diff options
Diffstat (limited to '')
-rw-r--r-- | fs/eventfd.c | 85 |
1 files changed, 51 insertions, 34 deletions
diff --git a/fs/eventfd.c b/fs/eventfd.c index 78e41c7c3d05..c0ffee99ad23 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -23,8 +23,7 @@ #include <linux/proc_fs.h> #include <linux/seq_file.h> #include <linux/idr.h> - -DEFINE_PER_CPU(int, eventfd_wake_count); +#include <linux/uio.h> static DEFINE_IDA(eventfd_ida); @@ -66,21 +65,21 @@ __u64 eventfd_signal(struct eventfd_ctx *ctx, __u64 n) * Deadlock or stack overflow issues can happen if we recurse here * through waitqueue wakeup handlers. If the caller users potentially * nested waitqueues with custom wakeup handlers, then it should - * check eventfd_signal_count() before calling this function. If - * it returns true, the eventfd_signal() call should be deferred to a + * check eventfd_signal_allowed() before calling this function. If + * it returns false, the eventfd_signal() call should be deferred to a * safe context. */ - if (WARN_ON_ONCE(this_cpu_read(eventfd_wake_count))) + if (WARN_ON_ONCE(current->in_eventfd)) return 0; spin_lock_irqsave(&ctx->wqh.lock, flags); - this_cpu_inc(eventfd_wake_count); + current->in_eventfd = 1; if (ULLONG_MAX - ctx->count < n) n = ULLONG_MAX - ctx->count; ctx->count += n; if (waitqueue_active(&ctx->wqh)) wake_up_locked_poll(&ctx->wqh, EPOLLIN); - this_cpu_dec(eventfd_wake_count); + current->in_eventfd = 0; spin_unlock_irqrestore(&ctx->wqh.lock, flags); return n; @@ -181,11 +180,14 @@ static __poll_t eventfd_poll(struct file *file, poll_table *wait) return events; } -static void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt) +void eventfd_ctx_do_read(struct eventfd_ctx *ctx, __u64 *cnt) { + lockdep_assert_held(&ctx->wqh.lock); + *cnt = (ctx->flags & EFD_SEMAPHORE) ? 1 : ctx->count; ctx->count -= *cnt; } +EXPORT_SYMBOL_GPL(eventfd_ctx_do_read); /** * eventfd_ctx_remove_wait_queue - Read the current counter and removes wait queue. @@ -216,32 +218,32 @@ int eventfd_ctx_remove_wait_queue(struct eventfd_ctx *ctx, wait_queue_entry_t *w } EXPORT_SYMBOL_GPL(eventfd_ctx_remove_wait_queue); -static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count, - loff_t *ppos) +static ssize_t eventfd_read(struct kiocb *iocb, struct iov_iter *to) { + struct file *file = iocb->ki_filp; struct eventfd_ctx *ctx = file->private_data; - ssize_t res; __u64 ucnt = 0; DECLARE_WAITQUEUE(wait, current); - if (count < sizeof(ucnt)) + if (iov_iter_count(to) < sizeof(ucnt)) return -EINVAL; - spin_lock_irq(&ctx->wqh.lock); - res = -EAGAIN; - if (ctx->count > 0) - res = sizeof(ucnt); - else if (!(file->f_flags & O_NONBLOCK)) { + if (!ctx->count) { + if ((file->f_flags & O_NONBLOCK) || + (iocb->ki_flags & IOCB_NOWAIT)) { + spin_unlock_irq(&ctx->wqh.lock); + return -EAGAIN; + } __add_wait_queue(&ctx->wqh, &wait); for (;;) { set_current_state(TASK_INTERRUPTIBLE); - if (ctx->count > 0) { - res = sizeof(ucnt); + if (ctx->count) break; - } if (signal_pending(current)) { - res = -ERESTARTSYS; - break; + __remove_wait_queue(&ctx->wqh, &wait); + __set_current_state(TASK_RUNNING); + spin_unlock_irq(&ctx->wqh.lock); + return -ERESTARTSYS; } spin_unlock_irq(&ctx->wqh.lock); schedule(); @@ -250,17 +252,16 @@ static ssize_t eventfd_read(struct file *file, char __user *buf, size_t count, __remove_wait_queue(&ctx->wqh, &wait); __set_current_state(TASK_RUNNING); } - if (likely(res > 0)) { - eventfd_ctx_do_read(ctx, &ucnt); - if (waitqueue_active(&ctx->wqh)) - wake_up_locked_poll(&ctx->wqh, EPOLLOUT); - } + eventfd_ctx_do_read(ctx, &ucnt); + current->in_eventfd = 1; + if (waitqueue_active(&ctx->wqh)) + wake_up_locked_poll(&ctx->wqh, EPOLLOUT); + current->in_eventfd = 0; spin_unlock_irq(&ctx->wqh.lock); - - if (res > 0 && put_user(ucnt, (__u64 __user *)buf)) + if (unlikely(copy_to_iter(&ucnt, sizeof(ucnt), to) != sizeof(ucnt))) return -EFAULT; - return res; + return sizeof(ucnt); } static ssize_t eventfd_write(struct file *file, const char __user *buf, size_t count, @@ -302,8 +303,10 @@ static ssize_t eventfd_write(struct file *file, const char __user *buf, size_t c } if (likely(res > 0)) { ctx->count += ucnt; + current->in_eventfd = 1; if (waitqueue_active(&ctx->wqh)) wake_up_locked_poll(&ctx->wqh, EPOLLIN); + current->in_eventfd = 0; } spin_unlock_irq(&ctx->wqh.lock); @@ -329,7 +332,7 @@ static const struct file_operations eventfd_fops = { #endif .release = eventfd_release, .poll = eventfd_poll, - .read = eventfd_read, + .read_iter = eventfd_read, .write = eventfd_write, .llseek = noop_llseek, }; @@ -406,6 +409,7 @@ EXPORT_SYMBOL_GPL(eventfd_ctx_fileget); static int do_eventfd(unsigned int count, int flags) { struct eventfd_ctx *ctx; + struct file *file; int fd; /* Check the EFD_* constants for consistency. */ @@ -425,11 +429,24 @@ static int do_eventfd(unsigned int count, int flags) ctx->flags = flags; ctx->id = ida_simple_get(&eventfd_ida, 0, 0, GFP_KERNEL); - fd = anon_inode_getfd("[eventfd]", &eventfd_fops, ctx, - O_RDWR | (flags & EFD_SHARED_FCNTL_FLAGS)); + flags &= EFD_SHARED_FCNTL_FLAGS; + flags |= O_RDWR; + fd = get_unused_fd_flags(flags); if (fd < 0) - eventfd_free_ctx(ctx); + goto err; + + file = anon_inode_getfile("[eventfd]", &eventfd_fops, ctx, flags); + if (IS_ERR(file)) { + put_unused_fd(fd); + fd = PTR_ERR(file); + goto err; + } + file->f_mode |= FMODE_NOWAIT; + fd_install(fd, file); + return fd; +err: + eventfd_free_ctx(ctx); return fd; } |