quda
quda copied to clipboard
lattice
QUDA is a library for performing calculations in lattice QCD on GPUs.
I'm compiling QUDA 1.1.0 using GCC 10.3.0., CUDA 11.3.1, OpenMPI, (external) Eigen, 3.3.9, and (external) NVSHMEM 2.4.1 on CentOS 8.4. The build is configured with: ``` cmake -DQUDA_GPU_ARCH=sm_80 -DQUDA _MPI=ON...
Make sure this does not break linking if we mix static / shared linkage.
See title, have the tunecache automatically save with some frequency either predetermined and/or settable via environment variable. As part of this we can hopefully remove many explicit tunecache save function...
Wanting to run `ctest` in parallel (`-j 16`) and wanting to replace the default binding (otherwise I get all processes bound to two cores on my test machine), I wanted...
We don't need to tune starting from grid size 1.
This is a self-reminder that the QIO tag needs to be updated once https://github.com/usqcd-software/qio/pull/14 is merged. Until that is completed you can not use the version of QIO and QUDA...
A comparison of perf numbers between release/1.1.x and GK shows there's a noteworthy regression in performance for typical staggered problem sizes and Nc values (24^4 local volume, Nc = 24;...
Requested by @jcosborn and @jxy. I'll do this post GK merge.
Per discussion in #1056 , after the GK merge we should decide on a consistent naming convention for `Float` types, etc, used throughout the library, and make the global change....