Spencer Bryngelson
Spencer Bryngelson
closed due to inactivity
Mode of failure using nvhpc22.9 on Bridges2 occurs at runtime (well, test-time): ```console [I]v025: bryngel-startup-proj/MFC $ ./mfc.sh test -j 1 --gpu --no-build -- -c bridges2 mfc: OK > (venv) Entered...
Actually, this also fails (Rogues Gallery, Violet1, A100): `nvhpc-hpcx-cuda11/23.7` with two-pass optimization via the same failure mode (passes with two-pass IPO set to false).
nvhpc23.11 works with two-pass IPO
This issue can be resolved by updating modules on common HPC systems like Bridges2 and Delta and disabling IPO/LTO if the detected NVHPC version is less than 23.11
I added @anandrdbz who knows the QBMM routines well
Updated for @wilfonba and @anandrdbz etc. Here's a Frontier run ```shell Test pre_process and simulation | from EA8FA07E to F8ADA51B (447 tests) tests/UUID (s) Summary 8A59E8E6 38.38 1D -> bc=-2...
The problem is not that the subroutines in some cases are necessarily slower than they should be. Though this is possible, it is an orthogonal issue that I think is...
A singularity container seems like a good idea, though there would have to be containers for GPU cases, CPU cases, as well as case-optimization of the benchmark case (in case...
Here is code for a docker container that works on my Mac: ```sh # syntax=docker/dockerfile:1 ############################################################ # MFC (Multiphase Flow Code) — portable CPU/GPU Dockerfile # ############################################################ # • Ubuntu ...