We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
1 parent fc3374d commit 19bc74dCopy full SHA for 19bc74d
kernel.cu
@@ -42,8 +42,8 @@ __global__ void weighted_minhash_cuda(
42
const uint32_t sample_offset = sample_index * sample_delta;
43
const uint32_t samples = blockDim.x * sample_delta;
44
extern __shared__ float shmem[];
45
- float *lnmins = &shmem[(threadIdx.y * blockDim.x + sample_index) * 3 * sample_delta];
46
- uint2 *dtmins = reinterpret_cast<uint2 *>(lnmins + sample_delta);
+ float *volatile lnmins = &shmem[(threadIdx.y * blockDim.x + sample_index) * 3 * sample_delta];
+ uint2 *volatile dtmins = reinterpret_cast<uint2 *>(lnmins + sample_delta);
47
int32_t row = -1;
48
for (uint32_t index = 0, border = 0;; index++) {
49
if (index >= border) {
0 commit comments