release v0.5.0 (#401)

Signed-off-by: Cyrill Leutwiler <bigcyrill@hotmail.com>

.cargo/config.toml

@@ -10,8 +10,11 @@ rustflags = [
     "-Clink-arg=-sWASM_ASYNC_COMPILATION=0",
     "-Clink-arg=-sDYNAMIC_EXECUTION=0",
     "-Clink-arg=-sALLOW_TABLE_GROWTH=1",
-    "-Clink-arg=--js-library=js/embed/soljson_interface.js",
+    "-Clink-arg=--js-library=js/emscripten/embed/soljson_interface.js",
-    "-Clink-arg=--pre-js=js/embed/pre.js",
+    "-Clink-arg=--pre-js=js/emscripten/embed/pre.js",
     "-Clink-arg=-sSTACK_SIZE=128kb",
     "-Clink-arg=-sNODEJS_CATCH_EXIT=0"
 ]
+
+[build]
+rustdocflags = ["-D", "warnings"]

.github/actions/get-emsdk/action.yml

@@ -3,7 +3,7 @@ inputs:
   version:
     description: ""
     required: false
-    default: "3.1.64"
+    default: "4.0.9"
 
 
 runs:

.github/actions/get-solc/action.yml

@@ -19,7 +19,7 @@ runs:
       shell: bash
       run: |
         mkdir -p solc
-        curl -sSL --output solc/solc https://github.com/ethereum/solidity/releases/download/v0.8.29/${SOLC_NAME}
+        curl -sSL --output solc/solc https://github.com/ethereum/solidity/releases/download/v0.8.30/${SOLC_NAME}
 
     - name: Make Solc Executable
       if: ${{ runner.os == 'Windows' }}

.github/scripts/json_generator.py

@@ -1,3 +1,13 @@
+#!/usr/bin/env python3
+"""
+This script generates JSON files for different platforms based on GitHub release data.
+It fetches release information from a specified GitHub repository and tag,
+parses the release assets, and generates JSON files for each platform with relevant metadata.
+It also handles checksum files and updates a list.json file for each platform.
+It requires the GITHUB_TOKEN environment variable to be set for authentication.
+Usage:
+    python json_generator.py <repo> <tag>
+"""
 import os
 import sys
 import json

.github/scripts/json_generator_nightly.py

@@ -0,0 +1,155 @@
+#!/usr/bin/env python3
+"""
+This script generates JSON files for different platforms based on GitHub data.
+Requires the GITHUB_SHA, FIRST_SOLC_VERSION, LAST_SOLC_VERSION, TAG and FILEPATH environment variables to be set.
+Usage:
+    python json_generator_nightly.py
+"""
+import os
+import sys
+import json
+from datetime import datetime
+
+def validate_env_variables():
+    """Validate that environment variables are set."""
+    if "GITHUB_SHA" not in os.environ:
+        print("Error: GITHUB_SHA environment variable is not set.")
+        sys.exit(1)
+    if "FIRST_SOLC_VERSION" not in os.environ:
+        print("Error: FIRST_SOLC_VERSION environment variable is not set.")
+        sys.exit(1)
+    if "LAST_SOLC_VERSION" not in os.environ:
+        print("Error: LAST_SOLC_VERSION environment variable is not set.")
+        sys.exit(1)
+    if "TAG" not in os.environ:
+        print("Error: TAG environment variable is not set.")
+        sys.exit(1)
+    if "FILEPATH" not in os.environ:
+        print("Error: FILEPATH environment variable is not set.")
+        sys.exit(1)
+
+
+def fetch_data_file():
+    """
+    Fetch the data.json file with artifacts urls and sha256 checksums
+    and parse it into a single dictionary mapping artifact names to their URLs and SHAs.
+    """
+    # read data.json file
+    artifacts_data = {}
+    data_file_path = os.environ["FILEPATH"]
+    if not os.path.exists(data_file_path):
+        print("Error: data.json file not found.")
+        sys.exit(1)
+    with open(data_file_path, 'r') as f:
+        try:
+            artifacts_data = json.load(f)
+        except json.JSONDecodeError:
+            print("Error: data.json file is not a valid JSON.")
+            sys.exit(1)
+
+    result = {}
+
+    for item in artifacts_data:
+        for key, value in item.items():
+            if key.endswith('_url'):
+                base_key = key.rsplit('_url', 1)[0]
+                if base_key not in result:
+                    result[base_key] = {}
+                result[base_key]['url'] = value
+            elif key.endswith('_sha'):
+                base_key = key.rsplit('_sha', 1)[0]
+                if base_key not in result:
+                    result[base_key] = {}
+                result[base_key]['sha'] = value
+
+    return result
+
+
+
+
+
+def extract_build_hash():
+    """Extract the first 8 characters of the commit hash."""
+    sha = os.environ.get("GITHUB_SHA")
+    return f"commit.{sha[:8]}"
+
+def generate_asset_json_nightly(name, url, checksum):
+    """Generate JSON for a specific asset."""
+    # Date in format YYYY-MM-DD
+    date = datetime.now().strftime("%Y.%-m.%-d")
+    last_version = os.environ.get("TAG").replace('v','')
+    version = f"{last_version}-nightly.{date}"
+    SHA = os.environ.get("GITHUB_SHA", "")[:8]
+    build = f"commit.{SHA}"
+    long_version = f"{version}+{build}"
+
+    return {
+        "name": name,
+        "version": version,
+        "build": build,
+        "longVersion": long_version,
+        "url": url,
+        "sha256": checksum,
+        "firstSolcVersion": os.environ.get("FIRST_SOLC_VERSION"),
+        "lastSolcVersion": os.environ.get("LAST_SOLC_VERSION")
+    }
+
+def save_platform_json(platform_folder, asset_json):
+    """Save asset JSON and update list.json for a specific platform."""
+    # Create platform folder if it doesn't exist
+    os.makedirs(platform_folder, exist_ok=True)
+
+    # Update or create list.json
+    list_file_path = os.path.join(platform_folder, "list.json")
+
+    if os.path.exists(list_file_path):
+        with open(list_file_path, 'r') as f:
+            try:
+                list_data = json.load(f)
+            except json.JSONDecodeError:
+                list_data = {"builds": [], "releases": {}, "latestRelease": ""}
+    else:
+        list_data = {"builds": [], "releases": {}, "latestRelease": ""}
+
+    # Remove any existing entry with the same path
+    list_data['builds'] = [
+        build for build in list_data['builds']
+        if build['version'] != asset_json['version']
+    ]
+    # Add the new build
+    list_data['builds'].append(asset_json)
+
+    # Update releases
+    version = asset_json['version']
+    list_data['releases'][version] = f"{asset_json['name']}+{asset_json['longVersion']}"
+
+    # Update latest release
+    list_data['latestRelease'] = version
+
+    with open(list_file_path, 'w') as f:
+        json.dump(list_data, f, indent=4)
+
+def main():
+
+    validate_env_variables()
+    data = fetch_data_file()
+
+    # Mapping of asset names to platform folders
+    platform_mapping = {
+        'resolc-x86_64-unknown-linux-musl': 'linux',
+        'resolc-universal-apple-darwin': 'macos',
+        'resolc-x86_64-pc-windows-msvc': 'windows',
+        'resolc-web.js': 'wasm'
+    }
+
+    # Process each asset
+    for asset in data.keys():
+        platform_name = platform_mapping.get(asset)
+        if platform_name:
+            platform_folder = os.path.join(platform_name)
+            asset_json = generate_asset_json_nightly(asset, data[asset]['url'], data[asset]['sha'])
+            save_platform_json(platform_folder, asset_json)
+            print(f"Processed {asset} for {platform_name}")
+
+if __name__ == "__main__":
+    main()

.github/workflows/release-llvm.yml

@@ -87,9 +87,14 @@ jobs:
         run: |
           brew install ninja
 
+      - name: Install Dependencies
+        if: ${{ matrix.host == 'windows' }}
+        run: |
+          choco install ninja
+
       - name: Install LLVM Builder
         run: |
-          cargo install --path crates/llvm-builder
+          cargo install --locked --force --path crates/llvm-builder
 
       - name: Clone LLVM
         run: |

.github/workflows/release-nightly.yml

@@ -0,0 +1,161 @@
+name: Nightly Release
+
+on:
+  schedule:
+    # Run every day at 01:00 UTC
+    - cron: "0 1 * * *"
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+env:
+  CARGO_TERM_COLOR: always
+
+jobs:
+  check_commits:
+    runs-on: ubuntu-latest
+    outputs:
+      has_commits: ${{ steps.check_commits.outputs.has_commits }}
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+          ref: "main"
+
+      - name: Check for commits from yesterday
+        id: check_commits
+        run: |
+          # Get yesterday's date in YYYY-MM-DD format
+          YESTERDAY=$(date -d "yesterday" +%Y-%m-%d)
+          echo "Checking for commits from: $YESTERDAY"
+
+          # Check if there were any commits yesterday
+          COMMIT_COUNT=$(git log --oneline --since="$YESTERDAY 00:00:00" --until="$YESTERDAY 23:59:59" | wc -l)
+
+          echo "Found $COMMIT_COUNT commits from yesterday"
+
+          if [ $COMMIT_COUNT -gt 0 ]; then
+            echo "has_commits=true" >> $GITHUB_OUTPUT
+            echo "✅ Found $COMMIT_COUNT commits from yesterday - continuing workflow"
+          else
+            echo "has_commits=false" >> $GITHUB_OUTPUT
+            echo "❌ No commits found from yesterday - skipping remaining steps"
+            echo "::notice::❌ No commits found from yesterday - skipping remaining steps"
+          fi
+
+  build:
+    if: ${{ needs.check_commits.outputs.has_commits == 'true' }}
+    needs: [check_commits]
+    uses: ./.github/workflows/reusable-build.yml
+    with:
+      is_release: false
+      retention_days: 40
+
+  create-macos-fat-binary:
+    if: ${{ needs.check_commits.outputs.has_commits == 'true' }}
+    needs: [build]
+    outputs:
+      resolc-universal-apple-darwin_url: ${{ steps.set-output.outputs.resolc-universal-apple-darwin_url }}
+      resolc-universal-apple-darwin_sha: ${{ steps.set-output.outputs.resolc-universal-apple-darwin_sha }}
+    runs-on: macos-14
+    steps:
+      - uses: actions/download-artifact@v4
+        with:
+          merge-multiple: true
+
+      - name: Create macOS Fat Binary
+        run: |
+          lipo resolc-aarch64-apple-darwin resolc-x86_64-apple-darwin -create -output resolc-universal-apple-darwin
+
+      - name: Make Executable
+        run: |
+          chmod +x resolc-universal-apple-darwin
+
+      - uses: actions/upload-artifact@v4
+        id: artifact-upload-step
+        with:
+          name: resolc-universal-apple-darwin
+          path: resolc-universal-apple-darwin
+          retention-days: 40
+
+      - name: Set output variables
+        id: set-output
+        env:
+          TARGET: resolc-universal-apple-darwin
+        run: |
+          echo "${TARGET}_url=${{ steps.artifact-upload-step.outputs.artifact-url }}" >> "$GITHUB_OUTPUT"
+          echo "${TARGET}_sha=${{ steps.artifact-upload-step.outputs.artifact-digest }}" >> "$GITHUB_OUTPUT"
+
+  generate-nightly-json:
+    runs-on: ubuntu-24.04
+    if: ${{ needs.check_commits.outputs.has_commits == 'true' }}
+    environment: tags
+    needs: [build, create-macos-fat-binary, check_commits]
+    steps:
+      - name: Checkout revive
+        uses: actions/checkout@v4
+        with:
+          path: revive
+
+      - name: Checkout resolc-bin
+        uses: actions/checkout@v4
+        with:
+          repository: paritytech/resolc-bin
+          path: resolc-bin
+
+      - name: Download Artifacts
+        uses: actions/download-artifact@v4
+        with:
+          merge-multiple: true
+          path: bins
+
+      - uses: actions/create-github-app-token@v1
+        id: app-token
+        with:
+          app-id: ${{ secrets.REVIVE_JSON_APP_ID }}
+          private-key: ${{ secrets.REVIVE_JSON_APP_KEY }}
+          owner: paritytech
+          repositories: resolc-bin
+
+      - name: Generate JSON
+        env:
+          GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          TOKEN: ${{ steps.app-token.outputs.token }}
+          APP_NAME: "paritytech-revive-json"
+          Green: "\e[32m"
+          NC: "\e[0m"
+        run: |
+          echo '[' > data.json
+          echo '${{ toJSON(needs.build.outputs) }}' >> data.json
+          echo ',' >> data.json
+          echo '${{ toJSON(needs.create-macos-fat-binary.outputs) }}' >> data.json
+          echo ']' >> data.json
+          chmod +x bins/resolc-x86_64-unknown-linux-musl
+          export FIRST_SOLC_VERSION=$(./bins/resolc-x86_64-unknown-linux-musl --supported-solc-versions | cut -f 1 -d "," | tr -d ">=")
+          export LAST_SOLC_VERSION=$(./bins/resolc-x86_64-unknown-linux-musl --supported-solc-versions | cut -f 2 -d "," | tr -d "<=")
+          export FILEPATH=$(readlink -f data.json)
+          export TAG=$(cd revive;gh release list --json name,isLatest --jq '.[] | select(.isLatest)|.name')
+          cd resolc-bin
+          mkdir -p nightly
+          cd nightly
+          python3 ../../revive/.github/scripts/json_generator_nightly.py
+          cd ..
+          git status
+
+          echo "${Green}Add new remote with gh app token${NC}"
+          git remote set-url origin $(git config remote.origin.url | sed "s/github.com/${APP_NAME}:${TOKEN}@github.com/g")
+
+          echo "${Green}Remove http section that causes issues with gh app auth token${NC}"
+          sed -i.bak '/\[http/d' ./.git/config
+          sed -i.bak '/extraheader/d' ./.git/config
+
+          git config user.email "ci@parity.io"
+          git config user.name "${APP_NAME}"
+
+          git add nightly/
+          git commit -m "Update nightly json"
+          git push origin main
+
+          echo "::notice::nightly info.list files were successfully published to https://github.com/paritytech/resolc-bin"

.github/workflows/release.yml

@@ -1,4 +1,5 @@
 name: Build & Release
+
 on:
   push:
     branches: ["main"]
@@ -14,7 +15,6 @@ concurrency:
 
 env:
   CARGO_TERM_COLOR: always
-  RUST_MUSL_CROSS_IMAGE: messense/rust-musl-cross@sha256:68b86bc7cb2867259e6b233415a665ff4469c28b57763e78c3bfea1c68091561
 
 jobs:
   check-version-changed:
@@ -28,33 +28,28 @@ jobs:
     steps:
       - uses: actions/checkout@v4
 
-      # Check that tag and version in Cargo.toml match
       - name: Check versions
         id: versions
         run: |
-          if [[ $CURRENT_TAG == 'main' ]];
+          if [[ $CURRENT_TAG == 'main' ]]; then
-            then
+            echo "::notice::Tag $CURRENT_TAG is not a release tag, skipping the check in the main branch"
-              echo "::notice::Tag $CURRENT_TAG is not a release tag, skipping the check in the main branch";
             exit 0
           fi
 
-          if [[ $CURRENT_TAG != "v"* ]];
+          if [[ $CURRENT_TAG != "v"* ]]; then
-            then
+            echo "::notice::Tag $CURRENT_TAG is not a release tag, skipping the check in a PR"
-              echo "::notice::Tag $CURRENT_TAG is not a release tag, skipping the check in a PR";
             exit 0
           fi
 
-          export PKG_VER=v$(cat Cargo.toml | grep -A 5 package] | grep version | cut -d '=' -f 2 | tr -d '"' | tr -d " ")
+          export PKG_VER=v$(cat crates/resolc/Cargo.toml | grep -A 5 package] | grep version | cut -d '=' -f 2 | tr -d '"' | tr -d " ")
           echo "Current tag $CURRENT_TAG"
           echo "Package version $PKG_VER"
-          #
+
-          if [[ $CURRENT_TAG != $PKG_VER  ]];
+          if [[ $CURRENT_TAG != $PKG_VER ]]; then
-            then
+            echo "::error::Tag $CURRENT_TAG doesn't match package version $PKG_VER in Cargo.toml, please fix"
-              echo "::error::Tag $CURRENT_TAG doesn't match package version $PKG_VER in Cargo.toml, please fix";
             exit 1
           fi
 
-          # Generating release notes early, in order to avoid checkout at the last step
           export RELEASE_NOTES="$(sed '/^## '${CURRENT_TAG}'/,/^## v/!d' CHANGELOG.md | sed -e '1d' -e '$d')"
 
           echo "Release notes:"
@@ -65,173 +60,15 @@ jobs:
           echo 'EOF' >> $GITHUB_OUTPUT
 
   build:
-    strategy:
-      matrix:
-        target:
-          [
-            x86_64-unknown-linux-musl,
-            aarch64-apple-darwin,
-            x86_64-apple-darwin,
-            x86_64-pc-windows-msvc,
-          ]
-        include:
-          - target: x86_64-unknown-linux-musl
-            type: musl
-            runner: ubuntu-24.04
-          - target: aarch64-apple-darwin
-            type: native
-            runner: macos-14
-          - target: x86_64-apple-darwin
-            type: native
-            runner: macos-13
-          - target: x86_64-pc-windows-msvc
-            type: native
-            runner: windows-2022
-    runs-on: ${{ matrix.runner }}
     needs: [check-version-changed]
-    steps:
+    uses: ./.github/workflows/reusable-build.yml
-      - uses: actions/checkout@v4
-      - uses: actions-rust-lang/setup-rust-toolchain@v1
     with:
-          # without this it will override our rust flags
+      is_release: true
-          rustflags: ""
+      retention_days: 1
-          cache-key: ${{ matrix.target }}
-
-      - name: Download LLVM
-        uses: ./.github/actions/get-llvm
-        with:
-          target: ${{ matrix.target }}
-
-      - name: Build
-        if: ${{ matrix.type == 'native' }}
-        shell: bash
-        run: |
-          export LLVM_SYS_181_PREFIX=$PWD/llvm-${{ matrix.target }}
-          make install-bin
-          mv target/release/resolc resolc-${{ matrix.target }} || mv target/release/resolc.exe resolc-${{ matrix.target }}.exe
-
-      - name: Build
-        if: ${{ matrix.type == 'musl' }}
-        run: |
-          docker run -v $PWD:/opt/revive $RUST_MUSL_CROSS_IMAGE /bin/bash -c "
-           cd /opt/revive
-           chown -R root:root .
-           apt update && apt upgrade -y && apt install -y pkg-config
-           export LLVM_SYS_181_PREFIX=/opt/revive/llvm-${{ matrix.target }}
-           make install-bin
-           mv target/${{ matrix.target }}/release/resolc resolc-${{ matrix.target }}
-           "
-          sudo chown -R $(id -u):$(id -g) .
-
-      - name: Install Solc
-        uses: ./.github/actions/get-solc
-
-      - name: Basic Sanity Check
-        shell: bash
-        run: |
-          result=$(./resolc-${{ matrix.target }} --bin crates/integration/contracts/flipper.sol)
-          echo $result
-          if [[ $result == *'0x50564d'* ]]; then exit 0; else exit 1; fi
-
-      - uses: actions/upload-artifact@v4
-        with:
-          name: resolc-${{ matrix.target }}
-          path: resolc-${{ matrix.target }}*
-          retention-days: 1
-
-  build-wasm:
-    runs-on: ubuntu-24.04
-    needs: [check-version-changed]
-    env:
-      RELEASE_RESOLC_WASM_URI: https://github.com/paritytech/revive-workflow-test/releases/download/${{ github.ref_name }}/resolc.wasm
-    steps:
-      - uses: actions/checkout@v4
-      - uses: actions-rust-lang/setup-rust-toolchain@v1
-        with:
-          target: wasm32-unknown-emscripten
-          # without this it will override our rust flags
-          rustflags: ""
-
-      - name: Download Host LLVM
-        uses: ./.github/actions/get-llvm
-        with:
-          target: x86_64-unknown-linux-gnu
-
-      - name: Download Wasm LLVM
-        uses: ./.github/actions/get-llvm
-        with:
-          target: wasm32-unknown-emscripten
-
-      - name: Download EMSDK
-        uses: ./.github/actions/get-emsdk
-
-      - name: Build
-        run: |
-          export LLVM_SYS_181_PREFIX=$PWD/llvm-x86_64-unknown-linux-gnu
-          export REVIVE_LLVM_TARGET_PREFIX=$PWD/llvm-wasm32-unknown-emscripten
-          source emsdk/emsdk_env.sh
-          make install-wasm
-          chmod -x ./target/wasm32-unknown-emscripten/release/resolc.wasm
-
-      - name: Set Up Node.js
-        uses: actions/setup-node@v3
-        with:
-          node-version: "20"
-
-      - name: Basic Sanity Check
-        run: |
-          mkdir -p solc
-          curl -sSLo solc/soljson.js https://github.com/ethereum/solidity/releases/download/v0.8.29/soljson.js
-          node -e "
-          const soljson = require('solc/soljson');
-          const createRevive = require('./target/wasm32-unknown-emscripten/release/resolc.js');
-
-          const compiler = createRevive();
-          compiler.soljson = soljson;
-
-          const standardJsonInput =
-          {
-            language: 'Solidity',
-            sources: {
-              'MyContract.sol': {
-                content: 'pragma solidity ^0.8.0; contract MyContract { function greet() public pure returns (string memory) { return \'Hello\'; } }',
-              },
-            },
-            settings: { optimizer: { enabled: false } }
-          };
-
-          compiler.writeToStdin(JSON.stringify(standardJsonInput));
-          compiler.callMain(['--standard-json']);
-
-          // Collect output
-          const stdout = compiler.readFromStdout();
-          const stderr = compiler.readFromStderr();
-
-          if (stderr) { console.error(stderr); process.exit(1); }
-
-          let out = JSON.parse(stdout);
-          let bytecode = out.contracts['MyContract.sol']['MyContract'].evm.bytecode.object
-          console.log(bytecode);
-
-          if(!bytecode.startsWith('50564d')) { process.exit(1); }
-          "
-
-      - name: Compress Artifact
-        run: |
-          mkdir -p resolc-wasm32-unknown-emscripten
-          mv ./target/wasm32-unknown-emscripten/release/resolc.js ./resolc-wasm32-unknown-emscripten/
-          mv ./target/wasm32-unknown-emscripten/release/resolc.wasm ./resolc-wasm32-unknown-emscripten/
-          mv ./target/wasm32-unknown-emscripten/release/resolc_web.js ./resolc-wasm32-unknown-emscripten/
-
-      - uses: actions/upload-artifact@v4
-        with:
-          name: resolc-wasm32-unknown-emscripten
-          path: resolc-wasm32-unknown-emscripten/*
-          retention-days: 1
 
   create-release:
     if: startsWith(github.ref_name, 'v')
-    needs: [check-version-changed, build-wasm]
+    needs: [check-version-changed, build]
     runs-on: macos-14
     environment: tags
     steps:
@@ -286,3 +123,43 @@ jobs:
             resolc.wasm
             resolc_web.js
             checksums.txt
+
+  npm-release:
+    needs: [create-release]
+    runs-on: macos-14
+    environment: tags
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Download Artifacts
+        uses: actions/download-artifact@v4
+        with:
+          merge-multiple: true
+
+      - name: Set Up Node.js
+        uses: actions/setup-node@v3
+        with:
+          node-version: "20"
+
+      - run: npm ci -w js/resolc
+
+      - name: Build
+        run: |
+          cp -f resolc.{wasm,js} js/resolc/src/resolc
+          npm -w js/resolc run build
+
+      - name: npm pack
+        run: npm -w js/resolc pack
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: npm_package
+          path: "parity-resolc-*.tgz"
+
+      - uses: octokit/request-action@bbedc70b1981e610d89f1f8de88311a1fc02fb83
+        with:
+          route: POST /repos/paritytech/npm_publish_automation/actions/workflows/publish.yml/dispatches
+          ref: main
+          inputs: '${{ format(''{{ "artifact_name": "npm_package", "repo": "{0}", "run_id": "{1}" }}'', github.repository, github.run_id) }}'
+        env:
+          GITHUB_TOKEN: ${{ secrets.NPM_PUBLISH_AUTOMATION_TOKEN }}

.github/workflows/reusable-build.yml

@@ -0,0 +1,262 @@
+name: Reusable Build
+
+on:
+  workflow_call:
+    inputs:
+      is_release:
+        description: "Whether this is a release build"
+        required: true
+        type: boolean
+      retention_days:
+        description: "Artifact retention days"
+        required: false
+        type: number
+        default: 1
+    outputs:
+      resolc-x86_64-unknown-linux-musl_url:
+        value: ${{ jobs.build.outputs.resolc-x86_64-unknown-linux-musl_url }}
+      resolc-x86_64-unknown-linux-musl_sha:
+        value: ${{ jobs.build.outputs.resolc-x86_64-unknown-linux-musl_sha }}
+      resolc-aarch64-apple-darwin_url:
+        value: ${{ jobs.build.outputs.resolc-aarch64-apple-darwin_url }}
+      resolc-aarch64-apple-darwin_sha:
+        value: ${{ jobs.build.outputs.resolc-aarch64-apple-darwin_sha }}
+      resolc-x86_64-apple-darwin_url:
+        value: ${{ jobs.build.outputs.resolc-x86_64-apple-darwin_url }}
+      resolc-x86_64-apple-darwin_sha:
+        value: ${{ jobs.build.outputs.resolc-x86_64-apple-darwin_sha }}
+      resolc-x86_64-pc-windows-msvc_url:
+        value: ${{ jobs.build.outputs.resolc-x86_64-pc-windows-msvc_url }}
+      resolc-x86_64-pc-windows-msvc_sha:
+        value: ${{ jobs.build.outputs.resolc-x86_64-pc-windows-msvc_sha }}
+      resolc-web_js_url:
+        value: ${{ jobs.build-wasm.outputs.resolc_web_js_url }}
+      resolc-web_js_sha:
+        value: ${{ jobs.build-wasm.outputs.resolc_web_js_sha }}
+
+env:
+  CARGO_TERM_COLOR: always
+  RUST_MUSL_CROSS_IMAGE: messense/rust-musl-cross@sha256:c0154e992adb791c3b848dd008939d19862549204f8cb26f5ca7a00f629e6067
+
+jobs:
+  build:
+    # github actions matrix jobs don't support multiple outputs
+    # ugly workaround from https://github.com/orgs/community/discussions/17245#discussioncomment-11222880
+    outputs:
+      resolc-x86_64-unknown-linux-musl_url: ${{ steps.set-output.outputs.resolc-x86_64-unknown-linux-musl_url }}
+      resolc-x86_64-unknown-linux-musl_sha: ${{ steps.set-output.outputs.resolc-x86_64-unknown-linux-musl_sha }}
+      resolc-aarch64-apple-darwin_url: ${{ steps.set-output.outputs.resolc-aarch64-apple-darwin_url }}
+      resolc-aarch64-apple-darwin_sha: ${{ steps.set-output.outputs.resolc-aarch64-apple-darwin_sha }}
+      resolc-x86_64-apple-darwin_url: ${{ steps.set-output.outputs.resolc-x86_64-apple-darwin_url }}
+      resolc-x86_64-apple-darwin_sha: ${{ steps.set-output.outputs.resolc-x86_64-apple-darwin_sha }}
+      resolc-x86_64-pc-windows-msvc_url: ${{ steps.set-output.outputs.resolc-x86_64-pc-windows-msvc_url }}
+      resolc-x86_64-pc-windows-msvc_sha: ${{ steps.set-output.outputs.resolc-x86_64-pc-windows-msvc_sha }}
+    strategy:
+      matrix:
+        target:
+          [
+            x86_64-unknown-linux-musl,
+            aarch64-apple-darwin,
+            x86_64-apple-darwin,
+            x86_64-pc-windows-msvc,
+          ]
+        include:
+          - target: x86_64-unknown-linux-musl
+            type: musl
+            runner: ubuntu-24.04
+          - target: aarch64-apple-darwin
+            type: native
+            runner: macos-14
+          - target: x86_64-apple-darwin
+            type: native
+            runner: macos-13
+          - target: x86_64-pc-windows-msvc
+            type: native
+            runner: windows-2022
+    runs-on: ${{ matrix.runner }}
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          rustflags: ""
+          cache-key: ${{ matrix.target }}
+
+      - name: Download LLVM
+        uses: ./.github/actions/get-llvm
+        with:
+          target: ${{ matrix.target }}
+
+      - name: Build (Native)
+        if: ${{ matrix.type == 'native' }}
+        shell: bash
+        run: |
+          export LLVM_SYS_181_PREFIX=$PWD/llvm-${{ matrix.target }}
+          make install-bin
+          mv target/release/resolc resolc-${{ matrix.target }} || mv target/release/resolc.exe resolc-${{ matrix.target }}.exe
+
+      - name: Build (MUSL)
+        if: ${{ matrix.type == 'musl' }}
+        run: |
+          docker run -v $PWD:/opt/revive $RUST_MUSL_CROSS_IMAGE /bin/bash -c "
+           cd /opt/revive
+           chown -R root:root .
+           apt update && apt upgrade -y && apt install -y pkg-config
+           export LLVM_SYS_181_PREFIX=/opt/revive/llvm-${{ matrix.target }}
+           make install-bin
+           mv target/${{ matrix.target }}/release/resolc resolc-${{ matrix.target }}
+           "
+          sudo chown -R $(id -u):$(id -g) .
+
+      - name: Install Solc
+        uses: ./.github/actions/get-solc
+
+      - name: Basic Sanity Check
+        shell: bash
+        run: |
+          result=$(./resolc-${{ matrix.target }} --bin crates/integration/contracts/flipper.sol)
+          echo $result
+          if [[ $result == *'50564d'* ]]; then exit 0; else exit 1; fi
+
+      - uses: actions/upload-artifact@v4
+        id: artifact-upload-step
+        with:
+          name: resolc-${{ matrix.target }}
+          path: resolc-${{ matrix.target }}*
+          retention-days: ${{ inputs.retention_days }}
+
+      - name: Set output variables
+        if: ${{ !inputs.is_release }}
+        id: set-output
+        shell: bash
+        run: |
+          echo "resolc-${{ matrix.target }}_url=${{ steps.artifact-upload-step.outputs.artifact-url }}" >> "$GITHUB_OUTPUT"
+          echo "resolc-${{ matrix.target }}_sha=${{ steps.artifact-upload-step.outputs.artifact-digest }}" >> "$GITHUB_OUTPUT"
+
+  build-wasm:
+    runs-on: ubuntu-24.04
+    outputs:
+      resolc_web_js_url: ${{ steps.set-output.outputs.resolc_web_js_url }}
+      resolc_web_js_sha: ${{ steps.set-output.outputs.resolc_web_js_sha }}
+    env:
+      RELEASE_RESOLC_WASM_URI: https://github.com/paritytech/revive/releases/download/${{ github.ref_name }}/resolc.wasm
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          target: wasm32-unknown-emscripten
+          rustflags: ""
+
+      - name: Download Host LLVM
+        uses: ./.github/actions/get-llvm
+        with:
+          target: x86_64-unknown-linux-gnu
+
+      - name: Download Wasm LLVM
+        uses: ./.github/actions/get-llvm
+        with:
+          target: wasm32-unknown-emscripten
+
+      - name: Download EMSDK
+        uses: ./.github/actions/get-emsdk
+
+      - name: Build
+        run: |
+          export LLVM_SYS_181_PREFIX=$PWD/llvm-x86_64-unknown-linux-gnu
+          export REVIVE_LLVM_TARGET_PREFIX=$PWD/llvm-wasm32-unknown-emscripten
+          source emsdk/emsdk_env.sh
+          make install-wasm
+          chmod -x ./target/wasm32-unknown-emscripten/release/resolc.wasm
+
+      - name: Set Up Node.js
+        uses: actions/setup-node@v3
+        with:
+          node-version: "20"
+
+      - name: Basic Sanity Check
+        run: |
+          mkdir -p solc
+          curl -sSLo solc/soljson.js https://github.com/ethereum/solidity/releases/download/v0.8.30/soljson.js
+          node -e "
+          const soljson = require('solc/soljson');
+          const createRevive = require('./target/wasm32-unknown-emscripten/release/resolc.js');
+
+          const compiler = createRevive();
+          compiler.soljson = soljson;
+
+          const standardJsonInput =
+          {
+            language: 'Solidity',
+            sources: {
+              'MyContract.sol': {
+                content: 'pragma solidity ^0.8.0; contract MyContract { function greet() public pure returns (string memory) { return \'Hello\'; } }',
+              },
+            },
+            settings: { optimizer: { enabled: false } }
+          };
+
+          compiler.writeToStdin(JSON.stringify(standardJsonInput));
+          compiler.callMain(['--standard-json']);
+
+          const stdout = compiler.readFromStdout();
+          const stderr = compiler.readFromStderr();
+
+          if (stderr) { console.error(stderr); process.exit(1); }
+
+          let out = JSON.parse(stdout);
+          let bytecode = out.contracts['MyContract.sol']['MyContract'].evm.bytecode.object
+          console.log(bytecode);
+
+          if(!bytecode.startsWith('50564d')) { process.exit(1); }
+          "
+
+      - name: Compress Artifact
+        run: |
+          mkdir -p resolc-wasm32-unknown-emscripten
+          mv ./target/wasm32-unknown-emscripten/release/resolc.js ./resolc-wasm32-unknown-emscripten/
+          mv ./target/wasm32-unknown-emscripten/release/resolc.wasm ./resolc-wasm32-unknown-emscripten/
+          mv ./target/wasm32-unknown-emscripten/release/resolc_web.js ./resolc-wasm32-unknown-emscripten/
+
+      - name: Upload artifacts (Release)
+        if: ${{ inputs.is_release }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: resolc-wasm32-unknown-emscripten
+          path: resolc-wasm32-unknown-emscripten/*
+          retention-days: ${{ inputs.retention_days }}
+
+      # There is no way to upload several files as several artifacts with a single upload-artifact step
+      # It's needed to have resolc_web.js separately for night builds for resolc-bin repo
+      # https://github.com/actions/upload-artifact/issues/331
+      - name: Upload artifact resolc.js (Nightly)
+        if: ${{ !inputs.is_release }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: resolc.js
+          path: resolc-wasm32-unknown-emscripten/resolc.js
+          retention-days: ${{ inputs.retention_days }}
+
+      - name: Upload artifacts resolc.wasm (Nightly)
+        if: ${{ !inputs.is_release }}
+        uses: actions/upload-artifact@v4
+        with:
+          name: resolc.wasm
+          path: resolc-wasm32-unknown-emscripten/resolc.wasm
+          retention-days: ${{ inputs.retention_days }}
+
+      - name: Upload artifacts resolc_web.js (Nightly)
+        if: ${{ !inputs.is_release }}
+        uses: actions/upload-artifact@v4
+        id: artifact-upload-step
+        with:
+          name: resolc_web.js
+          path: resolc-wasm32-unknown-emscripten/resolc_web.js
+          retention-days: ${{ inputs.retention_days }}
+
+      - name: Set output variables
+        if: ${{ !inputs.is_release }}
+        id: set-output
+        env:
+          TARGET: resolc_web_js
+        run: |
+          echo "${TARGET}_url=${{ steps.artifact-upload-step.outputs.artifact-url }}" >> "$GITHUB_OUTPUT"
+          echo "${TARGET}_sha=${{ steps.artifact-upload-step.outputs.artifact-digest }}" >> "$GITHUB_OUTPUT"

.github/workflows/test-llvm-builder.yml

@@ -7,6 +7,8 @@ on:
       - 'LLVM.lock'
       - 'crates/llvm-builder/**'
       - '.github/workflows/test-llvm-builder.yml'
+    paths-ignore:
+      - "**.md"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}

.github/workflows/test-wasm.yml

@@ -2,9 +2,13 @@ name: Test Wasm Version
 on:
   push:
     branches: ["main"]
+    paths-ignore:
+      - "**.md"
   pull_request:
     branches: ["main"]
     types: [opened, synchronize]
+    paths-ignore:
+      - "**.md"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
@@ -86,13 +90,18 @@ jobs:
       - name: Install Node Packages
         run: npm install
 
-      - name: Run Playwright tests
+      - name: Test emscripten
-        run: |
-          cd js
-          npx playwright install --with-deps
-          npx playwright test
-
-      - name: Test revive
         run: |
           echo "Running tests for ${{ matrix.os }}"
           npm run test:wasm
+
+      - name: Test @parity/resolc
+        run: |
+          echo "Running tests for ${{ matrix.os }}"
+          npm run -w js/resolc test
+
+      - name: Run Playwright tests
+        run: |
+          cd js/emscripten
+          npx playwright install --with-deps
+          npx playwright test

.github/workflows/test.yml

@@ -2,9 +2,13 @@ name: Test
 on:
   push:
     branches: ["main"]
+    paths-ignore:
+      - "**.md"
   pull_request:
     branches: ["main"]
     types: [opened, synchronize]
+    paths-ignore:
+      - "**.md"
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
@@ -22,6 +26,7 @@ jobs:
         with:
           # without this it will override our rust flags
           rustflags: ""
+          components: rustfmt, clippy
 
       - name: Install Solc
         uses: ./.github/actions/get-solc
@@ -53,5 +58,5 @@ jobs:
       - name: Test cargo workspace
         run: make test-workspace
 
-      - name: Test CLI
+      - name: Test docs
-        run: make test-cli
+        run: make doc

.gitignore

@@ -1,4 +1,5 @@
 /target
+/js/resolc/dist
 target-llvm
 *.dot
 .vscode/
@@ -11,7 +12,6 @@ target-llvm
 node_modules
 artifacts
 tmp
-package-lock.json
 /*.html
 /build
 soljson.js

.prettierrc.json

@@ -0,0 +1,6 @@
+{
+  "trailingComma": "es5",
+  "tabWidth": 2,
+  "semi": false,
+  "singleQuote": true
+}

CHANGELOG.md

@@ -4,14 +4,126 @@
 
 This is a development pre-release.
 
+Supported `polkadot-sdk` rev: `2509.0.0`
+
+## v0.5.0
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2509.0.0`
+
+### Added
+- Support for `SELFDESTRUCT`.
+
+### Changed
+- Emulated EVM heap memory accesses of zero length are never out of bounds.
+- Switched to newer and cheaper storage syscalls (omits reads and writes of `0` values).
+
+### Fixed
+- Introduced a workaround avoiding compiler crashes caused by a bug in LLVM affecting `SDIV`.
+- An off-by-one bug affecting `SDIV` overflow semantics.
+
+## v0.4.1
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2503.0.1`
+
+### Changed
+- The `ast` output is no longer pruned in standard JSON mode (required for foundry).
+- Support `standard_json.output_selection` to also look at per file settings.
+
+## v0.4.0
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2503.0.1`
+
+### Changed
+- Remove the broken `--llvm-ir` mode.
+- Remove the unused fallback for size optimization setting.
+- Unlinked contract binaries are emitted as raw ELF objects.
+
+### Added
+- Line debug information per YUL builtin and for `if` statements.
+- Column numbers in debug information.
+- Support for the YUL optimizer details in the standard json input definition.
+- The `revive-explorer` compiler utility.
+- `revive-yul`: The AST visitor interface.
+- The `--link` deploy time linking mode.
+
+### Fixed
+- The debug info source file matches the YUL path in `--debug-output-dir`, allowing tools to display the source line. 
+- Incosistent type forwarding in JSON output (empty string vs. null object).
+- The solc automatic import resolution.
+- Compiler panic on missing libraries definition.
+
+## v0.3.0
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2503.0.1`
+
+### Fixed
+
+- llvm-context: Bugfix the SAR YUL builtin translation.
+- runtime-api: Add the missing `memset` builtin.
+- npm package: Bugfix the exports field defined in the `package.json`.
+
+
+## v0.2.0
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2503.0.1`
+
+### Changed
+
+- Removed the license printer from the `resolc` binary.
+- EVM bytecode is no longer requested from solc (except in test utils) leading to less compilation work in the pipeline.
+
+### Fixed
+
+- solc-json-interface: Serializing of any custom key in the JSON input is only skipped if not provided.
+- npm package resolution no longer fails with an 'ERR_PACKAGE_PATH_NOT_EXPORTED' error for packages defining exports fields in the `package.json`.
+
+## v0.1.0
+
+This is a development pre-release.
+
+Supported `polkadot-sdk` rev: `2503.0.1`
+
+### Added
+
+- Add the PolkaVM heap size, stack size and debug info CLI compiler options to the standard JSON settings. This makes the standard JSON input succint for reproducible builds.
+
+### Changed
+
+- Supported `polkadot-sdk` version is now `2503.0.1`
+- The `emsdk` version is now `4.0.9`
+
+### Fixed
+
+## v0.1.0-dev.16
+
+This is a development pre-release.
+
 Supported `polkadot-sdk` rev:`c29e72a8628835e34deb6aa7db9a78a2e4eabcee`
 
 ### Added
 
+- Move the npm package from paritytech/js-revive, into this repo. The package `@parity/resolc` will be deployed to npm for each release.
+- Support for solc v0.8.30
+
 ### Changed
 
+- By default, heavy size optimizations are applied.
+
 ### Fixed
 
+- @parity/resolc: The solc dependency package is constrained to the latest supported version, preventing breaking the package ever time a new solc package was released. 
+- The resolc npm package no longer ignores the optimizer settings
+
 ## v0.1.0-dev.14
 
 This is a development pre-release.
@@ -19,13 +131,16 @@ This is a development pre-release.
 Supported `polkadot-sdk` rev:`c29e72a8628835e34deb6aa7db9a78a2e4eabcee`
 
 ### Added
+
 - The `revive-runner` helper utility binary which helps to run contracts locally without a blockchain node.
 - Allow configuration of the EVM heap memory size and stack size via CLI flags and JSON input settings.
 
 ### Changed
+
 - The default PVM stack memory size was increased from 16kb to 32kb.
 
 ### Fixed
+
 - Constructors avoid storing zero sized immutable data on exit.
 
 ## v0.1.0-dev.13
@@ -35,16 +150,19 @@ This is a development pre-release.
 Supported `polkadot-sdk` rev:`c29e72a8628835e34deb6aa7db9a78a2e4eabcee`
 
 ### Added
+
 - Support for solc v0.8.29
 - Decouples the solc JSON-input-output type definitions from the Solidity fronted and expose them via a dedicated crate.
 - `--supported-solc-versions` for `resolc` binary to return a `semver` range of supported `solc` versions.
 - Support for passing LLVM command line options via the prcoess input or providing one or more `--llvm-arg='..'` resolc CLI flag. This allows more fine-grained control over the LLVM backend configuration.
 
 ### Changed
+
 - Storage keys and values are big endian. This was a pre-mature optimization because for the contract itself it this is a no-op and thus not observable. However we should consider the storage layout as part of the contract ABI. The endianness of transient storage values are still kept as-is.
 - Running `resolc` using webkit is no longer supported.
 
 ### Fixed
+
 - A missing byte swap for the create2 salt value.
 
 ## v0.1.0-dev.12
@@ -54,10 +172,12 @@ This is a development pre-release.
 Supported `polkadot-sdk` rev: `21f6f0705e53c15aa2b8a5706b208200447774a9`
 
 ### Added
+
 - Per file output selection for `--standard-json` mode.
 - The `ir` output selection option for `--standard-json` mode.
 
 ### Changed
+
 - Improved code size: Large contracts compile to smaller code blobs when enabling aggressive size optimizations (`-Oz`).
 
 ### Fixed
@@ -73,6 +193,7 @@ Supported `polkadot-sdk` rev: `274a781e8ca1a9432c7ec87593bd93214abbff50`
 ### Changed
 
 ### Fixed
+
 - A bug causing incorrect loads from the emulated EVM linear memory.
 - A missing integer truncate after switching to 64bit.
 
@@ -83,10 +204,12 @@ This is a development pre-release.
 Supported `polkadot-sdk` rev: `274a781e8ca1a9432c7ec87593bd93214abbff50`
 
 ### Added
+
 - Support for the `coinbase` opcode.
 - The resolc web JS version.
 
 ### Changed
+
 - Missing the `--overwrite` flag emits an error instead of a warning.
 - The `resolc` executable prints the help by default.
 - Removed support for legacy EVM assembly (EVMLA) translation.
@@ -96,6 +219,7 @@ Supported `polkadot-sdk` rev: `274a781e8ca1a9432c7ec87593bd93214abbff50`
   If detected, the re-entrant call flag is not set and 0 deposit limit is endowed.
 
 ### Fixed
+
 - Solidity: Add the solc `--libraries` files to sources.
 - A data race in tests.
 - Fix `broken pipe` errors.
@@ -109,9 +233,11 @@ This is a development pre-release.
 ### Added
 
 ### Changed
+
 - Syscalls with more than 6 arguments now pack them into registers.
 
 ### Fixed
+
 - Remove reloading of the resolc.js file (fix issue with relative path in web worker)
 
 ## v0.1.0-dev.8
@@ -119,15 +245,18 @@ This is a development pre-release.
 This is a development pre-release.
 
 ### Added
+
 - The `revive-llvm-builder` crate with the `revive-llvm` helper utility for streamlined management of the LLVM framework dependency.
 - Initial support for running `resolc` in the browser.
 
 ### Changed
+
 - Suported contracts runtime is polkadot-sdk git version `d62a90c8c729acd98c7e9a5cab9803b8b211ffc5`.
 - The minimum supported Rust version is `1.81.0`.
 - Error out early instead of invoking `solc` with invalid base or include path flags.
 
 ### Fixed
+
 - Decouple the LLVM target dependency from the LLVM host dependency.
 - Do not error out if no files and no errors were produced. This aligns resolc closer to solc.
 - Fixes input normalization in the Wasm version.
@@ -137,17 +266,20 @@ This is a development pre-release.
 This is a development pre-release.
 
 ### Added
+
 - Implement the `GASPRICE` opcode.
 - Implement the `BASEFEE` opcode.
 - Implement the `GASLIMIT` opcode.
 
 ### Changed
+
 - The `GAS` opcode now returns the remaining `ref_time`.
 - Contracts can now be supplied call data input of arbitrary size.
 - Some syscalls now return the value in a register, slightly improving emitted contract code.
 - Calls forward maximum weight limits instead of 0, anticipating a change in polkadot-sdk where weight limits of 0 no longer interprets as uncapped limit.
 
 ### Fixed
+
 - A linker bug which was preventing certain contracts from linking with the PVM linker.
 - JS: Fix encoding conversion from JS string (UTF-16) to UTF-8.
 - The git commit hash slug is always displayed in the version string.
@@ -157,6 +289,7 @@ This is a development pre-release.
 This is a development pre-release.
 
 # Added
+
 - Implement the `BLOCKHASH` opcode.
 - Implement delegate calls.
 - Implement the `GASPRICE` opcode. Currently hard-coded to return `1`.
@@ -164,21 +297,24 @@ This is a development pre-release.
 - Initial support for emitting debug info (opt in via the `-g` flag)
 
 # Changed
+
 - resolc now emits 64bit PolkaVM blobs, reducing contract code size and execution time.
 - The RISC-V bit-manipulation target feature (`zbb`) is enabled.
 
 # Fixed
-- Compilation to Wasm (for usage in node and web browsers)
 
+- Compilation to Wasm (for usage in node and web browsers)
 
 ## v0.1.0-dev.5
 
 This is development pre-release.
 
 # Added
+
 - Implement the `CODESIZE` and `EXTCODESIZE` opcodes.
 
 # Changed
+
 - Include the full revive version in the contract metadata.
 
 # Fixed
@@ -188,9 +324,11 @@ This is development pre-release.
 This is development pre-release.
 
 # Added
+
 - Support the `ORIGIN` opcode.
 
 # Changed
+
 - Update polkavm to `v0.14.0`.
 - Enable the `a`, `fast-unaligned-access` and `xtheadcondmov` LLVM target features, decreasing the code size for some contracts.
 

Cargo.toml

@@ -3,7 +3,7 @@ resolver = "2"
 members = ["crates/*"]
 
 [workspace.package]
-version = "0.1.0-dev.14"
+version = "0.1.0"
 authors = [
     "Cyrill Leutwiler <cyrill@parity.io>",
     "Parity Technologies <admin@parity.io>",
@@ -11,76 +11,78 @@ authors = [
 license = "MIT/Apache-2.0"
 edition = "2021"
 repository = "https://github.com/paritytech/revive"
-rust-version = "1.81.0"
+rust-version = "1.85.0"
 
 [workspace.dependencies]
-revive-benchmarks = { version = "0.1.0-dev.14", path = "crates/benchmarks" }
+lld-sys = { version = "0.1.0", path = "crates/lld-sys" }
-revive-builtins = { version = "0.1.0-dev.14", path = "crates/builtins" }
+resolc = { version = "0.5.0", path = "crates/resolc", default-features = false }
-revive-common = { version = "0.1.0-dev.14", path = "crates/common" }
+revive-benchmarks = { version = "0.1.0", path = "crates/benchmarks" }
-revive-differential = { version = "0.1.0-dev.14", path = "crates/differential" }
+revive-build-utils = { version = "0.2.0", path = "crates/build-utils" }
-revive-integration = { version = "0.1.0-dev.14", path = "crates/integration" }
+revive-builtins = { version = "0.1.0", path = "crates/builtins" }
-revive-linker = { version = "0.1.0-dev.14", path = "crates/linker" }
+revive-common = { version = "0.2.1", path = "crates/common" }
-lld-sys = { version = "0.1.0-dev.14", path = "crates/lld-sys" }
+revive-differential = { version = "0.2.0", path = "crates/differential" }
-revive-llvm-context = { version = "0.1.0-dev.14", path = "crates/llvm-context" }
+revive-explorer = { version = "0.1.0", path = "crates/explore" }
-revive-runtime-api = { version = "0.1.0-dev.14", path = "crates/runtime-api" }
+revive-integration = { version = "0.3.0", path = "crates/integration" }
-revive-runner = { version = "0.1.0-dev.14", path = "crates/runner" }
+revive-linker = { version = "0.2.0", path = "crates/linker" }
-revive-solc-json-interface = { version = "0.1.0-dev.14", path = "crates/solc-json-interface" }
+revive-llvm-context = { version = "0.5.0", path = "crates/llvm-context" }
-revive-solidity = { version = "0.1.0-dev.14", path = "crates/solidity" }
+revive-runner = { version = "0.3.0", path = "crates/runner" }
-revive-stdlib = { version = "0.1.0-dev.14", path = "crates/stdlib" }
+revive-runtime-api = { version = "0.4.0", path = "crates/runtime-api" }
-revive-build-utils = { version = "0.1.0-dev.14", path = "crates/build-utils" }
+revive-solc-json-interface = { version = "0.4.0", path = "crates/solc-json-interface", default-features = false }
+revive-stdlib = { version = "0.2.0", path = "crates/stdlib" }
+revive-yul = { version = "0.4.0", path = "crates/yul" }
 
 hex = "0.4.3"
 cc = "1.2"
-libc = "0.2.169"
+libc = "0.2.172"
-tempfile = "3.17"
+tempfile = "3.20"
 anyhow = "1.0"
 semver = { version = "1.0", features = ["serde"] }
 itertools = "0.14"
 serde = { version = "1.0", features = ["derive"] }
 serde_json = { version = "1.0", features = ["arbitrary_precision"] }
-regex = "1.10"
+regex = "1.11"
-once_cell = "1.20"
+once_cell = "1.21"
 num = "0.4.3"
 sha1 = "0.10"
 sha3 = "0.10"
 thiserror = "2.0"
-which = "7.0"
+which = "8.0"
 path-slash = "0.2"
-rayon = "1.8"
+rayon = "1.10"
 clap = { version = "4", default-features = false, features = ["derive"] }
-polkavm-common = "0.21.0"
+polkavm-common = "0.29.0"
-polkavm-linker = "0.21.0"
+polkavm-linker = "0.29.0"
-polkavm-disassembler = "0.21.0"
+polkavm-disassembler = "0.29.0"
-polkavm = "0.21.0"
+polkavm = "0.29.0"
-alloy-primitives = { version = "0.8.21", features = ["serde"] }
+alloy-primitives = { version = "1.1", features = ["serde"] }
-alloy-sol-types = "0.8.21"
+alloy-sol-types = "1.1"
-alloy-genesis = "0.11.1"
+alloy-genesis = "1.0.41"
-alloy-serde = "0.11.1"
+alloy-serde = "1.0"
-env_logger = { version = "0.11.6", default-features = false }
+env_logger = { version = "0.11.8", default-features = false }
-serde_stacker = "0.1.11"
+serde_stacker = "0.1.12"
-criterion = { version = "0.5.1", features = ["html_reports"] }
+criterion = { version = "0.7", features = ["html_reports"] }
-log = { version = "0.4.25" }
+log = { version = "0.4.27" }
-git2 = { version = "0.20.0", default-features = false }
+git2 = { version = "0.20.2", default-features = false }
 downloader = "0.2.8"
-flate2 = "1.0.35"
+flate2 = "1.1"
-fs_extra = "1.3.0"
+fs_extra = "1.3"
 num_cpus = "1"
 tar = "0.4"
-toml = "0.8"
+toml = "0.9"
-assert_cmd = "2.0"
+assert_cmd = "2"
 assert_fs = "1.1"
+normpath = "1.3"
 
 # polkadot-sdk and friends
-codec = { version = "3.6.12", default-features = false, package = "parity-scale-codec" }
+codec = { version = "3.7.5", default-features = false, package = "parity-scale-codec" }
 scale-info = { version = "2.11.6", default-features = false }
-polkadot-sdk = { git = "https://github.com/paritytech/polkadot-sdk", rev = "c29e72a8628835e34deb6aa7db9a78a2e4eabcee" }
+polkadot-sdk = { version = "2509.0.0" }
 
 # llvm
 [workspace.dependencies.inkwell]
-git = "https://github.com/TheDan64/inkwell.git"
+version = "0.6.0"
-rev = "7b410298b6a93450adaa90b1841d5805a3038f12"
 default-features = false
-features = ["serde", "llvm18-0", "no-libffi-linking", "target-riscv"]
+features = ["serde", "llvm18-1", "no-libffi-linking", "target-riscv"]
 
 [profile.bench]
 inherits = "release"

Dockerfile

@@ -1,4 +1,4 @@
-FROM rust:1.84.0 AS llvm-builder
+FROM rust:1.85.0 AS llvm-builder
 WORKDIR /opt/revive
 
 RUN apt update && \ 
@@ -11,7 +11,7 @@ RUN make install-llvm-builder
 RUN revive-llvm --target-env musl clone
 RUN revive-llvm --target-env musl build  --llvm-projects lld --llvm-projects clang
 
-FROM messense/rust-musl-cross@sha256:68b86bc7cb2867259e6b233415a665ff4469c28b57763e78c3bfea1c68091561 AS resolc-builder
+FROM messense/rust-musl-cross@sha256:c0154e992adb791c3b848dd008939d19862549204f8cb26f5ca7a00f629e6067 AS resolc-builder
 WORKDIR /opt/revive
 
 RUN apt update && \ 

Makefile

@@ -6,14 +6,15 @@
 	install-llvm-builder \
 	install-llvm \
 	install-revive-runner \
+	install-revive-explorer \
 	format \
 	clippy \
+	doc \
 	machete \
 	test \
 	test-integration \
-	test-solidity \
+	test-resolc \
 	test-workspace \
-	test-cli \
 	test-wasm \
 	test-llvm-builder
 	bench \
@@ -24,24 +25,27 @@
 install: install-bin install-npm
 
 install-bin:
-	cargo install --locked --path crates/solidity
+	cargo install --force --locked --path crates/resolc
 
 install-npm:
 	npm install && npm fund
 
 install-wasm: install-npm
-	cargo build --target wasm32-unknown-emscripten -p revive-solidity --release --no-default-features
+	cargo build --target wasm32-unknown-emscripten -p resolc --release --no-default-features
 	npm run build:package
 
 install-llvm-builder:
-	cargo install --path crates/llvm-builder
+	cargo install --force --locked --path crates/llvm-builder
 
 install-llvm: install-llvm-builder
 	revive-llvm clone
 	revive-llvm build --llvm-projects lld --llvm-projects clang
 
 install-revive-runner:
-	cargo install --path crates/runner --no-default-features --locked
+	cargo install --locked --force --path crates/runner --no-default-features
+
+install-revive-explorer:
+	cargo install --locked --force --path crates/explorer --no-default-features
 
 format:
 	cargo fmt --all --check
@@ -49,24 +53,24 @@ format:
 clippy:
 	cargo clippy --all-features --workspace --tests --benches -- --deny warnings
 
+doc:
+	cargo doc --all-features --workspace --document-private-items --no-deps
+
 machete:
 	cargo install cargo-machete
 	cargo machete
 
-test: format clippy machete test-cli test-workspace install-revive-runner
+test: format clippy machete test-workspace install-revive-runner install-revive-explorer doc
 
 test-integration: install-bin
 	cargo test --package revive-integration
 
-test-solidity: install
+test-resolc: install
-	cargo test --package revive-solidity
+	cargo test --package resolc
 
 test-workspace: install
 	cargo test --workspace --exclude revive-llvm-builder
 
-test-cli: install
-	npm run test:cli
-
 test-wasm: install-wasm
 	npm run test:wasm
 
@@ -90,9 +94,6 @@ clean:
 	cargo clean ; \
 	revive-llvm clean ; \
 	rm -rf node_modules ; \
-	rm -rf crates/solidity/src/tests/cli-tests/artifacts ; \
+	rm -rf crates/resolc/src/tests/cli/artifacts ; \
-	cargo uninstall revive-solidity ; \
+	cargo uninstall resolc ; \
-	cargo uninstall revive-llvm-builder ; \
+	cargo uninstall revive-llvm-builder ;
-	rm -f package-lock.json ; \
-	rm -rf js/dist ; \
-	rm -f js/src/resolc.{wasm,js}

README.md

@@ -14,7 +14,39 @@ This is experimental software in active development and not ready just yet for p
 Discussion around the development is hosted on the [Polkadot Forum](https://forum.polkadot.network/t/contracts-update-solidity-on-polkavm/6949#a-new-solidity-compiler-1).
 
 ## Installation
-Please consult [the documentation](https://contracts.polkadot.io/revive_compiler/installation) for installation instructions.
+Building Solidity contracts for PolkaVM requires installing the following two compilers:
+- `resolc`: The revive Solidity compiler YUL frontend and PolkaVM code generator (provided by this repository).
+- `solc`: The [Ethereum Solidity reference compiler](https://github.com/ethereum/solidity/) implemenation.`resolc` uses `solc` during the compilation process, please refer to the [Ethereum Solidity documentation](https://docs.soliditylang.org/en/latest/installing-solidity.html) for installation instructions.
+
+### `resolc` binary releases
+`resolc` is distributed as a standalone binary (with `solc` as the only external dependency). Please download one of our [binary releases](https://github.com/paritytech/revive/releases) for your target platform and mind the platform specific instructions below.
+
+<details>
+  <summary>MacOS users</summary>
+
+> **MacOS** users need to clear the `downloaded` attribute from the binary and set the executable flag.
+> ```sh
+> xattr -rc resolc-universal-apple-darwin
+> chmod +x resolc-universal-apple-darwin
+> ```
+
+</details>
+
+<details>
+  <summary>Linux users</summary>
+
+> **Linux** users need to set the executable flag.
+> ```sh
+> chmod +x resolc-x86_64-unknown-linux-musl
+> ```
+
+</details>
+
+
+### `resolc` NPM package
+We distribute the revive compiler as [node.js module](https://www.npmjs.com/package/@parity/resolc) and [hardhat plugin](https://www.npmjs.com/package/@parity/hardhat-polkadot-resolc).
+
+Note: The `solc` dependency is bundled via NPM packaging and defaults to the latest supported version.
 
 ## Building from source
 
@@ -44,9 +76,7 @@ export LLVM_SYS_181_PREFIX=</path/to/the/extracted/archive>/target-llvm/gnu/targ
 <details>
   <summary>Building from source</summary>
 
- Use the provided [revive-llvm](crates/llvm-builder/README.md) utility to compile a compatible LLVM build locally and point `$LLVM_SYS_181_PREFIX` to the installation afterwards.
+The `Makefile` provides a shortcut target to obtain a compatible LLVM build, using the provided [revive-llvm](crates/llvm-builder/README.md) utility. Once installed, point `$LLVM_SYS_181_PREFIX` to the installation afterwards:
-
-The `Makefile` provides a shortcut target to obtain a compatible LLVM build:
 
 ```sh
 make install-llvm
@@ -99,8 +129,6 @@ Ensure that your branch passes `make test` locally when submitting a pull reques
 ### Design overview
 See the [relevant section in our documentation](https://contracts.polkadot.io/revive_compiler/architecture) to learn more about how the compiler works. 
 
-[Frontend](https://github.com/matter-labs/era-compiler-solidity) and [code generator](https://github.com/matter-labs/era-compiler-llvm-context) are based of ZKSync `zksolc` (the project started as a fork of the era compiler).
-
 ### Tests
 
 Before running the tests, ensure that Geth (Go Ethereum) is installed on your system. Follow the installation guide here: [Installing Geth](https://geth.ethereum.org/docs/getting-started/installing-geth).
@@ -109,3 +137,8 @@ Once Geth is installed, you can run the tests using the following command:
 ```sh
 make test
 ```
+# Acknowledgements
+
+The revive compiler project, after some early experiments with EVM bytecode translations, decided to fork the `era-compiler` framework.
+[Frontend](https://github.com/matter-labs/era-compiler-solidity), [code generator](https://github.com/matter-labs/era-compiler-llvm-context) and some supporting libraries are based of ZKSync `zksolc`. I'd like to express my gratitude and thank the original authors for providing a useable code base under a generous license.
+

RELEASE.md

@@ -4,15 +4,23 @@ Prior to the first stable release we neither have formal release processes nor d
 
 To create a new pre-release:
 
-1. Create a release PR which updates the `-dev.X` versions in the workspace `Cargo.toml` and updates the `CHANGELOG.md` accordingly.
+1. Create a release PR which, if necessary:
+  - Updates the versions in the workspace `Cargo.toml`
+  - Updates the version in each crate `Cargo.toml`
+  - Updates the version of the NPM package in `js/resolc/package.json`
+  - Updates the `CHANGELOG.md` to reflect all observable changes
 2. If the CI passes, merge the release PR.
-3. Push a tag that has the same `-dev.X` version as in `Cargo.toml`
+3. Push a `vX.Y.Z` tag that has the same version as in `Cargo.toml`
 4. The release workflow will attempt to build and publish a new pre-release if the latest tag does match the cargo package version.
-5. Wait for the `Release` workflow to finish. It should create the pre-release with the same `-dev.X` name.
+5. Wait for the `Release` workflow to finish. It should create the pre-release with the same name.
 6. Check that pre-release was created on the [Releases page](https://github.com/paritytech/revive/releases) with all artifacts.
 7. After the release is published, another workflow should start automatically and update json files in https://github.com/paritytech/resolc-bin. Check the changes.
 8. Update the [contract-docs](https://github.com/paritytech/contract-docs/) accordingly
 
+# `resolc` NPM package release
+
+Will happen automatically.
+
 # LLVM release
 
 To create a new LLVM release, run "Release LLVM" workflow. Use current LLVM version as parameter, e.g. `18.1.8`.

STATUS.md

@@ -1,30 +0,0 @@
-# Known issues
-
-The following is known and we are either working on it or it is a hard limitation. Please do not open a new issue.
-
-## Release
-
-`0.1.0-dev-2`
-
-## Missing features
-
-- [Libraries with public functions are not supported](https://github.com/paritytech/revive/issues/91)
-- [Automatic import resolution is not supported](https://github.com/paritytech/revive/issues/98)
-- The emulated EVM linear contract memory is limited to 64kb in size. Will be fixed with support for metered dynamic memory.
-- [The contract calldata is currently limited to 1kb in size](https://github.com/paritytech/revive/issues/57)
-- [EIP-4844 opcodes are not supported](https://github.com/paritytech/revive/issues/64)
-- [Delegate calls are not supported](https://github.com/paritytech/revive/issues/67)
-- [The `blockhash` opcode is not supported](https://github.com/paritytech/revive/issues/61)
-- [The `extcodesize` opcode is not supported](https://github.com/paritytech/revive/issues/58)
-- [The `origin` opcode is not supported](https://github.com/paritytech/revive/issues/59)
-- [Gas limits for contract calls are ignored](https://github.com/paritytech/revive/issues/60)
-- [Gas related opcodes are not supported](https://github.com/paritytech/revive/issues/60)
-- IPFS metadata hashes are not supported
-- [Compiled contract artifacts can exceed the pallet static memory limit and fail to deploy](https://github.com/paritytech/revive/issues/96).
-- [Transfers to inexistant accounts will fail if the transferred value lies below the ED.](https://github.com/paritytech/revive/issues/83) Will be fixed in the pallet to make the ED completely transparent for contracts.
-
-## Wontfix
-
-Please consult our documentation to learn more about Solidity and EVM features likely to remain unsupported (and why they will not be supported).
-
-TODO: Insert link to the relevant documentation section.

clippy.toml

@@ -0,0 +1 @@
+large-error-threshold = 192

crates/build-utils/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "revive-build-utils"
-version.workspace = true
+version = "0.2.0"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true

crates/build-utils/src/lib.rs

@@ -6,6 +6,10 @@ pub const REVIVE_LLVM_HOST_PREFIX: &str = "LLVM_SYS_181_PREFIX";
 /// The revive LLVM target dependency directory prefix environment variable.
 pub const REVIVE_LLVM_TARGET_PREFIX: &str = "REVIVE_LLVM_TARGET_PREFIX";
 
+/// The revive LLVM host tool help link.
+pub const REVIVE_LLVM_BUILDER_HELP_LINK: &str =
+    "https://github.com/paritytech/revive?tab=readme-ov-file#building-from-source";
+
 /// Constructs a path to the LLVM tool `name`.
 ///
 /// Respects the [`REVIVE_LLVM_HOST_PREFIX`] environment variable.
@@ -13,9 +17,7 @@ pub fn llvm_host_tool(name: &str) -> std::path::PathBuf {
     std::env::var_os(REVIVE_LLVM_HOST_PREFIX)
         .map(Into::<std::path::PathBuf>::into)
         .unwrap_or_else(|| {
-            panic!(
+            panic!("install LLVM using the revive-llvm builder and export '{REVIVE_LLVM_HOST_PREFIX}'; see also: {REVIVE_LLVM_BUILDER_HELP_LINK}")
-                "install LLVM using the revive-llvm builder and export {REVIVE_LLVM_HOST_PREFIX}",
-            )
         })
         .join("bin")
         .join(name)

crates/common/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "revive-common"
-version.workspace = true
+version = "0.2.1"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true
@@ -15,6 +15,8 @@ doctest = false
 
 [dependencies]
 anyhow = { workspace = true }
+hex = { workspace = true }
+sha3 = { workspace = true }
 serde = { workspace = true, features = ["derive"] }
 serde_json = { workspace = true, features = [ "arbitrary_precision", "unbounded_depth" ] }
 serde_stacker = { workspace = true }

crates/common/src/contract_identifier.rs

@@ -0,0 +1,33 @@
+//! The contract identifier helper library.
+
+use serde::{Deserialize, Serialize};
+
+/// This structure simplifies passing the contract identifiers through the compilation pipeline.
+#[derive(Debug, Clone, Serialize, Deserialize)]
+pub struct ContractIdentifier {
+    /// The absolute file path.
+    pub path: String,
+    /// The contract name.
+    /// Is set for Solidity contracts only. Otherwise it would be equal to the file name.
+    pub name: Option<String>,
+    /// The full contract identifier.
+    /// For Solidity, The format is `<absolute file path>:<contract name>`.
+    /// For other languages, `<absolute file path>`.
+    pub full_path: String,
+}
+
+impl ContractIdentifier {
+    /// A shortcut constructor.
+    pub fn new(path: String, name: Option<String>) -> Self {
+        let full_path = match name {
+            Some(ref name) => format!("{path}:{name}"),
+            None => path.clone(),
+        };
+
+        Self {
+            path,
+            name,
+            full_path,
+        }
+    }
+}

crates/common/src/extension.rs

@@ -37,4 +37,4 @@ pub static EXTENSION_POLKAVM_ASSEMBLY: &str = "pvmasm";
 pub static EXTENSION_POLKAVM_BINARY: &str = "pvm";
 
 /// The ELF shared object file extension.
-pub static EXTENSION_SHARED_OBJECT: &str = "so";
+pub static EXTENSION_OBJECT: &str = "o";

crates/common/src/keccak256.rs

@@ -0,0 +1,68 @@
+//! Keccak-256 hash utilities.
+
+use serde::{Deserialize, Serialize};
+use sha3::digest::FixedOutput;
+use sha3::Digest;
+
+pub const DIGEST_BYTES: usize = 32;
+
+/// Keccak-256 hash utilities.
+#[derive(Debug, Clone, PartialEq, Serialize, Deserialize)]
+pub struct Keccak256 {
+    /// Binary representation.
+    bytes: [u8; DIGEST_BYTES],
+    /// Hexadecimal string representation.
+    string: String,
+}
+
+impl Keccak256 {
+    /// Computes the `keccak256` hash for `preimage`.
+    pub fn from_slice(preimage: &[u8]) -> Self {
+        let bytes = sha3::Keccak256::digest(preimage).into();
+        let string = format!("0x{}", hex::encode(bytes));
+        Self { bytes, string }
+    }
+
+    /// Computes the `keccak256` hash for an array of `preimages`.
+    pub fn from_slices<R: AsRef<[u8]>>(preimages: &[R]) -> Self {
+        let mut hasher = sha3::Keccak256::new();
+        for preimage in preimages.iter() {
+            hasher.update(preimage);
+        }
+        let bytes: [u8; DIGEST_BYTES] = hasher.finalize_fixed().into();
+        let string = format!("0x{}", hex::encode(bytes));
+        Self { bytes, string }
+    }
+
+    /// Returns a reference to the 32-byte SHA-3 hash.
+    pub fn as_bytes(&self) -> &[u8] {
+        self.bytes.as_slice()
+    }
+
+    /// Returns a reference to the hexadecimal string representation.
+    pub fn as_str(&self) -> &str {
+        self.string.as_str()
+    }
+
+    /// Extracts the binary representation.
+    pub fn to_vec(&self) -> Vec<u8> {
+        self.bytes.to_vec()
+    }
+}
+
+impl std::fmt::Display for Keccak256 {
+    fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result {
+        write!(f, "{}", self.as_str())
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    #[test]
+    fn hash_and_stringify_works() {
+        assert_eq!(
+            super::Keccak256::from_slices(&["foo".as_bytes(), "bar".as_bytes(),]).as_str(),
+            "0x38d18acb67d25c8bb9942764b62f18e17054f66a817bd4295423adf9ed98873e"
+        );
+    }
+}

crates/common/src/lib.rs

@@ -3,9 +3,13 @@
 pub(crate) mod base;
 pub(crate) mod bit_length;
 pub(crate) mod byte_length;
+pub(crate) mod contract_identifier;
 pub(crate) mod evm_version;
 pub(crate) mod exit_code;
 pub(crate) mod extension;
+pub(crate) mod keccak256;
+pub(crate) mod metadata;
+pub(crate) mod object;
 pub(crate) mod utils;
 
 pub use self::base::*;
@@ -14,4 +18,8 @@ pub use self::byte_length::*;
 pub use self::evm_version::EVMVersion;
 pub use self::exit_code::*;
 pub use self::extension::*;
+pub use self::keccak256::*;
+pub use self::metadata::*;
+pub use self::object::*;
 pub use self::utils::*;
+pub use contract_identifier::*;

crates/common/src/metadata.rs

@@ -0,0 +1,42 @@
+//! The metadata hash type.
+
+use std::str::FromStr;
+
+use serde::{Deserialize, Serialize};
+
+/// The metadata hash type.
+#[derive(Debug, Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize)]
+pub enum MetadataHash {
+    /// Do not include bytecode hash.
+    #[serde(rename = "none")]
+    None,
+    /// Include the `ipfs` hash.
+    #[serde(rename = "ipfs")]
+    IPFS,
+    /// Include the `keccak256`` hash.
+    #[serde(rename = "keccak256")]
+    Keccak256,
+}
+
+impl FromStr for MetadataHash {
+    type Err = anyhow::Error;
+
+    fn from_str(string: &str) -> Result<Self, Self::Err> {
+        match string {
+            "none" => Ok(Self::None),
+            "ipfs" => Ok(Self::IPFS),
+            "keccak256" => Ok(Self::Keccak256),
+            string => anyhow::bail!("unknown bytecode hash mode: `{string}`"),
+        }
+    }
+}
+
+impl std::fmt::Display for MetadataHash {
+    fn fmt(&self, f: &mut std::fmt::Formatter) -> std::fmt::Result {
+        match self {
+            Self::None => write!(f, "none"),
+            Self::IPFS => write!(f, "ipfs"),
+            Self::Keccak256 => write!(f, "keccak256"),
+        }
+    }
+}

crates/common/src/object.rs

@@ -0,0 +1,61 @@
+//! The revive binary object helper module.
+
+use std::str::FromStr;
+
+use serde::{Deserialize, Serialize};
+
+/// The binary object format.
+///
+/// Unlinked contracts are stored in a different object format
+/// than final (linked) contract blobs.
+#[derive(Debug, Clone, Copy, PartialEq, Eq, Hash, Serialize, Deserialize)]
+pub enum ObjectFormat {
+    /// The unlinked ELF object format.
+    ELF,
+    /// The fully linked PVM format.
+    PVM,
+}
+
+impl ObjectFormat {
+    pub const PVM_MAGIC: [u8; 4] = [b'P', b'V', b'M', b'\0'];
+    pub const ELF_MAGIC: [u8; 4] = [0x7f, b'E', b'L', b'F'];
+}
+
+impl FromStr for ObjectFormat {
+    type Err = anyhow::Error;
+
+    fn from_str(value: &str) -> Result<Self, Self::Err> {
+        match value {
+            "ELF" => Ok(Self::ELF),
+            "PVM" => Ok(Self::PVM),
+            _ => anyhow::bail!(
+                "Unknown object format: {value}. Supported formats: {}, {}",
+                Self::ELF,
+                Self::PVM,
+            ),
+        }
+    }
+}
+
+impl TryFrom<&[u8]> for ObjectFormat {
+    type Error = &'static str;
+
+    fn try_from(value: &[u8]) -> Result<Self, Self::Error> {
+        if value.starts_with(&Self::PVM_MAGIC) {
+            return Ok(Self::PVM);
+        }
+        if value.starts_with(&Self::ELF_MAGIC) {
+            return Ok(Self::ELF);
+        }
+        Err("expected a contract object")
+    }
+}
+
+impl std::fmt::Display for ObjectFormat {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        match self {
+            Self::ELF => write!(f, "ELF"),
+            Self::PVM => write!(f, "PVM"),
+        }
+    }
+}

crates/common/src/utils.rs

@@ -1,25 +1,45 @@
 //! The compiler common utils.
 
 /// Deserializes a `serde_json` object from slice with the recursion limit disabled.
+///
 /// Must be used for all JSON I/O to avoid crashes due to the aforementioned limit.
 pub fn deserialize_from_slice<O>(input: &[u8]) -> anyhow::Result<O>
 where
     O: serde::de::DeserializeOwned,
 {
-    let mut deserializer = serde_json::Deserializer::from_slice(input);
+    let deserializer = serde_json::Deserializer::from_slice(input);
-    deserializer.disable_recursion_limit();
+    deserialize(deserializer)
-    let deserializer = serde_stacker::Deserializer::new(&mut deserializer);
-    let result = O::deserialize(deserializer)?;
-    Ok(result)
 }
 
 /// Deserializes a `serde_json` object from string with the recursion limit disabled.
+///
 /// Must be used for all JSON I/O to avoid crashes due to the aforementioned limit.
 pub fn deserialize_from_str<O>(input: &str) -> anyhow::Result<O>
 where
     O: serde::de::DeserializeOwned,
 {
-    let mut deserializer = serde_json::Deserializer::from_str(input);
+    let deserializer = serde_json::Deserializer::from_str(input);
+    deserialize(deserializer)
+}
+
+/// Deserializes a `serde_json` object from reader with the recursion limit disabled.
+///
+/// Must be used for all JSON I/O to avoid crashes due to the aforementioned limit.
+pub fn deserialize_from_reader<R, O>(reader: R) -> anyhow::Result<O>
+where
+    R: std::io::Read,
+    O: serde::de::DeserializeOwned,
+{
+    let deserializer = serde_json::Deserializer::from_reader(reader);
+    deserialize(deserializer)
+}
+
+/// Runs the generic deserializer.
+pub fn deserialize<'de, R, O>(mut deserializer: serde_json::Deserializer<R>) -> anyhow::Result<O>
+where
+    R: serde_json::de::Read<'de>,
+    O: serde::de::DeserializeOwned,
+{
     deserializer.disable_recursion_limit();
     let deserializer = serde_stacker::Deserializer::new(&mut deserializer);
     let result = O::deserialize(deserializer)?;

crates/differential/Cargo.toml

@@ -1,6 +1,7 @@
 [package]
 name = "revive-differential"
-version.workspace = true
+description = "utilities for differential testing the revive compiler against EVM"
+version = "0.2.0"
 license.workspace = true
 edition.workspace = true
 authors.workspace = true

crates/differential/src/go_duration.rs

@@ -3,8 +3,8 @@ use std::time::Duration;
 /// Parse a go formatted duration.
 ///
 /// Sources:
-/// - https://crates.io/crates/go-parse-duration (fixed an utf8 bug)
+/// - <https://crates.io/crates/go-parse-duration> (fixed an utf8 bug)
-/// - https://github.com/golang/go/blob/master/src/time/format.go
+/// - <https://github.com/golang/go/blob/master/src/time/format.go>
 pub fn parse_go_duration(value: &str) -> Result<Duration, String> {
     parse_duration(value).map(|ns| Duration::from_nanos(ns.unsigned_abs()))
 }

crates/explorer/Cargo.toml

@@ -0,0 +1,20 @@
+[package]
+name = "revive-explorer"
+version = "0.1.0"
+license.workspace = true
+edition.workspace = true
+repository.workspace = true
+authors.workspace = true
+description = "Helper utility to inspect debug builds"
+
+[[bin]]
+name = "revive-explorer"
+path = "src/main.rs"
+
+[dependencies]
+anyhow = { workspace = true }
+clap = { workspace = true, features = ["help", "std", "derive"] }
+num_cpus = { workspace = true }
+
+revive-yul = { workspace = true }
+

crates/explorer/README.md

@@ -0,0 +1,49 @@
+# revive-explorer
+
+The `revive-explorer` is a helper utility for exploring the compilers YUL lowering unit.
+
+It analyzes a given shared objects from the debug dump and outputs:
+- The count of each YUL statement translated.
+- A per YUL statement break-down of bytecode size contributed per.
+- Estimated `yul-phaser` cost parameters.
+
+Example:
+
+```
+statements count:
+	block 532
+	Caller 20
+	Not 73
+	Gas 24
+	Shr 2
+    ...
+	Shl 259
+	SetImmutable 2
+	CodeSize 1
+	CallDataLoad 87
+	Return 56
+bytes per statement:
+	Or 756
+	CodeCopy 158
+	Log3 620
+	Return 1562
+	MStore 36128
+	...
+	ReturnDataCopy 2854
+	DataOffset 28
+	assignment 1194
+	Number 540
+	CallValue 4258
+yul-phaser parameters:
+	--break-cost 1
+	--variable-declaration-cost 3
+	--function-call-cost 8
+	--if-cost 4
+	--expression-statement-cost 6
+	--function-definition-cost 11
+	--switch-cost 3
+	--block-cost 1
+	--leave-cost 1
+	--assignment-cost 1
+```
+

crates/explorer/src/dwarfdump.rs

@@ -0,0 +1,68 @@
+//! The `llvm-dwarfdump` utility helper library.
+
+use std::{
+    path::{Path, PathBuf},
+    process::{Command, Stdio},
+};
+
+pub static EXECUTABLE: &str = "llvm-dwarfdump";
+pub static DEBUG_LINES_ARGUMENTS: [&str; 1] = ["--debug-line"];
+pub static SOURCE_FILE_ARGUMENTS: [&str; 1] = ["--show-sources"];
+
+/// Calls the `llvm-dwarfdump` tool to extract debug line information
+/// from the shared object at `path`. Returns the output.
+///
+/// Provide `Some(dwarfdump_exectuable)` to override the default executable.
+pub fn debug_lines(
+    shared_object: &Path,
+    dwarfdump_executable: &Option<PathBuf>,
+) -> anyhow::Result<String> {
+    dwarfdump(shared_object, dwarfdump_executable, &DEBUG_LINES_ARGUMENTS)
+}
+
+/// Calls the `llvm-dwarfdump` tool to extract the source file name.
+/// Returns the source file path.
+///
+/// Provide `Some(dwarfdump_exectuable)` to override the default executable.
+pub fn source_file(
+    shared_object: &Path,
+    dwarfdump_executable: &Option<PathBuf>,
+) -> anyhow::Result<PathBuf> {
+    let output = dwarfdump(shared_object, dwarfdump_executable, &SOURCE_FILE_ARGUMENTS)?;
+    let output = output.trim();
+
+    if output.is_empty() {
+        anyhow::bail!(
+            "the shared object at path `{}` doesn't contain the source file name. Hint: compile with debug information (-g)?",
+            shared_object.display()
+        );
+    }
+
+    Ok(output.into())
+}
+
+/// The internal `llvm-dwarfdump` helper function.
+fn dwarfdump(
+    shared_object: &Path,
+    dwarfdump_executable: &Option<PathBuf>,
+    arguments: &[&str],
+) -> anyhow::Result<String> {
+    let executable = dwarfdump_executable
+        .to_owned()
+        .unwrap_or_else(|| PathBuf::from(EXECUTABLE));
+
+    let output = Command::new(executable)
+        .args(arguments)
+        .arg(shared_object)
+        .stdin(Stdio::null())
+        .stdout(Stdio::piped())
+        .stderr(Stdio::piped())
+        .spawn()?
+        .wait_with_output()?;
+
+    if !output.status.success() {
+        anyhow::bail!(String::from_utf8_lossy(&output.stderr).to_string());
+    }
+
+    Ok(String::from_utf8_lossy(&output.stdout).to_string())
+}

crates/explorer/src/dwarfdump_analyzer.rs

@@ -0,0 +1,222 @@
+//! The core dwarf dump analyzer library.
+
+use std::{
+    collections::HashMap,
+    path::{Path, PathBuf},
+};
+
+use revive_yul::lexer::token::location::Location;
+
+use crate::location_mapper::{self, LocationMapper};
+
+/// The dwarf dump analyzer.
+///
+/// Loads debug information from `llvm-dwarfdump` and calculates statistics
+/// about the compiled YUL statements:
+/// - Statements count
+/// - Per-statement
+#[derive(Debug, Default)]
+pub struct DwarfdumpAnalyzer {
+    /// The YUL source file path.
+    source: PathBuf,
+
+    /// The YUL location to statements map.
+    location_map: HashMap<Location, String>,
+
+    /// The `llvm-dwarfdump --debug-lines` output.
+    debug_lines: String,
+
+    /// The observed statements.
+    statements_count: HashMap<String, usize>,
+    /// The observed statement to instructions size.
+    statements_size: HashMap<String, u64>,
+}
+
+impl DwarfdumpAnalyzer {
+    /// The debug info analyzer constructor.
+    ///
+    /// `source` is the path to the YUL source file.
+    /// `debug_lines` is the `llvm-dwarfdump --debug-lines` output.
+    pub fn new(source: &Path, debug_lines: String) -> Self {
+        Self {
+            source: source.to_path_buf(),
+            debug_lines,
+            ..Default::default()
+        }
+    }
+
+    /// Run the analysis.
+    pub fn analyze(&mut self) -> anyhow::Result<()> {
+        self.map_locations()?;
+        self.analyze_statements()?;
+        Ok(())
+    }
+
+    /// Populate the maps so that we can always unwrap later.
+    fn map_locations(&mut self) -> anyhow::Result<()> {
+        self.location_map = LocationMapper::map_locations(&self.source)?;
+
+        self.statements_count = HashMap::with_capacity(self.location_map.len());
+        self.statements_size = HashMap::with_capacity(self.location_map.len());
+
+        for statement in self.location_map.values() {
+            if !self.statements_size.contains_key(statement) {
+                self.statements_size.insert(statement.clone(), 0);
+            }
+
+            *self.statements_count.entry(statement.clone()).or_insert(0) += 1;
+        }
+
+        Ok(())
+    }
+
+    /// Analyze how much bytes of insturctions each statement contributes.
+    fn analyze_statements(&mut self) -> anyhow::Result<()> {
+        let mut previous_offset = 0;
+        let mut previous_location = Location::new(0, 0);
+
+        for line in self
+            .debug_lines
+            .lines()
+            .skip_while(|line| !line.starts_with("Address"))
+            .skip(2)
+        {
+            let mut parts = line.split_whitespace();
+            let (Some(offset), Some(line), Some(column)) =
+                (parts.next(), parts.next(), parts.next())
+            else {
+                continue;
+            };
+
+            let current_offset = u64::from_str_radix(offset.trim_start_matches("0x"), 16)?;
+            let mut current_location = Location::new(line.parse()?, column.parse()?);
+
+            // TODO: A bug? Needs further investigation.
+            if current_location.line == 0 && current_location.column != 0 {
+                current_location.line = previous_location.line;
+            }
+
+            if let Some(statement) = self.location_map.get(&previous_location) {
+                let contribution = current_offset - previous_offset;
+                *self.statements_size.get_mut(statement).unwrap() += contribution;
+            }
+
+            previous_offset = current_offset;
+            previous_location = current_location;
+        }
+
+        Ok(())
+    }
+
+    /// Print the per-statement count break-down.
+    pub fn display_statement_count(&self) {
+        println!("statements count:");
+        for (statement, count) in self.statements_count.iter() {
+            println!("\t{statement} {count}");
+        }
+    }
+
+    /// Print the per-statement byte size contribution break-down.
+    pub fn display_statement_size(&self) {
+        println!("bytes per statement:");
+        for (statement, size) in self.statements_size.iter() {
+            println!("\t{statement} {size}");
+        }
+    }
+
+    /// Print the estimated `yul-phaser` cost parameters.
+    pub fn display_phaser_costs(&self, yul_phaser_scale: u64) {
+        println!("yul-phaser parameters:");
+        for (parameter, cost) in self.phaser_costs(yul_phaser_scale) {
+            println!("\t{parameter} {cost}");
+        }
+    }
+
+    /// Estimate the `yul-phaser` costs using the simplified weight function:
+    /// `Total size / toal count = cost`
+    pub fn phaser_costs(&self, yul_phaser_scale: u64) -> Vec<(String, u64)> {
+        let mut costs: HashMap<String, (usize, u64)> = HashMap::with_capacity(16);
+        for (statement, count) in self
+            .statements_count
+            .iter()
+            .filter(|(_, count)| **count > 0)
+        {
+            let size = self.statements_size.get(statement).unwrap();
+            let cost = match statement.as_str() {
+                location_mapper::FOR => "--for-loop-cost",
+                location_mapper::OTHER => continue,
+                location_mapper::INTERNAL => continue,
+                location_mapper::BLOCK => "--block-cost",
+                location_mapper::FUNCTION_CALL => "--function-call-cost",
+                location_mapper::IF => "--if-cost",
+                location_mapper::SWITCH => "--switch-cost",
+                location_mapper::DECLARATION => "--variable-declaration-cost",
+                location_mapper::ASSIGNMENT => "--assignment-cost",
+                location_mapper::FUNCTION_DEFINITION => "--function-definition-cost",
+                location_mapper::IDENTIFIER => "--identifier-cost",
+                location_mapper::LITERAL => "--literal-cost",
+                _ => "--expression-statement-cost",
+            };
+
+            let entry = costs.entry(cost.to_string()).or_default();
+            entry.0 += count;
+            entry.1 += size;
+        }
+
+        let costs = costs
+            .iter()
+            .map(|(cost, (count, size))| {
+                let ratio = *size / *count as u64;
+                (cost.to_string(), ratio.min(100))
+            })
+            .collect::<Vec<_>>();
+
+        let scaled_costs = scale_to(
+            costs
+                .iter()
+                .map(|(_, ratio)| *ratio)
+                .collect::<Vec<_>>()
+                .as_slice(),
+            yul_phaser_scale,
+        );
+
+        costs
+            .iter()
+            .zip(scaled_costs)
+            .map(|((cost, _), scaled_ratio)| (cost.to_string(), scaled_ratio))
+            .collect()
+    }
+}
+
+/// Given a slice of u64 values, returns a `Vec<u64>` where each element
+/// is linearly scaled into the closed interval [1, 10].
+fn scale_to(data: &[u64], scale_max: u64) -> Vec<u64> {
+    if data.is_empty() {
+        return Vec::new();
+    }
+
+    let mut min = data[0];
+    let mut max = data[0];
+    for &x in &data[1..] {
+        if x < min {
+            min = x;
+        }
+        if x > max {
+            max = x;
+        }
+    }
+    if max < scale_max {
+        return data.to_vec();
+    }
+
+    let range = max - min;
+    data.iter()
+        .map(|&x| {
+            if range == 0 {
+                1
+            } else {
+                1 + (x - min) * scale_max / range
+            }
+        })
+        .collect()
+}

crates/explorer/src/lib.rs

@@ -0,0 +1,6 @@
+//! The revive explorer leverages debug info to get insights into emitted code.
+
+pub mod dwarfdump;
+pub mod dwarfdump_analyzer;
+pub mod location_mapper;
+pub mod yul_phaser;

crates/explorer/src/location_mapper.rs

@@ -0,0 +1,123 @@
+//! The location mapper utility maps YUL source locations to AST statements.
+
+use std::{collections::HashMap, path::Path};
+
+use revive_yul::{
+    lexer::{token::location::Location, Lexer},
+    parser::{
+        identifier::Identifier,
+        statement::{
+            assignment::Assignment,
+            block::Block,
+            expression::{function_call::FunctionCall, literal::Literal},
+            for_loop::ForLoop,
+            function_definition::FunctionDefinition,
+            if_conditional::IfConditional,
+            object::Object,
+            switch::Switch,
+            variable_declaration::VariableDeclaration,
+        },
+    },
+    visitor::{AstNode, AstVisitor},
+};
+
+/// Code attributed to an unknown location.
+pub const OTHER: &str = "other";
+/// Code attributed to a compiler internal location.
+pub const INTERNAL: &str = "internal";
+/// Code attributed to a block.
+pub const BLOCK: &str = "block";
+/// Code attributed to a function call.
+pub const FUNCTION_CALL: &str = "function_call";
+/// Code attributed to a for loop.
+pub const FOR: &str = "for";
+/// Code attributed to an if statement.
+pub const IF: &str = "if";
+/// Code attributed to a switch statement.
+pub const SWITCH: &str = "switch";
+/// Code attributed to a variable declaration.
+pub const DECLARATION: &str = "let";
+/// Code attributed to a variable assignement.
+pub const ASSIGNMENT: &str = "assignment";
+/// Code attributed to a function definition.
+pub const FUNCTION_DEFINITION: &str = "function_definition";
+/// Code attributed to an identifier.
+pub const IDENTIFIER: &str = "identifier";
+/// Code attributed to a literal.
+pub const LITERAL: &str = "literal";
+
+/// The location to statements mapper.
+pub struct LocationMapper(HashMap<Location, String>);
+
+impl LocationMapper {
+    /// Construct a node location map from the given YUL `source` file.
+    pub fn map_locations(source: &Path) -> anyhow::Result<HashMap<Location, String>> {
+        let mut lexer = Lexer::new(std::fs::read_to_string(source)?);
+        let ast = Object::parse(&mut lexer, None).map_err(|error| {
+            anyhow::anyhow!("Contract `{}` parsing error: {:?}", source.display(), error)
+        })?;
+
+        let mut location_map = Self(Default::default());
+        ast.accept(&mut location_map);
+        location_map.0.insert(Location::new(0, 0), OTHER.into());
+        location_map.0.insert(Location::new(1, 0), INTERNAL.into());
+
+        Ok(location_map.0)
+    }
+}
+
+impl AstVisitor for LocationMapper {
+    fn visit(&mut self, node: &impl AstNode) {
+        node.visit_children(self);
+    }
+
+    fn visit_block(&mut self, node: &Block) {
+        node.visit_children(self);
+        self.0.insert(node.location, BLOCK.into());
+    }
+
+    fn visit_assignment(&mut self, node: &Assignment) {
+        node.visit_children(self);
+        self.0.insert(node.location, ASSIGNMENT.into());
+    }
+
+    fn visit_if_conditional(&mut self, node: &IfConditional) {
+        node.visit_children(self);
+        self.0.insert(node.location, IF.into());
+    }
+
+    fn visit_variable_declaration(&mut self, node: &VariableDeclaration) {
+        node.visit_children(self);
+        self.0.insert(node.location, DECLARATION.into());
+    }
+
+    fn visit_function_call(&mut self, node: &FunctionCall) {
+        node.visit_children(self);
+        self.0.insert(node.location, node.name.to_string());
+    }
+
+    fn visit_function_definition(&mut self, node: &FunctionDefinition) {
+        node.visit_children(self);
+        self.0.insert(node.location, FUNCTION_DEFINITION.into());
+    }
+
+    fn visit_identifier(&mut self, node: &Identifier) {
+        node.visit_children(self);
+        self.0.insert(node.location, IDENTIFIER.into());
+    }
+
+    fn visit_literal(&mut self, node: &Literal) {
+        node.visit_children(self);
+        self.0.insert(node.location, LITERAL.into());
+    }
+
+    fn visit_for_loop(&mut self, node: &ForLoop) {
+        node.visit_children(self);
+        self.0.insert(node.location, FOR.into());
+    }
+
+    fn visit_switch(&mut self, node: &Switch) {
+        node.visit_children(self);
+        self.0.insert(node.location, SWITCH.into());
+    }
+}

crates/explorer/src/main.rs

@@ -0,0 +1,59 @@
+use std::path::PathBuf;
+
+use clap::Parser;
+
+use revive_explorer::{dwarfdump, dwarfdump_analyzer::DwarfdumpAnalyzer, yul_phaser};
+
+/// The `revive-explorer` is a helper utility for exploring the compilers YUL lowering unit.
+///
+/// It analyzes a given shared objects from the debug dump and outputs:
+/// - The count of each YUL statement translated.
+/// - A per YUL statement break-down of bytecode size contributed per.
+/// - Estimated `yul-phaser` cost parameters.
+///
+/// Note: This tool might not be fully accurate, especially when the code was optimized.
+#[derive(Parser, Debug)]
+#[command(version, about, long_about = None)]
+struct Args {
+    /// Path of the dwarfdump executable.
+    #[arg(short, long)]
+    dwarfdump: Option<PathBuf>,
+
+    /// The YUL phaser cost scale maximum value.
+    #[arg(short, long, default_value_t = 10)]
+    cost_scale: u64,
+
+    /// Run the provided yul-phaser executable using the estimated costs.
+    #[arg(short, long)]
+    yul_phaser: Option<PathBuf>,
+
+    /// Path of the shared object to analyze.
+    /// It must have been compiled with debug info (-g).
+    file: PathBuf,
+}
+
+fn main() -> anyhow::Result<()> {
+    let args = Args::parse();
+
+    let source_file = dwarfdump::source_file(&args.file, &args.dwarfdump)?;
+    let debug_lines = dwarfdump::debug_lines(&args.file, &args.dwarfdump)?;
+    let mut analyzer = DwarfdumpAnalyzer::new(source_file.as_path(), debug_lines);
+
+    analyzer.analyze()?;
+
+    if let Some(path) = args.yul_phaser.as_ref() {
+        yul_phaser::run(
+            path,
+            source_file.as_path(),
+            analyzer.phaser_costs(args.cost_scale).as_slice(),
+            num_cpus::get() / 2, // TODO: should be configurable.
+        )?;
+        return Ok(());
+    }
+
+    analyzer.display_statement_count();
+    analyzer.display_statement_size();
+    analyzer.display_phaser_costs(args.cost_scale);
+
+    Ok(())
+}

crates/explorer/src/yul_phaser.rs

@@ -0,0 +1,79 @@
+//! The revive explorer YUL phaser utility library.
+//!
+//! This can be used to invoke the `yul-phaser` utility,
+//! used to find better YUL optimizer sequences.
+
+use std::{
+    path::{Path, PathBuf},
+    process::{Command, Stdio},
+    thread,
+    time::{SystemTime, UNIX_EPOCH},
+};
+
+/// The `yul-phaser` sane default arguments:
+/// - Less verbose output.
+/// - Sufficient rounds.
+/// - Sufficient random population start.
+const ARGUMENTS: [&str; 6] = [
+    "--hide-round",
+    "--rounds",
+    "1000",
+    "--random-population",
+    "100",
+    "--show-only-top-chromosome",
+];
+
+/// Run multiple YUL phaser executables in parallel.
+pub fn run(
+    executable: &Path,
+    source: &Path,
+    costs: &[(String, u64)],
+    n_threads: usize,
+) -> anyhow::Result<()> {
+    let mut handles = Vec::with_capacity(n_threads);
+
+    for n in 0..n_threads {
+        let executable = executable.to_path_buf();
+        let source = source.to_path_buf();
+        let costs = costs.to_vec();
+
+        handles.push(thread::spawn(move || {
+            spawn_process(executable, source, costs, n)
+        }));
+    }
+
+    for handle in handles {
+        let _ = handle.join();
+    }
+
+    Ok(())
+}
+
+/// The `yul-phaser` process spawning helper function.
+fn spawn_process(
+    executable: PathBuf,
+    source: PathBuf,
+    costs: Vec<(String, u64)>,
+    seed: usize,
+) -> anyhow::Result<()> {
+    let cost_parameters = costs
+        .iter()
+        .flat_map(|(parameter, cost)| vec![parameter.clone(), cost.to_string()]);
+
+    let secs = SystemTime::now()
+        .duration_since(UNIX_EPOCH)
+        .expect("Time went backwards")
+        .as_secs();
+
+    Command::new(executable)
+        .args(cost_parameters)
+        .args(ARGUMENTS)
+        .arg("--seed")
+        .arg((seed + secs as usize).to_string())
+        .arg(source)
+        .stdin(Stdio::null())
+        .spawn()?
+        .wait()?;
+
+    Ok(())
+}

crates/integration/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "revive-integration"
-version.workspace = true
+version = "0.3.0"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true
@@ -13,7 +13,7 @@ alloy-sol-types = { workspace = true }
 hex = { workspace = true }
 serde_json = { workspace = true }
 
-revive-solidity = { workspace = true }
+resolc = { workspace = true }
 revive-runner = { workspace = true }
 revive-llvm-context = { workspace = true }
 

crates/integration/codesize.json

@@ -1,10 +1,10 @@
 {
-  "Baseline": 939,
+  "Baseline": 914,
-  "Computation": 2282,
+  "Computation": 2295,
-  "DivisionArithmetics": 8849,
+  "DivisionArithmetics": 14496,
-  "ERC20": 18308,
+  "ERC20": 17482,
-  "Events": 1640,
+  "Events": 1674,
-  "FibonacciIterative": 1497,
+  "FibonacciIterative": 1490,
-  "Flipper": 2099,
+  "Flipper": 2086,
-  "SHA1": 8243
+  "SHA1": 8158
 }

crates/integration/contracts/AddModMulMod.sol

@@ -0,0 +1,81 @@
+// SPDX-License-Identifier: GPL-3.0
+
+pragma solidity ^0.8;
+
+/* runner.json
+{
+    "differential": true,
+    "actions": [
+        {
+            "Upload": {
+                "code": {
+                    "Solidity": {
+                        "contract": "AddModMulMod"
+                    }
+                }
+            }
+        },
+        {
+            "Instantiate": {
+                "value": 123123,
+                "code": {
+                    "Solidity": {
+                        "contract": "AddModMulModTester"
+                    }
+                }
+            }
+        },
+        {
+            "VerifyCall": {
+                "success": true
+            }
+        }
+    ]
+}
+*/
+
+contract AddModMulMod {
+    function test() public returns (uint256) {
+        // Note that this only works because computation on literals is done using
+        // unbounded integers.
+        if ((2**255 + 2**255) % 7 != addmod(2**255, 2**255, 7)) return 1;
+        if ((2**255 + 2**255) % 7 != addmod(2**255, 2**255, 7)) return 2;
+        return 0;
+    }
+
+    function f(uint256 d) public pure returns (uint256) {
+        addmod(1, 2, d);
+        return 2;
+    }
+
+    function g(uint256 d) public pure returns (uint256) {
+        mulmod(1, 2, d);
+        return 2;
+    }
+
+    function h() public pure returns (uint256) {
+        mulmod(0, 1, 2);
+        mulmod(1, 0, 2);
+        addmod(0, 1, 2);
+        addmod(1, 0, 2);
+        return 2;
+    }
+}
+
+contract AddModMulModTester {
+    constructor() payable {
+        AddModMulMod c = new AddModMulMod();
+
+        assert(c.test() == 0);
+
+        try c.f(0) returns (uint m) { revert(); } catch Panic(uint errorCode) {
+            assert(errorCode == 0x12);
+        }
+
+        try c.g(0) returns (uint m) { revert(); } catch Panic(uint errorCode) {
+            assert(errorCode == 0x12);
+        }
+
+        assert(c.h() == 2);
+    } 
+}

crates/integration/contracts/MemoryBounds.sol

@@ -0,0 +1,36 @@
+// SPDX-License-Identifier: MIT
+
+pragma solidity ^0.8.24;
+
+/* runner.json
+{
+  "differential": true,
+  "actions": [
+    {
+      "Instantiate": {
+        "code": {
+          "Solidity": {
+            "contract": "MemoryBounds"
+          }
+        }
+      }
+    },
+    {
+      "Call": {
+        "dest": {
+          "Instantiated": 0
+        }
+      }
+    }
+  ]
+}
+*/
+
+contract MemoryBounds {
+    fallback() external {
+        assembly {
+            // Accessing OOB offsets should always work when the length is 0.
+            return(100000, 0)
+        }
+    }
+}

crates/integration/contracts/SAR.sol

@@ -0,0 +1,49 @@
+// SPDX-License-Identifier: MIT
+pragma solidity ^0.8;
+
+/* runner.json
+{
+    "differential": true,
+    "actions": [
+        {
+            "Instantiate": {
+                "code": {
+                    "Solidity": {
+                        "contract": "SAR"
+                    }
+                }
+            }
+        }
+    ]
+}
+*/
+
+contract SAR {
+  constructor() payable {
+    assert(sar(0x03, 0x01) == 0x01);
+    assert(
+      sar(
+        0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff,
+        0x01
+      ) == 0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
+    );
+    assert(
+      sar(
+        0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff,
+        0xff
+      ) == 0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
+    );
+    assert(
+      sar(
+        0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff,
+        0x100
+      ) == 0xffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffffff
+    );
+  }
+
+  function sar(uint256 a, uint256 b) public pure returns (uint256 c) {
+    assembly {
+      c := sar(b, a)
+    }
+  }
+}

crates/integration/contracts/Selfdestruct.sol

@@ -0,0 +1,66 @@
+// SPDX-License-Identifier: MIT
+
+pragma solidity ^0.8;
+
+// TODO: This currently fails the differential test.
+// The pallet doesn't send the correct balance back.
+
+/* runner.json
+{
+    "differential": false,
+    "actions": [
+        {
+            "Upload": {
+                "code": {
+                    "Solidity": {
+                        "contract": "SelfdestructTester"
+                    }
+                }
+            }
+        },
+        {
+            "Instantiate": {
+                "code": {
+                    "Solidity": {
+                        "contract": "Selfdestruct"
+                    }
+                },
+                "value": 123456789
+            }
+        },
+        {
+            "Call": {
+                "dest": {
+                    "Instantiated": 0
+                }
+            }
+        }
+    ]
+}
+*/
+
+contract Selfdestruct {
+    address tester;
+    uint value;
+
+    constructor() payable {
+        require(msg.value > 0, "the test should have value");
+        value = msg.value;
+
+        SelfdestructTester s = new SelfdestructTester{value: msg.value}();
+        tester = address(s);
+    }
+
+    fallback() external {
+        (bool success, ) = tester.call(hex"");
+        require(success, "the call to the self destructing contract should succeed");
+    }
+}
+
+contract SelfdestructTester {
+    constructor() payable {}
+
+    fallback() external {
+        selfdestruct(payable(msg.sender));
+    }
+}

crates/integration/src/cases.rs

@@ -1,8 +1,8 @@
 use alloy_primitives::{Address, Bytes, I256, U256};
 use alloy_sol_types::{sol, SolCall, SolConstructor};
 
+use resolc::test_utils::*;
 use revive_llvm_context::OptimizerSettings;
-use revive_solidity::test_utils::*;
 
 #[derive(Clone)]
 pub struct Contract {
@@ -12,6 +12,26 @@ pub struct Contract {
     pub calldata: Vec<u8>,
 }
 
+impl Contract {
+    pub fn build(calldata: Vec<u8>, name: &'static str, code: &str) -> Self {
+        Self {
+            name,
+            evm_runtime: compile_evm_bin_runtime(name, code),
+            pvm_runtime: compile_blob(name, code),
+            calldata,
+        }
+    }
+
+    pub fn build_size_opt(calldata: Vec<u8>, name: &'static str, code: &str) -> Self {
+        Self {
+            name,
+            evm_runtime: compile_evm_bin_runtime(name, code),
+            pvm_runtime: compile_blob_with_options(name, code, true, OptimizerSettings::size()),
+            calldata,
+        }
+    }
+}
+
 macro_rules! case {
     // Arguments:
     //     1. The file name, expect to live under "../contracts/"
@@ -261,26 +281,6 @@ sol!(
 case!("AddressPredictor.sol", Predicted, constructorCall, predicted_constructor, salt: U256);
 case!("AddressPredictor.sol", AddressPredictor, constructorCall, address_predictor_constructor, salt: U256, bytecode: Bytes);
 
-impl Contract {
-    pub fn build(calldata: Vec<u8>, name: &'static str, code: &str) -> Self {
-        Self {
-            name,
-            evm_runtime: compile_evm_bin_runtime(name, code),
-            pvm_runtime: compile_blob(name, code),
-            calldata,
-        }
-    }
-
-    pub fn build_size_opt(calldata: Vec<u8>, name: &'static str, code: &str) -> Self {
-        Self {
-            name,
-            evm_runtime: compile_evm_bin_runtime(name, code),
-            pvm_runtime: compile_blob_with_options(name, code, true, OptimizerSettings::size()),
-            calldata,
-        }
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use rayon::iter::{IntoParallelIterator, ParallelIterator};

crates/integration/src/tests.rs

@@ -60,6 +60,10 @@ test_spec!(mload, "MLoad", "MLoad.sol");
 test_spec!(delegate_no_contract, "DelegateCaller", "DelegateCaller.sol");
 test_spec!(function_type, "FunctionType", "FunctionType.sol");
 test_spec!(layout_at, "LayoutAt", "LayoutAt.sol");
+test_spec!(shift_arithmetic_right, "SAR", "SAR.sol");
+test_spec!(add_mod_mul_mod, "AddModMulModTester", "AddModMulMod.sol");
+test_spec!(memory_bounds, "MemoryBounds", "MemoryBounds.sol");
+test_spec!(selfdestruct, "Selfdestruct", "Selfdestruct.sol");
 
 fn instantiate(path: &str, contract: &str) -> Vec<SpecsAction> {
     vec![Instantiate {
@@ -166,6 +170,8 @@ fn signed_division() {
         (minus_five, two),
         (I256::MINUS_ONE, I256::MIN),
         (one, I256::ZERO),
+        (I256::MIN, I256::MINUS_ONE),
+        (I256::MIN + I256::ONE, I256::MINUS_ONE),
     ] {
         actions.push(Call {
             origin: TestAddress::Alice,

crates/linker/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "revive-linker"
-version.workspace = true
+version = "0.2.0"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true
@@ -8,10 +8,10 @@ authors.workspace = true
 description = "revive compiler linker utils"
 
 [dependencies]
-tempfile = { workspace = true }
-polkavm-linker = { workspace = true }
-libc = { workspace = true }
 anyhow = { workspace = true }
+libc = { workspace = true }
+polkavm-linker = { workspace = true }
+tempfile = { workspace = true }
 
-revive-builtins = { workspace = true }
 lld-sys = { workspace = true }
+revive-builtins = { workspace = true }

crates/linker/src/elf.rs

@@ -0,0 +1,114 @@
+//! The revive ELF object linker library.
+
+use std::{ffi::CString, fs, path::PathBuf, sync::Mutex};
+
+use lld_sys::LLDELFLink;
+use tempfile::TempDir;
+
+use revive_builtins::COMPILER_RT;
+
+static GUARD: Mutex<()> = Mutex::new(());
+
+/// The revive ELF object linker.
+pub struct ElfLinker {
+    temporary_directory: TempDir,
+    output_path: PathBuf,
+    object_path: PathBuf,
+    symbols_path: PathBuf,
+    linker_script_path: PathBuf,
+}
+
+impl ElfLinker {
+    const LINKER_SCRIPT: &str = r#"
+SECTIONS {
+    .text : { KEEP(*(.text.polkavm_export)) *(.text .text.*) }
+}"#;
+
+    const BUILTINS_ARCHIVE_FILE: &str = "libclang_rt.builtins-riscv64.a";
+    const BUILTINS_LIB_NAME: &str = "clang_rt.builtins-riscv64";
+
+    /// The setup routine prepares a temporary working directory.
+    pub fn setup() -> anyhow::Result<Self> {
+        let temporary_directory = TempDir::new()?;
+        let object_path = temporary_directory.path().join("obj.o");
+        let output_path = temporary_directory.path().join("out.o");
+        let symbols_path = temporary_directory.path().join("sym.o");
+        let linker_script_path = temporary_directory.path().join("linker.ld");
+
+        fs::write(&linker_script_path, Self::LINKER_SCRIPT)
+            .map_err(|message| anyhow::anyhow!("{message} {linker_script_path:?}",))?;
+
+        let compiler_rt_path = temporary_directory.path().join(Self::BUILTINS_ARCHIVE_FILE);
+        fs::write(&compiler_rt_path, COMPILER_RT)
+            .map_err(|message| anyhow::anyhow!("{message} {compiler_rt_path:?}"))?;
+
+        Ok(Self {
+            temporary_directory,
+            output_path,
+            object_path,
+            symbols_path,
+            linker_script_path,
+        })
+    }
+
+    /// Link `input` with `symbols` and the `compiler_rt` via `LLD`.
+    pub fn link<T: AsRef<[u8]>>(self, input: T, symbols: T) -> anyhow::Result<Vec<u8>> {
+        fs::write(&self.object_path, input)
+            .map_err(|message| anyhow::anyhow!("{message} {:?}", self.object_path))?;
+
+        fs::write(&self.symbols_path, symbols)
+            .map_err(|message| anyhow::anyhow!("{message} {:?}", self.symbols_path))?;
+
+        if lld(self
+            .create_arguments()
+            .into_iter()
+            .map(|v| v.to_string())
+            .collect())
+        {
+            return Err(anyhow::anyhow!("ld.lld failed"));
+        }
+
+        Ok(fs::read(&self.output_path)?)
+    }
+
+    /// The argument creation helper function.
+    fn create_arguments(&self) -> Vec<String> {
+        [
+            "ld.lld",
+            "--error-limit=0",
+            "--relocatable",
+            "--emit-relocs",
+            "--no-relax",
+            "--unique",
+            "--gc-sections",
+            self.linker_script_path.to_str().expect("should be utf8"),
+            "-o",
+            self.output_path.to_str().expect("should be utf8"),
+            self.object_path.to_str().expect("should be utf8"),
+            self.symbols_path.to_str().expect("should be utf8"),
+            "--library-path",
+            self.temporary_directory
+                .path()
+                .to_str()
+                .expect("should be utf8"),
+            "--library",
+            Self::BUILTINS_LIB_NAME,
+        ]
+        .iter()
+        .map(ToString::to_string)
+        .collect()
+    }
+}
+
+/// The thread-safe LLD helper function.
+fn lld(arguments: Vec<String>) -> bool {
+    let c_strings = arguments
+        .into_iter()
+        .map(|arg| CString::new(arg).expect("ld.lld args should not contain null bytes"))
+        .collect::<Vec<_>>();
+
+    let args: Vec<*const libc::c_char> = c_strings.iter().map(|arg| arg.as_ptr()).collect();
+
+    let _lock = GUARD.lock().expect("ICE: linker mutex should not poison");
+    unsafe { LLDELFLink(args.as_ptr(), args.len()) == 0 }
+}

crates/linker/src/lib.rs

@@ -1,76 +1,4 @@
-use std::{env, ffi::CString, fs};
+//! The revive ELF object to PVM blob linker library.
 
-use lld_sys::LLDELFLink;
+pub mod elf;
-use revive_builtins::COMPILER_RT;
+pub mod pvm;
-
-const LINKER_SCRIPT: &str = r#"
-SECTIONS {
-    .text : { KEEP(*(.text.polkavm_export)) *(.text .text.*) }
-}"#;
-
-const BUILTINS_ARCHIVE_FILE: &str = "libclang_rt.builtins-riscv64.a";
-const BUILTINS_LIB_NAME: &str = "clang_rt.builtins-riscv64";
-
-fn invoke_lld(cmd_args: &[&str]) -> bool {
-    let c_strings = cmd_args
-        .iter()
-        .map(|arg| CString::new(*arg).expect("ld.lld args should not contain null bytes"))
-        .collect::<Vec<_>>();
-
-    let args: Vec<*const libc::c_char> = c_strings.iter().map(|arg| arg.as_ptr()).collect();
-
-    unsafe { LLDELFLink(args.as_ptr(), args.len()) == 0 }
-}
-
-pub fn polkavm_linker<T: AsRef<[u8]>>(code: T, strip_binary: bool) -> anyhow::Result<Vec<u8>> {
-    let mut config = polkavm_linker::Config::default();
-    config.set_strip(strip_binary);
-    config.set_optimize(true);
-
-    polkavm_linker::program_from_elf(config, code.as_ref())
-        .map_err(|reason| anyhow::anyhow!("polkavm linker failed: {}", reason))
-}
-
-pub fn link<T: AsRef<[u8]>>(input: T) -> anyhow::Result<Vec<u8>> {
-    let dir = tempfile::tempdir().expect("failed to create temp directory for linking");
-    let output_path = dir.path().join("out.so");
-    let object_path = dir.path().join("out.o");
-    let linker_script_path = dir.path().join("linker.ld");
-    let compiler_rt_path = dir.path().join(BUILTINS_ARCHIVE_FILE);
-
-    fs::write(&object_path, input).map_err(|msg| anyhow::anyhow!("{msg} {object_path:?}"))?;
-
-    if env::var("PVM_LINKER_DUMP_OBJ").is_ok() {
-        fs::copy(&object_path, "/tmp/out.o")?;
-    }
-
-    fs::write(&linker_script_path, LINKER_SCRIPT)
-        .map_err(|msg| anyhow::anyhow!("{msg} {linker_script_path:?}"))?;
-
-    fs::write(&compiler_rt_path, COMPILER_RT)
-        .map_err(|msg| anyhow::anyhow!("{msg} {compiler_rt_path:?}"))?;
-
-    let ld_args = [
-        "ld.lld",
-        "--error-limit=0",
-        "--relocatable",
-        "--emit-relocs",
-        "--no-relax",
-        "--unique",
-        "--gc-sections",
-        "--library-path",
-        dir.path().to_str().expect("should be utf8"),
-        "--library",
-        BUILTINS_LIB_NAME,
-        linker_script_path.to_str().expect("should be utf8"),
-        object_path.to_str().expect("should be utf8"),
-        "-o",
-        output_path.to_str().expect("should be utf8"),
-    ];
-
-    if invoke_lld(&ld_args) {
-        return Err(anyhow::anyhow!("ld.lld failed"));
-    }
-
-    Ok(fs::read(&output_path)?)
-}

crates/linker/src/pvm.rs

@@ -0,0 +1,10 @@
+//! The revive PVM blob linker library.
+
+pub fn polkavm_linker<T: AsRef<[u8]>>(code: T, strip_binary: bool) -> anyhow::Result<Vec<u8>> {
+    let mut config = polkavm_linker::Config::default();
+    config.set_strip(strip_binary);
+    config.set_optimize(true);
+
+    polkavm_linker::program_from_elf(config, code.as_ref())
+        .map_err(|reason| anyhow::anyhow!("polkavm linker failed: {}", reason))
+}

crates/llvm-builder/Cargo.toml

@@ -6,7 +6,7 @@ authors = [
     "Anton Baliasnikov <aba@matterlabs.dev>",
     "Cyrill Leutwiler <cyrill@parity.io>",
 ]
-version.workspace = true
+version = "0.4.0"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true

crates/llvm-builder/README.md

@@ -59,7 +59,7 @@ Obtain a compatible build for your host platform from the release section of thi
 
    * Install the builder using `cargo`:
       ```shell
-      cargo install --git https://github.com/paritytech/revive-llvm-builder --force --locked
+      cargo install --force --locked --path crates/llvm-builder
       ```
 
       > The builder is not the LLVM framework itself, but a tool that clones its repository and runs a sequence of build commands. By default it is installed in `~/.cargo/bin/`, which is recommended to be added to your `$PATH`.
@@ -88,7 +88,12 @@ Obtain a compatible build for your host platform from the release section of thi
 
       Build artifacts end up in the `./target-llvm/gnu/target-final/` directory by default.
       The `gnu` directory depends on the supported archticture and will either be `gnu`, `musl` or `emscripten`.
-      You now need to export the final target directory `$LLVM_SYS_181_PREFIX`: `export LLVM_SYS_181_PREFIX=${PWD}/target-llvm/gnu/target-final`
+      You now need to export the final target directory `$LLVM_SYS_181_PREFIX`:
+
+      ```shell
+      export LLVM_SYS_181_PREFIX=${PWD}/target-llvm/gnu/target-final
+      ```
+
       If built with the `--enable-tests` option, test tools will be in the `./target-llvm/gnu/build-final/` directory, along with copies of the build artifacts. For all supported build options, run `revive-llvm build --help`.
 
 </details>

crates/llvm-builder/src/build_type.rs

@@ -22,7 +22,7 @@ impl std::str::FromStr for BuildType {
             "Release" => Ok(Self::Release),
             "RelWithDebInfo" => Ok(Self::RelWithDebInfo),
             "MinSizeRel" => Ok(Self::MinSizeRel),
-            value => Err(format!("Unsupported build type: `{}`", value)),
+            value => Err(format!("Unsupported build type: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/builtins.rs

@@ -117,13 +117,7 @@ pub fn build(
     log::info!("building compiler-rt for rv64emac");
 
     crate::utils::check_presence("cmake")?;
-
-    let generator = if cfg!(target_os = "windows") {
-        "Visual Studio 17 2022"
-    } else {
     crate::utils::check_presence("ninja")?;
-        "Ninja"
-    };
 
     let llvm_module_compiler_rt = crate::LLVMPath::llvm_module_compiler_rt()?;
     let llvm_compiler_rt_build = crate::LLVMPath::llvm_build_compiler_rt()?;
@@ -136,7 +130,7 @@ pub fn build(
                 "-B",
                 llvm_compiler_rt_build.to_string_lossy().as_ref(),
                 "-G",
-                generator,
+                "Ninja",
             ])
             .args(CMAKE_STATIC_ARGS)
             .args(cmake_dynamic_args(build_type, target_env)?)

crates/llvm-builder/src/ccache_variant.rs

@@ -16,7 +16,7 @@ impl std::str::FromStr for CcacheVariant {
         match value {
             "ccache" => Ok(Self::Ccache),
             "sccache" => Ok(Self::Sccache),
-            value => Err(format!("Unsupported ccache variant: `{}`", value)),
+            value => Err(format!("Unsupported ccache variant: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/lib.rs

@@ -127,23 +127,22 @@ pub fn build(
     sanitizer: Option<sanitizer::Sanitizer>,
     enable_valgrind: bool,
 ) -> anyhow::Result<()> {
-    log::trace!("build type: {:?}", build_type);
+    log::trace!("build type: {build_type:?}");
-    log::trace!("target env: {:?}", target_env);
+    log::trace!("target env: {target_env:?}");
-    log::trace!("targets: {:?}", targets);
+    log::trace!("targets: {targets:?}");
-    log::trace!("llvm projects: {:?}", llvm_projects);
+    log::trace!("llvm projects: {llvm_projects:?}");
-    log::trace!("enable rtti: {:?}", enable_rtti);
+    log::trace!("enable rtti: {enable_rtti:?}");
-    log::trace!("default target: {:?}", default_target);
+    log::trace!("default target: {default_target:?}");
-    log::trace!("eneable tests: {:?}", enable_tests);
+    log::trace!("eneable tests: {enable_tests:?}");
-    log::trace!("enable_coverage: {:?}", enable_coverage);
+    log::trace!("enable_coverage: {enable_coverage:?}");
-    log::trace!("extra args: {:?}", extra_args);
+    log::trace!("extra args: {extra_args:?}");
-    log::trace!("sanitzer: {:?}", sanitizer);
+    log::trace!("sanitzer: {sanitizer:?}");
-    log::trace!("enable valgrind: {:?}", enable_valgrind);
+    log::trace!("enable valgrind: {enable_valgrind:?}");
 
     if !PathBuf::from(LLVMPath::DIRECTORY_LLVM_SOURCE).exists() {
         log::error!(
-            "LLVM project source directory {} does not exist (run `revive-llvm --target-env {} clone`)",
+            "LLVM project source directory {} does not exist (run `revive-llvm --target-env {target_env} clone`)",
-            LLVMPath::DIRECTORY_LLVM_SOURCE,
+            LLVMPath::DIRECTORY_LLVM_SOURCE
-            target_env
         )
     }
 

crates/llvm-builder/src/llvm_project.rs

@@ -22,7 +22,7 @@ impl std::str::FromStr for LLVMProject {
             "lld" => Ok(Self::LLD),
             "lldb" => Ok(Self::LLDB),
             "mlir" => Ok(Self::MLIR),
-            value => Err(format!("Unsupported LLVM project to enable: `{}`", value)),
+            value => Err(format!("Unsupported LLVM project to enable: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/lock.rs

@@ -30,7 +30,7 @@ impl TryFrom<&PathBuf> for Lock {
     fn try_from(path: &PathBuf) -> Result<Self, Self::Error> {
         let mut config_str = String::new();
         let mut config_file =
-            File::open(path).with_context(|| format!("Error opening {:?} file", path))?;
+            File::open(path).with_context(|| format!("Error opening {path:?} file"))?;
         config_file.read_to_string(&mut config_str)?;
         Ok(toml::from_str(&config_str)?)
     }

crates/llvm-builder/src/platforms/mod.rs

@@ -29,7 +29,7 @@ impl FromStr for Platform {
     fn from_str(value: &str) -> Result<Self, Self::Err> {
         match value {
             "PolkaVM" => Ok(Self::PolkaVM),
-            value => Err(format!("Unsupported platform: `{}`", value)),
+            value => Err(format!("Unsupported platform: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/revive_llvm/main.rs

@@ -69,8 +69,8 @@ fn main_inner() -> anyhow::Result<()> {
                 })
                 .collect();
 
-            log::debug!("extra_args: {:#?}", extra_args);
+            log::debug!("extra_args: {extra_args:#?}");
-            log::debug!("extra_args_unescaped: {:#?}", extra_args_unescaped);
+            log::debug!("extra_args_unescaped: {extra_args_unescaped:#?}");
 
             if let Some(ccache_variant) = ccache_variant {
                 revive_llvm_builder::utils::check_presence(ccache_variant.to_string().as_str())?;

crates/llvm-builder/src/sanitizer.rs

@@ -30,7 +30,7 @@ impl std::str::FromStr for Sanitizer {
             "thread" => Ok(Self::Thread),
             "dataflow" => Ok(Self::DataFlow),
             "address;undefined" => Ok(Self::AddressUndefined),
-            value => Err(format!("Unsupported sanitizer: `{}`", value)),
+            value => Err(format!("Unsupported sanitizer: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/target_env.rs

@@ -20,7 +20,7 @@ impl std::str::FromStr for TargetEnv {
             "gnu" => Ok(Self::GNU),
             "musl" => Ok(Self::MUSL),
             "emscripten" => Ok(Self::Emscripten),
-            value => Err(format!("Unsupported target environment: `{}`", value)),
+            value => Err(format!("Unsupported target environment: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/target_triple.rs

@@ -15,7 +15,7 @@ impl std::str::FromStr for TargetTriple {
     fn from_str(value: &str) -> Result<Self, Self::Err> {
         match value {
             "polkavm" => Ok(Self::PolkaVM),
-            value => Err(format!("Unsupported target triple: `{}`", value)),
+            value => Err(format!("Unsupported target triple: `{value}`")),
         }
     }
 }

crates/llvm-builder/src/utils.rs

@@ -38,7 +38,7 @@ pub const MUSL_SNAPSHOTS_URL: &str = "https://git.musl-libc.org/cgit/musl/snapsh
 pub const EMSDK_SOURCE_URL: &str = "https://github.com/emscripten-core/emsdk.git";
 
 /// The emscripten SDK version.
-pub const EMSDK_VERSION: &str = "3.1.64";
+pub const EMSDK_VERSION: &str = "4.0.9";
 
 /// The subprocess runner.
 ///
@@ -92,7 +92,7 @@ pub fn unpack_tar(filename: PathBuf, path: &str) -> anyhow::Result<()> {
 pub fn download_musl(name: &str) -> anyhow::Result<()> {
     log::info!("downloading musl {name}");
     let tar_file_name = format!("{name}.tar.gz");
-    let url = format!("{}/{tar_file_name}", MUSL_SNAPSHOTS_URL);
+    let url = format!("{MUSL_SNAPSHOTS_URL}/{tar_file_name}");
     let target_path = crate::llvm_path::DIRECTORY_LLVM_TARGET
         .get()
         .unwrap()
@@ -223,6 +223,6 @@ pub fn install_emsdk() -> anyhow::Result<()> {
 /// The LLVM target directory default path.
 pub fn directory_target_llvm(target_env: crate::target_env::TargetEnv) -> PathBuf {
     crate::llvm_path::DIRECTORY_LLVM_TARGET
-        .get_or_init(|| PathBuf::from(format!("./target-llvm/{}/", target_env)))
+        .get_or_init(|| PathBuf::from(format!("./target-llvm/{target_env}/")))
         .clone()
 }

crates/llvm-builder/tests/build.rs

@@ -2,20 +2,20 @@ pub mod common;
 
 use std::process::Command;
 
-use assert_cmd::prelude::*;
+use assert_cmd::{cargo, prelude::*};
 
 /// This test verifies that the LLVM repository can be successfully cloned, built, and cleaned.
 #[test]
 fn clone_build_and_clean() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("build")
         .arg("--llvm-projects")
@@ -25,13 +25,13 @@ fn clone_build_and_clean() -> anyhow::Result<()> {
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("builtins")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clean")
         .assert()
@@ -47,13 +47,13 @@ fn clone_build_and_clean() -> anyhow::Result<()> {
 fn clone_build_and_clean_musl() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .arg("clone")
         .current_dir(test_dir.path())
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("build")
         .arg("--llvm-projects")
@@ -63,7 +63,7 @@ fn clone_build_and_clean_musl() -> anyhow::Result<()> {
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .arg("--target-env")
         .arg("musl")
         .arg("build")
@@ -75,7 +75,7 @@ fn clone_build_and_clean_musl() -> anyhow::Result<()> {
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clean")
         .assert()
@@ -91,13 +91,13 @@ fn clone_build_and_clean_musl() -> anyhow::Result<()> {
 fn debug_build_with_tests_coverage() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("build")
         .arg("--enable-coverage")
@@ -120,13 +120,13 @@ fn debug_build_with_tests_coverage() -> anyhow::Result<()> {
 fn build_with_sanitizers() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("build")
         .arg("--sanitizer")
@@ -146,16 +146,16 @@ fn build_with_sanitizers() -> anyhow::Result<()> {
 #[cfg(target_os = "linux")]
 fn clone_build_and_clean_emscripten() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
-    let command = Command::cargo_bin(common::REVIVE_LLVM)?;
+    let command = Command::new(cargo::cargo_bin!("revive-llvm"));
     let program = command.get_program().to_string_lossy();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("build")
         .arg("--llvm-projects")
@@ -183,7 +183,7 @@ fn clone_build_and_clean_emscripten() -> anyhow::Result<()> {
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .arg("clean")
         .current_dir(test_dir.path())
         .assert()

crates/llvm-builder/tests/checkout.rs

@@ -2,7 +2,7 @@ pub mod common;
 
 use std::process::Command;
 
-use assert_cmd::prelude::*;
+use assert_cmd::{cargo, prelude::*};
 
 /// This test verifies that after cloning the LLVM repository, checking out a specific branch
 /// or reference works as expected.
@@ -10,13 +10,13 @@ use assert_cmd::prelude::*;
 fn checkout_after_clone() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("checkout")
         .assert()
@@ -31,13 +31,13 @@ fn checkout_after_clone() -> anyhow::Result<()> {
 fn force_checkout() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
         .success();
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("checkout")
         .arg("--force")

crates/llvm-builder/tests/clone.rs

@@ -2,7 +2,7 @@ pub mod common;
 
 use std::process::Command;
 
-use assert_cmd::prelude::*;
+use assert_cmd::{cargo, prelude::*};
 
 /// This test verifies that the LLVM repository can be successfully cloned using a specific branch
 /// and reference.
@@ -10,7 +10,7 @@ use assert_cmd::prelude::*;
 fn clone() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .assert()
@@ -25,7 +25,7 @@ fn clone() -> anyhow::Result<()> {
 fn clone_deep() -> anyhow::Result<()> {
     let test_dir = common::TestDir::with_lockfile(None)?;
 
-    Command::cargo_bin(common::REVIVE_LLVM)?
+    Command::new(cargo::cargo_bin!("revive-llvm"))
         .current_dir(test_dir.path())
         .arg("clone")
         .arg("--deep")

crates/llvm-context/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "revive-llvm-context"
-version.workspace = true
+version = "0.5.0"
 license.workspace = true
 edition.workspace = true
 repository.workspace = true
@@ -20,7 +20,6 @@ itertools = { workspace = true }
 serde = { workspace = true, features = ["derive"] }
 num = { workspace = true }
 hex = { workspace = true }
-sha3 = { workspace = true }
 inkwell = { workspace = true }
 libc = { workspace = true }
 polkavm-disassembler = { workspace = true }

crates/llvm-context/src/debug_config/ir_type.rs

@@ -1,5 +1,9 @@
 //! The debug IR type.
 
+use revive_common::{
+    EXTENSION_LLVM_SOURCE, EXTENSION_OBJECT, EXTENSION_POLKAVM_ASSEMBLY, EXTENSION_YUL,
+};
+
 /// The debug IR type.
 #[allow(clippy::upper_case_acronyms)]
 #[derive(Debug, Clone, Copy, PartialEq, Eq, Hash)]
@@ -11,22 +15,17 @@ pub enum IRType {
     /// Whether to dump the assembly code.
     Assembly,
     /// Whether to dump the ELF shared object
-    SO,
+    Object,
-    /// Whether to jump JSON
-    #[cfg(debug_assertions)]
-    JSON,
 }
 
 impl IRType {
     /// Returns the file extension for the specified IR.
     pub fn file_extension(&self) -> &'static str {
         match self {
-            Self::Yul => revive_common::EXTENSION_YUL,
+            Self::Yul => EXTENSION_YUL,
-            Self::LLVM => revive_common::EXTENSION_LLVM_SOURCE,
+            Self::LLVM => EXTENSION_LLVM_SOURCE,
-            Self::Assembly => revive_common::EXTENSION_POLKAVM_ASSEMBLY,
+            Self::Assembly => EXTENSION_POLKAVM_ASSEMBLY,
-            #[cfg(debug_assertions)]
+            Self::Object => EXTENSION_OBJECT,
-            Self::JSON => revive_common::EXTENSION_JSON,
-            Self::SO => revive_common::EXTENSION_SHARED_OBJECT,
         }
     }
 }

crates/llvm-context/src/debug_config/mod.rs

@@ -1,7 +1,5 @@
 //! The debug configuration.
 
-pub mod ir_type;
-
 use std::path::PathBuf;
 
 use serde::Deserialize;
@@ -9,6 +7,8 @@ use serde::Serialize;
 
 use self::ir_type::IRType;
 
+pub mod ir_type;
+
 /// The debug configuration.
 #[derive(Debug, Default, Serialize, Deserialize, Clone)]
 pub struct DebugConfig {
@@ -16,6 +16,8 @@ pub struct DebugConfig {
     pub output_directory: Option<PathBuf>,
     /// Whether debug info should be emitted.
     pub emit_debug_info: bool,
+    /// The YUL debug output file path.
+    pub contract_path: Option<PathBuf>,
 }
 
 impl DebugConfig {
@@ -24,15 +26,27 @@ impl DebugConfig {
         Self {
             output_directory,
             emit_debug_info,
+            contract_path: None,
         }
     }
 
+    /// Set the current contract path.
+    pub fn set_contract_path(&mut self, contract_path: &str) {
+        self.contract_path = self.yul_source_path(contract_path);
+    }
+
+    pub fn yul_source_path(&self, contract_path: &str) -> Option<PathBuf> {
+        self.output_directory.as_ref().map(|output_directory| {
+            let mut file_path = output_directory.to_owned();
+            let full_file_name = Self::full_file_name(contract_path, None, IRType::Yul);
+            file_path.push(full_file_name);
+            file_path
+        })
+    }
+
     /// Dumps the Yul IR.
     pub fn dump_yul(&self, contract_path: &str, code: &str) -> anyhow::Result<()> {
-        if let Some(output_directory) = self.output_directory.as_ref() {
+        if let Some(file_path) = self.yul_source_path(contract_path) {
-            let mut file_path = output_directory.to_owned();
-            let full_file_name = Self::full_file_name(contract_path, None, IRType::Yul);
-            file_path.push(full_file_name);
             std::fs::write(file_path, code)?;
         }
 
@@ -93,7 +107,7 @@ impl DebugConfig {
     pub fn dump_object(&self, contract_path: &str, code: &[u8]) -> anyhow::Result<()> {
         if let Some(output_directory) = self.output_directory.as_ref() {
             let mut file_path = output_directory.to_owned();
-            let full_file_name = Self::full_file_name(contract_path, None, IRType::SO);
+            let full_file_name = Self::full_file_name(contract_path, None, IRType::Object);
             file_path.push(full_file_name);
             std::fs::write(file_path, code)?;
         }
@@ -101,24 +115,6 @@ impl DebugConfig {
         Ok(())
     }
 
-    /// Dumps the stage output as a json file suitable for use with --recursive-process
-    #[cfg(debug_assertions)]
-    pub fn dump_stage_output(
-        &self,
-        contract_path: &str,
-        contract_suffix: Option<&str>,
-        stage_json: &Vec<u8>,
-    ) -> anyhow::Result<()> {
-        if let Some(output_directory) = self.output_directory.as_ref() {
-            let mut file_path = output_directory.to_owned();
-            let full_file_name = Self::full_file_name(contract_path, contract_suffix, IRType::JSON);
-            file_path.push(full_file_name);
-            std::fs::write(file_path, stage_json)?;
-        }
-
-        Ok(())
-    }
-
     /// Creates a full file name, given the contract full path, suffix, and extension.
     fn full_file_name(contract_path: &str, suffix: Option<&str>, ir_type: IRType) -> String {
         let mut full_file_name = contract_path.replace('/', "_").replace(':', ".");

crates/llvm-context/src/lib.rs

@@ -1,15 +1,16 @@
 //! The LLVM context library.
 
+#![allow(clippy::too_many_arguments)]
+
 use std::ffi::CString;
 use std::sync::OnceLock;
 
 pub use self::debug_config::ir_type::IRType as DebugConfigIR;
 pub use self::debug_config::DebugConfig;
-pub use self::memory::MemoryConfig;
 pub use self::optimizer::settings::size_level::SizeLevel as OptimizerSettingsSizeLevel;
 pub use self::optimizer::settings::Settings as OptimizerSettings;
 pub use self::optimizer::Optimizer;
-pub use self::polkavm::build_assembly_text as polkavm_build_assembly_text;
+pub use self::polkavm::build as polkavm_build;
 pub use self::polkavm::context::address_space::AddressSpace as PolkaVMAddressSpace;
 pub use self::polkavm::context::argument::Argument as PolkaVMArgument;
 pub use self::polkavm::context::attribute::Attribute as PolkaVMAttribute;
@@ -47,6 +48,7 @@ pub use self::polkavm::context::r#loop::Loop as PolkaVMLoop;
 pub use self::polkavm::context::solidity_data::SolidityData as PolkaVMContextSolidityData;
 pub use self::polkavm::context::yul_data::YulData as PolkaVMContextYulData;
 pub use self::polkavm::context::Context as PolkaVMContext;
+pub use self::polkavm::disassemble as polkavm_disassemble;
 pub use self::polkavm::evm::arithmetic as polkavm_evm_arithmetic;
 pub use self::polkavm::evm::bitwise as polkavm_evm_bitwise;
 pub use self::polkavm::evm::call as polkavm_evm_call;
@@ -67,16 +69,15 @@ pub use self::polkavm::evm::memory as polkavm_evm_memory;
 pub use self::polkavm::evm::r#return as polkavm_evm_return;
 pub use self::polkavm::evm::return_data as polkavm_evm_return_data;
 pub use self::polkavm::evm::storage as polkavm_evm_storage;
+pub use self::polkavm::hash as polkavm_hash;
+pub use self::polkavm::link as polkavm_link;
 pub use self::polkavm::r#const as polkavm_const;
-pub use self::polkavm::Dependency as PolkaVMDependency;
-pub use self::polkavm::DummyDependency as PolkaVMDummyDependency;
 pub use self::polkavm::DummyLLVMWritable as PolkaVMDummyLLVMWritable;
 pub use self::polkavm::WriteLLVM as PolkaVMWriteLLVM;
-pub use self::target_machine::target::Target;
+pub use self::target_machine::target::Target as PolkaVMTarget;
-pub use self::target_machine::TargetMachine;
+pub use self::target_machine::TargetMachine as PolkaVMTargetMachine;
 
 pub(crate) mod debug_config;
-pub(crate) mod memory;
 pub(crate) mod optimizer;
 pub(crate) mod polkavm;
 pub(crate) mod target_machine;
@@ -88,7 +89,7 @@ static DID_INITIALIZE: OnceLock<()> = OnceLock::new();
 /// This is a no-op if called subsequentially.
 ///
 /// `llvm_arguments` are passed as-is to the LLVM CL options parser.
-pub fn initialize_llvm(target: Target, name: &str, llvm_arguments: &[String]) {
+pub fn initialize_llvm(target: PolkaVMTarget, name: &str, llvm_arguments: &[String]) {
     let Ok(_) = DID_INITIALIZE.set(()) else {
         return; // Tests don't go through a recursive process
     };
@@ -111,6 +112,6 @@ pub fn initialize_llvm(target: Target, name: &str, llvm_arguments: &[String]) {
     inkwell::support::enable_llvm_pretty_stack_trace();
 
     match target {
-        Target::PVM => inkwell::targets::Target::initialize_riscv(&Default::default()),
+        PolkaVMTarget::PVM => inkwell::targets::Target::initialize_riscv(&Default::default()),
     }
 }

crates/llvm-context/src/memory.rs

@@ -1,21 +0,0 @@
-//! The compile time PolkaVM memory configuration settings.
-
-use serde::{Deserialize, Serialize};
-
-/// The PolkaVM memory configuration.
-#[derive(Clone, Copy, Debug, Deserialize, Serialize)]
-pub struct MemoryConfig {
-    /// The emulated EVM linear heap memory size in bytes.
-    pub heap_size: u32,
-    /// The PVM stack size in bytes.
-    pub stack_size: u32,
-}
-
-impl Default for MemoryConfig {
-    fn default() -> Self {
-        Self {
-            heap_size: 64 * 1024,
-            stack_size: 32 * 1024,
-        }
-    }
-}

crates/llvm-context/src/optimizer/mod.rs

@@ -1,7 +1,5 @@
 //! The LLVM optimizing tools.
 
-pub mod settings;
-
 use serde::Deserialize;
 use serde::Serialize;
 
@@ -9,6 +7,8 @@ use crate::target_machine::TargetMachine;
 
 use self::settings::Settings;
 
+pub mod settings;
+
 /// The LLVM optimizing tools.
 #[derive(Debug, Serialize, Deserialize)]
 pub struct Optimizer {

crates/llvm-context/src/optimizer/settings/mod.rs

@@ -1,8 +1,5 @@
 //! The LLVM optimizer settings.
 
-pub mod size_level;
-
-use revive_solc_json_interface::SolcStandardJsonInputSettingsOptimizer;
 use serde::Deserialize;
 use serde::Serialize;
 
@@ -10,6 +7,8 @@ use itertools::Itertools;
 
 use self::size_level::SizeLevel;
 
+pub mod size_level;
+
 /// The LLVM optimizer and code-gen settings.
 #[derive(Debug, Serialize, Deserialize, Clone, Eq)]
 pub struct Settings {
@@ -20,9 +19,6 @@ pub struct Settings {
     /// The back-end optimization level.
     pub level_back_end: inkwell::OptimizationLevel,
 
-    /// Fallback to optimizing for size if the bytecode is too large.
-    pub is_fallback_to_size_enabled: bool,
-
     /// Whether the LLVM `verify each` option is enabled.
     pub is_verify_each_enabled: bool,
     /// Whether the LLVM `debug logging` option is enabled.
@@ -41,8 +37,6 @@ impl Settings {
             level_middle_end_size,
             level_back_end,
 
-            is_fallback_to_size_enabled: false,
-
             is_verify_each_enabled: false,
             is_debug_logging_enabled: false,
         }
@@ -62,8 +56,6 @@ impl Settings {
             level_middle_end_size,
             level_back_end,
 
-            is_fallback_to_size_enabled: false,
-
             is_verify_each_enabled,
             is_debug_logging_enabled,
         }
@@ -197,16 +189,6 @@ impl Settings {
 
         combinations
     }
-
-    /// Sets the fallback to optimizing for size if the bytecode is too large.
-    pub fn enable_fallback_to_size(&mut self) {
-        self.is_fallback_to_size_enabled = true;
-    }
-
-    /// Whether the fallback to optimizing for size is enabled.
-    pub fn is_fallback_to_size_enabled(&self) -> bool {
-        self.is_fallback_to_size_enabled
-    }
 }
 
 impl PartialEq for Settings {
@@ -227,18 +209,3 @@ impl std::fmt::Display for Settings {
         )
     }
 }
-
-impl TryFrom<&SolcStandardJsonInputSettingsOptimizer> for Settings {
-    type Error = anyhow::Error;
-
-    fn try_from(value: &SolcStandardJsonInputSettingsOptimizer) -> Result<Self, Self::Error> {
-        let mut result = match value.mode {
-            Some(mode) => Self::try_from_cli(mode)?,
-            None => Self::cycles(),
-        };
-        if value.fallback_to_optimizing_for_size.unwrap_or_default() {
-            result.enable_fallback_to_size();
-        }
-        Ok(result)
-    }
-}

crates/llvm-context/src/polkavm/const/mod.rs

@@ -1,10 +1,12 @@
 //! The LLVM context constants.
 
+use revive_common::{BIT_LENGTH_X32, BYTE_LENGTH_WORD};
+
 /// The LLVM framework version.
 pub const LLVM_VERSION: semver::Version = semver::Version::new(18, 1, 4);
 
 /// The pointer width sized type.
-pub static XLEN: usize = revive_common::BIT_LENGTH_X32;
+pub static XLEN: usize = BIT_LENGTH_X32;
 
 /// The calldata size global variable name.
 pub static GLOBAL_CALLDATA_SIZE: &str = "calldatasize";
@@ -20,4 +22,4 @@ pub static GLOBAL_ADDRESS_SPILL_BUFFER: &str = "address_spill_buffer";
 
 /// The deployer call header size that consists of:
 /// - bytecode hash (32 bytes)
-pub const DEPLOYER_CALL_HEADER_SIZE: usize = revive_common::BYTE_LENGTH_WORD;
+pub const DEPLOYER_CALL_HEADER_SIZE: usize = BYTE_LENGTH_WORD;

crates/llvm-context/src/polkavm/context/argument.rs

@@ -66,9 +66,9 @@ impl<'ctx> Argument<'ctx> {
     /// Access the underlying value.
     ///
     /// Will emit a stack load if `self` is a pointer argument.
-    pub fn access<D: crate::polkavm::Dependency + Clone>(
+    pub fn access(
         &self,
-        context: &crate::polkavm::context::Context<'ctx, D>,
+        context: &crate::polkavm::context::Context<'ctx>,
     ) -> anyhow::Result<inkwell::values::BasicValueEnum<'ctx>> {
         match &self.value {
             Value::Register(value) => Ok(*value),
@@ -79,9 +79,9 @@ impl<'ctx> Argument<'ctx> {
     /// Access the underlying value.
     ///
     /// Will emit a stack load if `self` is a pointer argument.
-    pub fn as_pointer<D: crate::polkavm::Dependency + Clone>(
+    pub fn as_pointer(
         &self,
-        context: &crate::polkavm::context::Context<'ctx, D>,
+        context: &crate::polkavm::context::Context<'ctx>,
     ) -> anyhow::Result<crate::polkavm::context::Pointer<'ctx>> {
         match &self.value {
             Value::Register(value) => {

crates/llvm-context/src/polkavm/context/build.rs

@@ -1,7 +1,6 @@
 //! The LLVM module build.
 
-use std::collections::BTreeMap;
+use revive_common::BYTE_LENGTH_WORD;
-
 use serde::Deserialize;
 use serde::Serialize;
 
@@ -9,31 +8,23 @@ use serde::Serialize;
 #[derive(Debug, Serialize, Deserialize)]
 pub struct Build {
     /// The PolkaVM text assembly.
-    pub assembly_text: String,
+    pub assembly_text: Option<String>,
     /// The metadata hash.
-    pub metadata_hash: Option<[u8; revive_common::BYTE_LENGTH_WORD]>,
+    pub metadata_hash: Option<[u8; BYTE_LENGTH_WORD]>,
     /// The PolkaVM binary bytecode.
     pub bytecode: Vec<u8>,
-    /// The PolkaVM bytecode hash.
+    /// The PolkaVM bytecode hash. Unlinked builds don't have a hash yet.
-    pub bytecode_hash: String,
+    pub bytecode_hash: Option<[u8; BYTE_LENGTH_WORD]>,
-    /// The hash-to-full-path mapping of the contract factory dependencies.
-    pub factory_dependencies: BTreeMap<String, String>,
 }
 
 impl Build {
     /// A shortcut constructor.
-    pub fn new(
+    pub fn new(metadata_hash: Option<[u8; BYTE_LENGTH_WORD]>, bytecode: Vec<u8>) -> Self {
-        assembly_text: String,
-        metadata_hash: Option<[u8; revive_common::BYTE_LENGTH_WORD]>,
-        bytecode: Vec<u8>,
-        bytecode_hash: String,
-    ) -> Self {
         Self {
-            assembly_text,
+            assembly_text: None,
             metadata_hash,
             bytecode,
-            bytecode_hash,
+            bytecode_hash: None,
-            factory_dependencies: BTreeMap::new(),
         }
     }
 }

crates/llvm-context/src/polkavm/context/debug_info.rs

@@ -2,6 +2,8 @@
 
 use std::cell::RefCell;
 
+use revive_common::BIT_LENGTH_WORD;
+
 use inkwell::debug_info::AsDIScope;
 use inkwell::debug_info::DIScope;
 
@@ -51,11 +53,20 @@ pub struct DebugInfo<'ctx> {
 
 impl<'ctx> DebugInfo<'ctx> {
     /// A shortcut constructor.
-    pub fn new(module: &inkwell::module::Module<'ctx>) -> Self {
+    pub fn new(
+        module: &inkwell::module::Module<'ctx>,
+        debug_config: &crate::debug_config::DebugConfig,
+    ) -> Self {
+        let module_name = module.get_name().to_string_lossy();
+        let yul_name = debug_config
+            .contract_path
+            .as_ref()
+            .map(|path| path.display().to_string());
+
         let (builder, compile_unit) = module.create_debug_info_builder(
             true,
             inkwell::debug_info::DWARFSourceLanguage::C,
-            module.get_name().to_string_lossy().as_ref(),
+            yul_name.as_deref().unwrap_or_else(|| module_name.as_ref()),
             "",
             "",
             false,
@@ -155,7 +166,7 @@ impl<'ctx> DebugInfo<'ctx> {
         &self,
         flags: Option<inkwell::debug_info::DIFlags>,
     ) -> anyhow::Result<inkwell::debug_info::DIBasicType<'ctx>> {
-        self.create_primitive_type(revive_common::BIT_LENGTH_WORD, flags)
+        self.create_primitive_type(BIT_LENGTH_WORD, flags)
     }
 
     /// Return the DIBuilder.

crates/llvm-context/src/polkavm/context/function/llvm_runtime.rs

@@ -64,24 +64,13 @@ impl<'ctx> LLVMRuntime<'ctx> {
         }
     }
 
-    /// Declares an LLVM runtime function in the `module`,
-    pub fn declare(
-        module: &inkwell::module::Module<'ctx>,
-        name: &str,
-        r#type: inkwell::types::FunctionType<'ctx>,
-        linkage: Option<inkwell::module::Linkage>,
-    ) -> FunctionDeclaration<'ctx> {
-        let value = module.add_function(name, r#type, linkage);
-        FunctionDeclaration::new(r#type, value)
-    }
-
     /// Create the function definition from an existing symbol.
     pub fn define(
         module: &inkwell::module::Module<'ctx>,
         name: &str,
     ) -> Option<FunctionDeclaration<'ctx>> {
         let value = module.get_function(name)?;
-        value.set_linkage(inkwell::module::Linkage::External);
+        value.set_linkage(inkwell::module::Linkage::Private);
         FunctionDeclaration::new(value.get_type(), value).into()
     }
 }

crates/llvm-context/src/polkavm/context/function/mod.rs

@@ -1,12 +1,5 @@
 //! The LLVM IR generator function.
 
-pub mod declaration;
-pub mod intrinsics;
-pub mod llvm_runtime;
-pub mod r#return;
-pub mod runtime;
-pub mod yul_data;
-
 use std::collections::HashMap;
 
 use inkwell::debug_info::AsDIScope;
@@ -20,6 +13,13 @@ use self::declaration::Declaration;
 use self::r#return::Return;
 use self::yul_data::YulData;
 
+pub mod declaration;
+pub mod intrinsics;
+pub mod llvm_runtime;
+pub mod r#return;
+pub mod runtime;
+pub mod yul_data;
+
 /// The LLVM IR generator function.
 #[derive(Debug)]
 pub struct Function<'ctx> {
@@ -75,15 +75,6 @@ impl<'ctx> Function<'ctx> {
         self.name.as_str()
     }
 
-    /// Checks whether the function is defined outside of the front-end.
-    pub fn is_name_external(name: &str) -> bool {
-        name.starts_with("llvm.")
-            || (name.starts_with("__")
-                && name != self::runtime::FUNCTION_ENTRY
-                && name != self::runtime::FUNCTION_DEPLOY_CODE
-                && name != self::runtime::FUNCTION_RUNTIME_CODE)
-    }
-
     /// Returns the LLVM function declaration.
     pub fn declaration(&self) -> Declaration<'ctx> {
         self.declaration
@@ -223,30 +214,11 @@ impl<'ctx> Function<'ctx> {
         self.stack.get(name).copied()
     }
 
-    /// Removes the pointer to a stack variable.
-    pub fn remove_stack_pointer(&mut self, name: &str) {
-        self.stack.remove(name);
-    }
-
     /// Returns the return entity representation.
     pub fn r#return(&self) -> Return<'ctx> {
         self.r#return
     }
 
-    /// Returns the pointer to the function return value.
-    /// # Panics
-    /// If the pointer has not been set yet.
-    pub fn return_pointer(&self) -> Option<Pointer<'ctx>> {
-        self.r#return.return_pointer()
-    }
-
-    /// Returns the return data size in bytes, based on the default stack alignment.
-    /// # Panics
-    /// If the pointer has not been set yet.
-    pub fn return_data_size(&self) -> usize {
-        self.r#return.return_data_size()
-    }
-
     /// Returns the function entry block.
     pub fn entry_block(&self) -> inkwell::basic_block::BasicBlock<'ctx> {
         self.entry_block

crates/llvm-context/src/polkavm/context/function/runtime/arithmetics.rs

@@ -1,22 +1,19 @@
 //! Translates the arithmetic operations.
 
 use inkwell::values::BasicValue;
+use revive_common::BIT_LENGTH_WORD;
 
 use crate::polkavm::context::runtime::RuntimeFunction;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// Implements the division operator according to the EVM specification.
 pub struct Division;
 
-impl<D> RuntimeFunction<D> for Division
+impl RuntimeFunction for Division {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_division";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.word_type().fn_type(
             &[context.word_type().into(), context.word_type().into()],
             false,
@@ -25,7 +22,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let operand_1 = Self::paramater(context, 0).into_int_value();
         let operand_2 = Self::paramater(context, 1).into_int_value();
@@ -39,29 +36,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for Division
+impl WriteLLVM for Division {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Implements the signed division operator according to the EVM specification.
 pub struct SignedDivision;
 
-impl<D> RuntimeFunction<D> for SignedDivision
+impl RuntimeFunction for SignedDivision {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_signed_division";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.word_type().fn_type(
             &[context.word_type().into(), context.word_type().into()],
             false,
@@ -70,7 +61,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let operand_1 = Self::paramater(context, 0).into_int_value();
         let operand_2 = Self::paramater(context, 1).into_int_value();
@@ -96,14 +87,12 @@ where
 
         context.set_basic_block(block_overflow);
         let max_uint = context.builder().build_int_z_extend(
-            context
+            context.integer_type(BIT_LENGTH_WORD - 1).const_all_ones(),
-                .integer_type(revive_common::BIT_LENGTH_WORD - 1)
-                .const_all_ones(),
             context.word_type(),
             "max_uint",
         )?;
         let is_operand_1_overflow = context.builder().build_int_compare(
-            inkwell::IntPredicate::EQ,
+            inkwell::IntPredicate::SLT,
             operand_1,
             context.builder().build_int_neg(max_uint, "min_uint")?,
             "is_operand_1_overflow",
@@ -121,29 +110,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for SignedDivision
+impl WriteLLVM for SignedDivision {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Implements the remainder operator according to the EVM specification.
 pub struct Remainder;
 
-impl<D> RuntimeFunction<D> for Remainder
+impl RuntimeFunction for Remainder {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_remainder";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.word_type().fn_type(
             &[context.word_type().into(), context.word_type().into()],
             false,
@@ -152,7 +135,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let operand_1 = Self::paramater(context, 0).into_int_value();
         let operand_2 = Self::paramater(context, 1).into_int_value();
@@ -166,29 +149,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for Remainder
+impl WriteLLVM for Remainder {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Implements the signed remainder operator according to the EVM specification.
 pub struct SignedRemainder;
 
-impl<D> RuntimeFunction<D> for SignedRemainder
+impl RuntimeFunction for SignedRemainder {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_signed_remainder";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.word_type().fn_type(
             &[context.word_type().into(), context.word_type().into()],
             false,
@@ -197,7 +174,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let operand_1 = Self::paramater(context, 0).into_int_value();
         let operand_2 = Self::paramater(context, 1).into_int_value();
@@ -211,16 +188,13 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for SignedRemainder
+impl WriteLLVM for SignedRemainder {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
@@ -231,13 +205,12 @@ where
 ///
 /// The result is either the calculated quotient or zero,
 ///  selected at runtime.
-fn wrapped_division<'ctx, D, F, T>(
+fn wrapped_division<'ctx, F, T>(
-    context: &Context<'ctx, D>,
+    context: &Context<'ctx>,
     denominator: inkwell::values::IntValue<'ctx>,
     f: F,
 ) -> anyhow::Result<inkwell::values::BasicValueEnum<'ctx>>
 where
-    D: Dependency + Clone,
     F: FnOnce() -> anyhow::Result<T>,
     T: inkwell::values::IntMathValue<'ctx>,
 {

crates/llvm-context/src/polkavm/context/function/runtime/deploy_code.rs

@@ -1,63 +1,52 @@
 //! The deploy code function.
 
-use std::marker::PhantomData;
-
-use crate::polkavm::context::code_type::CodeType;
 use crate::polkavm::context::function::runtime;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// The deploy code function.
 /// Is a special function that is only used by the front-end generated code.
 #[derive(Debug)]
-pub struct DeployCode<B, D>
+pub struct DeployCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
     /// The deploy code AST representation.
     inner: B,
-    /// The `D` phantom data.
-    _pd: PhantomData<D>,
 }
 
-impl<B, D> DeployCode<B, D>
+impl<B> DeployCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
     /// A shortcut constructor.
     pub fn new(inner: B) -> Self {
-        Self {
+        Self { inner }
-            inner,
-            _pd: PhantomData,
-        }
     }
 }
 
-impl<B, D> WriteLLVM<D> for DeployCode<B, D>
+impl<B> WriteLLVM for DeployCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
         let function_type = context.function_type::<inkwell::types::BasicTypeEnum>(vec![], 0);
         context.add_function(
             runtime::FUNCTION_DEPLOY_CODE,
             function_type,
             0,
-            Some(inkwell::module::Linkage::External),
+            Some(inkwell::module::Linkage::Private),
+            None,
+            false,
         )?;
 
         self.inner.declare(context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        context.set_current_function(runtime::FUNCTION_DEPLOY_CODE, None)?;
+        context.set_current_function(runtime::FUNCTION_DEPLOY_CODE, None, false)?;
 
         context.set_basic_block(context.current_function().borrow().entry_block());
-        context.set_code_type(CodeType::Deploy);
 
         self.inner.into_llvm(context)?;
         context.set_debug_location(0, 0, None)?;
@@ -74,7 +63,6 @@ where
         }
 
         context.set_basic_block(context.current_function().borrow().return_block());
-        context.set_debug_location(0, 0, None)?;
         context.build_return(None);
 
         context.pop_debug_scope();

crates/llvm-context/src/polkavm/context/function/runtime/entry.rs

@@ -1,11 +1,16 @@
 //! The entry function.
 
 use inkwell::types::BasicType;
+use revive_common::BIT_LENGTH_ETH_ADDRESS;
+use revive_runtime_api::immutable_data::{
+    GLOBAL_IMMUTABLE_DATA_POINTER, GLOBAL_IMMUTABLE_DATA_SIZE,
+};
+use revive_runtime_api::polkavm_imports::CALL_DATA_SIZE;
+use revive_solc_json_interface::PolkaVMDefaultHeapMemorySize;
 
 use crate::polkavm::context::address_space::AddressSpace;
 use crate::polkavm::context::function::runtime;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// The entry function.
@@ -20,10 +25,7 @@ impl Entry {
 
     /// Initializes the global variables.
     /// The pointers are not initialized, because it's not possible to create a null pointer.
-    pub fn initialize_globals<D>(context: &mut Context<D>) -> anyhow::Result<()>
+    pub fn initialize_globals(context: &mut Context) -> anyhow::Result<()> {
-    where
-        D: Dependency + Clone,
-    {
         context.set_global(
             crate::polkavm::GLOBAL_CALLDATA_SIZE,
             context.xlen_type(),
@@ -38,9 +40,12 @@ impl Entry {
             context.xlen_type().const_zero(),
         );
 
-        let heap_memory_type = context
+        let heap_memory_type = context.byte_type().array_type(
-            .byte_type()
+            context
-            .array_type(context.memory_config.heap_size);
+                .memory_config
+                .heap_size
+                .unwrap_or(PolkaVMDefaultHeapMemorySize),
+        );
         context.set_global(
             crate::polkavm::GLOBAL_HEAP_MEMORY,
             heap_memory_type,
@@ -48,7 +53,7 @@ impl Entry {
             heap_memory_type.const_zero(),
         );
 
-        let address_type = context.integer_type(revive_common::BIT_LENGTH_ETH_ADDRESS);
+        let address_type = context.integer_type(BIT_LENGTH_ETH_ADDRESS);
         context.set_global(
             crate::polkavm::GLOBAL_ADDRESS_SPILL_BUFFER,
             address_type,
@@ -60,16 +65,13 @@ impl Entry {
     }
 
     /// Populate the calldata size global value.
-    pub fn load_calldata_size<D>(context: &mut Context<D>) -> anyhow::Result<()>
+    pub fn load_calldata_size(context: &mut Context) -> anyhow::Result<()> {
-    where
-        D: Dependency + Clone,
-    {
         let call_data_size_pointer = context
             .get_global(crate::polkavm::GLOBAL_CALLDATA_SIZE)?
             .value
             .as_pointer_value();
         let call_data_size_value = context
-            .build_runtime_call(revive_runtime_api::polkavm_imports::CALL_DATA_SIZE, &[])
+            .build_runtime_call(CALL_DATA_SIZE, &[])
             .expect("the call_data_size syscall method should return a value")
             .into_int_value();
         let call_data_size_value = context.builder().build_int_truncate(
@@ -86,10 +88,7 @@ impl Entry {
 
     /// Calls the deploy code if the first function argument was `1`.
     /// Calls the runtime code otherwise.
-    pub fn leave_entry<D>(context: &mut Context<D>) -> anyhow::Result<()>
+    pub fn leave_entry(context: &mut Context) -> anyhow::Result<()> {
-    where
-        D: Dependency + Clone,
-    {
         context.set_debug_location(0, 0, None)?;
 
         let is_deploy = context
@@ -129,11 +128,8 @@ impl Entry {
     }
 }
 
-impl<D> WriteLLVM<D> for Entry
+impl WriteLLVM for Entry {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
         let entry_arguments = vec![context.bool_type().as_basic_type_enum()];
         let entry_function_type = context.function_type(entry_arguments, 0);
         context.add_function(
@@ -141,16 +137,18 @@ where
             entry_function_type,
             0,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )?;
 
         context.declare_global(
-            revive_runtime_api::immutable_data::GLOBAL_IMMUTABLE_DATA_POINTER,
+            GLOBAL_IMMUTABLE_DATA_POINTER,
             context.word_type().array_type(0),
             AddressSpace::Stack,
         );
 
         context.declare_global(
-            revive_runtime_api::immutable_data::GLOBAL_IMMUTABLE_DATA_SIZE,
+            GLOBAL_IMMUTABLE_DATA_SIZE,
             context.xlen_type(),
             AddressSpace::Stack,
         );
@@ -161,9 +159,9 @@ where
     /// Instead of a single entrypoint, the runtime expects two exports: `call ` and `deploy`.
     /// `call` and `deploy` directly call `entry`, signaling a deploy if the first arg is `1`.
     /// The `entry` function loads calldata, sets globals and calls the runtime or deploy code.
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
         let entry = context
-            .get_function(runtime::FUNCTION_ENTRY)
+            .get_function(runtime::FUNCTION_ENTRY, false)
             .expect("the entry function should already be declared")
             .borrow()
             .declaration;
@@ -174,7 +172,7 @@ where
             true,
         );
 
-        context.set_current_function(runtime::FUNCTION_ENTRY, None)?;
+        context.set_current_function(runtime::FUNCTION_ENTRY, None, false)?;
         context.set_basic_block(context.current_function().borrow().entry_block());
 
         Self::initialize_globals(context)?;

crates/llvm-context/src/polkavm/context/function/runtime/revive.rs

@@ -5,7 +5,6 @@ use inkwell::values::BasicValue;
 use crate::polkavm::context::function::Attribute;
 use crate::polkavm::context::runtime::RuntimeFunction;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// Pointers are represented as opaque 256 bit integer values in EVM.
@@ -15,10 +14,7 @@ use crate::polkavm::WriteLLVM;
 /// (but wrong) pointers when truncated.
 pub struct WordToPointer;
 
-impl<D> RuntimeFunction<D> for WordToPointer
+impl RuntimeFunction for WordToPointer {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_int_truncate";
 
     const ATTRIBUTES: &'static [Attribute] = &[
@@ -27,7 +23,7 @@ where
         Attribute::AlwaysInline,
     ];
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context
             .xlen_type()
             .fn_type(&[context.word_type().into()], false)
@@ -35,7 +31,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let value = Self::paramater(context, 0).into_int_value();
         let truncated =
@@ -67,26 +63,20 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for WordToPointer
+impl WriteLLVM for WordToPointer {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// The revive runtime exit function.
 pub struct Exit;
 
-impl<D> RuntimeFunction<D> for Exit
+impl RuntimeFunction for Exit {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_exit";
 
     const ATTRIBUTES: &'static [Attribute] = &[
@@ -95,7 +85,7 @@ where
         Attribute::AlwaysInline,
     ];
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.void_type().fn_type(
             &[
                 context.xlen_type().into(),
@@ -108,7 +98,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let flags = Self::paramater(context, 0).into_int_value();
         let offset = Self::paramater(context, 1).into_int_value();
@@ -133,15 +123,12 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for Exit
+impl WriteLLVM for Exit {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }

crates/llvm-context/src/polkavm/context/function/runtime/runtime_code.rs

@@ -1,63 +1,52 @@
 //! The runtime code function.
 
-use std::marker::PhantomData;
-
-use crate::polkavm::context::code_type::CodeType;
 use crate::polkavm::context::function::runtime;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// The runtime code function.
 /// Is a special function that is only used by the front-end generated code.
 #[derive(Debug)]
-pub struct RuntimeCode<B, D>
+pub struct RuntimeCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
     /// The runtime code AST representation.
     inner: B,
-    /// The `D` phantom data.
-    _pd: PhantomData<D>,
 }
 
-impl<B, D> RuntimeCode<B, D>
+impl<B> RuntimeCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
     /// A shortcut constructor.
     pub fn new(inner: B) -> Self {
-        Self {
+        Self { inner }
-            inner,
-            _pd: PhantomData,
-        }
     }
 }
 
-impl<B, D> WriteLLVM<D> for RuntimeCode<B, D>
+impl<B> WriteLLVM for RuntimeCode<B>
 where
-    B: WriteLLVM<D>,
+    B: WriteLLVM,
-    D: Dependency + Clone,
 {
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
         let function_type = context.function_type::<inkwell::types::BasicTypeEnum>(vec![], 0);
         context.add_function(
             runtime::FUNCTION_RUNTIME_CODE,
             function_type,
             0,
-            Some(inkwell::module::Linkage::External),
+            Some(inkwell::module::Linkage::Private),
+            None,
+            false,
         )?;
 
         self.inner.declare(context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        context.set_current_function(runtime::FUNCTION_RUNTIME_CODE, None)?;
+        context.set_current_function(runtime::FUNCTION_RUNTIME_CODE, None, false)?;
 
         context.set_basic_block(context.current_function().borrow().entry_block());
-        context.set_code_type(CodeType::Runtime);
 
         self.inner.into_llvm(context)?;
         context.set_debug_location(0, 0, None)?;

crates/llvm-context/src/polkavm/context/function/runtime/sbrk.rs

@@ -1,11 +1,11 @@
 //! Emulates the linear EVM heap memory via a simulated `sbrk` system call.
 
 use inkwell::values::BasicValue;
+use revive_common::BYTE_LENGTH_WORD;
 
 use crate::polkavm::context::attribute::Attribute;
 use crate::polkavm::context::runtime::RuntimeFunction;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// Simulates the `sbrk` system call, reproducing the semantics of the EVM heap memory.
@@ -24,10 +24,7 @@ use crate::polkavm::WriteLLVM;
 /// - Maintains the total memory size (`msize`) in global heap size value.
 pub struct Sbrk;
 
-impl<D> RuntimeFunction<D> for Sbrk
+impl RuntimeFunction for Sbrk {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__sbrk_internal";
 
     const ATTRIBUTES: &'static [Attribute] = &[
@@ -36,7 +33,7 @@ where
         Attribute::WillReturn,
     ];
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.llvm().ptr_type(Default::default()).fn_type(
             &[context.xlen_type().into(), context.xlen_type().into()],
             false,
@@ -45,11 +42,22 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>> {
         let offset = Self::paramater(context, 0).into_int_value();
         let size = Self::paramater(context, 1).into_int_value();
 
+        let return_block = context.append_basic_block("return_pointer");
+        let body_block = context.append_basic_block("body");
+        let is_size_zero = context.builder().build_int_compare(
+            inkwell::IntPredicate::EQ,
+            size,
+            context.xlen_type().const_zero(),
+            "is_size_zero",
+        )?;
+        context.build_conditional_branch(is_size_zero, return_block, body_block)?;
+
+        context.set_basic_block(body_block);
         let trap_block = context.append_basic_block("trap");
         let offset_in_bounds_block = context.append_basic_block("offset_in_bounds");
         let is_offset_out_of_bounds = context.builder().build_int_compare(
@@ -71,7 +79,7 @@ where
         context.set_basic_block(offset_in_bounds_block);
         let mask = context
             .xlen_type()
-            .const_int(revive_common::BYTE_LENGTH_WORD as u64 - 1, false);
+            .const_int(BYTE_LENGTH_WORD as u64 - 1, false);
         let total_size = context
             .builder()
             .build_int_add(offset, size, "total_size")?;
@@ -94,7 +102,6 @@ where
         )?;
 
         context.set_basic_block(size_in_bounds_block);
-        let return_block = context.append_basic_block("return_pointer");
         let new_size_block = context.append_basic_block("new_size");
         let is_new_size = context.builder().build_int_compare(
             inkwell::IntPredicate::UGT,
@@ -130,15 +137,12 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for Sbrk
+impl WriteLLVM for Sbrk {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }

crates/llvm-context/src/polkavm/context/global.rs

@@ -5,7 +5,6 @@ use inkwell::values::BasicValue;
 
 use crate::polkavm::context::address_space::AddressSpace;
 use crate::polkavm::context::Context;
-use crate::PolkaVMDependency;
 
 /// The LLVM global value.
 #[derive(Debug, Clone, Copy)]
@@ -18,15 +17,14 @@ pub struct Global<'ctx> {
 
 impl<'ctx> Global<'ctx> {
     /// A shortcut constructor.
-    pub fn new<D, T, V>(
+    pub fn new<T, V>(
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
         r#type: T,
         address_space: AddressSpace,
         initializer: V,
         name: &str,
     ) -> Self
     where
-        D: PolkaVMDependency + Clone,
         T: BasicType<'ctx>,
         V: BasicValue<'ctx>,
     {
@@ -53,14 +51,13 @@ impl<'ctx> Global<'ctx> {
     }
 
     /// Construct an external global.
-    pub fn declare<D, T>(
+    pub fn declare<T>(
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
         r#type: T,
         address_space: AddressSpace,
         name: &str,
     ) -> Self
     where
-        D: PolkaVMDependency + Clone,
         T: BasicType<'ctx>,
     {
         let r#type = r#type.as_basic_type_enum();

crates/llvm-context/src/polkavm/context/mod.rs

@@ -1,22 +1,5 @@
 //! The LLVM IR generator context.
 
-pub mod address_space;
-pub mod argument;
-pub mod attribute;
-pub mod build;
-pub mod code_type;
-pub mod debug_info;
-pub mod function;
-pub mod global;
-pub mod r#loop;
-pub mod pointer;
-pub mod runtime;
-pub mod solidity_data;
-pub mod yul_data;
-
-#[cfg(test)]
-mod tests;
-
 use std::cell::RefCell;
 use std::collections::HashMap;
 use std::rc::Rc;
@@ -25,12 +8,14 @@ use inkwell::debug_info::AsDIScope;
 use inkwell::debug_info::DIScope;
 use inkwell::types::BasicType;
 use inkwell::values::BasicValue;
+use inkwell::values::InstructionOpcode;
+use revive_solc_json_interface::PolkaVMDefaultHeapMemorySize;
+use revive_solc_json_interface::PolkaVMDefaultStackMemorySize;
+use revive_solc_json_interface::SolcStandardJsonInputSettingsPolkaVMMemory;
 
-use crate::memory::MemoryConfig;
 use crate::optimizer::settings::Settings as OptimizerSettings;
 use crate::optimizer::Optimizer;
 use crate::polkavm::DebugConfig;
-use crate::polkavm::Dependency;
 use crate::target_machine::target::Target;
 use crate::target_machine::TargetMachine;
 use crate::PolkaVMLoadHeapWordFunction;
@@ -56,13 +41,27 @@ use self::runtime::RuntimeFunction;
 use self::solidity_data::SolidityData;
 use self::yul_data::YulData;
 
+pub mod address_space;
+pub mod argument;
+pub mod attribute;
+pub mod build;
+pub mod code_type;
+pub mod debug_info;
+pub mod function;
+pub mod global;
+pub mod r#loop;
+pub mod pointer;
+pub mod runtime;
+pub mod solidity_data;
+pub mod yul_data;
+
+#[cfg(test)]
+mod tests;
+
 /// The LLVM IR generator context.
 /// It is a not-so-big god-like object glueing all the compilers' complexity and act as an adapter
 /// and a superstructure over the inner `inkwell` LLVM context.
-pub struct Context<'ctx, D>
+pub struct Context<'ctx> {
-where
-    D: Dependency + Clone,
-{
     /// The inner LLVM context.
     llvm: &'ctx inkwell::context::Context,
     /// The inner LLVM context builder.
@@ -85,17 +84,9 @@ where
     current_function: Option<Rc<RefCell<Function<'ctx>>>>,
     /// The loop context stack.
     loop_stack: Vec<Loop<'ctx>>,
-    /// The extra LLVM arguments that were used during target initialization.
-    llvm_arguments: &'ctx [String],
     /// The PVM memory configuration.
-    memory_config: MemoryConfig,
+    memory_config: SolcStandardJsonInputSettingsPolkaVMMemory,
 
-    /// The project dependency manager. It can be any entity implementing the trait.
-    /// The manager is used to get information about contracts and their dependencies during
-    /// the multi-threaded compilation process.
-    dependency_manager: Option<D>,
-    /// Whether to append the metadata hash at the end of bytecode.
-    include_metadata_hash: bool,
     /// The debug info of the current module.
     debug_info: Option<DebugInfo<'ctx>>,
     /// The debug configuration telling whether to dump the needed IRs.
@@ -107,10 +98,7 @@ where
     yul_data: Option<YulData>,
 }
 
-impl<'ctx, D> Context<'ctx, D>
+impl<'ctx> Context<'ctx> {
-where
-    D: Dependency + Clone,
-{
     /// The functions hashmap default capacity.
     const FUNCTIONS_HASHMAP_INITIAL_CAPACITY: usize = 64;
 
@@ -219,27 +207,29 @@ where
     }
 
     /// Initializes a new LLVM context.
-    #[allow(clippy::too_many_arguments)]
     pub fn new(
         llvm: &'ctx inkwell::context::Context,
         module: inkwell::module::Module<'ctx>,
         optimizer: Optimizer,
-        dependency_manager: Option<D>,
-        include_metadata_hash: bool,
         debug_config: DebugConfig,
-        llvm_arguments: &'ctx [String],
+        memory_config: SolcStandardJsonInputSettingsPolkaVMMemory,
-        memory_config: MemoryConfig,
     ) -> Self {
         Self::set_data_layout(llvm, &module);
         Self::link_stdlib_module(llvm, &module);
         Self::link_polkavm_imports(llvm, &module);
-        Self::set_polkavm_stack_size(llvm, &module, memory_config.stack_size);
+        Self::set_polkavm_stack_size(
+            llvm,
+            &module,
+            memory_config
+                .stack_size
+                .unwrap_or(PolkaVMDefaultStackMemorySize),
+        );
         Self::set_module_flags(llvm, &module);
 
         let intrinsics = Intrinsics::new(llvm, &module);
         let llvm_runtime = LLVMRuntime::new(llvm, &module, &optimizer);
         let debug_info = debug_config.emit_debug_info.then(|| {
-            let debug_info = DebugInfo::new(&module);
+            let debug_info = DebugInfo::new(&module, &debug_config);
             debug_info.initialize_module(llvm, &module);
             debug_info
         });
@@ -256,12 +246,8 @@ where
             functions: HashMap::with_capacity(Self::FUNCTIONS_HASHMAP_INITIAL_CAPACITY),
             current_function: None,
             loop_stack: Vec::with_capacity(Self::LOOP_STACK_INITIAL_CAPACITY),
-            llvm_arguments,
             memory_config,
 
-            dependency_manager,
-            include_metadata_hash,
-
             debug_info,
             debug_config,
 
@@ -272,12 +258,10 @@ where
 
     /// Builds the LLVM IR module, returning the build artifacts.
     pub fn build(
-        mut self,
+        self,
         contract_path: &str,
-        metadata_hash: Option<[u8; revive_common::BYTE_LENGTH_WORD]>,
+        metadata_hash: Option<revive_common::Keccak256>,
     ) -> anyhow::Result<Build> {
-        let module_clone = self.module.clone();
-
         self.link_polkavm_exports(contract_path)?;
         self.link_immutable_data(contract_path)?;
 
@@ -305,6 +289,17 @@ where
                 )
             })?;
 
+        // Remove MinSize on functions that perform large integer div/rem to
+        // avoid compiler crash that happens when large integer div/rem by
+        // power-of-2 are not being expanded by ExpandLargeIntDivRem pass as
+        // it expects peephole from DAGCombine, which doesn't happen due to the
+        // MinSize attribute being set on the function.
+        // NOTE: As soon as it strips attribute from a function where large
+        // integer div/rem is used, it's crucial to call it after inlining.
+        // TODO: Remove this once LLVM fix is backported to LLVM 21 and we
+        // switch to corresponding inkwell version.
+        self.strip_minsize_for_divrem();
+
         self.debug_config
             .dump_llvm_ir_optimized(contract_path, self.module())?;
 
@@ -326,33 +321,16 @@ where
                 )
             })?;
 
-        let shared_object = revive_linker::link(buffer.as_slice())?;
+        let object = buffer.as_slice().to_vec();
 
-        self.debug_config
+        self.debug_config.dump_object(contract_path, &object)?;
-            .dump_object(contract_path, &shared_object)?;
 
-        let polkavm_bytecode =
+        crate::polkavm::build(
-            revive_linker::polkavm_linker(shared_object, !self.debug_config().emit_debug_info)?;
+            &object,
-
+            metadata_hash
-        let build = match crate::polkavm::build_assembly_text(
+                .as_ref()
-            contract_path,
+                .map(|hash| hash.as_bytes().try_into().unwrap()),
-            &polkavm_bytecode,
+        )
-            metadata_hash,
-            self.debug_config(),
-        ) {
-            Ok(build) => build,
-            Err(_error)
-                if self.optimizer.settings() != &OptimizerSettings::size()
-                    && self.optimizer.settings().is_fallback_to_size_enabled() =>
-            {
-                self.optimizer = Optimizer::new(OptimizerSettings::size());
-                self.module = module_clone;
-                self.build(contract_path, metadata_hash)?
-            }
-            Err(error) => Err(error)?,
-        };
-
-        Ok(build)
     }
 
     /// Verifies the current LLVM IR module.
@@ -429,11 +407,15 @@ where
         }
     }
 
-    /// Declare an external global.
+    /// Declare an external global. This is an idempotent method.
     pub fn declare_global<T>(&mut self, name: &str, r#type: T, address_space: AddressSpace)
     where
         T: BasicType<'ctx> + Clone + Copy,
     {
+        if self.globals.contains_key(name) {
+            return;
+        }
+
         let global = Global::declare(self, r#type, address_space, name);
         self.globals.insert(name.to_owned(), global);
     }
@@ -455,8 +437,16 @@ where
         r#type: inkwell::types::FunctionType<'ctx>,
         return_values_length: usize,
         linkage: Option<inkwell::module::Linkage>,
+        location: Option<(u32, u32)>,
+        is_frontend: bool,
     ) -> anyhow::Result<Rc<RefCell<Function<'ctx>>>> {
-        let value = self.module().add_function(name, r#type, linkage);
+        assert!(
+            self.get_function(name, is_frontend).is_none(),
+            "ICE: function '{name}' declared subsequentally"
+        );
+
+        let name = self.internal_function_name(name, is_frontend);
+        let value = self.module().add_function(&name, r#type, linkage);
 
         if self.debug_info().is_some() {
             self.builder().unset_current_debug_location();
@@ -470,7 +460,8 @@ where
                 Some(scp) => scp,
             };
             self.push_debug_scope(func_scope.as_debug_info_scope());
-            self.set_debug_location(0, 0, Some(func_scope.as_debug_info_scope()))?;
+            let (line, column) = location.unwrap_or((0, 0));
+            self.set_debug_location(line, column, Some(func_scope.as_debug_info_scope()))?;
         }
 
         let entry_block = self.llvm.append_basic_block(value, "entry");
@@ -496,7 +487,7 @@ where
         };
 
         let function = Function::new(
-            name.to_owned(),
+            name.clone(),
             FunctionDeclaration::new(r#type, value),
             r#return,
             entry_block,
@@ -504,7 +495,7 @@ where
         );
         Function::set_default_attributes(self.llvm, function.declaration(), &self.optimizer);
         let function = Rc::new(RefCell::new(function));
-        self.functions.insert(name.to_string(), function.clone());
+        self.functions.insert(name, function.clone());
 
         self.pop_debug_scope();
 
@@ -512,8 +503,14 @@ where
     }
 
     /// Returns a shared reference to the specified function.
-    pub fn get_function(&self, name: &str) -> Option<Rc<RefCell<Function<'ctx>>>> {
+    pub fn get_function(
-        self.functions.get(name).cloned()
+        &self,
+        name: &str,
+        is_frontend: bool,
+    ) -> Option<Rc<RefCell<Function<'ctx>>>> {
+        self.functions
+            .get(&self.internal_function_name(name, is_frontend))
+            .cloned()
     }
 
     /// Returns a shared reference to the current active function.
@@ -525,8 +522,13 @@ where
 
     /// Sets the current active function. If debug-info generation is enabled,
     /// constructs a debug-scope and pushes in on the scope-stack.
-    pub fn set_current_function(&mut self, name: &str, line: Option<u32>) -> anyhow::Result<()> {
+    pub fn set_current_function(
-        let function = self.functions.get(name).cloned().ok_or_else(|| {
+        &mut self,
+        name: &str,
+        location: Option<(u32, u32)>,
+        frontend: bool,
+    ) -> anyhow::Result<()> {
+        let function = self.get_function(name, frontend).ok_or_else(|| {
             anyhow::anyhow!("Failed to activate an undeclared function `{}`", name)
         })?;
         self.current_function = Some(function);
@@ -534,7 +536,8 @@ where
         if let Some(scope) = self.current_function().borrow().get_debug_scope() {
             self.push_debug_scope(scope);
         }
-        self.set_debug_location(line.unwrap_or_default(), 0, None)?;
+        let (line, column) = location.unwrap_or_default();
+        self.set_debug_location(line, column, None)?;
 
         Ok(())
     }
@@ -635,54 +638,6 @@ where
             .expect("The current context is not in a loop")
     }
 
-    /// Compiles a contract dependency, if the dependency manager is set.
-    pub fn compile_dependency(&mut self, name: &str) -> anyhow::Result<String> {
-        self.dependency_manager
-            .to_owned()
-            .ok_or_else(|| anyhow::anyhow!("The dependency manager is unset"))
-            .and_then(|manager| {
-                Dependency::compile(
-                    manager,
-                    name,
-                    self.optimizer.settings().to_owned(),
-                    self.include_metadata_hash,
-                    self.debug_config.clone(),
-                    self.llvm_arguments,
-                    self.memory_config,
-                )
-            })
-    }
-
-    /// Gets a full contract_path from the dependency manager.
-    pub fn resolve_path(&self, identifier: &str) -> anyhow::Result<String> {
-        self.dependency_manager
-            .to_owned()
-            .ok_or_else(|| anyhow::anyhow!("The dependency manager is unset"))
-            .and_then(|manager| {
-                let full_path = manager.resolve_path(identifier)?;
-                Ok(full_path)
-            })
-    }
-
-    /// Gets a deployed library address from the dependency manager.
-    pub fn resolve_library(&self, path: &str) -> anyhow::Result<inkwell::values::IntValue<'ctx>> {
-        self.dependency_manager
-            .to_owned()
-            .ok_or_else(|| anyhow::anyhow!("The dependency manager is unset"))
-            .and_then(|manager| {
-                let address = manager.resolve_library(path)?;
-                let address = self.word_const_str_hex(address.as_str());
-                Ok(address)
-            })
-    }
-
-    /// Extracts the dependency manager.
-    pub fn take_dependency_manager(&mut self) -> D {
-        self.dependency_manager
-            .take()
-            .expect("The dependency manager is unset")
-    }
-
     /// Returns the debug info.
     pub fn debug_info(&self) -> Option<&DebugInfo<'ctx>> {
         self.debug_info.as_ref()
@@ -715,21 +670,27 @@ where
         r#type: T,
         name: &str,
     ) -> Pointer<'ctx> {
-        let current_block = self.basic_block();
+        // TODO: Revisit. While at entry should be preferred in theory:
-        let entry_block = self.current_function().borrow().entry_block();
+        // - It has negligible code size impact on real word contracts.
+        // - Sometimes has negative impact on code size.
+        // - Messes up debug information used to analyze code size issues.
+        self.build_alloca(r#type, name)
 
-        match entry_block.get_first_instruction() {
+        // let current_block = self.basic_block();
-            Some(instruction) => self.builder().position_before(&instruction),
+        // let entry_block = self.current_function().borrow().entry_block();
-            None => self.builder().position_at_end(entry_block),
-        }
 
-        let pointer = self.build_alloca(r#type, name);
+        // match entry_block.get_first_instruction() {
-        self.set_basic_block(current_block);
+        //     Some(instruction) => self.builder().position_before(&instruction),
-        pointer
+        //     None => self.builder().position_at_end(entry_block),
+        // }
+
+        // let pointer = self.build_alloca(r#type, name);
+        // self.set_basic_block(current_block);
+        // pointer
     }
 
     /// Builds an aligned stack allocation at the current position.
-    /// Use this if [`build_alloca_at_entry`] might change program semantics.
+    /// Use this if [`Self::build_alloca_at_entry`] might change program semantics.
     /// Otherwise, alloca should always be built at the function prelude!
     pub fn build_alloca<T: BasicType<'ctx> + Clone + Copy>(
         &self,
@@ -787,9 +748,9 @@ where
     ) -> anyhow::Result<inkwell::values::BasicValueEnum<'ctx>> {
         match pointer.address_space {
             AddressSpace::Heap => {
-                let name = <PolkaVMLoadHeapWordFunction as RuntimeFunction<D>>::NAME;
+                let name = <PolkaVMLoadHeapWordFunction as RuntimeFunction>::NAME;
                 let declaration =
-                    <PolkaVMLoadHeapWordFunction as RuntimeFunction<D>>::declaration(self);
+                    <PolkaVMLoadHeapWordFunction as RuntimeFunction>::declaration(self);
                 let arguments = [self
                     .builder()
                     .build_ptr_to_int(pointer.value, self.xlen_type(), "offset_ptrtoint")?
@@ -825,7 +786,7 @@ where
         match pointer.address_space {
             AddressSpace::Heap => {
                 let declaration =
-                    <PolkaVMStoreHeapWordFunction as RuntimeFunction<D>>::declaration(self);
+                    <PolkaVMStoreHeapWordFunction as RuntimeFunction>::declaration(self);
                 let arguments = [
                     pointer.to_int(self).as_basic_value_enum(),
                     value.as_basic_value_enum(),
@@ -945,10 +906,7 @@ where
     pub fn build_runtime_call_to_getter(
         &self,
         import: &'static str,
-    ) -> anyhow::Result<inkwell::values::BasicValueEnum<'ctx>>
+    ) -> anyhow::Result<inkwell::values::BasicValueEnum<'ctx>> {
-    where
-        D: Dependency + Clone,
-    {
         let pointer = self.build_alloca_at_entry(self.word_type(), &format!("{import}_output"));
         self.build_runtime_call(import, &[pointer.to_int(self).into()]);
         self.build_load(pointer, import)
@@ -1043,7 +1001,7 @@ where
         length: inkwell::values::IntValue<'ctx>,
     ) -> anyhow::Result<()> {
         self.build_call(
-            <Exit as RuntimeFunction<D>>::declaration(self),
+            <Exit as RuntimeFunction>::declaration(self),
             &[flags.into(), offset.into(), length.into()],
             "exit",
         );
@@ -1067,14 +1025,14 @@ where
 
         Ok(self
             .build_call(
-                <WordToPointer as RuntimeFunction<D>>::declaration(self),
+                <WordToPointer as RuntimeFunction>::declaration(self),
                 &[value.into()],
                 "word_to_pointer",
             )
             .unwrap_or_else(|| {
                 panic!(
                     "revive runtime function {} should return a value",
-                    <WordToPointer as RuntimeFunction<D>>::NAME,
+                    <WordToPointer as RuntimeFunction>::NAME,
                 )
             })
             .into_int_value())
@@ -1090,7 +1048,7 @@ where
         size: inkwell::values::IntValue<'ctx>,
     ) -> anyhow::Result<inkwell::values::PointerValue<'ctx>> {
         let call_site_value = self.builder().build_call(
-            <PolkaVMSbrkFunction as RuntimeFunction<D>>::declaration(self).function_value(),
+            <PolkaVMSbrkFunction as RuntimeFunction>::declaration(self).function_value(),
             &[offset.into(), size.into()],
             "alloc_start",
         )?;
@@ -1112,7 +1070,7 @@ where
             .unwrap_or_else(|| {
                 panic!(
                     "revive runtime function {} should return a value",
-                    <PolkaVMSbrkFunction as RuntimeFunction<D>>::NAME,
+                    <PolkaVMSbrkFunction as RuntimeFunction>::NAME,
                 )
             })
             .into_pointer_value())
@@ -1412,19 +1370,8 @@ where
     /// Returns the Yul data reference.
     /// # Panics
     /// If the Yul data has not been initialized.
-    pub fn yul(&self) -> &YulData {
+    pub fn yul(&self) -> Option<&YulData> {
-        self.yul_data
+        self.yul_data.as_ref()
-            .as_ref()
-            .expect("The Yul data must have been initialized")
-    }
-
-    /// Returns the Yul data mutable reference.
-    /// # Panics
-    /// If the Yul data has not been initialized.
-    pub fn yul_mut(&mut self) -> &mut YulData {
-        self.yul_data
-            .as_mut()
-            .expect("The Yul data must have been initialized")
     }
 
     /// Returns the current number of immutables values in the contract.
@@ -1443,7 +1390,54 @@ where
     }
 
     pub fn heap_size(&self) -> inkwell::values::IntValue<'ctx> {
-        self.xlen_type()
+        self.xlen_type().const_int(
-            .const_int(self.memory_config.heap_size as u64, false)
+            self.memory_config
+                .heap_size
+                .unwrap_or(PolkaVMDefaultHeapMemorySize) as u64,
+            false,
+        )
+    }
+
+    /// Returns the internal function name.
+    fn internal_function_name(&self, name: &str, is_frontend: bool) -> String {
+        if is_frontend {
+            format!("{name}_{}", self.code_type().unwrap())
+        } else {
+            name.to_string()
+        }
+    }
+
+    /// Scans all functions in the module and removes the `MinSize` attribute
+    /// if the function contains any large sdiv, udiv, srem, urem instructions with either unknown
+    /// NOTE: The check here could be relaxed by checking denominator: if the denominator is
+    /// unknown or is a power-of-2 constant, then need to strip the `minsize` attribute; otherwise
+    /// instruction can be ignored as backend will expand it correctly.
+    fn strip_minsize_for_divrem(&self) {
+        self.module().get_functions().for_each(|func| {
+            let has_divrem = func.get_basic_block_iter().any(|b| {
+                b.get_instructions().any(|inst| match inst.get_opcode() {
+                    InstructionOpcode::SDiv
+                    | InstructionOpcode::UDiv
+                    | InstructionOpcode::SRem
+                    | InstructionOpcode::URem => {
+                        inst.get_type().into_int_type().get_bit_width() >= 256
+                    }
+                    _ => false,
+                })
+            });
+            if has_divrem
+                && func
+                    .get_enum_attribute(
+                        inkwell::attributes::AttributeLoc::Function,
+                        Attribute::MinSize as u32,
+                    )
+                    .is_some()
+            {
+                func.remove_enum_attribute(
+                    inkwell::attributes::AttributeLoc::Function,
+                    Attribute::MinSize as u32,
+                );
+            }
+        });
     }
 }

crates/llvm-context/src/polkavm/context/pointer/heap.rs

@@ -2,21 +2,20 @@
 
 use inkwell::values::BasicValueEnum;
 
+use revive_common::BYTE_LENGTH_BYTE;
+use revive_common::BYTE_LENGTH_WORD;
+
 use crate::polkavm::context::runtime::RuntimeFunction;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// Load a word size value from a heap pointer.
 pub struct LoadWord;
 
-impl<D> RuntimeFunction<D> for LoadWord
+impl RuntimeFunction for LoadWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_load_heap_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context
             .word_type()
             .fn_type(&[context.xlen_type().into()], false)
@@ -24,12 +23,12 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         let offset = Self::paramater(context, 0).into_int_value();
         let length = context
             .xlen_type()
-            .const_int(revive_common::BYTE_LENGTH_WORD as u64, false);
+            .const_int(BYTE_LENGTH_WORD as u64, false);
         let pointer = context.build_heap_gep(offset, length)?;
         let value = context
             .builder()
@@ -38,7 +37,7 @@ where
             .basic_block()
             .get_last_instruction()
             .expect("Always exists")
-            .set_alignment(revive_common::BYTE_LENGTH_BYTE as u32)
+            .set_alignment(BYTE_LENGTH_BYTE as u32)
             .expect("Alignment is valid");
 
         let swapped_value = context.build_byte_swap(value)?;
@@ -46,29 +45,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for LoadWord
+impl WriteLLVM for LoadWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Store a word size value through a heap pointer.
 pub struct StoreWord;
 
-impl<D> RuntimeFunction<D> for StoreWord
+impl RuntimeFunction for StoreWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_store_heap_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.void_type().fn_type(
             &[context.xlen_type().into(), context.word_type().into()],
             false,
@@ -77,12 +70,12 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         let offset = Self::paramater(context, 0).into_int_value();
         let length = context
             .xlen_type()
-            .const_int(revive_common::BYTE_LENGTH_WORD as u64, false);
+            .const_int(BYTE_LENGTH_WORD as u64, false);
         let pointer = context.build_heap_gep(offset, length)?;
 
         let value = context.build_byte_swap(Self::paramater(context, 1))?;
@@ -90,21 +83,18 @@ where
         context
             .builder()
             .build_store(pointer.value, value)?
-            .set_alignment(revive_common::BYTE_LENGTH_BYTE as u32)
+            .set_alignment(BYTE_LENGTH_BYTE as u32)
             .expect("Alignment is valid");
         Ok(None)
     }
 }
 
-impl<D> WriteLLVM<D> for StoreWord
+impl WriteLLVM for StoreWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }

crates/llvm-context/src/polkavm/context/pointer/mod.rs

@@ -5,7 +5,6 @@ use inkwell::types::BasicType;
 use crate::polkavm::context::address_space::AddressSpace;
 use crate::polkavm::context::global::Global;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 
 pub mod heap;
 pub mod storage;
@@ -39,13 +38,10 @@ impl<'ctx> Pointer<'ctx> {
     }
 
     /// Wraps a 256-bit primitive type pointer.
-    pub fn new_stack_field<D>(
+    pub fn new_stack_field(
-        context: &Context<'ctx, D>,
+        context: &Context<'ctx>,
         value: inkwell::values::PointerValue<'ctx>,
-    ) -> Self
+    ) -> Self {
-    where
-        D: Dependency + Clone,
-    {
         Self {
             r#type: context.word_type().as_basic_type_enum(),
             address_space: AddressSpace::Stack,
@@ -54,15 +50,14 @@ impl<'ctx> Pointer<'ctx> {
     }
 
     /// Creates a new pointer with the specified `offset`.
-    pub fn new_with_offset<D, T>(
+    pub fn new_with_offset<T>(
-        context: &Context<'ctx, D>,
+        context: &Context<'ctx>,
         address_space: AddressSpace,
         r#type: T,
         offset: inkwell::values::IntValue<'ctx>,
         name: &str,
     ) -> Self
     where
-        D: Dependency + Clone,
         T: BasicType<'ctx>,
     {
         assert_ne!(
@@ -92,25 +87,19 @@ impl<'ctx> Pointer<'ctx> {
     }
 
     /// Cast this pointer to a register sized integer value.
-    pub fn to_int<D>(&self, context: &Context<'ctx, D>) -> inkwell::values::IntValue<'ctx>
+    pub fn to_int(&self, context: &Context<'ctx>) -> inkwell::values::IntValue<'ctx> {
-    where
-        D: Dependency + Clone,
-    {
         context
             .builder()
             .build_ptr_to_int(self.value, context.xlen_type(), "ptr_to_xlen")
             .expect("we should be positioned")
     }
 
-    pub fn address_space_cast<D>(
+    pub fn address_space_cast(
         self,
-        context: &Context<'ctx, D>,
+        context: &Context<'ctx>,
         address_space: AddressSpace,
         name: &str,
-    ) -> anyhow::Result<Self>
+    ) -> anyhow::Result<Self> {
-    where
-        D: Dependency + Clone,
-    {
         let value = context.builder().build_address_space_cast(
             self.value,
             context.llvm().ptr_type(address_space.into()),

crates/llvm-context/src/polkavm/context/pointer/storage.rs

@@ -4,19 +4,15 @@ use inkwell::values::BasicValueEnum;
 
 use crate::polkavm::context::runtime::RuntimeFunction;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 use crate::polkavm::WriteLLVM;
 
 /// Load a word size value from a storage pointer.
 pub struct LoadWord;
 
-impl<D> RuntimeFunction<D> for LoadWord
+impl RuntimeFunction for LoadWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_load_storage_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context
             .word_type()
             .fn_type(&[context.llvm().ptr_type(Default::default()).into()], false)
@@ -24,7 +20,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         Ok(Some(emit_load(
             context,
@@ -34,29 +30,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for LoadWord
+impl WriteLLVM for LoadWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Load a word size value from a transient storage pointer.
 pub struct LoadTransientWord;
 
-impl<D> RuntimeFunction<D> for LoadTransientWord
+impl RuntimeFunction for LoadTransientWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_load_transient_storage_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context
             .word_type()
             .fn_type(&[context.llvm().ptr_type(Default::default()).into()], false)
@@ -64,35 +54,29 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         Ok(Some(emit_load(context, Self::paramater(context, 0), true)?))
     }
 }
 
-impl<D> WriteLLVM<D> for LoadTransientWord
+impl WriteLLVM for LoadTransientWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Store a word size value through a storage pointer.
 pub struct StoreWord;
 
-impl<D> RuntimeFunction<D> for StoreWord
+impl RuntimeFunction for StoreWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_store_storage_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.void_type().fn_type(
             &[
                 context.llvm().ptr_type(Default::default()).into(),
@@ -104,7 +88,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         emit_store(
             context,
@@ -117,29 +101,23 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for StoreWord
+impl WriteLLVM for StoreWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
 /// Store a word size value through a transient storage pointer.
 pub struct StoreTransientWord;
 
-impl<D> RuntimeFunction<D> for StoreTransientWord
+impl RuntimeFunction for StoreTransientWord {
-where
-    D: Dependency + Clone,
-{
     const NAME: &'static str = "__revive_store_transient_storage_word";
 
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx> {
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx> {
         context.void_type().fn_type(
             &[
                 context.llvm().ptr_type(Default::default()).into(),
@@ -151,7 +129,7 @@ where
 
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<BasicValueEnum<'ctx>>> {
         emit_store(
             context,
@@ -164,24 +142,25 @@ where
     }
 }
 
-impl<D> WriteLLVM<D> for StoreTransientWord
+impl WriteLLVM for StoreTransientWord {
-where
+    fn declare(&mut self, context: &mut Context) -> anyhow::Result<()> {
-    D: Dependency + Clone,
+        <Self as RuntimeFunction>::declare(self, context)
-{
-    fn declare(&mut self, context: &mut Context<D>) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::declare(self, context)
     }
 
-    fn into_llvm(self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn into_llvm(self, context: &mut Context) -> anyhow::Result<()> {
-        <Self as RuntimeFunction<_>>::emit(&self, context)
+        <Self as RuntimeFunction>::emit(&self, context)
     }
 }
 
-fn emit_load<'ctx, D: Dependency + Clone>(
+fn emit_load<'ctx>(
-    context: &mut Context<'ctx, D>,
+    context: &mut Context<'ctx>,
     key: BasicValueEnum<'ctx>,
     transient: bool,
 ) -> anyhow::Result<BasicValueEnum<'ctx>> {
+    let is_transient = context.xlen_type().const_int(transient as u64, false);
+    let key_pointer = context.build_alloca_at_entry(context.word_type(), "key_pointer");
+    let value_pointer = context.build_alloca_at_entry(context.word_type(), "value_pointer");
+
     let mut key = context.build_load(
         super::Pointer::new(
             context.word_type(),
@@ -193,33 +172,17 @@ fn emit_load<'ctx, D: Dependency + Clone>(
     if !transient {
         key = context.build_byte_swap(key)?;
     }
-
-    let key_pointer = context.build_alloca_at_entry(context.word_type(), "key_pointer");
-    let value_pointer = context.build_alloca_at_entry(context.word_type(), "value_pointer");
-    let length_pointer = context.build_alloca_at_entry(context.xlen_type(), "length_pointer");
-
     context.builder().build_store(key_pointer.value, key)?;
-    context.build_store(value_pointer, context.word_const(0))?;
-    context.build_store(
-        length_pointer,
-        context
-            .xlen_type()
-            .const_int(revive_common::BYTE_LENGTH_WORD as u64, false),
-    )?;
-
-    let is_transient = context.xlen_type().const_int(transient as u64, false);
 
     let arguments = [
         is_transient.into(),
         key_pointer.to_int(context).into(),
-        context.xlen_type().const_all_ones().into(),
         value_pointer.to_int(context).into(),
-        length_pointer.to_int(context).into(),
     ];
     context.build_runtime_call(revive_runtime_api::polkavm_imports::GET_STORAGE, &arguments);
 
     // We do not to check the return value: Solidity assumes infallible loads.
-    // If a key doesn't exist the "zero" value is returned (ensured by above write).
+    // If a key doesn't exist the syscall returns zero.
 
     let value = context.build_load(value_pointer, "storage_value")?;
     Ok(if transient {
@@ -229,12 +192,16 @@ fn emit_load<'ctx, D: Dependency + Clone>(
     })
 }
 
-fn emit_store<'ctx, D: Dependency + Clone>(
+fn emit_store<'ctx>(
-    context: &mut Context<'ctx, D>,
+    context: &mut Context<'ctx>,
     key: BasicValueEnum<'ctx>,
     value: BasicValueEnum<'ctx>,
     transient: bool,
 ) -> anyhow::Result<()> {
+    let is_transient = context.xlen_type().const_int(transient as u64, false);
+    let key_pointer = context.build_alloca_at_entry(context.word_type(), "key_pointer");
+    let value_pointer = context.build_alloca_at_entry(context.word_type(), "value_pointer");
+
     let mut key = context.build_load(
         super::Pointer::new(
             context.word_type(),
@@ -249,27 +216,20 @@ fn emit_store<'ctx, D: Dependency + Clone>(
             Default::default(),
             value.into_pointer_value(),
         ),
-        "key",
+        "value",
     )?;
     if !transient {
         key = context.build_byte_swap(key)?;
         value = context.build_byte_swap(value)?;
     }
 
-    let key_pointer = context.build_alloca_at_entry(context.word_type(), "key_pointer");
-    let value_pointer = context.build_alloca_at_entry(context.word_type(), "value_pointer");
-
     context.build_store(key_pointer, key)?;
     context.build_store(value_pointer, value)?;
 
-    let is_transient = context.xlen_type().const_int(transient as u64, false);
-
     let arguments = [
         is_transient.into(),
         key_pointer.to_int(context).into(),
-        context.xlen_type().const_all_ones().into(),
         value_pointer.to_int(context).into(),
-        context.integer_const(crate::polkavm::XLEN, 32).into(),
     ];
     context.build_runtime_call(revive_runtime_api::polkavm_imports::SET_STORAGE, &arguments);
 

crates/llvm-context/src/polkavm/context/runtime.rs

@@ -8,14 +8,10 @@ use crate::polkavm::context::function::declaration::Declaration;
 use crate::polkavm::context::function::Function;
 use crate::polkavm::context::Attribute;
 use crate::polkavm::context::Context;
-use crate::polkavm::Dependency;
 
 /// The revive runtime function interface simplifies declaring runtime functions
 /// and code emitting by providing helpful default implementations.
-pub trait RuntimeFunction<D>
+pub trait RuntimeFunction {
-where
-    D: Dependency + Clone,
-{
     /// The function name.
     const NAME: &'static str;
 
@@ -26,15 +22,17 @@ where
     ];
 
     /// The function type.
-    fn r#type<'ctx>(context: &Context<'ctx, D>) -> inkwell::types::FunctionType<'ctx>;
+    fn r#type<'ctx>(context: &Context<'ctx>) -> inkwell::types::FunctionType<'ctx>;
 
     /// Declare the function.
-    fn declare(&self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn declare(&self, context: &mut Context) -> anyhow::Result<()> {
         let function = context.add_function(
             Self::NAME,
             Self::r#type(context),
             0,
-            Some(inkwell::module::Linkage::External),
+            Some(inkwell::module::Linkage::LinkOnceODR),
+            None,
+            false,
         )?;
 
         let mut attributes = Self::ATTRIBUTES.to_vec();
@@ -48,22 +46,28 @@ where
             &attributes,
             true,
         );
+        let function = function.borrow().declaration().function_value();
+        let comdat = context
+            .module()
+            .get_or_insert_comdat(&format!("{}_comdat", Self::NAME));
+        comdat.set_selection_kind(inkwell::comdat::ComdatSelectionKind::NoDuplicates);
+        function.as_global_value().set_comdat(comdat);
 
         Ok(())
     }
 
     /// Get the function declaration.
-    fn declaration<'ctx>(context: &Context<'ctx, D>) -> Declaration<'ctx> {
+    fn declaration<'ctx>(context: &Context<'ctx>) -> Declaration<'ctx> {
         context
-            .get_function(Self::NAME)
+            .get_function(Self::NAME, false)
             .unwrap_or_else(|| panic!("runtime function {} should be declared", Self::NAME))
             .borrow()
             .declaration()
     }
 
     /// Emit the function.
-    fn emit(&self, context: &mut Context<D>) -> anyhow::Result<()> {
+    fn emit(&self, context: &mut Context) -> anyhow::Result<()> {
-        context.set_current_function(Self::NAME, None)?;
+        context.set_current_function(Self::NAME, None, false)?;
         context.set_basic_block(context.current_function().borrow().entry_block());
 
         let return_value = self.emit_body(context)?;
@@ -77,13 +81,13 @@ where
     /// Emit the function body.
     fn emit_body<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
     ) -> anyhow::Result<Option<inkwell::values::BasicValueEnum<'ctx>>>;
 
     /// Emit the function return instructions.
     fn emit_epilogue<'ctx>(
         &self,
-        context: &mut Context<'ctx, D>,
+        context: &mut Context<'ctx>,
         return_value: Option<inkwell::values::BasicValueEnum<'ctx>>,
     ) {
         let return_block = context.current_function().borrow().return_block();
@@ -97,12 +101,12 @@ where
 
     /// Get the nth function paramater.
     fn paramater<'ctx>(
-        context: &Context<'ctx, D>,
+        context: &Context<'ctx>,
         index: usize,
     ) -> inkwell::values::BasicValueEnum<'ctx> {
         let name = Self::NAME;
         context
-            .get_function(name)
+            .get_function(name, false)
             .unwrap_or_else(|| panic!("runtime function {name} should be declared"))
             .borrow()
             .declaration()

crates/llvm-context/src/polkavm/context/solidity_data.rs

@@ -2,6 +2,8 @@
 
 use std::collections::BTreeMap;
 
+use revive_common::BYTE_LENGTH_WORD;
+
 /// The LLVM IR generator Solidity data.
 /// Describes some data that is only relevant to Solidity.
 #[derive(Debug, Default)]
@@ -19,14 +21,14 @@ impl SolidityData {
 
     /// Returns the current size of immutable values in the contract.
     pub fn immutables_size(&self) -> usize {
-        self.immutables.len() * revive_common::BYTE_LENGTH_WORD
+        self.immutables.len() * BYTE_LENGTH_WORD
     }
 
     /// Allocates memory for an immutable value in the auxiliary heap.
     /// If the identifier is already known, just returns its offset.
     pub fn allocate_immutable(&mut self, identifier: &str) -> usize {
         let number_of_elements = self.immutables.len();
-        let new_offset = number_of_elements * revive_common::BYTE_LENGTH_WORD;
+        let new_offset = number_of_elements * BYTE_LENGTH_WORD;
         *self
             .immutables
             .entry(identifier.to_owned())

crates/llvm-context/src/polkavm/context/tests.rs

@@ -4,24 +4,21 @@ use crate::optimizer::settings::Settings as OptimizerSettings;
 use crate::optimizer::Optimizer;
 use crate::polkavm::context::attribute::Attribute;
 use crate::polkavm::context::Context;
-use crate::polkavm::DummyDependency;
+use crate::PolkaVMTarget;
 
 pub fn create_context(
     llvm: &inkwell::context::Context,
     optimizer_settings: OptimizerSettings,
-) -> Context<DummyDependency> {
+) -> Context<'_> {
-    crate::initialize_llvm(crate::Target::PVM, "resolc", Default::default());
+    crate::initialize_llvm(PolkaVMTarget::PVM, "resolc", Default::default());
 
     let module = llvm.create_module("test");
     let optimizer = Optimizer::new(optimizer_settings);
 
-    Context::<DummyDependency>::new(
+    Context::new(
         llvm,
         module,
         optimizer,
-        None,
-        true,
-        Default::default(),
         Default::default(),
         Default::default(),
     )
@@ -40,6 +37,8 @@ pub fn check_attribute_null_pointer_is_invalid() {
                 .fn_type(&[context.word_type().into()], false),
             1,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )
         .expect("Failed to add function");
     assert!(!function
@@ -63,6 +62,8 @@ pub fn check_attribute_optimize_for_size_mode_3() {
                 .fn_type(&[context.word_type().into()], false),
             1,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )
         .expect("Failed to add function");
     assert!(!function
@@ -86,6 +87,8 @@ pub fn check_attribute_optimize_for_size_mode_z() {
                 .fn_type(&[context.word_type().into()], false),
             1,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )
         .expect("Failed to add function");
     assert!(function
@@ -109,6 +112,8 @@ pub fn check_attribute_min_size_mode_3() {
                 .fn_type(&[context.word_type().into()], false),
             1,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )
         .expect("Failed to add function");
     assert!(!function
@@ -132,6 +137,8 @@ pub fn check_attribute_min_size_mode_z() {
                 .fn_type(&[context.word_type().into()], false),
             1,
             Some(inkwell::module::Linkage::External),
+            None,
+            false,
         )
         .expect("Failed to add function");
     assert!(function

crates/llvm-context/src/polkavm/context/yul_data.rs

@@ -2,60 +2,25 @@
 
 use std::collections::BTreeMap;
 
-use num::Zero;
-
 /// The LLVM IR generator Yul data.
-/// Describes some data that is only relevant to Yul.
+///
+/// Contains data that is only relevant to Yul.
 #[derive(Debug, Default)]
 pub struct YulData {
-    /// The list of constant arrays in the code section.
+    /// Mapping from Yul object identifiers to full contract paths.
-    /// It is a temporary storage used until the finalization method is called.
+    identifier_paths: BTreeMap<String, String>,
-    const_arrays: BTreeMap<u8, Vec<num::BigUint>>,
 }
 
 impl YulData {
-    /// Declares a temporary constant array representation.
+    /// A shorthand constructor.
-    pub fn const_array_declare(&mut self, index: u8, size: u16) -> anyhow::Result<()> {
+    pub fn new(identifier_paths: BTreeMap<String, String>) -> Self {
-        if self.const_arrays.contains_key(&index) {
+        Self { identifier_paths }
-            anyhow::bail!(
-                "The constant array with index {} is already declared",
-                index
-            );
     }
 
-        self.const_arrays
+    /// Resolves the full contract path by the Yul object identifier.
-            .insert(index, vec![num::BigUint::zero(); size as usize]);
+    pub fn resolve_path(&self, identifier: &str) -> Option<&str> {
-
+        self.identifier_paths
-        Ok(())
+            .get(identifier)
-    }
+            .map(|path| path.as_str())
-
-    /// Sets a value in the constant array representation.
-    pub fn const_array_set(
-        &mut self,
-        index: u8,
-        offset: u16,
-        value: num::BigUint,
-    ) -> anyhow::Result<()> {
-        let array = self.const_arrays.get_mut(&index).ok_or_else(|| {
-            anyhow::anyhow!("The constant array with index {} is not declared", index)
-        })?;
-        if offset >= array.len() as u16 {
-            anyhow::bail!(
-                "The constant array with index {} has size {} but the offset is {}",
-                index,
-                array.len(),
-                offset,
-            );
-        }
-        array[offset as usize] = value;
-
-        Ok(())
-    }
-
-    /// Finalizes the constant array declaration.
-    pub fn const_array_take(&mut self, index: u8) -> anyhow::Result<Vec<num::BigUint>> {
-        self.const_arrays.remove(&index).ok_or_else(|| {
-            anyhow::anyhow!("The constant array with index {} is not declared", index)
-        })
     }
 }