On Thu, Aug 6, 2009 at 9:03 AM, Iván Briano (Sachiel)<[email protected]> wrote: > On Thu, Aug 6, 2009 at 7:56 AM, Fernando Meira<[email protected]> wrote: >> Hi, >> >> about a month ago, or so, I've noticed (probably after some updates) that >> the battery module was using 100% of 1 cpu. >> The solution was to unload the module. >> Has anyone seen this too? Or has ideas of what can it be? Or better, how to >> solve it? >> > > Something changed in kernel 2.6.30 that screws up polling. More info on this > can be found somewhere in the archive for either this list or e-devel, I don't > remember. > Apparently, using hal as the backend for the module works just fine, > but I haven't > really tested.
yes, hal works fine. Someone (not me!) should either ask LKML, some friend kernel hacker or even look at HAL on how to do this. What E does is believe /sys files will block on read if data did not change, but instead 2.6.30 changed to READ always work but no data there, so we end reading nothing all the time and cpu gets abused. raster already said that changing this to period polling is not the correct solution and I agree, but a definitive answer should exist to "how to listen to uevent and /sys" properties changes. as Ivan pointed, in the list history/archive you should even find reference to exact kernel commit that broke, you could mail the author and ask it... we're quite busy these days :-/ -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -------------------------------------- MSN: [email protected] Skype: gsbarbieri Mobile: +55 (19) 9225-2202 ------------------------------------------------------------------------------ Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day trial. Simplify your report design, integration and deployment - and focus on what you do best, core application coding. Discover what's new with Crystal Reports now. http://p.sf.net/sfu/bobj-july _______________________________________________ enlightenment-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/enlightenment-users
