Code Coverage:

This package wraps the CUDA driver API. It is meant for users who need low-level access to the CUDA toolkit or the underlying GPU. For a wrapper of the higher-level runtime API, see CUDArt.

Requirements:

• Julia 0.5 or higher (use v0.1.0 of this package for compatibility with Julia 0.4)
• CUDA toolkit: tested up until v8.0, newer versions might work but can be incompatible
• CUDA hardware

Pkg.add("CUDAdrv")
Pkg.test("CUDAdrv")

In general, this wrapper tries to stay close to the abstraction level of the CUDA driver API. However, there are some additional features:

Except for the encapsulating context, destroy or unload calls are never needed. Objects are registered with the Julia garbage collector, and are automatically finalized when they go out of scope.

However, many CUDA API functions implicitly depend on global state, such as the current active context. The wrapper needs to model those dependencies in order for objects not to get destroyed before any dependent object is. If we fail to model these dependency relations, API calls might randomly fail, eg. in the case of a missing context dependency with a INVALID_CONTEXT or CONTEXT_IS_DESTROYED error message.

If this seems to be the case, re-run with TRACE=1 and file a bug report.

For extra information about what's happening behind the scenes, you can enable extra output by defining the DEBUG environment variable, or TRACE for even more information.

Note that these features are incompatible with precompilation as the debugging code is enabled statically in order to avoid any run-time overhead, so you need to wipe the compile cache or run Julia using --compilecache=no. Enabling colors with --color=yes is also recommended as it color-codes the output.