Boost C++ Libraries: Ticket #3837: thread_specific_ptr returns incorrect values if another one was constructed at the same address earlier

anonymous — Mon, 15 Feb 2010 23:36:53 GMT

More serious problems may arise if a thread_specific_ptr<X> and a thread_specific_ptr<Y> occupy the same address at different times.

anonymous — Fri, 11 Jun 2010 07:51:48 GMT

Why the assertion should not fail? You are using a global variable optr ion the two threads. The following interleaving seems possible

optr=in_place(); optr->reset(new int); main optr=none; optr=in_place(); other_thread assert(optr->get() == 0); main succeed

As the global variable has been re-assigned with in_place(). Am I missing something?

anonymous — Sat, 19 Jun 2010 19:53:03 GMT

the assertion should not fail because the global ts-ptr has been reconstructed, so it's ptr should be 0 in each thread. but it does fail, try it.

the current ts-ptr cant handle different ts-ptrs at the same address at different points in time.

status changed

Anthony Williams — Thu, 28 Oct 2010 10:30:43 GMT

status new → assigned

keywords set

viboes — Sun, 11 Dec 2011 16:40:31 GMT

keywords tss added

milestone changed

viboes — Mon, 28 May 2012 16:45:04 GMT

milestone Boost 1.42.0 → To Be Determined

viboes — Sat, 18 Aug 2012 17:38:33 GMT

If I understand correctly, you are expecting that the destructor of the thread_specific_ptr must reset the thread specific pointer for all the threads that are using this key.

IMO, the user must ensure that the thread_specific_pointer instance outlives all the threads that are using it. Otherwise, the library needs to maintain a list of all the threads that are using a specific key, which goes against the purpose of tss to reduce the contention between threads by accessing a thread specific and not a global data.

The documentation must state clearly this constraint (design principle). Note that pthread works already in this way.

owner, status changed; milestone deleted

viboes — Sun, 19 Aug 2012 15:29:31 GMT

owner changed from Anthony Williams to viboes
status assigned → new
milestone To Be Determined

status changed

viboes — Tue, 21 Aug 2012 05:39:17 GMT

status new → assigned

status changed; resolution set

viboes — Sat, 25 Aug 2012 12:31:30 GMT

status assigned → closed
resolution → invalid

Doc updated accordingly. Committed in trunk at revision [80198].