Dear list members,
I asked Mikhail Belous for a quick status update regarding the multithread
feature for the tor-daemon.
Here's what he said:
Good news:
My version of multi-thread tor passes the test and doesn't shows mem racing
problems.
Bad news:
1)Processing channel is only big enough operation to be paralysed. Tor
daemon uses a lot of global variables that should be locked. I have already
done this for some resources.
2) Multithread branch is behind master for over 1000 commits.
What should be done before release:
1) Merge master in better_workqueue branch.
2) Find all global resources, that are available from
command_proccess_cells. I already sent you a list, but it should be checked.
3) Lock all global resources.
4) Code review
5) Compare performance of one thread and multi-thread daemons.
--
Mit freundlichen Grüssen / Sincerely yours
Sebastian Urbach
-----------------------------------------
Definition of TOR:
10% luck, 20% skill, 15% concentrated
power of will, 5% pleasure, 50% pain and
100% reason to remember the name!
-----------------------------------------