fix bug for knnSearchGpu when knn!=ostride #374
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
fix bug for knnSearch when knn!=ostride
Now most nvidia gpu supports kernel concurrency, but flann always uses default stream, which means gpu can not process two indexs simutaneously when we use multithread. Creating cuda stream for each index can fix this problem.
fix errors in comment
knnSearch should be overrided. If not, Index initialize with KDTreeCuda3dIndexParams will call the default knnSearch, not the gpu one
flann::KDTreeCuda3dIndexParams params; params["gpu_stream"] = mystream;//cudaStream_t mystream params["gpu_stream"] = cudaStreamLegacy;//use default stream params["gpu_stream"] = cudaStreamPerThread;//one stream per thread if params["gpu_stream"] is not set, stream will be created each time calling buildIndex() note that there is no internal synchronization, that is, cudaStreamSynchronize() should be called manually if another stream is used to process the matrices (only in case that matrices_in_gpu_ram==true)
-gencode=arch=compute_52,code=sm_52
-gencode=arch=compute_52,code=\"sm_52,compute_52\"; -gencode=arch=compute_61,code=\"sm_61,compute_61\"
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
knnSearchGpu gives wrong results if knn!=ostride.
For example, when we define the output matrix as
flann::Matrix idx(data, rows, 3, 4);
which means we need to find 3 nearest neighbor, but we want to save the results to a matrix of stride=4, then the result is wrong.
This bug can be fixed changing only one place in kdtree_cuda_3d_index.cu.