X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fexit.c;h=08eee6d20cf89644a452942a9d257b3d983ec26b;hb=34a75f0025b9cf803b6a88db032e6ad6950c9313;hp=45384fd56958bcba01dc9faaca27fc8865f32c32;hpb=c7b5ebbddf7bcd3651947760f423e3783bbe6573;p=linux-2.6.git diff --git a/kernel/exit.c b/kernel/exit.c index 45384fd56..08eee6d20 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -10,10 +10,12 @@ #include #include #include +#include #include #include #include #include +#include #include #include #include @@ -24,7 +26,20 @@ #include #include #include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include /* for audit_free() */ #include +#include +#include +#include #include #include @@ -36,24 +51,91 @@ extern struct task_struct *child_reaper; int getrusage(struct task_struct *, int, struct rusage __user *); +static void exit_mm(struct task_struct * tsk); + static void __unhash_process(struct task_struct *p) { nr_threads--; - /* tasklist_lock is held, is this sufficient? */ - if (p->vx_info) { - atomic_dec(&p->vx_info->cvirt.nr_threads); - vx_nproc_dec(p); - } detach_pid(p, PIDTYPE_PID); - detach_pid(p, PIDTYPE_TGID); if (thread_group_leader(p)) { detach_pid(p, PIDTYPE_PGID); detach_pid(p, PIDTYPE_SID); - if (p->pid) - __get_cpu_var(process_counts)--; + + list_del_rcu(&p->tasks); + __get_cpu_var(process_counts)--; + } + list_del_rcu(&p->thread_group); + remove_parent(p); +} + +/* + * This function expects the tasklist_lock write-locked. + */ +static void __exit_signal(struct task_struct *tsk) +{ + struct signal_struct *sig = tsk->signal; + struct sighand_struct *sighand; + + BUG_ON(!sig); + BUG_ON(!atomic_read(&sig->count)); + + rcu_read_lock(); + sighand = rcu_dereference(tsk->sighand); + spin_lock(&sighand->siglock); + + posix_cpu_timers_exit(tsk); + if (atomic_dec_and_test(&sig->count)) + posix_cpu_timers_exit_group(tsk); + else { + /* + * If there is any task waiting for the group exit + * then notify it: + */ + if (sig->group_exit_task && atomic_read(&sig->count) == sig->notify_count) { + wake_up_process(sig->group_exit_task); + sig->group_exit_task = NULL; + } + if (tsk == sig->curr_target) + sig->curr_target = next_thread(tsk); + /* + * Accumulate here the counters for all threads but the + * group leader as they die, so they can be added into + * the process-wide totals when those are taken. + * The group leader stays around as a zombie as long + * as there are other threads. When it gets reaped, + * the exit.c code will add its counts into these totals. + * We won't ever get here for the group leader, since it + * will have been the last reference on the signal_struct. + */ + sig->utime = cputime_add(sig->utime, tsk->utime); + sig->stime = cputime_add(sig->stime, tsk->stime); + sig->min_flt += tsk->min_flt; + sig->maj_flt += tsk->maj_flt; + sig->nvcsw += tsk->nvcsw; + sig->nivcsw += tsk->nivcsw; + sig->sched_time += tsk->sched_time; + sig = NULL; /* Marker for below. */ + } + + __unhash_process(tsk); + + tsk->signal = NULL; + tsk->sighand = NULL; + spin_unlock(&sighand->siglock); + rcu_read_unlock(); + + __cleanup_sighand(sighand); + clear_tsk_thread_flag(tsk,TIF_SIGPENDING); + flush_sigqueue(&tsk->pending); + if (sig) { + flush_sigqueue(&sig->shared_pending); + __cleanup_signal(sig); } +} - REMOVE_LINKS(p); +static void delayed_put_task_struct(struct rcu_head *rhp) +{ + put_task_struct(container_of(rhp, struct task_struct, rcu)); } void release_task(struct task_struct * p) @@ -62,17 +144,14 @@ void release_task(struct task_struct * p) task_t *leader; struct dentry *proc_dentry; -repeat: +repeat: atomic_dec(&p->user->processes); spin_lock(&p->proc_lock); proc_dentry = proc_pid_unhash(p); write_lock_irq(&tasklist_lock); - if (unlikely(p->ptrace)) - __ptrace_unlink(p); + ptrace_unlink(p); BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children)); __exit_signal(p); - __exit_sighand(p); - __unhash_process(p); /* * If we are the last non-leader member of the thread @@ -81,7 +160,7 @@ repeat: */ zap_leader = 0; leader = p->group_leader; - if (leader != p && thread_group_empty(leader) && leader->state == TASK_ZOMBIE) { + if (leader != p && thread_group_empty(leader) && leader->exit_state == EXIT_ZOMBIE) { BUG_ON(leader->exit_signal == -1); do_notify_parent(leader, leader->exit_signal); /* @@ -100,28 +179,13 @@ repeat: spin_unlock(&p->proc_lock); proc_pid_flush(proc_dentry); release_thread(p); - put_task_struct(p); + call_rcu(&p->rcu, delayed_put_task_struct); p = leader; if (unlikely(zap_leader)) goto repeat; } -/* we are using it only for SMP init */ - -void unhash_process(struct task_struct *p) -{ - struct dentry *proc_dentry; - - spin_lock(&p->proc_lock); - proc_dentry = proc_pid_unhash(p); - write_lock_irq(&tasklist_lock); - __unhash_process(p); - write_unlock_irq(&tasklist_lock); - spin_unlock(&p->proc_lock); - proc_pid_flush(proc_dentry); -} - /* * This checks not only the pgrp, but falls back on the pid if no * satisfactory pgrp is found. I dunno - gdb doesn't work correctly @@ -163,7 +227,7 @@ static int will_become_orphaned_pgrp(int pgrp, task_t *ignored_task) do_each_task_pid(pgrp, PIDTYPE_PGID, p) { if (p == ignored_task - || p->state >= TASK_ZOMBIE + || p->exit_state || p->real_parent->pid == 1) continue; if (process_group(p->real_parent) != pgrp @@ -186,7 +250,7 @@ int is_orphaned_pgrp(int pgrp) return retval; } -static inline int has_stopped_jobs(int pgrp) +static int has_stopped_jobs(int pgrp) { int retval = 0; struct task_struct *p; @@ -212,7 +276,7 @@ static inline int has_stopped_jobs(int pgrp) } /** - * reparent_to_init() - Reparent the calling kernel thread to the init task. + * reparent_to_init - Reparent the calling kernel thread to the init task. * * If a kernel thread is launched as a result of a system call, or if * it ever exits, it should generally reparent itself to init so that @@ -223,37 +287,38 @@ static inline int has_stopped_jobs(int pgrp) * * NOTE that reparent_to_init() gives the caller full capabilities. */ -void reparent_to_init(void) +static void reparent_to_init(void) { write_lock_irq(&tasklist_lock); ptrace_unlink(current); /* Reparent to init */ - REMOVE_LINKS(current); - /* FIXME handle vchild_reaper/initpid */ + remove_parent(current); current->parent = child_reaper; current->real_parent = child_reaper; - SET_LINKS(current); + add_parent(current); /* Set the exit signal to SIGCHLD so we signal init on exit */ current->exit_signal = SIGCHLD; - if ((current->policy == SCHED_NORMAL) && (task_nice(current) < 0)) + if ((current->policy == SCHED_NORMAL || + current->policy == SCHED_BATCH) + && (task_nice(current) < 0)) set_user_nice(current, 0); /* cpus_allowed? */ /* rt_priority? */ /* signals? */ security_task_reparent_to_init(current); - memcpy(current->rlim, init_task.rlim, sizeof(*(current->rlim))); + memcpy(current->signal->rlim, init_task.signal->rlim, + sizeof(current->signal->rlim)); atomic_inc(&(INIT_USER->__count)); - switch_uid(INIT_USER); - write_unlock_irq(&tasklist_lock); + switch_uid(INIT_USER); } void __set_special_pids(pid_t session, pid_t pgrp) { - struct task_struct *curr = current; + struct task_struct *curr = current->group_leader; if (curr->signal->session != session) { detach_pid(curr, PIDTYPE_SID); @@ -281,7 +346,7 @@ void set_special_pids(pid_t session, pid_t pgrp) */ int allow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); @@ -302,7 +367,7 @@ EXPORT_SYMBOL(allow_signal); int disallow_signal(int sig) { - if (sig < 1 || sig > _NSIG) + if (!valid_signal(sig) || sig < 1) return -EINVAL; spin_lock_irq(¤t->sighand->siglock); @@ -337,7 +402,9 @@ void daemonize(const char *name, ...) exit_mm(current); set_special_pids(1, 1); + mutex_lock(&tty_mutex); current->signal->tty = NULL; + mutex_unlock(&tty_mutex); /* Block and flush all signals */ sigfillset(&blocked); @@ -350,6 +417,9 @@ void daemonize(const char *name, ...) fs = init_task.fs; current->fs = fs; atomic_inc(&fs->count); + exit_namespace(current); + current->namespace = init_task.namespace; + get_namespace(current->namespace); exit_files(current); current->files = init_task.files; atomic_inc(¤t->files->count); @@ -359,26 +429,35 @@ void daemonize(const char *name, ...) EXPORT_SYMBOL(daemonize); -static inline void close_files(struct files_struct * files) +static void close_files(struct files_struct * files) { int i, j; + struct fdtable *fdt; j = 0; + + /* + * It is safe to dereference the fd table without RCU or + * ->file_lock because this is the last reference to the + * files structure. + */ + fdt = files_fdtable(files); for (;;) { unsigned long set; i = j * __NFDBITS; - if (i >= files->max_fdset || i >= files->max_fds) + if (i >= fdt->max_fdset || i >= fdt->max_fds) break; - set = files->open_fds->fds_bits[j++]; + set = fdt->open_fds->fds_bits[j++]; while (set) { if (set & 1) { - struct file * file = xchg(&files->fd[i], NULL); + struct file * file = xchg(&fdt->fd[i], NULL); if (file) filp_close(file, files); - // vx_openfd_dec(i); + vx_openfd_dec(i); } i++; set >>= 1; + cond_resched(); } } } @@ -398,18 +477,22 @@ struct files_struct *get_files_struct(struct task_struct *task) void fastcall put_files_struct(struct files_struct *files) { + struct fdtable *fdt; + if (atomic_dec_and_test(&files->count)) { close_files(files); /* * Free the fd and fdset arrays if we expanded them. + * If the fdtable was embedded, pass files for freeing + * at the end of the RCU grace period. Otherwise, + * you can free files immediately. */ - if (files->fd != &files->fd_array[0]) - free_fd_array(files->fd, files->max_fds); - if (files->max_fdset > __FD_SETSIZE) { - free_fdset(files->open_fds, files->max_fdset); - free_fdset(files->close_on_exec, files->max_fdset); - } - kmem_cache_free(files_cachep, files); + fdt = files_fdtable(files); + if (fdt == &files->fdtab) + fdt->free_files = files; + else + kmem_cache_free(files_cachep, files); + free_fdtable(fdt); } } @@ -476,7 +559,7 @@ EXPORT_SYMBOL_GPL(exit_fs); * Turn us into a lazy TLB process if we * aren't already.. */ -static inline void __exit_mm(struct task_struct * tsk) +static void exit_mm(struct task_struct * tsk) { struct mm_struct *mm = tsk->mm; @@ -512,35 +595,30 @@ static inline void __exit_mm(struct task_struct * tsk) mmput(mm); } -void exit_mm(struct task_struct *tsk) +static inline void choose_new_parent(task_t *p, task_t *reaper) { - __exit_mm(tsk); -} - -EXPORT_SYMBOL(exit_mm); + /* check for reaper context */ + vxwprintk((p->xid != reaper->xid) && (reaper != child_reaper), + "rogue reaper: %p[%d,#%u] <> %p[%d,#%u]", + p, p->pid, p->xid, reaper, reaper->pid, reaper->xid); -static inline void choose_new_parent(task_t *p, task_t *reaper, task_t *child_reaper) -{ /* * Make sure we're not reparenting to ourselves and that * the parent is not a zombie. */ - BUG_ON(p == reaper || reaper->state >= TASK_ZOMBIE); + BUG_ON(p == reaper || reaper->exit_state); p->real_parent = reaper; - if (p->parent == p->real_parent) - BUG(); } -static inline void reparent_thread(task_t *p, task_t *father, int traced) +static void reparent_thread(task_t *p, task_t *father, int traced) { /* We don't want people slaying init. */ if (p->exit_signal != -1) p->exit_signal = SIGCHLD; - p->self_exec_id++; if (p->pdeath_signal) /* We already hold the tasklist_lock here. */ - group_send_sig_info(p->pdeath_signal, (void *) 0, p); + group_send_sig_info(p->pdeath_signal, SEND_SIG_NOINFO, p); /* Move the child from its dying parent to the new one. */ if (unlikely(traced)) { @@ -553,14 +631,14 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced) * anyway, so let go of it. */ p->ptrace = 0; - list_del_init(&p->sibling); + remove_parent(p); p->parent = p->real_parent; - list_add_tail(&p->sibling, &p->parent->children); + add_parent(p); /* If we'd notified the old parent about this child's death, * also notify the new parent. */ - if (p->state == TASK_ZOMBIE && p->exit_signal != -1 && + if (p->exit_state == EXIT_ZOMBIE && p->exit_signal != -1 && thread_group_empty(p)) do_notify_parent(p, p->exit_signal); else if (p->state == TASK_TRACED) { @@ -569,7 +647,7 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced) * a normal stop since it's no longer being * traced. */ - p->state = TASK_STOPPED; + ptrace_untrace(p); } } @@ -584,8 +662,8 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced) int pgrp = process_group(p); if (will_become_orphaned_pgrp(pgrp, NULL) && has_stopped_jobs(pgrp)) { - __kill_pg_info(SIGHUP, (void *)1, pgrp); - __kill_pg_info(SIGCONT, (void *)1, pgrp); + __kill_pg_info(SIGHUP, SEND_SIG_PRIV, pgrp); + __kill_pg_info(SIGCONT, SEND_SIG_PRIV, pgrp); } } } @@ -596,20 +674,19 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced) * group, and if no such member exists, give it to * the global child reaper process (ie "init") */ -static inline void forget_original_parent(struct task_struct * father, +static void forget_original_parent(struct task_struct * father, struct list_head *to_release) { struct task_struct *p, *reaper = father; struct list_head *_p, *_n; - /* FIXME handle vchild_reaper/initpid */ do { reaper = next_thread(reaper); if (reaper == father) { - reaper = child_reaper; + reaper = vx_child_reaper(father); break; } - } while (reaper->state >= TASK_ZOMBIE); + } while (reaper->exit_state); /* * There are only two places where our children can be: @@ -630,12 +707,12 @@ static inline void forget_original_parent(struct task_struct * father, if (father == p->real_parent) { /* reparent with a reaper, real father it's us */ - choose_new_parent(p, reaper, child_reaper); + choose_new_parent(p, vx_child_reaper(p)); reparent_thread(p, father, 0); } else { /* reparent ptraced task to its real parent */ __ptrace_unlink (p); - if (p->state == TASK_ZOMBIE && p->exit_signal != -1 && + if (p->exit_state == EXIT_ZOMBIE && p->exit_signal != -1 && thread_group_empty(p)) do_notify_parent(p, p->exit_signal); } @@ -646,12 +723,12 @@ static inline void forget_original_parent(struct task_struct * father, * zombie forever since we prevented it from self-reap itself * while it was being traced by us, to be able to see it in wait4. */ - if (unlikely(ptrace && p->state == TASK_ZOMBIE && p->exit_signal == -1)) + if (unlikely(ptrace && p->exit_state == EXIT_ZOMBIE && p->exit_signal == -1)) list_add(&p->ptrace_list, to_release); } list_for_each_safe(_p, _n, &father->ptrace_children) { p = list_entry(_p,struct task_struct,ptrace_list); - choose_new_parent(p, reaper, child_reaper); + choose_new_parent(p, reaper); reparent_thread(p, father, 1); } } @@ -666,7 +743,7 @@ static void exit_notify(struct task_struct *tsk) struct task_struct *t; struct list_head ptrace_dead, *_p, *_n; - if (signal_pending(tsk) && !tsk->signal->group_exit + if (signal_pending(tsk) && !(tsk->signal->flags & SIGNAL_GROUP_EXIT) && !thread_group_empty(tsk)) { /* * This occurs when there was a race between our exit @@ -721,8 +798,8 @@ static void exit_notify(struct task_struct *tsk) (t->signal->session == tsk->signal->session) && will_become_orphaned_pgrp(process_group(tsk), tsk) && has_stopped_jobs(process_group(tsk))) { - __kill_pg_info(SIGHUP, (void *)1, process_group(tsk)); - __kill_pg_info(SIGCONT, (void *)1, process_group(tsk)); + __kill_pg_info(SIGHUP, SEND_SIG_PRIV, process_group(tsk)); + __kill_pg_info(SIGCONT, SEND_SIG_PRIV, process_group(tsk)); } /* Let father know we died @@ -759,18 +836,12 @@ static void exit_notify(struct task_struct *tsk) do_notify_parent(tsk, SIGCHLD); } - state = TASK_ZOMBIE; - if (tsk->exit_signal == -1 && tsk->ptrace == 0) - state = TASK_DEAD; - tsk->state = state; - - /* - * Clear these here so that update_process_times() won't try to deliver - * itimer, profile or rlimit signals to this task while it is in late exit. - */ - tsk->it_virt_value = 0; - tsk->it_prof_value = 0; - tsk->rlim[RLIMIT_CPU].rlim_cur = RLIM_INFINITY; + state = EXIT_ZOMBIE; + if (tsk->exit_signal == -1 && + (likely(tsk->ptrace == 0) || + unlikely(tsk->parent->signal->flags & SIGNAL_GROUP_EXIT))) + state = EXIT_DEAD; + tsk->exit_state = state; write_unlock_irq(&tasklist_lock); @@ -781,69 +852,127 @@ static void exit_notify(struct task_struct *tsk) } /* If the process is dead, release it - nobody will wait for it */ - if (state == TASK_DEAD) + if (state == EXIT_DEAD) release_task(tsk); - - /* PF_DEAD causes final put_task_struct after we schedule. */ - preempt_disable(); - tsk->flags |= PF_DEAD; } -asmlinkage NORET_TYPE void do_exit(long code) +fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; + int group_dead; profile_task_exit(tsk); + WARN_ON(atomic_read(&tsk->fs_excl)); + if (unlikely(in_interrupt())) panic("Aiee, killing interrupt handler!"); if (unlikely(!tsk->pid)) panic("Attempted to kill the idle task!"); - if (unlikely(tsk->pid == 1)) + if (unlikely(tsk == child_reaper)) panic("Attempted to kill init!"); - if (tsk->io_context) - exit_io_context(); + + if (unlikely(current->ptrace & PT_TRACE_EXIT)) { + current->ptrace_message = code; + ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP); + } + + /* + * We're taking recursive faults here in do_exit. Safest is to just + * leave this task alone and wait for reboot. + */ + if (unlikely(tsk->flags & PF_EXITING)) { + printk(KERN_ALERT + "Fixing recursive fault but reboot is needed!\n"); + if (tsk->io_context) + exit_io_context(); + set_current_state(TASK_UNINTERRUPTIBLE); + schedule(); + } + tsk->flags |= PF_EXITING; - del_timer_sync(&tsk->real_timer); if (unlikely(in_atomic())) printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n", current->comm, current->pid, preempt_count()); - if (unlikely(current->ptrace & PT_TRACE_EXIT)) { - current->ptrace_message = code; - ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP); + acct_update_integrals(tsk); + if (tsk->mm) { + update_hiwater_rss(tsk->mm); + update_hiwater_vm(tsk->mm); } - - acct_process(code); - __exit_mm(tsk); + group_dead = atomic_dec_and_test(&tsk->signal->live); + if (group_dead) { + hrtimer_cancel(&tsk->signal->real_timer); + exit_itimers(tsk->signal); + acct_process(code); + if (current->tux_info) { +#ifdef CONFIG_TUX_DEBUG + printk("Possibly unexpected TUX-thread exit(%ld) at %p?\n", + code, __builtin_return_address(0)); +#endif + current->tux_exit(); + } + } + if (unlikely(tsk->robust_list)) + exit_robust_list(tsk); +#if defined(CONFIG_FUTEX) && defined(CONFIG_COMPAT) + if (unlikely(tsk->compat_robust_list)) + compat_exit_robust_list(tsk); +#endif + if (unlikely(tsk->audit_context)) + audit_free(tsk); + exit_mm(tsk); exit_sem(tsk); __exit_files(tsk); __exit_fs(tsk); exit_namespace(tsk); + exit_vx_info(tsk, code); + exit_nx_info(tsk); exit_thread(); + cpuset_exit(tsk); + exit_keys(tsk); - if (tsk->signal->leader) + if (group_dead && tsk->signal->leader) disassociate_ctty(1); - module_put(tsk->thread_info->exec_domain->module); + module_put(task_thread_info(tsk)->exec_domain->module); if (tsk->binfmt) module_put(tsk->binfmt->module); tsk->exit_code = code; + proc_exit_connector(tsk); exit_notify(tsk); #ifdef CONFIG_NUMA mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; #endif + /* + * If DEBUG_MUTEXES is on, make sure we are holding no locks: + */ + mutex_debug_check_no_locks_held(tsk); + + if (tsk->io_context) + exit_io_context(); + + if (tsk->splice_pipe) + __free_pipe_info(tsk->splice_pipe); + + /* PF_DEAD causes final put_task_struct after we schedule. */ + preempt_disable(); + BUG_ON(tsk->flags & PF_DEAD); + tsk->flags |= PF_DEAD; + schedule(); BUG(); /* Avoid "noreturn function does return". */ for (;;) ; } +EXPORT_SYMBOL_GPL(do_exit); + NORET_TYPE void complete_and_exit(struct completion *comp, long code) { if (comp) @@ -859,20 +988,6 @@ asmlinkage long sys_exit(int error_code) do_exit((error_code&0xff)<<8); } -task_t fastcall *next_thread(const task_t *p) -{ -#ifdef CONFIG_SMP - if (!p->sighand) - BUG(); - if (!spin_is_locked(&p->sighand->siglock) && - !rwlock_is_locked(&tasklist_lock)) - BUG(); -#endif - return pid_task(p->pids[PIDTYPE_TGID].pid_list.next, PIDTYPE_TGID); -} - -EXPORT_SYMBOL(next_thread); - /* * Take down every thread in the group. This is called by fatal signals * as well as by sys_exit_group (below). @@ -882,23 +997,20 @@ do_group_exit(int exit_code) { BUG_ON(exit_code & 0x80); /* core dumps don't get here */ - if (current->signal->group_exit) + if (current->signal->flags & SIGNAL_GROUP_EXIT) exit_code = current->signal->group_exit_code; else if (!thread_group_empty(current)) { struct signal_struct *const sig = current->signal; struct sighand_struct *const sighand = current->sighand; - read_lock(&tasklist_lock); spin_lock_irq(&sighand->siglock); - if (sig->group_exit) + if (sig->flags & SIGNAL_GROUP_EXIT) /* Another thread got here before we took the lock. */ exit_code = sig->group_exit_code; else { - sig->group_exit = 1; sig->group_exit_code = exit_code; zap_other_threads(current); } spin_unlock_irq(&sighand->siglock); - read_unlock(&tasklist_lock); } do_exit(exit_code); @@ -981,7 +1093,7 @@ static int wait_noreap_copyout(task_t *p, pid_t pid, uid_t uid, } /* - * Handle sys_wait4 work for one task in state TASK_ZOMBIE. We hold + * Handle sys_wait4 work for one task in state EXIT_ZOMBIE. We hold * read_lock(&tasklist_lock) on entry. If we return zero, we still hold * the lock and this task is uninteresting. If we return nonzero, we have * released the lock and the system call should return. @@ -1000,7 +1112,7 @@ static int wait_task_zombie(task_t *p, int noreap, int exit_code = p->exit_code; int why, status; - if (unlikely(p->state != TASK_ZOMBIE)) + if (unlikely(p->exit_state != EXIT_ZOMBIE)) return 0; if (unlikely(p->exit_signal == -1 && p->ptrace == 0)) return 0; @@ -1021,9 +1133,9 @@ static int wait_task_zombie(task_t *p, int noreap, * Try to move the task's state to DEAD * only one thread is allowed to do this: */ - state = xchg(&p->state, TASK_DEAD); - if (state != TASK_ZOMBIE) { - BUG_ON(state != TASK_DEAD); + state = xchg(&p->exit_state, EXIT_DEAD); + if (state != EXIT_ZOMBIE) { + BUG_ON(state != EXIT_DEAD); return 0; } if (unlikely(p->exit_signal == -1 && p->ptrace == 0)) { @@ -1035,6 +1147,9 @@ static int wait_task_zombie(task_t *p, int noreap, } if (likely(p->real_parent == p->parent) && likely(p->signal)) { + struct signal_struct *psig; + struct signal_struct *sig; + /* * The resource counters for the group leader are in its * own task_struct. Those for dead threads in the group @@ -1051,29 +1166,37 @@ static int wait_task_zombie(task_t *p, int noreap, * here reaping other children at the same time. */ spin_lock_irq(&p->parent->sighand->siglock); - p->parent->signal->cutime += - p->utime + p->signal->utime + p->signal->cutime; - p->parent->signal->cstime += - p->stime + p->signal->stime + p->signal->cstime; - p->parent->signal->cmin_flt += - p->min_flt + p->signal->min_flt + p->signal->cmin_flt; - p->parent->signal->cmaj_flt += - p->maj_flt + p->signal->maj_flt + p->signal->cmaj_flt; - p->parent->signal->cnvcsw += - p->nvcsw + p->signal->nvcsw + p->signal->cnvcsw; - p->parent->signal->cnivcsw += - p->nivcsw + p->signal->nivcsw + p->signal->cnivcsw; + psig = p->parent->signal; + sig = p->signal; + psig->cutime = + cputime_add(psig->cutime, + cputime_add(p->utime, + cputime_add(sig->utime, + sig->cutime))); + psig->cstime = + cputime_add(psig->cstime, + cputime_add(p->stime, + cputime_add(sig->stime, + sig->cstime))); + psig->cmin_flt += + p->min_flt + sig->min_flt + sig->cmin_flt; + psig->cmaj_flt += + p->maj_flt + sig->maj_flt + sig->cmaj_flt; + psig->cnvcsw += + p->nvcsw + sig->nvcsw + sig->cnvcsw; + psig->cnivcsw += + p->nivcsw + sig->nivcsw + sig->cnivcsw; spin_unlock_irq(&p->parent->sighand->siglock); } /* * Now we are sure this task is interesting, and no other - * thread can reap it because we set its state to TASK_DEAD. + * thread can reap it because we set its state to EXIT_DEAD. */ read_unlock(&tasklist_lock); retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0; - status = p->signal->group_exit + status = (p->signal->flags & SIGNAL_GROUP_EXIT) ? p->signal->group_exit_code : p->exit_code; if (!retval && stat_addr) retval = put_user(status, stat_addr); @@ -1100,7 +1223,8 @@ static int wait_task_zombie(task_t *p, int noreap, if (!retval && infop) retval = put_user(p->uid, &infop->si_uid); if (retval) { - p->state = TASK_ZOMBIE; + // TODO: is this safe? + p->exit_state = EXIT_ZOMBIE; return retval; } retval = p->pid; @@ -1109,7 +1233,8 @@ static int wait_task_zombie(task_t *p, int noreap, /* Double-check with lock held. */ if (p->real_parent != p->parent) { __ptrace_unlink(p); - p->state = TASK_ZOMBIE; + // TODO: is this safe? + p->exit_state = EXIT_ZOMBIE; /* * If this is not a detached task, notify the parent. * If it's still not detached after that, don't release @@ -1168,7 +1293,7 @@ static int wait_task_stopped(task_t *p, int delayed_group_leader, int noreap, exit_code = p->exit_code; if (unlikely(!exit_code) || - unlikely(p->state > TASK_STOPPED)) + unlikely(p->state & TASK_TRACED)) goto bail_ref; return wait_noreap_copyout(p, pid, uid, why, (exit_code << 8) | 0x7f, @@ -1180,13 +1305,13 @@ static int wait_task_stopped(task_t *p, int delayed_group_leader, int noreap, /* * This uses xchg to be atomic with the thread resuming and setting * it. It must also be done with the write lock held to prevent a - * race with the TASK_ZOMBIE case. + * race with the EXIT_ZOMBIE case. */ exit_code = xchg(&p->exit_code, 0); - if (unlikely(p->state >= TASK_ZOMBIE)) { + if (unlikely(p->exit_state)) { /* * The task resumed and then died. Let the next iteration - * catch it in TASK_ZOMBIE. Note that exit_code might + * catch it in EXIT_ZOMBIE. Note that exit_code might * already be zero here if it resumed and did _exit(0). * The task itself is dead and won't touch exit_code again; * other processors in this function are locked out. @@ -1202,13 +1327,20 @@ static int wait_task_stopped(task_t *p, int delayed_group_leader, int noreap, write_unlock_irq(&tasklist_lock); bail_ref: put_task_struct(p); - read_lock(&tasklist_lock); - return 0; + /* + * We are returning to the wait loop without having successfully + * removed the process and having released the lock. We cannot + * continue, since the "p" task pointer is potentially stale. + * + * Return -EAGAIN, and do_wait() will restart the loop from the + * beginning. Do _not_ re-acquire the lock. + */ + return -EAGAIN; } /* move to end of parent's list to avoid starvation */ remove_parent(p); - add_parent(p, p->parent); + add_parent(p); write_unlock_irq(&tasklist_lock); @@ -1237,6 +1369,75 @@ bail_ref: return retval; } +/* + * Handle do_wait work for one task in a live, non-stopped state. + * read_lock(&tasklist_lock) on entry. If we return zero, we still hold + * the lock and this task is uninteresting. If we return nonzero, we have + * released the lock and the system call should return. + */ +static int wait_task_continued(task_t *p, int noreap, + struct siginfo __user *infop, + int __user *stat_addr, struct rusage __user *ru) +{ + int retval; + pid_t pid; + uid_t uid; + + if (unlikely(!p->signal)) + return 0; + + if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) + return 0; + + spin_lock_irq(&p->sighand->siglock); + /* Re-check with the lock held. */ + if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) { + spin_unlock_irq(&p->sighand->siglock); + return 0; + } + if (!noreap) + p->signal->flags &= ~SIGNAL_STOP_CONTINUED; + spin_unlock_irq(&p->sighand->siglock); + + pid = p->pid; + uid = p->uid; + get_task_struct(p); + read_unlock(&tasklist_lock); + + if (!infop) { + retval = ru ? getrusage(p, RUSAGE_BOTH, ru) : 0; + put_task_struct(p); + if (!retval && stat_addr) + retval = put_user(0xffff, stat_addr); + if (!retval) + retval = p->pid; + } else { + retval = wait_noreap_copyout(p, pid, uid, + CLD_CONTINUED, SIGCONT, + infop, ru); + BUG_ON(retval == 0); + } + + return retval; +} + + +static inline int my_ptrace_child(struct task_struct *p) +{ + if (!(p->ptrace & PT_PTRACED)) + return 0; + if (!(p->ptrace & PT_ATTACHED)) + return 1; + /* + * This child was PTRACE_ATTACH'd. We should be seeing it only if + * we are the attacher. If we are the real parent, this is a race + * inside ptrace_attach. It is waiting for the tasklist_lock, + * which we have to switch the parent links, but has already set + * the flags in p->ptrace. + */ + return (p->parent != p->real_parent); +} + static long do_wait(pid_t pid, int options, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { @@ -1244,8 +1445,12 @@ static long do_wait(pid_t pid, int options, struct siginfo __user *infop, struct task_struct *tsk; int flag, retval; - add_wait_queue(¤t->wait_chldexit,&wait); + add_wait_queue(¤t->signal->wait_chldexit,&wait); repeat: + /* + * We will set this flag if we see any child that might later + * match our criteria, even if we are not able to reap it yet. + */ flag = 0; current->state = TASK_INTERRUPTIBLE; read_lock(&tasklist_lock); @@ -1261,66 +1466,75 @@ repeat: ret = eligible_child(pid, options, p); if (!ret) continue; - flag = 1; switch (p->state) { case TASK_TRACED: - if (!(p->ptrace & PT_PTRACED)) + /* + * When we hit the race with PTRACE_ATTACH, + * we will not report this child. But the + * race means it has not yet been moved to + * our ptrace_children list, so we need to + * set the flag here to avoid a spurious ECHILD + * when the race happens with the only child. + */ + flag = 1; + if (!my_ptrace_child(p)) continue; /*FALLTHROUGH*/ case TASK_STOPPED: + /* + * It's stopped now, so it might later + * continue, exit, or stop again. + */ + flag = 1; if (!(options & WUNTRACED) && - !(p->ptrace & PT_PTRACED)) + !my_ptrace_child(p)) continue; retval = wait_task_stopped(p, ret == 2, (options & WNOWAIT), infop, stat_addr, ru); + if (retval == -EAGAIN) + goto repeat; if (retval != 0) /* He released the lock. */ goto end; break; - case TASK_ZOMBIE: + default: + // case EXIT_DEAD: + if (p->exit_state == EXIT_DEAD) + continue; + // case EXIT_ZOMBIE: + if (p->exit_state == EXIT_ZOMBIE) { + /* + * Eligible but we cannot release + * it yet: + */ + if (ret == 2) + goto check_continued; + if (!likely(options & WEXITED)) + continue; + retval = wait_task_zombie( + p, (options & WNOWAIT), + infop, stat_addr, ru); + /* He released the lock. */ + if (retval != 0) + goto end; + break; + } +check_continued: /* - * Eligible but we cannot release it yet: + * It's running now, so it might later + * exit, stop, or stop and then continue. */ - if (ret == 2) - goto check_continued; - if (!likely(options & WEXITED)) + flag = 1; + if (!unlikely(options & WCONTINUED)) continue; - retval = wait_task_zombie( + retval = wait_task_continued( p, (options & WNOWAIT), infop, stat_addr, ru); if (retval != 0) /* He released the lock. */ goto end; break; - case TASK_DEAD: - continue; - default: -check_continued: - if (!unlikely(options & WCONTINUED)) - continue; - if (unlikely(!p->signal)) - continue; - spin_lock_irq(&p->sighand->siglock); - if (p->signal->stop_state < 0) { - pid_t pid; - uid_t uid; - - if (!(options & WNOWAIT)) - p->signal->stop_state = 0; - spin_unlock_irq(&p->sighand->siglock); - pid = p->pid; - uid = p->uid; - get_task_struct(p); - read_unlock(&tasklist_lock); - retval = wait_noreap_copyout(p, pid, - uid, CLD_CONTINUED, - SIGCONT, infop, ru); - BUG_ON(retval == 0); - goto end; - } - spin_unlock_irq(&p->sighand->siglock); - break; } } if (!flag) { @@ -1354,7 +1568,7 @@ check_continued: retval = -ECHILD; end: current->state = TASK_RUNNING; - remove_wait_queue(¤t->wait_chldexit,&wait); + remove_wait_queue(¤t->signal->wait_chldexit,&wait); if (infop) { if (retval > 0) retval = 0; @@ -1421,7 +1635,8 @@ asmlinkage long sys_wait4(pid_t pid, int __user *stat_addr, { long ret; - if (options & ~(WNOHANG|WUNTRACED|__WNOTHREAD|__WCLONE|__WALL)) + if (options & ~(WNOHANG|WUNTRACED|WCONTINUED| + __WNOTHREAD|__WCLONE|__WALL)) return -EINVAL; ret = do_wait(pid, options | WEXITED, NULL, stat_addr, ru);