Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR provides support for running CUDA compression / decompression on a CUDA stream. This allows for potential overlap between memory transfers and kernel execution, or memory transfers in different directions on devices that support it. Overlapping transfers and execution enables large performance gains in certain scenarios.
To that end I've introduced a
zfp_exec_params_cuda
struct (similar tozfp_exec_params_omp
) that contains acudaStream_t
member. Setting and querying the stream can be done using the newzfp_stream_set_cuda_stream
andzfp_stream_cuda_stream
functions.When no stream is specified, the default 0-stream is used. Since several CUDA calls in the code have been changed to their
*Async
equivalent, acudaStreamSynchronize(0)
is performed when stream 0 is used, to retain blocking behavior in that case.