deepseek-ai/DeepEP
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepEP.git synced 2026-11-05 00:47:32 +00:00
Code Issues 308 Projects Packages Wiki Activity

[GH-ISSUE #568] Compilation Issue with DeepEP under CUDA 13.0 #351

New issue
Closed
opened 2026-10-26 09:58:58 +00:00 by bsakaguchi · 1 comment
bsakaguchi commented 2026-10-26 09:58:58 +00:00
Owner
Copy link

Originally created by @zhihui1084 on GitHub (Jan 20, 2026).
Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/568

I encountered the following error when compiling and installing using python setup.py install after running pip install nvidia-nvshmem-cu13:

In file included from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/barrier.cuh:17,
                 from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/alltoall.cuh:13,
                 from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/defines.cuh:10,
                 from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/device/nvshmem_coll_defines.cuh:9,
                 from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/nvshmem.h:24,
                 from /home/dpsk_a2a/DeepEP/csrc/kernels/configs.cuh:86,
                 from /home/dpsk_a2a/DeepEP/csrc/kernels/api.cuh:5,
                 from /home/dpsk_a2a/DeepEP/csrc/config.hpp:3,
                 from /home/dpsk_a2a/DeepEP/csrc/deep_ep.hpp:15,
                 from /home/dpsk_a2a/DeepEP/csrc/deep_ep.cpp:1:
/usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/device_host/nvshmem_tensor.h:37:10: fatal error: cuda/std/tuple: No such file or directory
   37 | #include "cuda/std/tuple"
      |          ^~~~~~~~~~~~~~~~
compilation terminated.

Is there any idea about? Thanks.

Originally created by @zhihui1084 on GitHub (Jan 20, 2026). Original GitHub issue: https://github.com/deepseek-ai/DeepEP/issues/568 I encountered the following error when compiling and installing using `python setup.py install` after running `pip install nvidia-nvshmem-cu13`: ``` In file included from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/barrier.cuh:17, from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/alltoall.cuh:13, from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/non_abi/device/coll/defines.cuh:10, from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/device/nvshmem_coll_defines.cuh:9, from /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/nvshmem.h:24, from /home/dpsk_a2a/DeepEP/csrc/kernels/configs.cuh:86, from /home/dpsk_a2a/DeepEP/csrc/kernels/api.cuh:5, from /home/dpsk_a2a/DeepEP/csrc/config.hpp:3, from /home/dpsk_a2a/DeepEP/csrc/deep_ep.hpp:15, from /home/dpsk_a2a/DeepEP/csrc/deep_ep.cpp:1: /usr/local/lib/python3.12/dist-packages/nvidia/nvshmem/include/device_host/nvshmem_tensor.h:37:10: fatal error: cuda/std/tuple: No such file or directory 37 | #include "cuda/std/tuple" | ^~~~~~~~~~~~~~~~ compilation terminated. ``` Is there any idea about? Thanks.
bsakaguchi commented 2026-10-26 09:59:00 +00:00
Author
Owner
Copy link

@zhihui1084 commented on GitHub (Jan 20, 2026):

In my NGC 25.08 environment, the include directory is indeed missing the cuda/std/tuple.h file. The problem was resolved by executing the following steps and downgrading the nvshmem version to 3.4.5:

ln -s /usr/local/cuda/include/cccl/cuda /usr/local/cuda/include/cuda
pip install nvidia-nvshmem-cu13==3.4.5
<!-- gh-comment-id:3771845488 --> @zhihui1084 commented on GitHub (Jan 20, 2026): In my NGC 25.08 environment, the include directory is indeed missing the cuda/std/tuple.h file. The problem was resolved by executing the following steps and downgrading the nvshmem version to 3.4.5: ``` ln -s /usr/local/cuda/include/cccl/cuda /usr/local/cuda/include/cuda pip install nvidia-nvshmem-cu13==3.4.5 ```
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
hybrid-ep
chore/remove-clang-format-yapf
v2.5
nvDev
try-fix-cu128
mori-ep
antgroup-opt
compat-old-torch
add-linter
eager-rdma
v1.2.1
Labels
Clear labels   
pull-request
Mirrored from GitHub Pull Request

No labels
pull-request
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
deepseek-ai/DeepEP#351
No description provided.