RFE: Transition from TravisCI to Github Actions #329
Conversation
Disable Travis CI by deleting the .travis.yml file. Subsequent commits will enable the Github Actions continuous integration. Signed-off-by: Tom Hromatka <tom.hromatka@oracle.com>
Delete the unused variable 'len' from scmp_bpf_disasm.
scan-build identified the following two warnings:
scmp_bpf_disasm.c:304:10: warning: Although the value stored to 'len'
is used in the enclosing expression, the value is never actually read
from 'len'
while ((len = fread(&bpf, sizeof(bpf), 1, file))) {
scmp_bpf_disasm.c:441:10: warning: Although the value stored to 'len' is
used in the enclosing expression, the value is never actually read from
'len'
while ((len = fread(&bpf, sizeof(bpf), 1, file))) {
Signed-off-by: Tom Hromatka <tom.hromatka@oracle.com>
drakenclimber
changed the title
|
I haven't looked at the changes/commits yet, but +1000 on the ultra snazzy PR description! |
|
@drakenclimber out of curiosity, what is the code coverage percentage with the GH Actions approach? |
I added a link to each in the table above. |
drakenclimber
force-pushed
the
issues/githubactions
branch
from
76ce54c
to
c1c8b73
Compare
Add Github Actions workflow and actions to run the automated libseccomp tests and gather code coverage metrics. Signed-off-by: Tom Hromatka <tom.hromatka@oracle.com>
Travis CI has now been disabled. Delete the Travis CI build status badge and display the Github Actions continuous integration workflow badge. Signed-off-by: Tom Hromatka <tom.hromatka@oracle.com>
drakenclimber
force-pushed
the
issues/githubactions
branch
from
c1c8b73
to
5f9d9df
Compare
Arrrgh less than 90%?! Several years ago I spent an absurd amount of time one year getting the coverage above 90%, seeing it drop below 90% is a sad moment :) Although most of that was due to fixing various small bugs that cropped up during the additional testing so I can't be too upset. Although looking at the coverage data there are definitely a few things that look .... off. For example, the file list is definitely not correct (files missing, others added that shouldn't be there, etc.). Regardless, that's all something we can fix later. |
|
Merged at HEAD 04245d9. I'm thinking that maybe we keep this issue open until we sort out some of the remaining code coverage issues and backport everything to the release-2.5 branch, or should we close this out and create a new issue for that - thoughts? |
|
With this merged into the main branch and PR #331 backporting this to the release-2.5 branch I think we can close this out now, any remaining issues should be treated as new issues or handled in the backport PR. |
This pull request resolves Issue #299 and Issue #312.
This patchset transitions the libseccomp continuous integration from TravisCI to Github Actions. Here's a brief rundown of the pros/cons of each CI system:
1 It seems that the TravisCI VMs are more powerful than the Github Actions VMs, but Github Actions makes parallelization much easier. Thus the nontrivial run time differences.
2 There are a few ways to add support for additional architectures to Github Actions via either self-hosted runners or possibly this Docker container solution
3 The code coverage numbers differ slightly from the TravisCI implementation to the Github Actions implementation. I think the Github Actions numbers are more accurate, but I haven't dug too deeply. (Also, I can't get Github Actions code coverage to ignore
arch-syscall-check.c. It's kinda bugging me :).)