change LAMMPS easyblock to check for kokkos_arch before generic optarch#3945
Conversation
kokkos_arch before optarch
kokkos_arch before optarchkokkos_arch before generic optarch
|
@boegelbot please test @ jsc-zen3 EB_ARGS= LAMMPS-2Aug2023_update2-foss-2023a-kokkos.eb |
|
@ocaisa: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3334314169 processed Message to humans: this is just bookkeeping information for me, |
Hm, that build silently failed on jsc-zen3. I'll try manually. |
|
@boegelbot please test @ jsc-zen3 EB_ARGS="LAMMPS-2Aug2023_update2-foss-2023a-kokkos.eb" |
|
@ocaisa: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de PR test command '
Test results coming soon (I hope)... Details- notification for comment with ID 3337046268 processed Message to humans: this is just bookkeeping information for me, |
Most likely, I copy-pasted a search result where it turns out that GitHub cleans out the quotation marks |
|
Test report by @boegelbot Overview of tested easyconfigs (in order)
Build succeeded for 1 out of 1 (1 easyconfigs in total) |
Signed-off-by: Jan Andre Reuter <[email protected]>
(created using
eb --new-pr)Follow-up to #3943.
Instead of allowing to specify
kokkos_archwhen we build with a generic optarch, causingARMV80to be used, move the precedence ofkokkos_archto be considered first. The workaround implemented in #3943 only applied foraarch64and included code duplication for the else branch coming directly afterwards.With this, one can also use
kokkos_archon x86 platforms to work around potential issues, even when working with a generic optarch.