Thanks for the feedback. > Could that not lead into situations where chronyd randomly fails later after > start when it needs to allocate more memory?
With MCL_FUTURE specified, yes it could fail. The thought was to just abide by the previously set system limits (assuming that is why it failed), but that would cause more problems than its worth. I'll revert the behavior. With Solaris, since it does not have this artificial limit, we should be safe to use mlockall by itself (according to my understanding of the docs). >> + add_def POSIX > I don't like this definition very much. I didn't either, but I wasn't sure what the preferred style would be. I'll change the conditional calls to check for supported platforms individually. > it's a good opportunity to update the coding style to match the majority of > the code I'll do what I can, but please fill in the gaps on the final patch. > As for sending the patches, have you tried git send-email? That should avoid > problems with formatting. Maybe one day I'll learn how to copy and paste patches out of a terminal and into gmail. In the mean time, I'll try your suggestion. Regards, Stefan -- To unsubscribe email chrony-dev-requ...@chrony.tuxfamily.org with "unsubscribe" in the subject. For help email chrony-dev-requ...@chrony.tuxfamily.org with "help" in the subject. Trouble? Email listmas...@chrony.tuxfamily.org.
