Skip to content

[c10d][nvshmem] add nvshmem build rules and dependency for libtorch_cuda #159562

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

jushg
Copy link
Contributor

@jushg jushg commented Jul 31, 2025

Summary:
Add guarded build option for nvshmem-related c10d code with -c fbcode.caffe2_use_nvshmem

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:

  • torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu
  • torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu

Differential Revision: D78813568

cc @H-Huang @awgu @wanchaol @fegin @fduwjj @wz337 @wconstab @d4l3k @pragupta

Copy link

pytorch-bot bot commented Jul 31, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/159562

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit 810e75b with merge base 76a0609 (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

@pytorch-bot pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Jul 31, 2025
jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from b7383f9 to 63a7932 Compare August 4, 2025 10:02
@pytorch-bot pytorch-bot bot added ciflow/h100-symm-mem oncall: distributed Add this issue/PR to distributed oncall triage queue release notes: distributed (c10d) release notes category labels Aug 4, 2025
@jushg jushg force-pushed the export-D78813568 branch from 63a7932 to 0915a4e Compare August 4, 2025 10:03
jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from 0915a4e to 7b04fa3 Compare August 4, 2025 10:06
Copy link

linux-foundation-easycla bot commented Aug 4, 2025

CLA Signed

The committers listed above are authorized under a signed CLA.

  • ✅ login: jushg / name: Trong Tan (810e75b)

@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from 7b04fa3 to 008edf6 Compare August 4, 2025 10:10
jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from 008edf6 to b327369 Compare August 4, 2025 14:02
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

jushg added a commit to jushg/pytorch that referenced this pull request Aug 4, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch 2 times, most recently from 8594278 to cf69767 Compare August 5, 2025 09:47
pytorch-bot bot pushed a commit that referenced this pull request Aug 5, 2025
…uda (#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

jushg added a commit to jushg/pytorch that referenced this pull request Aug 5, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from cf69767 to 4417902 Compare August 5, 2025 09:51
@iremyux
Copy link
Collaborator

iremyux commented Aug 6, 2025

Adding ciflow/win-arm64 label to trigger Windows Arm64 CI and its test purposes - nothing about this PR specifically. (It should not effect the acceptance of the PR even if it fails.)

@iremyux iremyux added the ciflow/win-arm64 Trigger Windows Arm64 CI Workflows label Aug 6, 2025
@jushg jushg force-pushed the export-D78813568 branch from 4417902 to d4976c5 Compare August 6, 2025 15:10
pytorch-bot bot pushed a commit that referenced this pull request Aug 6, 2025
…uda (#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

@jushg jushg force-pushed the export-D78813568 branch from d4976c5 to 3bec3d6 Compare August 6, 2025 15:19
jushg added a commit to jushg/pytorch that referenced this pull request Aug 6, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch from 3bec3d6 to d02011f Compare August 6, 2025 19:51
jushg added a commit to jushg/pytorch that referenced this pull request Aug 6, 2025
…uda (pytorch#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

jushg added a commit to jushg/pytorch that referenced this pull request Aug 6, 2025
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@jushg jushg force-pushed the export-D78813568 branch 2 times, most recently from 7f21f66 to e7a150d Compare August 11, 2025 19:43
pytorch-bot bot pushed a commit that referenced this pull request Aug 11, 2025
…uda (#159562)

Summary:

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
…uda (pytorch#159562)

Summary:
Pull Request resolved: pytorch#159562

Add guarded build option for nvshmem-related c10d code with `-c fbcode.caffe2_use_nvshmem`

Guarded clause include nvshmem device + host code (static-linked) + these 2 files:
- `torch/csrc/distributed/c10d/symm_mem/NVSHMEMSymmetricMemory.cu`
-    `torch/csrc/distributed/c10d/symm_mem/nvshmem_extension.cu`

Differential Revision: D78813568
@facebook-github-bot
Copy link
Contributor

This pull request was exported from Phabricator. Differential Revision: D78813568

@jushg jushg force-pushed the export-D78813568 branch from e7a150d to 810e75b Compare August 11, 2025 19:47
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
ciflow/h100-symm-mem ciflow/trunk Trigger trunk jobs on your pull request ciflow/win-arm64 Trigger Windows Arm64 CI Workflows fb-exported oncall: distributed Add this issue/PR to distributed oncall triage queue release notes: distributed (c10d) release notes category
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants