On Fri, 20 Feb 2009, Paul Gortmaker wrote: > On Thu, Feb 19, 2009 at 7:11 AM, Thomas Pfaff <tpfaff@xxxxxxx> wrote: > > While testing the 2.6.26.8-rt16 kernel on an AT91SAM9260-EK Arm9 board i noticed > > a small regression to 2.6.24.7-rt27. > > > > The kernel compiles and boots fine but the average load when the cpu is idle > > tends to be 1, it starts at 0 when the system boots and goes to 1 or above after > > some time. > > > > All kernels have the at91 patches from http://maxim.org.za/sam9.html > > > > Following kernels are ok: > > > > 2.6.24.7-rt27 > > 2.6.26.8 without rt > > > > With the rt patch i have tested several combinations of the PREEMPT options: > > > > Without CONFIG_PREEMPT_HARDIRQ i get an oops on boot (see next mail). > > > > All other CONFIG_PREEMPT_[NONE|VOLUNTARY|DESKTOP|RT] options are working but > > with the wrong average load. > > Please test the following patch. I think you should find it fixes the problem. Thanks, I've queued this up for 26-rt as well as added it to my rt git tree. -- Steve -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
