Skip to content

hwloc: do not build hwloc CUDA support if --without-cuda used (and also always disable hwloc GL and OpenCL support) #4245

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Sep 21, 2017

Conversation

jsquyres
Copy link
Member

Fixes #4219.

@bgoglin FYI

Signed-off-by: Jeff Squyres [email protected]

@hppritcha This might be worth putting into v2.0.x...?

jsquyres and others added 2 commits September 21, 2017 08:24
Open MPI doesn't use GL or OpenCL OS devices, so just disable them in
hwloc.

Signed-off-by: Jeff Squyres <[email protected]>
@jsquyres jsquyres force-pushed the pr/disable-hwloc-cuda branch from 3f4cb5f to 84a721d Compare September 21, 2017 15:26
@jsquyres jsquyres changed the title hwloc: do not build hwloc CUDA support if --without-cuda used hwloc: do not build hwloc CUDA support if --without-cuda used (and also always disable hwloc GL and OpenCL support) Sep 21, 2017
@jsquyres
Copy link
Member Author

jsquyres commented Sep 21, 2017

@hjelmn @bwbarrett Looks like the OS X CI test failed, but there's no stdout/stderr to indicate why. Here's the output logs: https://jenkins.open-mpi.org/jenkins/job/open-mpi.build.platforms/1453/Platform=os-x/console

It just has:

3. Running autotools on top-level tree

==> Remove stale files
==> Writing m4 file with autogen.pl results
==> Running autoreconf
Running: autoreconf -ivf --warnings=all,no-obsolete,no-override -I config
autoreconf: Entering directory `.'
autoreconf: configure.ac: not using Gettext
autoreconf: running: aclocal -I config --force --warnings=all,no-obsolete,no-override -I ./config
FATAL: command execution failed
...Jenkins backtrace...

Any idea what happened here?

@jsquyres
Copy link
Member Author

Just for the helluvit, do the CS thing and try to make it fail again, just to see if it was some kind of transient / networking error to the OS testing box...

bot:ompi:retest

@bwbarrett
Copy link
Member

The root cause was:

Caused: java.io.IOException: Backing channel 'JNLP4-connect connection from c-67-160-8-108.hsd1.wa.comcast.net/67.160.8.108:17678' is disconnected.

Looks like the connection between the builder and the controller died. Not sure why this didn't auto-retry; add that to my todo list.

@jsquyres
Copy link
Member Author

@bwbarrett Ah, thanks for the explanation -- I missed that in the backtrace.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants