From bc1f20dd78e4b8079a4215d505584595b4542bca Mon Sep 17 00:00:00 2001
From: jgrusewski <jgrusewski@localhost>
Date: Fri, 3 Apr 2026 19:21:30 +0200
Subject: [PATCH] fix: add nvcc to ci-builder-cpu image for cudarc .cubin
 precompilation

cuda-nvcc-12-4 + cuda-cudart-dev-12-4 (~400MB) added to the CPU builder.
Reverts compile-and-deploy back to ci-builder-cpu (was briefly switched
to full ci-builder which is 6GB+ and unnecessary).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
---
 infra/docker/Dockerfile.ci-builder-cpu          | 11 ++++++++++-
 infra/k8s/argo/compile-and-deploy-template.yaml |  2 +-
 2 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/infra/docker/Dockerfile.ci-builder-cpu b/infra/docker/Dockerfile.ci-builder-cpu
index 07b39969e..b7948cfe2 100644
--- a/infra/docker/Dockerfile.ci-builder-cpu
+++ b/infra/docker/Dockerfile.ci-builder-cpu
@@ -71,5 +71,14 @@ RUN curl -fsSL https://github.com/gruntwork-io/terragrunt/releases/download/v0.7
 RUN curl -fsSL "https://dl.k8s.io/release/v1.31.4/bin/linux/amd64/kubectl" -o /usr/local/bin/kubectl \
     && chmod +x /usr/local/bin/kubectl
 
+# CUDA nvcc compiler only (no runtime/driver) — needed for cudarc .cubin precompilation
+# ~400MB vs ~6GB for the full cuda-devel image
+RUN curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/x86_64/cuda-keyring_1.1-1_all.deb -o /tmp/cuda-keyring.deb \
+    && dpkg -i /tmp/cuda-keyring.deb && rm /tmp/cuda-keyring.deb \
+    && apt-get update \
+    && apt-get install -y --no-install-recommends cuda-nvcc-12-4 cuda-cudart-dev-12-4 \
+    && rm -rf /var/lib/apt/lists/*
+ENV PATH="/usr/local/cuda-12.4/bin:${PATH}"
+
 # Verify
-RUN rustc --version && cargo --version && git --version && protoc --version && sccache --version && make --version && mold --version && rclone --version && tofu --version && terragrunt --version && kubectl version --client
+RUN rustc --version && cargo --version && git --version && protoc --version && sccache --version && make --version && mold --version && rclone --version && tofu --version && terragrunt --version && kubectl version --client && nvcc --version
diff --git a/infra/k8s/argo/compile-and-deploy-template.yaml b/infra/k8s/argo/compile-and-deploy-template.yaml
index b5efbdd8a..36fd98762 100644
--- a/infra/k8s/argo/compile-and-deploy-template.yaml
+++ b/infra/k8s/argo/compile-and-deploy-template.yaml
@@ -157,7 +157,7 @@ spec:
           operator: Exists
           effect: NoSchedule
       container:
-        image: gitlab-registry.foxhunt.svc.cluster.local:5000/root/foxhunt/ci-builder:latest
+        image: gitlab-registry.foxhunt.svc.cluster.local:5000/root/foxhunt/ci-builder-cpu:latest
         imagePullPolicy: Always
         command: ["/bin/sh", "-c"]
         env: