Strange behaviour of the profiling timer

Hello. I've sent a problem report recently on a strange behaviour of the profiling timer in FreeBSD 10.0, but there is still no response for about a week.
http://www.freebsd.org/cgi/query-pr.cgi?pr=187668

After this commit: https://github.com/freebsd/freebsd/comm ... 00faef91da setitimer does not work when the timer interval is set low (below the value of 1/kern.hz, I think). You can ensure that it is so by running a testing program I've attached.

I'm writing here, because problem reports 187669 and 187667 are already closed and my issue isn't even discussed. I do not mean to intrude, I just want to know if there is anything wrong with my PR. As I already said SBCL (maybe the most popular implementation of Common Lisp) cannot pass its tests because of it and therefore it cannot be installed from ports (as far as I can remember). Maybe there are more programs which can suffer from this behaviour of setitimer.

Thanks anyway for such a good OS. Vasily
 
Problem reports (PRs) are not processed in the order submitted, so the fact that PRs submitted after yours have been closed is irrelevant. If you look closely you'll see that PRs 187667 and 187669 were both related to security ports, and involved the routine upgrade of those ports in FreeBSD from their upstream source. These ports have the same maintainer. Port maintainers are separate from the developers who would most likely be assigned to look at your PR based on its severity and other open PRs at the same or higher severity level.

If you want to follow-up on your PR, you could post a message to the freebsd-stable@freebsd.org mailing list and inquire nicely. :)
 
Back
Top