Skip to content

[GCC/4.8.4] goolf/1.7.20 & gompi/1.7.20 (REVIEW)#1294

Merged
boegel merged 11 commits intoeasybuilders:developfrom
fgeorgatos:contrib_goolf-1.6.20
Feb 10, 2015
Merged

[GCC/4.8.4] goolf/1.7.20 & gompi/1.7.20 (REVIEW)#1294
boegel merged 11 commits intoeasybuilders:developfrom
fgeorgatos:contrib_goolf-1.6.20

Conversation

@fgeorgatos
Copy link
Copy Markdown
Contributor

@fgeorgatos fgeorgatos commented Jan 11, 2015

highlights:

  • using GCC/4.8.4, for maximum compatibility with 3rd party tools (debuggers, CUDA, intel* [1] [2] [3]... etc)
  • most recent/updated hwloc, openmpi, openblas
  • openblas over gcc w. NO_AFFINITY=1 (leave the mess for the resource manager & apps to handle :)

As others, I cannot fully interpret the effect of the last point, until it is tried across several systems...
so, provided "AS-IS" :)

[1] easybuilders/easybuild-framework#635 (comment)
[2] http://docs.nvidia.com/cuda/cuda-getting-started-guide-for-linux/index.html#axzz3RM915wwL
[3] http://docs.roguewave.com/totalview/8.14.1/html/index.html#page/General_TV_Documentation/totalviewpl-platforms.html
[4] http://nl.mathworks.com/support/compilers/R2014b/index.html?sec=glnxa64 # 2015b should be OK
[5] https://www.pgroup.com/products/pgiworkstation.htm
[6] http://docs.cray.com/books/S-9408-1409//S-9408-1409.pdf # N.B. CUDA/5.5 vs GCC/4.9.x
[7] https://openfoamwiki.net/index.php/Installation/Compatibility_Matrix
[8] http://www.reddit.com/r/linux/comments/2br67o/gcc_49_is_doing_some_seriously_crazy_shit

Also, from Installation Guide of Intel Parallel Studio XE 2015:

Linux Developer tools component installed, including gcc, g++ and related tools. (this is
the list of component versions tested by Intel; other versions may or may not work and
are not recommended - please refer to Technical Support if you have questions
o gcc 4.1-4.8
o binutils 2.17-2.24

ref. https://software.intel.com/sites/default/files/managed/14/61/Release_Notes_Studio_XE_2015_L_u3.pdf

@hpcugentbot
Copy link
Copy Markdown

Test PASSed.

@fgeorgatos fgeorgatos changed the title Contrib goolf/1.6.20, over gompi/1.6.20 [GCC/4.8.4] Contrib goolf/1.6.20, over gompi/1.6.20 (REVIEW) Jan 20, 2015
@fgeorgatos fgeorgatos changed the title [GCC/4.8.4] Contrib goolf/1.6.20, over gompi/1.6.20 (REVIEW) [GCC/4.8.4] goolf/1.6.20 & gompi/1.6.20 (REVIEW) Jan 20, 2015
@fgeorgatos fgeorgatos mentioned this pull request Jan 20, 2015
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

no need for this trick here, since comp is only used in one place, replace use of comp with ('GCC', '4.8.4')

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@boegel: done!

@boegel
Copy link
Copy Markdown
Member

boegel commented Jan 20, 2015

@fgeorgatos: tiny remarks, let's see a test report for this from your end?

@fgeorgatos
Copy link
Copy Markdown
Contributor Author

Test report by @fgeorgatos
SUCCESS
Build succeeded for 9 out of 9
Linux debian 6.0.10, Intel(R) Xeon(R) CPU L5640 @ 2.27GHz, Python 2.6.6
See https://gist.github.com/2a996d6b3925f8a775a5 for a full test report.

@boegel
Copy link
Copy Markdown
Member

boegel commented Jan 20, 2015

Test report by @boegel
SUCCESS
Build succeeded for 9 out of 9 (9 easyconfigs in this PR)
Linux SL 6.6, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/579a1e0885f965a607c2 for a full test report.

@hpcugentbot
Copy link
Copy Markdown

Test PASSed.

Signed-off-by: Fotis Georgatos <[email protected]>
@fgeorgatos
Copy link
Copy Markdown
Contributor Author

@boegel: in the meantime, I realised I could bump up the version of numactl at 2.0.10, which I did, via commit d7bb898.

I think that now this toolchain PR is more complete.
Also, I come to realise that toolchain components could benefit from some extra testing (building and running packages atop) which I admit to have not done at the level desired, yet. YMMV.

And, I also got hit by the numactl upstream tarball change, this month:
(do we have now two independent confirmations, to incriminate the source? :-)

$ md5sum ~/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz*682c38305b2596967881f3d77bc3fc9c  /home/users/fgeorgatos/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz
58443c6fe7e67126de07bd86e4d6b825  /home/users/fgeorgatos/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz.OLD
$ ll ~/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz*
-rw-r--r-- 1 fgeorgatos clusterusers  99040 Jan 23 09:10 /home/users/fgeorgatos/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz
-rw-r--r-- 1 fgeorgatos clusterusers 402082 Jan 11 00:13 /home/users/fgeorgatos/.local/easybuild/sources/n/numactl/numactl-2.0.10.tar.gz.OLD

@hpcugentbot
Copy link
Copy Markdown

Test PASSed.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

just use OpenMPI here directly

@hpcugentbot
Copy link
Copy Markdown

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/2299/
Test PASSed.

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

below sources please

@hpcugentbot
Copy link
Copy Markdown

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/2300/
Test PASSed.

@fgeorgatos
Copy link
Copy Markdown
Contributor Author

Test report by @fgeorgatos
SUCCESS
Build succeeded for 13 out of 13
Linux debian 6.0.10, Intel(R) Xeon(R) CPU L5640 @ 2.27GHz, Python 2.6.6
See https://gist.github.com/720bdb861d3837d094b4 for a full test report.

@fgeorgatos
Copy link
Copy Markdown
Contributor Author

Test report by @fgeorgatos
SUCCESS
Build succeeded for 13 out of 13
Linux debian 6.0.10, Intel(R) Xeon(R) CPU L5640 @ 2.27GHz, Python 2.6.6
See https://gist.github.com/ad03bed9b335e825e5f8 for a full test report.

@boegel
Copy link
Copy Markdown
Member

boegel commented Feb 10, 2015

Test report by @boegel
FAILED
Build succeeded for 9 out of 13 (13 easyconfigs in this PR)
Linux SL 6.6, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/7674e204a554d7088222 for a full test report.

@fgeorgatos fgeorgatos changed the title [GCC/4.8.4] goolf/1.6.20 & gompi/1.6.20 (REVIEW) [GCC/4.8.4] goolf/1.7.20 & gompi/1.7.20 (REVIEW) Feb 10, 2015
Signed-off-by: Fotis Georgatos <[email protected]>
@hpcugentbot
Copy link
Copy Markdown

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/2302/
Test PASSed.

@boegel
Copy link
Copy Markdown
Member

boegel commented Feb 10, 2015

Test report by @boegel
SUCCESS
Build succeeded for 13 out of 13 (13 easyconfigs in this PR)
Linux SL 6.6, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/9f409059c5c2018350c6 for a full test report.

boegel added a commit that referenced this pull request Feb 10, 2015
[GCC/4.8.4] goolf/1.7.20 & gompi/1.7.20 (REVIEW)
@boegel boegel merged commit 261362a into easybuilders:develop Feb 10, 2015
@boegel
Copy link
Copy Markdown
Member

boegel commented Feb 10, 2015

Thanks @fgeorgatos!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants