NASP/cuda-samples
85
0
Fork 0
mirror of https://github.com/NVIDIA/cuda-samples.git synced 2026-10-25 22:28:17 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge public changes to internal ToT

Browse Source
This commit is contained in:
Rob Armstrong
2025-05-21 09:28:35 -07:00
parent 2861e78272 adacf1cffd
commit fc88988f23
4 changed files with 14 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
Samples/6_Performance/transpose/transpose.cu
View File

@@ -597,17 +597,16 @@ int main(int argc, char **argv)
1,
TILE_DIM * BLOCK_ROWS);
// Reset d_odata to zero before starting the next loop iteration to avoid
// carrying over results from previous kernels. Without this reset, residual
// data from a prior kernel (e.g., 'copy') could make a subsequent
// kernel (e.g., 'copySharedMem') appear correct even if it performs no work,
// Reset d_odata to zero before starting the next loop iteration to avoid
// carrying over results from previous kernels. Without this reset, residual
// data from a prior kernel (e.g., 'copy') could make a subsequent
// kernel (e.g., 'copySharedMem') appear correct even if it performs no work,
// leading to false positives in compareData.
for (int i = 0; i < (size_x * size_y); ++i) {
h_odata[i] = 0;
}
// copy host data to device
checkCudaErrors(
cudaMemcpy(d_odata, h_odata, mem_size, cudaMemcpyHostToDevice));
checkCudaErrors(cudaMemcpy(d_odata, h_odata, mem_size, cudaMemcpyHostToDevice));
}
// cleanup