minke.yu c74a4037fb mooncake: probe pointers via cuda.bindings instead of ctypes libcudart ...

The ctypes-based probing loaded a second libcudart copy into the process,
which corrupted CUDA/runtime state and segfaulted the scheduler processes
shortly after registration (reproducible on prefill ranks, intermittent on
decode). cuda.bindings.runtime.cudaPointerGetAttributes is a properly
typed binding and boots cleanly on both roles.

2026-09-21 12:46:25 +08:00

..

sglang

mooncake: probe pointers via cuda.bindings instead of ctypes libcudart

2026-09-21 12:46:25 +08:00

MANIFEST.in

[Kernel] Move sgl-kernel under sglang.kernels.aot (#32648)

2026-07-29 17:25:00 +08:00

pyproject_cpu.toml

[CPU] Upgrade torch cpu dependencies to version 2.14. (#37939)

2026-09-17 14:01:43 +08:00

pyproject_npu.toml

[Deps] Raise smg-grpc-servicer floor to >=0.9.0 to unbreak SMG E2E CI (#38801)

2026-09-10 13:55:36 -07:00

pyproject_other.toml

[Deps] Raise smg-grpc-servicer floor to >=0.9.0 to unbreak SMG E2E CI (#38801)

2026-09-10 13:55:36 -07:00

pyproject_xpu.toml

[Intel XPU] Enable fused_moe_triton tuning on XPU and add tuned DeepSeek-OCR-2 configs (#28723)

2026-09-16 10:20:22 +08:00

pyproject.toml

[diffusion] chore: update Cache-DiT to 1.5.1 for DMD Calibrator, SVDQuant DQ, etc (#40104)

2026-09-19 11:14:33 +08:00

setup.py

[Rust] Rename mem-cache to sglang-radix-tree (#37290)

2026-09-01 03:03:44 -07:00