Bump ROCm Patch Version

README.md

@@ -40,9 +40,9 @@ The model is trained from a random initialization until convergence, which is de
         1. `ml cuda/12.9.1 gcc/13.3.1 mvapich2/2.3.7`
         1. `export LD_LIBRARY_PATH=/usr/lib64:$LD_LIBRARY_PATH`
     - ROCm (elcap):
-        1. `ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi`
+        1. `ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi`
             - If using WCI wheel:
-                1. `export LD_PRELOAD=/opt/rocm-7.1.0/llvm/lib/libomp.so` # for libomp.so
+                1. `export LD_PRELOAD=/opt/rocm-7.1.1/llvm/lib/libomp.so` # for libomp.so
 
 1. Install the benchmark in the python venv:
     - CUDA: `pip install --no-binary=mpi4py .[cuda] --prefix=.venvs/scaffoldvenv --extra-index-url https://download.pytorch.org/whl/cu129 2>&1 | tee install.log`
@@ -226,7 +226,7 @@ make && make install
 git clone https://github.com/LLNL/Caliper.git
 cd Caliper
 mkdir pybuild && cd pybuild
-ml rocm/7.1.0
+ml rocm/7.1.1
 ml cuda/12.9.1
 cmake -DWITH_PYTHON_BINDINGS=ON \
    -DWITH_ROCPROFILER=ON \

ScaFFold/utils/create_restart_script.py

@@ -98,7 +98,7 @@ def _get_env_setup() -> str:
 # --- Begin Environment Setup ---
 # Load Modules
 if command -v module &> /dev/null; then
-    ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
+    ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi
 fi
 
 # Activate Virtual Environment
@@ -111,7 +111,7 @@ def _get_env_setup() -> str:
 
 # Environment variables
 export SPINDLE_FLUXOPT=off
-export LD_PRELOAD=/opt/rocm-7.1.0/llvm/lib/libomp.so
+export LD_PRELOAD=/opt/rocm-7.1.1/llvm/lib/libomp.so
 
 export PROFILE_TORCH=ON
 # --- End Environment Setup ---

scripts/install-rccl.sh

@@ -6,7 +6,7 @@ if [ -d "aws-ofi-nccl.git" ]; then
   return 1 2>/dev/null || exit 1
 fi
 
-rocm_version=7.1.0
+rocm_version=7.1.1
 
 module swap PrgEnv-cray PrgEnv-gnu
 module load rocm/$rocm_version

scripts/install-tuolumne-torchpypi.sh

@@ -1,4 +1,4 @@
 . install-rccl.sh
 ml load python/3.11.5 && python3 -m venv .venvs/scaffoldvenv-tuo-pypi && source .venvs/scaffoldvenv-tuo-pypi/bin/activate && pip install --upgrade pip
-ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
+ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi
 pip install -e .[rocm] --prefix=.venvs/scaffoldvenv-tuo-pypi --extra-index-url https://download.pytorch.org/whl/rocm7.1 2>&1 | tee install.log

scripts/install-tuolumne.sh

@@ -1,5 +1,5 @@
 ml load python/3.11.5 && python3 -m venv .venvs/scaffoldvenv-tuo && source .venvs/scaffoldvenv-tuo/bin/activate && pip install --upgrade pip
-ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
+ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi
 pip install -e .[rocmwci] --prefix=.venvs/scaffoldvenv-tuo 2>&1 | tee install.log
 # Needed until new wheel exists for torch using mpich 9.1.0
 TORCH_LIB_DIR=".venvs/scaffoldvenv-tuo/lib/python3.11/site-packages/torch/lib"

scripts/scaffold-tuolumne-torchpypi.job

@@ -7,7 +7,7 @@
 # flux: -qpdebug
 # flux: -B fractale
 
-ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
+ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi
 
 . .venvs/scaffoldvenv-tuo-pypi/bin/activate
 

scripts/scaffold-tuolumne.job

@@ -7,13 +7,13 @@
 # flux: -qpdebug
 # flux: -B fractale
 
-ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.0 rccl/fast-env-slows-mpi
+ml cce/21.0.0 cray-mpich/9.1.0 rocm/7.1.1 rccl/fast-env-slows-mpi
 
 . .venvs/scaffoldvenv-tuo/bin/activate
 
 # (1) Avoid libmagma error
 # (2) Removing libmpi may cause segfault on mpi4py import
-export LD_PRELOAD="/opt/rocm-7.1.0/llvm/lib/libomp.so /opt/cray/pe/mpich/9.1.0/ofi/gnu/11.2/lib/libmpi_gnu.so.12"
+export LD_PRELOAD="/opt/rocm-7.1.1/llvm/lib/libomp.so /opt/cray/pe/mpich/9.1.0/ofi/gnu/11.2/lib/libmpi_gnu.so.12"
 
 torchrun-hpc -N 1 -n 1 $(which scaffold) generate_fractals -c $(pwd)/ScaFFold/configs/benchmark_default.yml