diff options
Diffstat (limited to 'src/kernels/level1/xswap.opencl')
| -rw-r--r-- | src/kernels/level1/xswap.opencl | 16 |
1 files changed, 8 insertions, 8 deletions
diff --git a/src/kernels/level1/xswap.opencl b/src/kernels/level1/xswap.opencl index f6487b58..267271c0 100644 --- a/src/kernels/level1/xswap.opencl +++ b/src/kernels/level1/xswap.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 Xswap(const int n, - __global real* 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 Xswap(const int n, + __global real* 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 @@ -40,10 +40,10 @@ __kernel void Xswap(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 XswapFast(const int n, - __global realV* xgm, - __global realV* ygm) { +__kernel __attribute__((reqd_work_group_size(WGS, 1, 1))) +void XswapFast(const int n, + __global realV* 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); |