diff options
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/configs/tiny.config | 1 | ||||
-rw-r--r-- | kernel/dma/debug.c | 9 | ||||
-rw-r--r-- | kernel/dma/direct.c | 3 | ||||
-rw-r--r-- | kernel/events/core.c | 85 | ||||
-rw-r--r-- | kernel/extable.c | 3 | ||||
-rw-r--r-- | kernel/fork.c | 10 | ||||
-rw-r--r-- | kernel/gcov/fs.c | 4 | ||||
-rw-r--r-- | kernel/gcov/gcc_3_4.c | 6 | ||||
-rw-r--r-- | kernel/gcov/gcc_4_7.c | 2 | ||||
-rw-r--r-- | kernel/kallsyms.c | 2 | ||||
-rw-r--r-- | kernel/kmod.c | 6 | ||||
-rw-r--r-- | kernel/locking/lockdep.c | 51 | ||||
-rw-r--r-- | kernel/locking/percpu-rwsem.c | 3 | ||||
-rw-r--r-- | kernel/module.c | 5 | ||||
-rw-r--r-- | kernel/pid.c | 1 | ||||
-rw-r--r-- | kernel/power/user.c | 101 | ||||
-rw-r--r-- | kernel/printk/internal.h | 5 | ||||
-rw-r--r-- | kernel/printk/printk.c | 34 | ||||
-rw-r--r-- | kernel/printk/printk_safe.c | 11 | ||||
-rw-r--r-- | kernel/sched/core.c | 10 | ||||
-rw-r--r-- | kernel/sched/debug.c | 44 | ||||
-rw-r--r-- | kernel/sched/fair.c | 48 | ||||
-rw-r--r-- | kernel/sched/sched.h | 7 | ||||
-rw-r--r-- | kernel/workqueue.c | 6 |
24 files changed, 253 insertions, 204 deletions
diff --git a/kernel/configs/tiny.config b/kernel/configs/tiny.config index 7fa0c4ae6394..8a44b93da0f3 100644 --- a/kernel/configs/tiny.config +++ b/kernel/configs/tiny.config @@ -6,7 +6,6 @@ CONFIG_CC_OPTIMIZE_FOR_SIZE=y CONFIG_KERNEL_XZ=y # CONFIG_KERNEL_LZO is not set # CONFIG_KERNEL_LZ4 is not set -CONFIG_OPTIMIZE_INLINING=y # CONFIG_SLAB is not set # CONFIG_SLUB is not set CONFIG_SLOB=y diff --git a/kernel/dma/debug.c b/kernel/dma/debug.c index 2031ed1ad7fa..9e1777c81f55 100644 --- a/kernel/dma/debug.c +++ b/kernel/dma/debug.c @@ -137,9 +137,12 @@ static const char *const maperr2str[] = { [MAP_ERR_CHECKED] = "dma map error checked", }; -static const char *type2name[5] = { "single", "page", - "scather-gather", "coherent", - "resource" }; +static const char *type2name[] = { + [dma_debug_single] = "single", + [dma_debug_sg] = "scather-gather", + [dma_debug_coherent] = "coherent", + [dma_debug_resource] = "resource", +}; static const char *dir2name[4] = { "DMA_BIDIRECTIONAL", "DMA_TO_DEVICE", "DMA_FROM_DEVICE", "DMA_NONE" }; diff --git a/kernel/dma/direct.c b/kernel/dma/direct.c index a8560052a915..8f4bbdaf965e 100644 --- a/kernel/dma/direct.c +++ b/kernel/dma/direct.c @@ -39,7 +39,8 @@ static inline struct page *dma_direct_to_page(struct device *dev, u64 dma_direct_get_required_mask(struct device *dev) { - u64 max_dma = phys_to_dma_direct(dev, (max_pfn - 1) << PAGE_SHIFT); + phys_addr_t phys = (phys_addr_t)(max_pfn - 1) << PAGE_SHIFT; + u64 max_dma = phys_to_dma_direct(dev, phys); return (1ULL << (fls64(max_dma) - 1)) * 2 - 1; } diff --git a/kernel/events/core.c b/kernel/events/core.c index 81e6d80cb219..bc9b98a9af9a 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -28,6 +28,7 @@ #include <linux/export.h> #include <linux/vmalloc.h> #include <linux/hardirq.h> +#include <linux/hugetlb.h> #include <linux/rculist.h> #include <linux/uaccess.h> #include <linux/syscalls.h> @@ -982,16 +983,10 @@ perf_cgroup_set_shadow_time(struct perf_event *event, u64 now) event->shadow_ctx_time = now - t->timestamp; } -/* - * Update cpuctx->cgrp so that it is set when first cgroup event is added and - * cleared when last cgroup event is removed. - */ static inline void -list_update_cgroup_event(struct perf_event *event, - struct perf_event_context *ctx, bool add) +perf_cgroup_event_enable(struct perf_event *event, struct perf_event_context *ctx) { struct perf_cpu_context *cpuctx; - struct list_head *cpuctx_entry; if (!is_cgroup_event(event)) return; @@ -1008,28 +1003,41 @@ list_update_cgroup_event(struct perf_event *event, * because if the first would mismatch, the second would not try again * and we would leave cpuctx->cgrp unset. */ - if (add && !cpuctx->cgrp) { + if (ctx->is_active && !cpuctx->cgrp) { struct perf_cgroup *cgrp = perf_cgroup_from_task(current, ctx); if (cgroup_is_descendant(cgrp->css.cgroup, event->cgrp->css.cgroup)) cpuctx->cgrp = cgrp; } - if (add && ctx->nr_cgroups++) + if (ctx->nr_cgroups++) return; - else if (!add && --ctx->nr_cgroups) + + list_add(&cpuctx->cgrp_cpuctx_entry, + per_cpu_ptr(&cgrp_cpuctx_list, event->cpu)); +} + +static inline void +perf_cgroup_event_disable(struct perf_event *event, struct perf_event_context *ctx) +{ + struct perf_cpu_context *cpuctx; + + if (!is_cgroup_event(event)) + return; + + /* + * Because cgroup events are always per-cpu events, + * @ctx == &cpuctx->ctx. + */ + cpuctx = container_of(ctx, struct perf_cpu_context, ctx); + + if (--ctx->nr_cgroups) return; - /* no cgroup running */ - if (!add) + if (ctx->is_active && cpuctx->cgrp) cpuctx->cgrp = NULL; - cpuctx_entry = &cpuctx->cgrp_cpuctx_entry; - if (add) - list_add(cpuctx_entry, - per_cpu_ptr(&cgrp_cpuctx_list, event->cpu)); - else - list_del(cpuctx_entry); + list_del(&cpuctx->cgrp_cpuctx_entry); } #else /* !CONFIG_CGROUP_PERF */ @@ -1095,11 +1103,14 @@ static inline u64 perf_cgroup_event_time(struct perf_event *event) } static inline void -list_update_cgroup_event(struct perf_event *event, - struct perf_event_context *ctx, bool add) +perf_cgroup_event_enable(struct perf_event *event, struct perf_event_context *ctx) { } +static inline void +perf_cgroup_event_disable(struct perf_event *event, struct perf_event_context *ctx) +{ +} #endif /* @@ -1790,13 +1801,14 @@ list_add_event(struct perf_event *event, struct perf_event_context *ctx) add_event_to_groups(event, ctx); } - list_update_cgroup_event(event, ctx, true); - list_add_rcu(&event->event_entry, &ctx->event_list); ctx->nr_events++; if (event->attr.inherit_stat) ctx->nr_stat++; + if (event->state > PERF_EVENT_STATE_OFF) + perf_cgroup_event_enable(event, ctx); + ctx->generation++; } @@ -1975,8 +1987,6 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) event->attach_state &= ~PERF_ATTACH_CONTEXT; - list_update_cgroup_event(event, ctx, false); - ctx->nr_events--; if (event->attr.inherit_stat) ctx->nr_stat--; @@ -1993,8 +2003,10 @@ list_del_event(struct perf_event *event, struct perf_event_context *ctx) * of error state is by explicit re-enabling * of the event */ - if (event->state > PERF_EVENT_STATE_OFF) + if (event->state > PERF_EVENT_STATE_OFF) { + perf_cgroup_event_disable(event, ctx); perf_event_set_state(event, PERF_EVENT_STATE_OFF); + } ctx->generation++; } @@ -2225,6 +2237,7 @@ event_sched_out(struct perf_event *event, if (READ_ONCE(event->pending_disable) >= 0) { WRITE_ONCE(event->pending_disable, -1); + perf_cgroup_event_disable(event, ctx); state = PERF_EVENT_STATE_OFF; } perf_event_set_state(event, state); @@ -2362,6 +2375,7 @@ static void __perf_event_disable(struct perf_event *event, event_sched_out(event, cpuctx, ctx); perf_event_set_state(event, PERF_EVENT_STATE_OFF); + perf_cgroup_event_disable(event, ctx); } /* @@ -2745,7 +2759,7 @@ static int __perf_install_in_context(void *info) } #ifdef CONFIG_CGROUP_PERF - if (is_cgroup_event(event)) { + if (event->state > PERF_EVENT_STATE_OFF && is_cgroup_event(event)) { /* * If the current cgroup doesn't match the event's * cgroup, we should not try to schedule it. @@ -2905,6 +2919,7 @@ static void __perf_event_enable(struct perf_event *event, ctx_sched_out(ctx, cpuctx, EVENT_TIME); perf_event_set_state(event, PERF_EVENT_STATE_INACTIVE); + perf_cgroup_event_enable(event, ctx); if (!ctx->is_active) return; @@ -3507,7 +3522,8 @@ static void cpu_ctx_sched_out(struct perf_cpu_context *cpuctx, static bool perf_less_group_idx(const void *l, const void *r) { - const struct perf_event *le = l, *re = r; + const struct perf_event *le = *(const struct perf_event **)l; + const struct perf_event *re = *(const struct perf_event **)r; return le->group_index < re->group_index; } @@ -3615,8 +3631,10 @@ static int merge_sched_in(struct perf_event *event, void *data) } if (event->state == PERF_EVENT_STATE_INACTIVE) { - if (event->attr.pinned) + if (event->attr.pinned) { + perf_cgroup_event_disable(event, ctx); perf_event_set_state(event, PERF_EVENT_STATE_ERROR); + } *can_add_hw = 0; ctx->rotate_necessary = 1; @@ -6916,9 +6934,12 @@ static u64 perf_virt_to_phys(u64 virt) * Try IRQ-safe __get_user_pages_fast first. * If failed, leave phys_addr as 0. */ - if ((current->mm != NULL) && - (__get_user_pages_fast(virt, 1, 0, &p) == 1)) - phys_addr = page_to_phys(p) + virt % PAGE_SIZE; + if (current->mm != NULL) { + pagefault_disable(); + if (__get_user_pages_fast(virt, 1, 0, &p) == 1) + phys_addr = page_to_phys(p) + virt % PAGE_SIZE; + pagefault_enable(); + } if (p) put_page(p); @@ -7973,7 +7994,7 @@ static void perf_event_mmap_event(struct perf_mmap_event *mmap_event) flags |= MAP_EXECUTABLE; if (vma->vm_flags & VM_LOCKED) flags |= MAP_LOCKED; - if (vma->vm_flags & VM_HUGETLB) + if (is_vm_hugetlb_page(vma)) flags |= MAP_HUGETLB; if (file) { diff --git a/kernel/extable.c b/kernel/extable.c index 7681f87e89dd..b0ea5eb0c3b4 100644 --- a/kernel/extable.c +++ b/kernel/extable.c @@ -34,7 +34,8 @@ u32 __initdata __visible main_extable_sort_needed = 1; /* Sort the kernel's built-in exception table */ void __init sort_main_extable(void) { - if (main_extable_sort_needed && __stop___ex_table > __start___ex_table) { + if (main_extable_sort_needed && + &__stop___ex_table > &__start___ex_table) { pr_notice("Sorting __ex_table...\n"); sort_extable(__start___ex_table, __stop___ex_table); } diff --git a/kernel/fork.c b/kernel/fork.c index d2a967bf85d5..4385f3d639f2 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -361,6 +361,7 @@ struct vm_area_struct *vm_area_dup(struct vm_area_struct *orig) if (new) { *new = *orig; INIT_LIST_HEAD(&new->anon_vma_chain); + new->vm_next = new->vm_prev = NULL; } return new; } @@ -553,14 +554,15 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm, if (retval) goto fail_nomem_anon_vma_fork; if (tmp->vm_flags & VM_WIPEONFORK) { - /* VM_WIPEONFORK gets a clean slate in the child. */ + /* + * VM_WIPEONFORK gets a clean slate in the child. + * Don't prepare anon_vma until fault since we don't + * copy page for current vma. + */ tmp->anon_vma = NULL; - if (anon_vma_prepare(tmp)) - goto fail_nomem_anon_vma_fork; } else if (anon_vma_fork(tmp, mpnt)) goto fail_nomem_anon_vma_fork; tmp->vm_flags &= ~(VM_LOCKED | VM_LOCKONFAULT); - tmp->vm_next = tmp->vm_prev = NULL; file = tmp->vm_file; if (file) { struct inode *inode = file_inode(file); diff --git a/kernel/gcov/fs.c b/kernel/gcov/fs.c index e5eb5ea7ea59..82babf5aa077 100644 --- a/kernel/gcov/fs.c +++ b/kernel/gcov/fs.c @@ -58,7 +58,7 @@ struct gcov_node { struct dentry *dentry; struct dentry **links; int num_loaded; - char name[0]; + char name[]; }; static const char objtree[] = OBJTREE; @@ -108,9 +108,9 @@ static void *gcov_seq_next(struct seq_file *seq, void *data, loff_t *pos) { struct gcov_iterator *iter = data; + (*pos)++; if (gcov_iter_next(iter)) return NULL; - (*pos)++; return iter; } diff --git a/kernel/gcov/gcc_3_4.c b/kernel/gcov/gcc_3_4.c index 801ee4b0b969..acb83558e5df 100644 --- a/kernel/gcov/gcc_3_4.c +++ b/kernel/gcov/gcc_3_4.c @@ -38,7 +38,7 @@ static struct gcov_info *gcov_info_head; struct gcov_fn_info { unsigned int ident; unsigned int checksum; - unsigned int n_ctrs[0]; + unsigned int n_ctrs[]; }; /** @@ -78,7 +78,7 @@ struct gcov_info { unsigned int n_functions; const struct gcov_fn_info *functions; unsigned int ctr_mask; - struct gcov_ctr_info counts[0]; + struct gcov_ctr_info counts[]; }; /** @@ -352,7 +352,7 @@ struct gcov_iterator { unsigned int count; int num_types; - struct type_info type_info[0]; + struct type_info type_info[]; }; static struct gcov_fn_info *get_func(struct gcov_iterator *iter) diff --git a/kernel/gcov/gcc_4_7.c b/kernel/gcov/gcc_4_7.c index ec37563674d6..908fdf5098c3 100644 --- a/kernel/gcov/gcc_4_7.c +++ b/kernel/gcov/gcc_4_7.c @@ -68,7 +68,7 @@ struct gcov_fn_info { unsigned int ident; unsigned int lineno_checksum; unsigned int cfg_checksum; - struct gcov_ctr_info ctrs[0]; + struct gcov_ctr_info ctrs[]; }; /** diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index a9b3f660dee7..16c8c605f4b0 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -175,7 +175,6 @@ unsigned long kallsyms_lookup_name(const char *name) } return module_kallsyms_lookup_name(name); } -EXPORT_SYMBOL_GPL(kallsyms_lookup_name); int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, unsigned long), @@ -194,7 +193,6 @@ int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, } return module_kallsyms_on_each_symbol(fn, data); } -EXPORT_SYMBOL_GPL(kallsyms_on_each_symbol); static unsigned long get_symbol_pos(unsigned long addr, unsigned long *symbolsize, diff --git a/kernel/kmod.c b/kernel/kmod.c index bc6addd9152b..37c3c4b97b8e 100644 --- a/kernel/kmod.c +++ b/kernel/kmod.c @@ -35,7 +35,7 @@ * (u64) THREAD_SIZE * 8UL); * * If you need less than 50 threads would mean we're dealing with systems - * smaller than 3200 pages. This assuems you are capable of having ~13M memory, + * smaller than 3200 pages. This assumes you are capable of having ~13M memory, * and this would only be an be an upper limit, after which the OOM killer * would take effect. Systems like these are very unlikely if modules are * enabled. @@ -120,7 +120,7 @@ out: * invoke it. * * If module auto-loading support is disabled then this function - * becomes a no-operation. + * simply returns -ENOENT. */ int __request_module(bool wait, const char *fmt, ...) { @@ -137,7 +137,7 @@ int __request_module(bool wait, const char *fmt, ...) WARN_ON_ONCE(wait && current_is_async()); if (!modprobe_path[0]) - return 0; + return -ENOENT; va_start(args, fmt); ret = vsnprintf(module_name, MODULE_NAME_LEN, fmt, args); diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c index 1511690e4de7..ac10db66cc63 100644 --- a/kernel/locking/lockdep.c +++ b/kernel/locking/lockdep.c @@ -3952,10 +3952,36 @@ static int mark_lock(struct task_struct *curr, struct held_lock *this, return ret; } +static inline short task_wait_context(struct task_struct *curr) +{ + /* + * Set appropriate wait type for the context; for IRQs we have to take + * into account force_irqthread as that is implied by PREEMPT_RT. + */ + if (curr->hardirq_context) { + /* + * Check if force_irqthreads will run us threaded. + */ + if (curr->hardirq_threaded || curr->irq_config) + return LD_WAIT_CONFIG; + + return LD_WAIT_SPIN; + } else if (curr->softirq_context) { + /* + * Softirqs are always threaded. + */ + return LD_WAIT_CONFIG; + } + + return LD_WAIT_MAX; +} + static int print_lock_invalid_wait_context(struct task_struct *curr, struct held_lock *hlock) { + short curr_inner; + if (!debug_locks_off()) return 0; if (debug_locks_silent) @@ -3971,6 +3997,10 @@ print_lock_invalid_wait_context(struct task_struct *curr, print_lock(hlock); pr_warn("other info that might help us debug this:\n"); + + curr_inner = task_wait_context(curr); + pr_warn("context-{%d:%d}\n", curr_inner, curr_inner); + lockdep_print_held_locks(curr); pr_warn("stack backtrace:\n"); @@ -4017,26 +4047,7 @@ static int check_wait_context(struct task_struct *curr, struct held_lock *next) } depth++; - /* - * Set appropriate wait type for the context; for IRQs we have to take - * into account force_irqthread as that is implied by PREEMPT_RT. - */ - if (curr->hardirq_context) { - /* - * Check if force_irqthreads will run us threaded. - */ - if (curr->hardirq_threaded || curr->irq_config) - curr_inner = LD_WAIT_CONFIG; - else - curr_inner = LD_WAIT_SPIN; - } else if (curr->softirq_context) { - /* - * Softirqs are always threaded. - */ - curr_inner = LD_WAIT_CONFIG; - } else { - curr_inner = LD_WAIT_MAX; - } + curr_inner = task_wait_context(curr); for (; depth < curr->lockdep_depth; depth++) { struct held_lock *prev = curr->held_locks + depth; diff --git a/kernel/locking/percpu-rwsem.c b/kernel/locking/percpu-rwsem.c index a008a1ba21a7..8bbafe3e5203 100644 --- a/kernel/locking/percpu-rwsem.c +++ b/kernel/locking/percpu-rwsem.c @@ -118,14 +118,15 @@ static int percpu_rwsem_wake_function(struct wait_queue_entry *wq_entry, unsigned int mode, int wake_flags, void *key) { - struct task_struct *p = get_task_struct(wq_entry->private); bool reader = wq_entry->flags & WQ_FLAG_CUSTOM; struct percpu_rw_semaphore *sem = key; + struct task_struct *p; /* concurrent against percpu_down_write(), can get stolen */ if (!__percpu_rwsem_trylock(sem, reader)) return 1; + p = get_task_struct(wq_entry->private); list_del_init(&wq_entry->entry); smp_store_release(&wq_entry->private, NULL); diff --git a/kernel/module.c b/kernel/module.c index 33569a01d6e1..646f1e2330d2 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -1515,7 +1515,7 @@ struct module_sect_attr { struct module_sect_attrs { struct attribute_group grp; unsigned int nsections; - struct module_sect_attr attrs[0]; + struct module_sect_attr attrs[]; }; static ssize_t module_sect_show(struct module_attribute *mattr, @@ -1608,7 +1608,7 @@ static void remove_sect_attrs(struct module *mod) struct module_notes_attrs { struct kobject *dir; unsigned int notes; - struct bin_attribute attrs[0]; + struct bin_attribute attrs[]; }; static ssize_t module_notes_read(struct file *filp, struct kobject *kobj, @@ -4355,6 +4355,7 @@ static int modules_open(struct inode *inode, struct file *file) } static const struct proc_ops modules_proc_ops = { + .proc_flags = PROC_ENTRY_PERMANENT, .proc_open = modules_open, .proc_read = seq_read, .proc_lseek = seq_lseek, diff --git a/kernel/pid.c b/kernel/pid.c index bc21c0fb26d8..c835b844aca7 100644 --- a/kernel/pid.c +++ b/kernel/pid.c @@ -256,6 +256,7 @@ struct pid *alloc_pid(struct pid_namespace *ns, pid_t *set_tid, get_pid_ns(ns); refcount_set(&pid->count, 1); + spin_lock_init(&pid->lock); for (type = 0; type < PIDTYPE_MAX; ++type) INIT_HLIST_HEAD(&pid->tasks[type]); diff --git a/kernel/power/user.c b/kernel/power/user.c index ef90eb1fb86e..7959449765d9 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -196,6 +196,50 @@ unlock: return res; } +struct compat_resume_swap_area { + compat_loff_t offset; + u32 dev; +} __packed; + +static int snapshot_set_swap_area(struct snapshot_data *data, + void __user *argp) +{ + sector_t offset; + dev_t swdev; + + if (swsusp_swap_in_use()) + return -EPERM; + + if (in_compat_syscall()) { + struct compat_resume_swap_area swap_area; + + if (copy_from_user(&swap_area, argp, sizeof(swap_area))) + return -EFAULT; + swdev = new_decode_dev(swap_area.dev); + offset = swap_area.offset; + } else { + struct resume_swap_area swap_area; + + if (copy_from_user(&swap_area, argp, sizeof(swap_area))) + return -EFAULT; + swdev = new_decode_dev(swap_area.dev); + offset = swap_area.offset; + } + + /* + * User space encodes device types as two-byte values, + * so we need to recode them + */ + if (!swdev) { + data->swap = -1; + return -EINVAL; + } + data->swap = swap_type_of(swdev, offset, NULL); + if (data->swap < 0) + return -ENODEV; + return 0; +} + static long snapshot_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -351,34 +395,7 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, break; case SNAPSHOT_SET_SWAP_AREA: - if (swsusp_swap_in_use()) { - error = -EPERM; - } else { - struct resume_swap_area swap_area; - dev_t swdev; - - error = copy_from_user(&swap_area, (void __user *)arg, - sizeof(struct resume_swap_area)); - if (error) { - error = -EFAULT; - break; - } - - /* - * User space encodes device types as two-byte values, - * so we need to recode them - */ - swdev = new_decode_dev(swap_area.dev); - if (swdev) { - offset = swap_area.offset; - data->swap = swap_type_of(swdev, offset, NULL); - if (data->swap < 0) - error = -ENODEV; - } else { - data->swap = -1; - error = -EINVAL; - } - } + error = snapshot_set_swap_area(data, (void __user *)arg); break; default: @@ -393,12 +410,6 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, } #ifdef CONFIG_COMPAT - -struct compat_resume_swap_area { - compat_loff_t offset; - u32 dev; -} __packed; - static long snapshot_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { @@ -409,33 +420,13 @@ snapshot_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case SNAPSHOT_AVAIL_SWAP_SIZE: case SNAPSHOT_ALLOC_SWAP_PAGE: case SNAPSHOT_CREATE_IMAGE: + case SNAPSHOT_SET_SWAP_AREA: return snapshot_ioctl(file, cmd, (unsigned long) compat_ptr(arg)); - - case SNAPSHOT_SET_SWAP_AREA: { - struct compat_resume_swap_area __user *u_swap_area = - compat_ptr(arg); - struct resume_swap_area swap_area; - mm_segment_t old_fs; - int err; - - err = get_user(swap_area.offset, &u_swap_area->offset); - err |= get_user(swap_area.dev, &u_swap_area->dev); - if (err) - return -EFAULT; - old_fs = get_fs(); - set_fs(KERNEL_DS); - err = snapshot_ioctl(file, SNAPSHOT_SET_SWAP_AREA, - (unsigned long) &swap_area); - set_fs(old_fs); - return err; - } - default: return snapshot_ioctl(file, cmd, arg); } } - #endif /* CONFIG_COMPAT */ static const struct file_operations snapshot_fops = { diff --git a/kernel/printk/internal.h b/kernel/printk/internal.h index c8e6ab689d42..b2b0f526f249 100644 --- a/kernel/printk/internal.h +++ b/kernel/printk/internal.h @@ -23,6 +23,9 @@ __printf(1, 0) int vprintk_func(const char *fmt, va_list args); void __printk_safe_enter(void); void __printk_safe_exit(void); +void printk_safe_init(void); +bool printk_percpu_data_ready(void); + #define printk_safe_enter_irqsave(flags) \ do { \ local_irq_save(flags); \ @@ -64,4 +67,6 @@ __printf(1, 0) int vprintk_func(const char *fmt, va_list args) { return 0; } #define printk_safe_enter_irq() local_irq_disable() #define printk_safe_exit_irq() local_irq_enable() +static inline void printk_safe_init(void) { } +static inline bool printk_percpu_data_ready(void) { return false; } #endif /* CONFIG_PRINTK */ diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 633f41a11d75..9a9b6156270b 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -460,6 +460,18 @@ static char __log_buf[__LOG_BUF_LEN] __aligned(LOG_ALIGN); static char *log_buf = __log_buf; static u32 log_buf_len = __LOG_BUF_LEN; +/* + * We cannot access per-CPU data (e.g. per-CPU flush irq_work) before + * per_cpu_areas are initialised. This variable is set to true when + * it's safe to access per-CPU data. + */ +static bool __printk_percpu_data_ready __read_mostly; + +bool printk_percpu_data_ready(void) +{ + return __printk_percpu_data_ready; +} + /* Return log buffer address */ char *log_buf_addr_get(void) { @@ -1146,12 +1158,28 @@ static void __init log_buf_add_cpu(void) static inline void log_buf_add_cpu(void) {} #endif /* CONFIG_SMP */ +static void __init set_percpu_data_ready(void) +{ + printk_safe_init(); + /* Make sure we set this flag only after printk_safe() init is done */ + barrier(); + __printk_percpu_data_ready = true; +} + void __init setup_log_buf(int early) { unsigned long flags; char *new_log_buf; unsigned int free; + /* + * Some archs call setup_log_buf() multiple times - first is very + * early, e.g. from setup_arch(), and second - when percpu_areas + * are initialised. + */ + if (!early) + set_percpu_data_ready(); + if (log_buf != __log_buf) return; @@ -2975,6 +3003,9 @@ static DEFINE_PER_CPU(struct irq_work, wake_up_klogd_work) = { void wake_up_klogd(void) { + if (!printk_percpu_data_ready()) + return; + preempt_disable(); if (waitqueue_active(&log_wait)) { this_cpu_or(printk_pending, PRINTK_PENDING_WAKEUP); @@ -2985,6 +3016,9 @@ void wake_up_klogd(void) void defer_console_output(void) { + if (!printk_percpu_data_ready()) + return; + preempt_disable(); __this_cpu_or(printk_pending, PRINTK_PENDING_OUTPUT); irq_work_queue(this_cpu_ptr(&wake_up_klogd_work)); diff --git a/kernel/printk/printk_safe.c b/kernel/printk/printk_safe.c index b4045e782743..d9a659a686f3 100644 --- a/kernel/printk/printk_safe.c +++ b/kernel/printk/printk_safe.c @@ -27,7 +27,6 @@ * There are situations when we want to make sure that all buffers * were handled or when IRQs are blocked. */ -static int printk_safe_irq_ready __read_mostly; #define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \ sizeof(atomic_t) - \ @@ -51,7 +50,7 @@ static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq); /* Get flushed in a more safe context. */ static void queue_flush_work(struct printk_safe_seq_buf *s) { - if (printk_safe_irq_ready) + if (printk_percpu_data_ready()) irq_work_queue(&s->work); } @@ -402,14 +401,6 @@ void __init printk_safe_init(void) #endif } - /* - * In the highly unlikely event that a NMI were to trigger at - * this moment. Make sure IRQ work is set up before this - * variable is set. - */ - barrier(); - printk_safe_irq_ready = 1; - /* Flush pending messages that did not have scheduled IRQ works. */ printk_safe_flush(); } diff --git a/kernel/sched/core.c b/kernel/sched/core.c index a2694ba82874..3a61a3b8eaa9 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2119,12 +2119,6 @@ int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags) return cpu; } -static void update_avg(u64 *avg, u64 sample) -{ - s64 diff = sample - *avg; - *avg += diff >> 3; -} - void sched_set_stop_task(int cpu, struct task_struct *stop) { struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; @@ -4126,7 +4120,8 @@ static inline void sched_submit_work(struct task_struct *tsk) * it wants to wake up a task to maintain concurrency. * As this function is called inside the schedule() context, * we disable preemption to avoid it calling schedule() again - * in the possible wakeup of a kworker. + * in the possible wakeup of a kworker and because wq_worker_sleeping() + * requires it. */ if (tsk->flags & (PF_WQ_WORKER | PF_IO_WORKER)) { preempt_disable(); @@ -6699,7 +6694,6 @@ void __init sched_init(void) rq_attach_root(rq, &def_root_domain); #ifdef CONFIG_NO_HZ_COMMON - rq->last_load_update_tick = jiffies; rq->last_blocked_load_update_tick = jiffies; atomic_set(&rq->nohz_flags, 0); #endif diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 8331bc04aea2..a562df57a86e 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -816,10 +816,12 @@ static int __init init_sched_debug_procfs(void) __initcall(init_sched_debug_procfs); -#define __P(F) SEQ_printf(m, "%-45s:%21Ld\n", #F, (long long)F) -#define P(F) SEQ_printf(m, "%-45s:%21Ld\n", #F, (long long)p->F) -#define __PN(F) SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)F)) -#define PN(F) SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)p->F)) +#define __PS(S, F) SEQ_printf(m, "%-45s:%21Ld\n", S, (long long)(F)) +#define __P(F) __PS(#F, F) +#define P(F) __PS(#F, p->F) +#define __PSN(S, F) SEQ_printf(m, "%-45s:%14Ld.%06ld\n", S, SPLIT_NS((long long)(F))) +#define __PN(F) __PSN(#F, F) +#define PN(F) __PSN(#F, p->F) #ifdef CONFIG_NUMA_BALANCING @@ -868,18 +870,9 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, SEQ_printf(m, "---------------------------------------------------------" "----------\n"); -#define __P(F) \ - SEQ_printf(m, "%-45s:%21Ld\n", #F, (long long)F) -#define P(F) \ - SEQ_printf(m, "%-45s:%21Ld\n", #F, (long long)p->F) -#define P_SCHEDSTAT(F) \ - SEQ_printf(m, "%-45s:%21Ld\n", #F, (long long)schedstat_val(p->F)) -#define __PN(F) \ - SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)F)) -#define PN(F) \ - SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)p->F)) -#define PN_SCHEDSTAT(F) \ - SEQ_printf(m, "%-45s:%14Ld.%06ld\n", #F, SPLIT_NS((long long)schedstat_val(p->F))) + +#define P_SCHEDSTAT(F) __PS(#F, schedstat_val(p->F)) +#define PN_SCHEDSTAT(F) __PSN(#F, schedstat_val(p->F)) PN(se.exec_start); PN(se.vruntime); @@ -939,10 +932,8 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, } __P(nr_switches); - SEQ_printf(m, "%-45s:%21Ld\n", - "nr_voluntary_switches", (long long)p->nvcsw); - SEQ_printf(m, "%-45s:%21Ld\n", - "nr_involuntary_switches", (long long)p->nivcsw); + __PS("nr_voluntary_switches", p->nvcsw); + __PS("nr_involuntary_switches", p->nivcsw); P(se.load.weight); #ifdef CONFIG_SMP @@ -956,6 +947,12 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, P(se.avg.util_est.ewma); P(se.avg.util_est.enqueued); #endif +#ifdef CONFIG_UCLAMP_TASK + __PS("uclamp.min", p->uclamp[UCLAMP_MIN].value); + __PS("uclamp.max", p->uclamp[UCLAMP_MAX].value); + __PS("effective uclamp.min", uclamp_eff_value(p, UCLAMP_MIN)); + __PS("effective uclamp.max", uclamp_eff_value(p, UCLAMP_MAX)); +#endif P(policy); P(prio); if (task_has_dl_policy(p)) { @@ -963,11 +960,7 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, P(dl.deadline); } #undef PN_SCHEDSTAT -#undef PN -#undef __PN #undef P_SCHEDSTAT -#undef P -#undef __P { unsigned int this_cpu = raw_smp_processor_id(); @@ -975,8 +968,7 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, t0 = cpu_clock(this_cpu); t1 = cpu_clock(this_cpu); - SEQ_printf(m, "%-45s:%21Ld\n", - "clock-delta", (long long)(t1-t0)); + __PS("clock-delta", t1-t0); } sched_show_numa(p, m); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index d7fb20adabeb..02f323b85b6d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2799,7 +2799,7 @@ static void task_numa_work(struct callback_head *work) * Skip inaccessible VMAs to avoid any confusion between * PROT_NONE and NUMA hinting ptes */ - if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))) + if (!vma_is_accessible(vma)) continue; do { @@ -4836,11 +4836,10 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) resched_curr(rq); } -static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, u64 remaining) +static void distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) { struct cfs_rq *cfs_rq; - u64 runtime; - u64 starting_runtime = remaining; + u64 runtime, remaining = 1; rcu_read_lock(); list_for_each_entry_rcu(cfs_rq, &cfs_b->throttled_cfs_rq, @@ -4855,10 +4854,13 @@ static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, u64 remaining) /* By the above check, this should never be true */ SCHED_WARN_ON(cfs_rq->runtime_remaining > 0); + raw_spin_lock(&cfs_b->lock); runtime = -cfs_rq->runtime_remaining + 1; - if (runtime > remaining) - runtime = remaining; - remaining -= runtime; + if (runtime > cfs_b->runtime) + runtime = cfs_b->runtime; + cfs_b->runtime -= runtime; + remaining = cfs_b->runtime; + raw_spin_unlock(&cfs_b->lock); cfs_rq->runtime_remaining += runtime; @@ -4873,8 +4875,6 @@ next: break; } rcu_read_unlock(); - - return starting_runtime - remaining; } /* @@ -4885,7 +4885,6 @@ next: */ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun, unsigned long flags) { - u64 runtime; int throttled; /* no need to continue the timer with no bandwidth constraint */ @@ -4914,24 +4913,17 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun, u cfs_b->nr_throttled += overrun; /* - * This check is repeated as we are holding onto the new bandwidth while - * we unthrottle. This can potentially race with an unthrottled group - * trying to acquire new bandwidth from the global pool. This can result - * in us over-using our runtime if it is all used during this loop, but - * only by limited amounts in that extreme case. + * This check is repeated as we release cfs_b->lock while we unthrottle. */ while (throttled && cfs_b->runtime > 0 && !cfs_b->distribute_running) { - runtime = cfs_b->runtime; cfs_b->distribute_running = 1; raw_spin_unlock_irqrestore(&cfs_b->lock, flags); /* we can't nest cfs_b->lock while distributing bandwidth */ - runtime = distribute_cfs_runtime(cfs_b, runtime); + distribute_cfs_runtime(cfs_b); raw_spin_lock_irqsave(&cfs_b->lock, flags); cfs_b->distribute_running = 0; throttled = !list_empty(&cfs_b->throttled_cfs_rq); - - lsub_positive(&cfs_b->runtime, runtime); } /* @@ -5065,10 +5057,9 @@ static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) if (!runtime) return; - runtime = distribute_cfs_runtime(cfs_b, runtime); + distribute_cfs_runtime(cfs_b); raw_spin_lock_irqsave(&cfs_b->lock, flags); - lsub_positive(&cfs_b->runtime, runtime); cfs_b->distribute_running = 0; raw_spin_unlock_irqrestore(&cfs_b->lock, flags); } @@ -6080,8 +6071,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask); struct sched_domain *this_sd; u64 avg_cost, avg_idle; - u64 time, cost; - s64 delta; + u64 time; int this = smp_processor_id(); int cpu, nr = INT_MAX; @@ -6119,9 +6109,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t } time = cpu_clock(this) - time; - cost = this_sd->avg_scan_cost; - delta = (s64)(time - cost) / 8; - this_sd->avg_scan_cost += delta; + update_avg(&this_sd->avg_scan_cost, time); return cpu; } @@ -9048,6 +9036,14 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s sds->avg_load = (sds->total_load * SCHED_CAPACITY_SCALE) / sds->total_capacity; + /* + * If the local group is more loaded than the selected + * busiest group don't try to pull any tasks. + */ + if (local->avg_load >= busiest->avg_load) { + env->imbalance = 0; + return; + } } /* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 0f616bf7bce3..db3a57675ccf 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -195,6 +195,12 @@ static inline int task_has_dl_policy(struct task_struct *p) #define cap_scale(v, s) ((v)*(s) >> SCHED_CAPACITY_SHIFT) +static inline void update_avg(u64 *avg, u64 sample) +{ + s64 diff = sample - *avg; + *avg += diff / 8; +} + /* * !! For sched_setattr_nocheck() (kernel) only !! * @@ -882,7 +888,6 @@ struct rq { #endif #ifdef CONFIG_NO_HZ_COMMON #ifdef CONFIG_SMP - unsigned long last_load_update_tick; unsigned long last_blocked_load_update_tick; unsigned int has_blocked_load; #endif /* CONFIG_SMP */ diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 3816a18c251e..891ccad5f271 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -858,7 +858,8 @@ void wq_worker_running(struct task_struct *task) * @task: task going to sleep * * This function is called from schedule() when a busy worker is - * going to sleep. + * going to sleep. Preemption needs to be disabled to protect ->sleeping + * assignment. */ void wq_worker_sleeping(struct task_struct *task) { @@ -875,7 +876,8 @@ void wq_worker_sleeping(struct task_struct *task) pool = worker->pool; - if (WARN_ON_ONCE(worker->sleeping)) + /* Return if preempted before wq_worker_running() was reached */ + if (worker->sleeping) return; worker->sleeping = 1; |