which fixes a real bug.

* thread.c (struct waiting_fd): declare
  (rb_thread_io_blocking_region): use on-stack list waiter
  (rb_notify_fd_close): walk vm->waiting_fds instead
  (call_without_gvl): remove old field setting
  (th_init): ditto
  [Feature #9632]

* vm_core.h (typedef struct rb_vm_struct): add waiting_fds list

* (typedef struct rb_thread_struct): remove waiting_fd field
  (rb_vm_living_threads_init): initialize waiting_fds list

This should fix bad interactions with test_race_gets_and_close
in test/ruby/test_io.rb since we ensure rb_notify_fd_close
continues returning the busy flag after enqueuing the interrupt.

* thread.c (rb_notify_fd_close): do not enqueue multiple interrupts
  [ruby-core:81581] [Bug #13632]

* test/ruby/test_io.rb (test_single_exception_on_close):
  new test based on script from Nikolay


git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/branches/ruby_2_3@59274 b2dd03c8-39d4-4d8f-98ff-823fe69b080e