On Dec 8, 2007 6:12 PM, Parag Warudkar <parag.warudkar@gmail.com> wrote:Looks like CPU_IDLE=y is necessary for the problem to show up. With CPU_IDLE=n HRT+NO_HZ+TICK_ONESHOT does not give soft lockup problems. (Actually with HIGH_RES_TIMERS=NO_HZ=TICK_ONESHOT=y I do see short freezes on ssh - when I cannot type anything for may be a second even under 100% idle. But Soft Lock up doesnt show up in dmesg with this configuration.) No Soft Lockup - CPU_IDLE=n 5+ Hours uptime --------------------------------------------------------------------------- HIGH_RES_TIMERS=NO_HZ=TICK_ONESHOT=n HIGH_RES_TIMERS=y NO_HZ=TICK_ONESHOT=n HIGH_RES_TIMERS=NO_HZ=y TICK_ONESHOT=n HIGH_RES_TIMERS=NO_HZ=TICK_ONESHOT=y - short freezes. Soft Lockup - CPU_IDLE=y 2 hours uptime -------------------------------------------------------------------------- HIGH_RES_TIMERS=NO_HZ=TICK_ONESHOT=y dmesg - BUG: soft lockup - CPU#1 stuck for 11s! [swapper:0] Pid: 0, comm: swapper Not tainted (2.6.24-rc4 #13) EIP: 0060:[<c0603eaa>] EFLAGS: 00000202 CPU: 1 EIP is at _spin_lock_irqsave+0x16/0x27 EAX: c06b4110 EBX: 00000001 ECX: f7873808 EDX: 00000293 ESI: 00000005 EDI: f7873808 EBP: 00000000 ESP: f7829f10 DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068 CR0: 8005003b CR2: 800f5958 CR3: 37284000 CR4: 000006d0 DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000 DR6: ffff0ff0 DR7: 00000400 [<c0438293>] tick_broadcast_oneshot_control+0x10/0xda [<c0437ce2>] tick_notify+0x1d4/0x2eb [<c04281bc>] get_next_timer_interrupt+0x143/0x1b4 [<c06058a1>] notifier_call_chain+0x2a/0x47 [<c04345c0>] raw_notifier_call_chain+0x17/0x1a [<c043781e>] clockevents_notify+0x19/0x4f [<c0533d23>] acpi_idle_enter_simple+0x183/0x1d0 [<c058cf03>] cpuidle_idle_call+0x53/0x78 [<c058ceb0>] cpuidle_idle_call+0x0/0x78 [<c0402575>] cpu_idle+0x97/0xb8 ======================= Parag --
| Tarkan Erimer | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Andrew Morton | -mm merge plans for 2.6.23 |
| James Bottomley | [Ksummit-2008-discuss] Fixing the Kernel Janitors project |
| Greg KH | [GIT PATCH] driver core patches against 2.6.24 |
git: | |
| Gerrit Renker | [PATCH 18/37] dccp: Support for Mandatory options |
| David Miller | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| David Miller | Re: [GIT]: Networking |
| Tantilov, Emil S | WARNING: at include/net/sock.h:417 udp_lib_unhash |
