[zeromq-dev] Race conditions again

Pieter Hintjens ph at imatix.com
Tue Jan 7 06:31:19 CET 2014


I've never used thread sanitizer. Two things strike me. First, if
there are *real* race conditions, you can help by tracking these down
and working with us to fix them. Secondly, if these are false
positives, there must be some "ignore" file, as we use for valgrind.

As for race conditions in your own code, if you do not use shared
mutable state, and only pass data between threads via messages, there
is no opportunity for race conditions in your own code.

-Pieter



On Tue, Jan 7, 2014 at 2:24 AM, Michi Henning <michi at triodia.com> wrote:
> Hi folks,
>
> I've been using zmq (3.2.3) with good success, but I'm getting a lot of noise from
> clang's thread sanitizer about race conditions.
>
> I've compiled both 3.2.3 and 4.0.3 with clang and thread sanitizer enabled.
> Running the test suite generates hundreds of reports of race conditions.
> It appears that many of these are harmless (at least, I haven't seen any crashes).
> But getting all this noise from thread sanitizer makes it very difficult to use the
> tool because any race conditions that I introduce in my own code are very hard
> to spot in all the noise. (Thread sanitizer does have a suppression mechanism,
> but it's quite coarse, nowhere near as fine-grained as, say, valgrind's.)
>
> In my experience, problems reported by thread sanitizer are generally worth
> paying attention to. (There are few false positives.)
>
> Is there any chance of getting a zmq release that runs clean with thread sanitizer?
>
> Thanks,
>
> Michi.
> _______________________________________________
> zeromq-dev mailing list
> zeromq-dev at lists.zeromq.org
> http://lists.zeromq.org/mailman/listinfo/zeromq-dev



More information about the zeromq-dev mailing list