Skip to content

feat: Detect when DeepEP is not compiled for H100 to refrain from using RDMA#1275

Merged
akoumpa merged 5 commits intoNVIDIA-NeMo:mainfrom
pzelasko:pzelasko/nano-v3-embed-with-a100-deepep
Feb 13, 2026
Merged

feat: Detect when DeepEP is not compiled for H100 to refrain from using RDMA#1275
akoumpa merged 5 commits intoNVIDIA-NeMo:mainfrom
pzelasko:pzelasko/nano-v3-embed-with-a100-deepep

Conversation

@pzelasko
Copy link
Copy Markdown
Contributor

What does this PR do ?

(re-opened #1264 but against main branch)

Detect when DeepEP is not compiled for H100 to refrain from using RDMA (specifically, when it's compiled with -DDISABLE_NVSHMEM)

Changelog

  • Add check for when DeeEP is not compiled for H100 (assuming A100)

Before your PR is "Ready for review"

Pre checks:

  • Make sure you read and followed Contributor guidelines
  • Did you write any new necessary tests?
  • Did you add or update any necessary documentation?

If you haven't finished some of the above items you can still open "Draft" PR.

Additional Information

  • Related to # (issue)

adil-a and others added 5 commits February 12, 2026 20:14
Signed-off-by: adil-a <adil.asif2000@hotmail.com>
Signed-off-by: adil-a <adil.asif2000@hotmail.com>
Signed-off-by: adil-a <adil.asif2000@hotmail.com>
Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>
@copy-pr-bot
Copy link
Copy Markdown

copy-pr-bot Bot commented Feb 13, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@akoumpa
Copy link
Copy Markdown
Contributor

akoumpa commented Feb 13, 2026

/ok to test 3b81ae0

@pzelasko pzelasko marked this pull request as ready for review February 13, 2026 15:51
Copy link
Copy Markdown
Contributor

@akoumpa akoumpa left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM, thank you @pzelasko

@akoumpa akoumpa merged commit f47b4e8 into NVIDIA-NeMo:main Feb 13, 2026
51 checks passed
linnanwang pushed a commit that referenced this pull request Apr 24, 2026
…ng RDMA (#1275)

* fix

Signed-off-by: adil-a <adil.asif2000@hotmail.com>

* rename

Signed-off-by: adil-a <adil.asif2000@hotmail.com>

* unit tests

Signed-off-by: adil-a <adil.asif2000@hotmail.com>

* Detect when DeepEP is not compiled for H100 to refrain from using RDMA

Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>

---------

Signed-off-by: adil-a <adil.asif2000@hotmail.com>
Signed-off-by: Piotr Żelasko <pzelasko@nvidia.com>
Co-authored-by: adil-a <adil.asif2000@hotmail.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants