> But I cant reproduce it atm...

OK, I need a break and get some sleep (but I realize 2.6 release is
is very soon).

I can reproduce [Bug #15383] with "rb_mutex_t.fork_gen" field
removed, but fork_gen shouldn't be necessary in mutex because of
rb_mutex_abandon_* functions and rb_mutex_cleanup_keeping_mutexes
already call list_head_init(waitq).

So there is a bug in one of the rb_mutex_{abandon,cleanup}*
functions which is hidden by the addition of rb_mutex_t.fork_gen.

Unsubscribe: <mailto:ruby-core-request / ruby-lang.org?subject=unsubscribe>
<http://lists.ruby-lang.org/cgi-bin/mailman/options/ruby-core>