X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fexit.c;h=d27b470ec5aa69d0e211795f740e35c178542806;hb=6a77f38946aaee1cd85eeec6cf4229b204c15071;hp=45384fd56958bcba01dc9faaca27fc8865f32c32;hpb=87fc8d1bb10cd459024a742c6a10961fefcef18f;p=linux-2.6.git diff --git a/kernel/exit.c b/kernel/exit.c index 45384fd56..d27b470ec 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -14,6 +14,7 @@ #include #include #include +#include #include #include #include @@ -24,6 +25,7 @@ #include #include #include +#include #include #include @@ -81,7 +83,7 @@ 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); /* @@ -100,6 +102,8 @@ repeat: spin_unlock(&p->proc_lock); proc_pid_flush(proc_dentry); release_thread(p); + if (p->vx_info) + release_vx_info(p->vx_info, p); put_task_struct(p); p = leader; @@ -163,7 +167,7 @@ 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 @@ -244,11 +248,11 @@ void reparent_to_init(void) /* 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) @@ -337,7 +341,9 @@ void daemonize(const char *name, ...) exit_mm(current); set_special_pids(1, 1); + down(&tty_sem); current->signal->tty = NULL; + up(&tty_sem); /* Block and flush all signals */ sigfillset(&blocked); @@ -375,7 +381,7 @@ static inline void close_files(struct files_struct * files) struct file * file = xchg(&files->fd[i], NULL); if (file) filp_close(file, files); - // vx_openfd_dec(i); + vx_openfd_dec(i); } i++; set >>= 1; @@ -476,7 +482,7 @@ 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) +void exit_mm(struct task_struct * tsk) { struct mm_struct *mm = tsk->mm; @@ -512,20 +518,13 @@ static inline void __exit_mm(struct task_struct * tsk) mmput(mm); } -void exit_mm(struct task_struct *tsk) -{ - __exit_mm(tsk); -} - -EXPORT_SYMBOL(exit_mm); - 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 >= EXIT_ZOMBIE); p->real_parent = reaper; if (p->parent == p->real_parent) BUG(); @@ -536,7 +535,6 @@ static inline 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. */ @@ -560,7 +558,7 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced) /* 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 +567,7 @@ 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); } } @@ -609,7 +607,7 @@ static inline void forget_original_parent(struct task_struct * father, reaper = child_reaper; break; } - } while (reaper->state >= TASK_ZOMBIE); + } while (reaper->exit_state); /* * There are only two places where our children can be: @@ -635,7 +633,7 @@ static inline void forget_original_parent(struct task_struct * father, } 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); } @@ -646,7 +644,7 @@ 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) { @@ -666,7 +664,7 @@ 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 @@ -759,18 +757,19 @@ 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; + 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; /* * 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; + tsk->it_virt_value = cputime_zero; + tsk->it_prof_value = cputime_zero; write_unlock_irq(&tasklist_lock); @@ -781,7 +780,7 @@ 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. */ @@ -789,9 +788,10 @@ static void exit_notify(struct task_struct *tsk) 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); @@ -803,6 +803,12 @@ asmlinkage NORET_TYPE void do_exit(long code) 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); + } + tsk->flags |= PF_EXITING; del_timer_sync(&tsk->real_timer); @@ -811,21 +817,21 @@ asmlinkage NORET_TYPE void do_exit(long code) 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_process(code); - __exit_mm(tsk); + acct_update_integrals(); + update_mem_hiwater(); + group_dead = atomic_dec_and_test(&tsk->signal->live); + if (group_dead) + acct_process(code); + exit_mm(tsk); exit_sem(tsk); __exit_files(tsk); __exit_fs(tsk); exit_namespace(tsk); exit_thread(); + exit_keys(tsk); - if (tsk->signal->leader) + if (group_dead && tsk->signal->leader) disassociate_ctty(1); module_put(tsk->thread_info->exec_domain->module); @@ -838,6 +844,8 @@ asmlinkage NORET_TYPE void do_exit(long code) mpol_free(tsk->mempolicy); tsk->mempolicy = NULL; #endif + + BUG_ON(!(current->flags & PF_DEAD)); schedule(); BUG(); /* Avoid "noreturn function does return". */ @@ -861,13 +869,6 @@ asmlinkage long sys_exit(int error_code) 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); } @@ -882,18 +883,18 @@ 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->flags = SIGNAL_GROUP_EXIT; sig->group_exit_code = exit_code; zap_other_threads(current); } @@ -981,7 +982,7 @@ 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 +1001,7 @@ 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 +1022,9 @@ 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)) { @@ -1051,10 +1052,16 @@ 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->cutime = + cputime_add(p->parent->signal->cutime, + cputime_add(p->utime, + cputime_add(p->signal->utime, + p->signal->cutime))); + p->parent->signal->cstime = + cputime_add(p->parent->signal->cstime, + cputime_add(p->stime, + cputime_add(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 += @@ -1068,12 +1075,12 @@ static int wait_task_zombie(task_t *p, int noreap, /* * 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 +1107,8 @@ 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 +1117,8 @@ 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 @@ -1180,13 +1189,13 @@ 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. @@ -1202,8 +1211,15 @@ 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 */ @@ -1237,6 +1253,75 @@ 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 +1329,12 @@ 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 +1350,66 @@ repeat: ret = eligible_child(pid, options, p); if (!ret) continue; - flag = 1; switch (p->state) { case TASK_TRACED: - if (!(p->ptrace & PT_PTRACED)) + 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 +1443,7 @@ 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 +1510,8 @@ 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);