On 8/14/20 8:45 AM, Przemek Klosowski via devel wrote:
That is correct. The "6:1" in the kworker case varies.On 8/14/20 8:33 AM, Wells, Roger K. via devel wrote:
That was not the cause. Now when it happens I have only three tasks running at 100% (same ones as reported earlier). Everything else, kerneloops, shutdown via power switch, etc, is as before.Could you repost to the list with more info? I think you're saying that after removing sadc you still are 100% CPU with
kworker/6:1+events_freezable dmesg systemd-journal
running full tilt. THis probably means that it's the kernel worker threads are doing something that causes lots of errors, and all the other processes (including sadc that is now gone) are just busy writing those errors. So, what is kernel doing? what does your system log say? type 'dmesg' or 'journalctl -e' and look at the end of the data.The following repeats at the end of journalctl:
In the last second of the journal there are 14688 individual entries.Aug 13 18:23:51 rwells-x280 kernel: Hardware name: LENOVO 20KFCTO1WW/20KFCTO1WW, BIOS N20ET55W (1.40 ) 06/01/2020
Aug 13 18:23:51 rwells-x280 kernel: Workqueue: events_freezable ieee80211_restart_work [mac80211]
Aug 13 18:23:51 rwells-x280 kernel: RIP: 0010:drv_sta_state+0x26e/0x3e0 [mac80211]
Aug 13 18:23:51 rwells-x280 kernel: Code: 0b 45 31 ed e9 44 fe ff ff 48 8b 83 78 04 00 00 48 8d b3 98 04 00 00 48 c7 c7 80 46 f2 c0 48 85 c0 48 0f 45 f0 e8 c9 bf 23 d2 <0f> 0b 41 bd fb ff ff ff e9 1b fe ff ff 65 8b 05 be fc 16 3f 89 c0
Aug 13 18:23:51 rwells-x280 systemd-journald[928]: Missed 44 kernel messages
Aug 13 18:23:51 rwells-x280 kernel: iTCO_wdt mac80211 iTCO_vendor_support snd_hda_intel irqbypass intel_rapl_msr snd_intel_dspcfg rapl snd_hda_codec libarc4 snd_hda_core uvcvideo intel_cstate iwlwifi snd_hwdep btusb videobuf2_vmalloc snd_seq intel_uncore videobuf2_memops btrtl videobuf2_v4l2 btbcm snd_seq_device btintel snd_pcm pcspkr cfg80211 mei_me wmi_bmof intel_wmi_thunderbolt videobuf2_common i2c_i801 snd_timer thunderbolt videodev bluetooth mei intel_pch_thermal mc thinkpad_acpi ucsi_acpi joydev typec_ucsi processor_thermal_device ecdh_generic intel_rapl_common ledtrig_audio ecc intel_xhci_usb_role_switch roles intel_soc_dts_iosf typec snd soundcore rfkill int3403_thermal int340x_thermal_zone int3400_thermal acpi_thermal_rel acpi_pad ip_tables dm_crypt i915 i2c_algo_bit cec crct10dif_pclmul crc32_pclmul crc32c_intel drm_kms_helper ghash_clmulni_intel nvme drm serio_raw e1000e uas nvme_core usb_storage wmi video hid_multitouch fuse
Aug 13 18:23:51 rwells-x280 kernel: CPU: 0 PID: 5 Comm: kworker/0:0 Tainted: G W 5.7.12-200.fc32.x86_64 #1
Also, the perf toolkit might help as explained in https://askubuntu.com/questions/33640/kworker-what-is-it-and-why-is-it-hogging-so-much-cpu
Install
perf:sudo apt-get install linux-tools-common linux-tools-3.11.0-15-generic(The second package must match your kernel version. You can first install just
linux-tools-commonand callperfto let it tell you which package it needs.)Record some 10 seconds of backtraces on all your CPUs:
sudo perf record -g -a sleep 10Analyse your recording:
sudo perf report
I'll give this a try
thanks
I did it but I should have remembered earlier:
When this happens all sudo prefaced terminal input just locks up and never (apparently) executes.
Maybe there is a clue here?
_______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
-- Roger Wells, P.E. leidos 221 Third St Newport, RI 02840 401-847-4210 (voice) 401-849-1585 (fax) roger.k.wells@xxxxxxxxxx
_______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
-- Roger Wells, P.E. leidos 221 Third St Newport, RI 02840 401-847-4210 (voice) 401-849-1585 (fax) roger.k.wells@xxxxxxxxxx
_______________________________________________ devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/ List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx
