| commit | 72a0e13408d527c2b93a66e88f23c2100d22ee43 | |
| author | Szilárd Páll <pall.szilard@gmail.com> | |
| Wed, 29 Mar 2017 01:26:11 +0000 (29 03:26 +0200) | ||
| committer | Mark Abraham <mark.j.abraham@gmail.com> | |
| Mon, 21 Aug 2017 12:27:29 +0000 (21 14:27 +0200) | ||
| tree | 4d7d38cef6a306df1f8f9a5510feef52e95211bf | treesnapshot (tar.gz zip) |
| parent | 25b6c3a473c93f32299f228fc7e52d234c73caae | commitdiff |
Add OpenCL pruning kernels and launch/timing logic
The kernels have been tested for correction on NVIDIA >=CC 3.5 and AMD
GCN devices. Tuning for AMD has been done on the old fglrx stack which
has limitations on the intra-workgroup parallelism, so choice of the
j4 concurrency parameter should be revisited at some later stage (using
the latest AMDGPU-PRO and hopefully ROCm).
A number of possible improvements have been also identified and noted as
comments in nbnxn_ocl.cpp.
Change-Id: I7129ec247706d33317df1256846943ee8b0d540c
The kernels have been tested for correction on NVIDIA >=CC 3.5 and AMD
GCN devices. Tuning for AMD has been done on the old fglrx stack which
has limitations on the intra-workgroup parallelism, so choice of the
j4 concurrency parameter should be revisited at some later stage (using
the latest AMDGPU-PRO and hopefully ROCm).
A number of possible improvements have been also identified and noted as
comments in nbnxn_ocl.cpp.
Change-Id: I7129ec247706d33317df1256846943ee8b0d540c