The patch titled ptrace: do_wait reorganization has been removed from the -mm tree. Its filename was ptrace-do_wait-reorganization.patch This patch was dropped because it was merged into mainline or a subsystem tree The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: ptrace: do_wait reorganization From: Roland McGrath <roland@xxxxxxxxxx> This breaks out the guts of do_wait into three subfunctions. The control flow is less nonobvious without so much goto. do_wait_thread and ptrace_do_wait contain the main work of the outer loop. wait_consider_task contains the main work of the inner loop. Signed-off-by: Roland McGrath <roland@xxxxxxxxxx> Cc: Oleg Nesterov <oleg@xxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- kernel/exit.c | 215 ++++++++++++++++++++++++++++++------------------ 1 file changed, 135 insertions(+), 80 deletions(-) diff -puN kernel/exit.c~ptrace-do_wait-reorganization kernel/exit.c --- a/kernel/exit.c~ptrace-do_wait-reorganization +++ a/kernel/exit.c @@ -1230,7 +1230,7 @@ static int wait_noreap_copyout(struct ta * 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(struct task_struct *p, int noreap, +static int wait_task_zombie(struct task_struct *p, int options, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { @@ -1238,7 +1238,10 @@ static int wait_task_zombie(struct task_ int retval, status, traced; pid_t pid = task_pid_vnr(p); - if (unlikely(noreap)) { + if (!likely(options & WEXITED)) + return 0; + + if (unlikely(options & WNOWAIT)) { uid_t uid = p->uid; int exit_code = p->exit_code; int why, status; @@ -1389,13 +1392,16 @@ static int wait_task_zombie(struct task_ * released the lock and the system call should return. */ static int wait_task_stopped(struct task_struct *p, - int noreap, struct siginfo __user *infop, + int options, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { int retval, exit_code, why; uid_t uid = 0; /* unneeded, required by compiler */ pid_t pid; + if (!(p->ptrace & PT_PTRACED) && !(options & WUNTRACED)) + return 0; + exit_code = 0; spin_lock_irq(&p->sighand->siglock); @@ -1413,7 +1419,7 @@ static int wait_task_stopped(struct task if (!exit_code) goto unlock_sig; - if (!noreap) + if (!unlikely(options & WNOWAIT)) p->exit_code = 0; uid = p->uid; @@ -1434,7 +1440,7 @@ unlock_sig: why = (p->ptrace & PT_PTRACED) ? CLD_TRAPPED : CLD_STOPPED; read_unlock(&tasklist_lock); - if (unlikely(noreap)) + if (unlikely(options & WNOWAIT)) return wait_noreap_copyout(p, pid, uid, why, exit_code, infop, ru); @@ -1468,7 +1474,7 @@ unlock_sig: * 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(struct task_struct *p, int noreap, +static int wait_task_continued(struct task_struct *p, int options, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { @@ -1476,6 +1482,9 @@ static int wait_task_continued(struct ta pid_t pid; uid_t uid; + if (!unlikely(options & WCONTINUED)) + return 0; + if (!(p->signal->flags & SIGNAL_STOP_CONTINUED)) return 0; @@ -1485,7 +1494,7 @@ static int wait_task_continued(struct ta spin_unlock_irq(&p->sighand->siglock); return 0; } - if (!noreap) + if (!unlikely(options & WNOWAIT)) p->signal->flags &= ~SIGNAL_STOP_CONTINUED; spin_unlock_irq(&p->sighand->siglock); @@ -1511,89 +1520,137 @@ static int wait_task_continued(struct ta return retval; } +/* + * Consider @p for a wait by @parent. + * + * -ECHILD should be in *@notask_error before the first call. + * Returns nonzero for a final return, when we have unlocked tasklist_lock. + * Returns zero if the search for a child should continue; + * then *@notask_error is 0 if @p is an eligible child, or still -ECHILD. + */ +static int wait_consider_task(struct task_struct *parent, + struct task_struct *p, int *notask_error, + enum pid_type type, struct pid *pid, int options, + struct siginfo __user *infop, + int __user *stat_addr, struct rusage __user *ru) +{ + int ret = eligible_child(type, pid, options, p); + if (ret <= 0) + return ret; + + if (p->exit_state == EXIT_DEAD) + return 0; + + /* + * We don't reap group leaders with subthreads. + */ + if (p->exit_state == EXIT_ZOMBIE && !delay_group_leader(p)) + return wait_task_zombie(p, options, infop, stat_addr, ru); + + /* + * It's stopped or running now, so it might + * later continue, exit, or stop again. + */ + *notask_error = 0; + + if (task_is_stopped_or_traced(p)) + return wait_task_stopped(p, options, infop, stat_addr, ru); + + return wait_task_continued(p, options, infop, stat_addr, ru); +} + +/* + * Do the work of do_wait() for one thread in the group, @tsk. + * + * -ECHILD should be in *@notask_error before the first call. + * Returns nonzero for a final return, when we have unlocked tasklist_lock. + * Returns zero if the search for a child should continue; then + * *@notask_error is 0 if there were any eligible children, or still -ECHILD. + */ +static int do_wait_thread(struct task_struct *tsk, int *notask_error, + enum pid_type type, struct pid *pid, int options, + struct siginfo __user *infop, int __user *stat_addr, + struct rusage __user *ru) +{ + struct task_struct *p; + + list_for_each_entry(p, &tsk->children, sibling) { + int ret = wait_consider_task(tsk, p, notask_error, + type, pid, options, + infop, stat_addr, ru); + if (ret) + return ret; + } + + return 0; +} + +static int ptrace_do_wait(struct task_struct *tsk, int *notask_error, + enum pid_type type, struct pid *pid, int options, + struct siginfo __user *infop, int __user *stat_addr, + struct rusage __user *ru) +{ + struct task_struct *p; + + /* + * If we never saw an eligile child, check for children stolen by + * ptrace. We don't leave -ECHILD in *@notask_error if there are any, + * because we will eventually be allowed to wait for them again. + */ + if (!*notask_error) + return 0; + + list_for_each_entry(p, &tsk->ptrace_children, ptrace_list) { + int ret = eligible_child(type, pid, options, p); + if (unlikely(ret < 0)) + return ret; + if (ret) { + *notask_error = 0; + return 0; + } + } + + return 0; +} + static long do_wait(enum pid_type type, struct pid *pid, int options, struct siginfo __user *infop, int __user *stat_addr, struct rusage __user *ru) { DECLARE_WAITQUEUE(wait, current); struct task_struct *tsk; - int flag, retval; + int retval; add_wait_queue(¤t->signal->wait_chldexit,&wait); repeat: - /* If there is nothing that can match our critier just get out */ + /* + * If there is nothing that can match our critiera just get out. + * We will clear @retval to zero if we see any child that might later + * match our criteria, even if we are not able to reap it yet. + */ retval = -ECHILD; if ((type < PIDTYPE_MAX) && (!pid || hlist_empty(&pid->tasks[type]))) goto end; - /* - * 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 = retval = 0; current->state = TASK_INTERRUPTIBLE; read_lock(&tasklist_lock); tsk = current; do { - struct task_struct *p; - - list_for_each_entry(p, &tsk->children, sibling) { - int ret = eligible_child(type, pid, options, p); - if (!ret) - continue; - - if (unlikely(ret < 0)) { - retval = ret; - } else if (task_is_stopped_or_traced(p)) { - /* - * It's stopped now, so it might later - * continue, exit, or stop again. - */ - flag = 1; - if (!(p->ptrace & PT_PTRACED) && - !(options & WUNTRACED)) - continue; - - retval = wait_task_stopped(p, - (options & WNOWAIT), infop, - stat_addr, ru); - } else if (p->exit_state == EXIT_ZOMBIE && - !delay_group_leader(p)) { - /* - * We don't reap group leaders with subthreads. - */ - if (!likely(options & WEXITED)) - continue; - retval = wait_task_zombie(p, - (options & WNOWAIT), infop, - stat_addr, ru); - } else if (p->exit_state != EXIT_DEAD) { - /* - * It's running now, so it might later - * exit, stop, or stop and then continue. - */ - flag = 1; - if (!unlikely(options & WCONTINUED)) - continue; - retval = wait_task_continued(p, - (options & WNOWAIT), infop, - stat_addr, ru); - } - if (retval != 0) /* tasklist_lock released */ - goto end; - } - if (!flag) { - list_for_each_entry(p, &tsk->ptrace_children, - ptrace_list) { - flag = eligible_child(type, pid, options, p); - if (!flag) - continue; - if (likely(flag > 0)) - break; - retval = flag; - goto end; - } + int tsk_result = do_wait_thread(tsk, &retval, + type, pid, options, + infop, stat_addr, ru); + if (!tsk_result) + tsk_result = ptrace_do_wait(tsk, &retval, + type, pid, options, + infop, stat_addr, ru); + if (tsk_result) { + /* + * tasklist_lock is unlocked and we have a final result. + */ + retval = tsk_result; + goto end; } + if (options & __WNOTHREAD) break; tsk = next_thread(tsk); @@ -1601,16 +1658,14 @@ repeat: } while (tsk != current); read_unlock(&tasklist_lock); - if (flag) { - if (options & WNOHANG) - goto end; + if (!retval && !(options & WNOHANG)) { retval = -ERESTARTSYS; - if (signal_pending(current)) - goto end; - schedule(); - goto repeat; + if (!signal_pending(current)) { + schedule(); + goto repeat; + } } - retval = -ECHILD; + end: current->state = TASK_RUNNING; remove_wait_queue(¤t->signal->wait_chldexit,&wait); _ Patches currently in -mm which might be from roland@xxxxxxxxxx are origin.patch markers-fix-duplicate-modpost-entry.patch linux-next.patch execve-filename-document-and-export-via-auxiliary-vector.patch flag-parameters-paccept-fix.patch flag-parameters-paccept-fix-flag-parameters-arch-specific-sock_nonblock.patch posix-timers-timer_delete-remove-the-bogus-it_process-=-null-check.patch posix-timers-release_posix_timer-kill-the-bogus-put_task_struct-it_process.patch signals-collect_signal-remove-the-unneeded-sigismember-check.patch signals-collect_signal-simplify-the-still_pending-logic.patch __exit_signal-dont-take-rcu-lock.patch signals-dequeue_signal-dont-check-signal_group_exit-when-setting-signal_stop_dequeued.patch signals-do_signal_stop-kill-the-signal_unkillable-check.patch coredump-zap_threads-comments-use-while_each_thread.patch signals-make-siginfo_t-si_utime-si_sstime-report-times-in-user_hz-not-hz.patch kernel-signalc-change-vars-pid-and-tgid-types-to-pid_t.patch include-asm-ptraceh-userspace-headers-cleanup.patch ptrace-give-more-respect-to-sigkill.patch ptrace-simplify-ptrace_stop-sigkill_pending-path.patch introduce-pf_kthread-flag.patch kill-pf_borrowed_mm-in-favour-of-pf_kthread.patch coredump-zap_threads-must-skip-kernel-threads.patch coredump-elf_core_dump-skip-kernel-threads.patch coredump-turn-mm-core_startup_done-into-the-pointer-to-struct-core_state.patch coredump-move-mm-core_waiters-into-struct-core_state.patch coredump-simplify-core_state-nr_threads-calculation.patch coredump-turn-core_state-nr_threads-into-atomic_t.patch coredump-make-mm-core_state-visible-to-core_dump.patch coredump-construct-the-list-of-coredumping-threads-at-startup-time.patch coredump-elf_core_dump-use-core_state-dumper-list.patch coredump-elf_fdpic_core_dump-use-core_state-dumper-list.patch coredump-kill-mm-core_done.patch coredump-binfmt_elf_fdpic-dont-use-sub-threads-mm.patch coredump-exit_mm-clear-mm-first-then-play-with-core_state.patch coredump-exit_mm-clear-mm-first-then-play-with-core_state-checkpatch-fixes.patch tracehook-add-linux-tracehookh.patch tracehook-exec.patch tracehook-unexport-ptrace_notify.patch tracehook-exit.patch tracehook-clone.patch tracehook-vfork-done.patch tracehook-release_task.patch tracehook-tracehook_tracer_task.patch tracehook-tracehook_expect_breakpoints.patch tracehook-tracehook_signal_handler.patch tracehook-tracehook_consider_ignored_signal.patch tracehook-tracehook_consider_fatal_signal.patch tracehook-syscall.patch tracehook-get_signal_to_deliver.patch tracehook-job-control.patch tracehook-death.patch tracehook-force-signal_pending.patch tracehook-tif_notify_resume.patch tracehook-asm-syscallh.patch tracehook-config_have_arch_tracehook.patch tracehook-wait_task_inactive.patch task_current_syscall.patch proc-pid-syscall.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html