xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-30 20:56:37 +08:00
Code Issues Packages Projects Releases Wiki Activity

[NIXL][Bugfix] Fix NIXL/RDMA registration failure over CuMemAllocator (#29569)

Browse Source 
Signed-off-by: Somoku <linsh0@protonmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
This commit is contained in:
Sheng Lin 2025-12-17 17:52:58 +08:00 committed by GitHub
parent 3b1d440ede
commit f4e884f222
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
csrc/cumem_allocator.cpp
View File

@ -107,6 +107,16 @@ void create_and_map(unsigned long long device, ssize_t size, CUdeviceptr d_mem,
prop.location.id = device; prop.location.id = device;
prop.allocFlags.compressionType = CU_MEM_ALLOCATION_COMP_NONE; prop.allocFlags.compressionType = CU_MEM_ALLOCATION_COMP_NONE;
#ifndef USE_ROCM
int flag = 0;
CUDA_CHECK(cuDeviceGetAttribute(
&flag, CU_DEVICE_ATTRIBUTE_GPU_DIRECT_RDMA_WITH_CUDA_VMM_SUPPORTED,
device));
if (flag) { // support GPUDirect RDMA if possible
prop.allocFlags.gpuDirectRDMACapable = 1;
}
#endif
#ifndef USE_ROCM #ifndef USE_ROCM
// Allocate memory using cuMemCreate // Allocate memory using cuMemCreate
CUDA_CHECK(cuMemCreate(p_memHandle, size, &prop, 0)); CUDA_CHECK(cuMemCreate(p_memHandle, size, &prop, 0));