unofficial linux-ck repo with cpu optimized packages is ONLINE

paneless · 2015-05-14 06:12:02

graysky wrote:

@paneless and Miblo - OK. And if you revert to 4.0.3-1-ck?

Three boots in, and no panics so far.

mauritiusdadd · 2015-05-14 07:34:22

I can confirm, after few hours of testing (and more than 30 reboots), that 4.0.2-6 works, on the other hand 4.0.3-2 does not as it panics both on shut-down/reboot and when the system is running. I'm going to test 4.0.3-1 now.

-- edit: 4.0.3-1 seems to work well.

-- edit2: typo

Last edited by mauritiusdadd (2015-05-14 08:55:33)

graysky · 2015-05-14 12:45:33

OK. It looks like: CK's three patches + NUMA enabled is stable under the 4.0 code base.

All - feel free to switch to the repo-hosted 4.0.3-1-ck build (or build your own from the AUR) and please report any issues you experience here.

Jellicent · 2015-05-14 14:43:25

This is not exactly a problem with the packages, but with the repository. Everytime I try to update I either get a time-out or reach the servers while updating my db and check for updates. But even if I don't have a time-out, I'm sure to get a time-out when I try to download the updates. I have been unable to update for five days now, and I've tried hourly. I have a feeling the server doesn't like me.

Any ideas? I have already added the repo server multiple times to the pacman.conf. It just keeps timing out.

I'll try to compile from AUR for now.

Nietzsche · 2015-05-14 15:36:15

Jellicent wrote:

This is not exactly a problem with the packages, but with the repository. Everytime I try to update I either get a time-out or reach the servers while updating my db and check for updates. But even if I don't have a time-out, I'm sure to get a time-out when I try to download the updates. I have been unable to update for five days now, and I've tried hourly. I have a feeling the server doesn't like me.
Any ideas? I have already added the repo server multiple times to the pacman.conf. It just keeps timing out.
I'll try to compile from AUR for now.

RTFM (https://wiki.archlinux.org/index.php/Repo-ck)

Miblo · 2015-05-14 15:44:28

Nietzsche wrote:

Jellicent wrote:
This is not exactly a problem with the packages, but with the repository. Everytime I try to update I either get a time-out or reach the servers while updating my db and check for updates. But even if I don't have a time-out, I'm sure to get a time-out when I try to download the updates. I have been unable to update for five days now, and I've tried hourly. I have a feeling the server doesn't like me.
Any ideas? I have already added the repo server multiple times to the pacman.conf. It just keeps timing out.
I'll try to compile from AUR for now.
RTFM (https://wiki.archlinux.org/index.php/Repo-ck)

That's a bit harsh. Jellicent has added the server multiple times, as suggested in TFM. The only further thing suggested there is to try changing pacman's downloader to wget.

Nietzsche · 2015-05-14 15:47:56

Miblo wrote:

Nietzsche wrote:
Jellicent wrote:
This is not exactly a problem with the packages, but with the repository. Everytime I try to update I either get a time-out or reach the servers while updating my db and check for updates. But even if I don't have a time-out, I'm sure to get a time-out when I try to download the updates. I have been unable to update for five days now, and I've tried hourly. I have a feeling the server doesn't like me.
Any ideas? I have already added the repo server multiple times to the pacman.conf. It just keeps timing out.
I'll try to compile from AUR for now.
RTFM (https://wiki.archlinux.org/index.php/Repo-ck)
That's a bit harsh. Jellicent has added the server multiple times, as suggested in TFM. The only further thing suggested there is to try changing pacman's downloader to wget.

Yeah, I guess. I think the wiki page should only mention the wget solution (even if the pacman output is a little messy with it).

Jellicent · 2015-05-14 19:20:21

Nietzsche wrote:

Miblo wrote:
Nietzsche wrote:
RTFM (https://wiki.archlinux.org/index.php/Repo-ck)
That's a bit harsh. Jellicent has added the server multiple times, as suggested in TFM. The only further thing suggested there is to try changing pacman's downloader to wget.
Yeah, I guess. I think the wiki page should only mention the wget solution (even if the pacman output is a little messy with it).

Thank you both.

I refrained from using wget because last time I tried it, which was a few weeks ago, it didn't work (as in nothing would download at all). Now it works as a solution though.

graysky · 2015-05-14 20:13:18

I think the problem is that godaddy get just swamped with d/l requests when I push new version to the repo since many Archers are pretty diligent in `pacman -Syu` ... or you could make the argument that they just suck at web hosting

keepitsimpleengineer · 2015-05-14 20:53:00

So far so good with 4.0.3-1-ck & BFS using schedtool SCHED_BATCH on i7-2600 Sandy Bridge Quad-Core 3.4GHz Turboboost hyperthreading. Four cores running around 3.5GHz, C0% all mostly > 90, no X windows... .. .
htop

graysky · 2015-05-14 23:20:53

OK guys. CK is seeking volunteers to help troubleshoot this issue.

1) You need to be affected by the kernel panics (note that 6 by my count have posted so far) experiencing kernel panics when NUMA is disabled under v4.0.x.
2) You need to understand that this release could result in data loss; see CK's email to me below.

If you qualify for both 1 and 2, please try it and help troubleshoot.

Con Kolivas wrote:

Okay get rid of the last test patch. Attached is a potentially dangerous patch which removes the crash that people are experiencing and replaces it with a warn message to try decipher better what the problem is. When you post this, PLEASE give a big fat warning saying that it might cause filesystem corruption and they should not use it on a system that has crucial data. If it runs for them, please get them to look for warnings in their system logs/dmesg output.

What's different from 4.0.3-1-ck which is now in the repo?
*NUMA is disabled in 4.0.3-3-ck.
*4.0.3-3-ck contains CK's new patch to help troubleshoot.
*Big fat warning, "this may cause filesystem corruptions; only use on a system that has does not have crucial data."

Files will be up in a few min.

Option 1 (roll your own):

http://repo-ck.com/PKG_source/next/testing/linux-ck-4.0.3-3.src.tar.gz

Option 2 (my builds):

pacman -U http://repo-ck.com/PKG_source/next/testing/linux-ck-4.0.3-3-x86_64.pkg.tar.xz
pacman -U http://repo-ck.com/PKG_source/next/testing/linux-ck-headers-4.0.3-3-x86_64.pkg.tar.xz

Last edited by graysky (2015-05-14 23:23:46)

eaufavor · 2015-05-15 03:22:14

Sorry about the delay but today I just find that linux-ck-4.0.3-2 with BFQ hangs several minutes after booted up. No kernel panic at all, everything just freezes. Need to hard reboot. journalctl shows nothing.

Everything was normal yesterday. Today linux-ck-4.0.3-2 hangs every single time. I feel the time until freeze is shorter each time I boot. Maybe that's why nothing happened yesterday.

No upgrade. i5 + radeon R9 open source driver +brtfs SSD

linux-ck-4.0.3-1 works fine.

eaufavor wrote:

linux-ck-4.0.3-1 and linux-ck-4.0.3-2 both work fine.
graysky wrote:
@nordic89, mauritiusdadd, paneless, miblo, inglor, and anyone else who has experienced a panic: glad to hear symptoms seem to have subsided. With no instant panics, I think it's safe to move this setting into the 4.0.3 code base now. I have compiled the corresponding v4.0.3 version with NUMA disabled for you to test.
What's different from 4.0.3-1-ck which is now in the repo?
*NUMA is disabled in 4.0.3-2-ck.
*Both 4.0.3-1-ck and 4.0.3-2-ck have the new patch from CK which seems to have fixed the panics.
Option 1 (roll your own):
http://repo-ck.com/PKG_source/next/testing/linux-ck-4.0.3-2.src.tar.gz
Option 2 (my builds):
pacman -U http://repo-ck.com/PKG_source/next/testing/linux-ck-4.0.3-2-x86_64.pkg.tar.xz
pacman -U http://repo-ck.com/PKG_source/next/testing/linux-ck-headers-4.0.3-2-x86_64.pkg.tar.xz

mauritiusdadd · 2015-05-15 05:17:19

graysky wrote:

OK guys. CK is seeking volunteers to help troubleshoot this issue.
1) You need to be affected by the kernel panics (note that 6 by my count have posted so far) experiencing kernel panics when NUMA is disabled under v4.0.x.
2) You need to understand that this release could result in data loss; see CK's email to me below.
If you qualify for both 1 and 2, please try it and help troubleshoot.
Con Kolivas wrote:
Okay get rid of the last test patch. Attached is a potentially dangerous patch which removes the crash that people are experiencing and replaces it with a warn message to try decipher better what the problem is. When you post this, PLEASE give a big fat warning saying that it might cause filesystem corruption and they should not use it on a system that has crucial data. If it runs for them, please get them to look for warnings in their system logs/dmesg output.

I just installed 4.0.3-3-ck and it still crashing but no message is written to the journal. I will continue to test it and I'll report back if I get some useful message.

--edit: This time I was able to take a picture of what was printed on screen during the shutdown http://i.imgur.com/hfaKZ9j.jpg, but due to the data corruption the journal was unreadable. After a second shutdown I was also able to retrieve this from the journal.

mag 15 07:18:40 530U3C systemd[1]: Reached target Shutdown.
mag 15 07:18:40 530U3C systemd[1]: Starting Shutdown.
mag 15 07:18:40 530U3C systemd[1]: Reached target Final Step.
mag 15 07:18:40 530U3C systemd[1]: Starting Final Step.
mag 15 07:18:40 530U3C systemd[1]: Starting Reboot...
mag 15 07:18:40 530U3C systemd[1]: Shutting down.
mag 15 07:18:40 530U3C systemd[1]: Hardware watchdog 'iTCO_wdt', version 0
mag 15 07:18:40 530U3C systemd[1]: Set hardware watchdog to 10min.
mag 15 07:18:40 530U3C kernel: watchdog watchdog0: watchdog did not stop!
mag 15 07:18:40 530U3C systemd-shutdown[1]: Sending SIGTERM to remaining processes...
mag 15 07:18:40 530U3C kernel: BUG: Dentry ffff8800ce9c8cc0{i=1,n=/}  still in use (2) [unmount of cgroup cgroup]
mag 15 07:18:40 530U3C kernel: ------------[ cut here ]------------
mag 15 07:18:40 530U3C kernel: WARNING: CPU: 2 PID: 142 at fs/dcache.c:1399 umount_check+0x7c/0x90()
mag 15 07:18:40 530U3C kernel: Modules linked in: uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev media joydev mousedev arc4 iwldvm coretemp hwmon intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_hdmi mac80211 kvm_intel kvm snd_hda_codec_realtek crct10dif_pclmul crc32_pclmul crc32c_intel snd_hda_codec_generic psmouse iTCO_wdt ghash_clmulni_intel iTCO_vendor_support serio_raw iwlwifi atkbd libps2 samsung_laptop led_class snd_hda_intel aesni_intel snd_hda_controller xhci_pci aes_x86_64 lrw cfg80211 gf128mul glue_helper xhci_hcd ablk_helper snd_hda_codec cryptd pcspkr snd_hwdep ehci_pci ehci_hcd r8169 snd_pcm snd_timer i2c_i801 lpc_ich mii rfkill usbcore snd mei_me mei soundcore shpchp usb_common thermal fan tpm_tis i8042 wmi serio tpm battery dell_smo8800 ac evdev
mag 15 07:18:40 530U3C kernel:  mac_hid processor sch_fq_codel ext4 crc16 mbcache jbd2 sd_mod i915 button intel_gtt i2c_algo_bit video drm_kms_helper drm i2c_core ahci libahci libata scsi_mod
mag 15 07:18:40 530U3C kernel: CPU: 2 PID: 142 Comm: systemd-udevd Not tainted 4.0.3-3-ck #1
mag 15 07:18:40 530U3C kernel: Hardware name: SAMSUNG ELECTRONICS CO., LTD. 530U3C/530U4C/SAMSUNG_NP1234567890, BIOS P14AAJ 04/15/2013
mag 15 07:18:40 530U3C kernel:  0000000000000000 00000000b88249c3 ffff8800ce58fc48 ffffffff8154afcd
mag 15 07:18:40 530U3C kernel:  0000000000000000 0000000000000000 ffff8800ce58fc88 ffffffff8107383a
mag 15 07:18:40 530U3C kernel:  ffff8800ce58fc68 ffff8800ce9c8cc0 ffff880196c38800 ffffffff81637000
mag 15 07:18:40 530U3C kernel: Call Trace:
mag 15 07:18:40 530U3C kernel:  [<ffffffff8154afcd>] dump_stack+0x4c/0x6e
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107383a>] warn_slowpath_common+0x8a/0xc0
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107396a>] warn_slowpath_null+0x1a/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c72dc>] umount_check+0x7c/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c4bea>] d_walk+0x6a/0x340
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c6fe5>] ? shrink_dcache_parent+0x65/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c7260>] ? d_invalidate+0x130/0x130
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c703a>] do_one_tree+0x2a/0x50
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c7c4f>] shrink_dcache_for_umount+0x2f/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b0e25>] generic_shutdown_super+0x25/0x100
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b11b6>] kill_anon_super+0x16/0x30
mag 15 07:18:40 530U3C kernel:  [<ffffffff81223c5f>] kernfs_kill_sb+0x6f/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff810e6b59>] cgroup_kill_sb+0x49/0x70
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b15a9>] deactivate_locked_super+0x49/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1a1c>] deactivate_super+0x6c/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce843>] cleanup_mnt+0x43/0xa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce8f2>] __cleanup_mnt+0x12/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff8108f684>] task_work_run+0xd4/0xf0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076099>] do_exit+0x2f9/0xaa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff810768d5>] do_group_exit+0x45/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076954>] SyS_exit_group+0x14/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff81550dc9>] system_call_fastpath+0x12/0x17
mag 15 07:18:40 530U3C kernel: ---[ end trace 653efa56b4f00368 ]---
mag 15 07:18:40 530U3C kernel: VFS: Busy inodes after unmount of cgroup. Self-destruct in 5 seconds.  Have a nice day...
mag 15 07:18:40 530U3C kernel: ------------[ cut here ]------------
mag 15 07:18:40 530U3C kernel: WARNING: CPU: 2 PID: 142 at fs/super.c:152 destroy_super+0xa9/0xb0()
mag 15 07:18:40 530U3C kernel: Modules linked in: uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev media joydev mousedev arc4 iwldvm coretemp hwmon intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_hdmi mac80211 kvm_intel kvm snd_hda_codec_realtek crct10dif_pclmul crc32_pclmul crc32c_intel snd_hda_codec_generic psmouse iTCO_wdt ghash_clmulni_intel iTCO_vendor_support serio_raw iwlwifi atkbd libps2 samsung_laptop led_class snd_hda_intel aesni_intel snd_hda_controller xhci_pci aes_x86_64 lrw cfg80211 gf128mul glue_helper xhci_hcd ablk_helper snd_hda_codec cryptd pcspkr snd_hwdep ehci_pci ehci_hcd r8169 snd_pcm snd_timer i2c_i801 lpc_ich mii rfkill usbcore snd mei_me mei soundcore shpchp usb_common thermal fan tpm_tis i8042 wmi serio tpm battery dell_smo8800 ac evdev
mag 15 07:18:40 530U3C kernel:  mac_hid processor sch_fq_codel ext4 crc16 mbcache jbd2 sd_mod i915 button intel_gtt i2c_algo_bit video drm_kms_helper drm i2c_core ahci libahci libata scsi_mod
mag 15 07:18:40 530U3C kernel: CPU: 2 PID: 142 Comm: systemd-udevd Tainted: G        W       4.0.3-3-ck #1
mag 15 07:18:40 530U3C kernel: Hardware name: SAMSUNG ELECTRONICS CO., LTD. 530U3C/530U4C/SAMSUNG_NP1234567890, BIOS P14AAJ 04/15/2013
mag 15 07:18:40 530U3C kernel:  0000000000000000 00000000b88249c3 ffff8800ce58fd78 ffffffff8154afcd
mag 15 07:18:40 530U3C kernel:  0000000000000000 0000000000000000 ffff8800ce58fdb8 ffffffff8107383a
mag 15 07:18:40 530U3C kernel:  ffffffff81844d20 ffff880196c38800 ffffffff81844d20 ffff880197acede0
mag 15 07:18:40 530U3C kernel: Call Trace:
mag 15 07:18:40 530U3C kernel:  [<ffffffff8154afcd>] dump_stack+0x4c/0x6e
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107383a>] warn_slowpath_common+0x8a/0xc0
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107396a>] warn_slowpath_null+0x1a/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b0df9>] destroy_super+0xa9/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1503>] __put_super.part.5+0x23/0x30
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b153e>] put_super+0x2e/0x50
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b15d1>] deactivate_locked_super+0x71/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1a1c>] deactivate_super+0x6c/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce843>] cleanup_mnt+0x43/0xa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce8f2>] __cleanup_mnt+0x12/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff8108f684>] task_work_run+0xd4/0xf0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076099>] do_exit+0x2f9/0xaa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff810768d5>] do_group_exit+0x45/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076954>] SyS_exit_group+0x14/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff81550dc9>] system_call_fastpath+0x12/0x17
mag 15 07:18:40 530U3C kernel: ---[ end trace 653efa56b4f00369 ]---
mag 15 07:18:40 530U3C kernel: BUG: Dentry ffff8800ce9c8f00{i=1,n=/}  still in use (2) [unmount of cgroup cgroup]
mag 15 07:18:40 530U3C kernel: ------------[ cut here ]------------
mag 15 07:18:40 530U3C kernel: WARNING: CPU: 2 PID: 142 at fs/dcache.c:1399 umount_check+0x7c/0x90()
mag 15 07:18:40 530U3C kernel: Modules linked in: uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev media joydev mousedev arc4 iwldvm coretemp hwmon intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_hdmi mac80211 kvm_intel kvm snd_hda_codec_realtek crct10dif_pclmul crc32_pclmul crc32c_intel snd_hda_codec_generic psmouse iTCO_wdt ghash_clmulni_intel iTCO_vendor_support serio_raw iwlwifi atkbd libps2 samsung_laptop led_class snd_hda_intel aesni_intel snd_hda_controller xhci_pci aes_x86_64 lrw cfg80211 gf128mul glue_helper xhci_hcd ablk_helper snd_hda_codec cryptd pcspkr snd_hwdep ehci_pci ehci_hcd r8169 snd_pcm snd_timer i2c_i801 lpc_ich mii rfkill usbcore snd mei_me mei soundcore shpchp usb_common thermal fan tpm_tis i8042 wmi serio tpm battery dell_smo8800 ac evdev
mag 15 07:18:40 530U3C kernel:  mac_hid processor sch_fq_codel ext4 crc16 mbcache jbd2 sd_mod i915 button intel_gtt i2c_algo_bit video drm_kms_helper drm i2c_core ahci libahci libata scsi_mod
mag 15 07:18:40 530U3C kernel: CPU: 2 PID: 142 Comm: systemd-udevd Tainted: G        W       4.0.3-3-ck #1
mag 15 07:18:40 530U3C kernel: Hardware name: SAMSUNG ELECTRONICS CO., LTD. 530U3C/530U4C/SAMSUNG_NP1234567890, BIOS P14AAJ 04/15/2013
mag 15 07:18:40 530U3C kernel:  0000000000000000 00000000b88249c3 ffff8800ce58fc48 ffffffff8154afcd
mag 15 07:18:40 530U3C kernel:  0000000000000000 0000000000000000 ffff8800ce58fc88 ffffffff8107383a
mag 15 07:18:40 530U3C kernel:  ffff8800ce58fc68 ffff8800ce9c8f00 ffff880196c39000 ffffffff81637000
mag 15 07:18:40 530U3C kernel: Call Trace:
mag 15 07:18:40 530U3C kernel:  [<ffffffff8154afcd>] dump_stack+0x4c/0x6e
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107383a>] warn_slowpath_common+0x8a/0xc0
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107396a>] warn_slowpath_null+0x1a/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c72dc>] umount_check+0x7c/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c4bea>] d_walk+0x6a/0x340
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c6fe5>] ? shrink_dcache_parent+0x65/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c7260>] ? d_invalidate+0x130/0x130
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c703a>] do_one_tree+0x2a/0x50
mag 15 07:18:40 530U3C kernel:  [<ffffffff811c7c4f>] shrink_dcache_for_umount+0x2f/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b0e25>] generic_shutdown_super+0x25/0x100
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b11b6>] kill_anon_super+0x16/0x30
mag 15 07:18:40 530U3C kernel:  [<ffffffff81223c5f>] kernfs_kill_sb+0x6f/0x90
mag 15 07:18:40 530U3C kernel:  [<ffffffff810e6b59>] cgroup_kill_sb+0x49/0x70
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b15a9>] deactivate_locked_super+0x49/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1a1c>] deactivate_super+0x6c/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce843>] cleanup_mnt+0x43/0xa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce8f2>] __cleanup_mnt+0x12/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff8108f684>] task_work_run+0xd4/0xf0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076099>] do_exit+0x2f9/0xaa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff810768d5>] do_group_exit+0x45/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076954>] SyS_exit_group+0x14/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff81550dc9>] system_call_fastpath+0x12/0x17
mag 15 07:18:40 530U3C kernel: ---[ end trace 653efa56b4f0036a ]---
mag 15 07:18:40 530U3C kernel: VFS: Busy inodes after unmount of cgroup. Self-destruct in 5 seconds.  Have a nice day...
mag 15 07:18:40 530U3C kernel: ------------[ cut here ]------------
mag 15 07:18:40 530U3C kernel: WARNING: CPU: 2 PID: 142 at fs/super.c:152 destroy_super+0xa9/0xb0()
mag 15 07:18:40 530U3C kernel: Modules linked in: uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core v4l2_common videodev media joydev mousedev arc4 iwldvm coretemp hwmon intel_rapl iosf_mbi x86_pkg_temp_thermal intel_powerclamp snd_hda_codec_hdmi mac80211 kvm_intel kvm snd_hda_codec_realtek crct10dif_pclmul crc32_pclmul crc32c_intel snd_hda_codec_generic psmouse iTCO_wdt ghash_clmulni_intel iTCO_vendor_support serio_raw iwlwifi atkbd libps2 samsung_laptop led_class snd_hda_intel aesni_intel snd_hda_controller xhci_pci aes_x86_64 lrw cfg80211 gf128mul glue_helper xhci_hcd ablk_helper snd_hda_codec cryptd pcspkr snd_hwdep ehci_pci ehci_hcd r8169 snd_pcm snd_timer i2c_i801 lpc_ich mii rfkill usbcore snd mei_me mei soundcore shpchp usb_common thermal fan tpm_tis i8042 wmi serio tpm battery dell_smo8800 ac evdev
mag 15 07:18:40 530U3C kernel:  mac_hid processor sch_fq_codel ext4 crc16 mbcache jbd2 sd_mod i915 button intel_gtt i2c_algo_bit video drm_kms_helper drm i2c_core ahci libahci libata scsi_mod
mag 15 07:18:40 530U3C kernel: CPU: 2 PID: 142 Comm: systemd-udevd Tainted: G        W       4.0.3-3-ck #1
mag 15 07:18:40 530U3C kernel: Hardware name: SAMSUNG ELECTRONICS CO., LTD. 530U3C/530U4C/SAMSUNG_NP1234567890, BIOS P14AAJ 04/15/2013
mag 15 07:18:40 530U3C kernel:  0000000000000000 00000000b88249c3 ffff8800ce58fd78 ffffffff8154afcd
mag 15 07:18:40 530U3C kernel:  0000000000000000 0000000000000000 ffff8800ce58fdb8 ffffffff8107383a
mag 15 07:18:40 530U3C kernel:  ffffffff81844d20 ffff880196c39000 ffffffff81844d20 ffff880197acede0
mag 15 07:18:40 530U3C kernel: Call Trace:
mag 15 07:18:40 530U3C kernel:  [<ffffffff8154afcd>] dump_stack+0x4c/0x6e
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107383a>] warn_slowpath_common+0x8a/0xc0
mag 15 07:18:40 530U3C kernel:  [<ffffffff8107396a>] warn_slowpath_null+0x1a/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b0df9>] destroy_super+0xa9/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1503>] __put_super.part.5+0x23/0x30
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b153e>] put_super+0x2e/0x50
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b15d1>] deactivate_locked_super+0x71/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811b1a1c>] deactivate_super+0x6c/0x80
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce843>] cleanup_mnt+0x43/0xa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff811ce8f2>] __cleanup_mnt+0x12/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff8108f684>] task_work_run+0xd4/0xf0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076099>] do_exit+0x2f9/0xaa0
mag 15 07:18:40 530U3C kernel:  [<ffffffff810768d5>] do_group_exit+0x45/0xb0
mag 15 07:18:40 530U3C kernel:  [<ffffffff81076954>] SyS_exit_group+0x14/0x20
mag 15 07:18:40 530U3C kernel:  [<ffffffff81550dc9>] system_call_fastpath+0x12/0x17
mag 15 07:18:40 530U3C kernel: ---[ end trace 653efa56b4f0036b ]---
mag 15 07:18:40 530U3C systemd-journal[112]: Journal stopped

NOTE: I was using BFQ, now I will switch to CFQ and see if there is something different.

Last edited by mauritiusdadd (2015-05-15 06:30:05)

rat5 · 2015-05-15 06:19:56

Hi,

I just want to say, that also for me the linux-ck 4.0.2 and 4.0.3 working fine

paneless · 2015-05-15 07:15:05

1) Yes I've seen the warning!

2) Installed 4.0.3-3-ck and currently have powered off / on four times without issue. BFQ enabled throughout. Will keep a close eye on the messages / journal and report any findings.

Last edited by paneless (2015-05-15 07:15:39)

mauritiusdadd · 2015-05-15 07:51:23

I don't know how the following information can be useful, but 4.0.3-3-ck panics also with CFQ and there is still no messages/warning in the journal so I forwarded its output to tty12 and these are the results: [1] [2]. They both happened when the system was running. I'm sorry that the head of the call trace is missing, but the messages are printed so fast on the screen that even capturing a video doesn't help.

Last edited by mauritiusdadd (2015-05-15 10:15:13)

graysky · 2015-05-15 08:01:30

Thanks all, see this recent note from Con as you test:

Con Kolivas wrote:

Thanks. Can you tell them to post the details of any warnings they see? Any kernel errors and warning. I want all the details.

paneless · 2015-05-15 08:45:15

The following is from this boot / session, no panics but these are the warnings in the journal.

May 15 09:21:23 localhost kernel: ACPI: RSDP 0x00000000000FAFB0 000024 (v02 HPQOEM)
May 15 09:21:23 localhost kernel: ACPI: XSDT 0x00000000CFFB0100 00006C (v01 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI: FACP 0x00000000CFFB0290 0000F4 (v03 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI BIOS Warning (bug): Optional FADT field Pm2ControlBlock has zero address or length: 0x0000000000000000/0x1 (20150204/tbfadt-649)
May 15 09:21:23 localhost kernel: ACPI BIOS Warning (bug): 32/64X length mismatch in FADT/Gpe0Block: 64/32 (20150204/tbfadt-618)
May 15 09:21:23 localhost kernel: ACPI: DSDT 0x00000000CFFB05D0 004732 (v01 HPQOEM SLIC-CPC 00000423 INTL 20051117)
May 15 09:21:23 localhost kernel: ACPI: FACS 0x00000000CFFBE000 000040
May 15 09:21:23 localhost kernel: ACPI: APIC 0x00000000CFFB0390 00007C (v01 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI: MCFG 0x00000000CFFB0410 00003C (v01 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI: SLIC 0x00000000CFFB0450 000176 (v01 HPQOEM SLIC-CPC 00000001 MSFT 00000001)
May 15 09:21:23 localhost kernel: ACPI: OEMB 0x00000000CFFBE040 000072 (v01 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI: SSDT 0x00000000CFFBA6B0 000843 (v01 HPQOEM SLIC-CPC 00000001 INTL 20051117)
May 15 09:21:23 localhost kernel: ACPI: SRAT 0x00000000CFFBAF00 0000C8 (v03 HPQOEM SLIC-CPC 00000002 AMD  00000001)
May 15 09:21:23 localhost kernel: ACPI: HPET 0x00000000CFFBAFD0 000038 (v01 HPQOEM SLIC-CPC 20100423 MSFT 00000097)
May 15 09:21:23 localhost kernel: ACPI: SSDT 0x00000000CFFBB010 000458 (v01 HPQOEM SLIC-CPC 00000001 AMD  00000001)

FWIW I'm using Ext4 and Athlon II processors.

paneless · 2015-05-15 10:54:27

Kernel panic on the boot previous to this one. I've tried all the journalctl options (--since date/time / -b -1 etc.) but I only have journal output for the successful boots.

Don't know if this is in anyway related, but at the point systemd-journal appears, the time jumps back a few seconds.

May 15 11:03:41 nix systemd-journal[175]: Journal started
May 15 11:03:37 nix systemd[1]: systemd 219 running in system mode. (+PAM -AUDIT -SELINUX -IMA -APPARMOR +SMACK -SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +S
May 15 11:03:37 nix systemd[1]: Detected architecture 'x86-64'.

edit) Another panic on boot but this time I have the journal up to that point.

May 15 13:11:23 localhost systemd[1]: Reached target Switch Root.
May 15 13:11:23 localhost systemd[1]: Starting Switch Root.
May 15 13:11:23 localhost systemd[1]: Starting Switch Root...
May 15 13:11:23 localhost systemd[1]: Switching root.
May 15 13:11:23 localhost systemd-journal[57]: Journal stopped
May 15 13:11:25 nix systemd-journal[166]: Runtime journal is using 8.0M (max allowed 184.9M, trying to leave 277.4M free of 1.7G available → current limit 184.9M).
May 15 13:11:25 nix systemd-journal[166]: Permanent journal is using 37.5M (max allowed 50.0M, trying to leave 2.9G free of 15.4G available → current limit 50.0M).
May 15 13:11:26 nix systemd-journal[166]: Time spent on flushing to /var is 453.941ms for 735 entries.
May 15 13:11:26 nix systemd-journald[57]: Received SIGTERM from PID 1 (systemd).
May 15 13:11:26 nix kernel: Switched to clocksource tsc
May 15 13:11:26 nix systemd-journal[166]: Journal started
May 15 13:11:24 nix systemd[1]: systemd 219 running in system mode. (+PAM -AUDIT -SELINUX -IMA -APPARMOR +SMACK -SYSVINIT +UTMP +LIBCRYPTSETUP +GCRYPT +GNUTLS +ACL +XZ +LZ4 +S
May 15 13:11:24 nix systemd[1]: Detected architecture 'x86-64'.
May 15 13:11:24 nix systemd[1]: Set hostname to <nix>.
May 15 13:11:24 nix systemd[1]: Unit type .busname is not supported on this system.
May 15 13:11:26 nix systemd[1]: Starting Flush Journal to Persistent Storage...
May 15 13:11:26 nix systemd-udevd[191]: starting version 219
lines 703-746/746 (END)

Again there is a time jump of 2 seconds back after Journal started, and this does not seem to happen on the journals for successful boots.

During the boot / panic, just after the Starting to flush journal to persistent storage I received the following:
general protection fault: 0000 [#1] PREEMPT SMP
skb_dequeue+0x4b/0x80

Last edited by paneless (2015-05-15 12:49:09)

mauritiusdadd · 2015-05-15 14:00:36

graysky wrote:

Thanks all, see this recent note from Con as you test:
Con Kolivas wrote:
Thanks. Can you tell them to post the details of any warnings they see? Any kernel errors and warning. I want all the details.

Ok, these are all the details I can provide at the moment, all the following files are available in the folder linux-ck-debug on my google-drive.

The following three files contain the full journal of three boots of linux 4.0.3-3-ck provided by the package built by graysky

trace-01.txt: this is the full journal of the boot from which I have extracted the calltrace I attached in my previous post.
trace-02.txt: this is the full journal of another boot of 4.0.3-3-ck.
journal-nodebug-1.txt: another full journal referring to 4.0.3-3-ck.

Since it seems that when a panic occurs, journald have no time to write the messages to the system log, I rebuilt linux 4.0.3-3-ck with debugging symbols appending the following entries to the configuration file

CONFIG_DEBUG_INFO=y
CONFIG_DEBUG_INFO_REDUCED=n
CONFIG_DEBUG_INFO_SPLIT=n
CONFIG_DEBUG_INFO_DWARF4=n
CONFIG_PROC_VMCORE=y
CONFIG_CRASH_DUMP=y
CONFIG_GDB_SCRIPTS=n

. The following files refers to my build of 4.0.3-3-ck:

vmlinux: this is the statically linked executable file containing the Linux kernel and the debugging symbols
vmcore-{1,2,3,4}*: virtual memory dumps captured by Kdump
vmcore-1.dump: partial** memory dump generated from vmcore-1 using makedumpfile
vmcore-2.dump: partial** memory dump generated from vmcore-2 using makedumpfile
vmcore-3.dump: partial** memory dump generated from vmcore-3 using makedumpfile
vmcore-4.dump: partial** memory dump generated from vmcore-4 using makedumpfile
dmesg-dump-1.txt: dmesg extracted from vmcore-1 using makedumpfile
dmesg-dump-2.txt: dmesg extracted from vmcore-2 using makedumpfile
dmesg-dump-3.txt: dmesg extracted from vmcore-3 using makedumpfile
dmesg-dump-4.txt: dmesg extracted from vmcore-4 using makedumpfile

NOTES:
* the kernel memory dumps generated by Kdump are huge (around 5.7GB for each dump). It is a bit difficult for me to upload them with my internet connection limited to 1Mbps in upload, for this reason I decide to use makedumpfile to generate a smaller version (see **).

** These are trimmed versions of the memory dumps ( from which have been excluded pages containing only zeroes, pages used to cache file contents (cache, cache private), pages belonging to user spaces processes, and free pages) generated using the following command :

makedumpfile -d 31 -c -x vmlinux vmcore-${n} vmcore-${n}.dump

Last edited by mauritiusdadd (2015-05-18 07:01:37)

paneless · 2015-05-15 17:24:16

Several successsful boots since the last panic and no further panics to report, but, and this is only a hunch and doesn't have anything but a bit of searching behind it, the latest systemd package was released the day before these panics were first reported in this thread. I don't know if CK4.0 / NUMA could be affected by the systemd upgrade or not, but just thought I would throw it out there.

ProFfeSsoRr · 2015-05-17 16:31:02

Read this topic and try again compile 4.0.3 with localmodconfig and enable NUMA, and with tp_smapi installed. No panics within 30 minutes (without NUMA 1-2 minutes before panic), i'm happy.

nomorewindows · 2015-05-17 17:59:53

Do you do something different with linux-ck? Because I tried nvidia-ck and now I get the nvidia logo whereas I don't get it with stock kernel and nvidia.

graysky · 2015-05-17 18:26:44

@nomore - Nothing modified that should display the logo or not.

Miblo · 2015-05-17 19:07:50

Sorry to take a back seat for this potentially dangerous testing, guys. I'm in the precarious position of not having backups of this system. Great work, those of you who are doing it.

Arch Linux

#2426 2015-05-14 06:12:02

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2427 2015-05-14 07:34:22

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2428 2015-05-14 12:45:33

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2429 2015-05-14 14:43:25

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2430 2015-05-14 15:36:15

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2431 2015-05-14 15:44:28

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2432 2015-05-14 15:47:56

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2433 2015-05-14 19:20:21

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2434 2015-05-14 20:13:18

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2435 2015-05-14 20:53:00

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2436 2015-05-14 23:20:53

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2437 2015-05-15 03:22:14

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2438 2015-05-15 05:17:19

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2439 2015-05-15 06:19:56

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2440 2015-05-15 07:15:05

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2441 2015-05-15 07:51:23

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2442 2015-05-15 08:01:30

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2443 2015-05-15 08:45:15

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2444 2015-05-15 10:54:27

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2445 2015-05-15 14:00:36

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2446 2015-05-15 17:24:16

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2447 2015-05-17 16:31:02

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2448 2015-05-17 17:59:53

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2449 2015-05-17 18:26:44

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

#2450 2015-05-17 19:07:50

Re: unofficial linux-ck repo with cpu optimized packages is ONLINE

Board footer