| commit | 8c3e9ab2275c8ccbdd35aa2ae193e3b2df4c4f34 | |
| author | Mark Abraham <mark.j.abraham@gmail.com> | |
| Sat, 27 Oct 2018 08:26:17 +0000 (27 10:26 +0200) | ||
| committer | Mark Abraham <mark.j.abraham@gmail.com> | |
| Tue, 30 Oct 2018 08:19:00 +0000 (30 09:19 +0100) | ||
| tree | 3a8643263c566d44dddfe8aba7f6fecf1a7e0d7e | treesnapshot (tar.gz zip) |
| parent | f81e2023610f40db37a35b74ee91694f4304512b | commitdiff |
Rename to GMX_OPENCL_NB_CLUSTER_SIZE
The Intel iGPU OpenCL support for NB kernels needs a different cluster
size from the support for AMD and NVIDIA devices. We do use GMX_OCL as
a prefix for various developer-facing aspects of the OpenCL support,
but we shouldn't expose the use of OCL as an abbreviation to users,
given that we use GMX_USE_OPENCL fully written out to control the use
of the OpenCL build configuration.
Change-Id: I9d9b43e2888d77575fc0a781e6c9577c73551b8a
The Intel iGPU OpenCL support for NB kernels needs a different cluster
size from the support for AMD and NVIDIA devices. We do use GMX_OCL as
a prefix for various developer-facing aspects of the OpenCL support,
but we shouldn't expose the use of OCL as an abbreviation to users,
given that we use GMX_USE_OPENCL fully written out to control the use
of the OpenCL build configuration.
Change-Id: I9d9b43e2888d77575fc0a781e6c9577c73551b8a