Merge branch 'main' into rjain/trtllm-spec-dec

Signed-off-by: jain-ria <riajain@NVIDIA.com>

Author: jain-ria

.devcontainer/post-create.sh

@@ -52,10 +52,6 @@ export CARGO_TARGET_DIR=$HOME/dynamo/.build/target
 cargo build --locked --profile dev --features mistralrs
 cargo doc --no-deps
 
-# create symlinks for the binaries in the deploy directory
-mkdir -p $HOME/dynamo/deploy/sdk/src/dynamo/sdk/cli/bin
-ln -sf $HOME/dynamo/.build/target/debug/dynamo-run $HOME/dynamo/deploy/sdk/src/dynamo/sdk/cli/bin/dynamo-run
-
 # install the python bindings
 cd $HOME/dynamo/lib/bindings/python && retry maturin develop
 

.github/workflows/pre-merge-rust.yml

@@ -47,16 +47,34 @@ jobs:
       run: |
         sudo apt-get update -y
         sudo apt-get install -y protobuf-compiler
-    - name: Cache Cargo Registry
-      uses: actions/cache@v3
-      with:
-        path: ~/.cargo/registry
-        key: ${{ runner.os }}-cargo-registry-${{ github.head_ref || github.ref_name }}-${{ hashFiles('**/Cargo.lock') }}
-    - name: Cache Cargo Tools
-      uses: actions/cache@v3
+
+    - name: Free runner disk space
+      run: |
+        echo "Disk space before:"
+        df -h
+        rm -rf \
+          /host_usr/share/dotnet /host_usr/local/lib/android /opt/ghc \
+          /host_usr/local/share/powershell /host_usr/share/swift /host_usr/local/.ghcup \
+          /host_usr/lib/jvm /opt/hostedtoolcache
+        echo "Disk space after:"
+        df -h
+
+    # TODO: Caching target/ dir handles most of the build caching improvements
+    #       currently, so sccache artifacts are mostly duplicates wasting disk space.
+    #       Revisit this to see if sccache can improve current caching behavior.
+    # TODO: Consider single target/ directory for all subdirectories/crates instead.
+    - name: Cache cargo artifacts
+      uses: actions/cache@v4
       with:
-        path: ~/.cargo/bin
-        key: ${{ runner.os }}-cargo-tools-${{ github.head_ref || github.ref_name }}-${{ hashFiles('**/Cargo.lock') }}
+        path: |
+          ~/.cargo/bin/
+          ~/.cargo/registry/index/
+          ~/.cargo/registry/cache/
+          ~/.cargo/git/db/
+          ${{ matrix.dir }}/target/
+        key: ${{ runner.os }}-cargo-${{ matrix.dir }}-${{ hashFiles('**/Cargo.lock') }}
+        restore-keys: ${{ runner.os }}-cargo-${{ matrix.dir }}-
+
     - name: Install Rust in dev environment
       # Install Rust only to run GitHub Local Actions in (dev environment) using the `ACT` environment variable.
       # See act usage: https://nektosact.com/introduction.html
@@ -68,10 +86,6 @@ jobs:
         echo "$HOME/.cargo/bin" >> $GITHUB_PATH
     - name: Set up Rust Toolchain Components
       run: rustup component add rustfmt clippy
-    - name: Run Cargo Check
-      working-directory: ${{ matrix.dir }}
-      run: cargo check --locked
-      timeout-minutes: 5
     - name: Verify Code Formatting
       working-directory: ${{ matrix.dir }}
       run: cargo fmt -- --check
@@ -83,6 +97,10 @@ jobs:
       run: |
         cargo-deny --version || cargo install cargo-deny@0.16.4
         cargo-deny --no-default-features check --hide-inclusion-graph licenses bans --config ${{ github.workspace }}/deny.toml
+    # Have an explicit step to build tests first to separate time spent on build vs execution.
+    - name: Compile Tests
+      working-directory: ${{ matrix.dir }}
+      run: cargo test --locked --no-run
     - name: Run Doc Tests
       working-directory: ${{ matrix.dir }}
       run: cargo doc --no-deps && cargo test --locked --doc

.github/workflows/trigger_ci.yml

@@ -47,11 +47,9 @@ jobs:
         filters: |
           vllm:
             - 'container/Dockerfile.vllm'
-            - 'container/Dockerfile.vllm_nixl'
-            - 'examples/python/llm/**'
-            - 'examples/python_rs/llm/**'
             - 'container/deps/requirements.vllm.txt'
             - 'container/deps/vllm/**'
+            - 'components/backends/vllm/**'
             - 'tests/serve/test_vllm.py'
           trtllm:
             - 'container/Dockerfile.tensorrt_llm'
@@ -62,6 +60,12 @@ jobs:
             - 'tests/serve/test_trtllm.py'
           sdk:
             - 'deploy/**'
+          sglang:
+            - 'container/Dockerfile.sglang'
+            - 'container/Dockerfile.sglang-deepep'
+            - 'components/backends/sglang/**'
+            - 'container/build.sh'
+            - 'tests/serve/test_sglang.py'
     - name: Check if Validation Workflow has run
       id: check_workflow
       uses: actions/github-script@v6
@@ -104,6 +108,10 @@ jobs:
           ci_variables["RUN_TENSORRTLLM"]="true"
         fi
 
+        if [ "${{ steps.src_changes.outputs.sglang }}" == "true" ]; then
+          ci_variables["RUN_SGLANG"]="true"
+        fi
+
         if [ "${{ steps.src_changes.outputs.sdk }}" == "true" ]; then
           ci_variables["RUN_SDK_CI"]="true"
         fi

.gitignore

@@ -89,6 +89,4 @@ generated-values.yaml
 TensorRT-LLM
 
 # Local build artifacts for devcontainer
-.build/
-# Copied binaries to ignore
-deploy/sdk/src/dynamo/sdk/cli/bin
+.build/