← GPU latency results

gpu/routine_full_coverage

Legacy report: ingestion provenance is unrecorded; no raw artifact is inferred.

Original report

GPU routine coverage

Source accounting: /home/shinaoka/tensor4all/tenferro-benchmark/.worktrees/gpu-refresh-20261017/data/results/nvidia-gpu/gpu/routine_selection/20261017_104441/case_status.json.

Manifest version: 1; coverage: full; effort: standard.

Measured coverage complete: no.

Verification-only, unsupported, missing and unselected rows are not measured coverage. Noise flags are descriptive screens, not regression conclusions.

expected

• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-trace
• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-eager
• gpu/dense:dense_matmul_f64_3072:pytorch-cuda
• gpu/dense:dense_matmul_f64_3072:jax-cuda
• gpu/dense:dense_matmul_f64_3072:cublaslt
• gpu/dense:dense_matmul_f64_3072:cutlass
• gpu/dense:dense_matmul_f64_3072:cusolver
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b1024_256:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:cublaslt
• gpu/dense:dense_batched_matmul_f64_b1024_256:cutlass
• gpu/dense:dense_batched_matmul_f64_b1024_256:cusolver
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-trace
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-eager
• gpu/dense:dense_qr_f64_1536:pytorch-cuda
• gpu/dense:dense_qr_f64_1536:jax-cuda
• gpu/dense:dense_qr_f64_1536:cublaslt
• gpu/dense:dense_qr_f64_1536:cutlass
• gpu/dense:dense_qr_f64_1536:cusolver
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_512_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_512_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_512_rhs16:cublaslt
• gpu/dense:dense_solve_f64_512_rhs16:cutlass
• gpu/dense:dense_solve_f64_512_rhs16:cusolver
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_1024_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:cublaslt
• gpu/dense:dense_solve_f64_1024_rhs16:cutlass
• gpu/dense:dense_solve_f64_1024_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs16:cutlass
• gpu/dense:dense_solve_f64_2048_rhs16:cusolver
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_4096_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:cublaslt
• gpu/dense:dense_solve_f64_4096_rhs16:cutlass
• gpu/dense:dense_solve_f64_4096_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs128:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs128:cutlass
• gpu/dense:dense_solve_f64_2048_rhs128:cusolver
• gpu/dense:dense_svd_f64_256:tenferro-cuda-trace
• gpu/dense:dense_svd_f64_256:tenferro-cuda-eager
• gpu/dense:dense_svd_f64_256:pytorch-cuda
• gpu/dense:dense_svd_f64_256:jax-cuda
• gpu/dense:dense_svd_f64_256:cublaslt
• gpu/dense:dense_svd_f64_256:cutlass
• gpu/dense:dense_svd_f64_256:cusolver
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-trace
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-eager
• gpu/dense:dense_eigh_f64_1024:pytorch-cuda
• gpu/dense:dense_eigh_f64_1024:jax-cuda
• gpu/dense:dense_eigh_f64_1024:cublaslt
• gpu/dense:dense_eigh_f64_1024:cutlass
• gpu/dense:dense_eigh_f64_1024:cusolver
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b16_32:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:cublaslt
• gpu/dense:dense_batched_matmul_f64_b16_32:cutlass
• gpu/dense:dense_batched_matmul_f64_b16_32:cusolver
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-trace
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-eager
• gpu/einsum:einsum_bin_matmul_3072_f64:pytorch-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:jax-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:cublaslt
• gpu/einsum:einsum_bin_matmul_3072_f64:cutlass
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:pytorch-cuda
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:cusparse
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:ginkgo
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:pytorch-cuda
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:cusparse
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:ginkgo
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-trace
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-eager
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:pytorch-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda-tree-jit

selected

• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-trace
• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-eager
• gpu/dense:dense_matmul_f64_3072:pytorch-cuda
• gpu/dense:dense_matmul_f64_3072:jax-cuda
• gpu/dense:dense_matmul_f64_3072:cublaslt
• gpu/dense:dense_matmul_f64_3072:cutlass
• gpu/dense:dense_matmul_f64_3072:cusolver
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b1024_256:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:cublaslt
• gpu/dense:dense_batched_matmul_f64_b1024_256:cutlass
• gpu/dense:dense_batched_matmul_f64_b1024_256:cusolver
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-trace
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-eager
• gpu/dense:dense_qr_f64_1536:pytorch-cuda
• gpu/dense:dense_qr_f64_1536:jax-cuda
• gpu/dense:dense_qr_f64_1536:cublaslt
• gpu/dense:dense_qr_f64_1536:cutlass
• gpu/dense:dense_qr_f64_1536:cusolver
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_512_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_512_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_512_rhs16:cublaslt
• gpu/dense:dense_solve_f64_512_rhs16:cutlass
• gpu/dense:dense_solve_f64_512_rhs16:cusolver
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_1024_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:cublaslt
• gpu/dense:dense_solve_f64_1024_rhs16:cutlass
• gpu/dense:dense_solve_f64_1024_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs16:cutlass
• gpu/dense:dense_solve_f64_2048_rhs16:cusolver
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_4096_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:cublaslt
• gpu/dense:dense_solve_f64_4096_rhs16:cutlass
• gpu/dense:dense_solve_f64_4096_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs128:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs128:cutlass
• gpu/dense:dense_solve_f64_2048_rhs128:cusolver
• gpu/dense:dense_svd_f64_256:tenferro-cuda-trace
• gpu/dense:dense_svd_f64_256:tenferro-cuda-eager
• gpu/dense:dense_svd_f64_256:pytorch-cuda
• gpu/dense:dense_svd_f64_256:jax-cuda
• gpu/dense:dense_svd_f64_256:cublaslt
• gpu/dense:dense_svd_f64_256:cutlass
• gpu/dense:dense_svd_f64_256:cusolver
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-trace
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-eager
• gpu/dense:dense_eigh_f64_1024:pytorch-cuda
• gpu/dense:dense_eigh_f64_1024:jax-cuda
• gpu/dense:dense_eigh_f64_1024:cublaslt
• gpu/dense:dense_eigh_f64_1024:cutlass
• gpu/dense:dense_eigh_f64_1024:cusolver
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b16_32:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:cublaslt
• gpu/dense:dense_batched_matmul_f64_b16_32:cutlass
• gpu/dense:dense_batched_matmul_f64_b16_32:cusolver
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-trace
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-eager
• gpu/einsum:einsum_bin_matmul_3072_f64:pytorch-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:jax-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:cublaslt
• gpu/einsum:einsum_bin_matmul_3072_f64:cutlass
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:pytorch-cuda
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:cusparse
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:ginkgo
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:pytorch-cuda
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:cusparse
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:ginkgo
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-trace
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-eager
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:pytorch-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda-tree-jit

executed

• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-trace
• gpu/dense:dense_matmul_f64_3072:tenferro-cuda-eager
• gpu/dense:dense_matmul_f64_3072:pytorch-cuda
• gpu/dense:dense_matmul_f64_3072:jax-cuda
• gpu/dense:dense_matmul_f64_3072:cublaslt
• gpu/dense:dense_matmul_f64_3072:cutlass
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b1024_256:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b1024_256:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b1024_256:cublaslt
• gpu/dense:dense_batched_matmul_f64_b1024_256:cutlass
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-trace
• gpu/dense:dense_qr_f64_1536:tenferro-cuda-eager
• gpu/dense:dense_qr_f64_1536:pytorch-cuda
• gpu/dense:dense_qr_f64_1536:jax-cuda
• gpu/dense:dense_qr_f64_1536:cusolver
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_512_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_512_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_512_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_512_rhs16:cusolver
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_1024_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_1024_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_1024_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs16:cusolver
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_4096_rhs16:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_4096_rhs16:pytorch-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:jax-cuda
• gpu/dense:dense_solve_f64_4096_rhs16:cusolver
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-trace
• gpu/dense:dense_solve_f64_2048_rhs128:tenferro-cuda-eager
• gpu/dense:dense_solve_f64_2048_rhs128:pytorch-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:jax-cuda
• gpu/dense:dense_solve_f64_2048_rhs128:cusolver
• gpu/dense:dense_svd_f64_256:tenferro-cuda-trace
• gpu/dense:dense_svd_f64_256:tenferro-cuda-eager
• gpu/dense:dense_svd_f64_256:pytorch-cuda
• gpu/dense:dense_svd_f64_256:jax-cuda
• gpu/dense:dense_svd_f64_256:cusolver
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-trace
• gpu/dense:dense_eigh_f64_1024:tenferro-cuda-eager
• gpu/dense:dense_eigh_f64_1024:pytorch-cuda
• gpu/dense:dense_eigh_f64_1024:jax-cuda
• gpu/dense:dense_eigh_f64_1024:cusolver
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-trace
• gpu/dense:dense_batched_matmul_f64_b16_32:tenferro-cuda-eager
• gpu/dense:dense_batched_matmul_f64_b16_32:pytorch-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:jax-cuda
• gpu/dense:dense_batched_matmul_f64_b16_32:cublaslt
• gpu/dense:dense_batched_matmul_f64_b16_32:cutlass
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-trace
• gpu/einsum:einsum_bin_matmul_3072_f64:tenferro-cuda-eager
• gpu/einsum:einsum_bin_matmul_3072_f64:pytorch-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:jax-cuda
• gpu/einsum:einsum_bin_matmul_3072_f64:cublaslt
• gpu/einsum:einsum_bin_matmul_3072_f64:cutlass
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:pytorch-cuda
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:cusparse
• gpu/sparse:sparse_synthetic_4m_64m_spmv_f64:ginkgo
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:pytorch-cuda
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:cusparse
• gpu/sparse:sparse_synthetic_64k_4m_spmm_f64_rhs1024:ginkgo
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-trace
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:tenferro-cuda-eager
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:pytorch-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda
• gpu/tensornetwork:tensornetwork_permutation_optimized_f32:jax-cuda-tree-jit

unsupported

• gpu/dense:dense_matmul_f64_3072:cusolver
• gpu/dense:dense_batched_matmul_f64_b1024_256:cusolver
• gpu/dense:dense_qr_f64_1536:cublaslt
• gpu/dense:dense_qr_f64_1536:cutlass
• gpu/dense:dense_solve_f64_512_rhs16:cublaslt
• gpu/dense:dense_solve_f64_512_rhs16:cutlass
• gpu/dense:dense_solve_f64_1024_rhs16:cublaslt
• gpu/dense:dense_solve_f64_1024_rhs16:cutlass
• gpu/dense:dense_solve_f64_2048_rhs16:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs16:cutlass
• gpu/dense:dense_solve_f64_4096_rhs16:cublaslt
• gpu/dense:dense_solve_f64_4096_rhs16:cutlass
• gpu/dense:dense_solve_f64_2048_rhs128:cublaslt
• gpu/dense:dense_solve_f64_2048_rhs128:cutlass
• gpu/dense:dense_svd_f64_256:cublaslt
• gpu/dense:dense_svd_f64_256:cutlass
• gpu/dense:dense_eigh_f64_1024:cublaslt
• gpu/dense:dense_eigh_f64_1024:cutlass
• gpu/dense:dense_batched_matmul_f64_b16_32:cusolver

failed

None.

missing

None.

verification_only

None.

raw_samples_unavailable

None.

noisy

None.

unselected

None.

duplicates

None.

unexpected

None.