All of lore.kernel.org
 help / color / mirror / Atom feed
From: Randy MacLeod <randy.macleod@windriver.com>
To: Paul Barker <paul@paulbarker.me.uk>
Cc: Yocto discussion list <yocto@yoctoproject.org>
Subject: Re: Python3 ptest and unittest
Date: Sun, 23 Feb 2014 16:50:47 -0500	[thread overview]
Message-ID: <530A6D37.3040606@windriver.com> (raw)
In-Reply-To: <CANyK_8cSbbgkbwojmpmUzM6-TzgTg7q7bpnuEvj0bvOgP74s0Q@mail.gmail.com>

On 14-02-22 08:07 PM, Paul Barker wrote:
> On 21 February 2014 16:08, Randy MacLeod <randy.macleod@windriver.com> wrote:
>> On 14-02-18 04:27 PM, Paul Barker wrote:
>>>
>>> I've just thrown together a couple of things which may be useful.
>>> They're currently slightly hackish but I could improve them and share
>>> them/submit them as patches if wanted:
>>>
>>> 1) I've wrote a custom python test runner which runs test suites and
>>> outputs the format expected by ptest natively instead of needing the
>>> sed magic in
>>> "openembedded-core/meta/recipes-devtools/python/python/run-ptest".
>>> I'm using it in my own project, might move the opkg test suite over to
>>> it if I have time, and it might be useful for any other python test
>>> suites. It's <50 lines of python :)
>>>
>>> 2) I've wrote a script which patches this test runner into python3's
>>> own testsuite then runs the suite. You don't even to patch the
>>> Makefile from the python source tree and install it (as in the python
>>> recipe in openembedded-core). It should run on anything with python3
>>> installed with the python standard library (as the standard library
>>> already includes all the tests). This may be a good option for adding
>>> ptest support to the python3 recipe - it'd just be a single 50-line
>>> 'run-ptest' script written in python.
>>>
>>> Does that sound interesting to anyone else?
>>
>> Yes, this seems to be very useful. Can you send a patch for review?
>>
>
> I've never worked with the ptest system before but I'm giving it a
> go... I'm mostly relying on https://wiki.yoctoproject.org/wiki/Ptest
> for info.
>
> It does look like the python3 test suite crashes with out-of-memory
> errors on qemux86 with the default configuration of 256 MB RAM. I've
> bumped it up to 1 GB and it seems to be working.

I've seen similar problems and dealt with it the same way.
Some tests deliberately try to cause an OOM and I assert that
they don't belong in ptest runs.

> Is there any way to
> note this in the ptest system so that this test suite is skipped if
> there isn't enough memory?

Not that I'm aware of. Add to docs?

>
> Also, I'm getting several failures within the testsuite which I'm not
> seeing when I run it on my desktop. Do we keep data on whether each
> ptest package is expected to pass or has known failures?

Not that I know of. We should.

That's something that I've been meaning to do but I won't get to it
for until another project is done so don't wait for me. :)

../Randy

>
> Cheers,
>


-- 
# Randy MacLeod. SMTS, Linux, Wind River
Direct: 613.963.1350


  reply	other threads:[~2014-02-23 21:50 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-18 21:27 Python3 ptest and unittest Paul Barker
2014-02-21 16:08 ` Randy MacLeod
2014-02-23  1:07   ` Paul Barker
2014-02-23 21:50     ` Randy MacLeod [this message]
2014-03-07  1:18       ` Paul Barker

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=530A6D37.3040606@windriver.com \
    --to=randy.macleod@windriver.com \
    --cc=paul@paulbarker.me.uk \
    --cc=yocto@yoctoproject.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.