Skip to content

Ignore iwlwifi, thermal errors on Dell laptop #343

Closed
Closed
Ignore iwlwifi, thermal errors on Dell laptop#343
Assignees
fredoh9
Labels
area:non-audio FailureFalse positives: failing when we don't want totype:bugSomething doesn't work as expected

Description

@fredoh9
fredoh9
opened on Aug 17, 2020

Need to filter out non audio errors in Dell laptop.

[    0.713554] tpm tpm0: tpm_try_transmit: send(): error -5
[    0.787321] platform regulatory.0: Direct firmware load for regulatory.db failed with error -2
[    0.787322] cfg80211: failed to load regulatory.db
[    1.281146] EXT4-fs (nvme0n1p6): re-mounted. Opts: errors=remount-ro
[    1.290704] usb 2-3: Enable of device-initiated U1 failed.
[    1.291010] usb 2-3: Enable of device-initiated U2 failed.
[    1.444017] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-56.ucode failed with error -2
[    1.444814] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-55.ucode failed with error -2
[    1.444863] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-54.ucode failed with error -2
[    1.444874] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-53.ucode failed with error -2
[    1.444882] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-52.ucode failed with error -2
[    1.444890] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-51.ucode failed with error -2
[    1.444898] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-50.ucode failed with error -2
[    1.444936] iwlwifi 0000:00:14.3: Direct firmware load for iwlwifi-QuZ-a0-hr-b0-49.ucode failed with error -2
[    1.446871] iwlwifi 0000:00:14.3: Direct firmware load for iwl-debug-yoyo.bin failed with error -2
[    1.631431] thermal thermal_zone10: failed to read out thermal zone (-61)

Activity

  1. fredoh9
    added
    type:bugSomething doesn't work as expected
    on Aug 17, 2020
  2. fredoh9
    self-assigned this
    on Aug 17, 2020

  3. fredoh9 commented on Aug 24, 2020

    @fredoh9
    fredoh9
    on Aug 24, 2020
    ContributorAuthor

    @plbossart
    For better filtering, I need sof-test logs for this, fail/errors in booting doesn't fail the test cases. Especially I need testing logs with

    • check-runtime-pm-status.sh
    • check-kmod-load-unload.sh

    Or let me allow SSH into the device once.

  4. plbossart commented on Aug 25, 2020

    @plbossart
    plbossart
    on Aug 25, 2020
    Member

    @fredoh9 the errors were copy/pasted from the sof-test logs. This is on a Dell CML-U w/ SoundWire which is NOT in CI for far.

  5. plbossart
    added 2 commits that reference this issue on Aug 27, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    984822e

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    366703f

  6. marc-hb commented on Aug 27, 2020

    @marc-hb
    marc-hb
    on Aug 27, 2020
    Collaborator

    The iwlwifi are warnings, not errors. We should really stop screen-scraping logs and use error levels some day.

  7. plbossart commented on Aug 27, 2020

    @plbossart
    plbossart
    on Aug 27, 2020
    Member

    The iwlwifi are warnings, not errors. We should really stop screen-scraping logs and use error levels some day.

    maybe, but this is real today.

    2020-08-26 22:04:29 UTC [ERROR] Caught dmesg error
===========================>>
Aug 26 17:00:21 plb-olympic kernel: [ 5456.059737] iwlwifi 0000:00:14.3: Microcode SW error detected. Restarting 0x0.
<<===========================
2020-08-26 22:04:29 UTC Sub-Test: [ERROR] Starting func_exit_handler(), exit status=1, FUNCNAME stack:
2020-08-26 22:04:29 UTC Sub-Test: [ERROR]  main()  @  ./multiple-pause-resume.sh
2020-08-26 22:04:30 UTC Sub-Test: [ERROR] sof-logger was already dead
2020-08-26 22:04:31 UTC Sub-Test: [INFO] Test Result: FAIL!
  8. plbossart
    added a commit that references this issue on Aug 27, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    ddea455

  9. plbossart commented on Aug 27, 2020

    @plbossart
    plbossart
    on Aug 27, 2020
    Member

    Actually @marc-hb you're right the filters are broken.

    We get an error on ICL RVP (and others) just because we will detect 'panic' in the following trace

    [  345.962670] sof-audio-pci 0000:00:1f.3: status = 0x00000000 panic = 0x00000000

    we really only want to track errors, then look for such keywords.
    Meh.

  10. plbossart commented on Aug 27, 2020

    @plbossart
    plbossart
    on Aug 27, 2020
    Member

    Can someone please explain to me what on earth this script is supposed to do. @fredoh9 @marc-hb @aiChaoSONG @xiulipan

    This script is causing false alarms in the kernel CI and needs to be fixed to use dmesg --level=err,warn

    begin_line=${1:-1}

# confirm begin_line is number, if it is not the number, direct using dmesg
[[ "${begin_line//[0-9]/}" ]] && begin_line=0
[[ "$begin_line" -eq 0 ]] && cmd="dmesg" || cmd="sed -n '$begin_line,\$p' /var/log/kern.log"

#echo "run $0 with parameter '$*' for check kernel message error"

if [ "$ignore_str" ]; then
    err=$(eval $cmd|grep 'Call Trace' -A5 -B3)$(eval $cmd | grep -E "$err_str"|grep -vE "$ignore_str")
else
    err=$(eval $cmd|grep 'Call Trace' -A5 -B3)$(eval $cmd | grep -E "$err_str")
fi

  11. marc-hb commented on Aug 28, 2020

    @marc-hb
    marc-hb
    on Aug 28, 2020
    Collaborator

    maybe, but this is real today.

    Agreed, happy to merge #343 except for the parentheses ( ) concern.

    Can someone please explain to me what on earth this script is supposed to do.

    The "begin_line" logic separates logs from consecutive tests.

    fixed to use dmesg --level=err,warn

    journalctl -p warning --since=... for two reasons:

    • no ring buffer exhaustion issue
    • user space errors matter too

    See initial attempts in PR #262. Bin-QA moved on but I believe @fredoh9 has been resuming this (he's on vacation now)

  12. plbossart
    added a commit that references this issue on Aug 28, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    fc264f1

  13. aiChaoSONG commented on Aug 28, 2020

    @aiChaoSONG
    aiChaoSONG
    on Aug 28, 2020

    @plbossart I saw your patch #354, I guess you already figured out what it does.

    • Before running a test case, record log point of /var/log/kern.log with func_lib_setup_kernel_last_line
    • After running the test case, collect log from the recorded log point with sof-kernel-log-check.sh "$KERNEL_LAST_LINE"

  14. plbossart commented on Aug 28, 2020

    @plbossart
    plbossart
    on Aug 28, 2020
    Member

    @plbossart I saw your patch #354, I guess you already figured out what it does.

    • Before running a test case, record log point of /var/log/kern.log with func_lib_setup_kernel_last_line
    • After running the test case, collect log from the recorded log point with sof-kernel-log-check.sh "$KERNEL_LAST_LINE"

    Yes, but this is not so good. We need to use a timestamp, that way we can filter with whatever level we want. Using the raw line number is a major limitation.

  15. plbossart
    added a commit that references this issue on Aug 28, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    7cfad1f
  16. marc-hb
    added a commit that references this issue on Sep 3, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    5d9e618

  17. marc-hb commented on Sep 4, 2020

    @marc-hb
    marc-hb
    on Sep 4, 2020
    Collaborator

    #354 was abandoned by @plbossart , I'm "rescuing" the filters part of it in #362 , please review.

  18. xiulipan
    added a commit that references this issue on Sep 8, 2020

    sof-kernel-log-check.sh: add Dell CML-U w/ SoundWire filters

    efe738f
  19. mengdonglin
    added
    area:non-audio FailureFalse positives: failing when we don't want to
    dmesg filter
    on Sep 22, 2020

  20. mengdonglin commented on Nov 2, 2020

    @mengdonglin
    mengdonglin
    on Nov 2, 2020
    Contributor

    #362 merged, close this issue

  21. mengdonglin
    closed this as completedon Nov 2, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

area:non-audio FailureFalse positives: failing when we don't want totype:bugSomething doesn't work as expected

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions