Message ID | 874kr1la5z.fsf_-_@x220.int.ebiederm.org (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | exec: s/group_exit_task/group_exec_task/ for clarity | expand |
diff --git a/include/linux/sched/signal.h b/include/linux/sched/signal.h index 5ff8697b21cd..43822e2b63e6 100644 --- a/include/linux/sched/signal.h +++ b/include/linux/sched/signal.h @@ -268,8 +268,9 @@ static inline void signal_set_stop_flags(struct signal_struct *sig, /* If true, all threads except ->group_exit_task have pending SIGKILL */ static inline int signal_group_exit(const struct signal_struct *sig) { - return (sig->flags & SIGNAL_GROUP_EXIT) || - (sig->group_exit_task != NULL); + return (sig->flags & (SIGNAL_GROUP_EXIT | + SIGNAL_GROUP_COREDUMP | + SIGNAL_GROUP_DETHREAD)); } extern void flush_signals(struct task_struct *);
There are two places where signal_group_exit are set. In the fs/exec.c:de_thread() and in fs/coredump.c:zap_threads(). The coredump usage of group_exit_task was explicitly added[1] so that signal_group_exit() would return true during a coredump. When examining the coredump usage it turns out that SIGNAL_GROUP_COREDUMP is set in all of the same places as group_exit_task. So signal_group_exit can test SIGNAL_GROUP_COREDUMP and achieve the same results with respect to coredumps as testing group_exit_task. Similarly the exec code sets and clears SIGNAL_GROUP_DETHREAD in all of the places where group_exit_task is set and cleared. So test SIGNAL_GROUP_COREDUMP | SIGNAL_GROUP_DETHREAD instead of group_exit_task. Cc: Oleg Nesterov <oleg@redhat.com> [1] 6cd8f0acae34 ("coredump: ensure that SIGKILL always kills the dumping thread") Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> --- include/linux/sched/signal.h | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)