4 ms·
Linux 3.15 and continuation of merge window
- SDGT 12y agoHere's the LKML listing, because that little asp classic site seems to be having some slowness, and they didn't really include much more than a small summary, which is about as big as the whole lkml article anyway. https://lkml.org/lkml/2014/6/8/70 https://lkml.org/lkml/2014/6/8/70
- mattwad 12y agoworse, it redirected me to VistaPrint on some sort of auto-timeout :(
- SDGT 12y agoYep that happened to me as well after spinning on the page for 30+seconds.
- dang 12y agoThanks; changed from [1]. HN strongly prefers original sources. 1. http://www.efytimes.com/e1/fullnews.asp?edid=140733 http://www.efytimes.com/e1/fullnews.asp?edid=140733
- rwmj 12y agoWith file locks that aren't broken by design! https://lwn.net/Articles/586904/ https://lwn.net/Articles/586904/
- co2 12y agoLinux 3.15 will also include a better renameat2 function with parameters that allow to write better pid file placing and handling without race conditions. http://man7.org/linux/man-pages/man2/renameat2.2.html http://man7.org/linux/man-pages/man2/renameat2.2.html
- quotemstr 12y agoOf course, it'd be nice if we could solve PID races entirely. The POSIX system is unbelievably broken compared to how NT does it. (No, "PID wraparound happens slow enough not to matter" is not a good foundation for a robust system.)
- lomnakkus 12y agoDoesn't e.g. systemd solve this problem? (PID files are just a bad idea.)