X-Git-Url: http://git.onelab.eu/?a=blobdiff_plain;f=kernel%2Fexit.c;h=99e79125ef031a26e6208aaac833b5e1fa493993;hb=16c70f8c1b54b61c3b951b6fb220df250fe09b32;hp=08eee6d20cf89644a452942a9d257b3d983ec26b;hpb=4e76c8a9fa413ccc09d3f7f664183dcce3555d57;p=linux-2.6.git diff --git a/kernel/exit.c b/kernel/exit.c index 08eee6d20..99e79125e 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -4,7 +4,6 @@ * Copyright (C) 1991, 1992 Linus Torvalds */ -#include #include #include #include @@ -22,10 +21,13 @@ #include #include #include +#include #include #include #include #include +#include +#include #include #include #include @@ -36,10 +38,10 @@ #include #include #include /* for audit_free() */ +#include #include #include #include -#include #include #include @@ -49,8 +51,6 @@ extern void sem_exit (void); 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) @@ -140,17 +140,13 @@ static void delayed_put_task_struct(struct rcu_head *rhp) void release_task(struct task_struct * p) { + struct task_struct *leader; int zap_leader; - task_t *leader; - struct dentry *proc_dentry; - repeat: + tracehook_release_task(p); atomic_dec(&p->user->processes); - spin_lock(&p->proc_lock); - proc_dentry = proc_pid_unhash(p); write_lock_irq(&tasklist_lock); - ptrace_unlink(p); - BUG_ON(!list_empty(&p->ptrace_list) || !list_empty(&p->ptrace_children)); + BUG_ON(tracehook_check_released(p)); __exit_signal(p); /* @@ -176,8 +172,7 @@ repeat: sched_exit(p); write_unlock_irq(&tasklist_lock); - spin_unlock(&p->proc_lock); - proc_pid_flush(proc_dentry); + proc_flush_task(p); release_thread(p); call_rcu(&p->rcu, delayed_put_task_struct); @@ -220,7 +215,7 @@ out: * * "I ask you, have you ever known what it is to be an orphan?" */ -static int will_become_orphaned_pgrp(int pgrp, task_t *ignored_task) +static int will_become_orphaned_pgrp(int pgrp, struct task_struct *ignored_task) { struct task_struct *p; int ret = 1; @@ -228,10 +223,10 @@ 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->exit_state - || p->real_parent->pid == 1) + || p->parent->pid == 1) continue; - if (process_group(p->real_parent) != pgrp - && p->real_parent->signal->session == p->signal->session) { + if (process_group(p->parent) != pgrp + && p->parent->signal->session == p->signal->session) { ret = 0; break; } @@ -259,16 +254,6 @@ static int has_stopped_jobs(int pgrp) if (p->state != TASK_STOPPED) continue; - /* If p is stopped by a debugger on a signal that won't - stop it, then don't count p as stopped. This isn't - perfect but it's a good approximation. */ - if (unlikely (p->ptrace) - && p->exit_code != SIGSTOP - && p->exit_code != SIGTSTP - && p->exit_code != SIGTTOU - && p->exit_code != SIGTTIN) - continue; - retval = 1; break; } while_each_task_pid(pgrp, PIDTYPE_PGID, p); @@ -291,11 +276,9 @@ static void reparent_to_init(void) { write_lock_irq(&tasklist_lock); - ptrace_unlink(current); /* Reparent to init */ remove_parent(current); current->parent = child_reaper; - current->real_parent = child_reaper; add_parent(current); /* Set the exit signal to SIGCHLD so we signal init on exit */ @@ -585,7 +568,7 @@ static void exit_mm(struct task_struct * tsk) down_read(&mm->mmap_sem); } atomic_inc(&mm->mm_count); - if (mm != tsk->active_mm) BUG(); + BUG_ON(mm != tsk->active_mm); /* more a memory barrier than a real lock */ task_lock(tsk); tsk->mm = NULL; @@ -595,7 +578,8 @@ static void exit_mm(struct task_struct * tsk) mmput(mm); } -static inline void choose_new_parent(task_t *p, task_t *reaper) +static inline void +choose_new_parent(struct task_struct *p, struct task_struct *reaper) { /* check for reaper context */ vxwprintk((p->xid != reaper->xid) && (reaper != child_reaper), @@ -607,10 +591,11 @@ static inline void choose_new_parent(task_t *p, task_t *reaper) * the parent is not a zombie. */ BUG_ON(p == reaper || reaper->exit_state); - p->real_parent = reaper; + p->parent = reaper; } -static void reparent_thread(task_t *p, task_t *father, int traced) +static void +reparent_thread(struct task_struct *p, struct task_struct *father) { /* We don't want people slaying init. */ if (p->exit_signal != -1) @@ -621,35 +606,14 @@ static void reparent_thread(task_t *p, task_t *father, int traced) 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)) { - /* Preserve ptrace links if someone else is tracing this child. */ - list_del_init(&p->ptrace_list); - if (p->parent != p->real_parent) - list_add(&p->ptrace_list, &p->real_parent->ptrace_children); - } else { - /* If this child is being traced, then we're the one tracing it - * anyway, so let go of it. - */ - p->ptrace = 0; - remove_parent(p); - p->parent = p->real_parent; - add_parent(p); + list_move_tail(&p->sibling, &p->parent->children); - /* If we'd notified the old parent about this child's death, - * also notify the new parent. - */ - 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) { - /* - * If it was at a trace stop, turn it into - * a normal stop since it's no longer being - * traced. - */ - ptrace_untrace(p); - } - } + /* If we'd notified the old parent about this child's death, + * also notify the new parent. + */ + if (p->exit_state == EXIT_ZOMBIE && p->exit_signal != -1 && + thread_group_empty(p)) + do_notify_parent(p, p->exit_signal); /* * process group orphan check @@ -674,8 +638,8 @@ static 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 void forget_original_parent(struct task_struct * father, - struct list_head *to_release) +static void +forget_original_parent(struct task_struct *father) { struct task_struct *p, *reaper = father; struct list_head *_p, *_n; @@ -688,48 +652,10 @@ static void forget_original_parent(struct task_struct * father, } } while (reaper->exit_state); - /* - * There are only two places where our children can be: - * - * - in our child list - * - in our ptraced child list - * - * Search them and reparent children. - */ list_for_each_safe(_p, _n, &father->children) { - int ptrace; - p = list_entry(_p,struct task_struct,sibling); - - ptrace = p->ptrace; - - /* if father isn't the real parent, then ptrace must be enabled */ - BUG_ON(father != p->real_parent && !ptrace); - - if (father == p->real_parent) { - /* reparent with a reaper, real father it's us */ - 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->exit_state == EXIT_ZOMBIE && p->exit_signal != -1 && - thread_group_empty(p)) - do_notify_parent(p, p->exit_signal); - } - - /* - * if the ptraced child is a zombie with exit_signal == -1 - * we must collect it before we exit, or it will remain - * 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->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); - reparent_thread(p, father, 1); + p = list_entry(_p, struct task_struct, sibling); + choose_new_parent(p, vx_child_reaper(p)); + reparent_thread(p, father); } } @@ -741,7 +667,8 @@ static void exit_notify(struct task_struct *tsk) { int state; struct task_struct *t; - struct list_head ptrace_dead, *_p, *_n; + int noreap; + void *cookie; if (signal_pending(tsk) && !(tsk->signal->flags & SIGNAL_GROUP_EXIT) && !thread_group_empty(tsk)) { @@ -777,10 +704,8 @@ static void exit_notify(struct task_struct *tsk) * jobs, send them a SIGHUP and then a SIGCONT. (POSIX 3.2.2.2) */ - INIT_LIST_HEAD(&ptrace_dead); - forget_original_parent(tsk, &ptrace_dead); + forget_original_parent(tsk); BUG_ON(!list_empty(&tsk->children)); - BUG_ON(!list_empty(&tsk->ptrace_children)); /* * Check to see if any process groups have become orphaned @@ -792,7 +717,7 @@ static void exit_notify(struct task_struct *tsk) * is about to become orphaned. */ - t = tsk->real_parent; + t = tsk->parent; if ((process_group(t) != process_group(tsk)) && (t->signal->session == tsk->signal->session) && @@ -824,32 +749,18 @@ static void exit_notify(struct task_struct *tsk) && !capable(CAP_KILL)) tsk->exit_signal = SIGCHLD; - - /* If something other than our normal parent is ptracing us, then - * send it a SIGCHLD instead of honoring exit_signal. exit_signal - * only has special meaning to our real parent. - */ - if (tsk->exit_signal != -1 && thread_group_empty(tsk)) { - int signal = tsk->parent == tsk->real_parent ? tsk->exit_signal : SIGCHLD; - do_notify_parent(tsk, signal); - } else if (tsk->ptrace) { - do_notify_parent(tsk, SIGCHLD); - } + if (!tracehook_notify_death(tsk, &noreap, &cookie) + && tsk->exit_signal != -1 && thread_group_empty(tsk)) + do_notify_parent(tsk, tsk->exit_signal); state = EXIT_ZOMBIE; - if (tsk->exit_signal == -1 && - (likely(tsk->ptrace == 0) || - unlikely(tsk->parent->signal->flags & SIGNAL_GROUP_EXIT))) + if (tsk->exit_signal == -1 && !noreap) state = EXIT_DEAD; tsk->exit_state = state; write_unlock_irq(&tasklist_lock); - list_for_each_safe(_p, _n, &ptrace_dead) { - list_del_init(_p); - t = list_entry(_p,struct task_struct,ptrace_list); - release_task(t); - } + tracehook_report_death(tsk, state, cookie); /* If the process is dead, release it - nobody will wait for it */ if (state == EXIT_DEAD) @@ -859,7 +770,9 @@ static void exit_notify(struct task_struct *tsk) fastcall NORET_TYPE void do_exit(long code) { struct task_struct *tsk = current; + struct taskstats *tidstats; int group_dead; + unsigned int mycpu; profile_task_exit(tsk); @@ -872,10 +785,7 @@ fastcall NORET_TYPE void do_exit(long code) if (unlikely(tsk == child_reaper)) panic("Attempted to kill init!"); - if (unlikely(current->ptrace & PT_TRACE_EXIT)) { - current->ptrace_message = code; - ptrace_notify((PTRACE_EVENT_EXIT << 8) | SIGTRAP); - } + tracehook_report_exit(&code); /* * We're taking recursive faults here in do_exit. Safest is to just @@ -892,11 +802,15 @@ fastcall NORET_TYPE void do_exit(long code) tsk->flags |= PF_EXITING; + ptrace_exit(tsk); + if (unlikely(in_atomic())) printk(KERN_INFO "note: %s[%d] exited with preempt_count %d\n", current->comm, current->pid, preempt_count()); + taskstats_exit_alloc(&tidstats, &mycpu); + acct_update_integrals(tsk); if (tsk->mm) { update_hiwater_rss(tsk->mm); @@ -906,15 +820,17 @@ fastcall NORET_TYPE void do_exit(long code) if (group_dead) { hrtimer_cancel(&tsk->signal->real_timer); exit_itimers(tsk->signal); - acct_process(code); - if (current->tux_info) { + } + + if (current->tux_info) { #ifdef CONFIG_TUX_DEBUG - printk("Possibly unexpected TUX-thread exit(%ld) at %p?\n", - code, __builtin_return_address(0)); + printk("Possibly unexpected TUX-thread exit(%ld) at %p?\n", + code, __builtin_return_address(0)); #endif - current->tux_exit(); - } + current->tux_exit(); } + + acct_collect(code, group_dead); if (unlikely(tsk->robust_list)) exit_robust_list(tsk); #if defined(CONFIG_FUTEX) && defined(CONFIG_COMPAT) @@ -923,14 +839,17 @@ fastcall NORET_TYPE void do_exit(long code) #endif if (unlikely(tsk->audit_context)) audit_free(tsk); + taskstats_exit_send(tsk, tidstats, group_dead, mycpu); + taskstats_exit_free(tidstats); + exit_mm(tsk); + if (group_dead) + acct_process(); 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); @@ -944,15 +863,25 @@ fastcall NORET_TYPE void do_exit(long code) tsk->exit_code = code; proc_exit_connector(tsk); + /* needs to stay before exit_notify() */ + exit_vx_info_early(tsk, code); 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: + * This must happen late, after the PID is not + * hashed anymore: */ - mutex_debug_check_no_locks_held(tsk); + if (unlikely(!list_empty(&tsk->pi_state_list))) + exit_pi_state_list(tsk); + if (unlikely(current->pi_state_cache)) + kfree(current->pi_state_cache); + /* + * Make sure we are holding no locks: + */ + debug_check_no_locks_held(tsk); if (tsk->io_context) exit_io_context(); @@ -960,6 +889,10 @@ fastcall NORET_TYPE void do_exit(long code) if (tsk->splice_pipe) __free_pipe_info(tsk->splice_pipe); + /* needs to stay after exit_notify() */ + exit_vx_info(tsk, code); + exit_nx_info(tsk); + /* PF_DEAD causes final put_task_struct after we schedule. */ preempt_disable(); BUG_ON(tsk->flags & PF_DEAD); @@ -1027,7 +960,7 @@ asmlinkage void sys_exit_group(int error_code) do_group_exit((error_code & 0xff) << 8); } -static int eligible_child(pid_t pid, int options, task_t *p) +static int eligible_child(pid_t pid, int options, struct task_struct *p) { if (pid > 0) { if (p->pid != pid) @@ -1041,10 +974,9 @@ static int eligible_child(pid_t pid, int options, task_t *p) } /* - * Do not consider detached threads that are - * not ptraced: + * Do not consider detached threads. */ - if (p->exit_signal == -1 && !p->ptrace) + if (p->exit_signal == -1) return 0; /* Wait for all children (clone and not) if __WALL is set; @@ -1059,7 +991,7 @@ static int eligible_child(pid_t pid, int options, task_t *p) * Do not consider thread group leaders that are * in a non-empty thread group: */ - if (current->tgid != p->tgid && delay_group_leader(p)) + if (delay_group_leader(p)) return 2; if (security_task_wait(p)) @@ -1068,12 +1000,13 @@ static int eligible_child(pid_t pid, int options, task_t *p) return 1; } -static int wait_noreap_copyout(task_t *p, pid_t pid, uid_t uid, +static int wait_noreap_copyout(struct task_struct *p, pid_t pid, uid_t uid, int why, int status, struct siginfo __user *infop, struct rusage __user *rusagep) { int retval = rusagep ? getrusage(p, RUSAGE_BOTH, rusagep) : 0; + put_task_struct(p); if (!retval) retval = put_user(SIGCHLD, &infop->si_signo); @@ -1098,7 +1031,7 @@ static int wait_noreap_copyout(task_t *p, pid_t pid, uid_t uid, * 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_zombie(task_t *p, int noreap, +static int wait_task_zombie(struct task_struct *p, int noreap, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { @@ -1114,7 +1047,7 @@ static int wait_task_zombie(task_t *p, int noreap, if (unlikely(p->exit_state != EXIT_ZOMBIE)) return 0; - if (unlikely(p->exit_signal == -1 && p->ptrace == 0)) + if (unlikely(p->exit_signal == -1)) return 0; get_task_struct(p); read_unlock(&tasklist_lock); @@ -1138,15 +1071,9 @@ static int wait_task_zombie(task_t *p, int noreap, BUG_ON(state != EXIT_DEAD); return 0; } - if (unlikely(p->exit_signal == -1 && p->ptrace == 0)) { - /* - * This can only happen in a race with a ptraced thread - * dying on another processor. - */ - return 0; - } + BUG_ON(p->exit_signal == -1); - if (likely(p->real_parent == p->parent) && likely(p->signal)) { + if (likely(p->signal)) { struct signal_struct *psig; struct signal_struct *sig; @@ -1228,28 +1155,8 @@ static int wait_task_zombie(task_t *p, int noreap, return retval; } retval = p->pid; - if (p->real_parent != p->parent) { - write_lock_irq(&tasklist_lock); - /* Double-check with lock held. */ - if (p->real_parent != p->parent) { - __ptrace_unlink(p); - // 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 - * it now. - */ - if (p->exit_signal != -1) { - do_notify_parent(p, p->exit_signal); - if (p->exit_signal != -1) - p = NULL; - } - } - write_unlock_irq(&tasklist_lock); - } - if (p != NULL) - release_task(p); + release_task(p); + BUG_ON(!retval); return retval; } @@ -1260,15 +1167,15 @@ static int wait_task_zombie(task_t *p, int noreap, * 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_stopped(task_t *p, int delayed_group_leader, int noreap, - struct siginfo __user *infop, +static int wait_task_stopped(struct task_struct *p, int delayed_group_leader, + int noreap, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { int retval, exit_code; if (!p->exit_code) return 0; - if (delayed_group_leader && !(p->ptrace & PT_PTRACED) && + if (delayed_group_leader && p->signal && p->signal->group_stop_count > 0) /* * A group stop is in progress and this is the group leader. @@ -1289,14 +1196,13 @@ static int wait_task_stopped(task_t *p, int delayed_group_leader, int noreap, if (unlikely(noreap)) { pid_t pid = p->pid; uid_t uid = p->uid; - int why = (p->ptrace & PT_PTRACED) ? CLD_TRAPPED : CLD_STOPPED; exit_code = p->exit_code; if (unlikely(!exit_code) || unlikely(p->state & TASK_TRACED)) goto bail_ref; - return wait_noreap_copyout(p, pid, uid, - why, (exit_code << 8) | 0x7f, + return wait_noreap_copyout(p, pid, uid, CLD_STOPPED, + (exit_code << 8) | 0x7f, infop, ru); } @@ -1352,9 +1258,7 @@ bail_ref: if (!retval && infop) retval = put_user(0, &infop->si_errno); if (!retval && infop) - retval = put_user((short)((p->ptrace & PT_PTRACED) - ? CLD_TRAPPED : CLD_STOPPED), - &infop->si_code); + retval = put_user((short)CLD_STOPPED, &infop->si_code); if (!retval && infop) retval = put_user(exit_code, &infop->si_status); if (!retval && infop) @@ -1375,7 +1279,7 @@ bail_ref: * 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, +static int wait_task_continued(struct task_struct *p, int noreap, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { @@ -1422,22 +1326,6 @@ static int wait_task_continued(task_t *p, int noreap, } -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) { @@ -1461,7 +1349,7 @@ repeat: int ret; list_for_each(_p,&tsk->children) { - p = list_entry(_p,struct task_struct,sibling); + p = list_entry(_p, struct task_struct, sibling); ret = eligible_child(pid, options, p); if (!ret) @@ -1469,26 +1357,17 @@ repeat: switch (p->state) { case TASK_TRACED: - /* - * 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*/ + continue; case TASK_STOPPED: /* * It's stopped now, so it might later * continue, exit, or stop again. */ flag = 1; - if (!(options & WUNTRACED) && - !my_ptrace_child(p)) + if (!(options & WUNTRACED)) + continue; + if (tracehook_inhibit_wait_stopped(p)) continue; retval = wait_task_stopped(p, ret == 2, (options & WNOWAIT), @@ -1513,6 +1392,10 @@ repeat: goto check_continued; if (!likely(options & WEXITED)) continue; + if (tracehook_inhibit_wait_zombie(p)) { + flag = 1; + continue; + } retval = wait_task_zombie( p, (options & WNOWAIT), infop, stat_addr, ru); @@ -1529,6 +1412,8 @@ check_continued: flag = 1; if (!unlikely(options & WCONTINUED)) continue; + if (tracehook_inhibit_wait_continued(p)) + continue; retval = wait_task_continued( p, (options & WNOWAIT), infop, stat_addr, ru); @@ -1537,21 +1422,19 @@ check_continued: break; } } - if (!flag) { - list_for_each(_p, &tsk->ptrace_children) { - p = list_entry(_p, struct task_struct, - ptrace_list); - if (!eligible_child(pid, options, p)) - continue; - flag = 1; - break; - } + + retval = ptrace_do_wait(tsk, pid, options, + infop, stat_addr, ru); + if (retval != -ECHILD) { + flag = 1; + if (retval != 0) /* He released the lock. */ + goto end; } + if (options & __WNOTHREAD) break; tsk = next_thread(tsk); - if (tsk->signal != current->signal) - BUG(); + BUG_ON(tsk->signal != current->signal); } while (tsk != current); read_unlock(&tasklist_lock); @@ -1571,7 +1454,7 @@ end: remove_wait_queue(¤t->signal->wait_chldexit,&wait); if (infop) { if (retval > 0) - retval = 0; + retval = 0; else { /* * For a WNOHANG return, clear out all the fields