Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Fix data incorrect for reduceSum function #43

Open
wants to merge 1 commit into
base: 1.2_25c2cd_merge
Choose a base branch
from
Open

Fix data incorrect for reduceSum function #43

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
18 changes: 18 additions & 0 deletions paddle/fluid/platform/cuda_device_function.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -101,6 +101,23 @@ HOSTDEVICE T Infinity() {
return INFINITY;
}

#ifdef PADDLE_WITH_HIP
template <typename T>
__device__ T reduceSum(T val, int tid, int len) {
const int warpSize = 32;
__shared__ T shm[warpSize*warpSize];
shm[tid] = val;

__syncthreads();

if (tid == 0 ) {
for (int i = 1 ; i < len ; i++)
val += shm[i];
}

return val;
}
#else
template <typename T>
__device__ T reduceSum(T val, int tid, int len) {
// NOTE(zcd): The warp size should be taken from the
Expand Down Expand Up @@ -134,6 +151,7 @@ __device__ T reduceSum(T val, int tid, int len) {
}
return val;
}
#endif

} // namespace platform
} // namespace paddle