STL::map

Accelerated Computing CUDA CUDA Programming and Performance
1

Hello,

I’m new to CUDA so forgive me if my question sounds a bit stupid. Is it possible to insert new elements in a STL::map in parallel ?
Thanks,
Do

2

Hello,

I’m new to CUDA so forgive me if my question sounds a bit stupid. Is it possible to insert new elements in a STL::map in parallel ?
Thanks,
Do

3

I’ve never heard of an implementation of std::map for device code. What are you using?

4

I’ve never heard of an implementation of std::map for device code. What are you using?

5

I use standard C++ as much as possible, seems to work pretty well for cards with CUDA capability 2.0 and up. Basically map is an associative array from the C++ Standard Template Library.

However I’m not sure it makes sense to try to write elements in a map in parallel.

Do

6

I use standard C++ as much as possible, seems to work pretty well for cards with CUDA capability 2.0 and up. Basically map is an associative array from the C++ Standard Template Library.

However I’m not sure it makes sense to try to write elements in a map in parallel.

Do

7

The STL doesn’t work in CUDA device code (host code is fine, of course), so I’m still confused.

8

The STL doesn’t work in CUDA device code (host code is fine, of course), so I’m still confused.

9

No it wouldn’t work and there doesn’t seem to be an easy way to port existing implementations to CUDA. Just writing an allocator isn’t enough AFAICT.