Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[REVIEW] Avoid useless copy when exchanging flag buffers in CSR WeakCC #3096

Merged
merged 4 commits into from
Nov 12, 2020
Merged

[REVIEW] Avoid useless copy when exchanging flag buffers in CSR WeakCC #3096

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
0eff96c
Avoid useless copy when exchanging flag buffers in CSR WeakCC
Nyrio Oct 30, 2020
e5c7fcb
Update changelog + fix clang-format
Nyrio Oct 30, 2020
9490b16
Use std::swap
Nyrio Nov 2, 2020
a6d4fdb
Merge branch 'branch-0.17' into enh-weakcc-flags
Nyrio Nov 12, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@
- PR #3044: Move leftover `linalg` and `stats` to RAFT namespaces
- PR #3067: Deleting prims moved to RAFT and updating header paths
- PR #3074: Reducing dask coordinate descent test runtime
- PR #3096: Avoid memory transfers in CSR WeakCC for DBSCAN
- PR #3088: More readable and robust FIL C++ test management
- PR #3052: Speeding up MNMG KNN Cl&Re testing
- PR #3115: Speeding up MNMG UMAP testing
Expand Down
11 changes: 2 additions & 9 deletions cpp/src_prims/sparse/csr.cuh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,7 @@
#include <cuda_runtime.h>
#include <stdio.h>

#include <algorithm>
#include <iostream>

namespace MLCommon {
Expand Down Expand Up @@ -802,8 +803,6 @@ void weak_cc_label_batched(Index_ *labels, const Index_ *row_ind,
"Index_ should be 4 or 8 bytes");

bool host_m;
bool *host_fa = (bool *)malloc(sizeof(bool) * N);
bool *host_xa = (bool *)malloc(sizeof(bool) * N);

dim3 blocks(raft::ceildiv(batchSize, Index_(TPB_X)));
dim3 threads(TPB_X);
Expand All @@ -824,20 +823,14 @@ void weak_cc_label_batched(Index_ *labels, const Index_ *row_ind,
CUDA_CHECK(cudaStreamSynchronize(stream));

//** swapping F1 and F2
raft::update_host(host_fa, state->fa, N, stream);
raft::update_host(host_xa, state->xa, N, stream);
raft::update_device(state->fa, host_xa, N, stream);
raft::update_device(state->xa, host_fa, N, stream);
std::swap(state->fa, state->xa);

//** Updating m *
raft::update_host(&host_m, state->m, 1, stream);
CUDA_CHECK(cudaStreamSynchronize(stream));

n_iters++;
} while (host_m);

free(host_fa);
free(host_xa);
}

/**
Expand Down