Skip to content

build(container): backport EFA build fixes (#9705, #9727) to release/1.2.1 - #10425

Merged
saturley-hall merged 3 commits into
release/1.2.1from
jihao/cherrypick-efa-build-fixes-1.2.1
Jun 11, 2026
Merged

build(container): backport EFA build fixes (#9705, #9727) to release/1.2.1#10425
saturley-hall merged 3 commits into
release/1.2.1from
jihao/cherrypick-efa-build-fixes-1.2.1

Conversation

@jh-nv

@jh-nv jh-nv commented Jun 8, 2026

Copy link
Copy Markdown
Contributor

Summary

Cherry-pick of two EFA container build fixes from main to release/1.2.1:

Together these fix the Dynamo EFA container build (GDRCopy kernel ≥6.15 compatibility, configurable libfabric repo, and the libfabric v2.5.1 overlay needed for the EFA RDMA path).

Original PRs

Conflicts resolved

Cherry-pick of #9727 conflicted in two files; resolved as follows:

  • container/context.yaml — took the incoming side (configurable nixl_libfabric_repo + nixl_libfabric_ref: v2.5.1); this is the core of the fix.
  • container/templates/wheel_builder.Dockerfile — combined both sides: kept release branch's git yasm (needed for the ffmpeg/libvpx build present on 1.2.x) and added feat(container): configurable libfabric repo + v2.5.1 overlay for EFA #9727's --setopt=tsflags=nocontexts.

#9705 applied cleanly.

Test plan

  • CI passes
  • EFA container build succeeds (libfabric v2.5.1 + GDRCopy v2.5.2)

🤖 Generated with Claude Code

…#9705)

Signed-off-by: Yifan Jiang <19356972+yifjiang@users.noreply.github.com>
Signed-off-by: Jie Hao <jihao@nvidia.com>
@jh-nv
jh-nv requested a review from a team as a code owner June 8, 2026 19:28
@jh-nv
jh-nv requested a review from a team June 8, 2026 19:28
@jh-nv
jh-nv requested review from a team as code owners June 8, 2026 19:28
Comment thread container/context.yaml Outdated
yifjiang and others added 2 commits June 8, 2026 16:16
…#9727)

Signed-off-by: Yifan Jiang <yifjiang@users.noreply.github.com>
Co-authored-by: Yifan Jiang <yifjiang@users.noreply.github.com>
Signed-off-by: Jie Hao <jihao@nvidia.com>
Signed-off-by: Erez Zarum <erezz@amazon.com>
Signed-off-by: Jie Hao <jihao@nvidia.com>
@jh-nv
jh-nv force-pushed the jihao/cherrypick-efa-build-fixes-1.2.1 branch from 4945715 to 6a3b62a Compare June 8, 2026 20:16

@dynamo-ops dynamo-ops left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Review: backport of EFA build fixes to release/1.2.1

The diff is clean. Cherry-picks of #9705, #9727, and #10047 applied correctly to the release branch. Conflict resolution in wheel_builder.Dockerfile (keeping yasm + adding --setopt=tsflags=nocontexts) is correct.

Two minor notes:

  1. PR description lists #9705 + #9727 only, but the actual diff includes #10047 (libfabric2.5.1 for EFA and TRTLLM) as well — the rm -rf /opt/amazon/efa && cp -Pfr replacement approach and the trtllm NIXL plugin block are from that PR. Consider updating the description for traceability.

  2. Dead exports in trtllm RUN block (cosmetic, non-blocking):

    export LD_PRELOAD=/opt/nvidia/nvda_nixl/lib64/libnixl.so && \
    export NIXL_PLUGIN_DIR=/opt/nvidia/nvda_nixl/lib64/plugins && \

    These have no effect inside RUN (the ENV statements below handle persistence). Could be removed to reduce confusion, but not a real issue since they match the subsequent ENV values.

@dynamo-ops dynamo-ops left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Previous review comments have been addressed. Approving.

@saturley-hall

Copy link
Copy Markdown
Member

Test failures are unrelated to change.

@saturley-hall
saturley-hall merged commit 82d7af9 into release/1.2.1 Jun 11, 2026
87 of 91 checks passed
@saturley-hall
saturley-hall deleted the jihao/cherrypick-efa-build-fixes-1.2.1 branch June 11, 2026 01:53
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants