Menu
SkillsMP a collecté 6 skills depuis rocm/mori. Ouvrez un skill pour examiner sa source et ses détails.
Affichage de 6 skills collectés sur 6.
Discover, diagram, and diagnose the RDMA/GPU network topology of a GPU cluster (rails, NICs, GID/RoCE config, GPU<->NIC PCIe affinity), determine whether the fabric supports cross-rail communication, and localize a failure to a fabric tier (NIC / leaf /…
Langue du texte source : anglais
Known environment issues that make MORI slow or behave oddly without being MORI bugs — currently: HIP VMM peer traffic silently falling off XGMI onto PCIe/host memory on Linux kernels built without CONFIG_DMABUF_MOVE_NOTIFY / CONFIG_PCI_P2PDMA, which hits…
Langue du texte source : anglais
Deploy and set up the MORI environment in a fresh Docker container or bare host: start the container, install ROCm dependencies, NIC userspace libraries (AINIC/Broadcom/Mellanox-NVIDIA), RDMA-core, and MORI itself. Use when the user asks to deploy MORI,…
Langue du texte source : anglais
Add a new storage medium to UMBP's distributed mode as a MediumBackend (HBM, SSD, CXL, a remote object store, a fake for tests). Covers the two shapes a backend can take — a PageMemorySource for anything page-addressable, or a full MediumBackend for a medium…
Langue du texte source : anglais
Add a new byte-moving path to UMBP's distributed mode as a TransferEngine (GPU copy, a second RDMA transport, a file/GDS engine, a fake for tests). Covers CanHandle pair-dispatch and why engines must be disjoint, the Plan/Submit/Wait split, bounce buffers,…
Langue du texte source : anglais
Write GPU kernels that move data with the CCO SDMA device API (ccoSdma: put/get/quiet/waitSignal/commit). Covers the build flags and env vars needed to turn the path on, which template arguments and coop scope to pick, how completion and signals work, and the…
Langue du texte source : anglais