT_SYMBOL_GPL(pm_qos_add_notifier); * @notifier: notifier block to be removed. * * will remove the notifier from the notification chain that gets called - * uppon changes to the pm_qos_class target value. + * upon changes to the pm_qos_class target value. */ int pm_qos_remove_notifier(int pm_qos_class, struct notifier_block *notifier) { diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c index 9a21681aa80..e36d5798cbf 100644 --- a/kernel/posix-timers.c +++ b/kernel/posix-timers.c @@ -289,21 +289,29 @@ void do_schedule_next_timer(struct siginfo *info) else schedule_next_timer(timr); - info->si_overrun = timr->it_overrun_last; + info->si_overrun += timr->it_overrun_last; } if (timr) unlock_timer(timr, flags); } -int posix_timer_event(struct k_itimer *timr,int si_private) +int posix_timer_event(struct k_itimer *timr, int si_private) { - memset(&timr->sigq->info, 0, sizeof(siginfo_t)); + /* + * FIXME: if ->sigq is queued we can race with + * dequeue_signal()->do_schedule_next_timer(). + * + * If dequeue_signal() sees the "right" value of + * si_sys_private it calls do_schedule_next_timer(). + * We re-queue ->sigq and drop ->it_lock(). + * do_schedule_next_timer() locks the timer + * and re-schedules it while ->sigq is pending. + * Not really bad, but not that we want. + */ timr->sigq->info.si_sys_private = si_private; - /* Send signal to the process that owns this timer.*/ timr->sigq->info.si_signo = timr->it_sigev_signo; - timr->sigq->info.si_errno = 0; timr->sigq->info.si_code = SI_TIMER; timr->sigq->info.si_tid = timr->it_id; timr->sigq->info.si_value = timr->it_sigev_value; @@ -435,6 +443,7 @@ static struct k_itimer * alloc_posix_timer(void) kmem_cache_free(posix_timers_cache, tmr); tmr = NULL; } + memset(&tmr->sigq->info, 0, sizeof(siginfo_t)); return tmr; } diff --git a/kernel/printk.c b/kernel/printk.c index a7f7559c5f6..b51b1567bb5 100644 --- a/kernel/printk.c +++ b/kernel/printk.c @@ -1309,14 +1309,14 @@ void tty_write_message(struct tty_struct *tty, char *msg) #if defined CONFIG_PRINTK -DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10); /* * printk rate limiting, lifted from the networking subsystem. * - * This enforces a rate limit: not more than one kernel message - * every printk_ratelimit_jiffies to make a denial-of-service - * attack impossible. + * This enforces a rate limit: not more than 10 kernel messages + * every 5s to make a denial-of-service attack impossible. */ +DEFINE_RATELIMIT_STATE(printk_ratelimit_state, 5 * HZ, 10); + int printk_ratelimit(void) { return __ratelimit(&printk_ratelimit_state); diff --git a/kernel/rcuclassic.c b/kernel/rcuclassic.c index 6f8696c502f..aad93cdc9f6 100644 --- a/kernel/rcuclassic.c +++ b/kernel/rcuclassic.c @@ -91,8 +91,8 @@ static void force_quiescent_state(struct rcu_data *rdp, * rdp->cpu is the current cpu. * * cpu_online_map is updated by the _cpu_down() - * using stop_machine_run(). Since we're in irqs disabled - * section, stop_machine_run() is not exectuting, hence + * using __stop_machine(). Since we're in irqs disabled + * section, __stop_machine() is not exectuting, hence * the cpu_online_map is stable. * * However, a cpu might have been offlined _just_ before diff --git a/kernel/relay.c b/kernel/relay.c index 04006ef970b..8d13a7855c0 100644 --- a/kernel/relay.c +++ b/kernel/relay.c @@ -944,6 +944,10 @@ static void relay_file_read_consume(struct rchan_buf *buf, size_t n_subbufs = buf->chan->n_subbufs; size_t read_subbuf; + if (buf->subbufs_produced == buf->subbufs_consumed && + buf->offset == buf->bytes_consumed) + return; + if (buf->bytes_consumed + bytes_consumed > subbuf_size) { relay_subbufs_consumed(buf->chan, buf->cpu, 1); buf->bytes_consumed = 0; @@ -975,6 +979,8 @@ static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos) relay_file_read_consume(buf, read_pos, 0); + consumed = buf->subbufs_consumed; + if (unlikely(buf->offset > subbuf_size)) { if (produced == consumed) return 0; @@ -993,8 +999,12 @@ static int relay_file_read_avail(struct rchan_buf *buf, size_t read_pos) if (consumed > produced) produced += n_subbufs * subbuf_size; - if (consumed == produced) + if (consumed == produced) { + if (buf->offset == subbuf_size && + buf->subbufs_produced > buf->subbufs_consumed) + return 1; return 0; + } return 1; } diff --git a/kernel/resource.c b/kernel/resource.c index 74af2d7cb5a..f5b518eabef 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -490,7 +490,7 @@ resource_size_t resource_alignment(struct resource *res) { switch (res->flags & (IORESOURCE_SIZEALIGN | IORESOURCE_STARTALIGN)) { case IORESOURCE_SIZEALIGN: - return res->end - res->start + 1; + return resource_size(res); case IORESOURCE_STARTALIGN: return res->start; default: diff --git a/kernel/sched.c b/kernel/sched.c index 0236958addc..d601fb0406c 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -600,7 +600,6 @@ struct rq { /* BKL stats */ unsigned int bkl_count; #endif - struct lock_class_key rq_lock_key; }; static DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); @@ -834,7 +833,7 @@ static inline u64 global_rt_period(void) static inline u64 global_rt_runtime(void) { - if (sysctl_sched_rt_period < 0) + if (sysctl_sched_rt_runtime < 0) return RUNTIME_INF; return (u64)sysctl_sched_rt_runtime * NSEC_PER_USEC; @@ -2759,10 +2758,10 @@ static void double_rq_lock(struct rq *rq1, struct rq *rq2) } else { if (rq1 < rq2) { spin_lock(&rq1->lock); - spin_lock(&rq2->lock); + spin_lock_nested(&rq2->lock, SINGLE_DEPTH_NESTING); } else { spin_lock(&rq2->lock); - spin_lock(&rq1->lock); + spin_lock_nested(&rq1->lock, SINGLE_DEPTH_NESTING); } } update_rq_clock(rq1); @@ -2805,14 +2804,21 @@ static int double_lock_balance(struct rq *this_rq, struct rq *busiest) if (busiest < this_rq) { spin_unlock(&this_rq->lock); spin_lock(&busiest->lock); - spin_lock(&this_rq->lock); + spin_lock_nested(&this_rq->lock, SINGLE_DEPTH_NESTING); ret = 1; } else - spin_lock(&busiest->lock); + spin_lock_nested(&busiest->lock, SINGLE_DEPTH_NESTING); } return ret; } +static void double_unlock_balance(struct rq *this_rq, struct rq *busiest) + __releases(busiest->lock) +{ + spin_unlock(&busiest->lock); + lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_); +} + /* * If dest_cpu is allowed for this process, migrate the task to it. * This is accomplished by forcing the cpu_allowed mask to only @@ -3637,7 +3643,7 @@ redo: ld_moved = move_tasks(this_rq, this_cpu, busiest, imbalance, sd, CPU_NEWLY_IDLE, &all_pinned); - spin_unlock(&busiest->lock); + double_unlock_balance(this_rq, busiest); if (unlikely(all_pinned)) { cpu_clear(cpu_of(busiest), *cpus); @@ -3752,7 +3758,7 @@ static void active_load_balance(struct rq *busiest_rq, int busiest_cpu) else schedstat_inc(sd, alb_failed); } - spin_unlock(&target_rq->lock); + double_unlock_balance(busiest_rq, target_rq); } #ifdef CONFIG_NO_HZ @@ -5004,19 +5010,21 @@ recheck: return -EPERM; } + if (user) { #ifdef CONFIG_RT_GROUP_SCHED - /* - * Do not allow realtime tasks into groups that have no runtime - * assigned. - */ - if (user - && rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0) - return -EPERM; + /* + * Do not allow realtime tasks into groups that have no runtime + * assigned. + */ + if (rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0) + return -EPERM; #endif - retval = security_task_setscheduler(p, policy, param); - if (retval) - return retval; + retval = security_task_setscheduler(p, policy, param); + if (retval) + return retval; + } + /* * make sure no PI-waiters arrive (or leave) while we are * changing the priority of the task: @@ -7671,34 +7679,34 @@ static ssize_t sched_power_savings_store(const char *buf, size_t count, int smt) } #ifdef CONFIG_SCHED_MC -static ssize_t sched_mc_power_savings_show(struct sys_device *dev, - struct sysdev_attribute *attr, char *page) +static ssize_t sched_mc_power_savings_show(struct sysdev_class *class, + char *page) { return sprintf(page, "%u\n", sched_mc_power_savings); } -static ssize_t sched_mc_power_savings_store(struct sys_device *dev, - struct sysdev_attribute *attr, +static ssize_t sched_mc_power_savings_store(struct sysdev_class *class, const char *buf, size_t count) { return sched_power_savings_store(buf, count, 0); } -static SYSDEV_ATTR(sched_mc_power_savings, 0644, sched_mc_power_savings_show, - sched_mc_power_savings_store); +static SYSDEV_CLASS_ATTR(sched_mc_power_savings, 0644, + sched_mc_power_savings_show, + sched_mc_power_savings_store); #endif #ifdef CONFIG_SCHED_SMT -static ssize_t sched_smt_power_savings_show(struct sys_device *dev, - struct sysdev_attribute *attr, char *page) +static ssize_t sched_smt_power_savings_show(struct sysdev_class *dev, + char *page) { return sprintf(page, "%u\n", sched_smt_power_savings); } -static ssize_t sched_smt_power_savings_store(struct sys_device *dev, - struct sysdev_attribute *attr, +static ssize_t sched_smt_power_savings_store(struct sysdev_class *dev, const char *buf, size_t count) { return sched_power_savings_store(buf, count, 1); } -static SYSDEV_ATTR(sched_smt_power_savings, 0644, sched_smt_power_savings_show, +static SYSDEV_CLASS_ATTR(sched_smt_power_savings, 0644, + sched_smt_power_savings_show, sched_smt_power_savings_store); #endif @@ -7998,7 +8006,6 @@ void __init sched_init(void) rq = cpu_rq(i); spin_lock_init(&rq->lock); - lockdep_set_class(&rq->lock, &rq->rq_lock_key); rq->nr_running = 0; init_cfs_rq(&rq->cfs, rq); init_rt_rq(&rq->rt, rq); diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index 22ed55d1167..204991a0bfa 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -32,13 +32,19 @@ #include <linux/ktime.h> #include <linux/module.h> +/* + * Scheduler clock - returns current time in nanosec units. + * This is default implementation. + * Architectures and sub-architectures can override this. + */ +unsigned long long __attribute__((weak)) sched_clock(void) +{ + return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); +} -#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK +static __read_mostly int sched_clock_running; -#define MULTI_SHIFT 15 -/* Max is double, Min is 1/2 */ -#define MAX_MULTI (2LL << MULTI_SHIFT) -#define MIN_MULTI (1LL << (MULTI_SHIFT-1)) +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK struct sched_clock_data { /* @@ -49,14 +55,9 @@ struct sched_clock_data { raw_spinlock_t lock; unsigned long tick_jiffies; - u64 prev_raw; u64 tick_raw; u64 tick_gtod; u64 clock; - s64 multi; -#ifdef CONFIG_NO_HZ - int check_max; -#endif }; static DEFINE_PER_CPU_SHARED_ALIGNED(struct sched_clock_data, sched_clock_data); @@ -71,8 +72,6 @@ static inline struct sched_clock_data *cpu_sdc(int cpu) return &per_cpu(sched_clock_data, cpu); } -static __read_mostly int sched_clock_running; - void sched_clock_init(void) { u64 ktime_now = ktime_to_ns(ktime_get()); @@ -84,90 +83,39 @@ void sched_clock_init(void) scd->lock = (raw_spinlock_t)__RAW_SPIN_LOCK_UNLOCKED; scd->tick_jiffies = now_jiffies; - scd->prev_raw = 0; scd->tick_raw = 0; scd->tick_gtod = ktime_now; scd->clock = ktime_now; - scd->multi = 1 << MULTI_SHIFT; -#ifdef CONFIG_NO_HZ - scd->check_max = 1; -#endif } sched_clock_running = 1; } -#ifdef CONFIG_NO_HZ -/* - * The dynamic ticks makes the delta jiffies inaccurate. This - * prevents us from checking the maximum time update. - * Disable the maximum check during stopped ticks. - */ -void sched_clock_tick_stop(int cpu) -{ - struct sched_clock_data *scd = cpu_sdc(cpu); - - scd->check_max = 0; -} - -void sched_clock_tick_start(int cpu) -{ - struct sched_clock_data *scd = cpu_sdc(cpu); - - scd->check_max = 1; -} - -static int check_max(struct sched_clock_data *scd) -{ - return scd->check_max; -} -#else -static int check_max(struct sched_clock_data *scd) -{ - return 1; -} -#endif /* CONFIG_NO_HZ */ - /* * update the percpu scd from the raw @now value * * - filter out backward motion * - use jiffies to generate a min,max window to clip the raw values */ -static void __update_sched_clock(struct sched_clock_data *scd, u64 now, u64 *time) +static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) { unsigned long now_jiffies = jiffies; long delta_jiffies = now_jiffies - scd->tick_jiffies; u64 clock = scd->clock; u64 min_clock, max_clock; - s64 delta = now - scd->prev_raw; + s64 delta = now - scd->tick_raw; WARN_ON_ONCE(!irqs_disabled()); - - /* - * At schedule tick the clock can be just under the gtod. We don't - * want to push it too prematurely. - */ - min_clock = scd->tick_gtod + (delta_jiffies * TICK_NSEC); - if (min_clock > TICK_NSEC) - min_clock -= TICK_NSEC / 2; + min_clock = scd->tick_gtod + delta_jiffies * TICK_NSEC; if (unlikely(delta < 0)) { clock++; goto out; } - /* - * The clock must stay within a jiffie of the gtod. - * But since we may be at the start of a jiffy or the end of one - * we add another jiffy buffer. - */ - max_clock = scd->tick_gtod + (2 + delta_jiffies) * TICK_NSEC; - - delta *= scd->multi; - delta >>= MULTI_SHIFT; + max_clock = min_clock + TICK_NSEC; - if (unlikely(clock + delta > max_clock) && check_max(scd)) { + if (unlikely(clock + delta > max_clock)) { if (clock < max_clock) clock = max_clock; else @@ -180,12 +128,10 @@ static void __update_sched_clock(struct sched_clock_data *scd, u64 now, u64 *tim if (unlikely(clock < min_clock)) clock = min_clock; - if (time) - *time = clock; - else { - scd->prev_raw = now; - scd->clock = clock; - } + scd->tick_jiffies = now_jiffies; + scd->clock = clock; + + return clock; } static void lock_double_clock(struct sched_clock_data *data1, @@ -203,7 +149,7 @@ static void lock_double_clock(struct sched_clock_data *data1, u64 sched_clock_cpu(int cpu) { struct sched_clock_data *scd = cpu_sdc(cpu); - u64 now, clock; + u64 now, clock, this_clock, remote_clock; if (unlikely(!sched_clock_running)) return 0ull; @@ -212,43 +158,44 @@ u64 sched_clock_cpu(int cpu) now = sched_clock(); if (cpu != raw_smp_processor_id()) { - /* - * in order to update a remote cpu's clock based on our - * unstable raw time rebase it against: - * tick_raw (offset between raw counters) - * tick_gotd (tick offset between cpus) - */ struct sched_clock_data *my_scd = this_scd(); lock_double_clock(scd, my_scd); - now -= my_scd->tick_raw; - now += scd->tick_raw; + this_clock = __update_sched_clock(my_scd, now); + remote_clock = scd->clock; - now += my_scd->tick_gtod; - now -= scd->tick_gtod; + /* + * Use the opportunity that we have both locks + * taken to couple the two clocks: we take the + * larger time as the latest time for both + * runqueues. (this creates monotonic movement) + */ + if (likely(remote_clock < this_clock)) { + clock = this_clock; + scd->clock = clock; + } else { + /* + * Should be rare, but possible: + */ + clock = remote_clock; + my_scd->clock = remote_clock; + } __raw_spin_unlock(&my_scd->lock); - - __update_sched_clock(scd, now, &clock); - - __raw_spin_unlock(&scd->lock); - } else { __raw_spin_lock(&scd->lock); - __update_sched_clock(scd, now, NULL); - clock = scd->clock; - __raw_spin_unlock(&scd->lock); + clock = __update_sched_clock(scd, now); } + __raw_spin_unlock(&scd->lock); + return clock; } void sched_clock_tick(void) { struct sched_clock_data *scd = this_scd(); - unsigned long now_jiffies = jiffies; - s64 mult, delta_gtod, delta_raw; u64 now, now_gtod; if (unlikely(!sched_clock_running)) @@ -260,29 +207,14 @@ void sched_clock_tick(void) now = sched_clock(); __raw_spin_lock(&scd->lock); - __update_sched_clock(scd, now, NULL); + __update_sched_clock(scd, now); /* * update tick_gtod after __update_sched_clock() because that will * already observe 1 new jiffy; adding a new tick_gtod to that would * increase the clock 2 jiffies. */ - delta_gtod = now_gtod - scd->tick_gtod; - delta_raw = now - scd->tick_raw; - - if ((long)delta_raw > 0) { - mult = delta_gtod << MULTI_SHIFT; - do_div(mult, delta_raw); - scd->multi = mult; - if (scd->multi > MAX_MULTI) - scd->multi = MAX_MULTI; - else if (scd->multi < MIN_MULTI) - scd->multi = MIN_MULTI; - } else - scd->multi = 1 << MULTI_SHIFT; - scd->tick_raw = now; scd->tick_gtod = now_gtod; - scd->tick_jiffies = now_jiffies; __raw_spin_unlock(&scd->lock); } @@ -301,7 +233,6 @@ EXPORT_SYMBOL_GPL(sched_clock_idle_sleep_event); void sched_clock_idle_wakeup_event(u64 delta_ns) { struct sched_clock_data *scd = this_scd(); - u64 now = sched_clock(); /* * Override the previous timestamp and ignore all @@ -310,27 +241,30 @@ void sched_clock_idle_wakeup_event(u64 delta_ns) * rq clock: */ __raw_spin_lock(&scd->lock); - scd->prev_raw = now; scd->clock += delta_ns; - scd->multi = 1 << MULTI_SHIFT; __raw_spin_unlock(&scd->lock); touch_softlockup_watchdog(); } EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); -#endif +#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ -/* - * Scheduler clock - returns current time in nanosec units. - * This is default implementation. - * Architectures and sub-architectures can override this. - */ -unsigned long long __attribute__((weak)) sched_clock(void) +void sched_clock_init(void) { - return (unsigned long long)jiffies * (NSEC_PER_SEC / HZ); + sched_clock_running = 1; } +u64 sched_clock_cpu(int cpu) +{ + if (unlikely(!sched_clock_running)) + return 0; + + return sched_clock(); +} + +#endif + unsigned long long cpu_clock(int cpu) { unsigned long long clock; diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c index cf2cd6ce4cb..fb8994c6d4b 100644 --- a/kernel/sched_fair.c +++ b/kernel/sched_fair.c @@ -899,7 +899,7 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p) * doesn't make sense. Rely on vruntime for fairness. */ if (rq->curr != p) - delta = max(10000LL, delta); + delta = max_t(s64, 10000LL, delta); hrtick_start(rq, delta); } @@ -1442,18 +1442,23 @@ __load_balance_iterator(struct cfs_rq *cfs_rq, struct list_head *next) struct task_struct *p = NULL; struct sched_entity *se; - while (next != &cfs_rq->tasks) { + if (next == &cfs_rq->tasks) + return NULL; + + /* Skip over entities that are not tasks */ + do { se = list_entry(next, struct sched_entity, group_node); next = next->next; + } while (next != &cfs_rq->tasks && !entity_is_task(se)); - /* Skip over entities that are not tasks */ - if (entity_is_task(se)) { - p = task_of(se); - break; - } - } + if (next == &cfs_rq->tasks) + return NULL; cfs_rq->balance_iterator = next; + + if (entity_is_task(se)) + p = task_of(se); + return p; } diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index 908c04f9dad..6163e4cf885 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -861,6 +861,8 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) #define RT_MAX_TRIES 3 static int double_lock_balance(struct rq *this_rq, struct rq *busiest); +static void double_unlock_balance(struct rq *this_rq, struct rq *busiest); + static void deactivate_task(struct rq *rq, struct task_struct *p, int sleep); static int pick_rt_task(struct rq *rq, struct task_struct *p, int cpu) @@ -1022,7 +1024,7 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) break; /* try again */ - spin_unlock(&lowest_rq->lock); + double_unlock_balance(rq, lowest_rq); lowest_rq = NULL; } @@ -1091,7 +1093,7 @@ static int push_rt_task(struct rq *rq) resched_task(lowest_rq->curr); - spin_unlock(&lowest_rq->lock); + double_unlock_balance(rq, lowest_rq); ret = 1; out: @@ -1197,7 +1199,7 @@ static int pull_rt_task(struct rq *this_rq) } skip: - spin_unlock(&src_rq->lock); + double_unlock_balance(this_rq, src_rq); } return ret; diff --git a/kernel/semaphore.c b/kernel/semaphore.c index aaaeae8244e..94a62c0d4ad 100644 --- a/kernel/semaphore.c +++ b/kernel/semaphore.c @@ -212,9 +212,7 @@ static inline int __sched __down_common(struct semaphore *sem, long state, waiter.up = 0; for (;;) { - if (state == TASK_INTERRUPTIBLE && signal_pending(task)) - goto interrupted; - if (state == TASK_KILLABLE && fatal_signal_pending(task)) + if (signal_pending_state(state, task)) goto interrupted; if (timeout <= 0) goto timed_out; diff --git a/kernel/signal.c b/kernel/signal.c index 954f77d7e3b..c539f60c6f4 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1304,6 +1304,7 @@ int send_sigqueue(struct sigqueue *q, struct task_struct *t, int group) q->info.si_overrun++; goto out; } + q->info.si_overrun = 0; signalfd_notify(t, sig); pending = group ? &t->signal->shared_pending : &t->pending; diff --git a/kernel/smp.c b/kernel/smp.c index 96fc7c0edc5..782e2b93e46 100644 --- a/kernel/smp.c +++ b/kernel/smp.c @@ -135,7 +135,8 @@ void generic_smp_call_function_interrupt(void) */ smp_wmb(); data->csd.flags &= ~CSD_FLAG_WAIT; - } else + } + if (data->csd.flags & CSD_FLAG_ALLOC) call_rcu(&data->rcu_head, rcu_free_call_data); } rcu_read_unlock(); @@ -260,6 +261,42 @@ void __smp_call_function_single(int cpu, struct call_single_data *data) generic_exec_single(cpu, data); } +/* Dummy function */ +static void quiesce_dummy(void *unused) +{ +} + +/* + * Ensure stack based data used in call function mask is safe to free. + * + * This is needed by smp_call_function_mask when using on-stack data, because + * a single call function queue is shared by all CPUs, and any CPU may pick up + * the data item on the queue at any time before it is deleted. So we need to + * ensure that all CPUs have transitioned through a quiescent state after + * this call. + * + * This is a very slow function, implemented by sending synchronous IPIs to + * all possible CPUs. For this reason, we have to alloc data rather than use + * stack based data even in the case of synchronous calls. The stack based + * data is then just used for deadlock/oom fallback which will be very rare. + * + * If a faster scheme can be made, we could go back to preferring stack based + * data -- the data allocation/free is non-zero cost. + */ +static void smp_call_function_mask_quiesce_stack(cpumask_t mask) +{ + struct call_single_data data; + int cpu; + + data.func = quiesce_dummy; + data.info = NULL; + + for_each_cpu_mask(cpu, mask) { + data.flags = CSD_FLAG_WAIT; + generic_exec_single(cpu, &data); + } +} + /** * smp_call_function_mask(): Run a function on a set of other CPUs. * @mask: The set of cpus to run on. @@ -285,6 +322,7 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info, cpumask_t allbutself; unsigned long flags; int cpu, num_cpus; + int slowpath = 0; /* Can deadlock when called with interrupts disabled */ WARN_ON(irqs_disabled()); @@ -306,15 +344,16 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info, return smp_call_function_single(cpu, func, info, wait); } - if (!wait) { - data = kmalloc(sizeof(*data), GFP_ATOMIC); - if (data) - data->csd.flags = CSD_FLAG_ALLOC; - } - if (!data) { + data = kmalloc(sizeof(*data), GFP_ATOMIC); + if (data) { + data->csd.flags = CSD_FLAG_ALLOC; + if (wait) + data->csd.flags |= CSD_FLAG_WAIT; + } else { data = &d; data->csd.flags = CSD_FLAG_WAIT; wait = 1; + slowpath = 1; } spin_lock_init(&data->lock); @@ -331,8 +370,11 @@ int smp_call_function_mask(cpumask_t mask, void (*func)(void *), void *info, arch_send_call_function_ipi(mask); /* optionally wait for the CPUs to complete */ - if (wait) + if (wait) { csd_flag_wait(&data->csd); + if (unlikely(slowpath)) + smp_call_function_mask_quiesce_stack(mask); + } return 0; } diff --git a/kernel/spinlock.c b/kernel/spinlock.c index a1fb54c93cd..44baeea94ab 100644 --- a/kernel/spinlock.c +++ b/kernel/spinlock.c @@ -292,6 +292,7 @@ void __lockfunc _spin_lock_nested(spinlock_t *lock, int subclass) } EXPORT_SYMBOL(_spin_lock_nested); + unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclass) { unsigned long flags; @@ -314,6 +315,16 @@ unsigned long __lockfunc _spin_lock_irqsave_nested(spinlock_t *lock, int subclas EXPORT_SYMBOL(_spin_lock_irqsave_nested); +void __lockfunc _spin_lock_nest_lock(spinlock_t *lock, + struct lockdep_map *nest_lock) +{ + preempt_disable(); + spin_acquire_nest(&lock->dep_map, 0, 0, nest_lock, _RET_IP_); + LOCK_CONTENDED(lock, _raw_spin_trylock, _raw_spin_lock); +} + +EXPORT_SYMBOL(_spin_lock_nest_lock); + #endif void __lockfunc _spin_unlock(spinlock_t *lock) diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c index 738b411ff2d..af3c7cea258 100644 --- a/kernel/stop_machine.c +++ b/kernel/stop_machine.c @@ -1,4 +1,4 @@ -/* Copyright 2005 Rusty Russell rusty@rustcorp.com.au IBM Corporation. +/* Copyright 2008, 2005 Rusty Russell rusty@rustcorp.com.au IBM Corporation. * GPL v2 and any later version. */ #include <linux/cpu.h> @@ -13,204 +13,177 @@ #include <asm/atomic.h> #include <asm/uaccess.h> -/* Since we effect priority and affinity (both of which are visible - * to, and settable by outside processes) we do indirection via a - * kthread. */ - -/* Thread to stop each CPU in user context. */ +/* This controls the threads on each CPU. */ enum stopmachine_state { - STOPMACHINE_WAIT, + /* Dummy starting state for thread. */ + STOPMACHINE_NONE, + /* Awaiting everyone to be scheduled. */ STOPMACHINE_PREPARE, + /* Disable interrupts. */ STOPMACHINE_DISABLE_IRQ, + /* Run the function */ + STOPMACHINE_RUN, + /* Exit */ STOPMACHINE_EXIT, }; +static enum stopmachine_state state; -static enum stopmachine_state stopmachine_state; -static unsigned int stopmachine_num_threads; -static atomic_t stopmachine_thread_ack; - -static int stopmachine(void *cpu) -{ - int irqs_disabled = 0; - int prepared = 0; - cpumask_of_cpu_ptr(cpumask, (int)(long)cpu); - - set_cpus_allowed_ptr(current, cpumask); - - /* Ack: we are alive */ - smp_mb(); /* Theoretically the ack = 0 might not be on this CPU yet. */ - atomic_inc(&stopmachine_thread_ack); - - /* Simple state machine */ - while (stopmachine_state != STOPMACHINE_EXIT) { - if (stopmachine_state == STOPMACHINE_DISABLE_IRQ - && !irqs_disabled) { - local_irq_disable(); - hard_irq_disable(); - irqs_disabled = 1; - /* Ack: irqs disabled. */ - smp_mb(); /* Must read state first. */ - atomic_inc(&stopmachine_thread_ack); - } else if (stopmachine_state == STOPMACHINE_PREPARE - && !prepared) { - /* Everyone is in place, hold CPU. */ - preempt_disable(); - prepared = 1; - smp_mb(); /* Must read state first. */ - atomic_inc(&stopmachine_thread_ack); - } - /* Yield in first stage: migration threads need to - * help our sisters onto their CPUs. */ - if (!prepared && !irqs_disabled) - yield(); - cpu_relax(); - } - - /* Ack: we are exiting. */ - smp_mb(); /* Must read state first. */ - atomic_inc(&stopmachine_thread_ack); - - if (irqs_disabled) - local_irq_enable(); - if (prepared) - preempt_enable(); +struct stop_machine_data { + int (*fn)(void *); + void *data; + int fnret; +}; - return 0; -} +/* Like num_online_cpus(), but hotplug cpu uses us, so we need this. */ +static unsigned int num_threads; +static atomic_t thread_ack; +static struct completion finished; +static DEFINE_MUTEX(lock); -/* Change the thread state */ -static void stopmachine_set_state(enum stopmachine_state state) +static void set_state(enum stopmachine_state newstate) { - atomic_set(&stopmachine_thread_ack, 0); + /* Reset ack counter. */ + atomic_set(&thread_ack, num_threads); smp_wmb(); - stopmachine_state = state; - while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) - cpu_relax(); + state = newstate; } -static int stop_machine(void) +/* Last one to ack a state moves to the next state. */ +static void ack_state(void) { - int i, ret = 0; - - atomic_set(&stopmachine_thread_ack, 0); - stopmachine_num_threads = 0; - stopmachine_state = STOPMACHINE_WAIT; - - for_each_online_cpu(i) { - if (i == raw_smp_processor_id()) - continue; - ret = kernel_thread(stopmachine, (void *)(long)i,CLONE_KERNEL); - if (ret < 0) - break; - stopmachine_num_threads++; - } - - /* Wait for them all to come to life. */ - while (atomic_read(&stopmachine_thread_ack) != stopmachine_num_threads) { - yield(); - cpu_relax(); + if (atomic_dec_and_test(&thread_ack)) { + /* If we're the last one to ack the EXIT, we're finished. */ + if (state == STOPMACHINE_EXIT) + complete(&finished); + else + set_state(state + 1); } +} - /* If some failed, kill them all. */ - if (ret < 0) { - stopmachine_set_state(STOPMACHINE_EXIT); - return ret; - } +/* This is the actual thread which stops the CPU. It exits by itself rather + * than waiting for kthread_stop(), because it's easier for hotplug CPU. */ +static int stop_cpu(struct stop_machine_data *smdata) +{ + enum stopmachine_state curstate = STOPMACHINE_NONE; - /* Now they are all started, make them hold the CPUs, ready. */ - preempt_disable(); - stopmachine_set_state(STOPMACHINE_PREPARE); + /* Simple state machine */ + do { + /* Chill out and ensure we re-read stopmachine_state. */ + cpu_relax(); + if (state != curstate) { + curstate = state; + switch (curstate) { + case STOPMACHINE_DISABLE_IRQ: + local_irq_disable(); + hard_irq_disable(); + break; + case STOPMACHINE_RUN: + /* |= allows error detection if functions on + * multiple CPUs. */ + smdata->fnret |= smdata->fn(smdata->data); + break; + default: + break; + } + ack_state(); + } + } while (curstate != STOPMACHINE_EXIT); - /* Make them disable irqs. */ - local_irq_disable(); - hard_irq_disable(); - stopmachine_set_state(STOPMACHINE_DISABLE_IRQ); + local_irq_enable(); + do_exit(0); +} +/* Callback for CPUs which aren't supposed to do anything. */ +static int chill(void *unused) +{ return 0; } -static void restart_machine(void) +int __stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus) { - stopmachine_set_state(STOPMACHINE_EXIT); - local_irq_enable(); - preempt_enable_no_resched(); -} + int i, err; + struct stop_machine_data active, idle; + struct task_struct **threads; + + active.fn = fn; + active.data = data; + active.fnret = 0; + idle.fn = chill; + idle.data = NULL; + + /* This could be too big for stack on large machines. */ + threads = kcalloc(NR_CPUS, sizeof(threads[0]), GFP_KERNEL); + if (!threads) + return -ENOMEM; + + /* Set up initial state. */ + mutex_lock(&lock); + init_completion(&finished); + num_threads = num_online_cpus(); + set_state(STOPMACHINE_PREPARE); -struct stop_machine_data { - int (*fn)(void *); - void *data; - struct completion done; -}; + for_each_online_cpu(i) { + struct stop_machine_data *smdata = &idle; + struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; -static int do_stop(void *_smdata) -{ - struct stop_machine_data *smdata = _smdata; - int ret; + if (!cpus) { + if (i == first_cpu(cpu_online_map)) + smdata = &active; + } else { + if (cpu_isset(i, *cpus)) + smdata = &active; + } - ret = stop_machine(); - if (ret == 0) { - ret = smdata->fn(smdata->data); - restart_machine(); - } + threads[i] = kthread_create((void *)stop_cpu, smdata, "kstop%u", + i); + if (IS_ERR(threads[i])) { + err = PTR_ERR(threads[i]); + threads[i] = NULL; + goto kill_threads; + } - /* We're done: you can kthread_stop us now */ - complete(&smdata->done); + /* Place it onto correct cpu. */ + kthread_bind(threads[i], i); - /* Wait for kthread_stop */ - set_current_state(TASK_INTERRUPTIBLE); - while (!kthread_should_stop()) { - schedule(); - set_current_state(TASK_INTERRUPTIBLE); + /* Make it highest prio. */ + if (sched_setscheduler_nocheck(threads[i], SCHED_FIFO, ¶m)) + BUG(); } - __set_current_state(TASK_RUNNING); - return ret; -} -struct task_struct *__stop_machine_run(int (*fn)(void *), void *data, - unsigned int cpu) -{ - static DEFINE_MUTEX(stopmachine_mutex); - struct stop_machine_data smdata; - struct task_struct *p; + /* We've created all the threads. Wake them all: hold this CPU so one + * doesn't hit this CPU until we're ready. */ + get_cpu(); + for_each_online_cpu(i) + wake_up_process(threads[i]); - smdata.fn = fn; - smdata.data = data; - init_completion(&smdata.done); + /* This will release the thread on our CPU. */ + put_cpu(); + wait_for_completion(&finished); + mutex_unlock(&lock); - mutex_lock(&stopmachine_mutex); + kfree(threads); - /* If they don't care which CPU fn runs on, bind to any online one. */ - if (cpu == NR_CPUS) - cpu = raw_smp_processor_id(); + return active.fnret; - p = kthread_create(do_stop, &smdata, "kstopmachine"); - if (!IS_ERR(p)) { - struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 }; +kill_threads: + for_each_online_cpu(i) + if (threads[i]) + kthread_stop(threads[i]); + mutex_unlock(&lock); - /* One high-prio thread per cpu. We'll do this one. */ - sched_setscheduler_nocheck(p, SCHED_FIFO, ¶m); - kthread_bind(p, cpu); - wake_up_process(p); - wait_for_completion(&smdata.done); - } - mutex_unlock(&stopmachine_mutex); - return p; + kfree(threads); + return err; } -int stop_machine_run(int (*fn)(void *), void *data, unsigned int cpu) +int stop_machine(int (*fn)(void *), void *data, const cpumask_t *cpus) { - struct task_struct *p; int ret; /* No CPUs can come up or down during this. */ get_online_cpus(); - p = __stop_machine_run(fn, data, cpu); - if (!IS_ERR(p)) - ret = kthread_stop(p); - else - ret = PTR_ERR(p); + ret = __stop_machine(fn, data, cpus); put_online_cpus(); return ret; } -EXPORT_SYMBOL_GPL(stop_machine_run); +EXPORT_SYMBOL_GPL(stop_machine); diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c index bf43284d685..80c4336f418 100644 --- a/kernel/time/tick-common.c +++ b/kernel/time/tick-common.c @@ -196,12 +196,10 @@ static int tick_check_new_device(struct clock_event_device *newdev) struct tick_device *td; int cpu, ret = NOTIFY_OK; unsigned long flags; - cpumask_of_cpu_ptr_declare(cpumask); spin_lock_irqsave(&tick_device_lock, flags); cpu = smp_processor_id(); - cpumask_of_cpu_ptr_next(cpumask, cpu); if (!cpu_isset(cpu, newdev->cpumask)) goto out_bc; @@ -209,7 +207,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) curdev = td->evtdev; /* cpu local device ? */ - if (!cpus_equal(newdev->cpumask, *cpumask)) { + if (!cpus_equal(newdev->cpumask, cpumask_of_cpu(cpu))) { /* * If the cpu affinity of the device interrupt can not @@ -222,7 +220,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) * If we have a cpu local device already, do not replace it * by a non cpu local device */ - if (curdev && cpus_equal(curdev->cpumask, *cpumask)) + if (curdev && cpus_equal(curdev->cpumask, cpumask_of_cpu(cpu))) goto out_bc; } @@ -254,7 +252,7 @@ static int tick_check_new_device(struct clock_event_device *newdev) curdev = NULL; } clockevents_exchange_device(curdev, newdev); - tick_setup_device(td, newdev, cpu, cpumask); + tick_setup_device(td, newdev, cpu, &cpumask_of_cpu(cpu)); if (newdev->features & CLOCK_EVT_FEAT_ONESHOT) tick_oneshot_notify(); diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c index 825b4c00fe4..f5da526424a 100644 --- a/kernel/time/tick-sched.c +++ b/kernel/time/tick-sched.c @@ -289,7 +289,6 @@ void tick_nohz_stop_sched_tick(int inidle) ts->tick_stopped = 1; ts->idle_jiffies = last_jiffies; rcu_enter_nohz(); - sched_clock_tick_stop(cpu); } /* @@ -392,7 +391,6 @@ void tick_nohz_restart_sched_tick(void) select_nohz_load_balancer(0); now = ktime_get(); tick_do_update_jiffies64(now); - sched_clock_tick_start(cpu); cpu_clear(cpu, nohz_cpu_mask); /* diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c index 4231a3dc224..f6e3af31b40 100644 --- a/kernel/trace/ftrace.c +++ b/kernel/trace/ftrace.c @@ -587,7 +587,7 @@ static int __ftrace_modify_code(void *data) static void ftrace_run_update_code(int command) { - stop_machine_run(__ftrace_modify_code, &command, NR_CPUS); + stop_machine(__ftrace_modify_code, &command, NULL); } void ftrace_disable_daemon(void) @@ -787,7 +787,7 @@ static int ftrace_update_code(void) !ftrace_enabled || !ftraced_trigger) return 0; - stop_machine_run(__ftrace_update_code, NULL, NR_CPUS); + stop_machine(__ftrace_update_code, NULL, NULL); return 1; } @@ -1564,7 +1564,7 @@ static int __init ftrace_dynamic_init(void) addr = (unsigned long)ftrace_record_ip; - stop_machine_run(ftrace_dyn_arch_init, &addr, NR_CPUS); + stop_machine(ftrace_dyn_arch_init, &addr, NULL); /* ftrace_dyn_arch_init places the return code in addr */ if (addr) { diff --git a/kernel/trace/trace_sysprof.c b/kernel/trace/trace_sysprof.c index ce2d723c10e..bb948e52ce2 100644 --- a/kernel/trace/trace_sysprof.c +++ b/kernel/trace/trace_sysprof.c @@ -213,9 +213,7 @@ static void start_stack_timers(void) int cpu; for_each_online_cpu(cpu) { - cpumask_of_cpu_ptr(new_mask, cpu); - - set_cpus_allowed_ptr(current, new_mask); + set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); start_stack_timer(cpu); } set_cpus_allowed_ptr(current, &saved_mask); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index ec7e4f62aaf..4048e92aa04 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -290,11 +290,11 @@ static void run_workqueue(struct cpu_workqueue_struct *cwq) BUG_ON(get_wq_data(work) != cwq); work_clear_pending(work); - lock_acquire(&cwq->wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_); - lock_acquire(&lockdep_map, 0, 0, 0, 2, _THIS_IP_); + lock_map_acquire(&cwq->wq->lockdep_map); + lock_map_acquire(&lockdep_map); f(work); - lock_release(&lockdep_map, 1, _THIS_IP_); - lock_release(&cwq->wq->lockdep_map, 1, _THIS_IP_); + lock_map_release(&lockdep_map); + lock_map_release(&cwq->wq->lockdep_map); if (unlikely(in_atomic() || lockdep_depth(current) > 0)) { printk(KERN_ERR "BUG: workqueue leaked lock or atomic: " @@ -413,8 +413,8 @@ void flush_workqueue(struct workqueue_struct *wq) int cpu; might_sleep(); - lock_acquire(&wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_); - lock_release(&wq->lockdep_map, 1, _THIS_IP_); + lock_map_acquire(&wq->lockdep_map); + lock_map_release(&wq->lockdep_map); for_each_cpu_mask_nr(cpu, *cpu_map) flush_cpu_workqueue(per_cpu_ptr(wq->cpu_wq, cpu)); } @@ -441,8 +441,8 @@ int flush_work(struct work_struct *work) if (!cwq) return 0; - lock_acquire(&cwq->wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_); - lock_release(&cwq->wq->lockdep_map, 1, _THIS_IP_); + lock_map_acquire(&cwq->wq->lockdep_map); + lock_map_release(&cwq->wq->lockdep_map); prev = NULL; spin_lock_irq(&cwq->lock); @@ -536,8 +536,8 @@ static void wait_on_work(struct work_struct *work) might_sleep(); - lock_acquire(&work->lockdep_map, 0, 0, 0, 2, _THIS_IP_); - lock_release(&work->lockdep_map, 1, _THIS_IP_); + lock_map_acquire(&work->lockdep_map); + lock_map_release(&work->lockdep_map); cwq = get_wq_data(work); if (!cwq) @@ -830,10 +830,21 @@ struct workqueue_struct *__create_workqueue_key(const char *name, start_workqueue_thread(cwq, -1); } else { cpu_maps_update_begin(); + /* + * We must place this wq on list even if the code below fails. + * cpu_down(cpu) can remove cpu from cpu_populated_map before + * destroy_workqueue() takes the lock, in that case we leak + * cwq[cpu]->thread. + */ spin_lock(&workqueue_lock); list_add(&wq->list, &workqueues); spin_unlock(&workqueue_lock); - + /* + * We must initialize cwqs for each possible cpu even if we + * are going to call destroy_workqueue() finally. Otherwise + * cpu_up() can hit the uninitialized cwq once we drop the + * lock. + */ for_each_possible_cpu(cpu) { cwq = init_cpu_workqueue(wq, cpu); if (err || !cpu_online(cpu)) @@ -861,8 +872,8 @@ static void cleanup_workqueue_thread(struct cpu_workqueue_struct *cwq) if (cwq->thread == NULL) return; - lock_acquire(&cwq->wq->lockdep_map, 0, 0, 0, 2, _THIS_IP_); - lock_release(&cwq->wq->lockdep_map, 1, _THIS_IP_); + lock_map_acquire(&cwq->wq->lockdep_map); + lock_map_release(&cwq->wq->lockdep_map); flush_cpu_workqueue(cwq); /* |