.. |
detail
|
Merge pull request #5852 from multitalentloes/cudagraph_experiment
|
2025-02-12 15:01:20 +01:00 |
device_management.cpp
|
fixed: division by zero if no gpu device was found
|
2024-11-01 10:29:51 +01:00 |
device_management.hpp
|
Print GPU used on every rank
|
2024-10-30 11:03:29 +01:00 |
gpu_resources.hpp
|
Added utility class to hold a GPU resource (stream, event, graph, etc).
|
2025-02-04 14:53:31 +01:00 |
gpu_smart_pointer.hpp
|
Fixes for NVCC older than 12.5.1
|
2025-02-06 10:52:49 +01:00 |
GpuBlockPreconditioner.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
GpuBuffer.cpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuBuffer.hpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuDILU.cpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuDILU.hpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuJac.cpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
GpuJac.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
GpuOwnerOverlapCopy.hpp
|
mark parameters/variables as const where appropriate
|
2025-01-23 07:57:44 +01:00 |
GpuSeqILU0.cpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
GpuSeqILU0.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
GpuSparseMatrix.cpp
|
implement PR feedback
|
2024-10-02 14:57:34 +02:00 |
GpuSparseMatrix.hpp
|
implement PR feedback
|
2024-10-02 14:57:34 +02:00 |
GpuVector.cpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuVector.hpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
GpuView.cpp
|
refactor opm_cuda_safe_call
|
2024-08-23 11:12:13 +02:00 |
GpuView.hpp
|
minor adjustments
|
2025-01-09 10:38:58 +01:00 |
OpmGpuILU0.cpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
OpmGpuILU0.hpp
|
use cudagraphs in gpu DILU and ILU0 apply
|
2025-02-12 11:10:40 +01:00 |
PreconditionerAdapter.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
PreconditionerConvertFieldTypeAdapter.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
PreconditionerHolder.hpp
|
refactor cuistl folder names
|
2024-08-22 15:32:21 +02:00 |
set_device.cpp
|
Print GPU used on every rank
|
2024-10-30 11:03:29 +01:00 |
set_device.hpp
|
Print GPU used on every rank
|
2024-10-30 11:03:29 +01:00 |
SolverAdapter.hpp
|
SolverAdapter is now more explicit on its communicator/MPI dependence. This fixes the MPI issues with CUDA/HIP/GPUISTL that were introduced with the GhostLastMatrixAdapter.
|
2024-09-24 10:03:23 +02:00 |