portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-10-31 03:54:33 +00:00
Code Releases Activity

Stream doesn't work

Browse Source
This commit is contained in:
Peter Boyle
2022-07-07 17:49:20 +01:00
parent 1f903d9296
commit 3544965f54
1 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
Grid/threads/Accelerator.h
View File

@@ -190,7 +190,7 @@ void Lambda6Apply(uint64_t num1, uint64_t num2, uint64_t num3,
#define accelerator_barrier(dummy) \ #define accelerator_barrier(dummy) \
{ \ { \
cudaStreamSynchronize(cpuStream); \ cudaDeviceSynchronize(); \
cudaError err = cudaGetLastError(); \ cudaError err = cudaGetLastError(); \
if ( cudaSuccess != err ) { \ if ( cudaSuccess != err ) { \
printf("accelerator_barrier(): Cuda error %s \n", \ printf("accelerator_barrier(): Cuda error %s \n", \
@@ -362,11 +362,11 @@ accelerator_inline int acceleratorSIMTlane(int Nsimd) {
dim3 hip_blocks ((num1+nt-1)/nt,num2,1); \ dim3 hip_blocks ((num1+nt-1)/nt,num2,1); \
if(hip_threads.x * hip_threads.y * hip_threads.z <= 64){ \ if(hip_threads.x * hip_threads.y * hip_threads.z <= 64){ \
hipLaunchKernelGGL(LambdaApply64,hip_blocks,hip_threads, \ hipLaunchKernelGGL(LambdaApply64,hip_blocks,hip_threads, \
0,cpuStream, \ 0,0/*cpuStream*/, \
num1,num2,nsimd, lambda); \ num1,num2,nsimd, lambda); \
} else { \ } else { \
hipLaunchKernelGGL(LambdaApply,hip_blocks,hip_threads, \ hipLaunchKernelGGL(LambdaApply,hip_blocks,hip_threads, \
0,cpuStream, \ 0,0/*cpuStream*/, \
num1,num2,nsimd, lambda); \ num1,num2,nsimd, lambda); \
} \ } \
} }