Hey Guys, I found the following issue with mrxvt: When the mrxvt window is completely obscured (e.g. completely behind a different window), then mrxvt starts "polling". This causes numerous interrupts, and CPU wake-ups per second (about 100 on my system), which is not desirable, especially on Laptops.
It took me quite some time to hunt this bug down, and I'm hoping that it
is fixed in the latest release in subversion fixes this. I'm pretty sure
that the problem is now solved, however I'd still appreciate some
testing on other WM's / systems. To test, you have to do one of the
following:
1. Run on a recent kernel compiled with CONFIG_TIMER_STATS=y.
2. Run powertop.
3. Run mrxvt, and leave some idle process in it (e.g. shell).
4. Cover the window / switch to other desktops / minimize / etc.
(Just perform whatever window operation you feel like to try and
"fool" the above patch).
5. Examine the powertop output, after your system is idle again. If
mrxvt is taking a large number of interrupts per second, then
please report it.
Just for comparison, if you want to see what "large" is, then completely
obscure, but not minimize, an mrxvt window from r292 or earlier. On my
system I get some 100 wakeups per second due to mrxvt. (You can see
exactly which instance of mrxvt is causing these by running "cat
/proc/timer_stats).
GI
--
100 THINGS I'D DO IF I EVER BECAME AN EVIL OVERLORD
37. If my trusted lieutenant tells me my Legions of Terror are losing a
battle, I will believe him. After all, he's my trusted lieutenant.
pgpxAdrDoWEIc.pgp
Description: PGP signature
------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php
_______________________________________________ Materm-usage mailing list Materm-usage@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/materm-usage Mrxvt home page: http://materm.sourceforge.net
