On Mon, May 13, 2013 at 12:26:05PM -0700, Stephen Boyd wrote: > On an SMP system with only one global clockevent and a dummy > clockevent per CPU we run into problems. We want the dummy > clockevents to be registered as the per CPU tick devices, but > we can only achieve that if we register the dummy clockevents > before the global clockevent or if we artificially inflate the > rating of the dummy clockevents to be higher than the rating > of the global clockevent. Failure to do so leads to boot > hangs when the dummy timers are registered on all other CPUs > besides the CPU that accepted the global clockevent as its tick > device and there is no broadcast timer to poke the dummy > devices. > > If we're registering multiple clockevents and one clockevent is > global and the other is local to a particular CPU we should > choose to use the local clockevent regardless of the rating of > the device. This way, if the clockevent is a dummy it will take > the tick device duty as long as there isn't a higher rated tick > device and any global clockevent will be bumped out into > broadcast mode, fixing the problem described above. > > Reported-by: Mark Rutland <mark.rutland@xxxxxxx> > Tested-by: Mark Rutland <mark.rutland@xxxxxxx> > Cc: John Stultz <john.stultz@xxxxxxxxxx> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> > Signed-off-by: Stephen Boyd <sboyd@xxxxxxxxxxxxxx> Tested-by: Sören Brinkmann <soren.brinkmann@xxxxxxxxxx> I tested this on Zynq. Zynq uses the ARM twd timer as clockevent devices when available. When I remove the twd nodes from DT the clockevent device is a timer common to both CPUs. Without this patch this seems to stall the second CPU resulting in the system to hang sooner or later. [ 65.360000] INFO: rcu_preempt detected stalls on CPUs/tasks: [ 65.360000] 1: (1 GPs behind) idle=56e/0/0 softirq=0/0 [ 65.360000] (detected by 0, t=6502 jiffies, g=4294966997, c=4294966996, q=6844) [ 65.360000] Task dump for CPU 1: [ 65.360000] swapper/1 R running 0 0 1 0x00000000 With this patch everything looks fine. Sören -- To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
