diff options
| author |  Axel Huebl <axel.huebl@plasma.ninja>
| 2022-06-08 18:26:58 -0700 |
|---|---|---|
| committer |  GitHub <noreply@github.com>
| 2022-06-09 01:26:58 +0000 |
| commit | 54a948e612b61a4d6db642723d6f03116ed46b1d (patch) | |
| tree | 5e52a95ec481b0ee24b865f875ef91a88a37ba15 /Source/Particles/Collision/BackgroundMCC/MCCProcess.cpp | |
| parent | cac7fd39629ad9b57ac2b1e329ddce31509f02aa (diff) | |
| download | WarpX-54a948e612b61a4d6db642723d6f03116ed46b1d.tar.gz WarpX-54a948e612b61a4d6db642723d6f03116ed46b1d.tar.zst WarpX-54a948e612b61a4d6db642723d6f03116ed46b1d.zip | |
Summit: Fix BLAS++/LAPACK++ Modules (#3172)
These modules are by default using GPU support, which points to
CUDA 11.0. Due to C++17 compiler bugs in that early CUDA release,
we cannot use that version and pull in circular dependencies to
two different CUDA toolkit versions as soon as we use BLAS++ &
LAPACK++: when compiling PSATD+RZ.
Nonetheless, we do not use BLAS++/LAPACK++ on GPU yet anyway. Thus,
we can simply use the CPU versions of those modules.
Diffstat (limited to 'Source/Particles/Collision/BackgroundMCC/MCCProcess.cpp')
0 files changed, 0 insertions, 0 deletions