dylanyunlon 93e498197a fix: thread_local reentrant guard — prevent cudaMalloc infinite recursion ...

CUB CachingDeviceAllocator 内部在 cache miss 时调 cudaMalloc，
被我们的 LD_PRELOAD 再次拦截 → DeviceAllocate → cudaMalloc → 无限递归 → segfault。

加 thread_local bool inside_cub 标志:
  外部调用 → CUB allocator (带缓存)
  CUB 内部调用 → 直接走 dlsym(RTLD_NEXT) 的真实 cudaMalloc

2026-08-13 09:42:26 +00:00

..

include

feat: CCCL CachingDeviceAllocator preload — 完整依赖链 288 files

2026-08-13 09:26:41 +00:00

build_cccl_preload.sh

fix: CUB_NS_QUALIFIER for wrapped namespace + drop _CCCL_COMPILER_GCC

2026-08-13 09:31:45 +00:00

cccl_allocator_preload.cu

fix: thread_local reentrant guard — prevent cudaMalloc infinite recursion

2026-08-13 09:42:26 +00:00

verify_preload.py

feat: CCCL CachingDeviceAllocator LD_PRELOAD — bypass CoreX expandable_segments ASSERT

2026-08-13 09:26:41 +00:00