Hello all,
Is it possible to call the CUB DeviceReduce (cub::DeviceReduce::reduce) from within a global kernel? If so, can anyone point me to how this is accomplished with some sort of simple example code, hint(s), etc.?
Thank you
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Call cuBLAS from device function | 1 | 704 | November 15, 2019 | |
| Calling cuBLAS from device? | 4 | 809 | April 26, 2023 | |
| cuSolver device API so that I can call from kernel? | 1 | 1284 | March 10, 2015 | |
| Reduction using CUDA CUB with complex numbers | 1 | 450 | May 22, 2020 | |
| Declare __shared__ in kernel and Use it in Device Function with CUB | 3 | 37 | January 16, 2025 | |
| calling a GPU kernel subroutine from a global subroutine | 6 | 3917 | July 9, 2015 | |
| CUBLAS call from a cuda kernel | 2 | 1485 | May 10, 2010 | |
| What is the state of device-side libraries? | 0 | 343 | August 15, 2019 | |
| Dynamic parallelisim and cusparse | 1 | 826 | April 26, 2015 | |
| Cublas within kernel | 1 | 1502 | July 28, 2009 |