http://lwn.net/Articles/223185/ This is a special feed containing comments posted to the individual LWN article titled "Clockevents and dyntick". hourly 2 http://lwn.net/Articles/231999/rss 2007-04-26T14:36:17+00:00 gyrovague I haven't seen benchmarks on this in particular, but I remember reading a piece a while ago where a guy claimed that his laptop battery lasts TWICE as long with a 100Hz tick than with a 1000Hz tick... So, I'm expecting great things...<br> http://lwn.net/Articles/224229/rss 2007-03-01T13:56:14+00:00 jmranger Has anyone seen any benchmarks on this ? I would expect no impact on performance, but I'm wondering how significant the power savings could be.<br> http://lwn.net/Articles/223641/rss 2007-02-24T06:14:07+00:00 jzbiciak IBM was pretty big on the tickless kernel idea quite awhile ago, probably due to their emphasis on hypervisors and virtualization. Separately, the ARM branch has been big on it for the embedded space.<br> <p> It looks like the major difference is that clockevents gives an abstraction layer for the clock. Once all the arches are synced up to that, I wonder if dyntick becomes the default?<br> <p> http://lwn.net/Articles/223367/rss 2007-02-22T11:12:42+00:00 arnd Just as a historic side note, support for disabling the timer tick during <br> idle was first merged three years ago for the s390 architecture, see <br> <a href="http://lwn.net/Articles/82378/">http://lwn.net/Articles/82378/</a>. The first lwn coverage I could find on the <br> topic is <a href="http://lwn.net/2001/0412/bigpage.php3#kernel">http://lwn.net/2001/0412/bigpage.php3#kernel</a>, another three years <br> earlier, with a patch against the 2.4.3 kernel.<br>