will_become_orphaned_pgrp: partially fix insufficient ->exit_state check
authorOleg Nesterov <oleg@tv-sign.ru>
Sun, 2 Mar 2008 18:44:42 +0000 (21:44 +0300)
committerLinus Torvalds <torvalds@woody.linux-foundation.org>
Mon, 3 Mar 2008 22:53:16 +0000 (14:53 -0800)
commit05e83df624fe682bb8571cdb2c6d5284a99c3066
treec55df1d1258f2fae4069aa7954a2c74ba1831192
parentf49ee505b1ecb5960984880740f09aba87f870dc
will_become_orphaned_pgrp: partially fix insufficient ->exit_state check

p->exit_state != 0 doesn't mean this process is dead, it may have
sub-threads.  Change the code to use "p->exit_state && thread_group_empty(p)"
instead.

Without this patch, ^Z doesn't deliver SIGTSTP to the foreground process
if the main thread has exited.

However, the new check is not perfect either.  There is a window when
exit_notify() drops tasklist and before release_task().  Suppose that
the last (non-leader) thread exits.  This means that entire group exits,
but thread_group_empty() is not true yet.

As Eric pointed out, is_global_init() is wrong as well, but I did not
dare to do other changes.

Just for the record, has_stopped_jobs() is absolutely wrong too.  But we
can't fix it now, we should first fix SIGNAL_STOP_STOPPED issues.

Even with this patch ^Z doesn't play well with the dead main thread.
The task is stopped correctly but do_wait(WSTOPPED) won't see it.  This
is another unrelated issue, will be (hopefully) fixed separately.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
kernel/exit.c