On 2/10/17 12:52 PM, Victoria Martinez de la Cruz wrote: > I'm reviewing it as well. At least trying to compile it on my own env. > Thanks. Based upon my own testing, I've pushed a new version today. Changes are to the second patch in the series, not your request.
For some odd reason, whenever the channel id was changed, the original code removed the xprt from the idle scan list -- but only for changing from a channel that wasn't the global default. It was obscure. It makes no sense. There isn't any code anywhere else in ntirpc or ganesha to re-add the xprt to the list. My code exposed this by eliminating the global default test. I've removed the senseless removal. Now, it is only removed during destroying the xprt (which was already in the original code). Also, because I've changed the refcounting from locking to atomic, the logging could crash during shutdown. No lock to prevent the xprt from disappearing. So I've moved the release after the logging (and put in a comment block explaining why). That will only apply to folks obsessively doing full debug logging. I'm feeling pretty confident about this was14oldbugs patch series, and think it should be pushed along with Swen's authgss_prot patch. I'll bring it up on the call today. ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ Nfs-ganesha-devel mailing list Nfs-ganesha-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs-ganesha-devel
