diff options
author | ko1 <ko1@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2017-08-02 00:50:42 +0000 |
---|---|---|
committer | ko1 <ko1@b2dd03c8-39d4-4d8f-98ff-823fe69b080e> | 2017-08-02 00:50:42 +0000 |
commit | 92f7813ae27ca7037f9ab6e5744675c01b3fa574 (patch) | |
tree | 9c4b43d074f0ac3382e78c67628acbfd345ffd40 /thread.c | |
parent | b7f205d54cd807419256ff829b416a8c2204aeec (diff) | |
download | ruby-92f7813ae27ca7037f9ab6e5744675c01b3fa574.tar.gz |
release VM stack properly.
* cont.c: r55766 change the handling method of Fiber's VM stack.
Resumed Fiber points NULL as VM stack and running Thread has
responsibility to manage it (marking and releasing).
However, thread_start_func_2()@thread.c and thread_free()@vm.c
doesn't free the VM stack if corresponding root Fiber is exist.
This causes memory leak. [Bug #13772]
* cont.c (root_fiber_alloc): fib->cont.saved_thread.ec.stack should be NULL
because running thread has responsibility to manage this stack.
* vm.c (rb_thread_recycle_stack_release): assert given stack is not NULL
(callers should care it).
git-svn-id: svn+ssh://ci.ruby-lang.org/ruby/trunk@59462 b2dd03c8-39d4-4d8f-98ff-823fe69b080e
Diffstat (limited to 'thread.c')
-rw-r--r-- | thread.c | 6 |
1 files changed, 2 insertions, 4 deletions
@@ -694,10 +694,8 @@ thread_start_func_2(rb_thread_t *th, VALUE *stack_start, VALUE *register_stack_s rb_threadptr_unlock_all_locking_mutexes(th); rb_check_deadlock(th->vm); - if (!th->root_fiber) { - rb_thread_recycle_stack_release(th->ec.stack); - th->ec.stack = 0; - } + rb_thread_recycle_stack_release(th->ec.stack); + th->ec.stack = NULL; } native_mutex_lock(&th->vm->thread_destruct_lock); /* make sure vm->running_thread never point me after this point.*/ |