You are not logged in.

#1 2010-03-22 19:02:26

pkerwien
Member
From: Sweden
Registered: 2009-07-06
Posts: 14

Server goes to sleep - related to hpet problem?

I recently changed my motherboard to a Gigabyte GA-790FXTA-UD5. First I used kernel-2.6.32.10 (amd64) and experienced lockup / freezes / sleep mode described in:

https://bugs.launchpad.net/ubuntu/+sour … bug/270798

I then upgraded to 2.6.33.1 (built via ABS) and the problem remains. Do any Arch users out there experience the same problem? If yes, how did you solve it?

Last edited by pkerwien (2010-03-22 19:04:45)


Linux is just like an indian tent: no Gates, no Windows and an Apache inside.

Offline

#2 2010-03-23 05:10:51

brenix
Member
From: California
Registered: 2008-03-05
Posts: 185

Re: Server goes to sleep - related to hpet problem?

Hmm.. Nice mobo btw! I got the same one, but haven't ran into any issues with lockups.

I'm currently running 2.6.33.1 (kernel26-ck x86_64 version) and haven't experienced any issues. I also haven't set any specific kernel options either... Let me know if you want me to try anything to see if I can reproduce the issue..

Offline

#3 2010-03-23 06:14:41

pkerwien
Member
From: Sweden
Registered: 2009-07-06
Posts: 14

Re: Server goes to sleep - related to hpet problem?

I usually have the "lockups" when looking at a video over NFS from the server. But I think it can happen anytime. Just pressing a button on the temporary connected USB keyboard, wakes up the server and network etc starts to work again.

Your Con Kolivas kernel, how it is configured? Could you post your kernel config file? Perhaps I should try to build my own kernel that I used during my long Gentoo time.

My PSU is a Seasonic 380W, which only has a 4-pin ATX12V connector. I hope that is not a problem with this motherboard. CPU is a Phenom II 955. The other plugged in peripherals are a Areca 1220 RAID controller, Radeon 7xxx PCI graphics card and a Intel Pro 1000 NIC.

When I removed the tickless support (NO_HZ) in the kernel, I see this now when the lockups occur:

BUG: soft lockup - CPU#3 stuck for 92s! [events/3:18]
Modules linked in: md5 xt_tcpudp nfsd exportfs nfs lockd fscache nfs_acl auth_rpcgss sunrpc iptable_filter ip_tables x_tables ext2 usbhid hid usb_storage ohci_hcd radeon ttm drm_kms_helper drm i2c_algo_bit ppdev amd64_edac_mod edac_core edac_mce_amd ehci_hcd xhci usbcore parport_pc shpchp pci_hotplug i2c_piix4 i2c_core sg k10temp e1000e lp parport button thermal serio_raw evdev cpufreq_ondemand powernow_k8 freq_table processor rtc_cmos rtc_core rtc_lib ext4 mbcache jbd2 crc16 dm_mod sd_mod pata_jmicron ata_generic pata_acpi pata_atiixp ahci arcmsr libata scsi_mod
CPU 3
Pid: 18, comm: events/3 Not tainted 2.6.33-ARCH #2 GA-790FXTA-UD5/GA-790FXTA-UD5
RIP: 0010:[<ffffffffa0061268>]  [<ffffffffa0061268>] arcmsr_interrupt+0x58/0x1e0 [arcmsr]
RSP: 0018:ffff880005583e78  EFLAGS: 00000246
RAX: ffffc90000070000 RBX: ffff880005583eb8 RCX: 0000000000000000
RDX: ffff880005580164 RSI: ffff880120b485a0 RDI: ffff880120b485a0
RBP: ffffffff8100a8d3 R08: 0000000000000000 R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000000 R12: ffff880005583df0
R13: 000000000000000e R14: ffff880120b485a0 R15: ffffffff81024af7
FS:  00007f2f1baa1700(0000) GS:ffff880005580000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 00007fd3be75b3a0 CR3: 00000001227a9000 CR4: 00000000000006e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process events/3 (pid: 18, threadinfo ffff8801278be000, task ffff8801278b1560)
Stack:
 0000000000000000 ffff880005583e90 ffffffff81259dfc ffff880120b485a0
<0> 0000000000000000 0000000000000000 0000000000000012 0000000000000001
<0> ffff880005583ed8 ffffffffa00616e1 ffffffff81057d08 ffff8801218c1000
Call Trace:
 <IRQ>
 [<ffffffff81259dfc>] ? uart_tasklet_action+0xc/0x10
 [<ffffffffa00616e1>] ? arcmsr_do_interrupt+0x21/0x40 [arcmsr]
 [<ffffffff81057d08>] ? __do_softirq+0x118/0x1f0
 [<ffffffff810aa578>] ? handle_IRQ_event+0x58/0x160
 [<ffffffff810ac5fe>] ? handle_fasteoi_irq+0x6e/0xe0
 [<ffffffff8100ae1c>] ? call_softirq+0x1c/0x30
 [<ffffffff8100d27d>] ? handle_irq+0x1d/0x30
 [<ffffffff8100c777>] ? do_IRQ+0x67/0xf0
 [<ffffffff8133fc13>] ? ret_from_intr+0x0/0x11
 <EOI>
 [<ffffffffa01b16e8>] ? e1000e_update_stats+0xe8/0x730 [e1000e]
 [<ffffffff81060733>] ? add_timer+0x13/0x20
 [<ffffffffa01b6507>] ? e1000_watchdog_task+0x77/0x5b0 [e1000e]
 [<ffffffff810e9d30>] ? vmstat_update+0x0/0x40
 [<ffffffffa01b6490>] ? e1000_watchdog_task+0x0/0x5b0 [e1000e]
 [<ffffffff8106ac0a>] ? worker_thread+0x15a/0x280
 [<ffffffff8106f690>] ? autoremove_wake_function+0x0/0x40
 [<ffffffff8106aab0>] ? worker_thread+0x0/0x280
 [<ffffffff8106f19e>] ? kthread+0x8e/0xa0
 [<ffffffff8100ad24>] ? kernel_thread_helper+0x4/0x10
 [<ffffffff8106f110>] ? kthread+0x0/0xa0
 [<ffffffff8100ad20>] ? kernel_thread_helper+0x0/0x10
Code: 00 00 b8 01 00 00 00 48 8b 5d d8 4c 8b 65 e0 4c 8b 6d e8 4c 8b 75 f0 4c 8b 7d f8 c9 c3 66 0f 1f 44 00 00 48 8b 47 28 44 8b 60 30 <44> 23 67 20 41 f6 c4 1f 74 72 44 89 60 30 41 f6 c4 04 74 20 48
Call Trace:
 <IRQ>  [<ffffffff81259dfc>] ? uart_tasklet_action+0xc/0x10
 [<ffffffffa00616e1>] ? arcmsr_do_interrupt+0x21/0x40 [arcmsr]
 [<ffffffff81057d08>] ? __do_softirq+0x118/0x1f0
 [<ffffffff810aa578>] ? handle_IRQ_event+0x58/0x160
 [<ffffffff810ac5fe>] ? handle_fasteoi_irq+0x6e/0xe0
 [<ffffffff8100ae1c>] ? call_softirq+0x1c/0x30
 [<ffffffff8100d27d>] ? handle_irq+0x1d/0x30
 [<ffffffff8100c777>] ? do_IRQ+0x67/0xf0
 [<ffffffff8133fc13>] ? ret_from_intr+0x0/0x11
 <EOI>  [<ffffffffa01b16e8>] ? e1000e_update_stats+0xe8/0x730 [e1000e]
 [<ffffffff81060733>] ? add_timer+0x13/0x20
 [<ffffffffa01b6507>] ? e1000_watchdog_task+0x77/0x5b0 [e1000e]
 [<ffffffff810e9d30>] ? vmstat_update+0x0/0x40
 [<ffffffffa01b6490>] ? e1000_watchdog_task+0x0/0x5b0 [e1000e]
 [<ffffffff8106ac0a>] ? worker_thread+0x15a/0x280
 [<ffffffff8106f690>] ? autoremove_wake_function+0x0/0x40
 [<ffffffff8106aab0>] ? worker_thread+0x0/0x280
 [<ffffffff8106f19e>] ? kthread+0x8e/0xa0
 [<ffffffff8100ad24>] ? kernel_thread_helper+0x4/0x10
 [<ffffffff8106f110>] ? kthread+0x0/0xa0
 [<ffffffff8100ad20>] ? kernel_thread_helper+0x0/0x10

I have now disabled the ondemand cpufreq governor, and the CPUs are now running at a constant freq of 800MHz. 8 hours so far without a lockup.

Not stable. The lockups are still there. But it took over 24h before it happened this time :-(

I'm now testing with everything as normal but with nomodeset, removed the modules radeon, drm, and everything kms related. So far 18h without a freeze.

Last edited by pkerwien (2010-03-24 18:29:18)


Linux is just like an indian tent: no Gates, no Windows and an Apache inside.

Offline

Board footer

Powered by FluxBB