[Bug nptl/13065] Race condition in pthread barriers

["triegel at redhat dot com" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=13065

Torvald Riegel <triegel at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |triegel at redhat dot com

--- Comment #2 from Torvald Riegel <triegel at redhat dot com> ---
This is conceptually related to Bug 13690, whose resolution depends on the
outcome of a POSIX request for clarification.  The same kind of wording that
needs to be clarified for that bug is not present in the barrier specification,
but it's essentially the same question of when POSIX synchronization objects
can be safely destroyed.  Therefore, I think it's good to wait for a result of
the clarification request.

Furthermore, destruction is somewhat more complex due to the standard leaving
it unspecified whether a thread that processes a signal will keep other waiters
block.  That is, if at least one of the other threads blocked on the barrier
may process signals, then one cannot destruct a barrier after returning from a
call to pthread_barrier_wait().

-- 
You are receiving this mail because:
You are on the CC list for the bug.