On 05/22/2013 11:47 AM, kunming yang wrote: > Thanks!!! > > There is still many problems. Why does the kernel hang after several > minutes? Every time, it takes almost the same time to "die". What might > cause that? Thanks for help!
The usual cause is if your hardware timer is of the "free-funning counter and match register" type, if you are programming the match register for a date in the past. In that case, the kernel will unblock when the free-running counter wraps. Another reason is if your tsc is not a 64 bits tsc, if you are not refreshing the tsc in kernel-space often enough (by calling __ipipe_tsc_update()). But there are many causes of lockups, the first reflex should be to add all the debugging issues. If the lockup is not replaced with an error message, please run cat /proc/interrupts and verify that all the interrupt controllers which appear in the dump have been checked as explained here: http://www.xenomai.org/index.php/I-pipe-core:ArmPorting#Interrupt_controller The interrupt controller name in the dump (third column if you only have one cpu), should help you find the code associated to this interrupt controller. -- Gilles. _______________________________________________ Xenomai mailing list [email protected] http://www.xenomai.org/mailman/listinfo/xenomai
