Merge pull request #884 from ruby-concurrency/thread-local

Do not iterate over hash which might conflict with new pair addition

lib/concurrent-ruby/concurrent/atomic/ruby_thread_local_var.rb

@@ -97,7 +97,10 @@ def self.thread_local_finalizer(index)
           # The cost of GC'ing a TLV is linear in the number of threads using TLVs
           # But that is natural! More threads means more storage is used per TLV
           # So naturally more CPU time is required to free more storage
-          THREAD_LOCAL_ARRAYS.each_value { |array| array[index] = nil }
+          #
+          # DO NOT use each_value which might conflict with new pair assignment
+          # into the hash in #value= method
+          THREAD_LOCAL_ARRAYS.values.each { |array| array[index] = nil }
           # free index has to be published after the arrays are cleared
           FREE.push(index)
         end