CUDA Unified Memory and use of std::vector in device code

Question

Back in the days, std::vector was not allowed in CUDA device code. Is that still true with current cuda 10.2 toolkit with Unified Memory?

I have a few public data members of type std::vector in a class that is passed reference to be used by a device kernel.

nvcc complains about calling a host function("std::vector...) ... from a global function("...) not allowed.

What is the correct way to use unified memory, if at all possible, to use on an std::vector? If it is not possible, is there an efficient work-around?

docs.nvidia.com/cuda/cuda-c-programming-guide/…

Robert Crovella
– Robert Crovella

2020-06-04 13:25:32 +00:00
Commented Jun 4, 2020 at 13:25 — Robert Crovella
– Robert Crovella, Commented Jun 4, 2020 at 13:25

2 revs · Accepted Answer · 2020-06-04 13:30:17Z

1

Back in the days, std::vector was not allowed in CUDA device code. Is that still true with current cuda 10.2 toolkit with Unified Memory?

Yes.

What is the correct way to use unified memory, if at all possible, to use on an std::vector?

There is not one. It isn't possible. There is no C++ standard library support on the device.

If it is not possible, is there an efficient work-around?

No there is not.

edited Jun 4, 2020 at 13:30

community wiki

2 revs
talonmies

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

CUDA Unified Memory and use of std::vector in device code

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related