Kenneth Hoste
Kenneth Hoste
When #241 and #260 are merged, the `bot/check-results.sh` script should be refined so it doesn't incorrectly conclude that a single `No missing installations` message in the output implies that there...
We should look into providing an `eessi-init` command that makes it easier/less awkward to initialize the EESSI environment. By default, it should simply run `source source /cvmfs/pilot.eessi-hpc.org/latest/init/bash`, but it could...
Some software versions are getting a bit old, so we should update to a newer toolchain. `2020b` would be a good candidate (lots of easyconfigs available), but perhaps `2021a` or...
I ran into the following problem when building `libfabric-1.12.1-GCCcore-10.3.0.eb` (in the 2021.12 pilot) for the `x86_64/generic` target: ``` In file included from /cvmfs/pilot.eessi-hpc.org/versions/2021.12/software/linux/x86_64/generic/software/GCCcore/10.3.0/lib/gcc/x86_64-pc-linux-gnu/10.3.0/include/immintrin.h:51, from prov/psm3/psm3/opa/opa_dwordcpy-x86_64.c:57: /cvmfs/pilot.eessi-hpc.org/versions/2021.12/software/linux/x86_64/generic/software/GCCcore/10.3.0/lib/gcc/x86_64-pc-linux-gnu/10.3.0/include/avxintrin.h:920:1: error: inlining failed in...
The `archspec.cpu.host` function was changed in archspec 0.1.3 (see https://github.com/archspec/archspec/pull/53) to return the best matching generic CPU microarchitecture level if there's no exact match. These levels are values like `x86-64-v4`,...
While installing `Perl/5.30.2-GCCcore-9.3.0` to `/cvmfs/pilot.eessi-hpc.org/versions/2021.12/software/linux/x86_64/intel/skylake_avx512/software/Perl/5.30.2-GCCcore-9.3.0`, I ran into the following issue: ``` failing sanity check for 'Inline' extension: command "perldoc -lm Inline " failed; output: /bin/bash: /cvmfs/pilot.eessi-hpc.org/versions/2021.12/software/linux/x86_64/intel/skylake_avx512/software/Perl/5.30.2-GCCcore-9.3.0/bin/perldoc: /cvmfs/pilot.eessi-hpc.org/versions/2021.12/software/linux/x86_64/intel/skylake_avx512/software/Perl/5.30.2-GCCcore-9.3.0/bin: bad interpreter:...
``` FAIL: opal_path_nfs =================== Failure : Mismatch: input "/vscmnt/leuven_icts", expected:0 got:1 Failure : Mismatch: input "/vscmnt/brussel_hydra_home", expected:0 got:1 ``` This looks very much like https://github.com/open-mpi/ompi/issues/318 I was able to work...
We added ReFrame to the 2021.06 compat layer (cfr. https://github.com/EESSI/gentoo-overlay/pull/41), and now the installation of ReFrame in the 2021.06 software layer is failing: ``` == 2021-08-26 14:33:05,527 build_log.py:169 ERROR EasyBuild...
We're currently always using a single node to build the software layer, it would be nice to have support for leveraging a pool of available nodes, which should significantly speed...