[LKP] [xarray] 4e99d4e957: BUG:soft_lockup-CPU##stuck_for#s

Rong Chen rong.a.chen at intel.com
Wed Dec 19 18:26:52 PST 2018



On 12/19/2018 05:54 PM, Rong Chen wrote:
>
>
> On 12/11/2018 10:29 PM, Matthew Wilcox wrote:
>> On Tue, Dec 11, 2018 at 05:54:41PM +0800, kernel test robot wrote:
>>> # To reproduce,
>>> # 1) save job-script and this script (both are attached in 0day 
>>> report email)
>>> # 2) run this script with your compiled kernel and optional env 
>>> $INSTALL_MOD_PATH
>>>
>>> kernel=$1
>>>
>>> initrds=(
>>>     /osimage/yocto/yocto-tiny-i386-2016-04-22.cgz
>>>     /lkp/lkp/lkp-i386.cgz
>>>     /osimage/deps/debian-x86_64-2016-08-31.cgz/run-ipconfig.i386_2016-09-03.cgz 
>>>
>>> )
>>>
>>> HTTP_PREFIX=https://download.01.org/0day-ci/lkp-qemu
>>> wget --timestamping "${initrds[@]/#/$HTTP_PREFIX}"
>>>
>>> {
>>>     cat "${initrds[@]//*\//}"
>>>     [[ $INSTALL_MOD_PATH ]] && (
>>>         cd "$INSTALL_MOD_PATH"
>>>
>>>
>>> To reproduce:
>>>
>>>          git clone https://github.com/intel/lkp-tests.git
>>>          cd lkp-tests
>>>          bin/lkp qemu -k <bzImage> job-script # job-script is 
>>> attached in this email
>> When I do this, the only output I get is:
>>
>> make: Entering directory '/home/willy/kernel/lkp-tests/bin/event'
>> gcc    -c -o wakeup.o wakeup.c
>> gcc  -static -o wakeup wakeup.o
>> rm -f wakeup.o
>> strip wakeup
>> make: Leaving directory '/home/willy/kernel/lkp-tests/bin/event'
>>
>> Running under sh -x, I get as far as:
>>
>> + exec /home/willy/kernel/lkp-tests/lkp-exec/qemu -k 
>> ../idrext/.build-yocto/arch/x86/boot/bzImage job-script
>>
>> Editing that script to set -x, I get to:
>>
>> + /home/willy/kernel/lkp-tests/sbin/pack -f -a x86_64 lkp-src
>> make: Entering directory '/home/willy/kernel/lkp-tests/bin/event'
>> gcc    -c -o wakeup.o wakeup.c
>> gcc  -static -o wakeup wakeup.o
>> rm -f wakeup.o
>> strip wakeup
>> make: Leaving directory '/home/willy/kernel/lkp-tests/bin/event'
>> + exit
>>
>> which would seem to indicate your 'pack' command failed with a non-zero
>> exit code?  Adding set -x to _that_ script gives me:
>>
>> + find . -type d -exec chmod g+rwx '{}' ';'
>> + chmod -R g+rw etc lib root usr
>> + '[' -d /osimage/addon-x86_64 ']'
>> + gzip -n -9 /home/willy/.lkp/cache/lkp-x86_64.cpio
>> + mv -f /home/willy/.lkp/cache/lkp-x86_64.cpio.gz 
>> /home/willy/.lkp/cache/lkp-x86_64.cgz
>> + [[ -n '' ]]
>> + [[ -n '' ]]
>> + exit
>>
>> I'm using your top-of-tree lkp-tests, commit
>> e28c726dfe85e3e595ba4776a4eb92f2559d3ac0
>>
>> By the way, I took this kernel config and ran it under one of your other
>> test cases (the one which runs Trinity) and didn't see a boot problem,
>> so either it's something specific to this setup, or the problem got
>> fixed between this commit and current top-of-tree.
>
> Please accept my apologies for taking your time, the reproduce tool 
> was broken for a long time, and it's fixed now.
> if it can't be reproduced, here is the original 
> bzImage(https://download.01.org/0day-ci/lkp-qemu/fbc/4e99d4e9579d3b950bf4b38d0d64eb1b9be78761/).

Hi,

It should be a false positive too, The test VM are usually overloaded 
much to improve test density, we will pay attention to the issue.

Best Regards,
Rong Chen


More information about the LKP mailing list