From: "Leonardo Brás" <leobras.c@gmail.com>
To: paulmck@kernel.org, Akira Yokosawa <akiyks@gmail.com>
Cc: perfbook@vger.kernel.org
Subject: Re: [PATCH v1] utilities/gitlab-ci: Change base machine type to 2-cpu to fix make check
Date: Tue, 07 Feb 2023 00:34:51 -0300 [thread overview]
Message-ID: <5a0ca273c8772e9854f836343816840e1ca31f00.camel@gmail.com> (raw)
In-Reply-To: <20230207011740.GV2948950@paulmck-ThinkPad-P17-Gen-1>
On Mon, 2023-02-06 at 17:17 -0800, Paul E. McKenney wrote:
> On Tue, Feb 07, 2023 at 09:19:09AM +0900, Akira Yokosawa wrote:
> > On Mon, 6 Feb 2023 17:55:15 -0300, Leonardo Bras wrote:
> > > `make -C CodeSamples/ check` currently fails on Gitlab-CI.
> > >
> > > Failure happen on sched_setaffinity:
> > > sched_setaffinity: Invalid argument
> > >
> > > Further debuging, it's possible to notice that some tests are trying to set
> > > the cpu affinity in CPUs 0 and 1, so at least 2 cpus are needed for the
> > > test to run properly.
> > >
> > > Current (default) saas-linux-small-amd64 instances used for testing on
> > > Gitlab-CI only offer a single CPU, which is not enough.
> > >
> > > For the test_codesamples pre-test, change the instance type to
> > > saas-linux-medium-amd64 so the tests can run properly.
> > >
> > > Fixes: 9217963c ("utilities/gitlab-ci.yml: Add 'make check' on
> > > test_codesamples")
> > > Signed-off-by: Leonardo Bras <leobras.c@gmail.com>
> >
> > Hi,
> >
> > I see that this change worked at:
> >
> > https://gitlab.com/linux-kernel/perfbook/-/jobs/3723917786
> >
> > So FWIW,
> >
> > Reviewed-by: Akira Yokosawa <akiyks@gmail.com>
>
> Queued and pushed, thank you both!
>
> > I don't have a list of minimum CPU count for programs under
> > CodeSamples/, but I guess there are a few which need more than
> > two CPUs.
> >
> > You might need to add conditionals in Makefile to exclude those
> > tests depending on available CPU count.
Sure,
I think my project group (linux-kernel) has an Ultimate subscription for being
open-source (I recall filling-in a form and getting for that in the past).
So, if it's up to 4 cpus, I think we can manage to test on the Gitlab-ci.
>
> In addition, some of the tests may need special command-line arguments.
> For example, most of the tests that take "--smoketest" probably want to
> use that argument for "make check".
For directories with those cases, we could write a script for calling every test
with their command-line arguments, and then call it from the directory's
Makefile.
>
> One thought is to take this one directory at a time. Any better ideas?
>
> Thanx, Paul
That works for me!
Every directory could have it's own `make check`, and we could call them from
the CodeSamples directory's `make check`
Thanks,
Leo
>
> > Thanks, Akira
> >
> > > ---
> > > utilities/gitlab-ci.yml | 1 +
> > > 1 file changed, 1 insertion(+)
> > >
> > > diff --git a/utilities/gitlab-ci.yml b/utilities/gitlab-ci.yml
> > > index 9545029e..42eda60f 100644
> > > --- a/utilities/gitlab-ci.yml
> > > +++ b/utilities/gitlab-ci.yml
> > > @@ -5,6 +5,7 @@ stages:
> > > - build
> > >
> > > test_codesamples:
> > > + tags: [ saas-linux-medium-amd64 ]
> > > stage: test
> > >
> > > before_script:
prev parent reply other threads:[~2023-02-07 3:34 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-06 20:55 [PATCH v1] utilities/gitlab-ci: Change base machine type to 2-cpu to fix make check Leonardo Bras
2023-02-07 0:19 ` Akira Yokosawa
2023-02-07 1:17 ` Paul E. McKenney
2023-02-07 3:34 ` Leonardo Brás [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5a0ca273c8772e9854f836343816840e1ca31f00.camel@gmail.com \
--to=leobras.c@gmail.com \
--cc=akiyks@gmail.com \
--cc=paulmck@kernel.org \
--cc=perfbook@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).