Hello, we use the kernel 3.2.16-rt27 on our powerpc board (MPC5200). At our load tests we got a kernel Oops message: [ 72.958944] sched: RT throttling activated [ 216.604657] BUG: scheduling while atomic: codesyscontrol/1165/0x00010001 [ 216.604675] Modules linked in: [ 216.604691] NIP: 101ceb6c LR: 484026e4 CTR: 101b83a4 [ 216.604711] REGS: c6a01f50 TRAP: 0901 Not tainted (3.2.16-rt27) [ 216.604724] MSR: 0000f932 <EE,PR,FP,ME,IR,DR> CR: 84002484 XER: 00000000 [ 216.604764] TASK = c6e20130[1165] 'codesyscontrol' THREAD: c6a00000 [ 216.604778] GPR00: 4826ab84 4826ab5c 48272920 4826ab84 0efe95d8 0efe95e8 0efe95f0 0efb9858 [ 216.604812] GPR08: 0efe95e0 4826ab84 000000f4 4826ab64 84002484 [ 216.604847] NIP [101ceb6c] 0x101ceb6c [ 216.604862] LR [484026e4] 0x484026e4 [ 216.604872] Call Trace: [ 216.604965] ------------[ cut here ]------------ [ 216.604990] Kernel BUG at c004f978 [verbose debug info unavailable] [ 216.605014] Oops: Exception in kernel mode, sig: 5 [#1] [ 216.605025] PREEMPT mpc5200-simple-platform [ 216.605039] Modules linked in: [ 216.605054] NIP: c004f978 LR: c004f970 CTR: c001db40 [ 216.605072] REGS: c6a01d40 TRAP: 0700 Not tainted (3.2.16-rt27) [ 216.605085] MSR: 00029032 <EE,ME,CE,IR,DR> CR: 24002424 XER: 00000000 [ 216.605120] TASK = c6e20130[1165] 'codesyscontrol' THREAD: c6a00000 [ 216.605134] GPR00: 00000001 c6a01df0 c6e20130 00000000 c74411f0 00000000 00000000 00000000 [ 216.605166] GPR08: c6e20130 0000005a 00000000 c0495c08 24002482 10279110 00000000 c4653600 [ 216.605199] GPR16: 7fffffff ffffffff 7fffffff c04951d8 c6a00000 c0494900 000000d8 23c36491 [ 216.605233] GPR24: c005eda0 ffffffff 3b9ac9ff 00000001 c6a01f50 00000000 00000001 c6e20130 [ 216.605295] NIP [c004f978] run_posix_cpu_timers+0x20/0x104 [ 216.605316] LR [c004f970] run_posix_cpu_timers+0x18/0x104 [ 216.605328] Call Trace: [ 216.605349] [c6a01df0] [c004f970] run_posix_cpu_timers+0x18/0x104 (unreliable) [ 216.605385] [c6a01e00] [c0039a28] update_process_times+0x48/0x5c [ 216.605424] [c6a01e20] [c005ee00] tick_sched_timer+0x60/0xd4 [ 216.605452] [c6a01e50] [c00507ec] __run_hrtimer+0x9c/0x2a8 [ 216.605480] [c6a01e80] [c0051a74] hrtimer_interrupt+0x158/0x4ac [ 216.605509] [c6a01f20] [c000c424] timer_interrupt+0x17c/0x2a8 [ 216.605544] [c6a01f40] [c00117e8] ret_from_except+0x0/0x14 [ 216.605560] Instruction dump: [ 216.605573] 419effec 38000000 90020000 4bffffac 9421fff0 7c0802a6 93e1000c 7c7f1b78 [ 216.605606] 90010014 481ab93d 7c0000a6 54008ffe <0f000000> 3d60c049 800b4894 392b4894 [ 216.826417] Kernel panic - not syncing: Fatal exception in interrupt The same in kernel 3.2.15-rt26... The problem occurs in overload situations only. So previously we also got the message that rt throttling is getting enabled. The problem seems to be dependent on the realtime patch because with a untouched kernel 3.2.16 we cant reproduce it. To locate the problem, additionally we try to disable switching to "throttling mode" by uncommenting "rt_rq->rt_throttled = 1;" in sched_rt.c but it still occurs. If the kernel is configured to be tickless the problem occurs less often. Any hints? any ideas? Thanks Harald Hannover Messe 2012 ? Berghof stellt aus : Industrial Automation Halle 9 / Stand D62 und MobiliTec Halle 25 / Stand K03/1 Neugierig auf: "Web- und Target-Visualisierungen ? Schneller als CoDeSys erlaubt" oder "Test- und Konditioniersysteme"? Infos und Ihr kostenloses persönliches Messeticket unter: www.berghof.com/HMI ______________________________________________________________________ This email has been scanned by the Symantec Email Security.cloud service. For more information please visit http://www.symanteccloud.com ______________________________________________________________________ -- 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
