cuda-samples · Issues· 126 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #409
Potential bug in simpleVulkan example
Updated Aug 28, 2026 - #411
Improve matrixMul Tiling to support non-multiple BLOCK_SIZE dimensions in matrixMul.cu .
Updated Aug 28, 2026 - #413
Missing check for CUDA failure in example
Updated Aug 28, 2026 - #417
Request for Official CUDA support for Ubuntu 26.04 LTS (release timeline for native installer/apt repo)
Updated Aug 28, 2026 - #418
Sample proposal: persistent kernel for tiny-model autoregressive inference (microGPT, ~4k MACs/token)
Updated Aug 28, 2026 - #433
CUDA 13.2.1 libcudart causes SIGSEGV in cuInit on WSL2 (Driver 596.49, Blackwell)
Updated Aug 28, 2026 - #437
helloTile: per-sample build fails on Ubuntu/Debian/Fedora — _FORTIFY_SOURCE rewrites printf to a non-tile-callable __printf_chk (v13.3)
Updated Aug 28, 2026 - #438
4_CUDA_Libraries: cmake -S cpp fails — CUDA language not enabled for the subtree, so cuda_std_17 is unresolvable for 24 leaves (v13.3)
Updated Aug 28, 2026 - #439
CUDA Tile 1.3: store_masked crashes on element-wise-computed tiles
Updated Aug 28, 2026 - #442
Problem building deviceQuery with Ubuntu 26.04 package cuda-toolkit-13-1
Updated Aug 28, 2026 - #444
Suggestion: Add roofline/arithmetic-intensity analysis to matrixMul samples
Updated Aug 28, 2026 - #446
cmake add_custom_command() is not passing the ccbin flag to nvcc for the Fatbin and PTX cuda samples
Updated Aug 28, 2026 - #447
Question: readiness for Windows ARM?
Updated Aug 28, 2026 - #449
p2pBandwidthLatencyTest shows large run-to-run variance with P2P enabled due to low repeat count
Updated Aug 28, 2026 - #64
Why do the Vulkan interop examples not do ownership transfers?
bugUpdated Aug 28, 2026