You are not logged in.
Hi,
I have hundreds of those messages here in the logfiles:
Apr 24 01:37:36 greg BUG: soft lockup - CPU#0 stuck for 61s! [swapper:0]
Apr 24 01:37:36 greg CPU 0:
Apr 24 01:37:36 greg Modules linked in: radeon drm cpufreq_ondemand nls_cp437 vfat fat joydev snd_seq_oss snd_seq_midi_event snd_seq firewire_ohci firewire_core snd_seq_device pcmcia crc_itu_t irda snd_atiixp sdhci snd_pcm_oss snd_atiixp_modem crc_ccitt psmouse tifm_7xx1 snd_mixer_oss mmc_core snd_ac97_codec ohci1394 pcspkr serio_raw i2c_piix4 tifm_core yenta_socket rsrc_nonstatic ac97_bus snd_pcm k8temp snd_timer snd ieee1394 pcmcia_core soundcore snd_page_alloc usbhid i2c_core hid ff_memless video output sg acer_wmi wmi led_class thermal evdev fan button battery ac powernow_k8 freq_table processor bcm43xx ieee80211softmac ieee80211 ieee80211_crypt r8169 rtc_cmos rtc_core rtc_lib ext3 jbd mbcache sr_mod ehci_hcd cdrom sd_mod ohci_hcd pata_acpi usbcore ata_generic pata_atiixp libata scsi_mod dock
Apr 24 01:37:36 greg Pid: 0, comm: swapper Not tainted 2.6.25-ARCH #1
Apr 24 01:37:36 greg RIP: 0010:[<ffffffff8045f3c7>] [<ffffffff8045f3c7>] _spin_unlock_irqrestore+0x7/0x40
Apr 24 01:37:36 greg RSP: 0018:ffffffff8055fe50 EFLAGS: 00000296
Apr 24 01:37:36 greg RAX: 0000000000000000 RBX: 00000000ffffffff RCX: 0000000000000020
Apr 24 01:37:36 greg RDX: 00000000000004b0 RSI: 0000000000000296 RDI: 0000000000000296
Apr 24 01:37:36 greg RBP: 00000001c7c4f1be R08: ffff8100010046c0 R09: 00000000022dd66f
Apr 24 01:37:36 greg R10: 0000000000000005 R11: ffffffff80220600 R12: ffffffff802558d2
Apr 24 01:37:36 greg R13: ffffffff8055fde8 R14: 0000000000bc74ba R15: ffffffff8055fdd8
Apr 24 01:37:36 greg FS: 00007f5c9b6d96f0(0000) GS:ffffffff80549000(0000) knlGS:0000000000000000
Apr 24 01:37:36 greg CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
Apr 24 01:37:36 greg CR2: 00007f1e1fcc7000 CR3: 000000003e9ca000 CR4: 00000000000006e0
Apr 24 01:37:36 greg DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Apr 24 01:37:36 greg DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Apr 24 01:37:36 greg
Apr 24 01:37:36 greg Call Trace:
Apr 24 01:37:36 greg [<ffffffff8025ae45>] ? tick_notify+0x2e5/0x430
Apr 24 01:37:36 greg [<ffffffff802562c7>] ? notifier_call_chain+0x37/0x70
Apr 24 01:37:36 greg [<ffffffff8025a7bb>] ? clockevents_notify+0x2b/0x90
Apr 24 01:37:36 greg [<ffffffff8812f57d>] ? :processor:acpi_idle_enter_simple+0x201/0x259
Apr 24 01:37:36 greg [<ffffffff803b9220>] ? cpuidle_idle_call+0x0/0x130
Apr 24 01:37:36 greg [<ffffffff803b92ff>] ? cpuidle_idle_call+0xdf/0x130
Apr 24 01:37:36 greg [<ffffffff8020b570>] ? default_idle+0x0/0x70
Apr 24 01:37:36 greg [<ffffffff803b9220>] ? cpuidle_idle_call+0x0/0x130
Apr 24 01:37:36 greg [<ffffffff8020b570>] ? default_idle+0x0/0x70
Apr 24 01:37:36 greg [<ffffffff8020b4e0>] ? cpu_idle+0x90/0x120
Apr 24 01:37:36 greg [<ffffffff80212900>] ? pda_init+0x30/0xb0They all look the same, concerning the call trace, and the instruction pointer always points at _spin_unlock_irqrestore. These errors appear sporadic, it could be every few minutes up to every few hours. And when they come up there are dozens of them at the same second.
The system becomes very unresponsive then, sometimes it catches itself, but every once a while only the magic sysrq key helps to get a clean reboot.
The kernel is 2.6.25-ARCH #1 SMP PREEMPT from testing, but I had the same problems with 2.6.24, too. Older kernels have worked so far, iirc.
The hardware is a 64 bit system with AMD Turion ML-34 processor. Graphics card is an ATI x700 mobility. I first suspected the catalyst driver, but a removal doesn't show any difference, as can be seen on the "Not tainted" note above.
After some websearching I found the hint to pass hpet=disable as kernel parameter, but this didn't make any difference either.
There were some similar posts here in the forum, but they all had a different call trace, so I decided to open a new thread.
Any ideas what could be the cause for this?
Regards,
Marc
Offline