On Sun, 2010-05-30 at 17:04 +0200, Marc Weber wrote: > Hi, > > Some weeks ago I got a new laptop running core -i5 processsor. > Everything seems to be working. However after resuming from suspend to > ram htop only shows one core having 100% utilization and I can't seem to > fix this putting cores off and online again using the sys cpuX/online > interface. Weird. pm-utils does not do anything special with the cpus across suspend/resume -- the kernel handles all that. I have gone ahead and cc'ed the linux-pm mailing list. > I've tried putting them offline before suspending as well. > > My test case: > > Makefile: > > a4: a1.x a2.x a3.x a4.x > all: a1.x a2.x a3.x a4.x a5.x a6.x a7.x a8.x a9.x a10.x a11.x b1.x b2.x b3.x b4.x > %.x: > ./hog > > hog is a just a c prog running a dummy loop which lasts about 4 secs. > > I made a screenshot of htop: > http://mawercer.de/~marc/htop.jpg > Make started 4 processes, however only one is running on a core. > > make -j4 doesn't finish faster than make -j1 either. > > Do you have any idea how to start debugging this issue? > > Of course everything is fine after rebooting. > > My kernel: > > Linux nixos 2.6.32.14 #1 SMP Sat May 29 22:06:43 UTC 2010 x86_64 GNU/Linux > > using pm-utils-1.3.0 > > Any ideas or hints are appreciated. > > Marc Weber > _______________________________________________ > Pm-utils mailing list > Pm-utils@xxxxxxxxxxxxxxxxxxxxx > http://lists.freedesktop.org/mailman/listinfo/pm-utils -- Victor Lowther LPIC2 UCP RHCE _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
