https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=195262

Xin LI <[email protected]> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|New                         |Open

--- Comment #7 from Xin LI <[email protected]> ---
(In reply to John Baldwin from comment #6)
The code in question have received a major refactor and the issue no longer
exists (in r262496: mutex is no longer held in random_kthread before sleeping).

In the new world order, the spin lock is only acquired as a leaf mutex in
random_harvestq_internal() and that would avoided all LOR cases (I would
probably move get_cyclecount() out from the locked scope but that's a different
issue).

So I think the _right_ fix is probably just merge the refactor back to
stable/10 instead.

-- 
You are receiving this mail because:
You are the assignee for the bug.
_______________________________________________
[email protected] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-bugs
To unsubscribe, send any mail to "[email protected]"

Reply via email to