This repository has been archived by the owner on Mar 21, 2024. It is now read-only.
reduce fails to compile with complex in CUDA 10.1 when compiling a separate library
#1044
Labels
Milestone
Comments
kwrobot
pushed a commit
to Kitware/VTK-m
that referenced
this issue
Jul 9, 2020
When you try to call the `Reduce` operation in the CUDA device adapter with a sufficently complex interator type, you get a compile error that says `error: cannot pass an argument with a user-provided copy-constructor to a device-side kernel launch`. This appears to be a bug in either nvcc or Thrust. I believe it is related to the following reported issues: * NVIDIA/thrust#928 * NVIDIA/thrust#1044 Work around this problem by making a special condition for calling `Reduce` with an `ArrayHandleMultiplexer` that calls the generic algorithm in `DeviceAdapterAlgorithmGeneral` instead of the algorithm in Thrust.
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Posted this in #928, but referencing here for visibility. Minimal reproducible example and details are here: #928 (comment)
This issue is also present in the reduction operation in CUB v1.8.0.
The text was updated successfully, but these errors were encountered: