kernel: update to 4.19.83

[egernst]

Fixes: 1816

Signed-off-by: Eric Ernst eric.ernst@intel.com

[grahamwhaley]

Your 'Fixes' has no leading # on the n, so fails the CI.
Also, that is not really a useful commit message - the git log will contain no useful explanation.
Having both the version bump and the code change in the same commit (especially without any explanatory text) looks strange as well.

[jodh-intel]

Also, #1816 was closed in September, so please can you raise a new issue and provide details of why this change is needed?

[egernst]

Pragmatism and laziness. Do we want a duplicate GitHub issue for bumps each time? The rationale for doing a bump should be very clear. If there are specific CVEs we open them seperately. WDYT?

[jodh-intel]

I personally still vote for discrete issues for each bump as referring to a closed issue is confusing.

The laziness factor can be handled by using hub:

$ echo -e "I am the issue title\n\nThis is the body of the description.\nMore body\nHub is cool" | hub issue create -F -

[egernst]

/test

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@552e940). Click here to learn what that means.
The diff coverage is n/a.

@@            Coverage Diff            @@
##             master    #2185   +/-   ##
=========================================
  Coverage          ?   50.72%           
=========================================
  Files             ?      111           
  Lines             ?    17370           
  Branches          ?        0           
=========================================
  Hits              ?     8811           
  Misses            ?     7468           
  Partials          ?     1091

[egernst]

Looks like ARM patches aren't applying.

@jongwu @Pennyzct can you PTAL?

[gnawux]

@Pennyzct your nightmare unfortunately 😿

[Pennyzct]

@egernst We will update asap.😊
@gnawux I'm REALLY LOOK FORWARD TO see the next LTS version. 😭

[jodh-intel]

Approving, but I do think it would be extremely useful for tracking purposes to specify in the commit message why this bump was done (CVE's? features? magic flying unicorns 🦄? That kinda stuff 😄).

[Pennyzct]

Hi @egernst plz bump the version to v4.19.84. It's the latest and will decrease the failed chunks.
I've raised the PR to do the update, and you could add the Depends-on: github.com/kata-containers/packaging#807 to see if ARM CI is working well again. ;)

[egernst]

/test

[amshinde]

@egernst CI does not seem to be very happy.

[egernst]

@Pennyzct ^ still unhappy with patches. Sorry.

[Pennyzct]

Hi @egernst
So sorry for the delay.
I have updated my PR to be also v4.19.86. And ARM CI is happy now. ;)

[egernst]

/test

[Pennyzct]

As to the failure on ARM CI:

not ok 1 Expose IP Address

I've addressed related issue kata-containers/ci#229 to fix, and it should be irrelevant to this patch. ;).

[egernst]

rerunning virtio-fs test. If it passes, we should be good to merge.