diff --git a/.github/ops-bot.yaml b/.github/ops-bot.yaml index 9a0b415503..2ed5231aef 100644 --- a/.github/ops-bot.yaml +++ b/.github/ops-bot.yaml @@ -6,3 +6,4 @@ branch_checker: true label_checker: true release_drafter: true recently_updated: true +forward_merger: true diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml index 2294b4193f..bd8b13d21e 100644 --- a/.github/workflows/build.yaml +++ b/.github/workflows/build.yaml @@ -28,7 +28,7 @@ concurrency: jobs: cpp-build: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -37,7 +37,7 @@ jobs: python-build: needs: [cpp-build] secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -46,7 +46,7 @@ jobs: upload-conda: needs: [cpp-build, python-build] secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-upload-packages.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-upload-packages.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -57,7 +57,7 @@ jobs: if: github.ref_type == 'branch' needs: python-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.04 with: arch: "amd64" branch: ${{ inputs.branch }} @@ -69,7 +69,7 @@ jobs: sha: ${{ inputs.sha }} wheel-build-pylibraft: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -79,7 +79,7 @@ jobs: wheel-publish-pylibraft: needs: wheel-build-pylibraft secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -89,7 +89,7 @@ jobs: wheel-build-raft-dask: needs: wheel-publish-pylibraft secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} @@ -99,7 +99,7 @@ jobs: wheel-publish-raft-dask: needs: wheel-build-raft-dask secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-publish.yaml@branch-24.04 with: build_type: ${{ inputs.build_type || 'branch' }} branch: ${{ inputs.branch }} diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml index d4cd51183c..82594909a9 100644 --- a/.github/workflows/pr.yaml +++ b/.github/workflows/pr.yaml @@ -25,29 +25,29 @@ jobs: - wheel-tests-raft-dask - devcontainer secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/pr-builder.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/pr-builder.yaml@branch-24.04 checks: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/checks.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/checks.yaml@branch-24.04 with: enable_check_generated_files: false conda-cpp-build: needs: checks secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-build.yaml@branch-24.04 with: build_type: pull-request node_type: cpu16 conda-cpp-tests: needs: conda-cpp-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@branch-24.04 with: build_type: pull-request conda-cpp-checks: needs: conda-cpp-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@branch-24.04 with: build_type: pull-request enable_check_symbols: true @@ -55,19 +55,19 @@ jobs: conda-python-build: needs: conda-cpp-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-python-build.yaml@branch-24.04 with: build_type: pull-request conda-python-tests: needs: conda-python-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.04 with: build_type: pull-request docs-build: needs: conda-python-build secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/custom-job.yaml@branch-24.04 with: build_type: pull-request node_type: "gpu-v100-latest-1" @@ -77,34 +77,34 @@ jobs: wheel-build-pylibraft: needs: checks secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.04 with: build_type: pull-request script: ci/build_wheel_pylibraft.sh wheel-tests-pylibraft: needs: wheel-build-pylibraft secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.04 with: build_type: pull-request script: ci/test_wheel_pylibraft.sh wheel-build-raft-dask: needs: wheel-tests-pylibraft secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-build.yaml@branch-24.04 with: build_type: pull-request script: "ci/build_wheel_raft_dask.sh" wheel-tests-raft-dask: needs: wheel-build-raft-dask secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.04 with: build_type: pull-request script: ci/test_wheel_raft_dask.sh devcontainer: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-24.04 with: build_command: | sccache -z; diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml index e2f956e11c..2a557a8b84 100644 --- a/.github/workflows/test.yaml +++ b/.github/workflows/test.yaml @@ -16,7 +16,7 @@ on: jobs: conda-cpp-checks: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-post-build-checks.yaml@branch-24.04 with: build_type: nightly branch: ${{ inputs.branch }} @@ -26,7 +26,7 @@ jobs: symbol_exclusions: _ZN\d+raft_cutlass conda-cpp-tests: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-cpp-tests.yaml@branch-24.04 with: build_type: nightly branch: ${{ inputs.branch }} @@ -34,7 +34,7 @@ jobs: sha: ${{ inputs.sha }} conda-python-tests: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/conda-python-tests.yaml@branch-24.04 with: build_type: nightly branch: ${{ inputs.branch }} @@ -42,7 +42,7 @@ jobs: sha: ${{ inputs.sha }} wheel-tests-pylibraft: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.04 with: build_type: nightly branch: ${{ inputs.branch }} @@ -51,7 +51,7 @@ jobs: script: ci/test_wheel_pylibraft.sh wheel-tests-raft-dask: secrets: inherit - uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@test-cuda-12.2 + uses: rapidsai/shared-workflows/.github/workflows/wheels-test.yaml@branch-24.04 with: build_type: nightly branch: ${{ inputs.branch }} diff --git a/cpp/include/raft/sparse/linalg/spmm.hpp b/cpp/include/raft/sparse/linalg/spmm.hpp index dd661c71ac..2eb1e58f1c 100644 --- a/cpp/include/raft/sparse/linalg/spmm.hpp +++ b/cpp/include/raft/sparse/linalg/spmm.hpp @@ -60,11 +60,19 @@ void spmm(raft::resources const& handle, { bool is_row_major = detail::is_row_major(y, z); + // WARNING: The following copy is working around a bug in cusparse which causes an alignment issue + // and incorrect results. This bug is fixed in CUDA 12.5+ so this workaround shouldn't be removed + // until that version is supported. + auto size = is_row_major ? (z.extent(0) - 1) * z.stride(0) + z.extent(1) + : (z.extent(1) - 1) * z.stride(1) + z.extent(0); + rmm::device_uvector z_tmp(size, raft::resource::get_cuda_stream(handle)); + raft::copy(z_tmp.data(), z.data_handle(), z_tmp.size(), raft::resource::get_cuda_stream(handle)); + auto z_tmp_view = is_row_major ? raft::make_device_strided_matrix_view( - z.data_handle(), z.extent(0), z.extent(1), z.stride(0)) + z_tmp.data(), z.extent(0), z.extent(1), z.stride(0)) : raft::make_device_strided_matrix_view( - z.data_handle(), z.extent(0), z.extent(1), z.stride(1)); + z_tmp.data(), z.extent(0), z.extent(1), z.stride(1)); auto descr_x = detail::create_descriptor(x); auto descr_y = detail::create_descriptor(y); @@ -74,10 +82,7 @@ void spmm(raft::resources const& handle, // WARNING: Do not remove the following copy unless you can, with certainty, say that // the underlying cuSPARSE issue affecting CUDA 12.2+ has been resolved. - raft::copy(z.data_handle(), - z_tmp_view.data_handle(), - z_tmp_view.size(), - raft::resource::get_cuda_stream(handle)); + raft::copy(z.data_handle(), z_tmp.data(), z_tmp.size(), raft::resource::get_cuda_stream(handle)); RAFT_CUSPARSE_TRY_NO_THROW(cusparseDestroySpMat(descr_x)); RAFT_CUSPARSE_TRY_NO_THROW(cusparseDestroyDnMat(descr_y)); RAFT_CUSPARSE_TRY_NO_THROW(cusparseDestroyDnMat(descr_z));