* feat: add Hygon DCU/DTK and Iluvatar CoreX platform support
Add build system and runtime support for two CUDA-compatible domestic
accelerator platforms:
- Hygon DCU with DTK SDK (USE_HYGON, /opt/dtk/cuda/cuda-11/)
- Iluvatar CoreX SDK (USE_COREX, /usr/local/corex/)
Both platforms expose CUDA-compatible APIs, so the integration follows
the same pattern as existing CUDA-like platforms (MUSA, MACA): add the
new macros to all platform guard chains and register SDK paths in CMake.
---------
Co-authored-by: KarmaD7 <KarmaD7@users.noreply.github.com>
* [Store] Add lock-free MmapArena allocator for buffer mmap path
Replace per-allocation mmap() syscalls in allocate_buffer_mmap_memory()
with a lock-free atomic bump allocator (MmapArena). Pre-allocates a
configurable pool (default 64GB) and serves allocations via CAS loop,
reducing allocation latency from ~1us (mmap syscall) to ~50ns (atomic).
Allocation lifecycle is static: all callers (ClientBufferAllocator,
global segments in RealClient::setup_internal) allocate at startup and
free at shutdown. The arena outlives all allocations, so the bump-only
(no individual free) design is correct for this usage pattern.
Feature-flagged via gflags:
--use_mmap_arena_allocator (default: true)
--mmap_arena_pool_size (default: 64GB)
Falls back to direct mmap() when arena is disabled, fails to init,
or is exhausted.
Cherry-picked from flow-ipc-poc branch (utils.cpp perf path only).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
* [Store] Fix three correctness issues in MmapArena
1. Honor caller's alignment contract: allocate() now accepts a
per-call alignment parameter and uses max(arena default, caller
request). allocate_buffer_mmap_memory() forwards its alignment
argument to the arena. Previously, the caller's alignment was
silently ignored — the arena always used 64-byte alignment
regardless of what the caller requested.
2. Remove MAP_POPULATE from arena pool mmap: the default pool is
64GB but callers typically use only a fraction (e.g. 4GB of
segments). MAP_POPULATE would pre-fault all 64GB of pages upfront,
causing seconds of startup delay and potentially triggering OOM
on machines with less physical memory. Pages now fault on demand.
3. Make alignment_ atomic and store it BEFORE the CAS on pool_base_:
previously alignment_ was a plain size_t written AFTER the release
CAS, so the store was not in the happens-before relationship
established by the acquire-release pair on pool_base_. Now both
alignment_ and pool_size_ are stored before the CAS with the
release fence guaranteeing their visibility to readers.
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
* [TRANSFER_ENGINE] align USE_MACA with MUSA GPU paths and docs
Extend CUDA/MUSA/HIP preprocessor guards to include USE_MACA across TE core, tests, and examples; add maca:0 to NIC priority matrix JSON; exclude USE_MACA from USE_MNNVL auto CUDA enablement. Document MACA build (MACA_HOME, USE_MACA, optional MACA_RUNTIME_LIBS) in English and Chinese build guides. Fix maca.h cudaPointerAttributes mapping to mcPointerAttribute_t for MACA SDK compatibility.
When building with USE_EFA=ON, auto_discover is disabled to prevent
RDMA transport installation (QP creation fails on EFA devices). This
means TCP transport is also not installed automatically. Add explicit
TCP transport installation for non-EFA protocols in the EFA build path.
Documentation changes:
- build.md: Add USE_EFA option and clarify USE_CUDA default/purpose
- supported-protocols.md: Add EFA as a supported protocol
- efa_transport.md: Add USE_CUDA=ON to build command, document GPU
memory requirement
Co-authored-by: whn09 <whn09@users.noreply.github.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Teng Ma <teng-ma@linux.alibaba.com>
* Add early mem backend detection method in NVLINK_allocator
Add early detection method for sglang NVLINK_allocator to avoid CuMemCreate
Use enumerate type to indicate mem backend type
format check use pre-commit
* Change enumerate type in allocator.py for different mem backend
* Isolate nvlink intraNode transport from nvlink_transport and modify corresponding transfer_engine_bench
* IntraNode transport isolation to be compatible with SGlang
* Modify the code style using pre-commit check
* isolate intraNode nvlink from previous nvlink_transport
* [TE] Revert to previous nvlink_transport
* Delete extra log and refine the code format
* Discard revert due to new related PR has been submitted
* Change the Code format to align with main branch
* Change to clang-format
* Modify transfer_engine_bench to be compatible with protocol --nvlink_intra
* Delete useless function in intranode_nvlink.cpp
* Used for rerun CI
* Modify the intraNode isolation to be compatible with transfer_bench and transfer_engine_impl
* isolate intraNode header file from mnnvl.h
* Delete specific instaince type for intraNode
---------
Co-authored-by: 百麒 <yaozhong.lyz@alibaba-inc.com>
* Improve AMD HIP support with hipify-perl
This commit improves AMD GPU support by migrating to hipify-perl
for automatic CUDA-to-HIP code conversion at build time, and extends HIP
compatibility to the NVLink transport layer.
* [TE] Add HIP support to nvlink-allocator with hipcc compilation
* Add USE_HIP to the documentation
* [TE/NVLINK] Check CU_DEVICE_ATTRIBUTE_HANDLE_TYPE_FABRIC_SUPPORTED only if USE_CUDA
* [TE/EXAMPLE] Fix compiler error if USE_MNNVL and USE_HIP
* Address review comments
* ci: add non-CUDA release workflow and update documentation
- Add release-non-cuda.yaml workflow for building non-CUDA version
- Modify build_wheel.sh to support dynamic package name modification
- Update README.md and docs with installation instructions for both versions
- CUDA version includes Mooncake-EP and GPU topology detection
- Non-CUDA version for environments without CUDA dependencies
* Update README.md
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
---------
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Aoi