Skip to content

unittests: don't powerdown node at the end#3309

Merged
miri64 merged 3 commits intoRIOT-OS:masterfrom
miri64:unittests/fix/stop
Jan 26, 2016
Merged

unittests: don't powerdown node at the end#3309
miri64 merged 3 commits intoRIOT-OS:masterfrom
miri64:unittests/fix/stop

Conversation

@miri64
Copy link
Copy Markdown
Member

@miri64 miri64 commented Jul 3, 2015

This currently only works for default output and OUTPUT=TEXT, but since we don't use OUTPUT=COMPILE and OUTPUT=XML anyways we should be fine.

@miri64 miri64 added Area: tests Area: tests and testing framework NSTF labels Jul 3, 2015
@miri64 miri64 added this to the Release 2015.06 milestone Jul 3, 2015
@miri64 miri64 added the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Jul 27, 2015
@OlegHahm OlegHahm removed the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Jul 28, 2015
@OlegHahm
Copy link
Copy Markdown
Member

What was the rationale for that?

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Aug 10, 2015

What was the rational for shutting it down? It's handy for native, but most other boards to my observence shut down before the output is complete.

@OlegHahm
Copy link
Copy Markdown
Member

Somewhere in the back of my head I remember that there was at some point already about this behavior. @LudwigOrtmann, @Kijewski, @kaspar030, do you recall anything? @phiros, @gebart, do you have an opinion?

@Kijewski
Copy link
Copy Markdown
Contributor

I don't recall, but git blames @authmillenon as the author of the shutdown. ;)

I am in favor of the PR, because the powerdown in here is quite arbitrary and forces you to implement LPM in order to run the test suite if I am not mistaken.

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Aug 11, 2015

@Kijewski I know that, I just don't know exactly why I did it any more... Probably out of inexperience :)

@OlegHahm OlegHahm modified the milestones: Release 2015.08, Release NEXT MAJOR Sep 2, 2015
@LudwigKnuepfer
Copy link
Copy Markdown
Member

Something about "handy with native" is the only thing I can come up with as well.

@OlegHahm OlegHahm assigned A-Paul and unassigned OlegHahm Nov 24, 2015
@OlegHahm
Copy link
Copy Markdown
Member

I have no opinion on this one, so I re-assign.

@OlegHahm OlegHahm modified the milestone: Release 2015.12 Dec 8, 2015
@miri64 miri64 added this to the Release 2016.03 milestone Jan 6, 2016
@miri64 miri64 added the Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties label Jan 23, 2016
@miri64 miri64 force-pushed the unittests/fix/stop branch from a5a8a50 to 4fb7939 Compare January 23, 2016 14:22
@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Jan 23, 2016

Rebased and adapted to current master.

@miri64 miri64 added the CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR label Jan 23, 2016
@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Jan 23, 2016

Note that at least for samr21-xpro and iotlab-m3 it seems that make reset isn't always working (and as such the output might not show up). But since I also tried it by hand I have the feeling this is an issue of make reset and not this script.

@A-Paul
Copy link
Copy Markdown
Member

A-Paul commented Jan 26, 2016

@authmillenon

as such the output might not show up). But since I also tried it by hand I have the feeling this is ...

When started 'by hand', reset also fails from time to time? Did I get this right?

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Jan 26, 2016

Yes, if it fails from time to time. This seems to be a problem with make reset, though.

@A-Paul
Copy link
Copy Markdown
Member

A-Paul commented Jan 26, 2016

I'm just asking because in an other test I experienced a significant impact by pexpect to the test result. It seems that foo.logfile = sys.stdout can cause a penalty to timing.

@A-Paul
Copy link
Copy Markdown
Member

A-Paul commented Jan 26, 2016

ACK!

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Jan 26, 2016

This seems to be an issue with the openocd version btw. On my arch machine I do not encounter this problem as much.

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Jan 26, 2016

I'm just asking because in an other test I experienced a significant impact by pexpect to the test result. It seems that foo.logfile = sys.stdout can cause a penalty to timing.

This can be.

@OlegHahm
Copy link
Copy Markdown
Member

Apparently this PR did more than the title says. @authmillenon, can you remember what was the exact problem with pexpect in Ubuntu?

@miri64
Copy link
Copy Markdown
Member Author

miri64 commented Feb 28, 2016

The processes in the children's process group will not be killed causing the older instances to be open in background.

@miri64 miri64 added the Area: network Area: Networking label Sep 30, 2018
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Area: network Area: Networking Area: tests Area: tests and testing framework CI: ready for build If set, CI server will compile all applications for all available boards for the labeled PR Community: Hack'n'ACK candidate This PR is a candidate for review and discussion during one of RIOT's monthly Hack'n'ACK parties

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants