Issues

• [BUG] Several runtime capability checks cache a process-wide result from the first queried device

  bug
  rapidsai/rmm#2572 · Filter by author bdice opened on Sep 29, 2026

• [FEA] Use CudaAsyncMemoryResource as the upstream for rmm.reinitialize(pool_allocator=True)

  feature requestPython
  rapidsai/rmm#2570 · Filter by author bdice opened on Sep 29, 2026
  

• [BUG] hwdecompress::is_supported() checks only the driver version, not device capability

  bugcpp
  rapidsai/rmm#2566 · Filter by author bdice opened on Sep 29, 2026
  

• [BUG] runtime_async_alloc::is_supported() caches a process-wide result from the first queried device

  bugcpp
  rapidsai/rmm#2564 · Filter by author bdice opened on Sep 29, 2026
  

• [FEA] Deprecate and remove resource ref aliases in favor of CCCL cuda::mr aliases

  breakingcppfeature request
  rapidsai/rmm#2562 · Filter by author bdice opened on Sep 28, 2026

• [FEA] Pilot nightly CI remediation

  ? - Needs Triagefeature request
  rapidsai/rmm#2556 · Filter by author bdice opened on Sep 18, 2026

• Add scheduled Lychee link checks

  rapidsai/rmm#2531 · Filter by author bdice opened on Aug 26, 2026

• [FEA] Use CUDA's current async pool as RMM's default memory resource

  feature request
  rapidsai/rmm#2497 · Filter by author bdice opened on Aug 3, 2026
  5

  

• [FEA] Improve pinned memory pool initialization time

  ? - Needs Triagefeature request
  rapidsai/rmm#2496 · Filter by author rishic3 opened on Jul 29, 2026
  2

• Handle cudaStreamGetId failures explicitly in stream_ordered_memory_resource

  rapidsai/rmm#2478 · Filter by author bdice opened on Jul 9, 2026
  1

• [FEA] Optionally compile rmm as object library

  ? - Needs Triagefeature request
  rapidsai/rmm#2463 · Filter by author baliika opened on Jul 3, 2026

• [FEA] Scope out components of rmm required to migrate all of RAPIDS to a CCCL-only dependency

  ? - Needs Triagefeature request
  rapidsai/rmm#2455 · Filter by author vyasr opened on Jun 25, 2026
  2

• CuPy missing from environment files?

  rapidsai/rmm#2428 · Filter by author jakirkham opened on May 29, 2026
  1

• Move system memory support check to runtime_capabilities.hpp

  rapidsai/rmm#2415 · Filter by author bdice opened on May 22, 2026
  

• Fix callback memory resource allocation API to match CCCL and support alignment

  rapidsai/rmm#2397 · Filter by author bdice opened on May 17, 2026

• Key arena per-stream state by CUDA stream ID

  bug
  rapidsai/rmm#2394 · Filter by author bdice opened on May 17, 2026

• Hide memory resource impl symbols

  rapidsai/rmm#2390 · Filter by author bdice opened on May 17, 2026

• [FEA] Expand RMM benchmarking

  feature request
  rapidsai/rmm#2385 · Filter by author bdice opened on May 15, 2026
  2

  

• [FEA] Use stream protocol and cudaStream_t for Cython bindings to rmm C++

  ? - Needs Triagefeature request
  rapidsai/rmm#2359 · Filter by author vyasr opened on Apr 14, 2026
  

• Replace cudf JNI custom tracking adaptor with statistics_resource_adaptor

  rapidsai/rmm#2358 · Filter by author bdice opened on Apr 14, 2026

• Enable stricter compiler warnings (-Wshadow, -Wnon-virtual-dtor, -Woverloaded-virtual)

  rapidsai/rmm#2338 · Filter by author maxwbuckley opened on Mar 25, 2026

• Enforce alignment in each memory resource

  rapidsai/rmm#2336 · Filter by author bdice opened on Mar 25, 2026

• [BUG] Managed memory allocated on gpu side doesnt show up in nvtop/nvidia-smi when used with pytorch+unsloth

  ? - Needs Triagebug
  rapidsai/rmm#2259 · Filter by author fish4terrisa-MSDSM opened on Feb 20, 2026
  3

• [BUG] cccl_async_resource_ref conversion operator triggers nvcc error 20013 with C++20 constexpr std::vector

  ? - Needs Triagebug
  rapidsai/rmm#2255 · Filter by author PointKernel opened on Feb 17, 2026

• [FEA] Migrate RAPIDS to CCCL cuda::stream_ref / cuda::stream in C++ and cuda.core streams in Python

  ? - Needs Triagefeature request
  rapidsai/rmm#2250 · Filter by author bdice opened on Feb 10, 2026
  1