| View previous topic :: View next topic |
| Author |
Message |
hvengel
Guru
Joined: 19 Sep 2004
Posts: 515
|
 Posted: Wed Nov 08, 2006 11:50 pm Post subject: |
 |
|
Your error is different from what I was seeing. I don't just freeze one app. When this happens to me every thing in X freezes other than the mouse pointer can still be moved. On my system it does not appear to be GTK related either since when I have seen this happen when no GTK apps were running.
I should add that since I turned off NUMA and voluntary preemption I have not seen a single freeze - knock on wood. Again it has only been about 4 days now so I can't proclaim that this is totally fixed just yet. But it for sure has been a huge improvement even if it turns not to completely fix the problem. To say the least this is the most stable my system has been in a long time.
The reason that glib was asked about (I think) is that this is the library that is used by many GTK apps that need to do multi-threaded things and that also need to be cross platform. glib has a threading API that hides the platform details so the programmers can write to a single API and have it work on all platforms. If the problem is related to threading then it could be glib that is being tripped up somehow. |
|
| Back to top |
|
 |
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2632
Location: Bay Area, CA
|
| Posted: Thu Nov 09, 2006 12:26 am Post subject: |
|
|
| fudge wrote: | | devsk wrote: | | fudge wrote: | | The only applications causing the hangs are GTK apps, firefox, eclipse and wireshark... |
Indeed, those are all gtk+ apps. Have you tried latest unstable versions of glib and gtk+ if you are running ARCH instead of ~ARCH? |
Already running unstable gtk+ and glib unstable is the same as stable. BTW, why do you think glib is involved here? I suspect it's in gtk+ *and* xorg. Why, because when I was running xorg-6.8.0.2 none of this was an issue. Since Gentoo decided that xorg-6.8 was obsolete, I had no real choice but to reluctantly move on to xorg-7.1. I've tried the nv driver as well, same result, sigh. |
sorry, for not seeing these messages. The forum notification is very flaky these days.
For you, the only option is to match the .config as closely as possible and disable PREEMPTION completely. Then, see what happens.
And since you are saying it worked with xorg 6.8.x, it is possible that latest xorg is no good for your machine. You can try 7.0 version and see if it helps. Also, we don't know anything about your hardware (mobo) and the problem at hand may be a hardware problem. What type of mouse are you using?
I think hvengel already covered your glib question well. Its like libc of the gtk apps, hence problems there will likely affect gtk apps. |
|
| Back to top |
|
|
fudge
Tux's lil' helper
Joined: 25 Jul 2002
Posts: 104
|
| Posted: Thu Nov 09, 2006 3:33 am Post subject: |
|
|
| devsk wrote: | sorry, for not seeing these messages. The forum notification is very flaky these days.
For you, the only option is to match the .config as closely as possible and disable PREEMPTION completely. Then, see what happens.
And since you are saying it worked with xorg 6.8.x, it is possible that latest xorg is no good for your machine. You can try 7.0 version and see if it helps. Also, we don't know anything about your hardware (mobo) and the problem at hand may be a hardware problem. What type of mouse are you using?
I think hvengel already covered your glib question well. Its like libc of the gtk apps, hence problems there will likely affect gtk apps. |
Are you suggesting turning off "Preempt Big Kernel Lock" as well as choosing the "No Force Preemption"?
Also, my mobo is the ASUS A8N32-SLI deluxe. I've disabled everything, including USB and am using PS2 mouse and keyboard. Once I've reached some sort of stability, I'll be switching things on one thing at a time. It's a bit frustrating, but I'm very glad kind souls on the forum, like yourselves, are willing to help and suggest things. Much obliged. |
|
| Back to top |
|
|
devsk
Advocate
Joined: 24 Oct 2003
Posts: 2632
Location: Bay Area, CA
|
| Posted: Thu Nov 09, 2006 6:25 am Post subject: |
|
|
| fudge wrote: | | Are you suggesting turning off "Preempt Big Kernel Lock" as well as choosing the "No Force Preemption"? |
You can leave PREEMPT_BKL in but choose "no forced preemption".
| fudge wrote: |
Also, my mobo is the ASUS A8N32-SLI deluxe. I've disabled everything, including USB and am using PS2 mouse and keyboard. Once I've reached some sort of stability, I'll be switching things on one thing at a time. It's a bit frustrating, but I'm very glad kind souls on the forum, like yourselves, are willing to help and suggest things. Much obliged. |
I hope you did some basic testing after you built or bought the system, like, running memtest+ overnight and running two or more prime95 instances under linux for extended period of time(like 24 hours), running badblocks (be VERY careful: use non-destructive mode if you can't lose data on disk and don't worry about the time it will take to finish) on each disk you added to the configuration. Basically, before software diagnostics, you have to make sure you have no "holes" in the hardware (minimally at least cpu, memory & disk). For example, I always do badblocks on every memory card I ever buy before using it. It just tells me if I need to RMA it. Same for new disks and RAM sticks. |
|
| Back to top |
|
|
fudge
Tux's lil' helper
Joined: 25 Jul 2002
Posts: 104
|
| Posted: Thu Nov 09, 2006 6:36 am Post subject: |
|
|
| devsk wrote: | | I hope you did some basic testing after you built or bought the system, like, running memtest+ overnight and running two or more prime95 instances under linux for extended period of time(like 24 hours), running badblocks (be VERY careful: use non-destructive mode if you can't lose data on disk and don't worry about the time it will take to finish) on each disk you added to the configuration. Basically, before software diagnostics, you have to make sure you have no "holes" in the hardware (minimally at least cpu, memory & disk). For example, I always do badblocks on every memory card I ever buy before using it. It just tells me if I need to RMA it. Same for new disks and RAM sticks. |
Ran memtest and cpuburn when I got the PC, so I'm confident that all that is OK. Remember xorg-6.8.0.2 gave me no problems, xorg-7.* was when the problems began. |
|
| Back to top |
|
|
Galahad
Tux's lil' helper
Joined: 12 Feb 2003
Posts: 126
|
| Posted: Mon Nov 13, 2006 11:17 pm Post subject: |
|
|
| On my system I think the nvidia 9629 drivers were to blame, at least I hope it's stable now with 8776. |
|
| Back to top |
|
|
Galahad
Tux's lil' helper
Joined: 12 Feb 2003
Posts: 126
|
| Posted: Tue Nov 21, 2006 10:19 pm Post subject: |
|
|
| 8776 is definitely stable, anything in the 90 series locks my system up hard after a very short time. |
|
| Back to top |
|
|
Nylan
n00b
Joined: 14 Jun 2003
Posts: 38
|
| Posted: Tue Feb 06, 2007 3:47 pm Post subject: |
|
|
Hello,
i think I'm having a similar problem.
My system randomly locks.
Error message from yesterday:
| Code: |
Feb 5 19:00:01 amber64 cron[15648]: (root) CMD (test -x /usr/sbin/run-crons && /usr/sbin/run-crons )
Feb 5 19:00:01 amber64 cron[15649]: (root) CMD (rm -f /var/spool/cron/lastrun/cron.hourly)
Feb 5 19:00:19 amber64 gdm(pam_unix)[12718]: session closed for user nylan
Feb 5 19:00:19 amber64 su(pam_unix)[14011]: session closed for user root
Feb 5 19:00:19 amber64 gdm[12718]: Error reinitilizing server
Feb 5 19:00:29 amber64 gdm(pam_unix)[15682]: session opened for user nylan by (uid=0)
Feb 5 19:00:33 amber64 (nylan-13738): Resolved address "xml:readwrite:/home/nylan/.gconf" to a writable configuration source at position 0
Feb 5 19:01:43 amber64 login(pam_unix)[13700]: authentication failure; logname= uid=0 euid=0 tty=tty1 ruser= rhost= user=root
Feb 5 19:01:47 amber64 login[13700]: FAILED LOGIN (1) on 'tty1' FOR `root', Authentication failure
Feb 5 19:01:49 amber64 login(pam_unix)[13700]: session opened for user root by (uid=0)
Feb 5 19:01:49 amber64 login[15963]: ROOT LOGIN on 'tty1'
Feb 5 19:03:59 amber64 BUG: using smp_processor_id() in preemptible [00000001] code: bug-buddy/15986
Feb 5 19:03:59 amber64 caller is oops_begin+0xb/0x6f
Feb 5 19:04:00 amber64
Feb 5 19:04:00 amber64 Call Trace:
Feb 5 19:04:00 amber64 [<ffffffff803459fa>] debug_smp_processor_id+0xaa/0xb8
Feb 5 19:04:00 amber64 [<ffffffff804e99ef>] oops_begin+0xb/0x6f
Feb 5 19:04:00 amber64 [<ffffffff8020ae13>] die+0x19/0x44
Feb 5 19:04:00 amber64 [<ffffffff8020b3e0>] do_invalid_op+0xad/0xb7
Feb 5 19:04:00 amber64 [<ffffffff8024d6aa>] generic_file_readonly_mmap+0xf/0x18
Feb 5 19:04:00 amber64 [<ffffffff8024fd3e>] generic_file_aio_read+0x15f/0x1b5
Feb 5 19:04:00 amber64 [<ffffffff804e961d>] error_exit+0x0/0x84
Feb 5 19:04:00 amber64 [<ffffffff8024d6aa>] generic_file_readonly_mmap+0xf/0x18
Feb 5 19:04:00 amber64 [<ffffffff8025e03d>] do_mmap_pgoff+0x480/0x75d
Feb 5 19:04:00 amber64 [<ffffffff804e91a0>] _spin_unlock_irq+0x15/0x30
Feb 5 19:04:00 amber64 [<ffffffff8020f15a>] sys_mmap+0x8b/0x110
Feb 5 19:04:00 amber64 [<ffffffff802095ae>] system_call+0x7e/0x83
Feb 5 19:04:00 amber64
Feb 5 19:04:00 amber64 invalid opcode: 0000 [1] PREEMPT SMP
Feb 5 19:04:00 amber64 CPU 0
Feb 5 19:04:00 amber64 Modules linked in: it87 hwmon_vid i2c_isa k8temp i2c_nforce2 nvidia(P) snd_seq_midi snd_emu10k1_synth snd_emux_synth snd_seq_virmidi snd_seq_midi_emul snd_pcm_oss snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_hda_intel snd_hda_codec 8139too skge snd_intel8x0 snd_emu10k1 snd_rawmidi snd_ac97_codec snd_ac97_bus snd_pcm snd_seq_device snd_timer snd_page_alloc snd_util_mem snd_hwdep snd sx8 sbp2 ohci1394 ieee1394 ohci_hcd uhci_hcd usb_storage usbhid ehci_hcd usbcore
Feb 5 19:04:00 amber64 Pid: 15986, comm: bug-buddy Tainted: P 2.6.19-gentoo-r5 #1
Feb 5 19:04:00 amber64 RIP: 0010:[<ffffffff8024d6aa>] [<ffffffff8024d6aa>] generic_file_readonly_mmap+0xf/0x18
Feb 5 19:04:00 amber64 RSP: 0018:ffff8100bae21e88 EFLAGS: 00010a06
Feb 5 19:04:00 amber64 RAX: ffffffff804fad80 RBX: ffff8100ce0071e0 RCX: 0000000000000000
Feb 5 19:04:00 amber64 RDX: 00000000fffffff8 RSI: ffff81010631f1c0 RDI: ffff810064f8bec4
Feb 5 19:04:00 amber64 RBP: 0000000000000075 R08: ffff8100bae21ef8 R09: 0000000000000000
Feb 5 19:04:00 amber64 R10: ffff8100ce0072b8 R11: ffff810113f6edb0 R12: 00000000ffffffea
Feb 5 19:04:00 amber64 R13: ffff810064f8bec0 R14: ffff810101372c68 R15: ffff81011b741bc0
Feb 5 19:04:00 amber64 FS: 00002b714effcab0(0000) GS:ffffffff80669000(0000) knlGS:00000000f75646b0
Feb 5 19:04:00 amber64 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
Feb 5 19:04:00 amber64 CR2: 00002b714b30e430 CR3: 000000010fac6000 CR4: 00000000000006e0
Feb 5 19:04:00 amber64 Process bug-buddy (pid: 15986, threadinfo ffff8100bae20000, task ffff81010523a180)
Feb 5 19:04:00 amber64 Stack: ffff8100ce0071e0 ffffffff8025e03d 0000000000000000 00000000455e29d9
Feb 5 19:04:00 amber64 0000000000000000 00000000455e29da 0000000000000000 0000000000000002
Feb 5 19:04:00 amber64 0000000000000000 0000000000000000 00002b714f31c000 0000000000000103
Feb 5 19:04:00 amber64 Call Trace:
Feb 5 19:04:00 amber64 [<ffffffff8025e03d>] do_mmap_pgoff+0x480/0x75d
Feb 5 19:04:00 amber64 [<ffffffff804e91a0>] _spin_unlock_irq+0x15/0x30
Feb 5 19:04:00 amber64 [<ffffffff8020f15a>] sys_mmap+0x8b/0x110
Feb 5 19:04:00 amber64 [<ffffffff802095ae>] system_call+0x7e/0x83
Feb 5 19:04:00 amber64
Feb 5 19:04:00 amber64
Feb 5 19:04:00 amber64 Code: ff ff ff b8 ea ff ff ff c3 41 56 49 89 d6 41 55 49 89 f5 41
Feb 5 19:04:00 amber64 RIP [<ffffffff8024d6aa>] generic_file_readonly_mmap+0xf/0x18
Feb 5 19:04:00 amber64 RSP <ffff8100bae21e88>
Feb 5 19:04:40 amber64 login(pam_unix)[13701]: session opened for user root by (uid=0)
Feb 5 19:04:40 amber64 login[15987]: ROOT LOGIN on 'tty2'
Feb 5 19:04:42 amber64 <6>emerge[15990] trap invalid opcode rip:2b1609a95678 rsp:7fffa1155e88 error:0
|
I've compiled 2.6.20 with PREEMPT and NUMA disabled. Same problems.
Hardware : Asus M2N32-SLI Deluxe
AMD X2 5000+
Geforce 7900 gs
Creative Audigy 2 ZS
I'll post informations/config as needed.
help appreciated  |
|
| Back to top |
|
|
guduri
Apprentice
Joined: 04 Apr 2005
Posts: 230
Location: Ann Arbor, Michigan
|
| Posted: Thu Feb 08, 2007 7:13 pm Post subject: |
|
|
I upgraded my agp graphics card from FX5500 to 6800 XT and I started noticing random freezes. Per Galahad, I downgraded my Nvidia drivers from 9xxx to 8776. My machine has not froze since. Looks like NVidia drivers could be unstable on some cards and stable on others.
_________________
Power is about what you can control. Freedom is about what you can unleash. |
|
| Back to top |
|
|
Nylan
n00b
Joined: 14 Jun 2003
Posts: 38
|
| Posted: Fri Feb 09, 2007 9:22 am Post subject: |
|
|
Culprit = Nvidia.
I downgraded my nvidia-drivers to 8776.
Short term solution. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Gentoo on AMD64
