Add Hermitian matrix tests

.gitignore

@@ -1,2 +1,4 @@
 *.x
 magma/
+.vscode/*
+*.code-workspace

README.md

@@ -1,5 +1,18 @@
 # Notes on testing MAGMA
 
+This is a test program for comparing the performance of common GPU eigensolvers.
+ - Flag `-DCUDA`: uses `cusolver`
+ - Flag `-DHIP`: uses `rocsolver`
+ - Flag `-DMAGMA`: uses the `MAGMA` library. Also needs either `-DCUDA` or `-DHIP`.
+
+Works with symmetric and Hermitian matrices.
+
+Needs compiler with C++17 support.
+
+## Summary of test results (LUMI/Mahti)
+
+![eigsolvers_cdouble](https://github.com/user-attachments/assets/a272d929-292e-4550-bbf5-abec422b6925)
+
 ## LUMI / MI250x (1 GCD)
 
 ### Installing MAGMA
@@ -23,10 +36,10 @@ ml partition/G
 ml rocm/6.0.3
 ml magma/2.8.0-cpeGNU-24.03-rocm
 
-hipcc -std=c++14 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.0.3.x
+hipcc -std=c++17 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.0.3.x -Wno-unused-result
 sbatch --partition=dev-g --nodes=1 --ntasks-per-node=1 --cpus-per-task=1 --gpus-per-node=1 --time=01:00:00 -o 'rocm6.0.3.out' --wrap='./rocm6.0.3.x 3,100,200,400,800,1600'
 
-hipcc -std=c++14 --offload-arch=gfx90a -O3 -DMAGMA -DHIP -lmagma eigh.cpp -o magma2.8.0_rocm6.0.3.x
+hipcc -std=c++17 --offload-arch=gfx90a -O3 -DMAGMA -DHIP -lmagma eigh.cpp -o magma2.8.0_rocm6.0.3.x -Wno-unused-result
 sbatch --partition=dev-g --nodes=1 --ntasks-per-node=1 --cpus-per-task=1 --gpus-per-node=1 --time=01:00:00 -o 'magma2.8.0_rocm6.0.3.out' --wrap='./magma2.8.0_rocm6.0.3.x 3,100,200,400,800,1600,3200,6400,12800'
 ```
 
@@ -37,7 +50,7 @@ ml LUMI/24.03
 ml partition/G
 ml rocm/6.2.2
 
-hipcc -std=c++14 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.2.2.x
+hipcc -std=c++17 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.2.2.x -Wno-unused-result
 sbatch --partition=dev-g --nodes=1 --ntasks-per-node=1 --cpus-per-task=1 --gpus-per-node=1 --time=01:00:00 -o 'rocm6.2.2.out' --wrap='./rocm6.2.2.x 3,100,200,400,800,1600,3200'
 ```
 
@@ -48,10 +61,10 @@ Container source [here](https://github.com/trossi/containers/tree/main/examples/
 ```bash
 export SINGULARITY_BIND="/pfs,/scratch,/projappl,/project,/flash,/appl"
 
-singularity exec rocm_magma.sif hipcc -std=c++14 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.3.2.x
+singularity exec rocm_magma.sif hipcc -std=c++17 --offload-arch=gfx90a -O3 -DHIP -lrocblas -lrocsolver eigh.cpp -o rocm6.3.2.x -Wno-unused-result
 sbatch --partition=dev-g --nodes=1 --ntasks-per-node=1 --cpus-per-task=1 --gpus-per-node=1 --time=01:00:00 -o 'rocm6.3.2.out' --wrap='singularity exec rocm_magma.sif ./rocm6.3.2.x 3,100,200,400,800,1600,3200,6400'
 
-singularity exec rocm_magma.sif hipcc -std=c++14 --offload-arch=gfx90a -O3 -DMAGMA -DHIP -lmagma eigh.cpp -o magma2.9.0_rocm6.3.2.x
+singularity exec rocm_magma.sif hipcc -std=c++17 --offload-arch=gfx90a -O3 -DMAGMA -DHIP -lmagma eigh.cpp -o magma2.9.0_rocm6.3.2.x -Wno-unused-result
 sbatch --partition=dev-g --nodes=1 --ntasks-per-node=1 --cpus-per-task=1 --gpus-per-node=1 --time=01:00:00 -o 'magma2.9.0_rocm6.3.2.out' --wrap='singularity exec rocm_magma.sif ./magma2.9.0_rocm6.3.2.x 3,100,200,400,800,1600,3200,6400,12800'
 ```
 
@@ -78,10 +91,10 @@ make -j128 lib/libmagma.so GPU_TARGET=Ampere OPENBLASDIR=$OPENBLAS_INSTALL_ROOT
 ```bash
 ml cuda/11.5.0
 
-nvcc -std=c++14 -arch=sm_80 -O3 -DCUDA -lcusolver eigh.cpp -o cuda11.5.0.x
+nvcc -std=c++17 -arch=sm_80 -O3 -DCUDA -lcusolver eigh.cpp -o cuda11.5.0.x
 sbatch -p gputest --nodes=1 --ntasks-per-node=1 --gres=gpu:a100:1 -t 0:15:00 -o cuda11.5.0.out --wrap='./cuda11.5.0.x 3,100,200,400,800,1600,3200,6400,12800'
 
-nvcc -std=c++14 -arch=sm_80 -O3 -DMAGMA -DCUDA -lmagma -I$PWD/magma/include -L$PWD/magma/lib -Xcompiler \"-Wl,-rpath,$PWD/magma/lib\" eigh.cpp -o magma2.8.0_cuda11.5.0.x
+nvcc -std=c++17 -arch=sm_80 -O3 -DMAGMA -DCUDA -lmagma -I$PWD/magma/include -L$PWD/magma/lib -Xcompiler \"-Wl,-rpath,$PWD/magma/lib\" eigh.cpp -o magma2.8.0_cuda11.5.0.x
 sbatch -p gputest --nodes=1 --ntasks-per-node=1 --gres=gpu:a100:1 -t 0:15:00 -o magma2.8.0_cuda11.5.0.out --wrap='./magma2.8.0_cuda11.5.0.x 3,100,200,400,800,1600,3200,6400,12800'
 ```
 
@@ -92,9 +105,9 @@ Container source [here](https://github.com/trossi/containers/tree/main/examples/
 ```bash
 export SINGULARITY_BIND="/scratch,/projappl,/appl"
 
-singularity exec -B /local_scratch cuda_magma.sif nvcc -std=c++14 -arch=sm_80 -O3 -DCUDA -lcusolver eigh.cpp -o cuda12.6.1.x
+singularity exec -B /local_scratch cuda_magma.sif nvcc -std=c++17 -arch=sm_80 -O3 -DCUDA -lcusolver eigh.cpp -o cuda12.6.1.x
 sbatch -p gputest --nodes=1 --ntasks-per-node=1 --gres=gpu:a100:1 -t 0:15:00 -o cuda12.6.1.out --wrap='singularity exec --nv cuda_magma.sif ./cuda12.6.1.x 3,100,200,400,800,1600,3200,6400,12800'
 
-singularity exec -B /local_scratch cuda_magma.sif nvcc -std=c++14 -arch=sm_80 -O3 -DMAGMA -DCUDA -lmagma eigh.cpp -o magma2.9.0_cuda12.6.1.x
+singularity exec -B /local_scratch cuda_magma.sif nvcc -std=c++17 -arch=sm_80 -O3 -DMAGMA -DCUDA -lmagma eigh.cpp -o magma2.9.0_cuda12.6.1.x
 sbatch -p gputest --nodes=1 --ntasks-per-node=1 --gres=gpu:a100:1 -t 0:15:00 -o magma2.9.0_cuda12.6.1.out --wrap='singularity exec --nv cuda_magma.sif ./magma2.9.0_cuda12.6.1.x 3,100,200,400,800,1600,3200,6400,12800'
 ```