Merge branch 'main' into codeql-ci/js/ml-powered-pack-release-0.3.3

This commit is contained in:
Henry Mercer
2022-09-13 15:15:56 +01:00
committed by GitHub
5550 changed files with 471715 additions and 181168 deletions

View File

@@ -1,30 +0,0 @@
{
"provide": [
"*/ql/src/qlpack.yml",
"*/ql/lib/qlpack.yml",
"*/ql/test/qlpack.yml",
"*/ql/examples/qlpack.yml",
"*/ql/consistency-queries/qlpack.yml",
"cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/qlpack.yml",
"go/ql/config/legacy-support/qlpack.yml",
"go/build/codeql-extractor-go/codeql-extractor.yml",
"javascript/ql/experimental/adaptivethreatmodeling/lib/qlpack.yml",
"javascript/ql/experimental/adaptivethreatmodeling/modelbuilding/qlpack.yml",
"javascript/ql/experimental/adaptivethreatmodeling/src/qlpack.yml",
"csharp/ql/campaigns/Solorigate/lib/qlpack.yml",
"csharp/ql/campaigns/Solorigate/src/qlpack.yml",
"csharp/ql/campaigns/Solorigate/test/qlpack.yml",
"misc/legacy-support/*/qlpack.yml",
"misc/suite-helpers/qlpack.yml",
"ruby/extractor-pack/codeql-extractor.yml",
"swift/extractor-pack/codeql-extractor.yml",
"ql/extractor-pack/codeql-extractor.yml"
],
"versionPolicies": {
"default": {
"requireChangeNotes": true,
"committedPrereleaseSuffix": "dev",
"committedVersion": "nextPatchRelease"
}
}
}

View File

@@ -0,0 +1,9 @@
# See here for image contents: https://github.com/microsoft/vscode-dev-containers/tree/v0.236.0/containers/cpp/.devcontainer/base.Dockerfile
# [Choice] Debian / Ubuntu version (use Debian 11, Ubuntu 18.04/22.04 on local arm64/Apple Silicon): debian-11, debian-10, ubuntu-22.04, ubuntu-20.04, ubuntu-18.04
FROM mcr.microsoft.com/vscode/devcontainers/cpp:0-ubuntu-22.04
USER root
ADD root.sh /tmp/root.sh
ADD update-codeql.sh /usr/local/bin/update-codeql
RUN bash /tmp/root.sh && rm /tmp/root.sh

View File

@@ -0,0 +1,25 @@
{
"extensions": [
"github.vscode-codeql",
"hbenl.vscode-test-explorer",
"ms-vscode.test-adapter-converter",
"slevesque.vscode-zipexplorer",
"ms-vscode.cpptools"
],
"settings": {
"files.watcherExclude": {
"**/target/**": true
},
"codeQL.runningQueries.memory": 2048
},
"build": {
"dockerfile": "Dockerfile",
},
"runArgs": [
"--cap-add=SYS_PTRACE",
"--security-opt",
"seccomp=unconfined"
],
"remoteUser": "vscode",
"onCreateCommand": ".devcontainer/swift/user.sh"
}

View File

@@ -0,0 +1,22 @@
set -xe
BAZELISK_VERSION=v1.12.0
BAZELISK_DOWNLOAD_SHA=6b0bcb2ea15bca16fffabe6fda75803440375354c085480fe361d2cbf32501db
apt-get update
export DEBIAN_FRONTEND=noninteractive
apt-get -y install --no-install-recommends \
zlib1g-dev \
uuid-dev \
python3-distutils \
python3-pip \
bash-completion
# Install Bazel
curl -fSsL -o /usr/local/bin/bazelisk https://github.com/bazelbuild/bazelisk/releases/download/${BAZELISK_VERSION}/bazelisk-linux-amd64
echo "${BAZELISK_DOWNLOAD_SHA} */usr/local/bin/bazelisk" | sha256sum --check -
chmod 0755 /usr/local/bin/bazelisk
ln -s bazelisk /usr/local/bin/bazel
# install latest codeql
update-codeql

View File

@@ -0,0 +1,20 @@
#!/bin/bash -e
URL=https://github.com/github/codeql-cli-binaries/releases
LATEST_VERSION=$(curl -L -s -H 'Accept: application/json' $URL/latest | sed -e 's/.*"tag_name":"\([^"]*\)".*/\1/')
CURRENT_VERSION=v$(codeql version 2>/dev/null | sed -ne 's/.*release \([0-9.]*\)\./\1/p')
if [[ $CURRENT_VERSION != $LATEST_VERSION ]]; then
if [[ $UID != 0 ]]; then
echo "update required, please run this script with sudo:"
echo " sudo $0"
exit 1
fi
ZIP=$(mktemp codeql.XXXX.zip)
curl -fSqL -o $ZIP $URL/download/$LATEST_VERSION/codeql-linux64.zip
unzip -q $ZIP -d /opt
rm $ZIP
ln -sf /opt/codeql/codeql /usr/local/bin/codeql
echo installed version $LATEST_VERSION
else
echo current version $CURRENT_VERSION is up-to-date
fi

13
.devcontainer/swift/user.sh Executable file
View File

@@ -0,0 +1,13 @@
set -xe
# add the workspace to the codeql search path
mkdir -p /home/vscode/.config/codeql
echo "--search-path /workspaces/codeql" > /home/vscode/.config/codeql/config
# create a swift extractor pack with the current state
cd /workspaces/codeql
bazel run swift/create-extractor-pack
#install and set up pre-commit
python3 -m pip install pre-commit --no-warn-script-location
$HOME/.local/bin/pre-commit install

View File

@@ -3,22 +3,12 @@ description: Fetches the latest version of CodeQL
runs:
using: composite
steps:
- name: Select platform - Linux
if: runner.os == 'Linux'
shell: bash
run: echo "GA_CODEQL_CLI_PLATFORM=linux64" >> $GITHUB_ENV
- name: Select platform - MacOS
if: runner.os == 'MacOS'
shell: bash
run: echo "GA_CODEQL_CLI_PLATFORM=osx64" >> $GITHUB_ENV
- name: Fetch CodeQL
shell: bash
run: |
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | grep -v beta | sort --version-sort | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-$GA_CODEQL_CLI_PLATFORM.zip "$LATEST"
unzip -q -d "${RUNNER_TEMP}" codeql-$GA_CODEQL_CLI_PLATFORM.zip
echo "${RUNNER_TEMP}/codeql" >> "${GITHUB_PATH}"
gh extension install github/gh-codeql
gh codeql set-channel nightly
gh codeql version
gh codeql version --format=json | jq -r .unpackedLocation >> "${GITHUB_PATH}"
env:
GITHUB_TOKEN: ${{ github.token }}

16
.github/labeler.yml vendored
View File

@@ -6,14 +6,23 @@
- csharp/**/*
- change-notes/**/*csharp*
Go:
- go/**/*
- change-notes/**/*go.*
Java:
- java/**/*
- any: [ 'java/**/*', '!java/kotlin-extractor/**/*', '!java/kotlin-explorer/**/*', '!java/ql/test/kotlin/**/*' ]
- change-notes/**/*java.*
JS:
- any: [ 'javascript/**/*', '!javascript/ql/experimental/adaptivethreatmodeling/**/*' ]
- change-notes/**/*javascript*
Kotlin:
- java/kotlin-extractor/**/*
- java/kotlin-explorer/**/*
- java/ql/test/kotlin/**/*
Python:
- python/**/*
- change-notes/**/*python*
@@ -21,7 +30,7 @@ Python:
Ruby:
- ruby/**/*
- change-notes/**/*ruby*
Swift:
- swift/**/*
- change-notes/**/*swift*
@@ -31,5 +40,6 @@ documentation:
- "**/*.md"
- docs/**/*
"QL-for-QL":
"QL-for-QL":
- ql/**/*
- .github/workflows/ql-for-ql*

View File

@@ -10,6 +10,7 @@ on:
- "*/ql/lib/**/*.qll"
- "!**/experimental/**"
- "!ql/**"
- "!swift/**"
- ".github/workflows/check-change-note.yml"
jobs:

View File

@@ -5,6 +5,7 @@ on:
paths:
- "*/ql/lib/**"
- .github/workflows/check-qldoc.yml
- .github/actions/fetch-codeql/action.yml
branches:
- main
- "rc/*"
@@ -14,27 +15,23 @@ jobs:
runs-on: ubuntu-latest
steps:
- name: Install CodeQL
run: |
gh extension install github/gh-codeql
gh codeql set-channel nightly
gh codeql version
env:
GITHUB_TOKEN: ${{ github.token }}
- uses: actions/checkout@v3
with:
fetch-depth: 2
- name: Install CodeQL
uses: ./.github/actions/fetch-codeql
- name: Check QLdoc coverage
shell: bash
run: |
EXIT_CODE=0
# TODO: remove the swift exception from the regex when we fix generated QLdoc
changed_lib_packs="$(git diff --name-only --diff-filter=ACMRT HEAD^ HEAD | { grep -Po '^(?!swift)[a-z]*/ql/lib' || true; } | sort -u)"
# TODO: remove the shared exception from the regex when coverage of qlpacks without dbschemes is supported
changed_lib_packs="$(git diff --name-only --diff-filter=ACMRT HEAD^ HEAD | { grep -Po '^(?!(swift|shared))[a-z]*/ql/lib' || true; } | sort -u)"
for pack_dir in ${changed_lib_packs}; do
lang="${pack_dir%/ql/lib}"
gh codeql generate library-doc-coverage --output="${RUNNER_TEMP}/${lang}-current.txt" --dir="${pack_dir}"
codeql generate library-doc-coverage --output="${RUNNER_TEMP}/${lang}-current.txt" --dir="${pack_dir}"
done
git checkout HEAD^
for pack_dir in ${changed_lib_packs}; do
@@ -42,7 +39,7 @@ jobs:
# In this case the right thing to do is to skip the check.
[[ ! -d "${pack_dir}" ]] && continue
lang="${pack_dir%/ql/lib}"
gh codeql generate library-doc-coverage --output="${RUNNER_TEMP}/${lang}-baseline.txt" --dir="${pack_dir}"
codeql generate library-doc-coverage --output="${RUNNER_TEMP}/${lang}-baseline.txt" --dir="${pack_dir}"
awk -F, '{gsub(/"/,""); if ($4==0 && $6=="public") print "\""$3"\"" }' "${RUNNER_TEMP}/${lang}-current.txt" | sort -u > "${RUNNER_TEMP}/current-undocumented.txt"
awk -F, '{gsub(/"/,""); if ($4==0 && $6=="public") print "\""$3"\"" }' "${RUNNER_TEMP}/${lang}-baseline.txt" | sort -u > "${RUNNER_TEMP}/baseline-undocumented.txt"
UNDOCUMENTED="$(grep -f <(comm -13 "${RUNNER_TEMP}/baseline-undocumented.txt" "${RUNNER_TEMP}/current-undocumented.txt") "${RUNNER_TEMP}/${lang}-current.txt" || true)"

View File

@@ -56,7 +56,7 @@ jobs:
# uses a compiled language
- run: |
dotnet build csharp /p:UseSharedCompilation=false
dotnet build csharp
- name: Perform CodeQL Analysis
uses: github/codeql-action/analyze@main

View File

@@ -12,6 +12,7 @@ on:
- main
paths:
- ".github/workflows/csv-coverage-metrics.yml"
- ".github/actions/fetch-codeql/action.yml"
jobs:
publish-java:
@@ -54,7 +55,7 @@ jobs:
DATABASE="${{ runner.temp }}/csharp-database"
PROJECT="${{ runner.temp }}/csharp-project"
dotnet new classlib --language=C# --output="$PROJECT"
codeql database create "$DATABASE" --language=csharp --source-root="$PROJECT" --command 'dotnet build /t:rebuild csharp-project.csproj /p:UseSharedCompilation=false'
codeql database create "$DATABASE" --language=csharp --source-root="$PROJECT" --command 'dotnet build /t:rebuild csharp-project.csproj'
- name: Capture coverage information
run: |
DATABASE="${{ runner.temp }}/csharp-database"

View File

@@ -3,18 +3,20 @@ name: Check framework coverage changes
on:
pull_request:
paths:
- '.github/workflows/csv-coverage-pr-comment.yml'
- '*/ql/src/**/*.ql'
- '*/ql/src/**/*.qll'
- '*/ql/lib/**/*.ql'
- '*/ql/lib/**/*.qll'
- 'misc/scripts/library-coverage/*.py'
- ".github/workflows/csv-coverage-pr-comment.yml"
- ".github/workflows/csv-coverage-pr-artifacts.yml"
- ".github/actions/fetch-codeql/action.yml"
- "*/ql/src/**/*.ql"
- "*/ql/src/**/*.qll"
- "*/ql/lib/**/*.ql"
- "*/ql/lib/**/*.qll"
- "misc/scripts/library-coverage/*.py"
# input data files
- '*/documentation/library-coverage/cwe-sink.csv'
- '*/documentation/library-coverage/frameworks.csv'
- "*/documentation/library-coverage/cwe-sink.csv"
- "*/documentation/library-coverage/frameworks.csv"
branches:
- main
- 'rc/*'
- "rc/*"
jobs:
generate:
@@ -23,77 +25,72 @@ jobs:
runs-on: ubuntu-latest
steps:
- name: Dump GitHub context
env:
GITHUB_CONTEXT: ${{ toJSON(github.event) }}
run: echo "$GITHUB_CONTEXT"
- name: Clone self (github/codeql) - MERGE
uses: actions/checkout@v3
with:
path: merge
- name: Clone self (github/codeql) - BASE
uses: actions/checkout@v3
with:
fetch-depth: 2
path: base
- run: |
git checkout HEAD^1
git log -1 --format='%H'
working-directory: base
- name: Set up Python 3.8
uses: actions/setup-python@v3
with:
python-version: 3.8
- name: Download CodeQL CLI
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh release download --repo "github/codeql-cli-binaries" --pattern "codeql-linux64.zip"
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Generate CSV files on merge commit of the PR
run: |
echo "Running generator on merge"
PATH="$PATH:codeql-cli/codeql" python merge/misc/scripts/library-coverage/generate-report.py ci merge merge
mkdir out_merge
cp framework-coverage-*.csv out_merge/
cp framework-coverage-*.rst out_merge/
- name: Generate CSV files on base commit of the PR
run: |
echo "Running generator on base"
PATH="$PATH:codeql-cli/codeql" python base/misc/scripts/library-coverage/generate-report.py ci base base
mkdir out_base
cp framework-coverage-*.csv out_base/
cp framework-coverage-*.rst out_base/
- name: Generate diff of coverage reports
run: |
python base/misc/scripts/library-coverage/compare-folders.py out_base out_merge comparison.md
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: csv-framework-coverage-merge
path: |
out_merge/framework-coverage-*.csv
out_merge/framework-coverage-*.rst
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: csv-framework-coverage-base
path: |
out_base/framework-coverage-*.csv
out_base/framework-coverage-*.rst
- name: Upload comparison results
uses: actions/upload-artifact@v3
with:
name: comparison
path: |
comparison.md
- name: Save PR number
run: |
mkdir -p pr
echo ${{ github.event.pull_request.number }} > pr/NR
- name: Upload PR number
uses: actions/upload-artifact@v3
with:
name: pr
path: pr/
- name: Dump GitHub context
env:
GITHUB_CONTEXT: ${{ toJSON(github.event) }}
run: echo "$GITHUB_CONTEXT"
- name: Clone self (github/codeql) - MERGE
uses: actions/checkout@v3
with:
path: merge
- name: Clone self (github/codeql) - BASE
uses: actions/checkout@v3
with:
fetch-depth: 2
path: base
- run: |
git checkout HEAD^1
git log -1 --format='%H'
working-directory: base
- name: Set up Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
- name: Download CodeQL CLI
uses: ./merge/.github/actions/fetch-codeql
- name: Generate CSV files on merge commit of the PR
run: |
echo "Running generator on merge"
python merge/misc/scripts/library-coverage/generate-report.py ci merge merge
mkdir out_merge
cp framework-coverage-*.csv out_merge/
cp framework-coverage-*.rst out_merge/
- name: Generate CSV files on base commit of the PR
run: |
echo "Running generator on base"
python base/misc/scripts/library-coverage/generate-report.py ci base base
mkdir out_base
cp framework-coverage-*.csv out_base/
cp framework-coverage-*.rst out_base/
- name: Generate diff of coverage reports
run: |
python base/misc/scripts/library-coverage/compare-folders.py out_base out_merge comparison.md
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: csv-framework-coverage-merge
path: |
out_merge/framework-coverage-*.csv
out_merge/framework-coverage-*.rst
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: csv-framework-coverage-base
path: |
out_base/framework-coverage-*.csv
out_base/framework-coverage-*.rst
- name: Upload comparison results
uses: actions/upload-artifact@v3
with:
name: comparison
path: |
comparison.md
- name: Save PR number
run: |
mkdir -p pr
echo ${{ github.event.pull_request.number }} > pr/NR
- name: Upload PR number
uses: actions/upload-artifact@v3
with:
name: pr
path: pr/

View File

@@ -22,7 +22,7 @@ jobs:
- name: Clone self (github/codeql)
uses: actions/checkout@v3
- name: Set up Python 3.8
uses: actions/setup-python@v3
uses: actions/setup-python@v4
with:
python-version: 3.8

View File

@@ -5,38 +5,29 @@ on:
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: script
- name: Clone self (github/codeql) for analysis
uses: actions/checkout@v3
with:
path: codeqlModels
fetch-depth: 0
- name: Set up Python 3.8
uses: actions/setup-python@v3
with:
python-version: 3.8
- name: Download CodeQL CLI
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh release download --repo "github/codeql-cli-binaries" --pattern "codeql-linux64.zip"
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Build modeled package list
run: |
CLI=$(realpath "codeql-cli/codeql")
echo $CLI
PATH="$PATH:$CLI" python script/misc/scripts/library-coverage/generate-timeseries.py codeqlModels
- name: Upload timeseries CSV
uses: actions/upload-artifact@v3
with:
name: framework-coverage-timeseries
path: framework-coverage-timeseries-*.csv
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: script
- name: Clone self (github/codeql) for analysis
uses: actions/checkout@v3
with:
path: codeqlModels
fetch-depth: 0
- name: Set up Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
- name: Download CodeQL CLI
uses: ./script/.github/actions/fetch-codeql
- name: Build modeled package list
run: |
python script/misc/scripts/library-coverage/generate-timeseries.py codeqlModels
- name: Upload timeseries CSV
uses: actions/upload-artifact@v3
with:
name: framework-coverage-timeseries
path: framework-coverage-timeseries-*.csv

View File

@@ -12,33 +12,27 @@ jobs:
runs-on: ubuntu-latest
steps:
- name: Dump GitHub context
env:
GITHUB_CONTEXT: ${{ toJSON(github.event) }}
run: echo "$GITHUB_CONTEXT"
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: ql
fetch-depth: 0
- name: Set up Python 3.8
uses: actions/setup-python@v3
with:
python-version: 3.8
- name: Download CodeQL CLI
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh release download --repo "github/codeql-cli-binaries" --pattern "codeql-linux64.zip"
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Dump GitHub context
env:
GITHUB_CONTEXT: ${{ toJSON(github.event) }}
run: echo "$GITHUB_CONTEXT"
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: ql
fetch-depth: 0
- name: Set up Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
- name: Download CodeQL CLI
uses: ./ql/.github/actions/fetch-codeql
- name: Generate coverage files
run: |
python ql/misc/scripts/library-coverage/generate-report.py ci ql ql
- name: Generate coverage files
run: |
PATH="$PATH:codeql-cli/codeql" python ql/misc/scripts/library-coverage/generate-report.py ci ql ql
- name: Create pull request with changes
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
python ql/misc/scripts/library-coverage/create-pr.py ql "$GITHUB_REPOSITORY"
- name: Create pull request with changes
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
python ql/misc/scripts/library-coverage/create-pr.py ql "$GITHUB_REPOSITORY"

View File

@@ -4,46 +4,39 @@ on:
workflow_dispatch:
inputs:
qlModelShaOverride:
description: 'github/codeql repo SHA used for looking up the CSV models'
description: "github/codeql repo SHA used for looking up the CSV models"
required: false
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: script
- name: Clone self (github/codeql) for analysis
uses: actions/checkout@v3
with:
path: codeqlModels
ref: ${{ github.event.inputs.qlModelShaOverride || github.ref }}
- name: Set up Python 3.8
uses: actions/setup-python@v3
with:
python-version: 3.8
- name: Download CodeQL CLI
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh release download --repo "github/codeql-cli-binaries" --pattern "codeql-linux64.zip"
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Build modeled package list
run: |
PATH="$PATH:codeql-cli/codeql" python script/misc/scripts/library-coverage/generate-report.py ci codeqlModels script
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: framework-coverage-csv
path: framework-coverage-*.csv
- name: Upload RST package list
uses: actions/upload-artifact@v3
with:
name: framework-coverage-rst
path: framework-coverage-*.rst
- name: Clone self (github/codeql)
uses: actions/checkout@v3
with:
path: script
- name: Clone self (github/codeql) for analysis
uses: actions/checkout@v3
with:
path: codeqlModels
ref: ${{ github.event.inputs.qlModelShaOverride || github.ref }}
- name: Set up Python 3.8
uses: actions/setup-python@v4
with:
python-version: 3.8
- name: Download CodeQL CLI
uses: ./script/.github/actions/fetch-codeql
- name: Build modeled package list
run: |
python script/misc/scripts/library-coverage/generate-report.py ci codeqlModels script
- name: Upload CSV package list
uses: actions/upload-artifact@v3
with:
name: framework-coverage-csv
path: framework-coverage-*.csv
- name: Upload RST package list
uses: actions/upload-artifact@v3
with:
name: framework-coverage-rst
path: framework-coverage-*.rst

View File

@@ -4,158 +4,111 @@ on:
paths:
- "go/**"
- .github/workflows/go-tests.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
jobs:
test-linux:
name: Test Linux (Ubuntu)
runs-on: ubuntu-latest
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Set up Go 1.18.1
uses: actions/setup-go@v3
with:
go-version: 1.18.1
id: go
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
run: |
echo "Removing old CodeQL Directory..."
rm -rf $HOME/codeql
echo "Done"
cd $HOME
echo "Downloading CodeQL CLI..."
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | sort --version-sort | grep -v beta | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-linux64.zip "$LATEST"
echo "Done"
echo "Unpacking CodeQL CLI..."
unzip -q codeql-linux64.zip
rm -f codeql-linux64.zip
echo "Done"
env:
GITHUB_TOKEN: ${{ github.token }}
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Check out code
uses: actions/checkout@v2
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
make
- name: Build
run: |
cd go
env PATH=$PATH:$HOME/codeql make
- name: Check that all QL and Go code is autoformatted
run: |
cd go
make check-formatting
- name: Check that all QL and Go code is autoformatted
run: |
cd go
env PATH=$PATH:$HOME/codeql make check-formatting
- name: Compile qhelp files to markdown
run: |
cd go
env QHELP_OUT_DIR=qhelp-out make qhelp-to-markdown
- name: Compile qhelp files to markdown
run: |
cd go
env PATH=$PATH:$HOME/codeql QHELP_OUT_DIR=qhelp-out make qhelp-to-markdown
- name: Upload qhelp markdown
uses: actions/upload-artifact@v2
with:
name: qhelp-markdown
path: go/qhelp-out/**/*.md
- name: Upload qhelp markdown
uses: actions/upload-artifact@v2
with:
name: qhelp-markdown
path: go/qhelp-out/**/*.md
- name: Test
run: |
cd go
env PATH=$PATH:$HOME/codeql make test
- name: Test
run: |
cd go
make test
test-mac:
name: Test MacOS
runs-on: macOS-latest
runs-on: macos-latest
steps:
- name: Set up Go 1.18.1
uses: actions/setup-go@v3
with:
go-version: 1.18.1
id: go
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Set up CodeQL CLI
run: |
echo "Removing old CodeQL Directory..."
rm -rf $HOME/codeql
echo "Done"
cd $HOME
echo "Downloading CodeQL CLI..."
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | sort --version-sort | grep -v beta | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-osx64.zip "$LATEST"
echo "Done"
echo "Unpacking CodeQL CLI..."
unzip -q codeql-osx64.zip
rm -f codeql-osx64.zip
echo "Done"
env:
GITHUB_TOKEN: ${{ github.token }}
- name: Check out code
uses: actions/checkout@v2
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
env PATH=$PATH:$HOME/codeql make
- name: Build
run: |
cd go
make
- name: Test
run: |
cd go
env PATH=$PATH:$HOME/codeql make test
- name: Test
run: |
cd go
make test
test-win:
name: Test Windows
runs-on: windows-2019
steps:
- name: Set up Go 1.18.1
uses: actions/setup-go@v3
with:
go-version: 1.18.1
id: go
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Set up CodeQL CLI
run: |
echo "Removing old CodeQL Directory..."
rm -rf $HOME/codeql
echo "Done"
cd "$HOME"
echo "Downloading CodeQL CLI..."
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | sort --version-sort | grep -v beta | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-win64.zip "$LATEST"
echo "Done"
echo "Unpacking CodeQL CLI..."
unzip -q -o codeql-win64.zip
unzip -q -o codeql-win64.zip codeql/codeql.exe
rm -f codeql-win64.zip
echo "Done"
env:
GITHUB_TOKEN: ${{ github.token }}
shell:
bash
- name: Check out code
uses: actions/checkout@v2
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
$Env:Path += ";$HOME\codeql"
cd go
make
- name: Build
run: |
cd go
make
- name: Test
run: |
$Env:Path += ";$HOME\codeql"
cd go
make test
- name: Test
run: |
cd go
make test

View File

@@ -5,6 +5,8 @@ on:
paths:
- "javascript/ql/experimental/adaptivethreatmodeling/**"
- .github/workflows/js-ml-tests.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
- "rc/*"
@@ -12,6 +14,9 @@ on:
paths:
- "javascript/ql/experimental/adaptivethreatmodeling/**"
- .github/workflows/js-ml-tests.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
workflow_dispatch:
defaults:
run:

View File

@@ -4,6 +4,9 @@ on:
jobs:
triage:
permissions:
contents: read
pull-requests: write
runs-on: ubuntu-latest
steps:
- uses: actions/labeler@v4

View File

@@ -61,7 +61,7 @@ jobs:
DATABASE=$2
cd codeql-$QL_VARIANT
SHORTNAME=`basename $DATABASE`
python java/ql/src/utils/model-generator/GenerateFlowModel.py $DATABASE $MODELS/${SHORTNAME}.qll
python java/ql/src/utils/model-generator/GenerateFlowModel.py --with-summaries --with-sinks $DATABASE $MODELS/${SHORTNAME}.qll
mv $MODELS/${SHORTNAME}.qll $MODELS/${SHORTNAME}Generated_${QL_VARIANT}.qll
cd ..
}

View File

@@ -9,6 +9,7 @@ on:
- main
paths:
- ".github/workflows/mad_regenerate-models.yml"
- ".github/actions/fetch-codeql/action.yml"
jobs:
regenerate-models:
@@ -20,7 +21,7 @@ jobs:
ref: ["placeholder"]
include:
- slug: "apache/commons-io"
ref: "8985de8fe74f6622a419b37a6eed0dbc484dc128"
ref: "13258ce2d07aa0e764bbaa8020af4dcd3a02a620"
exclude:
- slug: "placeholder"
ref: "placeholder"

View File

@@ -5,21 +5,28 @@ on:
branches: [main]
pull_request:
branches: [main]
paths:
- "ql/**"
- "**.qll"
- "**.ql"
- "**.dbscheme"
- "**/qlpack.yml"
- ".github/workflows/ql-for-ql-build.yml"
env:
CARGO_TERM_COLOR: always
jobs:
queries:
runs-on: ubuntu-latest
analyze:
runs-on: ubuntu-latest-xl
steps:
### Build the queries ###
- uses: actions/checkout@v3
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@aa93aea877e5fb8841bcb1193f672abf6e9f2980
uses: github/codeql-action/init@71a8b35ff4c80fcfcd05bc1cd932fe3c08f943ca
with:
languages: javascript # does not matter
tools: latest
- name: Get CodeQL version
id: get-codeql-version
run: |
@@ -27,37 +34,37 @@ jobs:
shell: bash
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Cache entire pack
id: cache-pack
uses: actions/cache@v3
with:
path: ${{ runner.temp }}/pack
key: ${{ runner.os }}-pack-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}-${{ hashFiles('ql/**/*.ql*') }}-${{ hashFiles('ql/**/qlpack.yml') }}-${{ hashFiles('ql/ql/src/ql.dbscheme*') }}-${{ steps.get-codeql-version.outputs.version }}--${{ hashFiles('.github/workflows/ql-for-ql-build.yml') }}
- name: Cache queries
if: steps.cache-pack.outputs.cache-hit != 'true'
id: cache-queries
uses: actions/cache@v3
with:
path: ${{ runner.temp }}/query-pack.zip
key: queries-${{ hashFiles('ql/**/*.ql*') }}-${{ hashFiles('ql/**/qlpack.yml') }}-${{ hashFiles('ql/ql/src/ql.dbscheme*') }}-${{ steps.get-codeql-version.outputs.version }}
path: ${{ runner.temp }}/queries
key: queries-${{ hashFiles('ql/**/*.ql*') }}-${{ hashFiles('ql/**/qlpack.yml') }}-${{ hashFiles('ql/ql/src/ql.dbscheme*') }}-${{ steps.get-codeql-version.outputs.version }}--${{ hashFiles('.github/workflows/ql-for-ql-build.yml') }}
- name: Build query pack
if: steps.cache-queries.outputs.cache-hit != 'true'
if: steps.cache-queries.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: |
cd ql/ql/src
"${CODEQL}" pack create
cd .codeql/pack/codeql/ql/0.0.0
zip "${PACKZIP}" -r .
"${CODEQL}" pack create -j 16
mv .codeql/pack/codeql/ql/0.0.0 ${{ runner.temp }}/queries
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Move cache queries to pack
if: steps.cache-pack.outputs.cache-hit != 'true'
run: |
cp -r ${{ runner.temp }}/queries ${{ runner.temp }}/pack
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
PACKZIP: ${{ runner.temp }}/query-pack.zip
- name: Upload query pack
uses: actions/upload-artifact@v3
with:
name: query-pack-zip
path: ${{ runner.temp }}/query-pack.zip
extractors:
strategy:
fail-fast: false
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
### Build the extractor ###
- name: Cache entire extractor
if: steps.cache-pack.outputs.cache-hit != 'true'
id: cache-extractor
uses: actions/cache@v3
with:
@@ -68,7 +75,7 @@ jobs:
ql/target/release/ql-extractor.exe
key: ${{ runner.os }}-extractor-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}
- name: Cache cargo
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
uses: actions/cache@v3
with:
path: |
@@ -77,127 +84,87 @@ jobs:
ql/target
key: ${{ runner.os }}-rust-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
- name: Check formatting
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: cd ql; cargo fmt --all -- --check
- name: Build
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: cd ql; cargo build --verbose
- name: Run tests
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: cd ql; cargo test --verbose
- name: Release build
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: cd ql; cargo build --release
- name: Generate dbscheme
if: steps.cache-extractor.outputs.cache-hit != 'true'
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: ql/target/release/ql-generator --dbscheme ql/ql/src/ql.dbscheme --library ql/ql/src/codeql_ql/ast/internal/TreeSitter.qll
- uses: actions/upload-artifact@v3
with:
name: extractor-ubuntu-latest
path: |
ql/target/release/ql-autobuilder
ql/target/release/ql-autobuilder.exe
ql/target/release/ql-extractor
ql/target/release/ql-extractor.exe
retention-days: 1
package:
runs-on: ubuntu-latest
needs:
- extractors
- queries
steps:
- uses: actions/checkout@v3
- uses: actions/download-artifact@v3
with:
name: query-pack-zip
path: query-pack-zip
- uses: actions/download-artifact@v3
with:
name: extractor-ubuntu-latest
path: linux64
- run: |
unzip query-pack-zip/*.zip -d pack
cp -r ql/codeql-extractor.yml ql/tools ql/ql/src/ql.dbscheme.stats pack/
mkdir -p pack/tools/linux64
if [[ -f linux64/ql-autobuilder ]]; then
cp linux64/ql-autobuilder pack/tools/linux64/autobuilder
chmod +x pack/tools/linux64/autobuilder
fi
if [[ -f linux64/ql-extractor ]]; then
cp linux64/ql-extractor pack/tools/linux64/extractor
chmod +x pack/tools/linux64/extractor
fi
cd pack
zip -rq ../codeql-ql.zip .
- uses: actions/upload-artifact@v3
with:
name: codeql-ql-pack
path: codeql-ql.zip
retention-days: 1
analyze:
runs-on: ubuntu-latest
strategy:
matrix:
folder: [cpp, csharp, java, javascript, python, ql, ruby, swift, go]
needs:
- package
steps:
- name: Download pack
uses: actions/download-artifact@v3
with:
name: codeql-ql-pack
path: ${{ runner.temp }}/codeql-ql-pack-artifact
- name: Prepare pack
### Package the queries and extractor ###
- name: Package pack
if: steps.cache-pack.outputs.cache-hit != 'true'
run: |
unzip "${PACK_ARTIFACT}/*.zip" -d "${PACK}"
cp -r ql/codeql-extractor.yml ql/tools ql/ql/src/ql.dbscheme.stats ${PACK}/
mkdir -p ${PACK}/tools/linux64
cp ql/target/release/ql-autobuilder ${PACK}/tools/linux64/autobuilder
cp ql/target/release/ql-extractor ${PACK}/tools/linux64/extractor
chmod +x ${PACK}/tools/linux64/autobuilder
chmod +x ${PACK}/tools/linux64/extractor
env:
PACK_ARTIFACT: ${{ runner.temp }}/codeql-ql-pack-artifact
PACK: ${{ runner.temp }}/pack
### Run the analysis ###
- name: Hack codeql-action options
run: |
JSON=$(jq -nc --arg pack "${PACK}" '.database."run-queries"=["--search-path", $pack] | .resolve.queries=["--search-path", $pack] | .resolve.extractor=["--search-path", $pack] | .database.init=["--search-path", $pack]')
JSON=$(jq -nc --arg pack "${PACK}" '.database."run-queries"=["--search-path", $pack] | .resolve.queries=["--search-path", $pack] | .resolve.extractor=["--search-path", $pack] | .resolve.languages=["--search-path", $pack] | .database.init=["--search-path", $pack]')
echo "CODEQL_ACTION_EXTRA_OPTIONS=${JSON}" >> ${GITHUB_ENV}
env:
PACK: ${{ runner.temp }}/pack
- name: Checkout repository
uses: actions/checkout@v3
- name: Create CodeQL config file
run: |
echo "paths:" > ${CONF}
echo " - ${FOLDER}" >> ${CONF}
echo "paths-ignore:" >> ${CONF}
echo " - ql/ql/test" >> ${CONF}
echo " - ql/ql/test" >> ${CONF}
echo " - \"*/ql/lib/upgrades/\"" >> ${CONF}
echo "disable-default-queries: true" >> ${CONF}
echo "packs:" >> ${CONF}
echo " - codeql/ql" >> ${CONF}
echo "queries:" >> ${CONF}
echo " - uses: ./ql/ql/src/codeql-suites/ql-code-scanning.qls" >> ${CONF}
echo "Config file: "
cat ${CONF}
env:
env:
CONF: ./ql-for-ql-config.yml
FOLDER: ${{ matrix.folder }}
- name: Initialize CodeQL
uses: github/codeql-action/init@aa93aea877e5fb8841bcb1193f672abf6e9f2980
uses: github/codeql-action/init@71a8b35ff4c80fcfcd05bc1cd932fe3c08f943ca
with:
languages: ql
db-location: ${{ runner.temp }}/db
config-file: ./ql-for-ql-config.yml
tools: latest
- name: Move pack cache
run: |
cp -r ${PACK}/.cache ql/ql/src/.cache
env:
PACK: ${{ runner.temp }}/pack
- name: Perform CodeQL Analysis
uses: github/codeql-action/analyze@aa93aea877e5fb8841bcb1193f672abf6e9f2980
with:
category: "ql-for-ql-${{ matrix.folder }}"
uses: github/codeql-action/analyze@71a8b35ff4c80fcfcd05bc1cd932fe3c08f943ca
with:
category: "ql-for-ql"
- name: Copy sarif file to CWD
run: cp ../results/ql.sarif ./${{ matrix.folder }}.sarif
run: cp ../results/ql.sarif ./ql-for-ql.sarif
- name: Fixup the $scema in sarif # Until https://github.com/microsoft/sarif-vscode-extension/pull/436/ is part in a stable release
run: |
sed -i 's/\$schema.*/\$schema": "https:\/\/raw.githubusercontent.com\/oasis-tcs\/sarif-spec\/master\/Schemata\/sarif-schema-2.1.0",/' ql-for-ql.sarif
- name: Sarif as artifact
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.folder }}.sarif
path: ${{ matrix.folder }}.sarif
name: ql-for-ql.sarif
path: ql-for-ql.sarif
- name: Split out the sarif file into langs
run: |
mkdir split-sarif
node ./ql/scripts/split-sarif.js ql-for-ql.sarif split-sarif
- name: Upload langs as artifacts
uses: actions/upload-artifact@v3
with:
name: ql-for-ql-langs
path: split-sarif
retention-days: 1

View File

@@ -25,7 +25,7 @@ jobs:
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@aa93aea877e5fb8841bcb1193f672abf6e9f2980
uses: github/codeql-action/init@71a8b35ff4c80fcfcd05bc1cd932fe3c08f943ca
with:
languages: javascript # does not matter
- uses: actions/cache@v3
@@ -36,7 +36,7 @@ jobs:
ql/target
key: ${{ runner.os }}-qltest-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
- name: Build Extractor
run: cd ql; env "PATH=$PATH:`dirname ${CODEQL}`" ./create-extractor-pack.sh
run: cd ql; env "PATH=$PATH:`dirname ${CODEQL}`" ./scripts/create-extractor-pack.sh
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Checkout ${{ matrix.repo }}

View File

@@ -5,10 +5,12 @@ on:
branches: [main]
paths:
- "ql/**"
- codeql-workspace.yml
pull_request:
branches: [main]
paths:
- "ql/**"
- codeql-workspace.yml
env:
CARGO_TERM_COLOR: always
@@ -20,7 +22,7 @@ jobs:
- uses: actions/checkout@v3
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@aa93aea877e5fb8841bcb1193f672abf6e9f2980
uses: github/codeql-action/init@71a8b35ff4c80fcfcd05bc1cd932fe3c08f943ca
with:
languages: javascript # does not matter
- uses: actions/cache@v3
@@ -34,7 +36,7 @@ jobs:
run: |
cd ql;
codeqlpath=$(dirname ${{ steps.find-codeql.outputs.codeql-path }});
env "PATH=$PATH:$codeqlpath" ./create-extractor-pack.sh
env "PATH=$PATH:$codeqlpath" ./scripts/create-extractor-pack.sh
- name: Run QL tests
run: |
"${CODEQL}" test run --check-databases --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --search-path "${{ github.workspace }}/ql/extractor-pack" --consistency-queries ql/ql/consistency-queries ql/ql/test
@@ -42,7 +44,7 @@ jobs:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Check QL formatting
run: |
find ql/ql "(" -name "*.ql" -or -name "*.qll" ")" -print0 | xargs -0 "${CODEQL}" query format --check-only
find ql/ql/src "(" -name "*.ql" -or -name "*.qll" ")" -print0 | xargs -0 "${CODEQL}" query format --check-only
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Check QL compilation

View File

@@ -5,9 +5,12 @@ on:
branches:
- main
- 'rc/**'
tags:
- 'codeql-cli/*'
pull_request:
paths:
- '.github/workflows/query-list.yml'
- '.github/actions/fetch-codeql/action.yml'
- 'misc/scripts/generate-code-scanning-query-list.py'
jobs:
@@ -21,14 +24,12 @@ jobs:
with:
path: codeql
- name: Set up Python 3.8
uses: actions/setup-python@v3
uses: actions/setup-python@v4
with:
python-version: 3.8
- name: Download CodeQL CLI
# Look under the `codeql` directory, as this is where we checked out the `github/codeql` repo
uses: ./codeql/.github/actions/fetch-codeql
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Build code scanning query list
run: |
python codeql/misc/scripts/generate-code-scanning-query-list.py > code-scanning-query-list.csv

View File

@@ -5,6 +5,8 @@ on:
paths:
- "ruby/**"
- .github/workflows/ruby-build.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
- "rc/*"
@@ -12,6 +14,8 @@ on:
paths:
- "ruby/**"
- .github/workflows/ruby-build.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
- "rc/*"
@@ -88,19 +92,15 @@ jobs:
steps:
- uses: actions/checkout@v3
- name: Fetch CodeQL
run: |
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | grep -v beta | sort --version-sort | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-linux64.zip "$LATEST"
unzip -q codeql-linux64.zip
env:
GITHUB_TOKEN: ${{ github.token }}
uses: ./.github/actions/fetch-codeql
- name: Build Query Pack
run: |
codeql/codeql pack create ql/lib --output target/packs
codeql/codeql pack install ql/src
codeql/codeql pack create ql/src --output target/packs
codeql pack create ../shared/ssa --output target/packs
codeql pack create ql/lib --output target/packs
codeql pack install ql/src
codeql pack create ql/src --output target/packs
PACK_FOLDER=$(readlink -f target/packs/codeql/ruby-queries/*)
codeql/codeql generate query-help --format=sarifv2.1.0 --output="${PACK_FOLDER}/rules.sarif" ql/src
codeql generate query-help --format=sarifv2.1.0 --output="${PACK_FOLDER}/rules.sarif" ql/src
(cd ql/src; find queries \( -name '*.qhelp' -o -name '*.rb' -o -name '*.erb' \) -exec bash -c 'mkdir -p "'"${PACK_FOLDER}"'/$(dirname "{}")"' \; -exec cp "{}" "${PACK_FOLDER}/{}" \;)
- uses: actions/upload-artifact@v3
with:
@@ -177,19 +177,15 @@ jobs:
runs-on: ${{ matrix.os }}
needs: [package]
steps:
- uses: actions/checkout@v3
- name: Fetch CodeQL
uses: ./.github/actions/fetch-codeql
- uses: actions/checkout@v3
with:
repository: Shopify/example-ruby-app
ref: 67a0decc5eb550f3a9228eda53925c3afd40dfe9
- name: Fetch CodeQL
shell: bash
run: |
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | grep -v beta | sort --version-sort | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql.zip "$LATEST"
unzip -q codeql.zip
env:
GITHUB_TOKEN: ${{ github.token }}
working-directory: ${{ runner.temp }}
- name: Download Ruby bundle
uses: actions/download-artifact@v3
with:
@@ -213,12 +209,12 @@ jobs:
- name: Run QL test
shell: bash
run: |
"${{ runner.temp }}/codeql/codeql" test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" .
codeql test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" .
- name: Create database
shell: bash
run: |
"${{ runner.temp }}/codeql/codeql" database create --search-path "${{ runner.temp }}/ruby-bundle" --language ruby --source-root . ../database
codeql database create --search-path "${{ runner.temp }}/ruby-bundle" --language ruby --source-root . ../database
- name: Analyze database
shell: bash
run: |
"${{ runner.temp }}/codeql/codeql" database analyze --search-path "${{ runner.temp }}/ruby-bundle" --format=sarifv2.1.0 --output=out.sarif ../database ruby-code-scanning.qls
codeql database analyze --search-path "${{ runner.temp }}/ruby-bundle" --format=sarifv2.1.0 --output=out.sarif ../database ruby-code-scanning.qls

View File

@@ -5,6 +5,8 @@ on:
paths:
- "ruby/**"
- .github/workflows/ruby-qltest.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
- "rc/*"
@@ -12,6 +14,8 @@ on:
paths:
- "ruby/**"
- .github/workflows/ruby-qltest.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
- "rc/*"

View File

@@ -5,6 +5,7 @@ on:
paths:
- "swift/**"
- .github/workflows/swift-codegen.yml
- .github/actions/fetch-codeql/action.yml
branches:
- main
@@ -15,18 +16,22 @@ jobs:
- uses: actions/checkout@v3
- uses: ./.github/actions/fetch-codeql
- uses: bazelbuild/setup-bazelisk@v2
- uses: actions/setup-python@v3
- uses: pre-commit/action@v3.0.0
name: Check that python code is properly formatted
with:
extra_args: autopep8 --all-files
- name: Run unit tests
run: |
bazel test //swift/codegen/test --test_output=errors
- name: Check that QL generated code was checked in
run: |
bazel run //swift/codegen
git add swift
git diff --exit-code --stat HEAD
- uses: pre-commit/action@v3.0.0
name: Check that QL generated code was checked in
with:
extra_args: swift-codegen --all-files
- name: Generate C++ files
run: |
bazel run //swift/codegen:cppcodegen -- --cpp-output=$PWD/swift-generated-headers
bazel run //swift/codegen:codegen -- --generate=trap,cpp --cpp-output=$PWD/swift-generated-cpp-files
- uses: actions/upload-artifact@v3
with:
name: swift-generated-headers
path: swift-generated-headers/*.h
name: swift-generated-cpp-files
path: swift-generated-cpp-files/**

View File

@@ -0,0 +1,35 @@
name: "Swift: Run Integration Tests"
on:
pull_request:
paths:
- "swift/**"
- .github/workflows/swift-integration-tests.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
defaults:
run:
working-directory: swift
jobs:
integration-tests:
runs-on: ${{ matrix.os }}
strategy:
fail-fast: false
matrix:
os:
- ubuntu-20.04
# - macos-latest TODO
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/fetch-codeql
- uses: bazelbuild/setup-bazelisk@v2
- uses: actions/setup-python@v3
- name: Build Swift extractor
run: |
bazel run //swift:create-extractor-pack
- name: Run integration tests
run: |
python integration-tests/runner.py

View File

@@ -5,6 +5,8 @@ on:
paths:
- "swift/**"
- .github/workflows/swift-qltest.yml
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
branches:
- main
defaults:

View File

@@ -5,6 +5,7 @@ on:
paths:
- "*/ql/*/change-notes/**/*"
- ".github/workflows/validate-change-notes.yml"
- ".github/actions/fetch-codeql/action.yml"
branches:
- main
- "rc/*"
@@ -12,6 +13,7 @@ on:
paths:
- "*/ql/*/change-notes/**/*"
- ".github/workflows/validate-change-notes.yml"
- ".github/actions/fetch-codeql/action.yml"
jobs:
check-change-note:

6
.gitignore vendored
View File

@@ -55,3 +55,9 @@ go/tools/win64
go/tools/tokenizer.jar
go/main
# node_modules folders except in the JS test suite
node_modules/
!/javascript/ql/test/**/node_modules/
# Temporary folders for working with generated models
.model-temp

View File

@@ -15,6 +15,12 @@ repos:
- id: clang-format
files: ^swift/.*\.(h|c|cpp)$
- repo: https://github.com/pre-commit/mirrors-autopep8
rev: v1.6.0
hooks:
- id: autopep8
files: ^swift/codegen/.*\.py
- repo: local
hooks:
- id: codeql-format
@@ -25,7 +31,7 @@ repos:
- id: sync-files
name: Fix files required to be identical
files: \.(qll?|qhelp)$
files: \.(qll?|qhelp|swift)$
language: system
entry: python3 config/sync-files.py --latest
pass_filenames: false
@@ -40,7 +46,7 @@ repos:
name: Run Swift checked in code generation
files: ^swift/(codegen/|.*/generated/|ql/lib/(swift\.dbscheme$|codeql/swift/elements))
language: system
entry: bazel run //swift/codegen
entry: bazel run //swift/codegen -- --quiet
pass_filenames: false
- id: swift-codegen-unit-tests

View File

@@ -28,8 +28,8 @@
# QL for QL reviewers
/ql/ @github/codeql-ql-for-ql-reviewers
# Bazel
**/*.bazel @github/codeql-ci-reviewers
# Bazel (excluding BUILD.bazel files)
WORKSPACE.bazel @github/codeql-ci-reviewers
**/*.bzl @github/codeql-ci-reviewers
# Documentation etc
@@ -38,6 +38,8 @@
# Workflows
/.github/workflows/ @github/codeql-ci-reviewers
/.github/workflows/go-* @github/codeql-go
/.github/workflows/js-ml-tests.yml @github/codeql-ml-powered-queries-reviewers
/.github/workflows/ql-for-ql-* @github/codeql-ql-for-ql-reviewers
/.github/workflows/ruby-* @github/codeql-ruby
/.github/workflows/swift-* @github/codeql-c

33
codeql-workspace.yml Normal file
View File

@@ -0,0 +1,33 @@
provide:
- "*/ql/src/qlpack.yml"
- "*/ql/lib/qlpack.yml"
- "*/ql/test/qlpack.yml"
- "*/ql/examples/qlpack.yml"
- "*/ql/consistency-queries/qlpack.yml"
- "shared/*/qlpack.yml"
- "cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/qlpack.yml"
- "go/ql/config/legacy-support/qlpack.yml"
- "go/build/codeql-extractor-go/codeql-extractor.yml"
- "javascript/ql/experimental/adaptivethreatmodeling/lib/qlpack.yml"
# This pack is explicitly excluded from the workspace since most users
# will want to use a version of this pack from the package cache. Internal
# users can uncomment the following line and place a custom ML model
# in the corresponding pack to test a custom ML model within their local
# checkout.
# - "javascript/ql/experimental/adaptivethreatmodeling/model/qlpack.yml"
- "javascript/ql/experimental/adaptivethreatmodeling/modelbuilding/qlpack.yml"
- "javascript/ql/experimental/adaptivethreatmodeling/src/qlpack.yml"
- "csharp/ql/campaigns/Solorigate/lib/qlpack.yml"
- "csharp/ql/campaigns/Solorigate/src/qlpack.yml"
- "csharp/ql/campaigns/Solorigate/test/qlpack.yml"
- "misc/legacy-support/*/qlpack.yml"
- "misc/suite-helpers/qlpack.yml"
- "ruby/extractor-pack/codeql-extractor.yml"
- "swift/extractor-pack/codeql-extractor.yml"
- "ql/extractor-pack/codeql-extractor.ym"
versionPolicies:
default:
requireChangeNotes: true
committedPrereleaseSuffix: dev
committedVersion: nextPatchRelease

View File

@@ -22,13 +22,16 @@
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl3.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl5.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplForContentDataFlow.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl2.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl3.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImpl4.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImpl.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImpl2.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplForLibraries.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplForLibraries.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplForHttpClientLibraries.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImpl.qll"
],
"DataFlow Java/C++/C#/Python Common": [
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImplCommon.qll",
@@ -36,7 +39,8 @@
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImplCommon.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplCommon.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImplCommon.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplCommon.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplCommon.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImplCommon.qll"
],
"TaintTracking::Configuration Java/C++/C#/Python": [
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
@@ -57,7 +61,8 @@
"python/ql/lib/semmle/python/dataflow/new/internal/tainttracking3/TaintTrackingImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/tainttracking4/TaintTrackingImpl.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/tainttrackingforlibraries/TaintTrackingImpl.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/tainttrackingforlibraries/TaintTrackingImpl.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/tainttracking1/TaintTrackingImpl.qll"
],
"DataFlow Java/C++/C#/Python Consistency checks": [
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImplConsistency.qll",
@@ -65,12 +70,14 @@
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImplConsistency.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplConsistency.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImplConsistency.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplConsistency.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplConsistency.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImplConsistency.qll"
],
"DataFlow Java/C# Flow Summaries": [
"java/ql/lib/semmle/code/java/dataflow/internal/FlowSummaryImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/FlowSummaryImpl.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/FlowSummaryImpl.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/FlowSummaryImpl.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/FlowSummaryImpl.qll"
],
"SsaReadPosition Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/internal/rangeanalysis/SsaReadPositionCommon.qll",
@@ -385,7 +392,9 @@
"java/ql/test/TestUtilities/InlineExpectationsTest.qll",
"python/ql/test/TestUtilities/InlineExpectationsTest.qll",
"ruby/ql/test/TestUtilities/InlineExpectationsTest.qll",
"ql/ql/test/TestUtilities/InlineExpectationsTest.qll"
"ql/ql/test/TestUtilities/InlineExpectationsTest.qll",
"go/ql/test/TestUtilities/InlineExpectationsTest.qll",
"swift/ql/test/TestUtilities/InlineExpectationsTest.qll"
],
"C++ ExternalAPIs": [
"cpp/ql/src/Security/CWE/CWE-020/ExternalAPIs.qll",
@@ -446,19 +455,11 @@
"python/ql/src/Lexical/CommentedOutCodeReferences.inc.qhelp"
],
"IDE Contextual Queries": [
"cpp/ql/src/IDEContextual.qll",
"csharp/ql/src/IDEContextual.qll",
"java/ql/src/IDEContextual.qll",
"javascript/ql/src/IDEContextual.qll",
"python/ql/src/analysis/IDEContextual.qll"
],
"SSA C#": [
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/SsaImplCommon.qll",
"csharp/ql/lib/semmle/code/csharp/controlflow/internal/pressa/SsaImplCommon.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/basessa/SsaImplCommon.qll",
"csharp/ql/lib/semmle/code/cil/internal/SsaImplCommon.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/SsaImplCommon.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/SsaImplCommon.qll"
"cpp/ql/lib/IDEContextual.qll",
"csharp/ql/lib/IDEContextual.qll",
"java/ql/lib/IDEContextual.qll",
"javascript/ql/lib/IDEContextual.qll",
"python/ql/lib/analysis/IDEContextual.qll"
],
"CryptoAlgorithms Python/JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/CryptoAlgorithms.qll",
@@ -476,28 +477,39 @@
"ruby/ql/lib/codeql/ruby/security/internal/SensitiveDataHeuristics.qll"
],
"ReDoS Util Python/JS/Ruby/Java": [
"javascript/ql/lib/semmle/javascript/security/performance/ReDoSUtil.qll",
"python/ql/lib/semmle/python/security/performance/ReDoSUtil.qll",
"ruby/ql/lib/codeql/ruby/security/performance/ReDoSUtil.qll",
"java/ql/lib/semmle/code/java/security/performance/ReDoSUtil.qll"
"javascript/ql/lib/semmle/javascript/security/regexp/NfaUtils.qll",
"python/ql/lib/semmle/python/security/regexp/NfaUtils.qll",
"ruby/ql/lib/codeql/ruby/security/regexp/NfaUtils.qll",
"java/ql/lib/semmle/code/java/security/regexp/NfaUtils.qll"
],
"ReDoS Exponential Python/JS/Ruby/Java": [
"javascript/ql/lib/semmle/javascript/security/performance/ExponentialBackTracking.qll",
"python/ql/lib/semmle/python/security/performance/ExponentialBackTracking.qll",
"ruby/ql/lib/codeql/ruby/security/performance/ExponentialBackTracking.qll",
"java/ql/lib/semmle/code/java/security/performance/ExponentialBackTracking.qll"
"javascript/ql/lib/semmle/javascript/security/regexp/ExponentialBackTracking.qll",
"python/ql/lib/semmle/python/security/regexp/ExponentialBackTracking.qll",
"ruby/ql/lib/codeql/ruby/security/regexp/ExponentialBackTracking.qll",
"java/ql/lib/semmle/code/java/security/regexp/ExponentialBackTracking.qll"
],
"ReDoS Polynomial Python/JS/Ruby/Java": [
"javascript/ql/lib/semmle/javascript/security/performance/SuperlinearBackTracking.qll",
"python/ql/lib/semmle/python/security/performance/SuperlinearBackTracking.qll",
"ruby/ql/lib/codeql/ruby/security/performance/SuperlinearBackTracking.qll",
"java/ql/lib/semmle/code/java/security/performance/SuperlinearBackTracking.qll"
"javascript/ql/lib/semmle/javascript/security/regexp/SuperlinearBackTracking.qll",
"python/ql/lib/semmle/python/security/regexp/SuperlinearBackTracking.qll",
"ruby/ql/lib/codeql/ruby/security/regexp/SuperlinearBackTracking.qll",
"java/ql/lib/semmle/code/java/security/regexp/SuperlinearBackTracking.qll"
],
"RegexpMatching Python/JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/regexp/RegexpMatching.qll",
"python/ql/lib/semmle/python/security/regexp/RegexpMatching.qll",
"ruby/ql/lib/codeql/ruby/security/regexp/RegexpMatching.qll"
],
"BadTagFilterQuery Python/JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/BadTagFilterQuery.qll",
"python/ql/lib/semmle/python/security/BadTagFilterQuery.qll",
"ruby/ql/lib/codeql/ruby/security/BadTagFilterQuery.qll"
],
"OverlyLargeRange Python/JS/Ruby/Java": [
"javascript/ql/lib/semmle/javascript/security/OverlyLargeRangeQuery.qll",
"python/ql/lib/semmle/python/security/OverlyLargeRangeQuery.qll",
"ruby/ql/lib/codeql/ruby/security/OverlyLargeRangeQuery.qll",
"java/ql/lib/semmle/code/java/security/OverlyLargeRangeQuery.qll"
],
"CFG": [
"csharp/ql/lib/semmle/code/csharp/controlflow/internal/ControlFlowGraphImplShared.qll",
"ruby/ql/lib/codeql/ruby/controlflow/internal/ControlFlowGraphImplShared.qll",
@@ -519,7 +531,9 @@
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/AccessPathSyntax.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/AccessPathSyntax.qll",
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/AccessPathSyntax.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/AccessPathSyntax.qll"
"ruby/ql/lib/codeql/ruby/dataflow/internal/AccessPathSyntax.qll",
"python/ql/lib/semmle/python/frameworks/data/internal/AccessPathSyntax.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/AccessPathSyntax.qll"
],
"IncompleteUrlSubstringSanitization": [
"javascript/ql/src/Security/CWE-020/IncompleteUrlSubstringSanitization.qll",
@@ -537,7 +551,8 @@
],
"ApiGraphModels": [
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/ApiGraphModels.qll",
"ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModels.qll"
"ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModels.qll",
"python/ql/lib/semmle/python/frameworks/data/internal/ApiGraphModels.qll"
],
"TaintedFormatStringQuery Ruby/JS": [
"javascript/ql/lib/semmle/javascript/security/dataflow/TaintedFormatStringQuery.qll",
@@ -558,5 +573,25 @@
"Typo database": [
"javascript/ql/src/Expressions/TypoDatabase.qll",
"ql/ql/src/codeql_ql/style/TypoDatabase.qll"
],
"Swift declarations test file": [
"swift/ql/test/extractor-tests/declarations/declarations.swift",
"swift/ql/test/library-tests/ast/declarations.swift"
],
"Swift statements test file": [
"swift/ql/test/extractor-tests/statements/statements.swift",
"swift/ql/test/library-tests/ast/statements.swift"
],
"Swift expressions test file": [
"swift/ql/test/extractor-tests/expressions/expressions.swift",
"swift/ql/test/library-tests/ast/expressions.swift"
],
"Swift patterns test file": [
"swift/ql/test/extractor-tests/patterns/patterns.swift",
"swift/ql/test/library-tests/ast/patterns.swift"
],
"IncompleteMultiCharacterSanitization JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/IncompleteMultiCharacterSanitizationQuery.qll",
"ruby/ql/lib/codeql/ruby/security/IncompleteMultiCharacterSanitizationQuery.qll"
]
}
}

View File

@@ -299,7 +299,7 @@ namespace Semmle.Autobuild.Cpp.Tests
{
Actions.RunProcess[@"cmd.exe /C nuget restore C:\Project\test.sln -DisableParallelProcessing"] = 1;
Actions.RunProcess[@"cmd.exe /C C:\Project\.nuget\nuget.exe restore C:\Project\test.sln -DisableParallelProcessing"] = 0;
Actions.RunProcess[@"cmd.exe /C CALL ^""C:\Program Files ^(x86^)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat^"" && set Platform=&& type NUL && C:\odasa\tools\odasa index --auto msbuild C:\Project\test.sln /p:UseSharedCompilation=false /t:rebuild /p:Platform=""x86"" /p:Configuration=""Release"" /p:MvcBuildViews=true"] = 0;
Actions.RunProcess[@"cmd.exe /C CALL ^""C:\Program Files ^(x86^)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat^"" && set Platform=&& type NUL && C:\odasa\tools\odasa index --auto msbuild C:\Project\test.sln /t:rebuild /p:Platform=""x86"" /p:Configuration=""Release"" /p:MvcBuildViews=true"] = 0;
Actions.RunProcessOut[@"C:\Program Files (x86)\Microsoft Visual Studio\Installer\vswhere.exe -prerelease -legacy -property installationPath"] = "";
Actions.RunProcess[@"C:\Program Files (x86)\Microsoft Visual Studio\Installer\vswhere.exe -prerelease -legacy -property installationPath"] = 1;
Actions.RunProcess[@"C:\Program Files (x86)\Microsoft Visual Studio\Installer\vswhere.exe -prerelease -legacy -property installationVersion"] = 0;

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,3 @@
description: Add relation for tracking C++ braced initializers
compatibility: full
braced_initialisers.rel: delete

View File

@@ -0,0 +1,17 @@
class Expr extends @expr {
string toString() { none() }
}
class Location extends @location_expr {
string toString() { none() }
}
predicate isExprWithNewBuiltin(Expr expr) {
exists(int kind | exprs(expr, kind, _) | 330 <= kind and kind <= 334)
}
from Expr expr, int kind, int kind_new, Location location
where
exprs(expr, kind, location) and
if isExprWithNewBuiltin(expr) then kind_new = 0 else kind_new = kind
select expr, kind_new, location

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,3 @@
description: Add new builtin operations
compatibility: partial
exprs.rel: run exprs.qlo

View File

@@ -0,0 +1,17 @@
class AttributeArgument extends @attribute_arg {
string toString() { none() }
}
class Attribute extends @attribute {
string toString() { none() }
}
class LocationDefault extends @location_default {
string toString() { none() }
}
from AttributeArgument arg, int kind, Attribute attr, int index, LocationDefault location
where
attribute_args(arg, kind, attr, index, location) and
not arg instanceof @attribute_arg_constant_expr
select arg, kind, attr, index, location

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,4 @@
description: Support all constant attribute arguments
compatibility: backwards
attribute_arg_constant.rel: delete
attribute_args.rel: run attribute_args.qlo

View File

@@ -0,0 +1,17 @@
class Expr extends @expr {
string toString() { none() }
}
class Location extends @location_expr {
string toString() { none() }
}
predicate isExprWithNewBuiltin(Expr expr) {
exists(int kind | exprs(expr, kind, _) | 336 <= kind and kind <= 362)
}
from Expr expr, int kind, int kind_new, Location location
where
exprs(expr, kind, location) and
if isExprWithNewBuiltin(expr) then kind_new = 1 else kind_new = kind
select expr, kind_new, location

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,3 @@
description: Add new builtin operations
compatibility: partial
exprs.rel: run exprs.qlo

View File

@@ -0,0 +1,13 @@
class Expr extends @expr {
string toString() { none() }
}
class Location extends @location_expr {
string toString() { none() }
}
from Expr expr, int kind, int kind_new, Location location
where
exprs(expr, kind, location) and
if expr instanceof @blockassignexpr then kind_new = 0 else kind_new = kind
select expr, kind_new, location

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,3 @@
description: Support block assignment
compatibility: partial
exprs.rel: run exprs.qlo

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,3 @@
description: Add relation for orphaned local variables
compatibility: full
orphaned_variables.rel: delete

View File

@@ -1,3 +1,69 @@
## 0.3.4
### Deprecated APIs
* Many classes/predicates/modules with upper-case acronyms in their name have been renamed to follow our style-guide.
The old name still exists as a deprecated alias.
### New Features
* Added support for getting the link targets of global and namespace variables.
* Added a `BlockAssignExpr` class, which models a `memcpy`-like operation used in compiler generated copy/move constructors and assignment operations.
### Minor Analysis Improvements
* All deprecated predicates/classes/modules that have been deprecated for over a year have been deleted.
## 0.3.3
### New Features
* Added a predicate `getValueConstant` to `AttributeArgument` that yields the argument value as an `Expr` when the value is a constant expression.
* A new class predicate `MustFlowConfiguration::allowInterproceduralFlow` has been added to the `semmle.code.cpp.ir.dataflow.MustFlow` library. The new predicate can be overridden to disable interprocedural flow.
* Added subclasses of `BuiltInOperations` for `__builtin_bit_cast`, `__builtin_shuffle`, `__has_unique_object_representations`, `__is_aggregate`, and `__is_assignable`.
### Major Analysis Improvements
* The IR dataflow library now includes flow through global variables. This enables new findings in many scenarios.
## 0.3.2
### Bug Fixes
* Under certain circumstances a variable declaration that is not also a definition could be associated with a `Variable` that did not have the definition as a `VariableDeclarationEntry`. This is now fixed, and a unique `Variable` will exist that has both the declaration and the definition as a `VariableDeclarationEntry`.
## 0.3.1
### Minor Analysis Improvements
* `AnalysedExpr::isNullCheck` and `AnalysedExpr::isValidCheck` have been updated to handle variable accesses on the left-hand side of the C++ logical "and", and variable declarations in conditions.
## 0.3.0
### Deprecated APIs
* The `BarrierGuard` class has been deprecated. Such barriers and sanitizers can now instead be created using the new `BarrierGuard` parameterized module.
### Bug Fixes
* `UserType.getADeclarationEntry()` now yields all forward declarations when the user type is a `class`, `struct`, or `union`.
## 0.2.3
### New Features
* An `isBraced` predicate was added to the `Initializer` class which holds when a C++ braced initializer was used in the initialization.
## 0.2.2
### Deprecated APIs
* The `AnalysedString` class in the `StringAnalysis` module has been replaced with `AnalyzedString`, to follow our style guide. The old name still exists as a deprecated alias.
### New Features
* A `getInitialization` predicate was added to the `ConstexprIfStmt`, `IfStmt`, and `SwitchStmt` classes that yields the C++17-style initializer of the `if` or `switch` statement when it exists.
## 0.2.1
## 0.2.0

View File

@@ -1,4 +0,0 @@
---
category: feature
---
* A `getInitialization` predicate was added to the `ConstexprIfStmt`, `IfStmt`, and `SwitchStmt` classes that yields the C++17-style initializer of the `if` or `switch` statement when it exists.

View File

@@ -1,4 +0,0 @@
---
category: deprecated
---
* The `AnalysedString` class in the `StringAnalysis` module has been replaced with `AnalyzedString`, to follow our style guide. The old name still exists as a deprecated alias.

View File

@@ -0,0 +1,4 @@
---
category: feature
---
* Added subclasses of `BuiltInOperations` for `__is_same`, `__is_function`, `__is_layout_compatible`, `__is_pointer_interconvertible_base_of`, `__is_array`, `__array_rank`, `__array_extent`, `__is_arithmetic`, `__is_complete_type`, `__is_compound`, `__is_const`, `__is_floating_point`, `__is_fundamental`, `__is_integral`, `__is_lvalue_reference`, `__is_member_function_pointer`, `__is_member_object_pointer`, `__is_member_pointer`, `__is_object`, `__is_pointer`, `__is_reference`, `__is_rvalue_reference`, `__is_scalar`, `__is_signed`, `__is_unsigned`, `__is_void`, and `__is_volatile`.

View File

@@ -0,0 +1,4 @@
---
category: fix
---
* Fixed an issue in the taint tracking analysis where implicit reads were not allowed by default in sinks or additional taint steps that used flow states.

View File

@@ -0,0 +1,9 @@
## 0.2.2
### Deprecated APIs
* The `AnalysedString` class in the `StringAnalysis` module has been replaced with `AnalyzedString`, to follow our style guide. The old name still exists as a deprecated alias.
### New Features
* A `getInitialization` predicate was added to the `ConstexprIfStmt`, `IfStmt`, and `SwitchStmt` classes that yields the C++17-style initializer of the `if` or `switch` statement when it exists.

View File

@@ -0,0 +1,5 @@
## 0.2.3
### New Features
* An `isBraced` predicate was added to the `Initializer` class which holds when a C++ braced initializer was used in the initialization.

View File

@@ -0,0 +1,9 @@
## 0.3.0
### Deprecated APIs
* The `BarrierGuard` class has been deprecated. Such barriers and sanitizers can now instead be created using the new `BarrierGuard` parameterized module.
### Bug Fixes
* `UserType.getADeclarationEntry()` now yields all forward declarations when the user type is a `class`, `struct`, or `union`.

View File

@@ -0,0 +1,5 @@
## 0.3.1
### Minor Analysis Improvements
* `AnalysedExpr::isNullCheck` and `AnalysedExpr::isValidCheck` have been updated to handle variable accesses on the left-hand side of the C++ logical "and", and variable declarations in conditions.

View File

@@ -0,0 +1,5 @@
## 0.3.2
### Bug Fixes
* Under certain circumstances a variable declaration that is not also a definition could be associated with a `Variable` that did not have the definition as a `VariableDeclarationEntry`. This is now fixed, and a unique `Variable` will exist that has both the declaration and the definition as a `VariableDeclarationEntry`.

View File

@@ -0,0 +1,11 @@
## 0.3.3
### New Features
* Added a predicate `getValueConstant` to `AttributeArgument` that yields the argument value as an `Expr` when the value is a constant expression.
* A new class predicate `MustFlowConfiguration::allowInterproceduralFlow` has been added to the `semmle.code.cpp.ir.dataflow.MustFlow` library. The new predicate can be overridden to disable interprocedural flow.
* Added subclasses of `BuiltInOperations` for `__builtin_bit_cast`, `__builtin_shuffle`, `__has_unique_object_representations`, `__is_aggregate`, and `__is_assignable`.
### Major Analysis Improvements
* The IR dataflow library now includes flow through global variables. This enables new findings in many scenarios.

View File

@@ -0,0 +1,15 @@
## 0.3.4
### Deprecated APIs
* Many classes/predicates/modules with upper-case acronyms in their name have been renamed to follow our style-guide.
The old name still exists as a deprecated alias.
### New Features
* Added support for getting the link targets of global and namespace variables.
* Added a `BlockAssignExpr` class, which models a `memcpy`-like operation used in compiler generated copy/move constructors and assignment operations.
### Minor Analysis Improvements
* All deprecated predicates/classes/modules that have been deprecated for over a year have been deleted.

View File

@@ -1,2 +1,2 @@
---
lastReleaseVersion: 0.2.1
lastReleaseVersion: 0.3.4

View File

@@ -0,0 +1,136 @@
import semmle.code.cpp.ir.dataflow.DataFlow
import semmle.code.cpp.ir.dataflow.DataFlow2
module ProductFlow {
abstract class Configuration extends string {
bindingset[this]
Configuration() { any() }
/**
* Holds if `(source1, source2)` is a relevant data flow source.
*
* `source1` and `source2` must belong to the same callable.
*/
abstract predicate isSourcePair(DataFlow::Node source1, DataFlow::Node source2);
/**
* Holds if `(sink1, sink2)` is a relevant data flow sink.
*
* `sink1` and `sink2` must belong to the same callable.
*/
abstract predicate isSinkPair(DataFlow::Node sink1, DataFlow::Node sink2);
predicate hasFlowPath(
DataFlow::PathNode source1, DataFlow2::PathNode source2, DataFlow::PathNode sink1,
DataFlow2::PathNode sink2
) {
reachable(this, source1, source2, sink1, sink2)
}
}
private import Internal
module Internal {
class Conf1 extends DataFlow::Configuration {
Conf1() { this = "Conf1" }
override predicate isSource(DataFlow::Node source) {
exists(Configuration conf | conf.isSourcePair(source, _))
}
override predicate isSink(DataFlow::Node sink) {
exists(Configuration conf | conf.isSinkPair(sink, _))
}
}
class Conf2 extends DataFlow2::Configuration {
Conf2() { this = "Conf2" }
override predicate isSource(DataFlow::Node source) {
exists(Configuration conf, DataFlow::Node source1 |
conf.isSourcePair(source1, source) and
any(Conf1 c).hasFlow(source1, _)
)
}
override predicate isSink(DataFlow::Node sink) {
exists(Configuration conf, DataFlow::Node sink1 |
conf.isSinkPair(sink1, sink) and any(Conf1 c).hasFlow(_, sink1)
)
}
}
}
private predicate reachableInterprocEntry(
Configuration conf, DataFlow::PathNode source1, DataFlow2::PathNode source2,
DataFlow::PathNode node1, DataFlow2::PathNode node2
) {
conf.isSourcePair(node1.getNode(), node2.getNode()) and
node1 = source1 and
node2 = source2
or
exists(
DataFlow::PathNode midEntry1, DataFlow2::PathNode midEntry2, DataFlow::PathNode midExit1,
DataFlow2::PathNode midExit2
|
reachableInterprocEntry(conf, source1, source2, midEntry1, midEntry2) and
interprocEdgePair(midExit1, midExit2, node1, node2) and
localPathStep1*(midEntry1, midExit1) and
localPathStep2*(midEntry2, midExit2)
)
}
private predicate localPathStep1(DataFlow::PathNode pred, DataFlow::PathNode succ) {
DataFlow::PathGraph::edges(pred, succ) and
pragma[only_bind_out](pred.getNode().getEnclosingCallable()) =
pragma[only_bind_out](succ.getNode().getEnclosingCallable())
}
private predicate localPathStep2(DataFlow2::PathNode pred, DataFlow2::PathNode succ) {
DataFlow2::PathGraph::edges(pred, succ) and
pragma[only_bind_out](pred.getNode().getEnclosingCallable()) =
pragma[only_bind_out](succ.getNode().getEnclosingCallable())
}
pragma[nomagic]
private predicate interprocEdge1(
Declaration predDecl, Declaration succDecl, DataFlow::PathNode pred1, DataFlow::PathNode succ1
) {
DataFlow::PathGraph::edges(pred1, succ1) and
predDecl != succDecl and
pred1.getNode().getEnclosingCallable() = predDecl and
succ1.getNode().getEnclosingCallable() = succDecl
}
pragma[nomagic]
private predicate interprocEdge2(
Declaration predDecl, Declaration succDecl, DataFlow2::PathNode pred2, DataFlow2::PathNode succ2
) {
DataFlow2::PathGraph::edges(pred2, succ2) and
predDecl != succDecl and
pred2.getNode().getEnclosingCallable() = predDecl and
succ2.getNode().getEnclosingCallable() = succDecl
}
private predicate interprocEdgePair(
DataFlow::PathNode pred1, DataFlow2::PathNode pred2, DataFlow::PathNode succ1,
DataFlow2::PathNode succ2
) {
exists(Declaration predDecl, Declaration succDecl |
interprocEdge1(predDecl, succDecl, pred1, succ1) and
interprocEdge2(predDecl, succDecl, pred2, succ2)
)
}
private predicate reachable(
Configuration conf, DataFlow::PathNode source1, DataFlow2::PathNode source2,
DataFlow::PathNode sink1, DataFlow2::PathNode sink2
) {
exists(DataFlow::PathNode mid1, DataFlow2::PathNode mid2 |
reachableInterprocEntry(conf, source1, source2, mid1, mid2) and
conf.isSinkPair(sink1.getNode(), sink2.getNode()) and
localPathStep1*(mid1, sink1) and
localPathStep2*(mid2, sink2)
)
}
}

View File

@@ -0,0 +1,26 @@
/**
* Provides a library for local (intra-procedural) and global (inter-procedural)
* data flow analysis: deciding whether data can flow from a _source_ to a
* _sink_. This library differs from the one in `semmle.code.cpp.dataflow` in that
* this library uses the IR (Intermediate Representation) library, which provides
* a more precise semantic representation of the program, whereas the other dataflow
* library uses the more syntax-oriented ASTs. This library should provide more accurate
* results than the AST-based library in most scenarios.
*
* Unless configured otherwise, _flow_ means that the exact value of
* the source may reach the sink. We do not track flow across pointer
* dereferences or array indexing.
*
* To use global (interprocedural) data flow, extend the class
* `DataFlow::Configuration` as documented on that class. To use local
* (intraprocedural) data flow between expressions, call
* `DataFlow::localExprFlow`. For more general cases of local data flow, call
* `DataFlow::localFlow` or `DataFlow::localFlowStep` with arguments of type
* `DataFlow::Node`.
*/
import cpp
module DataFlow {
import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowImpl
}

View File

@@ -0,0 +1,16 @@
/**
* Provides a `DataFlow2` module, which is a copy of the `DataFlow` module. Use
* this class when data-flow configurations must depend on each other. Two
* classes extending `DataFlow::Configuration` should never depend on each
* other, but one of them should instead depend on a
* `DataFlow2::Configuration`, a `DataFlow3::Configuration`, or a
* `DataFlow4::Configuration`.
*
* See `semmle.code.cpp.ir.dataflow.DataFlow` for the full documentation.
*/
import cpp
module DataFlow2 {
import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowImpl2
}

View File

@@ -0,0 +1,16 @@
/**
* Provides a `DataFlow3` module, which is a copy of the `DataFlow` module. Use
* this class when data-flow configurations must depend on each other. Two
* classes extending `DataFlow::Configuration` should never depend on each
* other, but one of them should instead depend on a
* `DataFlow2::Configuration`, a `DataFlow3::Configuration`, or a
* `DataFlow4::Configuration`.
*
* See `semmle.code.cpp.ir.dataflow.DataFlow` for the full documentation.
*/
import cpp
module DataFlow3 {
import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowImpl3
}

View File

@@ -0,0 +1,16 @@
/**
* Provides a `DataFlow4` module, which is a copy of the `DataFlow` module. Use
* this class when data-flow configurations must depend on each other. Two
* classes extending `DataFlow::Configuration` should never depend on each
* other, but one of them should instead depend on a
* `DataFlow2::Configuration`, a `DataFlow3::Configuration`, or a
* `DataFlow4::Configuration`.
*
* See `semmle.code.cpp.ir.dataflow.DataFlow` for the full documentation.
*/
import cpp
module DataFlow4 {
import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowImpl4
}

View File

@@ -0,0 +1,23 @@
/**
* Provides a predicate for non-contextual virtual dispatch and function
* pointer resolution.
*/
import cpp
private import semmle.code.cpp.ir.ValueNumbering
private import internal.DataFlowDispatch
private import semmle.code.cpp.ir.IR
/**
* Resolve potential target function(s) for `call`.
*
* If `call` is a call through a function pointer (`ExprCall`) or its target is
* a virtual member function, simple data flow analysis is performed in order
* to identify the possible target(s).
*/
Function resolveCall(Call call) {
exists(CallInstruction callInstruction |
callInstruction.getAst() = call and
result = viableCallable(callInstruction)
)
}

View File

@@ -0,0 +1,23 @@
/**
* Provides classes for performing local (intra-procedural) and
* global (inter-procedural) taint-tracking analyses.
*
* We define _taint propagation_ informally to mean that a substantial part of
* the information from the source is preserved at the sink. For example, taint
* propagates from `x` to `x + 100`, but it does not propagate from `x` to `x >
* 100` since we consider a single bit of information to be too little.
*
* To use global (interprocedural) taint tracking, extend the class
* `TaintTracking::Configuration` as documented on that class. To use local
* (intraprocedural) taint tracking between expressions, call
* `TaintTracking::localExprTaint`. For more general cases of local taint
* tracking, call `TaintTracking::localTaint` or
* `TaintTracking::localTaintStep` with arguments of type `DataFlow::Node`.
*/
import semmle.code.cpp.ir.dataflow.DataFlow
import semmle.code.cpp.ir.dataflow.DataFlow2
module TaintTracking {
import experimental.semmle.code.cpp.ir.dataflow.internal.tainttracking1.TaintTrackingImpl
}

View File

@@ -0,0 +1,15 @@
/**
* Provides a `TaintTracking2` module, which is a copy of the `TaintTracking`
* module. Use this class when data-flow configurations or taint-tracking
* configurations must depend on each other. Two classes extending
* `DataFlow::Configuration` should never depend on each other, but one of them
* should instead depend on a `DataFlow2::Configuration`, a
* `DataFlow3::Configuration`, or a `DataFlow4::Configuration`. The
* `TaintTracking::Configuration` class extends `DataFlow::Configuration`, and
* `TaintTracking2::Configuration` extends `DataFlow2::Configuration`.
*
* See `semmle.code.cpp.ir.dataflow.TaintTracking` for the full documentation.
*/
module TaintTracking2 {
import experimental.semmle.code.cpp.ir.dataflow.internal.tainttracking2.TaintTrackingImpl
}

View File

@@ -0,0 +1,15 @@
/**
* Provides a `TaintTracking3` module, which is a copy of the `TaintTracking`
* module. Use this class when data-flow configurations or taint-tracking
* configurations must depend on each other. Two classes extending
* `DataFlow::Configuration` should never depend on each other, but one of them
* should instead depend on a `DataFlow2::Configuration`, a
* `DataFlow3::Configuration`, or a `DataFlow4::Configuration`. The
* `TaintTracking::Configuration` class extends `DataFlow::Configuration`, and
* `TaintTracking2::Configuration` extends `DataFlow2::Configuration`.
*
* See `semmle.code.cpp.ir.dataflow.TaintTracking` for the full documentation.
*/
module TaintTracking3 {
import experimental.semmle.code.cpp.ir.dataflow.internal.tainttracking3.TaintTrackingImpl
}

View File

@@ -0,0 +1,273 @@
private import cpp
private import semmle.code.cpp.ir.IR
private import experimental.semmle.code.cpp.ir.dataflow.DataFlow
private import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
private import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import DataFlowImplCommon as DataFlowImplCommon
/**
* Gets a function that might be called by `call`.
*/
cached
Function viableCallable(CallInstruction call) {
DataFlowImplCommon::forceCachingInSameStage() and
result = call.getStaticCallTarget()
or
// If the target of the call does not have a body in the snapshot, it might
// be because the target is just a header declaration, and the real target
// will be determined at run time when the caller and callee are linked
// together by the operating system's dynamic linker. In case a _unique_
// function with the right signature is present in the database, we return
// that as a potential callee.
exists(string qualifiedName, int nparams |
callSignatureWithoutBody(qualifiedName, nparams, call) and
functionSignatureWithBody(qualifiedName, nparams, result) and
strictcount(Function other | functionSignatureWithBody(qualifiedName, nparams, other)) = 1
)
or
// Virtual dispatch
result = call.(VirtualDispatch::DataSensitiveCall).resolve()
}
/**
* Provides virtual dispatch support compatible with the original
* implementation of `semmle.code.cpp.security.TaintTracking`.
*/
private module VirtualDispatch {
/** A call that may dispatch differently depending on the qualifier value. */
abstract class DataSensitiveCall extends DataFlowCall {
/**
* Gets the node whose value determines the target of this call. This node
* could be the qualifier of a virtual dispatch or the function-pointer
* expression in a call to a function pointer. What they have in common is
* that we need to find out which data flows there, and then it's up to the
* `resolve` predicate to stitch that information together and resolve the
* call.
*/
abstract DataFlow::Node getDispatchValue();
/** Gets a candidate target for this call. */
abstract Function resolve();
/**
* Whether `src` can flow to this call.
*
* Searches backwards from `getDispatchValue()` to `src`. The `allowFromArg`
* parameter is true when the search is allowed to continue backwards into
* a parameter; non-recursive callers should pass `_` for `allowFromArg`.
*/
predicate flowsFrom(DataFlow::Node src, boolean allowFromArg) {
src = this.getDispatchValue() and allowFromArg = true
or
exists(DataFlow::Node other, boolean allowOtherFromArg |
this.flowsFrom(other, allowOtherFromArg)
|
// Call argument
exists(DataFlowCall call, Position i |
other
.(DataFlow::ParameterNode)
.isParameterOf(pragma[only_bind_into](call).getStaticCallTarget(), i) and
src.(ArgumentNode).argumentOf(call, pragma[only_bind_into](pragma[only_bind_out](i)))
) and
allowOtherFromArg = true and
allowFromArg = true
or
// Call return
exists(DataFlowCall call, ReturnKind returnKind |
other = getAnOutNode(call, returnKind) and
returnNodeWithKindAndEnclosingCallable(src, returnKind, call.getStaticCallTarget())
) and
allowFromArg = false
or
// Local flow
DataFlow::localFlowStep(src, other) and
allowFromArg = allowOtherFromArg
or
// Flow from global variable to load.
exists(LoadInstruction load, GlobalOrNamespaceVariable var |
var = src.asVariable() and
other.asInstruction() = load and
addressOfGlobal(load.getSourceAddress(), var) and
// The `allowFromArg` concept doesn't play a role when `src` is a
// global variable, so we just set it to a single arbitrary value for
// performance.
allowFromArg = true
)
or
// Flow from store to global variable.
exists(StoreInstruction store, GlobalOrNamespaceVariable var |
var = other.asVariable() and
store = src.asInstruction() and
storeIntoGlobal(store, var) and
// Setting `allowFromArg` to `true` like in the base case means we
// treat a store to a global variable like the dispatch itself: flow
// may come from anywhere.
allowFromArg = true
)
)
}
}
pragma[noinline]
private predicate storeIntoGlobal(StoreInstruction store, GlobalOrNamespaceVariable var) {
addressOfGlobal(store.getDestinationAddress(), var)
}
/** Holds if `addressInstr` is an instruction that produces the address of `var`. */
private predicate addressOfGlobal(Instruction addressInstr, GlobalOrNamespaceVariable var) {
// Access directly to the global variable
addressInstr.(VariableAddressInstruction).getAstVariable() = var
or
// Access to a field on a global union
exists(FieldAddressInstruction fa |
fa = addressInstr and
fa.getObjectAddress().(VariableAddressInstruction).getAstVariable() = var and
fa.getField().getDeclaringType() instanceof Union
)
}
/**
* A ReturnNode with its ReturnKind and its enclosing callable.
*
* Used to fix a join ordering issue in flowsFrom.
*/
pragma[noinline]
private predicate returnNodeWithKindAndEnclosingCallable(
ReturnNode node, ReturnKind kind, DataFlowCallable callable
) {
node.getKind() = kind and
node.getEnclosingCallable() = callable
}
/** Call through a function pointer. */
private class DataSensitiveExprCall extends DataSensitiveCall {
DataSensitiveExprCall() { not exists(this.getStaticCallTarget()) }
override DataFlow::Node getDispatchValue() { result.asInstruction() = this.getCallTarget() }
override Function resolve() {
exists(FunctionInstruction fi |
this.flowsFrom(DataFlow::instructionNode(fi), _) and
result = fi.getFunctionSymbol()
) and
(
this.getNumberOfArguments() <= result.getEffectiveNumberOfParameters() and
this.getNumberOfArguments() >= result.getEffectiveNumberOfParameters()
or
result.isVarargs()
)
}
}
/** Call to a virtual function. */
private class DataSensitiveOverriddenFunctionCall extends DataSensitiveCall {
DataSensitiveOverriddenFunctionCall() {
exists(this.getStaticCallTarget().(VirtualFunction).getAnOverridingFunction())
}
override DataFlow::Node getDispatchValue() { result.asInstruction() = this.getThisArgument() }
override MemberFunction resolve() {
exists(Class overridingClass |
this.overrideMayAffectCall(overridingClass, result) and
this.hasFlowFromCastFrom(overridingClass)
)
}
/**
* Holds if `this` is a virtual function call whose static target is
* overridden by `overridingFunction` in `overridingClass`.
*/
pragma[noinline]
private predicate overrideMayAffectCall(Class overridingClass, MemberFunction overridingFunction) {
overridingFunction.getAnOverriddenFunction+() = this.getStaticCallTarget().(VirtualFunction) and
overridingFunction.getDeclaringType() = overridingClass
}
/**
* Holds if the qualifier of `this` has flow from an upcast from
* `derivedClass`.
*/
pragma[noinline]
private predicate hasFlowFromCastFrom(Class derivedClass) {
exists(ConvertToBaseInstruction toBase |
this.flowsFrom(DataFlow::instructionNode(toBase), _) and
derivedClass = toBase.getDerivedClass()
)
}
}
}
/**
* Holds if `f` is a function with a body that has name `qualifiedName` and
* `nparams` parameter count. See `functionSignature`.
*/
private predicate functionSignatureWithBody(string qualifiedName, int nparams, Function f) {
functionSignature(f, qualifiedName, nparams) and
exists(f.getBlock())
}
/**
* Holds if the target of `call` is a function _with no definition_ that has
* name `qualifiedName` and `nparams` parameter count. See `functionSignature`.
*/
pragma[noinline]
private predicate callSignatureWithoutBody(string qualifiedName, int nparams, CallInstruction call) {
exists(Function target |
target = call.getStaticCallTarget() and
not exists(target.getBlock()) and
functionSignature(target, qualifiedName, nparams)
)
}
/**
* Holds if `f` has name `qualifiedName` and `nparams` parameter count. This is
* an approximation of its signature for the purpose of matching functions that
* might be the same across link targets.
*/
private predicate functionSignature(Function f, string qualifiedName, int nparams) {
qualifiedName = f.getQualifiedName() and
nparams = f.getNumberOfParameters() and
not f.isStatic()
}
/**
* Holds if the set of viable implementations that can be called by `call`
* might be improved by knowing the call context.
*/
predicate mayBenefitFromCallContext(CallInstruction call, Function f) {
mayBenefitFromCallContext(call, f, _)
}
/**
* Holds if `call` is a call through a function pointer, and the pointer
* value is given as the `arg`'th argument to `f`.
*/
private predicate mayBenefitFromCallContext(
VirtualDispatch::DataSensitiveCall call, Function f, int arg
) {
f = pragma[only_bind_out](call).getEnclosingCallable() and
exists(InitializeParameterInstruction init |
not exists(call.getStaticCallTarget()) and
init.getEnclosingFunction() = f and
call.flowsFrom(DataFlow::instructionNode(init), _) and
init.getParameter().getIndex() = arg
)
}
/**
* Gets a viable dispatch target of `call` in the context `ctx`. This is
* restricted to those `call`s for which a context might make a difference.
*/
Function viableImplInCallContext(CallInstruction call, CallInstruction ctx) {
result = viableCallable(call) and
exists(int i, Function f |
mayBenefitFromCallContext(pragma[only_bind_into](call), f, i) and
f = ctx.getStaticCallTarget() and
result = ctx.getArgument(i).getUnconvertedResultExpression().(FunctionAccess).getTarget()
)
}
/** Holds if arguments at position `apos` match parameters at position `ppos`. */
pragma[inline]
predicate parameterMatch(ParameterPosition ppos, ArgumentPosition apos) { ppos = apos }

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,212 @@
/**
* Provides consistency queries for checking invariants in the language-specific
* data-flow classes and predicates.
*/
private import DataFlowImplSpecific::Private
private import DataFlowImplSpecific::Public
private import tainttracking1.TaintTrackingParameter::Private
private import tainttracking1.TaintTrackingParameter::Public
module Consistency {
private newtype TConsistencyConfiguration = MkConsistencyConfiguration()
/** A class for configuring the consistency queries. */
class ConsistencyConfiguration extends TConsistencyConfiguration {
string toString() { none() }
/** Holds if `n` should be excluded from the consistency test `uniqueEnclosingCallable`. */
predicate uniqueEnclosingCallableExclude(Node n) { none() }
/** Holds if `n` should be excluded from the consistency test `uniqueNodeLocation`. */
predicate uniqueNodeLocationExclude(Node n) { none() }
/** Holds if `n` should be excluded from the consistency test `missingLocation`. */
predicate missingLocationExclude(Node n) { none() }
/** Holds if `n` should be excluded from the consistency test `postWithInFlow`. */
predicate postWithInFlowExclude(Node n) { none() }
/** Holds if `n` should be excluded from the consistency test `argHasPostUpdate`. */
predicate argHasPostUpdateExclude(ArgumentNode n) { none() }
/** Holds if `n` should be excluded from the consistency test `reverseRead`. */
predicate reverseReadExclude(Node n) { none() }
}
private class RelevantNode extends Node {
RelevantNode() {
this instanceof ArgumentNode or
this instanceof ParameterNode or
this instanceof ReturnNode or
this = getAnOutNode(_, _) or
simpleLocalFlowStep(this, _) or
simpleLocalFlowStep(_, this) or
jumpStep(this, _) or
jumpStep(_, this) or
storeStep(this, _, _) or
storeStep(_, _, this) or
readStep(this, _, _) or
readStep(_, _, this) or
defaultAdditionalTaintStep(this, _) or
defaultAdditionalTaintStep(_, this)
}
}
query predicate uniqueEnclosingCallable(Node n, string msg) {
exists(int c |
n instanceof RelevantNode and
c = count(nodeGetEnclosingCallable(n)) and
c != 1 and
not any(ConsistencyConfiguration conf).uniqueEnclosingCallableExclude(n) and
msg = "Node should have one enclosing callable but has " + c + "."
)
}
query predicate uniqueType(Node n, string msg) {
exists(int c |
n instanceof RelevantNode and
c = count(getNodeType(n)) and
c != 1 and
msg = "Node should have one type but has " + c + "."
)
}
query predicate uniqueNodeLocation(Node n, string msg) {
exists(int c |
c =
count(string filepath, int startline, int startcolumn, int endline, int endcolumn |
n.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
) and
c != 1 and
not any(ConsistencyConfiguration conf).uniqueNodeLocationExclude(n) and
msg = "Node should have one location but has " + c + "."
)
}
query predicate missingLocation(string msg) {
exists(int c |
c =
strictcount(Node n |
not exists(string filepath, int startline, int startcolumn, int endline, int endcolumn |
n.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
) and
not any(ConsistencyConfiguration conf).missingLocationExclude(n)
) and
msg = "Nodes without location: " + c
)
}
query predicate uniqueNodeToString(Node n, string msg) {
exists(int c |
c = count(n.toString()) and
c != 1 and
msg = "Node should have one toString but has " + c + "."
)
}
query predicate missingToString(string msg) {
exists(int c |
c = strictcount(Node n | not exists(n.toString())) and
msg = "Nodes without toString: " + c
)
}
query predicate parameterCallable(ParameterNode p, string msg) {
exists(DataFlowCallable c | isParameterNode(p, c, _) and c != nodeGetEnclosingCallable(p)) and
msg = "Callable mismatch for parameter."
}
query predicate localFlowIsLocal(Node n1, Node n2, string msg) {
simpleLocalFlowStep(n1, n2) and
nodeGetEnclosingCallable(n1) != nodeGetEnclosingCallable(n2) and
msg = "Local flow step does not preserve enclosing callable."
}
private DataFlowType typeRepr() { result = getNodeType(_) }
query predicate compatibleTypesReflexive(DataFlowType t, string msg) {
t = typeRepr() and
not compatibleTypes(t, t) and
msg = "Type compatibility predicate is not reflexive."
}
query predicate unreachableNodeCCtx(Node n, DataFlowCall call, string msg) {
isUnreachableInCall(n, call) and
exists(DataFlowCallable c |
c = nodeGetEnclosingCallable(n) and
not viableCallable(call) = c
) and
msg = "Call context for isUnreachableInCall is inconsistent with call graph."
}
query predicate localCallNodes(DataFlowCall call, Node n, string msg) {
(
n = getAnOutNode(call, _) and
msg = "OutNode and call does not share enclosing callable."
or
n.(ArgumentNode).argumentOf(call, _) and
msg = "ArgumentNode and call does not share enclosing callable."
) and
nodeGetEnclosingCallable(n) != call.getEnclosingCallable()
}
// This predicate helps the compiler forget that in some languages
// it is impossible for a result of `getPreUpdateNode` to be an
// instance of `PostUpdateNode`.
private Node getPre(PostUpdateNode n) {
result = n.getPreUpdateNode()
or
none()
}
query predicate postIsNotPre(PostUpdateNode n, string msg) {
getPre(n) = n and
msg = "PostUpdateNode should not equal its pre-update node."
}
query predicate postHasUniquePre(PostUpdateNode n, string msg) {
exists(int c |
c = count(n.getPreUpdateNode()) and
c != 1 and
msg = "PostUpdateNode should have one pre-update node but has " + c + "."
)
}
query predicate uniquePostUpdate(Node n, string msg) {
1 < strictcount(PostUpdateNode post | post.getPreUpdateNode() = n) and
msg = "Node has multiple PostUpdateNodes."
}
query predicate postIsInSameCallable(PostUpdateNode n, string msg) {
nodeGetEnclosingCallable(n) != nodeGetEnclosingCallable(n.getPreUpdateNode()) and
msg = "PostUpdateNode does not share callable with its pre-update node."
}
private predicate hasPost(Node n) { exists(PostUpdateNode post | post.getPreUpdateNode() = n) }
query predicate reverseRead(Node n, string msg) {
exists(Node n2 | readStep(n, _, n2) and hasPost(n2) and not hasPost(n)) and
not any(ConsistencyConfiguration conf).reverseReadExclude(n) and
msg = "Origin of readStep is missing a PostUpdateNode."
}
query predicate argHasPostUpdate(ArgumentNode n, string msg) {
not hasPost(n) and
not any(ConsistencyConfiguration c).argHasPostUpdateExclude(n) and
msg = "ArgumentNode is missing PostUpdateNode."
}
// This predicate helps the compiler forget that in some languages
// it is impossible for a `PostUpdateNode` to be the target of
// `simpleLocalFlowStep`.
private predicate isPostUpdateNode(Node n) { n instanceof PostUpdateNode or none() }
query predicate postWithInFlow(Node n, string msg) {
isPostUpdateNode(n) and
not clearsContent(n, _) and
simpleLocalFlowStep(_, n) and
not any(ConsistencyConfiguration c).postWithInFlowExclude(n) and
msg = "PostUpdateNode should not be the target of local flow."
}
}

View File

@@ -0,0 +1,11 @@
/**
* Provides IR-specific definitions for use in the data flow library.
*/
module Private {
import DataFlowPrivate
import DataFlowDispatch
}
module Public {
import DataFlowUtil
}

View File

@@ -0,0 +1,560 @@
private import cpp as Cpp
private import DataFlowUtil
private import semmle.code.cpp.ir.IR
private import DataFlowDispatch
private import DataFlowImplConsistency
private import semmle.code.cpp.ir.internal.IRCppLanguage
private import SsaInternals as Ssa
/** Gets the callable in which this node occurs. */
DataFlowCallable nodeGetEnclosingCallable(Node n) { result = n.getEnclosingCallable() }
/** Holds if `p` is a `ParameterNode` of `c` with position `pos`. */
predicate isParameterNode(ParameterNode p, DataFlowCallable c, ParameterPosition pos) {
p.isParameterOf(c, pos)
}
/** Holds if `arg` is an `ArgumentNode` of `c` with position `pos`. */
predicate isArgumentNode(ArgumentNode arg, DataFlowCall c, ArgumentPosition pos) {
arg.argumentOf(c, pos)
}
/**
* A data flow node that occurs as the argument of a call and is passed as-is
* to the callable. Instance arguments (`this` pointer) and read side effects
* on parameters are also included.
*/
abstract class ArgumentNode extends Node {
/**
* Holds if this argument occurs at the given position in the given call.
* The instance argument is considered to have index `-1`.
*/
abstract predicate argumentOf(DataFlowCall call, ArgumentPosition pos);
/** Gets the call in which this node is an argument. */
DataFlowCall getCall() { this.argumentOf(result, _) }
}
/**
* A data flow node that occurs as the argument to a call, or an
* implicit `this` pointer argument.
*/
private class PrimaryArgumentNode extends ArgumentNode, OperandNode {
override ArgumentOperand op;
PrimaryArgumentNode() { exists(CallInstruction call | op = call.getAnArgumentOperand()) }
override predicate argumentOf(DataFlowCall call, ArgumentPosition pos) {
op = call.getArgumentOperand(pos.(DirectPosition).getIndex())
}
override string toStringImpl() { result = argumentOperandToString(op) }
}
private string argumentOperandToString(ArgumentOperand op) {
exists(Expr unconverted |
unconverted = op.getDef().getUnconvertedResultExpression() and
result = unconverted.toString()
)
or
// Certain instructions don't map to an unconverted result expression. For these cases
// we fall back to a simpler naming scheme. This can happen in IR-generated constructors.
not exists(op.getDef().getUnconvertedResultExpression()) and
(
result = "Argument " + op.(PositionalArgumentOperand).getIndex()
or
op instanceof ThisArgumentOperand and result = "Argument this"
)
}
private class SideEffectArgumentNode extends ArgumentNode, SideEffectOperandNode {
override predicate argumentOf(DataFlowCall dfCall, ArgumentPosition pos) {
this.getCallInstruction() = dfCall and
pos.(IndirectionPosition).getArgumentIndex() = this.getArgumentIndex() and
pos.(IndirectionPosition).getIndirectionIndex() = super.getIndirectionIndex()
}
override string toStringImpl() {
result = argumentOperandToString(this.getAddressOperand()) + " indirection"
}
}
/** A parameter position represented by an integer. */
class ParameterPosition = Position;
/** An argument position represented by an integer. */
class ArgumentPosition = Position;
class Position extends TPosition {
abstract string toString();
}
class DirectPosition extends Position, TDirectPosition {
int index;
DirectPosition() { this = TDirectPosition(index) }
override string toString() { if index = -1 then result = "this" else result = index.toString() }
int getIndex() { result = index }
}
class IndirectionPosition extends Position, TIndirectionPosition {
int argumentIndex;
int indirectionIndex;
IndirectionPosition() { this = TIndirectionPosition(argumentIndex, indirectionIndex) }
override string toString() {
if argumentIndex = -1
then if indirectionIndex > 0 then result = "this indirection" else result = "this"
else
if indirectionIndex > 0
then result = argumentIndex.toString() + " indirection"
else result = argumentIndex.toString()
}
int getArgumentIndex() { result = argumentIndex }
int getIndirectionIndex() { result = indirectionIndex }
}
newtype TPosition =
TDirectPosition(int index) { exists(any(CallInstruction c).getArgument(index)) } or
TIndirectionPosition(int argumentIndex, int indirectionIndex) {
hasOperandAndIndex(_, any(CallInstruction call).getArgumentOperand(argumentIndex),
indirectionIndex)
}
private newtype TReturnKind =
TNormalReturnKind(int index) {
exists(IndirectReturnNode return |
return.getAddressOperand() = any(ReturnValueInstruction r).getReturnAddressOperand() and
index = return.getIndirectionIndex() - 1 // We subtract one because the return loads the value.
)
} or
TIndirectReturnKind(int argumentIndex, int indirectionIndex) {
exists(IndirectReturnNode return, ReturnIndirectionInstruction returnInd |
returnInd.hasIndex(argumentIndex) and
return.getAddressOperand() = returnInd.getSourceAddressOperand() and
indirectionIndex = return.getIndirectionIndex() - 1 // We subtract one because the return loads the value.
)
}
/**
* A return kind. A return kind describes how a value can be returned
* from a callable. For C++, this is simply a function return.
*/
class ReturnKind extends TReturnKind {
/** Gets a textual representation of this return kind. */
abstract string toString();
}
private class NormalReturnKind extends ReturnKind, TNormalReturnKind {
int index;
NormalReturnKind() { this = TNormalReturnKind(index) }
override string toString() { result = "indirect return" }
}
private class IndirectReturnKind extends ReturnKind, TIndirectReturnKind {
int argumentIndex;
int indirectionIndex;
IndirectReturnKind() { this = TIndirectReturnKind(argumentIndex, indirectionIndex) }
override string toString() { result = "indirect outparam[" + argumentIndex.toString() + "]" }
}
/** A data flow node that occurs as the result of a `ReturnStmt`. */
class ReturnNode extends Node instanceof IndirectReturnNode {
/** Gets the kind of this returned value. */
abstract ReturnKind getKind();
}
/**
* This predicate represents an annoying hack that we have to do. We use the
* `ReturnIndirectionInstruction` to determine which variables need flow back
* out of a function. However, the IR will unconditionally create those for a
* variable passed to a function even though the variable was never updated by
* the function. And if a function has too many `ReturnNode`s the dataflow
* library lowers its precision for that function by disabling field flow.
*
* So we those eliminate `ReturnNode`s that would have otherwise been created
* by this unconditional `ReturnIndirectionInstruction` by requiring that there
* must exist an SSA definition of the IR variable in the function.
*/
private predicate hasNonInitializeParameterDef(IRVariable v) {
exists(Ssa::Def def |
not def.getDefiningInstruction() instanceof InitializeParameterInstruction and
v = def.getSourceVariable().getBaseVariable().(Ssa::BaseIRVariable).getIRVariable()
)
}
class ReturnIndirectionNode extends IndirectReturnNode, ReturnNode {
override ReturnKind getKind() {
exists(int argumentIndex, ReturnIndirectionInstruction returnInd |
returnInd.hasIndex(argumentIndex) and
this.getAddressOperand() = returnInd.getSourceAddressOperand() and
result = TIndirectReturnKind(argumentIndex, this.getIndirectionIndex() - 1) and
hasNonInitializeParameterDef(returnInd.getIRVariable())
)
or
this.getAddressOperand() = any(ReturnValueInstruction r).getReturnAddressOperand() and
result = TNormalReturnKind(this.getIndirectionIndex() - 1)
}
}
private Operand fullyConvertedCallStep(Operand op) {
not exists(getANonConversionUse(op)) and
exists(Instruction instr |
conversionFlow(op, instr, _) and
result = getAUse(instr)
)
}
/**
* Gets the instruction that uses this operand, if the instruction is not
* ignored for dataflow purposes.
*/
private Instruction getUse(Operand op) {
result = op.getUse() and
not Ssa::ignoreOperand(op)
}
/** Gets a use of the instruction `instr` that is not ignored for dataflow purposes. */
Operand getAUse(Instruction instr) {
result = instr.getAUse() and
not Ssa::ignoreOperand(result)
}
/**
* Gets a use of `operand` that is:
* - not ignored for dataflow purposes, and
* - not a conversion-like instruction.
*/
private Instruction getANonConversionUse(Operand operand) {
result = getUse(operand) and
not conversionFlow(_, result, _)
}
/**
* Gets the operand that represents the first use of the value of `call` following
* a sequnce of conversion-like instructions.
*/
predicate operandForfullyConvertedCall(Operand operand, CallInstruction call) {
exists(getANonConversionUse(operand)) and
(
operand = getAUse(call)
or
operand = fullyConvertedCallStep*(getAUse(call))
)
}
/**
* Gets the instruction that represents the first use of the value of `call` following
* a sequnce of conversion-like instructions.
*
* This predicate only holds if there is no suitable operand (i.e., no operand of a non-
* conversion instruction) to use to represent the value of `call` after conversions.
*/
predicate instructionForfullyConvertedCall(Instruction instr, CallInstruction call) {
not operandForfullyConvertedCall(_, call) and
(
// If there is no use of the call then we pick the call instruction
not exists(getAUse(call)) and
instr = call
or
// Otherwise, flow to the first non-conversion use.
exists(Operand operand | operand = fullyConvertedCallStep*(getAUse(call)) |
instr = getANonConversionUse(operand)
)
)
}
/** Holds if `node` represents the output node for `call`. */
private predicate simpleOutNode(Node node, CallInstruction call) {
operandForfullyConvertedCall(node.asOperand(), call)
or
instructionForfullyConvertedCall(node.asInstruction(), call)
}
/** A data flow node that represents the output of a call. */
class OutNode extends Node {
OutNode() {
// Return values not hidden behind indirections
simpleOutNode(this, _)
or
// Return values hidden behind indirections
this instanceof IndirectReturnOutNode
or
// Modified arguments hidden behind indirections
this instanceof IndirectArgumentOutNode
}
/** Gets the underlying call. */
abstract DataFlowCall getCall();
abstract ReturnKind getReturnKind();
}
private class DirectCallOutNode extends OutNode {
CallInstruction call;
DirectCallOutNode() { simpleOutNode(this, call) }
override DataFlowCall getCall() { result = call }
override ReturnKind getReturnKind() { result = TNormalReturnKind(0) }
}
private class IndirectCallOutNode extends OutNode, IndirectReturnOutNode {
override DataFlowCall getCall() { result = this.getCallInstruction() }
override ReturnKind getReturnKind() { result = TNormalReturnKind(this.getIndirectionIndex()) }
}
private class SideEffectOutNode extends OutNode, IndirectArgumentOutNode {
override DataFlowCall getCall() { result = this.getCallInstruction() }
override ReturnKind getReturnKind() {
result = TIndirectReturnKind(this.getArgumentIndex(), this.getIndirectionIndex())
}
}
/**
* Gets a node that can read the value returned from `call` with return kind
* `kind`.
*/
OutNode getAnOutNode(DataFlowCall call, ReturnKind kind) {
result.getCall() = call and
result.getReturnKind() = kind
}
/**
* Holds if data can flow from `node1` to `node2` in a way that loses the
* calling context. For example, this would happen with flow through a
* global or static variable.
*/
predicate jumpStep(Node n1, Node n2) {
exists(Cpp::GlobalOrNamespaceVariable v |
v =
n1.asInstruction()
.(StoreInstruction)
.getResultAddress()
.(VariableAddressInstruction)
.getAstVariable() and
v = n2.asVariable()
or
v =
n2.asInstruction()
.(LoadInstruction)
.getSourceAddress()
.(VariableAddressInstruction)
.getAstVariable() and
v = n1.asVariable()
)
}
/**
* Holds if data can flow from `node1` to `node2` via an assignment to `f`.
* Thus, `node2` references an object with a field `f` that contains the
* value of `node1`.
*/
predicate storeStep(Node node1, Content c, PostFieldUpdateNode node2) {
exists(int indirectionIndex1, int numberOfLoads, StoreInstruction store |
nodeHasInstruction(node1, store, pragma[only_bind_into](indirectionIndex1)) and
node2.getIndirectionIndex() = 0 and
numberOfLoadsFromOperand(node2.getFieldAddress(), store.getDestinationAddressOperand(),
numberOfLoads)
|
exists(FieldContent fc | fc = c |
fc.getField() = node2.getUpdatedField() and
fc.getIndirectionIndex() = 1 + indirectionIndex1 + numberOfLoads
)
or
exists(UnionContent uc | uc = c |
uc.getAField() = node2.getUpdatedField() and
uc.getIndirectionIndex() = 1 + indirectionIndex1 + numberOfLoads
)
)
}
/**
* Holds if `operandFrom` flows to `operandTo` using a sequence of conversion-like
* operations and exactly `n` `LoadInstruction` operations.
*/
private predicate numberOfLoadsFromOperandRec(Operand operandFrom, Operand operandTo, int ind) {
exists(LoadInstruction load | load.getSourceAddressOperand() = operandFrom |
operandTo = operandFrom and ind = 0
or
numberOfLoadsFromOperand(load.getAUse(), operandTo, ind - 1)
)
or
exists(Operand op, Instruction instr |
instr = op.getDef() and
conversionFlow(operandFrom, instr, _) and
numberOfLoadsFromOperand(op, operandTo, ind)
)
}
/**
* Holds if `operandFrom` flows to `operandTo` using a sequence of conversion-like
* operations and exactly `n` `LoadInstruction` operations.
*/
private predicate numberOfLoadsFromOperand(Operand operandFrom, Operand operandTo, int n) {
numberOfLoadsFromOperandRec(operandFrom, operandTo, n)
or
not any(LoadInstruction load).getSourceAddressOperand() = operandFrom and
not conversionFlow(operandFrom, _, _) and
operandFrom = operandTo and
n = 0
}
// Needed to join on both an operand and an index at the same time.
pragma[noinline]
predicate nodeHasOperand(Node node, Operand operand, int indirectionIndex) {
node.asOperand() = operand and indirectionIndex = 0
or
hasOperandAndIndex(node, operand, indirectionIndex)
}
// Needed to join on both an instruction and an index at the same time.
pragma[noinline]
predicate nodeHasInstruction(Node node, Instruction instr, int indirectionIndex) {
node.asInstruction() = instr and indirectionIndex = 0
or
hasInstructionAndIndex(node, instr, indirectionIndex)
}
/**
* Holds if data can flow from `node1` to `node2` via a read of `f`.
* Thus, `node1` references an object with a field `f` whose value ends up in
* `node2`.
*/
predicate readStep(Node node1, Content c, Node node2) {
exists(FieldAddress fa1, Operand operand, int numberOfLoads, int indirectionIndex2 |
nodeHasOperand(node2, operand, indirectionIndex2) and
nodeHasOperand(node1, fa1.getObjectAddressOperand(), _) and
numberOfLoadsFromOperand(fa1, operand, numberOfLoads)
|
exists(FieldContent fc | fc = c |
fc.getField() = fa1.getField() and
fc.getIndirectionIndex() = indirectionIndex2 + numberOfLoads
)
or
exists(UnionContent uc | uc = c |
uc.getAField() = fa1.getField() and
uc.getIndirectionIndex() = indirectionIndex2 + numberOfLoads
)
)
}
/**
* Holds if values stored inside content `c` are cleared at node `n`.
*/
predicate clearsContent(Node n, Content c) {
none() // stub implementation
}
/**
* Holds if the value that is being tracked is expected to be stored inside content `c`
* at node `n`.
*/
predicate expectsContent(Node n, ContentSet c) { none() }
/** Gets the type of `n` used for type pruning. */
IRType getNodeType(Node n) {
suppressUnusedNode(n) and
result instanceof IRVoidType // stub implementation
}
/** Gets a string representation of a type returned by `getNodeType`. */
string ppReprType(IRType t) { none() } // stub implementation
/**
* Holds if `t1` and `t2` are compatible, that is, whether data can flow from
* a node of type `t1` to a node of type `t2`.
*/
pragma[inline]
predicate compatibleTypes(IRType t1, IRType t2) {
any() // stub implementation
}
private predicate suppressUnusedNode(Node n) { any() }
//////////////////////////////////////////////////////////////////////////////
// Java QL library compatibility wrappers
//////////////////////////////////////////////////////////////////////////////
/** A node that performs a type cast. */
class CastNode extends Node {
CastNode() { none() } // stub implementation
}
/**
* A function that may contain code or a variable that may contain itself. When
* flow crosses from one _enclosing callable_ to another, the interprocedural
* data-flow library discards call contexts and inserts a node in the big-step
* relation used for human-readable path explanations.
*/
class DataFlowCallable = Cpp::Declaration;
class DataFlowExpr = Expr;
class DataFlowType = IRType;
/** A function call relevant for data flow. */
class DataFlowCall extends CallInstruction {
Function getEnclosingCallable() { result = this.getEnclosingFunction() }
}
predicate isUnreachableInCall(Node n, DataFlowCall call) { none() } // stub implementation
int accessPathLimit() { result = 5 }
/**
* Holds if access paths with `c` at their head always should be tracked at high
* precision. This disables adaptive access path precision for such access paths.
*/
predicate forceHighPrecision(Content c) { none() }
/** The unit type. */
private newtype TUnit = TMkUnit()
/** The trivial type with a single element. */
class Unit extends TUnit {
/** Gets a textual representation of this element. */
string toString() { result = "unit" }
}
/** Holds if `n` should be hidden from path explanations. */
predicate nodeIsHidden(Node n) { n instanceof OperandNode and not n instanceof ArgumentNode }
class LambdaCallKind = Unit;
/** Holds if `creation` is an expression that creates a lambda of kind `kind` for `c`. */
predicate lambdaCreation(Node creation, LambdaCallKind kind, DataFlowCallable c) { none() }
/** Holds if `call` is a lambda call of kind `kind` where `receiver` is the lambda expression. */
predicate lambdaCall(DataFlowCall call, LambdaCallKind kind, Node receiver) { none() }
/** Extra data-flow steps needed for lambda flow analysis. */
predicate additionalLambdaFlowStep(Node nodeFrom, Node nodeTo, boolean preservesValue) { none() }
/**
* Holds if flow is allowed to pass from parameter `p` and back to itself as a
* side-effect, resulting in a summary from `p` to itself.
*
* One example would be to allow flow like `p.foo = p.bar;`, which is disallowed
* by default as a heuristic.
*/
predicate allowParameterReturnInSelf(ParameterNode p) { none() }
private class MyConsistencyConfiguration extends Consistency::ConsistencyConfiguration {
override predicate argHasPostUpdateExclude(ArgumentNode n) {
// The rules for whether an IR argument gets a post-update node are too
// complex to model here.
any()
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -0,0 +1,93 @@
/**
* Provides predicates for mapping the `FunctionInput` and `FunctionOutput`
* classes used in function models to the corresponding instructions.
*/
private import semmle.code.cpp.ir.IR
private import experimental.semmle.code.cpp.ir.dataflow.DataFlow
private import experimental.semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import SsaInternals as Ssa
/**
* Gets the instruction that goes into `input` for `call`.
*/
DataFlow::Node callInput(CallInstruction call, FunctionInput input) {
// An argument or qualifier
exists(int index |
result.asOperand() = call.getArgumentOperand(index) and
input.isParameterOrQualifierAddress(index)
)
or
// A value pointed to by an argument or qualifier
exists(int index, int indirectionIndex |
hasOperandAndIndex(result, call.getArgumentOperand(index), indirectionIndex) and
input.isParameterDerefOrQualifierObject(index, indirectionIndex)
)
or
exists(int ind |
result = getIndirectReturnOutNode(call, ind) and
input.isReturnValueDeref(ind)
)
}
/**
* Gets the instruction that holds the `output` for `call`.
*/
Node callOutput(CallInstruction call, FunctionOutput output) {
// The return value
result.asInstruction() = call and
output.isReturnValue()
or
// The side effect of a call on the value pointed to by an argument or qualifier
exists(int index, int indirectionIndex |
result.(IndirectArgumentOutNode).getArgumentIndex() = index and
result.(IndirectArgumentOutNode).getIndirectionIndex() + 1 = indirectionIndex and
result.(IndirectArgumentOutNode).getCallInstruction() = call and
output.isParameterDerefOrQualifierObject(index, indirectionIndex)
)
or
exists(int ind |
result = getIndirectReturnOutNode(call, ind) and
output.isReturnValueDeref(ind)
)
}
DataFlow::Node callInput(CallInstruction call, FunctionInput input, int d) {
exists(DataFlow::Node n | n = callInput(call, input) and d > 0 |
// An argument or qualifier
hasOperandAndIndex(result, n.asOperand(), d)
or
exists(Operand operand, int indirectionIndex |
// A value pointed to by an argument or qualifier
hasOperandAndIndex(n, operand, indirectionIndex) and
hasOperandAndIndex(result, operand, indirectionIndex + d)
)
)
}
private IndirectReturnOutNode getIndirectReturnOutNode(CallInstruction call, int d) {
result.getCallInstruction() = call and
result.getIndirectionIndex() = d
}
/**
* Gets the instruction that holds the `output` for `call`.
*/
bindingset[d]
Node callOutput(CallInstruction call, FunctionOutput output, int d) {
exists(DataFlow::Node n | n = callOutput(call, output) and d > 0 |
// The return value
result = getIndirectReturnOutNode(n.asInstruction(), d)
or
// If there isn't an indirect out node for the call with indirection `d` then
// we conflate this with the underlying `CallInstruction`.
not exists(getIndirectReturnOutNode(call, d)) and
n.asInstruction() = result.asInstruction()
or
// The side effect of a call on the value pointed to by an argument or qualifier
exists(Operand operand, int indirectionIndex |
Ssa::outNodeHasAddressAndIndex(n, operand, indirectionIndex) and
Ssa::outNodeHasAddressAndIndex(result, operand, indirectionIndex + d)
)
)
}

View File

@@ -0,0 +1,136 @@
private import cpp
// The `ValueNumbering` library has to be imported right after `cpp` to ensure
// that the cached IR gets the same checksum here as it does in queries that use
// `ValueNumbering` without `DataFlow`.
private import semmle.code.cpp.ir.ValueNumbering
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.DataFlow
private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import PrintIRUtilities
/**
* Gets the local dataflow from other nodes in the same function to this node.
*/
private string getFromFlow(DataFlow::Node useNode, int order1, int order2) {
exists(DataFlow::Node defNode, string prefix |
(
simpleLocalFlowStep(defNode, useNode) and prefix = ""
or
any(DataFlow::Configuration cfg).isAdditionalFlowStep(defNode, useNode) and
defNode.getEnclosingCallable() = useNode.getEnclosingCallable() and
prefix = "+"
) and
if defNode.asInstruction() = useNode.asOperand().getAnyDef()
then
// Shorthand for flow from the def of this operand.
result = prefix + "def" and
order1 = -1 and
order2 = 0
else
if defNode.asOperand().getUse() = useNode.asInstruction()
then
// Shorthand for flow from an operand of this instruction
result = prefix + defNode.asOperand().getDumpId() and
order1 = -1 and
order2 = defNode.asOperand().getDumpSortOrder()
else result = prefix + nodeId(defNode, order1, order2)
)
}
/**
* Gets the local dataflow from this node to other nodes in the same function.
*/
private string getToFlow(DataFlow::Node defNode, int order1, int order2) {
exists(DataFlow::Node useNode, string prefix |
(
simpleLocalFlowStep(defNode, useNode) and prefix = ""
or
any(DataFlow::Configuration cfg).isAdditionalFlowStep(defNode, useNode) and
defNode.getEnclosingCallable() = useNode.getEnclosingCallable() and
prefix = "+"
) and
if useNode.asInstruction() = defNode.asOperand().getUse()
then
// Shorthand for flow to this operand's instruction.
result = prefix + "result" and
order1 = -1 and
order2 = 0
else result = prefix + nodeId(useNode, order1, order2)
)
}
/**
* Gets the properties of the dataflow node `node`.
*/
private string getNodeProperty(DataFlow::Node node, string key) {
// List dataflow into and out of this node. Flow into this node is printed as `src->@`, and flow
// out of this node is printed as `@->dest`.
key = "flow" and
result =
strictconcat(string flow, boolean to, int order1, int order2 |
flow = getFromFlow(node, order1, order2) + "->@" and to = false
or
flow = "@->" + getToFlow(node, order1, order2) and to = true
|
flow, ", " order by to, order1, order2, flow
)
or
// Is this node a dataflow sink?
key = "sink" and
any(DataFlow::Configuration cfg).isSink(node) and
result = "true"
or
// Is this node a dataflow source?
key = "source" and
any(DataFlow::Configuration cfg).isSource(node) and
result = "true"
or
// Is this node a dataflow barrier, and if so, what kind?
key = "barrier" and
result =
strictconcat(string kind |
any(DataFlow::Configuration cfg).isBarrier(node) and kind = "full"
or
any(DataFlow::Configuration cfg).isBarrierIn(node) and kind = "in"
or
any(DataFlow::Configuration cfg).isBarrierOut(node) and kind = "out"
|
kind, ", "
)
or
// Is there partial flow from a source to this node?
// This property will only be emitted if partial flow is enabled by overriding
// `DataFlow::Configration::explorationLimit()`.
key = "pflow" and
result =
strictconcat(DataFlow::PartialPathNode sourceNode, DataFlow::PartialPathNode destNode, int dist,
int order1, int order2 |
any(DataFlow::Configuration cfg).hasPartialFlow(sourceNode, destNode, dist) and
destNode.getNode() = node and
// Only print flow from a source in the same function.
sourceNode.getNode().getEnclosingCallable() = node.getEnclosingCallable()
|
nodeId(sourceNode.getNode(), order1, order2) + "+" + dist.toString(), ", "
order by
order1, order2, dist desc
)
}
/**
* Property provider for local IR dataflow.
*/
class LocalFlowPropertyProvider extends IRPropertyProvider {
override string getOperandProperty(Operand operand, string key) {
exists(DataFlow::Node node |
operand = node.asOperand() and
result = getNodeProperty(node, key)
)
}
override string getInstructionProperty(Instruction instruction, string key) {
exists(DataFlow::Node node |
instruction = node.asInstruction() and
result = getNodeProperty(node, key)
)
}
}

View File

@@ -0,0 +1,33 @@
/**
* Print the dataflow local store steps in IR dumps.
*/
private import cpp
// The `ValueNumbering` library has to be imported right after `cpp` to ensure
// that the cached IR gets the same checksum here as it does in queries that use
// `ValueNumbering` without `DataFlow`.
private import semmle.code.cpp.ir.ValueNumbering
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.DataFlow
private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
private import PrintIRUtilities
/**
* Property provider for local IR dataflow store steps.
*/
class LocalFlowPropertyProvider extends IRPropertyProvider {
override string getInstructionProperty(Instruction instruction, string key) {
exists(DataFlow::Node objectNode, Content content |
key = "content[" + content.toString() + "]" and
instruction = objectNode.asInstruction() and
result =
strictconcat(string element, DataFlow::Node fieldNode |
storeStep(fieldNode, content, objectNode) and
element = nodeId(fieldNode, _, _)
|
element, ", "
)
)
}
}

View File

@@ -0,0 +1,39 @@
/**
* Shared utilities used when printing dataflow annotations in IR dumps.
*/
private import cpp
// The `ValueNumbering` library has to be imported right after `cpp` to ensure
// that the cached IR gets the same checksum here as it does in queries that use
// `ValueNumbering` without `DataFlow`.
private import semmle.code.cpp.ir.ValueNumbering
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.DataFlow
/**
* Gets a short ID for an IR dataflow node.
* - For `Instruction`s, this is just the result ID of the instruction (e.g. `m128`).
* - For `Operand`s, this is the label of the operand, prefixed with the result ID of the
* instruction and a dot (e.g. `m128.left`).
* - For `Variable`s, this is the qualified name of the variable.
*/
string nodeId(DataFlow::Node node, int order1, int order2) {
exists(Instruction instruction | instruction = node.asInstruction() |
result = instruction.getResultId() and
order1 = instruction.getBlock().getDisplayIndex() and
order2 = instruction.getDisplayIndexInBlock()
)
or
exists(Operand operand, Instruction instruction |
operand = node.asOperand() and
instruction = operand.getUse()
|
result = instruction.getResultId() + "." + operand.getDumpId() and
order1 = instruction.getBlock().getDisplayIndex() and
order2 = instruction.getDisplayIndexInBlock()
)
or
result = "var(" + node.asVariable().getQualifiedName() + ")" and
order1 = 1000000 and
order2 = 0
}

View File

@@ -0,0 +1,547 @@
private import codeql.ssa.Ssa as SsaImplCommon
private import semmle.code.cpp.ir.IR
private import DataFlowUtil
private import DataFlowImplCommon as DataFlowImplCommon
private import semmle.code.cpp.models.interfaces.Allocation as Alloc
private import semmle.code.cpp.models.interfaces.DataFlow as DataFlow
private import semmle.code.cpp.ir.internal.IRCppLanguage
private import DataFlowPrivate
private import ssa0.SsaInternals as SsaInternals0
import SsaInternalsCommon
private module SourceVariables {
int getMaxIndirectionForIRVariable(IRVariable var) {
exists(Type type, boolean isGLValue |
var.getLanguageType().hasType(type, isGLValue) and
if isGLValue = true
then result = 1 + getMaxIndirectionsForType(type)
else result = getMaxIndirectionsForType(type)
)
}
class BaseSourceVariable = SsaInternals0::BaseSourceVariable;
class BaseIRVariable = SsaInternals0::BaseIRVariable;
class BaseCallVariable = SsaInternals0::BaseCallVariable;
cached
private newtype TSourceVariable =
TSourceIRVariable(BaseIRVariable baseVar, int ind) {
ind = [0 .. getMaxIndirectionForIRVariable(baseVar.getIRVariable())]
} or
TCallVariable(AllocationInstruction call, int ind) {
ind = [0 .. countIndirectionsForCppType(getResultLanguageType(call))]
}
abstract class SourceVariable extends TSourceVariable {
int ind;
bindingset[ind]
SourceVariable() { any() }
abstract string toString();
int getIndirection() { result = ind }
abstract BaseSourceVariable getBaseVariable();
}
class SourceIRVariable extends SourceVariable, TSourceIRVariable {
BaseIRVariable var;
SourceIRVariable() { this = TSourceIRVariable(var, ind) }
IRVariable getIRVariable() { result = var.getIRVariable() }
override BaseIRVariable getBaseVariable() { result.getIRVariable() = this.getIRVariable() }
override string toString() {
ind = 0 and
result = this.getIRVariable().toString()
or
ind > 0 and
result = this.getIRVariable().toString() + " indirection"
}
}
class CallVariable extends SourceVariable, TCallVariable {
AllocationInstruction call;
CallVariable() { this = TCallVariable(call, ind) }
AllocationInstruction getCall() { result = call }
override BaseCallVariable getBaseVariable() { result.getCallInstruction() = call }
override string toString() {
ind = 0 and
result = "Call"
or
ind > 0 and
result = "Call indirection"
}
}
}
import SourceVariables
predicate hasIndirectOperand(Operand op, int indirectionIndex) {
exists(CppType type, int m |
not ignoreOperand(op) and
type = getLanguageType(op) and
m = countIndirectionsForCppType(type) and
indirectionIndex = [1 .. m]
)
}
predicate hasIndirectInstruction(Instruction instr, int indirectionIndex) {
exists(CppType type, int m |
not ignoreInstruction(instr) and
type = getResultLanguageType(instr) and
m = countIndirectionsForCppType(type) and
indirectionIndex = [1 .. m]
)
}
cached
private newtype TDefOrUseImpl =
TDefImpl(Operand address, int indirectionIndex) {
isDef(_, _, address, _, _, indirectionIndex) and
// We only include the definition if the SSA pruning stage
// concluded that the definition is live after the write.
any(SsaInternals0::Def def).getAddressOperand() = address
} or
TUseImpl(Operand operand, int indirectionIndex) {
isUse(_, operand, _, _, indirectionIndex) and
not isDef(_, _, operand, _, _, _)
}
abstract private class DefOrUseImpl extends TDefOrUseImpl {
/** Gets a textual representation of this element. */
abstract string toString();
/** Gets the block of this definition or use. */
abstract IRBlock getBlock();
/** Holds if this definition or use has index `index` in block `block`. */
abstract predicate hasIndexInBlock(IRBlock block, int index);
final predicate hasIndexInBlock(IRBlock block, int index, SourceVariable sv) {
this.hasIndexInBlock(block, index) and
sv = this.getSourceVariable()
}
/** Gets the location of this element. */
abstract Cpp::Location getLocation();
/**
* Gets the index (i.e., the number of loads required) of this
* definition or use.
*
* Note that this is _not_ the definition's (or use's) index in
* the enclosing basic block. To obtain this index, use
* `DefOrUseImpl::hasIndexInBlock/2` or `DefOrUseImpl::hasIndexInBlock/3`.
*/
abstract int getIndirectionIndex();
/**
* Gets the instruction that computes the base of this definition or use.
* This is always a `VariableAddressInstruction` or an `AllocationInstruction`.
*/
abstract Instruction getBase();
final BaseSourceVariable getBaseSourceVariable() {
exists(IRVariable var |
result.(BaseIRVariable).getIRVariable() = var and
instructionHasIRVariable(this.getBase(), var)
)
or
result.(BaseCallVariable).getCallInstruction() = this.getBase()
}
/** Gets the variable that is defined or used. */
final SourceVariable getSourceVariable() {
exists(BaseSourceVariable v, int ind |
sourceVariableHasBaseAndIndex(result, v, ind) and
defOrUseHasSourceVariable(this, v, ind)
)
}
}
pragma[noinline]
private predicate instructionHasIRVariable(VariableAddressInstruction vai, IRVariable var) {
vai.getIRVariable() = var
}
private predicate defOrUseHasSourceVariable(DefOrUseImpl defOrUse, BaseSourceVariable bv, int ind) {
defHasSourceVariable(defOrUse, bv, ind)
or
useHasSourceVariable(defOrUse, bv, ind)
}
pragma[noinline]
private predicate defHasSourceVariable(DefImpl def, BaseSourceVariable bv, int ind) {
bv = def.getBaseSourceVariable() and
ind = def.getIndirection()
}
pragma[noinline]
private predicate useHasSourceVariable(UseImpl use, BaseSourceVariable bv, int ind) {
bv = use.getBaseSourceVariable() and
ind = use.getIndirection()
}
pragma[noinline]
private predicate sourceVariableHasBaseAndIndex(SourceVariable v, BaseSourceVariable bv, int ind) {
v.getBaseVariable() = bv and
v.getIndirection() = ind
}
class DefImpl extends DefOrUseImpl, TDefImpl {
Operand address;
int ind;
DefImpl() { this = TDefImpl(address, ind) }
override Instruction getBase() { isDef(_, _, address, result, _, _) }
Operand getAddressOperand() { result = address }
int getIndirection() { isDef(_, _, address, _, result, ind) }
override int getIndirectionIndex() { result = ind }
Instruction getDefiningInstruction() { isDef(_, result, address, _, _, _) }
override string toString() { result = "DefImpl" }
override IRBlock getBlock() { result = this.getDefiningInstruction().getBlock() }
override Cpp::Location getLocation() { result = this.getDefiningInstruction().getLocation() }
final override predicate hasIndexInBlock(IRBlock block, int index) {
this.getDefiningInstruction() = block.getInstruction(index)
}
predicate isCertain() { isDef(true, _, address, _, _, ind) }
}
class UseImpl extends DefOrUseImpl, TUseImpl {
Operand operand;
int ind;
UseImpl() { this = TUseImpl(operand, ind) }
Operand getOperand() { result = operand }
override string toString() { result = "UseImpl" }
final override predicate hasIndexInBlock(IRBlock block, int index) {
operand.getUse() = block.getInstruction(index)
}
final override IRBlock getBlock() { result = operand.getUse().getBlock() }
final override Cpp::Location getLocation() { result = operand.getLocation() }
final int getIndirection() { isUse(_, operand, _, result, ind) }
override int getIndirectionIndex() { result = ind }
override Instruction getBase() { isUse(_, operand, result, _, ind) }
predicate isCertain() { isUse(true, operand, _, _, ind) }
}
/**
* Holds if `defOrUse1` is a definition which is first read by `use`,
* or if `defOrUse1` is a use and `use` is a next subsequent use.
*
* In both cases, `use` can either be an explicit use written in the
* source file, or it can be a phi node as computed by the SSA library.
*/
predicate adjacentDefRead(DefOrUse defOrUse1, UseOrPhi use) {
exists(IRBlock bb1, int i1, SourceVariable v |
defOrUse1.asDefOrUse().hasIndexInBlock(bb1, i1, v)
|
exists(IRBlock bb2, int i2 |
adjacentDefRead(_, pragma[only_bind_into](bb1), pragma[only_bind_into](i1),
pragma[only_bind_into](bb2), pragma[only_bind_into](i2))
|
use.asDefOrUse().(UseImpl).hasIndexInBlock(bb2, i2, v)
)
or
exists(PhiNode phi |
lastRefRedef(_, bb1, i1, phi) and
use.asPhi() = phi and
phi.getSourceVariable() = pragma[only_bind_into](v)
)
)
}
private predicate useToNode(UseOrPhi use, Node nodeTo) {
exists(UseImpl useImpl |
useImpl = use.asDefOrUse() and
nodeHasOperand(nodeTo, useImpl.getOperand(), useImpl.getIndirectionIndex())
)
or
nodeTo.(SsaPhiNode).getPhiNode() = use.asPhi()
}
pragma[noinline]
predicate outNodeHasAddressAndIndex(
IndirectArgumentOutNode out, Operand address, int indirectionIndex
) {
out.getAddressOperand() = address and
out.getIndirectionIndex() = indirectionIndex
}
private predicate defToNode(Node nodeFrom, Def def) {
nodeHasInstruction(nodeFrom, def.getDefiningInstruction(), def.getIndirectionIndex())
}
private predicate nodeToDefOrUse(Node nodeFrom, SsaDefOrUse defOrUse) {
// Node -> Def
defToNode(nodeFrom, defOrUse)
or
// Node -> Use
useToNode(defOrUse, nodeFrom)
}
/**
* Perform a single conversion-like step from `nFrom` to `nTo`. This relation
* only holds when there is no use-use relation out of `nTo`.
*/
private predicate indirectConversionFlowStep(Node nFrom, Node nTo) {
not exists(UseOrPhi defOrUse |
nodeToDefOrUse(nTo, defOrUse) and
adjacentDefRead(defOrUse, _)
) and
exists(Operand op1, Operand op2, int indirectionIndex, Instruction instr |
hasOperandAndIndex(nFrom, op1, pragma[only_bind_into](indirectionIndex)) and
hasOperandAndIndex(nTo, op2, pragma[only_bind_into](indirectionIndex)) and
instr = op2.getDef() and
conversionFlow(op1, instr, _)
)
}
/**
* The reason for this predicate is a bit annoying:
* We cannot mark a `PointerArithmeticInstruction` that computes an offset based on some SSA
* variable `x` as a use of `x` since this creates taint-flow in the following example:
* ```c
* int x = array[source]
* sink(*array)
* ```
* This is because `source` would flow from the operand of `PointerArithmeticInstruction` to the
* result of the instruction, and into the `IndirectOperand` that represents the value of `*array`.
* Then, via use-use flow, flow will arrive at `*array` in `sink(*array)`.
*
* So this predicate recurses back along conversions and `PointerArithmeticInstruction`s to find the
* first use that has provides use-use flow, and uses that target as the target of the `nodeFrom`.
*/
private predicate adjustForPointerArith(Node nodeFrom, UseOrPhi use) {
nodeFrom = any(PostUpdateNode pun).getPreUpdateNode() and
exists(DefOrUse defOrUse, Node adjusted |
indirectConversionFlowStep*(adjusted, nodeFrom) and
nodeToDefOrUse(adjusted, defOrUse) and
adjacentDefRead(defOrUse, use)
)
}
/** Holds if there is def-use or use-use flow from `nodeFrom` to `nodeTo`. */
predicate ssaFlow(Node nodeFrom, Node nodeTo) {
// `nodeFrom = any(PostUpdateNode pun).getPreUpdateNode()` is implied by adjustedForPointerArith.
exists(UseOrPhi use |
adjustForPointerArith(nodeFrom, use) and
useToNode(use, nodeTo)
)
or
not nodeFrom = any(PostUpdateNode pun).getPreUpdateNode() and
exists(DefOrUse defOrUse1, UseOrPhi use |
nodeToDefOrUse(nodeFrom, defOrUse1) and
adjacentDefRead(defOrUse1, use) and
useToNode(use, nodeTo)
)
}
/** Holds if `nodeTo` receives flow from the phi node `nodeFrom`. */
predicate fromPhiNode(SsaPhiNode nodeFrom, Node nodeTo) {
exists(PhiNode phi, SourceVariable sv, IRBlock bb1, int i1, UseOrPhi use |
phi = nodeFrom.getPhiNode() and
phi.definesAt(sv, bb1, i1) and
useToNode(use, nodeTo)
|
exists(IRBlock bb2, int i2 |
use.asDefOrUse().hasIndexInBlock(bb2, i2, sv) and
adjacentDefRead(phi, bb1, i1, bb2, i2)
)
or
exists(PhiNode phiTo |
lastRefRedef(phi, _, _, phiTo) and
nodeTo.(SsaPhiNode).getPhiNode() = phiTo
)
)
}
private SsaInternals0::SourceVariable getOldSourceVariable(SourceVariable v) {
v.getBaseVariable().(BaseIRVariable).getIRVariable() =
result.getBaseVariable().(SsaInternals0::BaseIRVariable).getIRVariable()
or
v.getBaseVariable().(BaseCallVariable).getCallInstruction() =
result.getBaseVariable().(SsaInternals0::BaseCallVariable).getCallInstruction()
}
/**
* Holds if there is a write at index `i` in basic block `bb` to variable `v` that's
* subsequently read (as determined by the SSA pruning stage).
*/
private predicate variableWriteCand(IRBlock bb, int i, SourceVariable v) {
exists(SsaInternals0::Def def, SsaInternals0::SourceVariable v0 |
def.asDefOrUse().hasIndexInBlock(bb, i, v0) and
v0 = getOldSourceVariable(v)
)
}
private module SsaInput implements SsaImplCommon::InputSig {
import InputSigCommon
import SourceVariables
/**
* Holds if the `i`'th write in block `bb` writes to the variable `v`.
* `certain` is `true` if the write is guaranteed to overwrite the entire variable.
*/
predicate variableWrite(IRBlock bb, int i, SourceVariable v, boolean certain) {
DataFlowImplCommon::forceCachingInSameStage() and
variableWriteCand(bb, i, v) and
exists(DefImpl def | def.hasIndexInBlock(bb, i, v) |
if def.isCertain() then certain = true else certain = false
)
}
/**
* Holds if the `i`'th read in block `bb` reads to the variable `v`.
* `certain` is `true` if the read is guaranteed. For C++, this is always the case.
*/
predicate variableRead(IRBlock bb, int i, SourceVariable v, boolean certain) {
exists(UseImpl use | use.hasIndexInBlock(bb, i, v) |
if use.isCertain() then certain = true else certain = false
)
}
}
/**
* The final SSA predicates used for dataflow purposes.
*/
cached
module SsaCached {
/**
* Holds if `def` is accessed at index `i1` in basic block `bb1` (either a read
* or a write), `def` is read at index `i2` in basic block `bb2`, and there is a
* path between them without any read of `def`.
*/
cached
predicate adjacentDefRead(Definition def, IRBlock bb1, int i1, IRBlock bb2, int i2) {
SsaImpl::adjacentDefRead(def, bb1, i1, bb2, i2)
}
/**
* Holds if the node at index `i` in `bb` is a last reference to SSA definition
* `def`. The reference is last because it can reach another write `next`,
* without passing through another read or write.
*/
cached
predicate lastRefRedef(Definition def, IRBlock bb, int i, Definition next) {
SsaImpl::lastRefRedef(def, bb, i, next)
}
}
cached
private newtype TSsaDefOrUse =
TDefOrUse(DefOrUseImpl defOrUse) {
defOrUse instanceof UseImpl
or
// Like in the pruning stage, we only include definition that's live after the
// write as the final definitions computed by SSA.
exists(Definition def, SourceVariable sv, IRBlock bb, int i |
def.definesAt(sv, bb, i) and
defOrUse.(DefImpl).hasIndexInBlock(bb, i, sv)
)
} or
TPhi(PhiNode phi)
abstract private class SsaDefOrUse extends TSsaDefOrUse {
string toString() { none() }
DefOrUseImpl asDefOrUse() { none() }
PhiNode asPhi() { none() }
abstract Location getLocation();
}
class DefOrUse extends TDefOrUse, SsaDefOrUse {
DefOrUseImpl defOrUse;
DefOrUse() { this = TDefOrUse(defOrUse) }
final override DefOrUseImpl asDefOrUse() { result = defOrUse }
final override Location getLocation() { result = defOrUse.getLocation() }
final SourceVariable getSourceVariable() { result = defOrUse.getSourceVariable() }
override string toString() { result = defOrUse.toString() }
}
class Phi extends TPhi, SsaDefOrUse {
PhiNode phi;
Phi() { this = TPhi(phi) }
final override PhiNode asPhi() { result = phi }
final override Location getLocation() { result = phi.getBasicBlock().getLocation() }
override string toString() { result = "Phi" }
}
class UseOrPhi extends SsaDefOrUse {
UseOrPhi() {
this.asDefOrUse() instanceof UseImpl
or
this instanceof Phi
}
final override Location getLocation() {
result = this.asDefOrUse().getLocation() or result = this.(Phi).getLocation()
}
}
class Def extends DefOrUse {
override DefImpl defOrUse;
Operand getAddressOperand() { result = defOrUse.getAddressOperand() }
Instruction getAddress() { result = this.getAddressOperand().getDef() }
/**
* This predicate ensures that joins go from `defOrUse` to the result
* instead of the other way around.
*/
pragma[inline]
int getIndirectionIndex() {
pragma[only_bind_into](result) = pragma[only_bind_out](defOrUse).getIndirectionIndex()
}
Instruction getDefiningInstruction() { result = defOrUse.getDefiningInstruction() }
}
private module SsaImpl = SsaImplCommon::Make<SsaInput>;
class PhiNode = SsaImpl::PhiNode;
class Definition = SsaImpl::Definition;
import SsaCached

View File

@@ -0,0 +1,268 @@
import cpp as Cpp
import semmle.code.cpp.ir.IR
import semmle.code.cpp.ir.internal.IRCppLanguage
private import semmle.code.cpp.ir.implementation.raw.internal.SideEffects as SideEffects
private import DataFlowImplCommon as DataFlowImplCommon
private import DataFlowUtil
/**
* Holds if `operand` is an operand that is not used by the dataflow library.
* Ignored operands are not recognizd as uses by SSA, and they don't have a
* corresponding `(Indirect)OperandNode`.
*/
predicate ignoreOperand(Operand operand) {
operand = any(Instruction instr | ignoreInstruction(instr)).getAnOperand()
}
/**
* Holds if `instr` is an instruction that is not used by the dataflow library.
* Ignored instructions are not recognized as reads/writes by SSA, and they
* don't have a corresponding `(Indirect)InstructionNode`.
*/
predicate ignoreInstruction(Instruction instr) {
DataFlowImplCommon::forceCachingInSameStage() and
(
instr instanceof WriteSideEffectInstruction or
instr instanceof PhiInstruction or
instr instanceof ReadSideEffectInstruction or
instr instanceof ChiInstruction or
instr instanceof InitializeIndirectionInstruction
)
}
/**
* Gets the C++ type of `this` in the member function `f`.
* The result is a glvalue if `isGLValue` is true, and
* a prvalue if `isGLValue` is false.
*/
bindingset[isGLValue]
private CppType getThisType(Cpp::MemberFunction f, boolean isGLValue) {
result.hasType(f.getTypeOfThis(), isGLValue)
}
/**
* Gets the C++ type of the instruction `i`.
*
* This is equivalent to `i.getResultLanguageType()` with the exception
* of instructions that directly references a `this` IRVariable. In this
* case, `i.getResultLanguageType()` gives an unknown type, whereas the
* predicate gives the expected type (i.e., a potentially cv-qualified
* type `A*` where `A` is the declaring type of the member function that
* contains `i`).
*/
cached
CppType getResultLanguageType(Instruction i) {
if i.(VariableAddressInstruction).getIRVariable() instanceof IRThisVariable
then
if i.isGLValue()
then result = getThisType(i.getEnclosingFunction(), true)
else result = getThisType(i.getEnclosingFunction(), false)
else result = i.getResultLanguageType()
}
/**
* Gets the C++ type of the operand `operand`.
* This is equivalent to the type of the operand's defining instruction.
*
* See `getResultLanguageType` for a description of this behavior.
*/
CppType getLanguageType(Operand operand) { result = getResultLanguageType(operand.getDef()) }
/**
* Gets the maximum number of indirections a glvalue of type `type` can have.
* For example:
* - If `type = int`, the result is 1
* - If `type = MyStruct`, the result is 1
* - If `type = char*`, the result is 2
*/
int getMaxIndirectionsForType(Type type) {
result = countIndirectionsForCppType(getTypeForGLValue(type))
}
/**
* Gets the maximum number of indirections a value of type `type` can have.
*
* Note that this predicate is intended to be called on unspecified types
* (i.e., `countIndirections(e.getUnspecifiedType())`).
*/
private int countIndirections(Type t) {
result =
1 +
countIndirections([t.(Cpp::PointerType).getBaseType(), t.(Cpp::ReferenceType).getBaseType()])
or
not t instanceof Cpp::PointerType and
not t instanceof Cpp::ReferenceType and
result = 0
}
/**
* Gets the maximum number of indirections a value of C++
* type `langType` can have.
*/
int countIndirectionsForCppType(LanguageType langType) {
exists(Type type | langType.hasType(type, true) |
result = 1 + countIndirections(type.getUnspecifiedType())
)
or
exists(Type type | langType.hasType(type, false) |
result = countIndirections(type.getUnspecifiedType())
)
}
/**
* A `CallInstruction` that calls an allocation function such
* as `malloc` or `operator new`.
*/
class AllocationInstruction extends CallInstruction {
AllocationInstruction() { this.getStaticCallTarget() instanceof Cpp::AllocationFunction }
}
/**
* Holds if `i` is a base instruction that starts a sequence of uses
* of some variable that SSA can handle.
*
* This is either when `i` is a `VariableAddressInstruction` or when
* `i` is a fresh allocation produced by an `AllocationInstruction`.
*/
private predicate isSourceVariableBase(Instruction i) {
i instanceof VariableAddressInstruction or i instanceof AllocationInstruction
}
/**
* Holds if the value pointed to by `operand` can potentially be
* modified be the caller.
*/
predicate isModifiableByCall(ArgumentOperand operand) {
exists(CallInstruction call, int index, CppType type |
type = getLanguageType(operand) and
call.getArgumentOperand(index) = operand and
if index = -1
then not call.getStaticCallTarget() instanceof Cpp::ConstMemberFunction
else not SideEffects::isConstPointerLike(any(Type t | type.hasType(t, _)))
)
}
cached
private module Cached {
/**
* Holds if `op` is a use of an SSA variable rooted at `base` with `ind` number
* of indirections.
*
* `certain` is `true` if the operand is guaranteed to read the variable, and
* `indirectionIndex` specifies the number of loads required to read the variable.
*/
cached
predicate isUse(boolean certain, Operand op, Instruction base, int ind, int indirectionIndex) {
not ignoreOperand(op) and
certain = true and
exists(LanguageType type, int m, int ind0 |
type = getLanguageType(op) and
m = countIndirectionsForCppType(type) and
isUseImpl(op, base, ind0) and
ind = ind0 + [0 .. m] and
indirectionIndex = ind - ind0
)
}
/**
* Holds if `operand` is a use of an SSA variable rooted at `base`, and the
* path from `base` to `operand` passes through `ind` load-like instructions.
*/
private predicate isUseImpl(Operand operand, Instruction base, int ind) {
DataFlowImplCommon::forceCachingInSameStage() and
ind = 0 and
operand.getDef() = base and
isSourceVariableBase(base)
or
exists(Operand mid, Instruction instr |
isUseImpl(mid, base, ind) and
instr = operand.getDef() and
conversionFlow(mid, instr, false)
)
or
exists(int ind0 |
isUseImpl(operand.getDef().(LoadInstruction).getSourceAddressOperand(), base, ind0)
or
isUseImpl(operand.getDef().(InitializeParameterInstruction).getAnOperand(), base, ind0)
|
ind0 = ind - 1
)
}
/**
* Holds if `address` is an address of an SSA variable rooted at `base`,
* and `instr` is a definition of the SSA variable with `ind` number of indirections.
*
* `certain` is `true` if `instr` is guaranteed to write to the variable, and
* `indirectionIndex` specifies the number of loads required to read the variable
* after the write operation.
*/
cached
predicate isDef(
boolean certain, Instruction instr, Operand address, Instruction base, int ind,
int indirectionIndex
) {
certain = true and
exists(int ind0, CppType type, int m |
address =
[
instr.(StoreInstruction).getDestinationAddressOperand(),
instr.(InitializeParameterInstruction).getAnOperand(),
instr.(InitializeDynamicAllocationInstruction).getAllocationAddressOperand(),
instr.(UninitializedInstruction).getAnOperand()
]
|
isDefImpl(address, base, ind0) and
type = getLanguageType(address) and
m = countIndirectionsForCppType(type) and
ind = ind0 + [1 .. m] and
indirectionIndex = ind - (ind0 + 1)
)
}
/**
* Holds if `address` is a use of an SSA variable rooted at `base`, and the
* path from `base` to `address` passes through `ind` load-like instructions.
*
* Note: Unlike `isUseImpl`, this predicate recurses through pointer-arithmetic
* instructions.
*/
private predicate isDefImpl(Operand address, Instruction base, int ind) {
DataFlowImplCommon::forceCachingInSameStage() and
ind = 0 and
address.getDef() = base and
isSourceVariableBase(base)
or
exists(Operand mid, Instruction instr |
isDefImpl(mid, base, ind) and
instr = address.getDef() and
conversionFlow(mid, instr, _)
)
or
exists(int ind0 |
isDefImpl(address.getDef().(LoadInstruction).getSourceAddressOperand(), base, ind0)
or
isDefImpl(address.getDef().(InitializeParameterInstruction).getAnOperand(), base, ind0)
|
ind0 = ind - 1
)
}
}
import Cached
/**
* Inputs to the shared SSA library's parameterized module that is shared
* between the SSA pruning stage, and the final SSA stage.
*/
module InputSigCommon {
class BasicBlock = IRBlock;
BasicBlock getImmediateBasicBlockDominator(BasicBlock bb) { result.immediatelyDominates(bb) }
BasicBlock getABasicBlockSuccessor(BasicBlock bb) { result = bb.getASuccessor() }
class ExitBasicBlock extends IRBlock {
ExitBasicBlock() { this.getLastInstruction() instanceof ExitFunctionInstruction }
}
}

Some files were not shown because too many files have changed in this diff Show More