diff options
author | Cedric Nugteren <web@cedricnugteren.nl> | 2016-09-13 21:14:51 +0200 |
---|---|---|
committer | GitHub <noreply@github.com> | 2016-09-13 21:14:51 +0200 |
commit | f07ac22f5b57d22756d779d2e53620f988d786ee (patch) | |
tree | e8bcbc331683ca6fd807f5a5b83bb05c6e6fed69 /src/kernels/level1/xcopy.opencl | |
parent | 7c13bacf129291e3e295ecb6e833788477085fa0 (diff) | |
parent | 4b94afda941a86f363064ff02f97e21eb9618794 (diff) |
Merge pull request #99 from CNugteren/development
Update to version 0.9.0
Diffstat (limited to 'src/kernels/level1/xcopy.opencl')
-rw-r--r-- | src/kernels/level1/xcopy.opencl | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/src/kernels/level1/xcopy.opencl b/src/kernels/level1/xcopy.opencl index 97c27ccf..228e0735 100644 --- a/src/kernels/level1/xcopy.opencl +++ b/src/kernels/level1/xcopy.opencl @@ -22,10 +22,10 @@ R"( // ================================================================================================= // Full version of the kernel with offsets and strided accesses -__attribute__((reqd_work_group_size(WGS, 1, 1))) -__kernel void Xcopy(const int n, - const __global real* restrict xgm, const int x_offset, const int x_inc, - __global real* ygm, const int y_offset, const int y_inc) { +__kernel __attribute__((reqd_work_group_size(WGS, 1, 1))) +void Xcopy(const int n, + const __global real* restrict xgm, const int x_offset, const int x_inc, + __global real* ygm, const int y_offset, const int y_inc) { // Loops over the work that needs to be done (allows for an arbitrary number of threads) #pragma unroll @@ -38,10 +38,10 @@ __kernel void Xcopy(const int n, // Faster version of the kernel without offsets and strided accesses. Also assumes that 'n' is // dividable by 'VW', 'WGS' and 'WPT'. -__attribute__((reqd_work_group_size(WGS, 1, 1))) -__kernel void XcopyFast(const int n, - const __global realV* restrict xgm, - __global realV* ygm) { +__kernel __attribute__((reqd_work_group_size(WGS, 1, 1))) +void XcopyFast(const int n, + const __global realV* restrict xgm, + __global realV* ygm) { #pragma unroll for (int w=0; w<WPT; ++w) { const int id = w*get_global_size(0) + get_global_id(0); |