Compare commits

..

3 Commits

Author SHA1 Message Date
Arthur Baars
f4f81886d7 Java: update @id of experimental ExecTainted.ql query 2021-06-28 13:18:25 +02:00
Arthur Baars
e7a3ca2ed4 Revert "Java: remove duplicate query"
This reverts commit 0b59e408ba.
2021-06-28 13:15:10 +02:00
Arthur Baars
0b59e408ba Java: remove duplicate query 2021-06-28 12:29:54 +02:00
5938 changed files with 209573 additions and 512411 deletions

View File

@@ -1,6 +1,4 @@
{ "provide": [ "ruby/.codeqlmanifest.json",
"*/ql/src/qlpack.yml",
"*/ql/lib/qlpack.yml",
{ "provide": [ "*/ql/src/qlpack.yml",
"*/ql/test/qlpack.yml",
"cpp/ql/test/query-tests/Security/CWE/CWE-190/semmle/tainted/qlpack.yml",
"*/ql/examples/qlpack.yml",

View File

@@ -1,14 +1,9 @@
{
"extensions": [
"rust-lang.rust",
"bungcip.better-toml",
"github.vscode-codeql",
"slevesque.vscode-zipexplorer"
],
"settings": {
"files.watcherExclude": {
"**/target/**": true
},
"codeQL.runningQueries.memory": 2048
}
}

3
.gitattributes vendored
View File

@@ -48,6 +48,3 @@
*.gif -text
*.dll -text
*.pdb -text
java/ql/test/stubs/**/*.java linguist-generated=true
java/ql/test/experimental/stubs/**/*.java linguist-generated=true

View File

@@ -1,14 +0,0 @@
name: Fetch CodeQL
description: Fetches the latest version of CodeQL
runs:
using: composite
steps:
- name: Fetch CodeQL
shell: bash
run: |
LATEST=$(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | grep -v beta | sort --version-sort | tail -1)
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-linux64.zip "$LATEST"
unzip -q codeql-linux64.zip
echo "${{ github.workspace }}/codeql" >> $GITHUB_PATH
env:
GITHUB_TOKEN: ${{ github.token }}

View File

@@ -1,18 +0,0 @@
version: 2
updates:
- package-ecosystem: "cargo"
directory: "ruby/node-types"
schedule:
interval: "daily"
- package-ecosystem: "cargo"
directory: "ruby/generator"
schedule:
interval: "daily"
- package-ecosystem: "cargo"
directory: "ruby/extractor"
schedule:
interval: "daily"
- package-ecosystem: "cargo"
directory: "ruby/autobuilder"
schedule:
interval: "daily"

View File

@@ -11,8 +11,6 @@ on:
- 'rc/*'
paths:
- 'csharp/**'
- '.github/codeql/**'
- '.github/workflows/codeql-analysis.yml'
schedule:
- cron: '0 9 * * 1'
@@ -40,8 +38,8 @@ jobs:
# Autobuild attempts to build any compiled languages (C/C++, C#, or Java).
# If this step fails, then you should remove it and run the build manually (see below)
#- name: Autobuild
# uses: github/codeql-action/autobuild@main
- name: Autobuild
uses: github/codeql-action/autobuild@main
# Command-line programs to run using the OS shell.
# 📚 https://git.io/JvXDl
@@ -50,8 +48,9 @@ jobs:
# and modify them (or add more) to build your code if your project
# uses a compiled language
- run: |
dotnet build csharp
#- run: |
# make bootstrap
# make release
- name: Perform CodeQL Analysis
uses: github/codeql-action/analyze@main

View File

@@ -49,23 +49,19 @@ jobs:
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
- name: Generate CSV files on merge and base 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@v2
with:
@@ -80,12 +76,6 @@ jobs:
path: |
out_base/framework-coverage-*.csv
out_base/framework-coverage-*.rst
- name: Upload comparison results
uses: actions/upload-artifact@v2
with:
name: comparison
path: |
comparison.md
- name: Save PR number
run: |
mkdir -p pr

View File

@@ -26,9 +26,40 @@ jobs:
with:
python-version: 3.8
- name: Check coverage difference file and comment
# download artifacts from the PR job:
- name: Download artifact - MERGE
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
RUN_ID: ${{ github.event.workflow_run.id }}
run: |
python misc/scripts/library-coverage/comment-pr.py "$GITHUB_REPOSITORY" "$RUN_ID"
gh run download --name "csv-framework-coverage-merge" --dir "out_merge" "$RUN_ID"
- name: Download artifact - BASE
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
RUN_ID: ${{ github.event.workflow_run.id }}
run: |
gh run download --name "csv-framework-coverage-base" --dir "out_base" "$RUN_ID"
- name: Download artifact - PR
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
RUN_ID: ${{ github.event.workflow_run.id }}
run: |
gh run download --name "pr" --dir "pr" "$RUN_ID"
- name: Check coverage files
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
RUN_ID: ${{ github.event.workflow_run.id }}
run: |
PR=$(cat "pr/NR")
python misc/scripts/library-coverage/compare-files-comment-pr.py \
out_base out_merge comparison.md "$GITHUB_REPOSITORY" "$PR" "$RUN_ID"
- name: Upload comparison results
uses: actions/upload-artifact@v2
with:
name: comparison
path: |
comparison.md

View File

@@ -1,44 +0,0 @@
name: Update framework coverage reports
on:
workflow_dispatch:
schedule:
- cron: "0 0 * * *"
jobs:
update:
name: Update framework coverage report
if: github.repository == 'github/codeql'
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@v2
with:
path: ql
fetch-depth: 0
- name: Set up Python 3.8
uses: actions/setup-python@v2
with:
python-version: 3.8
- name: Download CodeQL CLI
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh release download --repo "github/codeql-cli-binaries" --pattern "codeql-linux64.zip"
- name: Unzip CodeQL CLI
run: unzip -d codeql-cli codeql-linux64.zip
- name: Generate coverage files
run: |
PATH="$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"

View File

@@ -1,39 +0,0 @@
name: Query help preview
on:
pull_request:
branches:
- main
- 'rc/*'
paths:
- "ruby/**/*.qhelp"
jobs:
qhelp:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
with:
fetch-depth: 2
- name: Determine changed files
id: changes
run: |
echo -n "::set-output name=qhelp_files::"
(git diff --name-only --diff-filter=ACMRT HEAD~1 HEAD | grep .qhelp$ | grep -v .inc.qhelp;
git diff --name-only --diff-filter=ACMRT HEAD~1 HEAD | grep .inc.qhelp$ | xargs -d '\n' -rn1 basename | xargs -d '\n' -rn1 git grep -l) |
sort -u | xargs -d '\n' -n1 printf "'%s' "
- uses: ./.github/actions/fetch-codeql
- name: QHelp preview
if: ${{ steps.changes.outputs.qhelp_files }}
run: |
( echo "QHelp previews:";
for path in ${{ steps.changes.outputs.qhelp_files }} ; do
echo "<details> <summary>${path}</summary>"
echo
codeql generate query-help --format=markdown ${path}
echo "</details>"
done) | gh pr comment "${{ github.event.pull_request.number }}" -F -
env:
GITHUB_TOKEN: ${{ github.token }}

View File

@@ -1,232 +0,0 @@
name: "Ruby: Build"
on:
push:
paths:
- 'ruby/**'
branches:
- main
- 'rc/*'
pull_request:
paths:
- 'ruby/**'
branches:
- main
- 'rc/*'
workflow_dispatch:
inputs:
tag:
description: "Version tag to create"
required: false
env:
CARGO_TERM_COLOR: always
defaults:
run:
working-directory: ruby
jobs:
build:
strategy:
fail-fast: false
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v2
- name: Install GNU tar
if: runner.os == 'macOS'
run: |
brew install gnu-tar
echo "/usr/local/opt/gnu-tar/libexec/gnubin" >> $GITHUB_PATH
- uses: actions/cache@v2
with:
path: |
~/.cargo/registry
~/.cargo/git
ruby/target
key: ${{ runner.os }}-rust-cargo-${{ hashFiles('**/Cargo.lock') }}
- name: Check formatting
run: cargo fmt --all -- --check
- name: Build
run: cargo build --verbose
- name: Run tests
run: cargo test --verbose
- name: Release build
run: cargo build --release
- name: Generate dbscheme
if: ${{ matrix.os == 'ubuntu-latest' }}
run: target/release/ruby-generator --dbscheme ql/lib/ruby.dbscheme --library ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
- uses: actions/upload-artifact@v2
if: ${{ matrix.os == 'ubuntu-latest' }}
with:
name: ruby.dbscheme
path: ruby/ql/lib/ruby.dbscheme
- uses: actions/upload-artifact@v2
if: ${{ matrix.os == 'ubuntu-latest' }}
with:
name: TreeSitter.qll
path: ruby/ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
- uses: actions/upload-artifact@v2
with:
name: extractor-${{ matrix.os }}
path: |
ruby/target/release/ruby-autobuilder
ruby/target/release/ruby-autobuilder.exe
ruby/target/release/ruby-extractor
ruby/target/release/ruby-extractor.exe
retention-days: 1
compile-queries:
runs-on: ubuntu-latest
env:
CODEQL_THREADS: 4 # TODO: remove this once it's set by the CLI
steps:
- uses: actions/checkout@v2
- 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: ${{ github.token }}
- name: Build Query Pack
run: |
codeql/codeql pack create ql/lib --output target/packs
codeql/codeql pack install ql/src
codeql/codeql pack create ql/src --output target/packs
PACK_FOLDER=$(readlink -f target/packs/codeql/ruby-queries/*)
codeql/codeql generate query-help --format=sarifv2.1.0 --output="${PACK_FOLDER}/rules.sarif" ql/src
(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}/{}" \;)
- name: Compile with previous CodeQL versions
run: |
for version in $(gh release list --repo https://github.com/github/codeql-cli-binaries | cut -f 1 | sort --version-sort | tail -3 | head -2); do
rm -f codeql-linux64.zip
gh release download --repo https://github.com/github/codeql-cli-binaries --pattern codeql-linux64.zip "$version"
rm -rf codeql; unzip -q codeql-linux64.zip
codeql/codeql query compile target/packs/*
done
env:
GITHUB_TOKEN: ${{ github.token }}
- uses: actions/upload-artifact@v2
with:
name: codeql-ruby-queries
path: |
ruby/target/packs/*
retention-days: 1
package:
runs-on: ubuntu-latest
needs: [build, compile-queries]
steps:
- uses: actions/checkout@v2
- uses: actions/download-artifact@v2
with:
name: ruby.dbscheme
path: ruby/ruby
- uses: actions/download-artifact@v2
with:
name: extractor-ubuntu-latest
path: ruby/linux64
- uses: actions/download-artifact@v2
with:
name: extractor-windows-latest
path: ruby/win64
- uses: actions/download-artifact@v2
with:
name: extractor-macos-latest
path: ruby/osx64
- run: |
mkdir -p ruby
cp -r codeql-extractor.yml tools ql/lib/ruby.dbscheme.stats ruby/
mkdir -p ruby/tools/{linux64,osx64,win64}
cp linux64/ruby-autobuilder ruby/tools/linux64/autobuilder
cp osx64/ruby-autobuilder ruby/tools/osx64/autobuilder
cp win64/ruby-autobuilder.exe ruby/tools/win64/autobuilder.exe
cp linux64/ruby-extractor ruby/tools/linux64/extractor
cp osx64/ruby-extractor ruby/tools/osx64/extractor
cp win64/ruby-extractor.exe ruby/tools/win64/extractor.exe
chmod +x ruby/tools/{linux64,osx64}/{autobuilder,extractor}
zip -rq codeql-ruby.zip ruby
- uses: actions/upload-artifact@v2
with:
name: codeql-ruby-pack
path: ruby/codeql-ruby.zip
retention-days: 1
- uses: actions/download-artifact@v2
with:
name: codeql-ruby-queries
path: ruby/qlpacks
- run: |
echo '{
"provide": [
"ruby/codeql-extractor.yml",
"qlpacks/*/*/*/qlpack.yml"
]
}' > .codeqlmanifest.json
zip -rq codeql-ruby-bundle.zip .codeqlmanifest.json ruby qlpacks
- uses: actions/upload-artifact@v2
with:
name: codeql-ruby-bundle
path: ruby/codeql-ruby-bundle.zip
retention-days: 1
test:
defaults:
run:
working-directory: ${{ github.workspace }}
strategy:
fail-fast: false
matrix:
os: [ubuntu-latest, macos-latest, windows-latest]
runs-on: ${{ matrix.os }}
needs: [package]
steps:
- uses: actions/checkout@v2
with:
repository: Shopify/example-ruby-app
ref: 67a0decc5eb550f3a9228eda53925c3afd40dfe9
- name: Fetch CodeQL
shell: bash
run: |
LATEST=$(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@v2
with:
name: codeql-ruby-bundle
path: ${{ runner.temp }}
- name: Unzip Ruby bundle
shell: bash
run: unzip -q -d "${{ runner.temp }}/ruby-bundle" "${{ runner.temp }}/codeql-ruby-bundle.zip"
- name: Prepare test files
shell: bash
run: |
echo "import ruby select count(File f)" > "test.ql"
echo "| 4 |" > "test.expected"
echo 'name: sample-tests
version: 0.0.0
dependencies:
codeql/ruby-all: 0.0.1
extractor: ruby
tests: .
' > qlpack.yml
- name: Run QL test
shell: bash
run: |
"${{ runner.temp }}/codeql/codeql" test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" .
- name: Create database
shell: bash
run: |
"${{ runner.temp }}/codeql/codeql" database create --search-path "${{ 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

View File

@@ -1,71 +0,0 @@
name: "Ruby: Collect database stats"
on:
push:
branches:
- main
- 'rc/*'
paths:
- ruby/ql/lib/ruby.dbscheme
pull_request:
branches:
- main
- 'rc/*'
paths:
- ruby/ql/lib/ruby.dbscheme
workflow_dispatch:
jobs:
measure:
env:
CODEQL_THREADS: 4 # TODO: remove this once it's set by the CLI
strategy:
fail-fast: false
matrix:
repo: [rails/rails, discourse/discourse, spree/spree]
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- uses: ./.github/actions/fetch-codeql
- uses: ./ruby/actions/create-extractor-pack
- name: Checkout ${{ matrix.repo }}
uses: actions/checkout@v2
with:
repository: ${{ matrix.repo }}
path: ${{ github.workspace }}/repo
- name: Create database
run: |
codeql database create \
--search-path "${{ github.workspace }}/ruby" \
--threads 4 \
--language ruby --source-root "${{ github.workspace }}/repo" \
"${{ runner.temp }}/database"
- name: Measure database
run: |
mkdir -p "stats/${{ matrix.repo }}"
codeql dataset measure --threads 4 --output "stats/${{ matrix.repo }}/stats.xml" "${{ runner.temp }}/database/db-ruby"
- uses: actions/upload-artifact@v2
with:
name: measurements
path: stats
retention-days: 1
merge:
runs-on: ubuntu-latest
needs: measure
steps:
- uses: actions/checkout@v2
- uses: actions/download-artifact@v2
with:
name: measurements
path: stats
- run: |
python -m pip install --user lxml
find stats -name 'stats.xml' | sort | xargs python ruby/scripts/merge_stats.py --output ruby/ql/lib/ruby.dbscheme.stats --normalise ruby_tokeninfo
- uses: actions/upload-artifact@v2
with:
name: ruby.dbscheme.stats
path: ruby/ql/lib/ruby.dbscheme.stats

View File

@@ -1,48 +0,0 @@
name: "Ruby: Run QL Tests"
on:
push:
paths:
- 'ruby/**'
branches:
- main
- 'rc/*'
pull_request:
paths:
- 'ruby/**'
branches:
- main
- 'rc/*'
env:
CARGO_TERM_COLOR: always
defaults:
run:
working-directory: ruby
jobs:
qltest:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- uses: ./.github/actions/fetch-codeql
- uses: ./ruby/actions/create-extractor-pack
- 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 "${{ github.workspace }}/ruby" --additional-packs "${{ github.workspace }}" --consistency-queries ql/consistency-queries ql/test
env:
GITHUB_TOKEN: ${{ github.token }}
- name: Check QL formatting
run: find ql "(" -name "*.ql" -or -name "*.qll" ")" -print0 | xargs -0 codeql query format --check-only
- name: Check QL compilation
run: |
codeql query compile --check-only --threads=4 --warnings=error --search-path "${{ github.workspace }}/ruby" --additional-packs "${{ github.workspace }}" "ql/src" "ql/examples"
env:
GITHUB_TOKEN: ${{ github.token }}
- name: Check DB upgrade scripts
run: |
echo >empty.trap
codeql dataset import -S ql/lib/upgrades/initial/ruby.dbscheme testdb empty.trap
codeql dataset upgrade testdb --additional-packs ql/lib/upgrades
diff -q testdb/ruby.dbscheme ql/lib/ruby.dbscheme

View File

@@ -1,20 +0,0 @@
name: Check synchronized files
on:
push:
branches:
- main
- 'rc/*'
pull_request:
branches:
- main
- 'rc/*'
jobs:
sync:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v2
- name: Check synchronized files
run: python config/sync-files.py

View File

@@ -3,7 +3,6 @@
/java/ @github/codeql-java
/javascript/ @github/codeql-javascript
/python/ @github/codeql-python
/ruby/ @github/codeql-ruby
# Make @xcorail (GitHub Security Lab) a code owner for experimental queries so he gets pinged when we promote a query out of experimental
/cpp/**/experimental/**/* @github/codeql-c-analysis @xcorail
@@ -11,7 +10,6 @@
/java/**/experimental/**/* @github/codeql-java @xcorail
/javascript/**/experimental/**/* @github/codeql-javascript @xcorail
/python/**/experimental/**/* @github/codeql-python @xcorail
/ruby/**/experimental/**/* @github/codeql-ruby @xcorail
# Notify members of codeql-go about PRs to the shared data-flow library files
/java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl.qll @github/codeql-java @github/codeql-go
@@ -19,9 +17,3 @@
/java/ql/src/semmle/code/java/dataflow/internal/DataFlowImplCommon.qll @github/codeql-java @github/codeql-go
/java/ql/src/semmle/code/java/dataflow/internal/tainttracking1/TaintTrackingImpl.qll @github/codeql-java @github/codeql-go
/java/ql/src/semmle/code/java/dataflow/internal/tainttracking2/TaintTrackingImpl.qll @github/codeql-java @github/codeql-go
# CodeQL tools and associated docs
/docs/codeql-cli/ @github/codeql-cli-reviewers
/docs/codeql-for-visual-studio-code/ @github/codeql-vscode-reviewers
/docs/ql-language-reference/ @github/codeql-frontend-reviewers
/docs/query-*-style-guide.md @github/codeql-analysis-reviewers

View File

@@ -1,338 +1,332 @@
{
"DataFlow Java/C++/C#/Python": [
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl2.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl3.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl4.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl5.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImpl6.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImplForSerializability.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImpl.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImpl2.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImpl3.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImpl4.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImplLocal.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl2.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl3.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl2.qll",
"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",
"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"
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl.qll",
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl2.qll",
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl3.qll",
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl4.qll",
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl5.qll",
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl6.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImpl.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImpl2.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImpl3.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImpl4.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImplLocal.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl2.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl3.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImpl2.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImpl3.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImpl5.qll",
"python/ql/src/semmle/python/dataflow/new/internal/DataFlowImpl.qll",
"python/ql/src/semmle/python/dataflow/new/internal/DataFlowImpl2.qll",
"python/ql/src/semmle/python/dataflow/new/internal/DataFlowImpl3.qll",
"python/ql/src/semmle/python/dataflow/new/internal/DataFlowImpl4.qll"
],
"DataFlow Java/C++/C#/Python Common": [
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImplCommon.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImplCommon.qll",
"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"
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImplCommon.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImplCommon.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImplCommon.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImplCommon.qll",
"python/ql/src/semmle/python/dataflow/new/internal/DataFlowImplCommon.qll"
],
"TaintTracking::Configuration Java/C++/C#/Python": [
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"cpp/ql/lib/semmle/code/cpp/ir/dataflow/internal/tainttracking3/TaintTrackingImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/tainttracking3/TaintTrackingImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/tainttracking4/TaintTrackingImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/tainttracking5/TaintTrackingImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/tainttracking1/TaintTrackingImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/tainttracking2/TaintTrackingImpl.qll",
"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"
"cpp/ql/src/semmle/code/cpp/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/tainttracking3/TaintTrackingImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/tainttracking3/TaintTrackingImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/tainttracking4/TaintTrackingImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/tainttracking5/TaintTrackingImpl.qll",
"java/ql/src/semmle/code/java/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"java/ql/src/semmle/code/java/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"python/ql/src/semmle/python/dataflow/new/internal/tainttracking1/TaintTrackingImpl.qll",
"python/ql/src/semmle/python/dataflow/new/internal/tainttracking2/TaintTrackingImpl.qll",
"python/ql/src/semmle/python/dataflow/new/internal/tainttracking3/TaintTrackingImpl.qll",
"python/ql/src/semmle/python/dataflow/new/internal/tainttracking4/TaintTrackingImpl.qll"
],
"DataFlow Java/C++/C#/Python Consistency checks": [
"java/ql/lib/semmle/code/java/dataflow/internal/DataFlowImplConsistency.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/DataFlowImplConsistency.qll",
"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"
"java/ql/src/semmle/code/java/dataflow/internal/DataFlowImplConsistency.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/DataFlowImplConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/dataflow/internal/DataFlowImplConsistency.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/DataFlowImplConsistency.qll",
"python/ql/src/semmle/python/dataflow/new/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"
"java/ql/src/semmle/code/java/dataflow/internal/FlowSummaryImpl.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/FlowSummaryImpl.qll"
],
"SsaReadPosition Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/internal/rangeanalysis/SsaReadPositionCommon.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/SsaReadPositionCommon.qll"
"java/ql/src/semmle/code/java/dataflow/internal/rangeanalysis/SsaReadPositionCommon.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/rangeanalysis/SsaReadPositionCommon.qll"
],
"Sign Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/internal/rangeanalysis/Sign.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/Sign.qll"
"java/ql/src/semmle/code/java/dataflow/internal/rangeanalysis/Sign.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/rangeanalysis/Sign.qll"
],
"SignAnalysis Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/internal/rangeanalysis/SignAnalysisCommon.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/SignAnalysisCommon.qll"
"java/ql/src/semmle/code/java/dataflow/internal/rangeanalysis/SignAnalysisCommon.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/rangeanalysis/SignAnalysisCommon.qll"
],
"Bound Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/Bound.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/Bound.qll"
"java/ql/src/semmle/code/java/dataflow/Bound.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/Bound.qll"
],
"ModulusAnalysis Java/C#": [
"java/ql/lib/semmle/code/java/dataflow/ModulusAnalysis.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/ModulusAnalysis.qll"
"java/ql/src/semmle/code/java/dataflow/ModulusAnalysis.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/ModulusAnalysis.qll"
],
"C++ SubBasicBlocks": [
"cpp/ql/lib/semmle/code/cpp/controlflow/SubBasicBlocks.qll",
"cpp/ql/lib/semmle/code/cpp/dataflow/internal/SubBasicBlocks.qll"
"cpp/ql/src/semmle/code/cpp/controlflow/SubBasicBlocks.qll",
"cpp/ql/src/semmle/code/cpp/dataflow/internal/SubBasicBlocks.qll"
],
"IR Instruction": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/Instruction.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/Instruction.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/Instruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/Instruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/Instruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/Instruction.qll",
"csharp/ql/src/experimental/ir/implementation/raw/Instruction.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/Instruction.qll"
],
"IR IRBlock": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IRBlock.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IRBlock.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IRBlock.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/IRBlock.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/IRBlock.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/IRBlock.qll",
"csharp/ql/src/experimental/ir/implementation/raw/IRBlock.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/IRBlock.qll"
],
"IR IRVariable": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IRVariable.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IRVariable.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IRVariable.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/IRVariable.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/IRVariable.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/IRVariable.qll",
"csharp/ql/src/experimental/ir/implementation/raw/IRVariable.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/IRVariable.qll"
],
"IR IRFunction": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IRFunction.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IRFunction.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IRFunction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/IRFunction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/IRFunction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/IRFunction.qll",
"csharp/ql/src/experimental/ir/implementation/raw/IRFunction.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/IRFunction.qll"
],
"IR Operand": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/Operand.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/Operand.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/Operand.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/Operand.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/Operand.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/Operand.qll",
"csharp/ql/src/experimental/ir/implementation/raw/Operand.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/Operand.qll"
],
"IR IRType": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/IRType.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/IRType.qll",
"csharp/ql/src/experimental/ir/implementation/IRType.qll"
],
"IR IRConfiguration": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/IRConfiguration.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/IRConfiguration.qll",
"csharp/ql/src/experimental/ir/implementation/IRConfiguration.qll"
],
"IR UseSoundEscapeAnalysis": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/UseSoundEscapeAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/UseSoundEscapeAnalysis.qll",
"csharp/ql/src/experimental/ir/implementation/UseSoundEscapeAnalysis.qll"
],
"IR IRFunctionBase": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/internal/IRFunctionBase.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/internal/IRFunctionBase.qll",
"csharp/ql/src/experimental/ir/implementation/internal/IRFunctionBase.qll"
],
"IR Operand Tag": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/internal/OperandTag.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/internal/OperandTag.qll",
"csharp/ql/src/experimental/ir/implementation/internal/OperandTag.qll"
],
"IR TInstruction": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/internal/TInstruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/internal/TInstruction.qll",
"csharp/ql/src/experimental/ir/implementation/internal/TInstruction.qll"
],
"IR TIRVariable": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/internal/TIRVariable.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/internal/TIRVariable.qll",
"csharp/ql/src/experimental/ir/implementation/internal/TIRVariable.qll"
],
"IR IR": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IR.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IR.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/IR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/IR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/IR.qll",
"csharp/ql/src/experimental/ir/implementation/raw/IR.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/IR.qll"
],
"IR IRConsistency": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/IRConsistency.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/IRConsistency.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/IRConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/IRConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/IRConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/IRConsistency.qll",
"csharp/ql/src/experimental/ir/implementation/raw/IRConsistency.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/IRConsistency.qll"
],
"IR PrintIR": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/PrintIR.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/PrintIR.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/PrintIR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/PrintIR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/PrintIR.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/PrintIR.qll",
"csharp/ql/src/experimental/ir/implementation/raw/PrintIR.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/PrintIR.qll"
],
"IR IntegerConstant": [
"cpp/ql/lib/semmle/code/cpp/ir/internal/IntegerConstant.qll",
"cpp/ql/src/semmle/code/cpp/ir/internal/IntegerConstant.qll",
"csharp/ql/src/experimental/ir/internal/IntegerConstant.qll"
],
"IR IntegerInteval": [
"cpp/ql/lib/semmle/code/cpp/ir/internal/IntegerInterval.qll",
"cpp/ql/src/semmle/code/cpp/ir/internal/IntegerInterval.qll",
"csharp/ql/src/experimental/ir/internal/IntegerInterval.qll"
],
"IR IntegerPartial": [
"cpp/ql/lib/semmle/code/cpp/ir/internal/IntegerPartial.qll",
"cpp/ql/src/semmle/code/cpp/ir/internal/IntegerPartial.qll",
"csharp/ql/src/experimental/ir/internal/IntegerPartial.qll"
],
"IR Overlap": [
"cpp/ql/lib/semmle/code/cpp/ir/internal/Overlap.qll",
"cpp/ql/src/semmle/code/cpp/ir/internal/Overlap.qll",
"csharp/ql/src/experimental/ir/internal/Overlap.qll"
],
"IR EdgeKind": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/EdgeKind.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/EdgeKind.qll",
"csharp/ql/src/experimental/ir/implementation/EdgeKind.qll"
],
"IR MemoryAccessKind": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/MemoryAccessKind.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/MemoryAccessKind.qll",
"csharp/ql/src/experimental/ir/implementation/MemoryAccessKind.qll"
],
"IR TempVariableTag": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/TempVariableTag.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/TempVariableTag.qll",
"csharp/ql/src/experimental/ir/implementation/TempVariableTag.qll"
],
"IR Opcode": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/Opcode.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/Opcode.qll",
"csharp/ql/src/experimental/ir/implementation/Opcode.qll"
],
"IR SSAConsistency": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConsistency.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConsistency.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConsistency.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/SSAConsistency.qll"
],
"C++ IR InstructionImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/InstructionImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/InstructionImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/InstructionImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/InstructionImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/InstructionImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/InstructionImports.qll"
],
"C++ IR IRImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/IRImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/IRImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRImports.qll"
],
"C++ IR IRBlockImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/IRBlockImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRBlockImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRBlockImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/IRBlockImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRBlockImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRBlockImports.qll"
],
"C++ IR IRFunctionImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/IRFunctionImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRFunctionImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRFunctionImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/IRFunctionImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRFunctionImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRFunctionImports.qll"
],
"C++ IR IRVariableImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/IRVariableImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRVariableImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRVariableImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/IRVariableImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/IRVariableImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/IRVariableImports.qll"
],
"C++ IR OperandImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/OperandImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/OperandImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/OperandImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/OperandImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/OperandImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/OperandImports.qll"
],
"C++ IR PrintIRImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/PrintIRImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintIRImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintIRImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/PrintIRImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintIRImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintIRImports.qll"
],
"C++ SSA SSAConstructionImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstructionImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstructionImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstructionImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstructionImports.qll"
],
"SSA AliasAnalysis": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/AliasAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/AliasAnalysis.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/AliasAnalysis.qll"
],
"SSA PrintAliasAnalysis": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintAliasAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintAliasAnalysis.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintAliasAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintAliasAnalysis.qll"
],
"C++ SSA AliasAnalysisImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasAnalysisImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/AliasAnalysisImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasAnalysisImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/AliasAnalysisImports.qll"
],
"C++ IR ValueNumberingImports": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/gvn/internal/ValueNumberingImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingImports.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/internal/ValueNumberingImports.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/gvn/internal/ValueNumberingImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingImports.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/internal/ValueNumberingImports.qll"
],
"IR SSA SimpleSSA": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SimpleSSA.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SimpleSSA.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/SimpleSSA.qll"
],
"IR AliasConfiguration (unaliased_ssa)": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasConfiguration.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/AliasConfiguration.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/AliasConfiguration.qll"
],
"IR SSA SSAConstruction": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstruction.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/SSAConstruction.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/SSAConstruction.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/SSAConstruction.qll"
],
"IR SSA PrintSSA": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintSSA.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintSSA.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/PrintSSA.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/internal/PrintSSA.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/internal/PrintSSA.qll"
],
"IR ValueNumberInternal": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/gvn/internal/ValueNumberingInternal.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingInternal.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/internal/ValueNumberingInternal.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/gvn/internal/ValueNumberingInternal.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingInternal.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/internal/ValueNumberingInternal.qll",
"csharp/ql/src/experimental/ir/implementation/raw/gvn/internal/ValueNumberingInternal.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingInternal.qll"
],
"C++ IR ValueNumber": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/gvn/ValueNumbering.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/ValueNumbering.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/ValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/gvn/ValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/ValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/ValueNumbering.qll",
"csharp/ql/src/experimental/ir/implementation/raw/gvn/ValueNumbering.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/gvn/ValueNumbering.qll"
],
"C++ IR PrintValueNumbering": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/gvn/PrintValueNumbering.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/PrintValueNumbering.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/PrintValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/gvn/PrintValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/gvn/PrintValueNumbering.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/gvn/PrintValueNumbering.qll",
"csharp/ql/src/experimental/ir/implementation/raw/gvn/PrintValueNumbering.qll",
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/gvn/PrintValueNumbering.qll"
],
"C++ IR ConstantAnalysis": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/constant/ConstantAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/constant/ConstantAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/constant/ConstantAnalysis.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/constant/ConstantAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/constant/ConstantAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/constant/ConstantAnalysis.qll"
],
"C++ IR PrintConstantAnalysis": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/constant/PrintConstantAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/constant/PrintConstantAnalysis.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/aliased_ssa/constant/PrintConstantAnalysis.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/constant/PrintConstantAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/constant/PrintConstantAnalysis.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/aliased_ssa/constant/PrintConstantAnalysis.qll"
],
"C++ IR ReachableBlock": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/reachability/ReachableBlock.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/ReachableBlock.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/reachability/ReachableBlock.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/ReachableBlock.qll"
],
"C++ IR PrintReachableBlock": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/reachability/PrintReachableBlock.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/PrintReachableBlock.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/reachability/PrintReachableBlock.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/PrintReachableBlock.qll"
],
"C++ IR Dominance": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/reachability/Dominance.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/Dominance.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/reachability/Dominance.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/Dominance.qll"
],
"C++ IR PrintDominance": [
"cpp/ql/lib/semmle/code/cpp/ir/implementation/raw/internal/reachability/PrintDominance.qll",
"cpp/ql/lib/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/PrintDominance.qll"
"cpp/ql/src/semmle/code/cpp/ir/implementation/raw/internal/reachability/PrintDominance.qll",
"cpp/ql/src/semmle/code/cpp/ir/implementation/unaliased_ssa/internal/reachability/PrintDominance.qll"
],
"C# IR InstructionImports": [
"csharp/ql/src/experimental/ir/implementation/raw/internal/InstructionImports.qll",
@@ -367,14 +361,13 @@
"csharp/ql/src/experimental/ir/implementation/unaliased_ssa/gvn/internal/ValueNumberingImports.qll"
],
"C# ControlFlowReachability": [
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/ControlFlowReachability.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/ControlFlowReachability.qll"
"csharp/ql/src/semmle/code/csharp/dataflow/internal/ControlFlowReachability.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/rangeanalysis/ControlFlowReachability.qll"
],
"Inline Test Expectations": [
"cpp/ql/test/TestUtilities/InlineExpectationsTest.qll",
"java/ql/test/TestUtilities/InlineExpectationsTest.qll",
"python/ql/test/TestUtilities/InlineExpectationsTest.qll",
"ruby/ql/test/TestUtilities/InlineExpectationsTest.qll"
"python/ql/test/TestUtilities/InlineExpectationsTest.qll"
],
"C++ ExternalAPIs": [
"cpp/ql/src/Security/CWE/CWE-020/ExternalAPIs.qll",
@@ -385,11 +378,11 @@
"cpp/ql/src/Security/CWE/CWE-020/ir/SafeExternalAPIFunction.qll"
],
"XML": [
"cpp/ql/lib/semmle/code/cpp/XML.qll",
"csharp/ql/lib/semmle/code/csharp/XML.qll",
"java/ql/lib/semmle/code/xml/XML.qll",
"javascript/ql/lib/semmle/javascript/XML.qll",
"python/ql/lib/semmle/python/xml/XML.qll"
"cpp/ql/src/semmle/code/cpp/XML.qll",
"csharp/ql/src/semmle/code/csharp/XML.qll",
"java/ql/src/semmle/code/xml/XML.qll",
"javascript/ql/src/semmle/javascript/XML.qll",
"python/ql/src/semmle/python/xml/XML.qll"
],
"DuplicationProblems.inc.qhelp": [
"cpp/ql/src/Metrics/Files/DuplicationProblems.inc.qhelp",
@@ -443,39 +436,17 @@
"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"
"csharp/ql/src/semmle/code/csharp/dataflow/internal/SsaImplCommon.qll",
"csharp/ql/src/semmle/code/csharp/controlflow/internal/pressa/SsaImplCommon.qll",
"csharp/ql/src/semmle/code/csharp/dataflow/internal/basessa/SsaImplCommon.qll",
"csharp/ql/src/semmle/code/cil/internal/SsaImplCommon.qll"
],
"CryptoAlgorithms Python/JS": [
"javascript/ql/lib/semmle/javascript/security/CryptoAlgorithms.qll",
"python/ql/lib/semmle/python/concepts/CryptoAlgorithms.qll"
"javascript/ql/src/semmle/javascript/security/CryptoAlgorithms.qll",
"python/ql/src/semmle/python/concepts/CryptoAlgorithms.qll"
],
"SensitiveDataHeuristics Python/JS": [
"javascript/ql/lib/semmle/javascript/security/internal/SensitiveDataHeuristics.qll",
"python/ql/lib/semmle/python/security/internal/SensitiveDataHeuristics.qll"
],
"ReDoS Util Python/JS": [
"javascript/ql/lib/semmle/javascript/security/performance/ReDoSUtil.qll",
"python/ql/lib/semmle/python/security/performance/ReDoSUtil.qll"
],
"ReDoS Exponential Python/JS": [
"javascript/ql/lib/semmle/javascript/security/performance/ExponentialBackTracking.qll",
"python/ql/lib/semmle/python/security/performance/ExponentialBackTracking.qll"
],
"ReDoS Polynomial Python/JS": [
"javascript/ql/lib/semmle/javascript/security/performance/SuperlinearBackTracking.qll",
"python/ql/lib/semmle/python/security/performance/SuperlinearBackTracking.qll",
"ruby/ql/lib/codeql/ruby/regexp/SuperlinearBackTracking.qll"
],
"CFG": [
"csharp/ql/lib/semmle/code/csharp/controlflow/internal/ControlFlowGraphImplShared.qll",
"ruby/ql/lib/codeql/ruby/controlflow/internal/ControlFlowGraphImplShared.qll"
],
"TypeTracker": [
"python/ql/lib/semmle/python/dataflow/new/internal/TypeTracker.qll",
"ruby/ql/lib/codeql/ruby/typetracking/TypeTracker.qll"
"javascript/ql/src/semmle/javascript/security/internal/SensitiveDataHeuristics.qll",
"python/ql/src/semmle/python/security/internal/SensitiveDataHeuristics.qll"
]
}

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* The 'Uncontrolled data in SQL query' (cpp/sql-injection) query now supports the `libpqxx` library.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* The DataFlow libraries have been augmented with support for `Configuration`-specific in-place read steps at, for example, sinks and custom taint steps. This means that it is now possible to specify sinks that accept flow with non-empty access paths.

View File

@@ -1,2 +0,0 @@
lgtm
* The 'Uncontrolled data in arithmetic expression' (cpp/uncontrolled-arithmetic) query now recognizes more sources of randomness.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* The 'Wrong type of arguments to formatting function' (cpp/wrong-type-format-argument) query is now more accepting of the string and character formatting differences between Microsoft and non-Microsoft platforms. There are now fewer false positive results.

View File

@@ -1,3 +0,0 @@
lgtm,codescanning
* The "Cleartext storage of sensitive information in file" (cpp/cleartext-storage-file) query now uses dataflow to produce additional results.
* Heuristics in the SensitiveExprs.qll library have been improved, making the "Cleartext storage of sensitive information in file" (cpp/cleartext-storage-file), "Cleartext storage of sensitive information in buffer" (cpp/cleartext-storage-buffer) and "Cleartext storage of sensitive information in an SQLite" (cpp/cleartext-storage-database) queries more accurate.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Improvements have been made to the `cpp/toctou-race-condition` query, both to find more correct results and fewer false positive results.

View File

@@ -1,2 +0,0 @@
lgtm
* Improvements made to the (`cpp/uncontrolled-arithmetic`) query, reducing the frequency of false positive results.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Virtual function specifiers are now accessible via the new predicates on `Function` (`.isDeclaredVirtual`, `.isOverride`, and `.isFinal`).

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Added `Function.hasTrailingReturnType` predicate to check whether a function was declared with a trailing return type.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Added `RoutineType.hasCLinkage` predicate to check whether a function type has "C" language linkage.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Lowered the precision of `cpp/potentially-dangerous-function` so it is run but not displayed on LGTM by default and so it's only run and displayed on Code Scanning if a broader suite like `cpp-security-extended` is opted into.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* Added `Element.getPrimaryQlClasses()` predicate, which gets a comma-separated list of the names of the primary CodeQL classes to which this element belongs.

View File

@@ -1,2 +0,0 @@
lgtm,codescanning
* The query `cpp/implicit-bitfield-downcast` now accounts for C++ reference types, which leads to more true positive results.

View File

@@ -1,4 +0,0 @@
lgtm,codescanning
* The `SimpleRangeAnalysis` library includes information from the
immediate guard for determining the upper bound of a stack
variable for improved accuracy.

View File

@@ -1,4 +0,0 @@
lgtm,codescanning
* The `memberMayBeVarSize` predicate considers more fields to be variable size.
As a result, the "Static buffer overflow" query (cpp/static-buffer-overflow)
produces fewer false positives.

View File

@@ -1,4 +0,0 @@
---
dependencies: {}
compiled: false
lockVersion: 1.0.0

View File

@@ -1,4 +1,3 @@
name: codeql/cpp-examples
version: 0.0.2
dependencies:
codeql/cpp-all: "*"
name: codeql-cpp-examples
version: 0.0.0
libraryPathDependencies: codeql-cpp

View File

@@ -1,4 +0,0 @@
---
dependencies: {}
compiled: false
lockVersion: 1.0.0

View File

@@ -1,7 +0,0 @@
name: codeql/cpp-all
version: 0.0.2
dbscheme: semmlecode.cpp.dbscheme
extractor: cpp
library: true
dependencies:
codeql/cpp-upgrades: 0.0.2

View File

@@ -1,297 +0,0 @@
/**
* Provides the `Element` class, which is the base class for all classes representing C or C++
* program elements.
*/
import semmle.code.cpp.Location
private import semmle.code.cpp.Enclosing
private import semmle.code.cpp.internal.ResolveClass
/**
* Get the `Element` that represents this `@element`.
* Normally this will simply be a cast of `e`, but sometimes it is not.
* For example, for an incomplete struct `e` the result may be a
* complete struct with the same name.
*/
pragma[inline]
Element mkElement(@element e) { unresolveElement(result) = e }
/**
* INTERNAL: Do not use.
*
* Gets an `@element` that resolves to the `Element`. This should
* normally only be called from member predicates, where `e` is not
* `this` and you need the result for an argument to a database
* extensional.
* See `underlyingElement` for when `e` is `this`.
*/
pragma[inline]
@element unresolveElement(Element e) {
not result instanceof @usertype and
result = e
or
e = resolveClass(result)
}
/**
* INTERNAL: Do not use.
*
* Gets the `@element` that this `Element` extends. This should normally
* only be called from member predicates, where `e` is `this` and you
* need the result for an argument to a database extensional.
* See `unresolveElement` for when `e` is not `this`.
*/
@element underlyingElement(Element e) { result = e }
/**
* A C/C++ element with no member predicates other than `toString`. Not for
* general use. This class does not define a location, so classes wanting to
* change their location without affecting other classes can extend
* `ElementBase` instead of `Element` to create a new rootdef for `getURL`,
* `getLocation`, or `hasLocationInfo`.
*/
class ElementBase extends @element {
/** Gets a textual representation of this element. */
cached
string toString() { none() }
/** DEPRECATED: use `getAPrimaryQlClass` instead. */
deprecated string getCanonicalQLClass() { result = this.getAPrimaryQlClass() }
/**
* Gets a comma-separated list of the names of the primary CodeQL classes to which this element belongs.
*/
final string getPrimaryQlClasses() { result = concat(getAPrimaryQlClass(), ",") }
/**
* Gets the name of a primary CodeQL class to which this element belongs.
*
* For most elements, this is simply the most precise syntactic category to
* which they belong; for example, `AddExpr` is a primary class, but
* `BinaryOperation` is not.
*
* This predicate can have multiple results if multiple primary classes match.
* For some elements, this predicate may not have a result.
*/
string getAPrimaryQlClass() { none() }
}
/**
* A C/C++ element. This class is the base class for all C/C++
* elements, such as functions, classes, expressions, and so on.
*/
class Element extends ElementBase {
/** Gets the primary file where this element occurs. */
File getFile() { result = this.getLocation().getFile() }
/**
* Holds if this element may be from source. This predicate holds for all
* elements, except for those in the dummy file, whose name is the empty string.
* The dummy file contains declarations that are built directly into the compiler.
*/
predicate fromSource() { this.getFile().fromSource() }
/**
* Holds if this element may be from a library.
*
* DEPRECATED: always true.
*/
deprecated predicate fromLibrary() { this.getFile().fromLibrary() }
/** Gets the primary location of this element. */
Location getLocation() { none() }
/**
* Gets the source of this element: either itself or a macro that expanded
* to this element.
*
* If the element is not in a macro expansion, then the "root" is just
* the element itself. Otherwise, it is the definition of the innermost
* macro whose expansion the element is in.
*
* This method is useful for filtering macro results in checks: simply
* blame `e.findRootCause` rather than `e`. This will report only bugs
* that are not in macros, and in addition report macros that (somewhere)
* expand to a bug.
*/
Element findRootCause() {
if exists(MacroInvocation mi | this = mi.getAGeneratedElement())
then
exists(MacroInvocation mi |
this = mi.getAGeneratedElement() and
not exists(MacroInvocation closer |
this = closer.getAGeneratedElement() and
mi = closer.getParentInvocation+()
) and
result = mi.getMacro()
)
else result = this
}
/**
* Gets the parent scope of this `Element`, if any.
* A scope is a `Type` (`Class` / `Enum`), a `Namespace`, a `BlockStmt`, a `Function`,
* or certain kinds of `Statement`.
*/
Element getParentScope() {
// result instanceof class
exists(Declaration m |
m = this and
result = m.getDeclaringType() and
not this instanceof EnumConstant
)
or
exists(TemplateClass tc | this = tc.getATemplateArgument() and result = tc)
or
// result instanceof namespace
exists(Namespace n | result = n and n.getADeclaration() = this)
or
exists(FriendDecl d, Namespace n | this = d and n.getADeclaration() = d and result = n)
or
exists(Namespace n | this = n and result = n.getParentNamespace())
or
// result instanceof stmt
exists(LocalVariable v |
this = v and
exists(DeclStmt ds | ds.getADeclaration() = v and result = ds.getParent())
)
or
exists(Parameter p | this = p and result = p.getFunction())
or
exists(GlobalVariable g, Namespace n | this = g and n.getADeclaration() = g and result = n)
or
exists(EnumConstant e | this = e and result = e.getDeclaringEnum())
or
// result instanceof block|function
exists(BlockStmt b | this = b and blockscope(unresolveElement(b), unresolveElement(result)))
or
exists(TemplateFunction tf | this = tf.getATemplateArgument() and result = tf)
or
// result instanceof stmt
exists(ControlStructure s | this = s and result = s.getParent())
or
using_container(unresolveElement(result), underlyingElement(this))
}
/**
* Holds if this element comes from a macro expansion. Only elements that
* are entirely generated by a macro are included - for elements that
* partially come from a macro, see `isAffectedByMacro`.
*/
predicate isInMacroExpansion() { inMacroExpansion(this) }
/**
* Holds if this element is affected in any way by a macro. All elements
* that are totally or partially generated by a macro are included, so
* this is a super-set of `isInMacroExpansion`.
*/
predicate isAffectedByMacro() { affectedByMacro(this) }
private Element getEnclosingElementPref() {
enclosingfunction(underlyingElement(this), unresolveElement(result)) or
result.(Function) = stmtEnclosingElement(this) or
this.(LocalScopeVariable).getFunction() = result or
enumconstants(underlyingElement(this), unresolveElement(result), _, _, _, _) or
derivations(underlyingElement(this), unresolveElement(result), _, _, _) or
stmtparents(underlyingElement(this), _, unresolveElement(result)) or
exprparents(underlyingElement(this), _, unresolveElement(result)) or
namequalifiers(underlyingElement(this), unresolveElement(result), _, _) or
initialisers(underlyingElement(this), unresolveElement(result), _, _) or
exprconv(unresolveElement(result), underlyingElement(this)) or
param_decl_bind(underlyingElement(this), _, unresolveElement(result)) or
using_container(unresolveElement(result), underlyingElement(this)) or
static_asserts(unresolveElement(this), _, _, _, underlyingElement(result))
}
/** Gets the closest `Element` enclosing this one. */
cached
Element getEnclosingElement() {
result = getEnclosingElementPref()
or
not exists(getEnclosingElementPref()) and
(
this = result.(Class).getAMember()
or
result = exprEnclosingElement(this)
or
var_decls(underlyingElement(this), unresolveElement(result), _, _, _)
)
}
/**
* Holds if this `Element` is a part of a template instantiation (but not
* the template itself).
*/
predicate isFromTemplateInstantiation(Element instantiation) {
exists(Element e | isFromTemplateInstantiationRec(e, instantiation) |
this = e or
this.(DeclarationEntry).getDeclaration() = e
)
}
/**
* Holds if this `Element` is part of a template `template` (not if it is
* part of an instantiation of `template`). This means it is represented in
* the database purely as syntax and without guarantees on the presence or
* correctness of type-based operations such as implicit conversions.
*
* If an element is nested within several templates, this predicate holds with
* a value of `template` for each containing template.
*/
predicate isFromUninstantiatedTemplate(Element template) {
exists(Element e | isFromUninstantiatedTemplateRec(e, template) |
this = e or
this.(DeclarationEntry).getDeclaration() = e
)
}
}
private predicate isFromTemplateInstantiationRec(Element e, Element instantiation) {
instantiation.(Function).isConstructedFrom(_) and
e = instantiation
or
instantiation.(Class).isConstructedFrom(_) and
e = instantiation
or
instantiation.(Variable).isConstructedFrom(_) and
e = instantiation
or
isFromTemplateInstantiationRec(e.getEnclosingElement(), instantiation)
}
private predicate isFromUninstantiatedTemplateRec(Element e, Element template) {
is_class_template(unresolveElement(template)) and
e = template
or
is_function_template(unresolveElement(template)) and
e = template
or
is_variable_template(unresolveElement(template)) and
e = template
or
isFromUninstantiatedTemplateRec(e.getEnclosingElement(), template)
}
/**
* A C++11 `static_assert` or C11 `_Static_assert` construct. For example each
* line in the following example contains a static assert:
* ```
* static_assert(sizeof(MyStruct) <= 4096);
* static_assert(sizeof(MyStruct) <= 4096, "MyStruct is too big!");
* ```
*/
class StaticAssert extends Locatable, @static_assert {
override string toString() { result = "static_assert(..., \"" + getMessage() + "\")" }
/**
* Gets the expression which this static assertion ensures is true.
*/
Expr getCondition() { static_asserts(underlyingElement(this), unresolveElement(result), _, _, _) }
/**
* Gets the message which will be reported by the compiler if this static assertion fails.
*/
string getMessage() { static_asserts(underlyingElement(this), _, result, _, _) }
override Location getLocation() { static_asserts(underlyingElement(this), _, _, result, _) }
}

View File

@@ -1,448 +0,0 @@
/**
* Provides classes representing files and folders.
*/
import semmle.code.cpp.Element
import semmle.code.cpp.Declaration
import semmle.code.cpp.metrics.MetricFile
/** A file or folder. */
class Container extends Locatable, @container {
/**
* Gets the absolute, canonical path of this container, using forward slashes
* as path separator.
*
* The path starts with a _root prefix_ followed by zero or more _path
* segments_ separated by forward slashes.
*
* The root prefix is of one of the following forms:
*
* 1. A single forward slash `/` (Unix-style)
* 2. An upper-case drive letter followed by a colon and a forward slash,
* such as `C:/` (Windows-style)
* 3. Two forward slashes, a computer name, and then another forward slash,
* such as `//FileServer/` (UNC-style)
*
* Path segments are never empty (that is, absolute paths never contain two
* contiguous slashes, except as part of a UNC-style root prefix). Also, path
* segments never contain forward slashes, and no path segment is of the
* form `.` (one dot) or `..` (two dots).
*
* Note that an absolute path never ends with a forward slash, except if it is
* a bare root prefix, that is, the path has no path segments. A container
* whose absolute path has no segments is always a `Folder`, not a `File`.
*/
string getAbsolutePath() { none() } // overridden by subclasses
/**
* DEPRECATED: Use `getLocation` instead.
* Gets a URL representing the location of this container.
*
* For more information see [Providing URLs](https://help.semmle.com/QL/learn-ql/ql/locations.html#providing-urls).
*/
deprecated string getURL() { none() } // overridden by subclasses
/**
* Gets the relative path of this file or folder from the root folder of the
* analyzed source location. The relative path of the root folder itself is
* the empty string.
*
* This has no result if the container is outside the source root, that is,
* if the root folder is not a reflexive, transitive parent of this container.
*/
string getRelativePath() {
exists(string absPath, string pref |
absPath = getAbsolutePath() and sourceLocationPrefix(pref)
|
absPath = pref and result = ""
or
absPath = pref.regexpReplaceAll("/$", "") + "/" + result and
not result.matches("/%")
)
}
/**
* Gets the base name of this container including extension, that is, the last
* segment of its absolute path, or the empty string if it has no segments.
*
* Here are some examples of absolute paths and the corresponding base names
* (surrounded with quotes to avoid ambiguity):
*
* <table border="1">
* <tr><th>Absolute path</th><th>Base name</th></tr>
* <tr><td>"/tmp/tst.js"</td><td>"tst.js"</td></tr>
* <tr><td>"C:/Program Files (x86)"</td><td>"Program Files (x86)"</td></tr>
* <tr><td>"/"</td><td>""</td></tr>
* <tr><td>"C:/"</td><td>""</td></tr>
* <tr><td>"D:/"</td><td>""</td></tr>
* <tr><td>"//FileServer/"</td><td>""</td></tr>
* </table>
*/
string getBaseName() {
result = getAbsolutePath().regexpCapture(".*/(([^/]*?)(?:\\.([^.]*))?)", 1)
}
/**
* Gets the extension of this container, that is, the suffix of its base name
* after the last dot character, if any.
*
* In particular,
*
* - if the name does not include a dot, there is no extension, so this
* predicate has no result;
* - if the name ends in a dot, the extension is the empty string;
* - if the name contains multiple dots, the extension follows the last dot.
*
* Here are some examples of absolute paths and the corresponding extensions
* (surrounded with quotes to avoid ambiguity):
*
* <table border="1">
* <tr><th>Absolute path</th><th>Extension</th></tr>
* <tr><td>"/tmp/tst.js"</td><td>"js"</td></tr>
* <tr><td>"/tmp/.classpath"</td><td>"classpath"</td></tr>
* <tr><td>"/bin/bash"</td><td>not defined</td></tr>
* <tr><td>"/tmp/tst2."</td><td>""</td></tr>
* <tr><td>"/tmp/x.tar.gz"</td><td>"gz"</td></tr>
* </table>
*/
string getExtension() { result = getAbsolutePath().regexpCapture(".*/([^/]*?)(\\.([^.]*))?", 3) }
/**
* Gets the stem of this container, that is, the prefix of its base name up to
* (but not including) the last dot character if there is one, or the entire
* base name if there is not.
*
* Here are some examples of absolute paths and the corresponding stems
* (surrounded with quotes to avoid ambiguity):
*
* <table border="1">
* <tr><th>Absolute path</th><th>Stem</th></tr>
* <tr><td>"/tmp/tst.js"</td><td>"tst"</td></tr>
* <tr><td>"/tmp/.classpath"</td><td>""</td></tr>
* <tr><td>"/bin/bash"</td><td>"bash"</td></tr>
* <tr><td>"/tmp/tst2."</td><td>"tst2"</td></tr>
* <tr><td>"/tmp/x.tar.gz"</td><td>"x.tar"</td></tr>
* </table>
*/
string getStem() { result = getAbsolutePath().regexpCapture(".*/([^/]*?)(?:\\.([^.]*))?", 1) }
/** Gets the parent container of this file or folder, if any. */
Container getParentContainer() {
containerparent(unresolveElement(result), underlyingElement(this))
}
/** Gets a file or sub-folder in this container. */
Container getAChildContainer() { this = result.getParentContainer() }
/** Gets a file in this container. */
File getAFile() { result = getAChildContainer() }
/** Gets the file in this container that has the given `baseName`, if any. */
File getFile(string baseName) {
result = getAFile() and
result.getBaseName() = baseName
}
/** Gets a sub-folder in this container. */
Folder getAFolder() { result = getAChildContainer() }
/** Gets the sub-folder in this container that has the given `baseName`, if any. */
Folder getFolder(string baseName) {
result = getAFolder() and
result.getBaseName() = baseName
}
/**
* Gets a textual representation of the path of this container.
*
* This is the absolute path of the container.
*/
override string toString() { result = getAbsolutePath() }
}
/**
* A folder that was observed on disk during the build process.
*
* For the example folder name of "/usr/home/me", the path decomposes to:
*
* 1. "/usr/home" - see `getParentContainer`.
* 2. "me" - see `getBaseName`.
*
* To get the full path, use `getAbsolutePath`.
*/
class Folder extends Container, @folder {
override string getAbsolutePath() { folders(underlyingElement(this), result) }
override Location getLocation() {
result.getContainer() = this and
result.hasLocationInfo(_, 0, 0, 0, 0)
}
override string getAPrimaryQlClass() { result = "Folder" }
/**
* DEPRECATED: Use `getLocation` instead.
* Gets the URL of this folder.
*/
deprecated override string getURL() { result = "file://" + this.getAbsolutePath() + ":0:0:0:0" }
/**
* DEPRECATED: use `getAbsolutePath` instead.
* Gets the name of this folder.
*/
deprecated string getName() { folders(underlyingElement(this), result) }
/**
* DEPRECATED: use `getAbsolutePath` instead.
* Holds if this element is named `name`.
*/
deprecated predicate hasName(string name) { name = this.getName() }
/**
* DEPRECATED: use `getAbsolutePath` instead.
* Gets the full name of this folder.
*/
deprecated string getFullName() { result = this.getName() }
/**
* DEPRECATED: use `getBaseName` instead.
* Gets the last part of the folder name.
*/
deprecated string getShortName() { result = this.getBaseName() }
/**
* DEPRECATED: use `getParentContainer` instead.
* Gets the parent folder.
*/
deprecated Folder getParent() {
containerparent(unresolveElement(result), underlyingElement(this))
}
}
/**
* A file that was observed on disk during the build process.
*
* For the example filename of "/usr/home/me/myprogram.c", the filename
* decomposes to:
*
* 1. "/usr/home/me" - see `getParentContainer`.
* 2. "myprogram.c" - see `getBaseName`.
*
* The base name further decomposes into the _stem_ and _extension_ -- see
* `getStem` and `getExtension`. To get the full path, use `getAbsolutePath`.
*/
class File extends Container, @file {
override string getAbsolutePath() { files(underlyingElement(this), result) }
override string toString() { result = Container.super.toString() }
override string getAPrimaryQlClass() { result = "File" }
override Location getLocation() {
result.getContainer() = this and
result.hasLocationInfo(_, 0, 0, 0, 0)
}
/**
* DEPRECATED: Use `getLocation` instead.
* Gets the URL of this file.
*/
deprecated override string getURL() { result = "file://" + this.getAbsolutePath() + ":0:0:0:0" }
/** Holds if this file was compiled as C (at any point). */
predicate compiledAsC() { fileannotations(underlyingElement(this), 1, "compiled as c", "1") }
/** Holds if this file was compiled as C++ (at any point). */
predicate compiledAsCpp() { fileannotations(underlyingElement(this), 1, "compiled as c++", "1") }
/**
* Holds if this file was compiled by a Microsoft compiler (at any point).
*
* Note: currently unreliable - on some projects only some of the files that
* are compiled by a Microsoft compiler are detected by this predicate.
*/
predicate compiledAsMicrosoft() {
exists(File f, Compilation c |
c.getAFileCompiled() = f and
(
c.getAnArgument() = "--microsoft" or
c.getAnArgument()
.toLowerCase()
.replaceAll("\\", "/")
.matches(["%/cl.exe", "%/clang-cl.exe"])
) and
f.getAnIncludedFile*() = this
)
}
/** Gets a top-level element declared in this file. */
Declaration getATopLevelDeclaration() { result.getAFile() = this and result.isTopLevel() }
/** Gets a declaration in this file. */
Declaration getADeclaration() { result.getAFile() = this }
/** Holds if this file uses the given macro. */
predicate usesMacro(Macro m) {
exists(MacroInvocation mi |
mi.getFile() = this and
mi.getMacro() = m
)
}
/**
* Gets a file that is directly included from this file (using a
* pre-processor directive like `#include`).
*/
File getAnIncludedFile() {
exists(Include i | i.getFile() = this and i.getIncludedFile() = result)
}
/**
* Holds if this file may be from source. This predicate holds for all files
* except the dummy file, whose name is the empty string, which contains
* declarations that are built into the compiler.
*/
override predicate fromSource() { numlines(underlyingElement(this), _, _, _) }
/**
* Holds if this file may be from a library.
*
* DEPRECATED: For historical reasons this is true for any file.
*/
deprecated override predicate fromLibrary() { any() }
/** Gets the metric file. */
MetricFile getMetrics() { result = this }
/**
* Gets the remainder of the base name after the first dot character. Note
* that the name of this predicate is in plural form, unlike `getExtension`,
* which gets the remainder of the base name after the _last_ dot character.
*
* Predicates `getStem` and `getExtension` should be preferred over
* `getShortName` and `getExtensions` since the former pair is compatible
* with the file libraries of other languages.
* Note the slight difference between this predicate and `getStem`:
* for example, for "file.tar.gz", this predicate will have the result
* "tar.gz", while `getExtension` will have the result "gz".
*/
string getExtensions() {
exists(string name, int firstDotPos |
name = this.getBaseName() and
firstDotPos = min([name.indexOf("."), name.length() - 1]) and
result = name.suffix(firstDotPos + 1)
)
}
/**
* Gets the short name of this file, that is, the prefix of its base name up
* to (but not including) the first dot character if there is one, or the
* entire base name if there is not. For example, if the full name is
* "/path/to/filename.a.bcd" then the short name is "filename".
*
* Predicates `getStem` and `getExtension` should be preferred over
* `getShortName` and `getExtensions` since the former pair is compatible
* with the file libraries of other languages.
* Note the slight difference between this predicate and `getStem`:
* for example, for "file.tar.gz", this predicate will have the result
* "file", while `getStem` will have the result "file.tar".
*/
string getShortName() {
exists(string name, int firstDotPos |
name = this.getBaseName() and
firstDotPos = min([name.indexOf("."), name.length()]) and
result = name.prefix(firstDotPos)
)
or
this.getAbsolutePath() = "" and
result = ""
}
}
/**
* Holds if any file was compiled by a Microsoft compiler.
*/
predicate anyFileCompiledAsMicrosoft() { any(File f).compiledAsMicrosoft() }
/**
* A C/C++ header file, as determined (mainly) by file extension.
*
* For the related notion of whether a file is included anywhere (using a
* pre-processor directive like `#include`), use `Include.getIncludedFile`.
*/
class HeaderFile extends File {
HeaderFile() {
this.getExtension().toLowerCase() =
["h", "r", "hpp", "hxx", "h++", "hh", "hp", "tcc", "tpp", "txx", "t++"]
or
not exists(this.getExtension()) and
exists(Include i | i.getIncludedFile() = this)
}
override string getAPrimaryQlClass() { result = "HeaderFile" }
/**
* Holds if this header file does not contain any declaration entries or top level
* declarations. For example it might be:
* - a file containing only preprocessor directives and/or comments
* - an empty file
* - a file that contains non-top level code or data that's included in an
* unusual way
*/
predicate noTopLevelCode() {
not exists(DeclarationEntry de | de.getFile() = this) and
not exists(Declaration d | d.getFile() = this and d.isTopLevel()) and
not exists(UsingEntry ue | ue.getFile() = this)
}
}
/**
* A C source file, as determined by file extension.
*
* For the related notion of whether a file is compiled as C code, use
* `File.compiledAsC`.
*/
class CFile extends File {
CFile() { this.getExtension().toLowerCase() = ["c", "i"] }
override string getAPrimaryQlClass() { result = "CFile" }
}
/**
* A C++ source file, as determined by file extension.
*
* For the related notion of whether a file is compiled as C++ code, use
* `File.compiledAsCpp`.
*/
class CppFile extends File {
CppFile() {
this.getExtension().toLowerCase() =
["cpp", "cxx", "c++", "cc", "cp", "icc", "ipp", "ixx", "i++", "ii"]
// Note: .C files are indistinguishable from .c files on some
// file systems, so we just treat them as CFile's.
}
override string getAPrimaryQlClass() { result = "CppFile" }
}
/**
* DEPRECATED: Objective-C is no longer supported.
* An Objective C source file, as determined by file extension.
*
* For the related notion of whether a file is compiled as Objective C
* code, use `File.compiledAsObjC`.
*/
deprecated class ObjCFile extends File {
ObjCFile() { none() }
}
/**
* DEPRECATED: Objective-C is no longer supported.
* An Objective C++ source file, as determined by file extension.
*
* For the related notion of whether a file is compiled as Objective C++
* code, use `File.compiledAsObjCpp`.
*/
deprecated class ObjCppFile extends File {
ObjCppFile() { none() }
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,608 +0,0 @@
/**
* Provides classes for modeling variables and their declarations.
*/
import semmle.code.cpp.Element
import semmle.code.cpp.exprs.Access
import semmle.code.cpp.Initializer
private import semmle.code.cpp.internal.ResolveClass
/**
* A C/C++ variable. For example, in the following code there are four
* variables, `a`, `b`, `c` and `d`:
* ```
* extern int a;
* int a;
*
* void myFunction(int b) {
* int c;
* }
*
* namespace N {
* extern int d;
* int d = 1;
* }
* ```
*
* For local variables, there is a one-to-one correspondence between
* `Variable` and `VariableDeclarationEntry`.
*
* For other types of variable, there is a one-to-many relationship between
* `Variable` and `VariableDeclarationEntry`. For example, a `Parameter`
* can have multiple declarations.
*/
class Variable extends Declaration, @variable {
override string getAPrimaryQlClass() { result = "Variable" }
/** Gets the initializer of this variable, if any. */
Initializer getInitializer() { result.getDeclaration() = this }
/** Holds if this variable has an initializer. */
predicate hasInitializer() { exists(this.getInitializer()) }
/** Gets an access to this variable. */
VariableAccess getAnAccess() { result.getTarget() = this }
/**
* Gets a specifier of this variable. This includes `extern`, `static`,
* `auto`, `private`, `protected`, `public`. Specifiers of the *type* of
* this variable, such as `const` and `volatile`, are instead accessed
* through `this.getType().getASpecifier()`.
*/
override Specifier getASpecifier() {
varspecifiers(underlyingElement(this), unresolveElement(result))
}
/** Gets an attribute of this variable. */
Attribute getAnAttribute() { varattributes(underlyingElement(this), unresolveElement(result)) }
/** Holds if this variable is `const`. */
predicate isConst() { this.getType().isConst() }
/** Holds if this variable is `volatile`. */
predicate isVolatile() { this.getType().isVolatile() }
/** Gets the name of this variable. */
override string getName() { none() }
/** Gets the type of this variable. */
Type getType() { none() }
/** Gets the type of this variable, after typedefs have been resolved. */
Type getUnderlyingType() { result = this.getType().getUnderlyingType() }
/**
* Gets the type of this variable, after specifiers have been deeply
* stripped and typedefs have been resolved.
*/
Type getUnspecifiedType() { result = this.getType().getUnspecifiedType() }
/**
* Gets the type of this variable prior to deduction caused by the C++11
* `auto` keyword.
*
* If the type of this variable was not declared with the C++11 `auto`
* keyword, then this predicate does not hold.
*
* If the type of this variable is completely `auto`, then `result` is an
* instance of `AutoType`. For example:
*
* `auto four = 4;`
*
* If the type of this variable is partially `auto`, then a descendant of
* `result` is an instance of `AutoType`. For example:
*
* `const auto& c = container;`
*/
Type getTypeWithAuto() { autoderivation(underlyingElement(this), unresolveElement(result)) }
/**
* Holds if the type of this variable is declared using the C++ `auto`
* keyword.
*/
predicate declaredUsingAutoType() { autoderivation(underlyingElement(this), _) }
override VariableDeclarationEntry getADeclarationEntry() { result.getDeclaration() = this }
override Location getADeclarationLocation() { result = getADeclarationEntry().getLocation() }
override VariableDeclarationEntry getDefinition() {
result = getADeclarationEntry() and
result.isDefinition()
}
override Location getDefinitionLocation() { result = getDefinition().getLocation() }
override Location getLocation() {
if exists(getDefinition())
then result = this.getDefinitionLocation()
else result = this.getADeclarationLocation()
}
/**
* Gets an expression that is assigned to this variable somewhere in the
* program.
*/
Expr getAnAssignedValue() {
result = this.getInitializer().getExpr()
or
exists(ConstructorFieldInit cfi | cfi.getTarget() = this and result = cfi.getExpr())
or
exists(AssignExpr ae | ae.getLValue().(Access).getTarget() = this and result = ae.getRValue())
or
exists(ClassAggregateLiteral l | result = l.getFieldExpr(this))
}
/**
* Gets an assignment expression that assigns to this variable.
* For example: `x=...` or `x+=...`.
*
* This does _not_ include the initialization of the variable. Use
* `Variable.getInitializer()` to get the variable's initializer,
* or use `Variable.getAnAssignedValue()` to get an expression that
* is the right-hand side of an assignment or an initialization of
* the varible.
*/
Assignment getAnAssignment() { result.getLValue() = this.getAnAccess() }
/**
* Holds if this variable is `constexpr`.
*/
predicate isConstexpr() { this.hasSpecifier("is_constexpr") }
/**
* Holds if this variable is declared `constinit`.
*/
predicate isConstinit() { this.hasSpecifier("declared_constinit") }
/**
* Holds if this variable is `thread_local`.
*/
predicate isThreadLocal() { this.hasSpecifier("is_thread_local") }
/**
* Holds if this variable is constructed from `v` as a result
* of template instantiation. If so, it originates either from a template
* variable or from a variable nested in a template class.
*/
predicate isConstructedFrom(Variable v) {
variable_instantiation(underlyingElement(this), unresolveElement(v))
}
/**
* Holds if this is a compiler-generated variable. For example, a
* [range-based for loop](http://en.cppreference.com/w/cpp/language/range-for)
* typically has three compiler-generated variables, named `__range`,
* `__begin`, and `__end`:
*
* `for (char c : str) { ... }`
*/
predicate isCompilerGenerated() { compgenerated(underlyingElement(this)) }
}
/**
* A particular declaration or definition of a C/C++ variable. For example, in
* the following code there are six variable declaration entries - two each for
* `a` and `d`, and one each for `b` and `c`:
* ```
* extern int a;
* int a;
*
* void myFunction(int b) {
* int c;
* }
*
* namespace N {
* extern int d;
* int d = 1;
* }
* ```
*/
class VariableDeclarationEntry extends DeclarationEntry, @var_decl {
override Variable getDeclaration() { result = getVariable() }
override string getAPrimaryQlClass() { result = "VariableDeclarationEntry" }
/**
* Gets the variable which is being declared or defined.
*/
Variable getVariable() { var_decls(underlyingElement(this), unresolveElement(result), _, _, _) }
/**
* Gets the name, if any, used for the variable at this declaration or
* definition.
*
* In most cases, this will be the name of the variable itself. The only
* case in which it can differ is in a parameter declaration entry,
* because the parameter may have a different name in the declaration
* than in the definition. For example:
*
* ```
* // Declaration. Parameter is named "x".
* int f(int x);
*
* // Definition. Parameter is named "y".
* int f(int y) { return y; }
* ```
*/
override string getName() { var_decls(underlyingElement(this), _, _, result, _) and result != "" }
/**
* Gets the type of the variable which is being declared or defined.
*/
override Type getType() { var_decls(underlyingElement(this), _, unresolveElement(result), _, _) }
override Location getLocation() { var_decls(underlyingElement(this), _, _, _, result) }
/**
* Holds if this is a definition of a variable.
*
* This always holds for local variables and member variables, but need
* not hold for global variables. In the case of function parameters,
* this holds precisely when the enclosing `FunctionDeclarationEntry` is
* a definition.
*/
override predicate isDefinition() { var_def(underlyingElement(this)) }
override string getASpecifier() { var_decl_specifiers(underlyingElement(this), result) }
}
/**
* A parameter as described within a particular declaration or definition
* of a C/C++ function. For example the declaration of `a` in the following
* code:
* ```
* void myFunction(int a) {
* int b;
* }
* ```
*/
class ParameterDeclarationEntry extends VariableDeclarationEntry {
ParameterDeclarationEntry() { param_decl_bind(underlyingElement(this), _, _) }
override string getAPrimaryQlClass() { result = "ParameterDeclarationEntry" }
/**
* Gets the function declaration or definition which this parameter
* description is part of.
*/
FunctionDeclarationEntry getFunctionDeclarationEntry() {
param_decl_bind(underlyingElement(this), _, unresolveElement(result))
}
/**
* Gets the zero-based index of this parameter.
*/
int getIndex() { param_decl_bind(underlyingElement(this), result, _) }
private string getAnonymousParameterDescription() {
not exists(getName()) and
exists(string idx |
idx =
((getIndex() + 1).toString() + "th")
.replaceAll("1th", "1st")
.replaceAll("2th", "2nd")
.replaceAll("3th", "3rd")
.replaceAll("11st", "11th")
.replaceAll("12nd", "12th")
.replaceAll("13rd", "13th") and
if exists(getCanonicalName())
then result = "declaration of " + getCanonicalName() + " as anonymous " + idx + " parameter"
else result = "declaration of " + idx + " parameter"
)
}
override string toString() {
isDefinition() and
result = "definition of " + getName()
or
not isDefinition() and
if getName() = getCanonicalName()
then result = "declaration of " + getName()
else result = "declaration of " + getCanonicalName() + " as " + getName()
or
result = getAnonymousParameterDescription()
}
/**
* Gets the name of this `ParameterDeclarationEntry` including it's type.
*
* For example: "int p".
*/
string getTypedName() {
exists(string typeString, string nameString |
(if exists(getType().getName()) then typeString = getType().getName() else typeString = "") and
(if exists(getName()) then nameString = getName() else nameString = "") and
if typeString != "" and nameString != ""
then result = typeString + " " + nameString
else result = typeString + nameString
)
}
}
/**
* A C/C++ variable with block scope [N4140 3.3.3]. In other words, a local
* variable or a function parameter. For example, the variables `a`, `b` and
* `c` in the following code:
* ```
* void myFunction(int a) {
* int b;
* static int c;
* }
* ```
*
* See also `StackVariable`, which is the class of local-scope variables
* without statics and thread-locals.
*/
class LocalScopeVariable extends Variable, @localscopevariable {
/** Gets the function to which this variable belongs. */
Function getFunction() { none() } // overridden in subclasses
}
/**
* A C/C++ variable with _automatic storage duration_. In other words, a
* function parameter or a local variable that is not static or thread-local.
* For example, the variables `a` and `b` in the following code.
* ```
* void myFunction(int a) {
* int b;
* static int c;
* }
* ```
*/
class StackVariable extends LocalScopeVariable {
StackVariable() {
not this.isStatic() and
not this.isThreadLocal()
}
}
/**
* A C/C++ local variable. In other words, any variable that has block
* scope [N4140 3.3.3], but is not a parameter of a `Function` or `CatchBlock`.
* For example the variables `b` and `c` in the following code:
* ```
* void myFunction(int a) {
* int b;
* static int c;
* }
* ```
*
* Local variables can be static; use the `isStatic` member predicate to detect
* those.
*
* A local variable can be declared by a `DeclStmt` or a `ConditionDeclExpr`.
*/
class LocalVariable extends LocalScopeVariable, @localvariable {
override string getAPrimaryQlClass() { result = "LocalVariable" }
override string getName() { localvariables(underlyingElement(this), _, result) }
override Type getType() { localvariables(underlyingElement(this), unresolveElement(result), _) }
override Function getFunction() {
exists(DeclStmt s | s.getADeclaration() = this and s.getEnclosingFunction() = result)
or
exists(ConditionDeclExpr e | e.getVariable() = this and e.getEnclosingFunction() = result)
}
}
/**
* A variable whose contents always have static storage duration. This can be a
* global variable, a namespace variable, a static local variable, or a static
* member variable.
*/
class StaticStorageDurationVariable extends Variable {
StaticStorageDurationVariable() {
this instanceof GlobalOrNamespaceVariable
or
this.(LocalVariable).isStatic()
or
this.(MemberVariable).isStatic()
}
/**
* Holds if the initializer for this variable is evaluated at runtime.
*/
predicate hasDynamicInitialization() {
runtimeExprInStaticInitializer(this.getInitializer().getExpr())
}
}
/**
* Holds if `e` is an expression in a static initializer that must be evaluated
* at run time. This predicate computes "is non-const" instead of "is const"
* since computing "is const" for an aggregate literal with many children would
* either involve recursion through `forall` on those children or an iteration
* through the rank numbers of the children, both of which can be slow.
*/
private predicate runtimeExprInStaticInitializer(Expr e) {
inStaticInitializer(e) and
if e instanceof AggregateLiteral // in sync with the cast in `inStaticInitializer`
then runtimeExprInStaticInitializer(e.getAChild())
else not e.getFullyConverted().isConstant()
}
/**
* Holds if `e` is the initializer of a `StaticStorageDurationVariable`, either
* directly or below some top-level `AggregateLiteral`s.
*/
private predicate inStaticInitializer(Expr e) {
exists(StaticStorageDurationVariable var | e = var.getInitializer().getExpr())
or
// The cast to `AggregateLiteral` ensures we only compute what'll later be
// needed by `runtimeExprInStaticInitializer`.
inStaticInitializer(e.getParent().(AggregateLiteral))
}
/**
* A C++ local variable declared as `static`.
*/
class StaticLocalVariable extends LocalVariable, StaticStorageDurationVariable { }
/**
* A C/C++ variable which has global scope or namespace scope. For example the
* variables `a` and `b` in the following code:
* ```
* int a;
*
* namespace N {
* int b;
* }
* ```
*/
class GlobalOrNamespaceVariable extends Variable, @globalvariable {
override string getName() { globalvariables(underlyingElement(this), _, result) }
override Type getType() { globalvariables(underlyingElement(this), unresolveElement(result), _) }
override Element getEnclosingElement() { none() }
}
/**
* A C/C++ variable which has namespace scope. For example the variable `b`
* in the following code:
* ```
* int a;
*
* namespace N {
* int b;
* }
* ```
*/
class NamespaceVariable extends GlobalOrNamespaceVariable {
NamespaceVariable() {
exists(Namespace n | namespacembrs(unresolveElement(n), underlyingElement(this)))
}
override string getAPrimaryQlClass() { result = "NamespaceVariable" }
}
/**
* A C/C++ variable which has global scope. For example the variable `a`
* in the following code:
* ```
* int a;
*
* namespace N {
* int b;
* }
* ```
*
* Note that variables declared in anonymous namespaces have namespace scope,
* even though they are accessed in the same manner as variables declared in
* the enclosing scope of said namespace (which may be the global scope).
*/
class GlobalVariable extends GlobalOrNamespaceVariable {
GlobalVariable() { not this instanceof NamespaceVariable }
override string getAPrimaryQlClass() { result = "GlobalVariable" }
}
/**
* A C structure member or C++ member variable. For example the member
* variables `m` and `s` in the following code:
* ```
* class MyClass {
* public:
* int m;
* static int s;
* };
* ```
*
* This includes static member variables in C++. To exclude static member
* variables, use `Field` instead of `MemberVariable`.
*/
class MemberVariable extends Variable, @membervariable {
MemberVariable() { this.isMember() }
override string getAPrimaryQlClass() { result = "MemberVariable" }
/** Holds if this member is private. */
predicate isPrivate() { this.hasSpecifier("private") }
/** Holds if this member is protected. */
predicate isProtected() { this.hasSpecifier("protected") }
/** Holds if this member is public. */
predicate isPublic() { this.hasSpecifier("public") }
override string getName() { membervariables(underlyingElement(this), _, result) }
override Type getType() {
if strictcount(this.getAType()) = 1
then result = this.getAType()
else
// In rare situations a member variable may have multiple types in
// different translation units. In that case, we return the unspecified
// type.
result = this.getAType().getUnspecifiedType()
}
/** Holds if this member is mutable. */
predicate isMutable() { getADeclarationEntry().hasSpecifier("mutable") }
private Type getAType() { membervariables(underlyingElement(this), unresolveElement(result), _) }
}
/**
* A C/C++ function pointer variable.
*
* DEPRECATED: use `Variable.getType() instanceof FunctionPointerType` instead.
*/
deprecated class FunctionPointerVariable extends Variable {
FunctionPointerVariable() { this.getType() instanceof FunctionPointerType }
}
/**
* A C/C++ function pointer member variable.
*
* DEPRECATED: use `MemberVariable.getType() instanceof FunctionPointerType` instead.
*/
deprecated class FunctionPointerMemberVariable extends MemberVariable {
FunctionPointerMemberVariable() { this instanceof FunctionPointerVariable }
}
/**
* A C++14 variable template. For example, in the following code the variable
* template `v` defines a family of variables:
* ```
* template<class T>
* T v;
* ```
*/
class TemplateVariable extends Variable {
TemplateVariable() { is_variable_template(underlyingElement(this)) }
/**
* Gets an instantiation of this variable template.
*/
Variable getAnInstantiation() { result.isConstructedFrom(this) }
}
/**
* A non-static local variable or parameter that is not part of an
* uninstantiated template. Uninstantiated templates are purely syntax, and
* only on instantiation will they be complete with information about types,
* conversions, call targets, etc. For example in the following code, the
* variables `a` in `myFunction` and `b` in the instantiation
* `myTemplateFunction<int>`, but not `b` in the template
* `myTemplateFunction<T>`:
* ```
* void myFunction() {
* float a;
* }
*
* template<typename T>
* void myTemplateFunction() {
* T b;
* }
*
* ...
*
* myTemplateFunction<int>();
* ```
*/
class SemanticStackVariable extends StackVariable {
SemanticStackVariable() { not this.isFromUninstantiatedTemplate(_) }
}

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

@@ -1,285 +0,0 @@
private import cpp
private import DataFlowUtil
private import DataFlowDispatch
private import FlowVar
/** Gets the instance argument of a non-static call. */
private Node getInstanceArgument(Call call) {
result.asExpr() = call.getQualifier()
or
result.(PreObjectInitializerNode).getExpr().(ConstructorCall) = call
// This does not include the implicit `this` argument on auto-generated
// base class destructor calls as those do not have an AST element.
}
/** An argument to a call. */
private class Argument extends Expr {
Call call;
int pos;
Argument() { call.getArgument(pos) = this }
/** Gets the call that has this argument. */
Call getCall() { result = call }
/** Gets the position of this argument. */
int getPosition() { result = pos }
}
/**
* A data flow node that occurs as the argument of a call and is passed as-is
* to the callable. Arguments that are wrapped in an implicit varargs array
* creation are not included, but the implicitly created array is.
* Instance arguments are also included.
*/
class ArgumentNode extends Node {
ArgumentNode() {
exists(Argument arg | this.asExpr() = arg) or
this = getInstanceArgument(_)
}
/**
* Holds if this argument occurs at the given position in the given call.
* The instance argument is considered to have index `-1`.
*/
predicate argumentOf(DataFlowCall call, int pos) {
exists(Argument arg | this.asExpr() = arg | call = arg.getCall() and pos = arg.getPosition())
or
pos = -1 and this = getInstanceArgument(call)
}
/** Gets the call in which this node is an argument. */
DataFlowCall getCall() { this.argumentOf(result, _) }
}
private newtype TReturnKind =
TNormalReturnKind() or
TRefReturnKind(int i) { exists(Parameter parameter | i = parameter.getIndex()) }
/**
* 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. */
string toString() {
this instanceof TNormalReturnKind and
result = "return"
or
this instanceof TRefReturnKind and
result = "ref"
}
}
/** A data flow node that represents a returned value in the called function. */
abstract class ReturnNode extends Node {
/** Gets the kind of this returned value. */
abstract ReturnKind getKind();
}
/** A `ReturnNode` that occurs as the result of a `ReturnStmt`. */
private class NormalReturnNode extends ReturnNode, ExprNode {
NormalReturnNode() { exists(ReturnStmt ret | this.getExpr() = ret.getExpr()) }
/** Gets the kind of this returned value. */
override ReturnKind getKind() { result = TNormalReturnKind() }
}
/**
* A `ReturnNode` that occurs as a result of a definition of a reference
* parameter reaching the end of a function body.
*/
private class RefReturnNode extends ReturnNode, RefParameterFinalValueNode {
/** Gets the kind of this returned value. */
override ReturnKind getKind() { result = TRefReturnKind(this.getParameter().getIndex()) }
}
/** A data flow node that represents the output of a call at the call site. */
abstract class OutNode extends Node {
/** Gets the underlying call. */
abstract DataFlowCall getCall();
}
private class ExprOutNode extends OutNode, ExprNode {
ExprOutNode() { this.getExpr() instanceof Call }
/** Gets the underlying call. */
override DataFlowCall getCall() { result = this.getExpr() }
}
private class RefOutNode extends OutNode, DefinitionByReferenceOrIteratorNode {
/** Gets the underlying call. */
override DataFlowCall getCall() { result = this.getArgument().getParent() }
}
/**
* Gets a node that can read the value returned from `call` with return kind
* `kind`.
*/
OutNode getAnOutNode(DataFlowCall call, ReturnKind kind) {
result = call.getNode() and
kind = TNormalReturnKind()
or
exists(int i |
result.(DefinitionByReferenceOrIteratorNode).getArgument() = call.getArgument(i) and
kind = TRefReturnKind(i)
)
}
/**
* 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) { none() }
/**
* 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 f, PostUpdateNode node2) {
exists(ClassAggregateLiteral aggr, Field field |
// The following line requires `node2` to be both an `ExprNode` and a
// `PostUpdateNode`, which means it must be an `ObjectInitializerNode`.
node2.asExpr() = aggr and
f.(FieldContent).getField() = field and
aggr.getFieldExpr(field) = node1.asExpr()
)
or
exists(FieldAccess fa |
exists(Assignment a |
node1.asExpr() = a and
a.getLValue() = fa
) and
node2.getPreUpdateNode().asExpr() = fa.getQualifier() and
f.(FieldContent).getField() = fa.getTarget()
)
or
exists(ConstructorFieldInit cfi |
node2.getPreUpdateNode().(PreConstructorInitThis).getConstructorFieldInit() = cfi and
f.(FieldContent).getField() = cfi.getTarget() and
node1.asExpr() = cfi.getExpr()
)
}
/**
* 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 f, Node node2) {
exists(FieldAccess fr |
node1.asExpr() = fr.getQualifier() and
fr.getTarget() = f.(FieldContent).getField() and
fr = node2.asExpr() and
not fr = any(AssignExpr a).getLValue()
)
}
/**
* Holds if values stored inside content `c` are cleared at node `n`.
*/
predicate clearsContent(Node n, Content c) {
none() // stub implementation
}
/** Gets the type of `n` used for type pruning. */
Type getNodeType(Node n) {
suppressUnusedNode(n) and
result instanceof VoidType // stub implementation
}
/** Gets a string representation of a type returned by `getNodeType`. */
string ppReprType(Type 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(Type t1, Type 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
}
class DataFlowCallable = Function;
class DataFlowExpr = Expr;
class DataFlowType = Type;
/** A function call relevant for data flow. */
class DataFlowCall extends Expr {
DataFlowCall() { this instanceof Call }
/**
* Gets the nth argument for this call.
*
* The range of `n` is from `0` to `getNumberOfArguments() - 1`.
*/
Expr getArgument(int n) { result = this.(Call).getArgument(n) }
/** Gets the data flow node corresponding to this call. */
ExprNode getNode() { result.getExpr() = this }
/** Gets the enclosing callable of this call. */
Function getEnclosingCallable() { result = this.getEnclosingFunction() }
}
predicate isUnreachableInCall(Node n, DataFlowCall call) { none() } // stub implementation
int accessPathLimit() { result = 5 }
/** 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` does not require a `PostUpdateNode` as it either cannot be
* modified or its modification cannot be observed, for example if it is a
* freshly created object that is not saved in a variable.
*
* This predicate is only used for consistency checks.
*/
predicate isImmutableOrUnobservable(Node n) {
// Is the null pointer (or something that's not really a pointer)
exists(n.asExpr().getValue())
or
// Isn't a pointer or is a pointer to const
forall(DerivedType dt | dt = n.asExpr().getActualType() |
dt.getBaseType().isConst()
or
dt.getBaseType() instanceof RoutineType
)
// The above list of cases isn't exhaustive, but it narrows down the
// consistency alerts enough that most of them are interesting.
}
/** Holds if `n` should be hidden from path explanations. */
predicate nodeIsHidden(Node n) { none() }
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() }

View File

@@ -1,841 +0,0 @@
/**
* Provides C++-specific definitions for use in the data flow library.
*/
private import cpp
private import semmle.code.cpp.dataflow.internal.FlowVar
private import semmle.code.cpp.models.interfaces.DataFlow
private import semmle.code.cpp.controlflow.Guards
private import semmle.code.cpp.dataflow.internal.AddressFlow
cached
private newtype TNode =
TExprNode(Expr e) or
TPartialDefinitionNode(PartialDefinition pd) or
TPreObjectInitializerNode(Expr e) {
e instanceof ConstructorCall
or
e instanceof ClassAggregateLiteral
} or
TExplicitParameterNode(Parameter p) { exists(p.getFunction().getBlock()) } or
TInstanceParameterNode(MemberFunction f) { exists(f.getBlock()) and not f.isStatic() } or
TPreConstructorInitThis(ConstructorFieldInit cfi) or
TPostConstructorInitThis(ConstructorFieldInit cfi) or
TInnerPartialDefinitionNode(Expr e) {
exists(PartialDefinition def, Expr outer |
def.definesExpressions(e, outer) and
// This condition ensures that we don't get two post-update nodes sharing
// the same pre-update node.
e != outer
)
} or
TUninitializedNode(LocalVariable v) { not v.hasInitializer() } or
TRefParameterFinalValueNode(Parameter p) { exists(FlowVar var | var.reachesRefParameter(p)) }
/**
* A node in a data flow graph.
*
* A node can be either an expression, a parameter, or an uninitialized local
* variable. Such nodes are created with `DataFlow::exprNode`,
* `DataFlow::parameterNode`, and `DataFlow::uninitializedNode` respectively.
*/
class Node extends TNode {
/** Gets the function to which this node belongs. */
Function getFunction() { none() } // overridden in subclasses
/**
* INTERNAL: Do not use. Alternative name for `getFunction`.
*/
final Function getEnclosingCallable() { result = this.getFunction() }
/** Gets the type of this node. */
Type getType() { none() } // overridden in subclasses
/**
* Gets the expression corresponding to this node, if any. This predicate
* only has a result on nodes that represent the value of evaluating the
* expression. For data flowing _out of_ an expression, like when an
* argument is passed by reference, use `asDefiningArgument` instead of
* `asExpr`.
*/
Expr asExpr() { result = this.(ExprNode).getExpr() }
/** Gets the parameter corresponding to this node, if any. */
Parameter asParameter() { result = this.(ExplicitParameterNode).getParameter() }
/**
* Gets the argument that defines this `DefinitionByReferenceNode`, if any.
* This predicate should be used instead of `asExpr` when referring to the
* value of a reference argument _after_ the call has returned. For example,
* in `f(&x)`, this predicate will have `&x` as its result for the `Node`
* that represents the new value of `x`.
*/
Expr asDefiningArgument() { result = this.(DefinitionByReferenceNode).getArgument() }
/**
* Gets the expression that is partially defined by this node, if any.
*
* Partial definitions are created for field stores (`x.y = taint();` is a partial
* definition of `x`), and for calls that may change the value of an object (so
* `x.set(taint())` is a partial definition of `x`, and `transfer(&x, taint())` is
* a partial definition of `&x`).
*/
Expr asPartialDefinition() {
this.(PartialDefinitionNode).getPartialDefinition().definesExpressions(_, result)
}
/**
* Gets the uninitialized local variable corresponding to this node, if
* any.
*/
LocalVariable asUninitialized() { result = this.(UninitializedNode).getLocalVariable() }
/** Gets a textual representation of this element. */
string toString() { none() } // overridden by subclasses
/** Gets the location of this element. */
Location getLocation() { none() } // overridden by subclasses
/**
* Holds if this element is at the specified location.
* The location spans column `startcolumn` of line `startline` to
* column `endcolumn` of line `endline` in file `filepath`.
* For more information, see
* [Locations](https://help.semmle.com/QL/learn-ql/ql/locations.html).
*/
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
getLocation().hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
/**
* Gets an upper bound on the type of this node.
*/
Type getTypeBound() { result = getType() }
}
/**
* An expression, viewed as a node in a data flow graph.
*/
class ExprNode extends Node, TExprNode {
Expr expr;
ExprNode() { this = TExprNode(expr) }
override Function getFunction() { result = expr.getEnclosingFunction() }
override Type getType() { result = expr.getType() }
override string toString() { result = expr.toString() }
override Location getLocation() { result = expr.getLocation() }
/** Gets the expression corresponding to this node. */
Expr getExpr() { result = expr }
}
abstract class ParameterNode extends Node, TNode {
/**
* Holds if this node is the parameter of `c` at the specified (zero-based)
* position. The implicit `this` parameter is considered to have index `-1`.
*/
abstract predicate isParameterOf(Function f, int i);
}
/**
* The value of a parameter at function entry, viewed as a node in a data
* flow graph.
*/
class ExplicitParameterNode extends ParameterNode, TExplicitParameterNode {
Parameter param;
ExplicitParameterNode() { this = TExplicitParameterNode(param) }
override Function getFunction() { result = param.getFunction() }
override Type getType() { result = param.getType() }
override string toString() { result = param.toString() }
override Location getLocation() { result = param.getLocation() }
/** Gets the parameter corresponding to this node. */
Parameter getParameter() { result = param }
override predicate isParameterOf(Function f, int i) { f.getParameter(i) = param }
}
class ImplicitParameterNode extends ParameterNode, TInstanceParameterNode {
MemberFunction f;
ImplicitParameterNode() { this = TInstanceParameterNode(f) }
override Function getFunction() { result = f }
override Type getType() { result = f.getDeclaringType() }
override string toString() { result = "this" }
override Location getLocation() { result = f.getLocation() }
override predicate isParameterOf(Function fun, int i) { f = fun and i = -1 }
}
/**
* INTERNAL: do not use.
*
* A node that represents the value of a variable after a function call that
* may have changed the variable because it's passed by reference or because an
* iterator for it was passed by value or by reference.
*/
class DefinitionByReferenceOrIteratorNode extends PartialDefinitionNode {
Expr inner;
Expr argument;
DefinitionByReferenceOrIteratorNode() {
this.getPartialDefinition().definesExpressions(inner, argument) and
(
this.getPartialDefinition() instanceof DefinitionByReference
or
this.getPartialDefinition() instanceof DefinitionByIterator
)
}
override Function getFunction() { result = inner.getEnclosingFunction() }
override Type getType() { result = inner.getType() }
override Location getLocation() { result = argument.getLocation() }
override ExprNode getPreUpdateNode() { result.getExpr() = argument }
/** Gets the argument corresponding to this node. */
Expr getArgument() { result = argument }
/** Gets the parameter through which this value is assigned. */
Parameter getParameter() {
exists(FunctionCall call, int i |
argument = call.getArgument(i) and
result = call.getTarget().getParameter(i)
)
}
}
/**
* A node that represents the value of a variable after a function call that
* may have changed the variable because it's passed by reference.
*
* A typical example would be a call `f(&x)`. Firstly, there will be flow into
* `x` from previous definitions of `x`. Secondly, there will be a
* `DefinitionByReferenceNode` to represent the value of `x` after the call has
* returned. This node will have its `getArgument()` equal to `&x`.
*/
class DefinitionByReferenceNode extends DefinitionByReferenceOrIteratorNode {
override VariablePartialDefinition pd;
override string toString() { result = "ref arg " + argument.toString() }
}
/**
* The value of an uninitialized local variable, viewed as a node in a data
* flow graph.
*/
class UninitializedNode extends Node, TUninitializedNode {
LocalVariable v;
UninitializedNode() { this = TUninitializedNode(v) }
override Function getFunction() { result = v.getFunction() }
override Type getType() { result = v.getType() }
override string toString() { result = v.toString() }
override Location getLocation() { result = v.getLocation() }
/** Gets the uninitialized local variable corresponding to this node. */
LocalVariable getLocalVariable() { result = v }
}
/** INTERNAL: do not use. The final value of a non-const ref parameter. */
class RefParameterFinalValueNode extends Node, TRefParameterFinalValueNode {
Parameter p;
RefParameterFinalValueNode() { this = TRefParameterFinalValueNode(p) }
override Function getFunction() { result = p.getFunction() }
override Type getType() { result = p.getType() }
override string toString() { result = p.toString() }
override Location getLocation() { result = p.getLocation() }
Parameter getParameter() { result = p }
}
/**
* A node associated with an object after an operation that might have
* changed its state.
*
* This can be either the argument to a callable after the callable returns
* (which might have mutated the argument), or the qualifier of a field after
* an update to the field.
*
* Nodes corresponding to AST elements, for example `ExprNode`, usually refer
* to the value before the update with the exception of `ClassInstanceExpr`,
* which represents the value after the constructor has run.
*/
abstract class PostUpdateNode extends Node {
/**
* Gets the node before the state update.
*/
abstract Node getPreUpdateNode();
override Function getFunction() { result = getPreUpdateNode().getFunction() }
override Type getType() { result = getPreUpdateNode().getType() }
override Location getLocation() { result = getPreUpdateNode().getLocation() }
}
abstract private class PartialDefinitionNode extends PostUpdateNode, TPartialDefinitionNode {
PartialDefinition pd;
PartialDefinitionNode() { this = TPartialDefinitionNode(pd) }
override Location getLocation() { result = pd.getActualLocation() }
PartialDefinition getPartialDefinition() { result = pd }
override string toString() { result = getPreUpdateNode().toString() + " [post update]" }
}
private class VariablePartialDefinitionNode extends PartialDefinitionNode {
override VariablePartialDefinition pd;
override Node getPreUpdateNode() { pd.definesExpressions(_, result.asExpr()) }
}
/**
* INTERNAL: do not use.
*
* A synthetic data flow node used for flow into a collection when an iterator
* write occurs in a callee.
*/
private class IteratorPartialDefinitionNode extends PartialDefinitionNode {
override IteratorPartialDefinition pd;
override Node getPreUpdateNode() { pd.definesExpressions(_, result.asExpr()) }
}
/**
* A post-update node on the `e->f` in `f(&e->f)` (and other forms).
*/
private class InnerPartialDefinitionNode extends TInnerPartialDefinitionNode, PostUpdateNode {
Expr e;
InnerPartialDefinitionNode() { this = TInnerPartialDefinitionNode(e) }
override ExprNode getPreUpdateNode() { result.getExpr() = e }
override Function getFunction() { result = e.getEnclosingFunction() }
override Type getType() { result = e.getType() }
override string toString() { result = e.toString() + " [inner post update]" }
override Location getLocation() { result = e.getLocation() }
}
/**
* A node representing the temporary value of an object that was just
* constructed by a constructor call or an aggregate initializer. This is only
* for objects, not for pointers to objects.
*
* These expressions are their own post-update nodes but instead have synthetic
* pre-update nodes.
*/
private class ObjectInitializerNode extends PostUpdateNode, TExprNode {
PreObjectInitializerNode pre;
ObjectInitializerNode() {
// If a `Node` is associated with a `PreObjectInitializerNode`, then it's
// an `ObjectInitializerNode`.
pre.getExpr() = this.asExpr()
}
override PreObjectInitializerNode getPreUpdateNode() { result = pre }
// No override of `toString` since these nodes already have a `toString` from
// their overlap with `ExprNode`.
}
/**
* INTERNAL: do not use.
*
* A synthetic data-flow node that plays the role of a temporary object that
* has not yet been initialized.
*/
class PreObjectInitializerNode extends Node, TPreObjectInitializerNode {
Expr getExpr() { this = TPreObjectInitializerNode(result) }
override Function getFunction() { result = getExpr().getEnclosingFunction() }
override Type getType() { result = getExpr().getType() }
override Location getLocation() { result = getExpr().getLocation() }
override string toString() { result = getExpr().toString() + " [pre init]" }
}
/**
* A synthetic data-flow node that plays the role of the post-update `this`
* pointer in a `ConstructorFieldInit`. For example, the `x(1)` in
* `C() : x(1) { }` is roughly equivalent to `this.x = 1`, and this node is
* equivalent to the `this` _after_ the field has been assigned.
*/
private class PostConstructorInitThis extends PostUpdateNode, TPostConstructorInitThis {
override PreConstructorInitThis getPreUpdateNode() {
this = TPostConstructorInitThis(result.getConstructorFieldInit())
}
override string toString() {
result = getPreUpdateNode().getConstructorFieldInit().toString() + " [post-this]"
}
}
/**
* INTERNAL: do not use.
*
* A synthetic data-flow node that plays the role of the pre-update `this`
* pointer in a `ConstructorFieldInit`. For example, the `x(1)` in
* `C() : x(1) { }` is roughly equivalent to `this.x = 1`, and this node is
* equivalent to the `this` _before_ the field has been assigned.
*/
class PreConstructorInitThis extends Node, TPreConstructorInitThis {
ConstructorFieldInit getConstructorFieldInit() { this = TPreConstructorInitThis(result) }
override Constructor getFunction() { result = getConstructorFieldInit().getEnclosingFunction() }
override PointerType getType() {
result.getBaseType() = getConstructorFieldInit().getEnclosingFunction().getDeclaringType()
}
override Location getLocation() { result = getConstructorFieldInit().getLocation() }
override string toString() { result = getConstructorFieldInit().toString() + " [pre-this]" }
}
/**
* Gets the `Node` corresponding to the value of evaluating `e`. For data
* flowing _out of_ an expression, like when an argument is passed by
* reference, use `definitionByReferenceNodeFromArgument` instead.
*/
ExprNode exprNode(Expr e) { result.getExpr() = e }
/**
* Gets the `Node` corresponding to the value of `p` at function entry.
*/
ParameterNode parameterNode(Parameter p) { result.(ExplicitParameterNode).getParameter() = p }
/**
* Gets the `Node` corresponding to a definition by reference of the variable
* that is passed as `argument` of a call.
*/
DefinitionByReferenceNode definitionByReferenceNodeFromArgument(Expr argument) {
result.getArgument() = argument
}
/**
* Gets the `Node` corresponding to the value of an uninitialized local
* variable `v`.
*/
UninitializedNode uninitializedNode(LocalVariable v) { result.getLocalVariable() = v }
private module ThisFlow {
/**
* Gets the 0-based index of `thisNode` in `b`, where `thisNode` is an access
* to `this` that may or may not have an associated `PostUpdateNode`. To make
* room for synthetic nodes that access `this`, the index may not correspond
* to an actual `ControlFlowNode`.
*/
private int basicBlockThisIndex(BasicBlock b, Node thisNode) {
// The implicit `this` parameter node is given a very negative offset to
// make space for any `ConstructorFieldInit`s there may be between it and
// the block contents.
thisNode.(ImplicitParameterNode).getFunction().getBlock() = b and
result = -2147483648
or
// Place the synthetic `this` node for a `ConstructorFieldInit` at a
// negative offset in the first basic block, between the
// `ImplicitParameterNode` and the first statement.
exists(Constructor constructor, int i |
thisNode.(PreConstructorInitThis).getConstructorFieldInit() = constructor.getInitializer(i) and
result = -2147483648 + 1 + i and
b = thisNode.getFunction().getBlock()
)
or
b.getNode(result) = thisNode.asExpr().(ThisExpr)
}
private int thisRank(BasicBlock b, Node thisNode) {
thisNode = rank[result](Node n, int i | i = basicBlockThisIndex(b, n) | n order by i)
}
private int lastThisRank(BasicBlock b) { result = max(thisRank(b, _)) }
private predicate thisAccessBlockReaches(BasicBlock b1, BasicBlock b2) {
exists(basicBlockThisIndex(b1, _)) and b2 = b1.getASuccessor()
or
exists(BasicBlock mid |
thisAccessBlockReaches(b1, mid) and
b2 = mid.getASuccessor() and
not exists(basicBlockThisIndex(mid, _))
)
}
predicate adjacentThisRefs(Node n1, Node n2) {
exists(BasicBlock b | thisRank(b, n1) + 1 = thisRank(b, n2))
or
exists(BasicBlock b1, BasicBlock b2 |
lastThisRank(b1) = thisRank(b1, n1) and
thisAccessBlockReaches(b1, b2) and
thisRank(b2, n2) = 1
)
}
}
/**
* Holds if data flows from `nodeFrom` to `nodeTo` in exactly one local
* (intra-procedural) step.
*/
cached
predicate localFlowStep(Node nodeFrom, Node nodeTo) {
simpleLocalFlowStep(nodeFrom, nodeTo)
or
// Field flow is not strictly a "step" but covers the whole function
// transitively. There's no way to get a step-like relation out of the global
// data flow library, so we just have to accept some big steps here.
FieldFlow::fieldFlow(nodeFrom, nodeTo)
}
/**
* INTERNAL: do not use.
*
* This is the local flow predicate that's used as a building block in global
* data flow. It may have less flow than the `localFlowStep` predicate.
*/
predicate simpleLocalFlowStep(Node nodeFrom, Node nodeTo) {
// Expr -> Expr
exprToExprStep_nocfg(nodeFrom.asExpr(), nodeTo.asExpr())
or
// Assignment -> LValue post-update node
//
// This is used for assignments whose left-hand side is not a variable
// assignment or a storeStep but is still modeled by other means. It could be
// a call to `operator*` or `operator[]` where taint should flow to the
// post-update node of the qualifier.
exists(AssignExpr assign |
nodeFrom.asExpr() = assign and
nodeTo.(PostUpdateNode).getPreUpdateNode().asExpr() = assign.getLValue()
)
or
// Node -> FlowVar -> VariableAccess
exists(FlowVar var |
(
exprToVarStep(nodeFrom.asExpr(), var)
or
varSourceBaseCase(var, nodeFrom.asParameter())
or
varSourceBaseCase(var, nodeFrom.asUninitialized())
or
var.definedPartiallyAt(nodeFrom.asPartialDefinition())
) and
varToNodeStep(var, nodeTo)
)
or
// Expr -> DefinitionByReferenceNode
exprToDefinitionByReferenceStep(nodeFrom.asExpr(), nodeTo.asDefiningArgument())
or
// `this` -> adjacent-`this`
ThisFlow::adjacentThisRefs(nodeFrom, nodeTo)
or
// post-update-`this` -> following-`this`-ref
ThisFlow::adjacentThisRefs(nodeFrom.(PostUpdateNode).getPreUpdateNode(), nodeTo)
or
// In `f(&x->a)`, this step provides the flow from post-`&` to post-`x->a`,
// from which there is field flow to `x` via reverse read.
exists(PartialDefinition def, Expr inner, Expr outer |
def.definesExpressions(inner, outer) and
inner = nodeTo.(InnerPartialDefinitionNode).getPreUpdateNode().asExpr() and
outer = nodeFrom.(PartialDefinitionNode).getPreUpdateNode().asExpr()
)
or
// Reverse flow: data that flows from the post-update node of a reference
// returned by a function call, back into the qualifier of that function.
// This allows data to flow 'in' through references returned by a modeled
// function such as `operator[]`.
exists(DataFlowFunction f, Call call, FunctionInput inModel, FunctionOutput outModel |
call.getTarget() = f and
inModel.isReturnValueDeref() and
outModel.isQualifierObject() and
f.hasDataFlow(inModel, outModel) and
nodeFrom.(PostUpdateNode).getPreUpdateNode().asExpr() = call and
nodeTo.asDefiningArgument() = call.getQualifier()
)
}
/**
* Holds if data flows from `source` to `sink` in zero or more local
* (intra-procedural) steps.
*/
predicate localFlow(Node source, Node sink) { localFlowStep*(source, sink) }
/**
* Holds if data can flow from `e1` to `e2` in zero or more
* local (intra-procedural) steps.
*/
predicate localExprFlow(Expr e1, Expr e2) { localFlow(exprNode(e1), exprNode(e2)) }
/**
* Holds if the initial value of `v`, if it is a source, flows to `var`.
*/
private predicate varSourceBaseCase(FlowVar var, Variable v) { var.definedByInitialValue(v) }
/**
* Holds if `var` is defined by an assignment-like operation that causes flow
* directly from `assignedExpr` to `var`, _and_ `assignedExpr` evaluates to
* the same value as what is assigned to `var`.
*/
private predicate exprToVarStep(Expr assignedExpr, FlowVar var) {
exists(ControlFlowNode operation |
var.definedByExpr(assignedExpr, operation) and
not operation instanceof PostfixCrementOperation
)
}
/**
* Holds if the node `n` is an access of the variable `var`.
*/
private predicate varToNodeStep(FlowVar var, Node n) {
n.asExpr() = var.getAnAccess()
or
var.reachesRefParameter(n.(RefParameterFinalValueNode).getParameter())
}
/**
* Holds if data flows from `fromExpr` to `toExpr` directly, in the case
* where `toExpr` is the immediate AST parent of `fromExpr`. For example,
* data flows from `x` and `y` to `b ? x : y`.
*/
private predicate exprToExprStep_nocfg(Expr fromExpr, Expr toExpr) {
toExpr = any(ConditionalExpr cond | fromExpr = cond.getThen() or fromExpr = cond.getElse())
or
toExpr = any(AssignExpr assign | fromExpr = assign.getRValue())
or
toExpr = any(CommaExpr comma | fromExpr = comma.getRightOperand())
or
toExpr = any(PostfixCrementOperation op | fromExpr = op.getOperand())
or
toExpr = any(StmtExpr stmtExpr | fromExpr = stmtExpr.getResultExpr())
or
toExpr.(AddressOfExpr).getOperand() = fromExpr
or
// This rule enables flow from an array to its elements. Example: `a` to
// `a[i]` or `*a`, where `a` is an array type. It does not enable flow from a
// pointer to its indirection as in `p[i]` where `p` is a pointer type.
exists(Expr toConverted |
variablePartiallyAccessed(fromExpr, toConverted) and
toExpr = toConverted.getUnconverted() and
not toExpr = fromExpr
)
or
toExpr.(BuiltInOperationBuiltInAddressOf).getOperand() = fromExpr
or
// The following case is needed to track the qualifier object for flow
// through fields. It gives flow from `T(x)` to `new T(x)`. That's not
// strictly _data_ flow but _taint_ flow because the type of `fromExpr` is
// `T` while the type of `toExpr` is `T*`.
//
// This discrepancy is an artifact of how `new`-expressions are represented
// in the database in a way that slightly varies from what the standard
// specifies. In the C++ standard, there is no constructor call expression
// `T(x)` after `new`. Instead there is a type `T` and an optional
// initializer `(x)`.
toExpr.(NewExpr).getInitializer() = fromExpr
or
// A lambda expression (`[captures](params){body}`) is just a thin wrapper
// around the desugared closure creation in the form of a
// `ClassAggregateLiteral` (`{ capture1, ..., captureN }`).
toExpr.(LambdaExpression).getInitializer() = fromExpr
or
// Data flow through a function model.
toExpr =
any(Call call |
exists(DataFlowFunction f, FunctionInput inModel, FunctionOutput outModel |
f.hasDataFlow(inModel, outModel) and
(
exists(int iIn |
inModel.isParameterDeref(iIn) and
call.passesByReference(iIn, fromExpr)
)
or
exists(int iIn |
inModel.isParameter(iIn) and
fromExpr = call.getArgument(iIn)
)
or
inModel.isQualifierObject() and
fromExpr = call.getQualifier()
or
inModel.isQualifierAddress() and
fromExpr = call.getQualifier()
) and
call.getTarget() = f and
// AST dataflow treats a reference as if it were the referred-to object, while the dataflow
// models treat references as pointers. If the return type of the call is a reference, then
// look for data flow the the referred-to object, rather than the reference itself.
if call.getType().getUnspecifiedType() instanceof ReferenceType
then outModel.isReturnValueDeref()
else outModel.isReturnValue()
)
)
}
private predicate exprToDefinitionByReferenceStep(Expr exprIn, Expr argOut) {
exists(DataFlowFunction f, Call call, FunctionOutput outModel, int argOutIndex |
call.getTarget() = f and
argOut = call.getArgument(argOutIndex) and
outModel.isParameterDeref(argOutIndex) and
exists(int argInIndex, FunctionInput inModel | f.hasDataFlow(inModel, outModel) |
inModel.isParameterDeref(argInIndex) and
call.passesByReference(argInIndex, exprIn)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
)
}
private module FieldFlow {
private import DataFlowImplCommon
private import DataFlowImplLocal
private import DataFlowPrivate
/**
* A configuration for finding local-only flow through fields. This uses the
* `Configuration` class in the dedicated `DataFlowImplLocal` copy of the
* shared library that's not user-exposed directly.
*
* To keep the flow local to a single function, we put barriers on parameters
* and return statements. Sources and sinks are the values that go into and
* out of fields, respectively.
*/
private class FieldConfiguration extends Configuration {
FieldConfiguration() { this = "FieldConfiguration" }
override predicate isSource(Node source) {
storeStep(source, _, _)
or
// Also mark `foo(a.b);` as a source when `a.b` may be overwritten by `foo`.
readStep(_, _, any(Node node | node.asExpr() = source.asDefiningArgument()))
}
override predicate isSink(Node sink) { readStep(_, _, sink) }
override predicate isBarrier(Node node) { node instanceof ParameterNode }
override predicate isBarrierOut(Node node) {
node.asExpr().getParent() instanceof ReturnStmt
or
node.asExpr().getParent() instanceof ThrowExpr
}
}
predicate fieldFlow(Node node1, Node node2) {
exists(FieldConfiguration cfg | cfg.hasFlow(node1, node2)) and
// This configuration should not be able to cross function boundaries, but
// we double-check here just to be sure.
getNodeEnclosingCallable(node1) = getNodeEnclosingCallable(node2)
}
}
VariableAccess getAnAccessToAssignedVariable(Expr assign) {
(
assign instanceof Assignment
or
assign instanceof CrementOperation
) and
exists(FlowVar var |
var.definedByExpr(_, assign) and
result = var.getAnAccess()
)
}
private newtype TContent =
TFieldContent(Field f) or
TCollectionContent() or
TArrayContent()
/**
* A description of the way data may be stored inside an object. Examples
* include instance fields, the contents of a collection object, or the contents
* of an array.
*/
class Content extends TContent {
/** Gets a textual representation of this element. */
abstract string toString();
predicate hasLocationInfo(string path, int sl, int sc, int el, int ec) {
path = "" and sl = 0 and sc = 0 and el = 0 and ec = 0
}
}
/** A reference through an instance field. */
class FieldContent extends Content, TFieldContent {
Field f;
FieldContent() { this = TFieldContent(f) }
Field getField() { result = f }
override string toString() { result = f.toString() }
override predicate hasLocationInfo(string path, int sl, int sc, int el, int ec) {
f.getLocation().hasLocationInfo(path, sl, sc, el, ec)
}
}
/** A reference through an array. */
private class ArrayContent extends Content, TArrayContent {
override string toString() { result = "[]" }
}
/** A reference through the contents of some collection-like container. */
private class CollectionContent extends Content, TCollectionContent {
override string toString() { result = "<element>" }
}
/**
* A guard that validates some expression.
*
* To use this in a configuration, extend the class and provide a
* characteristic predicate precisely specifying the guard, and override
* `checks` to specify what is being validated and in which branch.
*
* It is important that all extending classes in scope are disjoint.
*/
class BarrierGuard extends GuardCondition {
/** Override this predicate to hold if this guard validates `e` upon evaluating to `b`. */
abstract predicate checks(Expr e, boolean b);
/** Gets a node guarded by this guard. */
final ExprNode getAGuardedNode() {
exists(SsaDefinition def, Variable v, boolean branch |
result.getExpr() = def.getAUse(v) and
this.checks(def.getAUse(v), branch) and
this.controls(result.getExpr().getBasicBlock(), branch)
)
}
}

View File

@@ -1,279 +0,0 @@
/**
* 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.
*/
private import semmle.code.cpp.models.interfaces.DataFlow
private import semmle.code.cpp.models.interfaces.Taint
private import semmle.code.cpp.models.interfaces.Iterator
private import semmle.code.cpp.models.interfaces.PointerWrapper
private module DataFlow {
import semmle.code.cpp.dataflow.internal.DataFlowUtil
}
/**
* Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
* (intra-procedural) step.
*/
predicate localTaintStep(DataFlow::Node src, DataFlow::Node sink) {
DataFlow::localFlowStep(src, sink) or
localAdditionalTaintStep(src, sink)
}
/**
* Holds if the additional step from `src` to `sink` should be included in all
* global taint flow configurations.
*/
predicate defaultAdditionalTaintStep(DataFlow::Node src, DataFlow::Node sink) {
localAdditionalTaintStep(src, sink)
}
/**
* Holds if default `TaintTracking::Configuration`s should allow implicit reads
* of `c` at sinks and inputs to additional taint steps.
*/
bindingset[node]
predicate defaultImplicitTaintRead(DataFlow::Node node, DataFlow::Content c) { none() }
/**
* Holds if `node` should be a sanitizer in all global taint flow configurations
* but not in local taint.
*/
predicate defaultTaintSanitizer(DataFlow::Node node) { none() }
/**
* Holds if taint can flow in one local step from `nodeFrom` to `nodeTo` excluding
* local data flow steps. That is, `nodeFrom` and `nodeTo` are likely to represent
* different objects.
*/
cached
predicate localAdditionalTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
// Taint can flow through expressions that alter the value but preserve
// more than one bit of it _or_ expressions that follow data through
// pointer indirections.
exists(Expr exprFrom, Expr exprTo |
exprFrom = nodeFrom.asExpr() and
exprTo = nodeTo.asExpr()
|
exprFrom = exprTo.getAChild() and
not noParentExprFlow(exprFrom, exprTo) and
not noFlowFromChildExpr(exprTo)
or
// Taint can flow from the `x` variable in `x++` to all subsequent
// accesses to the unmodified `x` variable.
//
// `DataFlow` without taint specifies flow from `++x` and `x += 1` into the
// variable `x` and thus into subsequent accesses because those expressions
// compute the same value as `x`. This is not the case for `x++`, which
// computes a different value, so we have to add that ourselves for taint
// tracking. The flow from expression `x` into `x++` etc. is handled in the
// case above.
exprTo = DataFlow::getAnAccessToAssignedVariable(exprFrom.(PostfixCrementOperation))
or
// In `for (char c : s) { ... c ... }`, this rule propagates taint from `s`
// to `c`.
exists(RangeBasedForStmt rbf |
exprFrom = rbf.getRange() and
// It's guaranteed up to at least C++20 that the range-based for loop
// desugars to a variable with an initializer.
exprTo = rbf.getVariable().getInitializer().getExpr()
)
)
or
// Taint can flow through modeled functions
exprToExprStep(nodeFrom.asExpr(), nodeTo.asExpr())
or
exprToDefinitionByReferenceStep(nodeFrom.asExpr(), nodeTo.asDefiningArgument())
or
exprToPartialDefinitionStep(nodeFrom.asExpr(), nodeTo.asPartialDefinition())
or
// Reverse taint: taint that flows from the post-update node of a reference
// returned by a function call, back into the qualifier of that function.
// This allows taint to flow 'in' through references returned by a modeled
// function such as `operator[]`.
exists(TaintFunction f, Call call, FunctionInput inModel, FunctionOutput outModel |
call.getTarget() = f and
inModel.isReturnValueDeref() and
nodeFrom.(DataFlow::PostUpdateNode).getPreUpdateNode().asExpr() = call and
f.hasTaintFlow(inModel, outModel) and
(
outModel.isQualifierObject() and
nodeTo.asDefiningArgument() = call.getQualifier()
or
exists(int argOutIndex |
outModel.isParameterDeref(argOutIndex) and
nodeTo.asDefiningArgument() = call.getArgument(argOutIndex)
)
)
)
}
/**
* Holds if taint may propagate from `source` to `sink` in zero or more local
* (intra-procedural) steps.
*/
predicate localTaint(DataFlow::Node source, DataFlow::Node sink) { localTaintStep*(source, sink) }
/**
* Holds if taint can flow from `e1` to `e2` in zero or more
* local (intra-procedural) steps.
*/
predicate localExprTaint(Expr e1, Expr e2) {
localTaint(DataFlow::exprNode(e1), DataFlow::exprNode(e2))
}
/**
* Holds if we do not propagate taint from `fromExpr` to `toExpr`
* even though `toExpr` is the AST parent of `fromExpr`.
*/
private predicate noParentExprFlow(Expr fromExpr, Expr toExpr) {
fromExpr = toExpr.(ConditionalExpr).getCondition()
or
fromExpr = toExpr.(CommaExpr).getLeftOperand()
or
fromExpr = toExpr.(AssignExpr).getLValue() // LHS of `=`
}
/**
* Holds if we do not propagate taint from a child of `e` to `e` itself.
*/
private predicate noFlowFromChildExpr(Expr e) {
e instanceof ComparisonOperation
or
e instanceof LogicalAndExpr
or
e instanceof LogicalOrExpr
or
// Allow taint from `operator*` on smart pointers.
exists(Call call | e = call |
not call.getTarget() = any(PointerWrapper wrapper).getAnUnwrapperFunction()
)
or
e instanceof SizeofOperator
or
e instanceof AlignofOperator
or
e instanceof ClassAggregateLiteral
or
e instanceof FieldAccess
}
private predicate exprToExprStep(Expr exprIn, Expr exprOut) {
exists(DataFlowFunction f, Call call, FunctionOutput outModel |
call.getTarget() = f and
exprOut = call and
outModel.isReturnValueDeref() and
exists(int argInIndex, FunctionInput inModel | f.hasDataFlow(inModel, outModel) |
// Taint flows from a pointer to a dereference, which DataFlow does not handle
// dest_ptr = strdup(tainted_ptr)
inModel.isParameterDeref(argInIndex) and
exprIn = call.getArgument(argInIndex)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
)
or
exists(TaintFunction f, Call call, FunctionInput inModel, FunctionOutput outModel |
call.getTarget() = f and
(
exprOut = call and
outModel.isReturnValueDeref()
or
exprOut = call and
outModel.isReturnValue()
) and
f.hasTaintFlow(inModel, outModel) and
(
exists(int argInIndex |
inModel.isParameterDeref(argInIndex) and
exprIn = call.getArgument(argInIndex)
or
inModel.isParameterDeref(argInIndex) and
call.passesByReference(argInIndex, exprIn)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
or
inModel.isQualifierObject() and
exprIn = call.getQualifier()
)
)
}
private predicate exprToDefinitionByReferenceStep(Expr exprIn, Expr argOut) {
exists(DataFlowFunction f, Call call, FunctionOutput outModel, int argOutIndex |
call.getTarget() = f and
argOut = call.getArgument(argOutIndex) and
outModel.isParameterDeref(argOutIndex) and
exists(int argInIndex, FunctionInput inModel | f.hasDataFlow(inModel, outModel) |
// Taint flows from a pointer to a dereference, which DataFlow does not handle
// memcpy(&dest_var, tainted_ptr, len)
inModel.isParameterDeref(argInIndex) and
exprIn = call.getArgument(argInIndex)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
)
or
exists(
TaintFunction f, Call call, FunctionInput inModel, FunctionOutput outModel, int argOutIndex
|
call.getTarget() = f and
argOut = call.getArgument(argOutIndex) and
outModel.isParameterDeref(argOutIndex) and
f.hasTaintFlow(inModel, outModel) and
(
exists(int argInIndex |
inModel.isParameterDeref(argInIndex) and
exprIn = call.getArgument(argInIndex)
or
inModel.isParameterDeref(argInIndex) and
call.passesByReference(argInIndex, exprIn)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
or
inModel.isQualifierObject() and
exprIn = call.getQualifier()
)
)
}
private predicate exprToPartialDefinitionStep(Expr exprIn, Expr exprOut) {
exists(TaintFunction f, Call call, FunctionInput inModel, FunctionOutput outModel |
call.getTarget() = f and
(
exprOut = call.getQualifier() and
outModel.isQualifierObject()
) and
f.hasTaintFlow(inModel, outModel) and
exists(int argInIndex |
inModel.isParameterDeref(argInIndex) and
exprIn = call.getArgument(argInIndex)
or
inModel.isParameterDeref(argInIndex) and
call.passesByReference(argInIndex, exprIn)
or
inModel.isParameter(argInIndex) and
exprIn = call.getArgument(argInIndex)
)
)
or
exists(Assignment a |
iteratorDereference(exprOut) and
a.getLValue() = exprOut and
a.getRValue() = exprIn
)
}
private predicate iteratorDereference(Call c) { c.getTarget() instanceof IteratorReferenceFunction }

View File

@@ -1,120 +0,0 @@
/**
* Provides an implementation of global (interprocedural) taint tracking.
* This file re-exports the local (intraprocedural) taint-tracking analysis
* from `TaintTrackingParameter::Public` and adds a global analysis, mainly
* exposed through the `Configuration` class. For some languages, this file
* exists in several identical copies, allowing queries to use multiple
* `Configuration` classes that depend on each other without introducing
* mutual recursion among those configurations.
*/
import TaintTrackingParameter::Public
private import TaintTrackingParameter::Private
/**
* A configuration of interprocedural taint tracking analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the taint tracking library must define its own unique extension of
* this abstract class.
*
* A taint-tracking configuration is a special data flow configuration
* (`DataFlow::Configuration`) that allows for flow through nodes that do not
* necessarily preserve values but are still relevant from a taint tracking
* perspective. (For example, string concatenation, where one of the operands
* is tainted.)
*
* To create a configuration, extend this class with a subclass whose
* characteristic predicate is a unique singleton string. For example, write
*
* ```ql
* class MyAnalysisConfiguration extends TaintTracking::Configuration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isSanitizer`.
* // Optionally override `isSanitizerIn`.
* // Optionally override `isSanitizerOut`.
* // Optionally override `isSanitizerGuard`.
* // Optionally override `isAdditionalTaintStep`.
* }
* ```
*
* Then, to query whether there is flow between some `source` and `sink`,
* write
*
* ```ql
* exists(MyAnalysisConfiguration cfg | cfg.hasFlow(source, sink))
* ```
*
* Multiple configurations can coexist, but it is unsupported to depend on
* another `TaintTracking::Configuration` or a `DataFlow::Configuration` in the
* overridden predicates that define sources, sinks, or additional steps.
* Instead, the dependency should go to a `TaintTracking2::Configuration` or a
* `DataFlow2::Configuration`, `DataFlow3::Configuration`, etc.
*/
abstract class Configuration extends DataFlow::Configuration {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant taint source.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSource(DataFlow::Node source);
/**
* Holds if `sink` is a relevant taint sink.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSink(DataFlow::Node sink);
/** Holds if the node `node` is a taint sanitizer. */
predicate isSanitizer(DataFlow::Node node) { none() }
final override predicate isBarrier(DataFlow::Node node) {
isSanitizer(node) or
defaultTaintSanitizer(node)
}
/** Holds if taint propagation into `node` is prohibited. */
predicate isSanitizerIn(DataFlow::Node node) { none() }
final override predicate isBarrierIn(DataFlow::Node node) { isSanitizerIn(node) }
/** Holds if taint propagation out of `node` is prohibited. */
predicate isSanitizerOut(DataFlow::Node node) { none() }
final override predicate isBarrierOut(DataFlow::Node node) { isSanitizerOut(node) }
/** Holds if taint propagation through nodes guarded by `guard` is prohibited. */
predicate isSanitizerGuard(DataFlow::BarrierGuard guard) { none() }
final override predicate isBarrierGuard(DataFlow::BarrierGuard guard) { isSanitizerGuard(guard) }
/**
* Holds if the additional taint propagation step from `node1` to `node2`
* must be taken into account in the analysis.
*/
predicate isAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) { none() }
final override predicate isAdditionalFlowStep(DataFlow::Node node1, DataFlow::Node node2) {
isAdditionalTaintStep(node1, node2) or
defaultAdditionalTaintStep(node1, node2)
}
override predicate allowImplicitRead(DataFlow::Node node, DataFlow::Content c) {
(this.isSink(node) or this.isAdditionalTaintStep(node, _)) and
defaultImplicitTaintRead(node, c)
}
/**
* Holds if taint may flow from `source` to `sink` for this configuration.
*/
// overridden to provide taint-tracking specific qldoc
override predicate hasFlow(DataFlow::Node source, DataFlow::Node sink) {
super.hasFlow(source, sink)
}
}

View File

@@ -1,120 +0,0 @@
/**
* Provides an implementation of global (interprocedural) taint tracking.
* This file re-exports the local (intraprocedural) taint-tracking analysis
* from `TaintTrackingParameter::Public` and adds a global analysis, mainly
* exposed through the `Configuration` class. For some languages, this file
* exists in several identical copies, allowing queries to use multiple
* `Configuration` classes that depend on each other without introducing
* mutual recursion among those configurations.
*/
import TaintTrackingParameter::Public
private import TaintTrackingParameter::Private
/**
* A configuration of interprocedural taint tracking analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the taint tracking library must define its own unique extension of
* this abstract class.
*
* A taint-tracking configuration is a special data flow configuration
* (`DataFlow::Configuration`) that allows for flow through nodes that do not
* necessarily preserve values but are still relevant from a taint tracking
* perspective. (For example, string concatenation, where one of the operands
* is tainted.)
*
* To create a configuration, extend this class with a subclass whose
* characteristic predicate is a unique singleton string. For example, write
*
* ```ql
* class MyAnalysisConfiguration extends TaintTracking::Configuration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isSanitizer`.
* // Optionally override `isSanitizerIn`.
* // Optionally override `isSanitizerOut`.
* // Optionally override `isSanitizerGuard`.
* // Optionally override `isAdditionalTaintStep`.
* }
* ```
*
* Then, to query whether there is flow between some `source` and `sink`,
* write
*
* ```ql
* exists(MyAnalysisConfiguration cfg | cfg.hasFlow(source, sink))
* ```
*
* Multiple configurations can coexist, but it is unsupported to depend on
* another `TaintTracking::Configuration` or a `DataFlow::Configuration` in the
* overridden predicates that define sources, sinks, or additional steps.
* Instead, the dependency should go to a `TaintTracking2::Configuration` or a
* `DataFlow2::Configuration`, `DataFlow3::Configuration`, etc.
*/
abstract class Configuration extends DataFlow::Configuration {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant taint source.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSource(DataFlow::Node source);
/**
* Holds if `sink` is a relevant taint sink.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSink(DataFlow::Node sink);
/** Holds if the node `node` is a taint sanitizer. */
predicate isSanitizer(DataFlow::Node node) { none() }
final override predicate isBarrier(DataFlow::Node node) {
isSanitizer(node) or
defaultTaintSanitizer(node)
}
/** Holds if taint propagation into `node` is prohibited. */
predicate isSanitizerIn(DataFlow::Node node) { none() }
final override predicate isBarrierIn(DataFlow::Node node) { isSanitizerIn(node) }
/** Holds if taint propagation out of `node` is prohibited. */
predicate isSanitizerOut(DataFlow::Node node) { none() }
final override predicate isBarrierOut(DataFlow::Node node) { isSanitizerOut(node) }
/** Holds if taint propagation through nodes guarded by `guard` is prohibited. */
predicate isSanitizerGuard(DataFlow::BarrierGuard guard) { none() }
final override predicate isBarrierGuard(DataFlow::BarrierGuard guard) { isSanitizerGuard(guard) }
/**
* Holds if the additional taint propagation step from `node1` to `node2`
* must be taken into account in the analysis.
*/
predicate isAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) { none() }
final override predicate isAdditionalFlowStep(DataFlow::Node node1, DataFlow::Node node2) {
isAdditionalTaintStep(node1, node2) or
defaultAdditionalTaintStep(node1, node2)
}
override predicate allowImplicitRead(DataFlow::Node node, DataFlow::Content c) {
(this.isSink(node) or this.isAdditionalTaintStep(node, _)) and
defaultImplicitTaintRead(node, c)
}
/**
* Holds if taint may flow from `source` to `sink` for this configuration.
*/
// overridden to provide taint-tracking specific qldoc
override predicate hasFlow(DataFlow::Node source, DataFlow::Node sink) {
super.hasFlow(source, sink)
}
}

View File

@@ -1,265 +0,0 @@
private import cpp
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.DataFlow
private import semmle.code.cpp.ir.dataflow.internal.DataFlowPrivate
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, int i |
other.(DataFlow::ParameterNode).isParameterOf(call.getStaticCallTarget(), i) and
src.(ArgumentNode).argumentOf(call, 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.
*/
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()
)
}

View File

@@ -1,503 +0,0 @@
private import cpp
private import DataFlowUtil
private import semmle.code.cpp.ir.IR
private import DataFlowDispatch
/**
* 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 OperandNode {
/**
* 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, int 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 {
override ArgumentOperand op;
PrimaryArgumentNode() { exists(CallInstruction call | op = call.getAnArgumentOperand()) }
override predicate argumentOf(DataFlowCall call, int pos) { op = call.getArgumentOperand(pos) }
override string toString() {
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"
)
}
}
/**
* A data flow node representing the read side effect of a call on a
* specific parameter.
*/
private class SideEffectArgumentNode extends ArgumentNode {
override SideEffectOperand op;
ReadSideEffectInstruction read;
SideEffectArgumentNode() { op = read.getSideEffectOperand() }
override predicate argumentOf(DataFlowCall call, int pos) {
read.getPrimaryInstruction() = call and
pos = getArgumentPosOfSideEffect(read.getIndex())
}
override string toString() {
result = read.getArgumentDef().getUnconvertedResultExpression().toString() + " indirection"
or
// Some 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(read.getArgumentDef().getUnconvertedResultExpression()) and
(
if read.getIndex() = -1
then result = "Argument this indirection"
else result = "Argument " + read.getIndex() + " indirection"
)
}
}
private newtype TReturnKind =
TNormalReturnKind() or
TIndirectReturnKind(ParameterIndex index)
/**
* 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 {
override string toString() { result = "return" }
}
private class IndirectReturnKind extends ReturnKind, TIndirectReturnKind {
ParameterIndex index;
IndirectReturnKind() { this = TIndirectReturnKind(index) }
override string toString() { result = "outparam[" + index.toString() + "]" }
}
/** A data flow node that occurs as the result of a `ReturnStmt`. */
class ReturnNode extends InstructionNode {
Instruction primary;
ReturnNode() {
exists(ReturnValueInstruction ret | instr = ret.getReturnValue() and primary = ret)
or
exists(ReturnIndirectionInstruction rii |
instr = rii.getSideEffectOperand().getAnyDef() and primary = rii
)
}
/** Gets the kind of this returned value. */
abstract ReturnKind getKind();
}
class ReturnValueNode extends ReturnNode {
override ReturnValueInstruction primary;
override ReturnKind getKind() { result = TNormalReturnKind() }
}
class ReturnIndirectionNode extends ReturnNode {
override ReturnIndirectionInstruction primary;
override ReturnKind getKind() {
exists(int index |
primary.hasIndex(index) and
result = TIndirectReturnKind(index)
)
}
}
/** A data flow node that represents the output of a call. */
class OutNode extends InstructionNode {
OutNode() {
instr instanceof CallInstruction or
instr instanceof WriteSideEffectInstruction
}
/** Gets the underlying call. */
abstract DataFlowCall getCall();
abstract ReturnKind getReturnKind();
}
private class CallOutNode extends OutNode {
override CallInstruction instr;
override DataFlowCall getCall() { result = instr }
override ReturnKind getReturnKind() { result instanceof NormalReturnKind }
}
private class SideEffectOutNode extends OutNode {
override WriteSideEffectInstruction instr;
override DataFlowCall getCall() { result = instr.getPrimaryInstruction() }
override ReturnKind getReturnKind() { result = TIndirectReturnKind(instr.getIndex()) }
}
/**
* Gets a node that can read the value returned from `call` with return kind
* `kind`.
*/
OutNode getAnOutNode(DataFlowCall call, ReturnKind kind) {
// There should be only one `OutNode` for a given `(call, kind)` pair. Showing the optimizer that
// this is true helps it make better decisions downstream, especially in virtual dispatch.
result =
unique(OutNode outNode |
outNode.getCall() = call and
outNode.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) { none() }
private predicate fieldStoreStepNoChi(Node node1, FieldContent f, PostUpdateNode node2) {
exists(StoreInstruction store, Class c |
store = node2.asInstruction() and
store.getSourceValueOperand() = node1.asOperand() and
getWrittenField(store, f.(FieldContent).getAField(), c) and
f.hasOffset(c, _, _)
)
}
private FieldAddressInstruction getFieldInstruction(Instruction instr) {
result = instr or
result = instr.(CopyValueInstruction).getUnary()
}
pragma[noinline]
private predicate getWrittenField(Instruction instr, Field f, Class c) {
exists(FieldAddressInstruction fa |
fa =
getFieldInstruction([
instr.(StoreInstruction).getDestinationAddress(),
instr.(WriteSideEffectInstruction).getDestinationAddress()
]) and
f = fa.getField() and
c = f.getDeclaringType()
)
}
private predicate fieldStoreStepChi(Node node1, FieldContent f, PostUpdateNode node2) {
exists(ChiPartialOperand operand, ChiInstruction chi |
chi.getPartialOperand() = operand and
node1.asOperand() = operand and
node2.asInstruction() = chi and
exists(Class c |
c = chi.getResultType() and
exists(int startBit, int endBit |
chi.getUpdatedInterval(startBit, endBit) and
f.hasOffset(c, startBit, endBit)
)
or
getWrittenField(operand.getDef(), f.getAField(), c) and
f.hasOffset(c, _, _)
)
)
}
private predicate arrayStoreStepChi(Node node1, ArrayContent a, PostUpdateNode node2) {
exists(a) and
exists(ChiPartialOperand operand, ChiInstruction chi, StoreInstruction store |
chi.getPartialOperand() = operand and
store = operand.getDef() and
node1.asOperand() = operand and
// This `ChiInstruction` will always have a non-conflated result because both `ArrayStoreNode`
// and `PointerStoreNode` require it in their characteristic predicates.
node2.asInstruction() = chi and
(
// `x[i] = taint()`
// This matches the characteristic predicate in `ArrayStoreNode`.
store.getDestinationAddress() instanceof PointerAddInstruction
or
// `*p = taint()`
// This matches the characteristic predicate in `PointerStoreNode`.
store.getDestinationAddress().(CopyValueInstruction).getUnary() instanceof LoadInstruction
)
)
}
/**
* 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 f, PostUpdateNode node2) {
fieldStoreStepNoChi(node1, f, node2) or
fieldStoreStepChi(node1, f, node2) or
arrayStoreStepChi(node1, f, node2) or
fieldStoreStepAfterArraySuppression(node1, f, node2)
}
// This predicate pushes the correct `FieldContent` onto the access path when the
// `suppressArrayRead` predicate has popped off an `ArrayContent`.
private predicate fieldStoreStepAfterArraySuppression(
Node node1, FieldContent f, PostUpdateNode node2
) {
exists(WriteSideEffectInstruction write, ChiInstruction chi, Class c |
not chi.isResultConflated() and
node1.asInstruction() = chi and
node2.asInstruction() = chi and
chi.getPartial() = write and
getWrittenField(write, f.getAField(), c) and
f.hasOffset(c, _, _)
)
}
bindingset[result, i]
private int unbindInt(int i) { i <= result and i >= result }
pragma[noinline]
private predicate getLoadedField(LoadInstruction load, Field f, Class c) {
exists(FieldAddressInstruction fa |
fa = load.getSourceAddress() and
f = fa.getField() and
c = f.getDeclaringType()
)
}
/**
* 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`.
*/
private predicate fieldReadStep(Node node1, FieldContent f, Node node2) {
exists(LoadOperand operand |
node2.asOperand() = operand and
node1.asInstruction() = operand.getAnyDef() and
exists(Class c |
c = operand.getAnyDef().getResultType() and
exists(int startBit, int endBit |
operand.getUsedInterval(unbindInt(startBit), unbindInt(endBit)) and
f.hasOffset(c, startBit, endBit)
)
or
getLoadedField(operand.getUse(), f.getAField(), c) and
f.hasOffset(c, _, _)
)
)
}
/**
* When a store step happens in a function that looks like an array write such as:
* ```cpp
* void f(int* pa) {
* pa = source();
* }
* ```
* it can be a write to an array, but it can also happen that `f` is called as `f(&a.x)`. If that is
* the case, the `ArrayContent` that was written by the call to `f` should be popped off the access
* path, and a `FieldContent` containing `x` should be pushed instead.
* So this case pops `ArrayContent` off the access path, and the `fieldStoreStepAfterArraySuppression`
* predicate in `storeStep` ensures that we push the right `FieldContent` onto the access path.
*/
predicate suppressArrayRead(Node node1, ArrayContent a, Node node2) {
exists(a) and
exists(WriteSideEffectInstruction write, ChiInstruction chi |
node1.asInstruction() = write and
node2.asInstruction() = chi and
chi.getPartial() = write and
getWrittenField(write, _, _)
)
}
private class ArrayToPointerConvertInstruction extends ConvertInstruction {
ArrayToPointerConvertInstruction() {
this.getUnary().getResultType() instanceof ArrayType and
this.getResultType() instanceof PointerType
}
}
private Instruction skipOneCopyValueInstructionRec(CopyValueInstruction copy) {
copy.getUnary() = result and not result instanceof CopyValueInstruction
or
result = skipOneCopyValueInstructionRec(copy.getUnary())
}
private Instruction skipCopyValueInstructions(Operand op) {
not result instanceof CopyValueInstruction and result = op.getDef()
or
result = skipOneCopyValueInstructionRec(op.getDef())
}
private predicate arrayReadStep(Node node1, ArrayContent a, Node node2) {
exists(a) and
// Explicit dereferences such as `*p` or `p[i]` where `p` is a pointer or array.
exists(LoadOperand operand, Instruction address |
operand.isDefinitionInexact() and
node1.asInstruction() = operand.getAnyDef() and
operand = node2.asOperand() and
address = skipCopyValueInstructions(operand.getAddressOperand()) and
(
address instanceof LoadInstruction or
address instanceof ArrayToPointerConvertInstruction or
address instanceof PointerOffsetInstruction
)
)
}
/**
* In cases such as:
* ```cpp
* void f(int* pa) {
* *pa = source();
* }
* ...
* int x;
* f(&x);
* use(x);
* ```
* the load on `x` in `use(x)` will exactly overlap with its definition (in this case the definition
* is a `WriteSideEffect`). This predicate pops the `ArrayContent` (pushed by the store in `f`)
* from the access path.
*/
private predicate exactReadStep(Node node1, ArrayContent a, Node node2) {
exists(a) and
exists(WriteSideEffectInstruction write, ChiInstruction chi |
not chi.isResultConflated() and
chi.getPartial() = write and
node1.asInstruction() = write and
node2.asInstruction() = chi and
// To distinquish this case from the `arrayReadStep` case we require that the entire variable was
// overwritten by the `WriteSideEffectInstruction` (i.e., there is a load that reads the
// entire variable).
exists(LoadInstruction load | load.getSourceValue() = chi)
)
}
/**
* 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 f, Node node2) {
fieldReadStep(node1, f, node2) or
arrayReadStep(node1, f, node2) or
exactReadStep(node1, f, node2) or
suppressArrayRead(node1, f, node2)
}
/**
* Holds if values stored inside content `c` are cleared at node `n`.
*/
predicate clearsContent(Node n, Content c) {
none() // stub implementation
}
/** 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 InstructionNode {
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 = 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 }
/** 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` does not require a `PostUpdateNode` as it either cannot be
* modified or its modification cannot be observed, for example if it is a
* freshly created object that is not saved in a variable.
*
* This predicate is only used for consistency checks.
*/
predicate isImmutableOrUnobservable(Node n) {
// The rules for whether an IR argument gets a post-update node are too
// complex to model here.
any()
}
/** 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() }

View File

@@ -1,879 +0,0 @@
/**
* Provides C++-specific definitions for use in the data flow library.
*/
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.controlflow.IRGuards
private import semmle.code.cpp.models.interfaces.DataFlow
cached
private module Cached {
cached
newtype TIRDataFlowNode =
TInstructionNode(Instruction i) or
TOperandNode(Operand op) or
TVariableNode(Variable var)
cached
predicate localFlowStepCached(Node nodeFrom, Node nodeTo) {
simpleLocalFlowStep(nodeFrom, nodeTo)
}
}
private import Cached
/**
* A node in a data flow graph.
*
* A node can be either an expression, a parameter, or an uninitialized local
* variable. Such nodes are created with `DataFlow::exprNode`,
* `DataFlow::parameterNode`, and `DataFlow::uninitializedNode` respectively.
*/
class Node extends TIRDataFlowNode {
/**
* INTERNAL: Do not use.
*/
Declaration getEnclosingCallable() { none() } // overridden in subclasses
/** Gets the function to which this node belongs, if any. */
Function getFunction() { none() } // overridden in subclasses
/** Gets the type of this node. */
IRType getType() { none() } // overridden in subclasses
/** Gets the instruction corresponding to this node, if any. */
Instruction asInstruction() { result = this.(InstructionNode).getInstruction() }
/** Gets the operands corresponding to this node, if any. */
Operand asOperand() { result = this.(OperandNode).getOperand() }
/**
* Gets the non-conversion expression corresponding to this node, if any.
* This predicate only has a result on nodes that represent the value of
* evaluating the expression. For data flowing _out of_ an expression, like
* when an argument is passed by reference, use `asDefiningArgument` instead
* of `asExpr`.
*
* If this node strictly (in the sense of `asConvertedExpr`) corresponds to
* a `Conversion`, then the result is the underlying non-`Conversion` base
* expression.
*/
Expr asExpr() { result = this.(ExprNode).getExpr() }
/**
* Gets the expression corresponding to this node, if any. The returned
* expression may be a `Conversion`.
*/
Expr asConvertedExpr() { result = this.(ExprNode).getConvertedExpr() }
/**
* Gets the argument that defines this `DefinitionByReferenceNode`, if any.
* This predicate should be used instead of `asExpr` when referring to the
* value of a reference argument _after_ the call has returned. For example,
* in `f(&x)`, this predicate will have `&x` as its result for the `Node`
* that represents the new value of `x`.
*/
Expr asDefiningArgument() { result = this.(DefinitionByReferenceNode).getArgument() }
/** Gets the positional parameter corresponding to this node, if any. */
Parameter asParameter() { result = this.(ExplicitParameterNode).getParameter() }
/**
* Gets the variable corresponding to this node, if any. This can be used for
* modeling flow in and out of global variables.
*/
Variable asVariable() { result = this.(VariableNode).getVariable() }
/**
* Gets the expression that is partially defined by this node, if any.
*
* Partial definitions are created for field stores (`x.y = taint();` is a partial
* definition of `x`), and for calls that may change the value of an object (so
* `x.set(taint())` is a partial definition of `x`, and `transfer(&x, taint())` is
* a partial definition of `&x`).
*/
Expr asPartialDefinition() { result = this.(PartialDefinitionNode).getDefinedExpr() }
/**
* DEPRECATED: See UninitializedNode.
*
* Gets the uninitialized local variable corresponding to this node, if
* any.
*/
deprecated LocalVariable asUninitialized() { none() }
/**
* Gets an upper bound on the type of this node.
*/
IRType getTypeBound() { result = getType() }
/** Gets the location of this element. */
Location getLocation() { none() } // overridden by subclasses
/**
* Holds if this element is at the specified location.
* The location spans column `startcolumn` of line `startline` to
* column `endcolumn` of line `endline` in file `filepath`.
* For more information, see
* [Locations](https://help.semmle.com/QL/learn-ql/ql/locations.html).
*/
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.getLocation().hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
/** Gets a textual representation of this element. */
string toString() { none() } // overridden by subclasses
}
/**
* An instruction, viewed as a node in a data flow graph.
*/
class InstructionNode extends Node, TInstructionNode {
Instruction instr;
InstructionNode() { this = TInstructionNode(instr) }
/** Gets the instruction corresponding to this node. */
Instruction getInstruction() { result = instr }
override Declaration getEnclosingCallable() { result = this.getFunction() }
override Function getFunction() { result = instr.getEnclosingFunction() }
override IRType getType() { result = instr.getResultIRType() }
override Location getLocation() { result = instr.getLocation() }
override string toString() {
// This predicate is overridden in subclasses. This default implementation
// does not use `Instruction.toString` because that's expensive to compute.
result = this.getInstruction().getOpcode().toString()
}
}
/**
* An operand, viewed as a node in a data flow graph.
*/
class OperandNode extends Node, TOperandNode {
Operand op;
OperandNode() { this = TOperandNode(op) }
/** Gets the operand corresponding to this node. */
Operand getOperand() { result = op }
override Declaration getEnclosingCallable() { result = this.getFunction() }
override Function getFunction() { result = op.getUse().getEnclosingFunction() }
override IRType getType() { result = op.getIRType() }
override Location getLocation() { result = op.getLocation() }
override string toString() { result = this.getOperand().toString() }
}
/**
* An expression, viewed as a node in a data flow graph.
*/
class ExprNode extends InstructionNode {
ExprNode() { exists(instr.getConvertedResultExpression()) }
/**
* Gets the non-conversion expression corresponding to this node, if any. If
* this node strictly (in the sense of `getConvertedExpr`) corresponds to a
* `Conversion`, then the result is that `Conversion`'s non-`Conversion` base
* expression.
*/
Expr getExpr() { result = instr.getUnconvertedResultExpression() }
/**
* Gets the expression corresponding to this node, if any. The returned
* expression may be a `Conversion`.
*/
Expr getConvertedExpr() { result = instr.getConvertedResultExpression() }
override string toString() { result = this.asConvertedExpr().toString() }
}
/**
* INTERNAL: do not use. Translates a parameter/argument index into a negative
* number that denotes the index of its side effect (pointer indirection).
*/
bindingset[index]
int getArgumentPosOfSideEffect(int index) {
// -1 -> -2
// 0 -> -3
// 1 -> -4
// ...
result = -3 - index
}
/**
* The value of a parameter at function entry, viewed as a node in a data
* flow graph. This includes both explicit parameters such as `x` in `f(x)`
* and implicit parameters such as `this` in `x.f()`.
*
* To match a specific kind of parameter, consider using one of the subclasses
* `ExplicitParameterNode`, `ThisParameterNode`, or
* `ParameterIndirectionNode`.
*/
class ParameterNode extends InstructionNode {
ParameterNode() {
// To avoid making this class abstract, we enumerate its values here
instr instanceof InitializeParameterInstruction
or
instr instanceof InitializeIndirectionInstruction
}
/**
* Holds if this node is the parameter of `f` at the specified position. The
* implicit `this` parameter is considered to have position `-1`, and
* pointer-indirection parameters are at further negative positions.
*/
predicate isParameterOf(Function f, int pos) { none() } // overridden by subclasses
}
/** An explicit positional parameter, not including `this` or `...`. */
private class ExplicitParameterNode extends ParameterNode {
override InitializeParameterInstruction instr;
ExplicitParameterNode() { exists(instr.getParameter()) }
override predicate isParameterOf(Function f, int pos) {
f.getParameter(pos) = instr.getParameter()
}
/** Gets the `Parameter` associated with this node. */
Parameter getParameter() { result = instr.getParameter() }
override string toString() { result = instr.getParameter().toString() }
}
/** An implicit `this` parameter. */
class ThisParameterNode extends ParameterNode {
override InitializeParameterInstruction instr;
ThisParameterNode() { instr.getIRVariable() instanceof IRThisVariable }
override predicate isParameterOf(Function f, int pos) {
pos = -1 and instr.getEnclosingFunction() = f
}
override string toString() { result = "this" }
}
/** A synthetic parameter to model the pointed-to object of a pointer parameter. */
class ParameterIndirectionNode extends ParameterNode {
override InitializeIndirectionInstruction instr;
override predicate isParameterOf(Function f, int pos) {
exists(int index |
instr.getEnclosingFunction() = f and
instr.hasIndex(index)
|
pos = getArgumentPosOfSideEffect(index)
)
}
override string toString() { result = "*" + instr.getIRVariable().toString() }
}
/**
* DEPRECATED: Data flow was never an accurate way to determine what
* expressions might be uninitialized. It errs on the side of saying that
* everything is uninitialized, and this is even worse in the IR because the IR
* doesn't use syntactic hints to rule out variables that are definitely
* initialized.
*
* The value of an uninitialized local variable, viewed as a node in a data
* flow graph.
*/
deprecated class UninitializedNode extends Node {
UninitializedNode() { none() }
LocalVariable getLocalVariable() { none() }
}
/**
* A node associated with an object after an operation that might have
* changed its state.
*
* This can be either the argument to a callable after the callable returns
* (which might have mutated the argument), or the qualifier of a field after
* an update to the field.
*
* Nodes corresponding to AST elements, for example `ExprNode`, usually refer
* to the value before the update with the exception of `ClassInstanceExpr`,
* which represents the value after the constructor has run.
*
* This class exists to match the interface used by Java. There are currently no non-abstract
* classes that extend it. When we implement field flow, we can revisit this.
*/
abstract class PostUpdateNode extends InstructionNode {
/**
* Gets the node before the state update.
*/
abstract Node getPreUpdateNode();
}
/**
* The base class for nodes that perform "partial definitions".
*
* In contrast to a normal "definition", which provides a new value for
* something, a partial definition is an expression that may affect a
* value, but does not necessarily replace it entirely. For example:
* ```
* x.y = 1; // a partial definition of the object `x`.
* x.y.z = 1; // a partial definition of the object `x.y`.
* x.setY(1); // a partial definition of the object `x`.
* setY(&x); // a partial definition of the object `x`.
* ```
*/
abstract private class PartialDefinitionNode extends PostUpdateNode {
abstract Expr getDefinedExpr();
}
private class ExplicitFieldStoreQualifierNode extends PartialDefinitionNode {
override ChiInstruction instr;
StoreInstruction store;
ExplicitFieldStoreQualifierNode() {
not instr.isResultConflated() and
instr.getPartial() = store and
(
instr.getUpdatedInterval(_, _) or
store.getDestinationAddress() instanceof FieldAddressInstruction
)
}
// By using an operand as the result of this predicate we avoid the dataflow inconsistency errors
// caused by having multiple nodes sharing the same pre update node. This inconsistency error can cause
// a tuple explosion in the big step dataflow relation since it can make many nodes be the entry node
// into a big step.
override Node getPreUpdateNode() { result.asOperand() = instr.getTotalOperand() }
override Expr getDefinedExpr() {
result =
store
.getDestinationAddress()
.(FieldAddressInstruction)
.getObjectAddress()
.getUnconvertedResultExpression()
}
}
/**
* Not every store instruction generates a chi instruction that we can attach a PostUpdateNode to.
* For instance, an update to a field of a struct containing only one field. Even if the store does
* have a chi instruction, a subsequent use of the result of the store may be linked directly to the
* result of the store as an inexact definition if the store totally overlaps the use. For these
* cases we attach the PostUpdateNode to the store instruction. There's no obvious pre update node
* for this case (as the entire memory is updated), so `getPreUpdateNode` is implemented as
* `none()`.
*/
private class ExplicitSingleFieldStoreQualifierNode extends PartialDefinitionNode {
override StoreInstruction instr;
ExplicitSingleFieldStoreQualifierNode() {
(
instr.getAUse().isDefinitionInexact()
or
not exists(ChiInstruction chi | chi.getPartial() = instr)
) and
// Without this condition any store would create a `PostUpdateNode`.
instr.getDestinationAddress() instanceof FieldAddressInstruction
}
override Node getPreUpdateNode() { none() }
override Expr getDefinedExpr() {
result =
instr
.getDestinationAddress()
.(FieldAddressInstruction)
.getObjectAddress()
.getUnconvertedResultExpression()
}
}
private FieldAddressInstruction getFieldInstruction(Instruction instr) {
result = instr or
result = instr.(CopyValueInstruction).getUnary()
}
/**
* The target of a `fieldStoreStepAfterArraySuppression` store step, which is used to convert
* an `ArrayContent` to a `FieldContent` when the `WriteSideEffect` instruction stores
* into a field. See the QLDoc for `suppressArrayRead` for an example of where such a conversion
* is inserted.
*/
private class WriteSideEffectFieldStoreQualifierNode extends PartialDefinitionNode {
override ChiInstruction instr;
WriteSideEffectInstruction write;
FieldAddressInstruction field;
WriteSideEffectFieldStoreQualifierNode() {
not instr.isResultConflated() and
instr.getPartial() = write and
field = getFieldInstruction(write.getDestinationAddress())
}
override Node getPreUpdateNode() { result.asOperand() = instr.getTotalOperand() }
override Expr getDefinedExpr() {
result = field.getObjectAddress().getUnconvertedResultExpression()
}
}
/**
* The `PostUpdateNode` that is the target of a `arrayStoreStepChi` store step. The overriden
* `ChiInstruction` corresponds to the instruction represented by `node2` in `arrayStoreStepChi`.
*/
private class ArrayStoreNode extends PartialDefinitionNode {
override ChiInstruction instr;
PointerAddInstruction add;
ArrayStoreNode() {
not instr.isResultConflated() and
exists(StoreInstruction store |
instr.getPartial() = store and
add = store.getDestinationAddress()
)
}
override Node getPreUpdateNode() { result.asOperand() = instr.getTotalOperand() }
override Expr getDefinedExpr() { result = add.getLeft().getUnconvertedResultExpression() }
}
/**
* The `PostUpdateNode` that is the target of a `arrayStoreStepChi` store step. The overriden
* `ChiInstruction` corresponds to the instruction represented by `node2` in `arrayStoreStepChi`.
*/
private class PointerStoreNode extends PostUpdateNode {
override ChiInstruction instr;
PointerStoreNode() {
not instr.isResultConflated() and
exists(StoreInstruction store |
instr.getPartial() = store and
store.getDestinationAddress().(CopyValueInstruction).getUnary() instanceof LoadInstruction
)
}
override Node getPreUpdateNode() { result.asOperand() = instr.getTotalOperand() }
}
/**
* A node that represents the value of a variable after a function call that
* may have changed the variable because it's passed by reference.
*
* A typical example would be a call `f(&x)`. Firstly, there will be flow into
* `x` from previous definitions of `x`. Secondly, there will be a
* `DefinitionByReferenceNode` to represent the value of `x` after the call has
* returned. This node will have its `getArgument()` equal to `&x` and its
* `getVariableAccess()` equal to `x`.
*/
class DefinitionByReferenceNode extends InstructionNode {
override WriteSideEffectInstruction instr;
/** Gets the unconverted argument corresponding to this node. */
Expr getArgument() {
result =
instr
.getPrimaryInstruction()
.(CallInstruction)
.getArgument(instr.getIndex())
.getUnconvertedResultExpression()
}
/** Gets the parameter through which this value is assigned. */
Parameter getParameter() {
exists(CallInstruction ci | result = ci.getStaticCallTarget().getParameter(instr.getIndex()))
}
override string toString() {
// This string should be unique enough to be helpful but common enough to
// avoid storing too many different strings.
result =
instr.getPrimaryInstruction().(CallInstruction).getStaticCallTarget().getName() +
" output argument"
or
not exists(instr.getPrimaryInstruction().(CallInstruction).getStaticCallTarget()) and
result = "output argument"
}
}
/**
* A `Node` corresponding to a variable in the program, as opposed to the
* value of that variable at some particular point. This can be used for
* modeling flow in and out of global variables.
*/
class VariableNode extends Node, TVariableNode {
Variable v;
VariableNode() { this = TVariableNode(v) }
/** Gets the variable corresponding to this node. */
Variable getVariable() { result = v }
override Function getFunction() { none() }
override Declaration getEnclosingCallable() {
// 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.
// Therefore we want a distinct enclosing callable for each `VariableNode`,
// and that can be the `Variable` itself.
result = v
}
override IRType getType() { result.getCanonicalLanguageType().hasUnspecifiedType(v.getType(), _) }
override Location getLocation() { result = v.getLocation() }
override string toString() { result = v.toString() }
}
/**
* Gets the node corresponding to `instr`.
*/
InstructionNode instructionNode(Instruction instr) { result.getInstruction() = instr }
/**
* DEPRECATED: use `definitionByReferenceNodeFromArgument` instead.
*
* Gets the `Node` corresponding to a definition by reference of the variable
* that is passed as `argument` of a call.
*/
deprecated DefinitionByReferenceNode definitionByReferenceNode(Expr e) { result.getArgument() = e }
/**
* Gets the `Node` corresponding to the value of evaluating `e` or any of its
* conversions. There is no result if `e` is a `Conversion`. For data flowing
* _out of_ an expression, like when an argument is passed by reference, use
* `definitionByReferenceNodeFromArgument` instead.
*/
ExprNode exprNode(Expr e) { result.getExpr() = e }
/**
* Gets the `Node` corresponding to the value of evaluating `e`. Here, `e` may
* be a `Conversion`. For data flowing _out of_ an expression, like when an
* argument is passed by reference, use
* `definitionByReferenceNodeFromArgument` instead.
*/
ExprNode convertedExprNode(Expr e) { result.getConvertedExpr() = e }
/**
* Gets the `Node` corresponding to the value of `p` at function entry.
*/
ExplicitParameterNode parameterNode(Parameter p) { result.getParameter() = p }
/**
* Gets the `Node` corresponding to a definition by reference of the variable
* that is passed as unconverted `argument` of a call.
*/
DefinitionByReferenceNode definitionByReferenceNodeFromArgument(Expr argument) {
result.getArgument() = argument
}
/** Gets the `VariableNode` corresponding to the variable `v`. */
VariableNode variableNode(Variable v) { result.getVariable() = v }
/**
* DEPRECATED: See UninitializedNode.
*
* Gets the `Node` corresponding to the value of an uninitialized local
* variable `v`.
*/
Node uninitializedNode(LocalVariable v) { none() }
/**
* Holds if data flows from `nodeFrom` to `nodeTo` in exactly one local
* (intra-procedural) step.
*/
predicate localFlowStep = localFlowStepCached/2;
/**
* INTERNAL: do not use.
*
* This is the local flow predicate that's used as a building block in global
* data flow. It may have less flow than the `localFlowStep` predicate.
*/
predicate simpleLocalFlowStep(Node nodeFrom, Node nodeTo) {
// Operand -> Instruction flow
simpleInstructionLocalFlowStep(nodeFrom.asOperand(), nodeTo.asInstruction())
or
// Instruction -> Operand flow
simpleOperandLocalFlowStep(nodeFrom.asInstruction(), nodeTo.asOperand())
}
pragma[noinline]
private predicate getFieldSizeOfClass(Class c, Type type, int size) {
exists(Field f |
f.getDeclaringType() = c and
f.getUnderlyingType() = type and
type.getSize() = size
)
}
private predicate isSingleFieldClass(Type type, Operand op) {
exists(int size, Class c |
c = op.getType().getUnderlyingType() and
c.getSize() = size and
getFieldSizeOfClass(c, type, size)
)
}
private predicate simpleOperandLocalFlowStep(Instruction iFrom, Operand opTo) {
// Propagate flow from an instruction to its exact uses.
opTo.getDef() = iFrom
or
opTo = any(ReadSideEffectInstruction read).getSideEffectOperand() and
not iFrom.isResultConflated() and
iFrom = opTo.getAnyDef()
or
// Loading a single `int` from an `int *` parameter is not an exact load since
// the parameter may point to an entire array rather than a single `int`. The
// following rule ensures that any flow going into the
// `InitializeIndirectionInstruction`, even if it's for a different array
// element, will propagate to a load of the first element.
//
// Since we're linking `InitializeIndirectionInstruction` and
// `LoadInstruction` together directly, this rule will break if there's any
// reassignment of the parameter indirection, including a conditional one that
// leads to a phi node.
exists(InitializeIndirectionInstruction init |
iFrom = init and
opTo.(LoadOperand).getAnyDef() = init and
// Check that the types match. Otherwise we can get flow from an object to
// its fields, which leads to field conflation when there's flow from other
// fields to the object elsewhere.
init.getParameter().getType().getUnspecifiedType().(DerivedType).getBaseType() =
opTo.getType().getUnspecifiedType()
)
or
// Flow from stores to structs with a single field to a load of that field.
exists(LoadInstruction load |
load.getSourceValueOperand() = opTo and
opTo.getAnyDef() = iFrom and
isSingleFieldClass(pragma[only_bind_out](pragma[only_bind_out](iFrom).getResultType()), opTo)
)
}
private predicate simpleInstructionLocalFlowStep(Operand opFrom, Instruction iTo) {
iTo.(CopyInstruction).getSourceValueOperand() = opFrom
or
iTo.(PhiInstruction).getAnInputOperand() = opFrom
or
// Treat all conversions as flow, even conversions between different numeric types.
iTo.(ConvertInstruction).getUnaryOperand() = opFrom
or
iTo.(CheckedConvertOrNullInstruction).getUnaryOperand() = opFrom
or
iTo.(InheritanceConversionInstruction).getUnaryOperand() = opFrom
or
// A chi instruction represents a point where a new value (the _partial_
// operand) may overwrite an old value (the _total_ operand), but the alias
// analysis couldn't determine that it surely will overwrite every bit of it or
// that it surely will overwrite no bit of it.
//
// By allowing flow through the total operand, we ensure that flow is not lost
// due to shortcomings of the alias analysis. We may get false flow in cases
// where the data is indeed overwritten.
//
// Flow through the partial operand belongs in the taint-tracking libraries
// for now.
iTo.getAnOperand().(ChiTotalOperand) = opFrom
or
// Add flow from write side-effects to non-conflated chi instructions through their
// partial operands. From there, a `readStep` will find subsequent reads of that field.
// Consider the following example:
// ```
// void setX(Point* p, int new_x) {
// p->x = new_x;
// }
// ...
// setX(&p, taint());
// ```
// Here, a `WriteSideEffectInstruction` will provide a new definition for `p->x` after the call to
// `setX`, which will be melded into `p` through a chi instruction.
exists(ChiInstruction chi | chi = iTo |
opFrom.getAnyDef() instanceof WriteSideEffectInstruction and
chi.getPartialOperand() = opFrom and
not chi.isResultConflated() and
// In a call such as `set_value(&x->val);` we don't want the memory representing `x` to receive
// dataflow by a simple step. Instead, this is handled by field flow. If we add a simple step here
// we can get field-to-object flow.
not chi.isPartialUpdate()
)
or
// Flow through modeled functions
modelFlow(opFrom, iTo)
}
private predicate modelFlow(Operand opFrom, Instruction iTo) {
exists(
CallInstruction call, DataFlowFunction func, FunctionInput modelIn, FunctionOutput modelOut
|
call.getStaticCallTarget() = func and
func.hasDataFlow(modelIn, modelOut)
|
(
modelOut.isReturnValue() and
iTo = call
or
// TODO: Add write side effects for return values
modelOut.isReturnValueDeref() and
iTo = call
or
exists(int index, WriteSideEffectInstruction outNode |
modelOut.isParameterDerefOrQualifierObject(index) and
iTo = outNode and
outNode = getSideEffectFor(call, index)
)
) and
(
exists(int index |
modelIn.isParameterOrQualifierAddress(index) and
opFrom = call.getArgumentOperand(index)
)
or
exists(int index, ReadSideEffectInstruction read |
modelIn.isParameterDerefOrQualifierObject(index) and
read = getSideEffectFor(call, index) and
opFrom = read.getSideEffectOperand()
)
)
)
}
/**
* Holds if the result is a side effect for instruction `call` on argument
* index `argument`. This helper predicate makes it easy to join on both of
* these columns at once, avoiding pathological join orders in case the
* argument index should get joined first.
*/
pragma[noinline]
SideEffectInstruction getSideEffectFor(CallInstruction call, int argument) {
call = result.getPrimaryInstruction() and
argument = result.(IndexedInstruction).getIndex()
}
/**
* Holds if data flows from `source` to `sink` in zero or more local
* (intra-procedural) steps.
*/
predicate localFlow(Node source, Node sink) { localFlowStep*(source, sink) }
/**
* Holds if data can flow from `i1` to `i2` in zero or more
* local (intra-procedural) steps.
*/
predicate localInstructionFlow(Instruction e1, Instruction e2) {
localFlow(instructionNode(e1), instructionNode(e2))
}
/**
* Holds if data can flow from `e1` to `e2` in zero or more
* local (intra-procedural) steps.
*/
predicate localExprFlow(Expr e1, Expr e2) { localFlow(exprNode(e1), exprNode(e2)) }
/**
* Gets a field corresponding to the bit range `[startBit..endBit)` of class `c`, if any.
*/
private Field getAField(Class c, int startBit, int endBit) {
result.getDeclaringType() = c and
startBit = 8 * result.getByteOffset() and
endBit = 8 * result.getType().getSize() + startBit
or
exists(Field f, Class cInner |
f = c.getAField() and
cInner = f.getUnderlyingType() and
result = getAField(cInner, startBit - 8 * f.getByteOffset(), endBit - 8 * f.getByteOffset())
)
}
private newtype TContent =
TFieldContent(Class c, int startBit, int endBit) { exists(getAField(c, startBit, endBit)) } or
TCollectionContent() or
TArrayContent()
/**
* A description of the way data may be stored inside an object. Examples
* include instance fields, the contents of a collection object, or the contents
* of an array.
*/
class Content extends TContent {
/** Gets a textual representation of this element. */
abstract string toString();
predicate hasLocationInfo(string path, int sl, int sc, int el, int ec) {
path = "" and sl = 0 and sc = 0 and el = 0 and ec = 0
}
}
/** A reference through an instance field. */
class FieldContent extends Content, TFieldContent {
Class c;
int startBit;
int endBit;
FieldContent() { this = TFieldContent(c, startBit, endBit) }
// Ensure that there's just 1 result for `toString`.
override string toString() { result = min(Field f | f = getAField() | f.toString()) }
predicate hasOffset(Class cl, int start, int end) { cl = c and start = startBit and end = endBit }
Field getAField() { result = getAField(c, startBit, endBit) }
}
/** A reference through an array. */
class ArrayContent extends Content, TArrayContent {
override string toString() { result = "[]" }
}
/** A reference through the contents of some collection-like container. */
private class CollectionContent extends Content, TCollectionContent {
override string toString() { result = "<element>" }
}
/**
* A guard that validates some instruction.
*
* To use this in a configuration, extend the class and provide a
* characteristic predicate precisely specifying the guard, and override
* `checks` to specify what is being validated and in which branch.
*
* It is important that all extending classes in scope are disjoint.
*/
class BarrierGuard extends IRGuardCondition {
/** Override this predicate to hold if this guard validates `instr` upon evaluating to `b`. */
predicate checksInstr(Instruction instr, boolean b) { none() }
/** Override this predicate to hold if this guard validates `expr` upon evaluating to `b`. */
predicate checks(Expr e, boolean b) { none() }
/** Gets a node guarded by this guard. */
final Node getAGuardedNode() {
exists(ValueNumber value, boolean edge |
(
this.checksInstr(value.getAnInstruction(), edge)
or
this.checks(value.getAnInstruction().getConvertedResultExpression(), edge)
) and
result.asInstruction() = value.getAnInstruction() and
this.controls(result.asInstruction().getBlock(), edge)
)
}
}

View File

@@ -1,223 +0,0 @@
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.DataFlow
private import ModelUtil
private import semmle.code.cpp.models.interfaces.DataFlow
private import semmle.code.cpp.models.interfaces.SideEffect
/**
* Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
* (intra-procedural) step.
*/
predicate localTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
DataFlow::localFlowStep(nodeFrom, nodeTo)
or
localAdditionalTaintStep(nodeFrom, nodeTo)
}
/**
* Holds if taint can flow in one local step from `nodeFrom` to `nodeTo` excluding
* local data flow steps. That is, `nodeFrom` and `nodeTo` are likely to represent
* different objects.
*/
cached
predicate localAdditionalTaintStep(DataFlow::Node nodeFrom, DataFlow::Node nodeTo) {
operandToInstructionTaintStep(nodeFrom.asOperand(), nodeTo.asInstruction())
or
instructionToOperandTaintStep(nodeFrom.asInstruction(), nodeTo.asOperand())
}
private predicate instructionToOperandTaintStep(Instruction fromInstr, Operand toOperand) {
// Propagate flow from the definition of an operand to the operand, even when the overlap is inexact.
// We only do this in certain cases:
// 1. The instruction's result must not be conflated, and
// 2. The instruction's result type is one the types where we expect element-to-object flow. Currently
// this is array types and union types. This matches the other two cases of element-to-object flow in
// `DefaultTaintTracking`.
toOperand.getAnyDef() = fromInstr and
not fromInstr.isResultConflated() and
(
fromInstr.getResultType() instanceof ArrayType or
fromInstr.getResultType() instanceof Union
)
or
exists(ReadSideEffectInstruction readInstr |
fromInstr = readInstr.getArgumentDef() and
toOperand = readInstr.getSideEffectOperand()
)
or
toOperand.(LoadOperand).getAnyDef() = fromInstr
}
/**
* Holds if taint propagates from `nodeFrom` to `nodeTo` in exactly one local
* (intra-procedural) step.
*/
private predicate operandToInstructionTaintStep(Operand opFrom, Instruction instrTo) {
// Taint can flow through expressions that alter the value but preserve
// more than one bit of it _or_ expressions that follow data through
// pointer indirections.
instrTo.getAnOperand() = opFrom and
(
instrTo instanceof ArithmeticInstruction
or
instrTo instanceof BitwiseInstruction
or
instrTo instanceof PointerArithmeticInstruction
or
// The `CopyInstruction` case is also present in non-taint data flow, but
// that uses `getDef` rather than `getAnyDef`. For taint, we want flow
// from a definition of `myStruct` to a `myStruct.myField` expression.
instrTo instanceof CopyInstruction
)
or
// Unary instructions tend to preserve enough information in practice that we
// want taint to flow through.
// The exception is `FieldAddressInstruction`. Together with the rules below for
// `LoadInstruction`s and `ChiInstruction`s, flow through `FieldAddressInstruction`
// could cause flow into one field to come out an unrelated field.
// This would happen across function boundaries, where the IR would not be able to
// match loads to stores.
instrTo.(UnaryInstruction).getUnaryOperand() = opFrom and
(
not instrTo instanceof FieldAddressInstruction
or
instrTo.(FieldAddressInstruction).getField().getDeclaringType() instanceof Union
)
or
instrTo.(LoadInstruction).getSourceAddressOperand() = opFrom
or
// Flow from an element to an array or union that contains it.
instrTo.(ChiInstruction).getPartialOperand() = opFrom and
not instrTo.isResultConflated() and
exists(Type t | instrTo.getResultLanguageType().hasType(t, false) |
t instanceof Union
or
t instanceof ArrayType
)
or
// Until we have flow through indirections across calls, we'll take flow out
// of the indirection and into the argument.
// When we get proper flow through indirections across calls, this code can be
// moved to `adjusedSink` or possibly into the `DataFlow::ExprNode` class.
exists(ReadSideEffectInstruction read |
read.getSideEffectOperand() = opFrom and
read.getArgumentDef() = instrTo
)
or
// Until we have from through indirections across calls, we'll take flow out
// of the parameter and into its indirection.
// `InitializeIndirectionInstruction` only has a single operand: the address of the
// value whose indirection we are initializing. When initializing an indirection of a parameter `p`,
// the IR looks like this:
// ```
// m1 = InitializeParameter[p] : &r1
// r2 = Load[p] : r2, m1
// m3 = InitializeIndirection[p] : &r2
// ```
// So by having flow from `r2` to `m3` we're enabling flow from `m1` to `m3`. This relies on the
// `LoadOperand`'s overlap being exact.
instrTo.(InitializeIndirectionInstruction).getAnOperand() = opFrom
or
modeledTaintStep(opFrom, instrTo)
}
/**
* Holds if taint may propagate from `source` to `sink` in zero or more local
* (intra-procedural) steps.
*/
predicate localTaint(DataFlow::Node source, DataFlow::Node sink) { localTaintStep*(source, sink) }
/**
* Holds if taint can flow from `i1` to `i2` in zero or more
* local (intra-procedural) steps.
*/
predicate localInstructionTaint(Instruction i1, Instruction i2) {
localTaint(DataFlow::instructionNode(i1), DataFlow::instructionNode(i2))
}
/**
* Holds if taint can flow from `e1` to `e2` in zero or more
* local (intra-procedural) steps.
*/
predicate localExprTaint(Expr e1, Expr e2) {
localTaint(DataFlow::exprNode(e1), DataFlow::exprNode(e2))
}
/**
* Holds if the additional step from `src` to `sink` should be included in all
* global taint flow configurations.
*/
predicate defaultAdditionalTaintStep(DataFlow::Node src, DataFlow::Node sink) {
localAdditionalTaintStep(src, sink)
}
/**
* Holds if default `TaintTracking::Configuration`s should allow implicit reads
* of `c` at sinks and inputs to additional taint steps.
*/
bindingset[node]
predicate defaultImplicitTaintRead(DataFlow::Node node, DataFlow::Content c) { none() }
/**
* Holds if `node` should be a sanitizer in all global taint flow configurations
* but not in local taint.
*/
predicate defaultTaintSanitizer(DataFlow::Node node) { none() }
/**
* Holds if taint can flow from `instrIn` to `instrOut` through a call to a
* modeled function.
*/
predicate modeledTaintStep(Operand nodeIn, Instruction nodeOut) {
exists(CallInstruction call, TaintFunction func, FunctionInput modelIn, FunctionOutput modelOut |
(
nodeIn = callInput(call, modelIn)
or
exists(int n |
modelIn.isParameterDerefOrQualifierObject(n) and
if n = -1
then nodeIn = callInput(call, any(InQualifierObject inQualifier))
else nodeIn = callInput(call, any(InParameter inParam | inParam.getIndex() = n))
)
) and
nodeOut = callOutput(call, modelOut) and
call.getStaticCallTarget() = func and
func.hasTaintFlow(modelIn, modelOut)
)
or
// Taint flow from one argument to another and data flow from an argument to a
// return value. This happens in functions like `strcat` and `memcpy`. We
// could model this flow in two separate steps, but that would add reverse
// flow from the write side-effect to the call instruction, which may not be
// desirable.
exists(
CallInstruction call, Function func, FunctionInput modelIn, OutParameterDeref modelMidOut,
int indexMid, InParameter modelMidIn, OutReturnValue modelOut
|
nodeIn = callInput(call, modelIn) and
nodeOut = callOutput(call, modelOut) and
call.getStaticCallTarget() = func and
func.(TaintFunction).hasTaintFlow(modelIn, modelMidOut) and
func.(DataFlowFunction).hasDataFlow(modelMidIn, modelOut) and
modelMidOut.isParameterDeref(indexMid) and
modelMidIn.isParameter(indexMid)
)
or
// Taint flow from a pointer argument to an output, when the model specifies flow from the deref
// to that output, but the deref is not modeled in the IR for the caller.
exists(
CallInstruction call, ReadSideEffectInstruction read, Function func, FunctionInput modelIn,
FunctionOutput modelOut
|
read.getSideEffectOperand() = callInput(call, modelIn) and
read.getArgumentDef() = nodeIn.getDef() and
not read.getSideEffect().isResultModeled() and
call.getStaticCallTarget() = func and
(
func.(DataFlowFunction).hasDataFlow(modelIn, modelOut)
or
func.(TaintFunction).hasTaintFlow(modelIn, modelOut)
) and
nodeOut = callOutput(call, modelOut)
)
}

View File

@@ -1,120 +0,0 @@
/**
* Provides an implementation of global (interprocedural) taint tracking.
* This file re-exports the local (intraprocedural) taint-tracking analysis
* from `TaintTrackingParameter::Public` and adds a global analysis, mainly
* exposed through the `Configuration` class. For some languages, this file
* exists in several identical copies, allowing queries to use multiple
* `Configuration` classes that depend on each other without introducing
* mutual recursion among those configurations.
*/
import TaintTrackingParameter::Public
private import TaintTrackingParameter::Private
/**
* A configuration of interprocedural taint tracking analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the taint tracking library must define its own unique extension of
* this abstract class.
*
* A taint-tracking configuration is a special data flow configuration
* (`DataFlow::Configuration`) that allows for flow through nodes that do not
* necessarily preserve values but are still relevant from a taint tracking
* perspective. (For example, string concatenation, where one of the operands
* is tainted.)
*
* To create a configuration, extend this class with a subclass whose
* characteristic predicate is a unique singleton string. For example, write
*
* ```ql
* class MyAnalysisConfiguration extends TaintTracking::Configuration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isSanitizer`.
* // Optionally override `isSanitizerIn`.
* // Optionally override `isSanitizerOut`.
* // Optionally override `isSanitizerGuard`.
* // Optionally override `isAdditionalTaintStep`.
* }
* ```
*
* Then, to query whether there is flow between some `source` and `sink`,
* write
*
* ```ql
* exists(MyAnalysisConfiguration cfg | cfg.hasFlow(source, sink))
* ```
*
* Multiple configurations can coexist, but it is unsupported to depend on
* another `TaintTracking::Configuration` or a `DataFlow::Configuration` in the
* overridden predicates that define sources, sinks, or additional steps.
* Instead, the dependency should go to a `TaintTracking2::Configuration` or a
* `DataFlow2::Configuration`, `DataFlow3::Configuration`, etc.
*/
abstract class Configuration extends DataFlow::Configuration {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant taint source.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSource(DataFlow::Node source);
/**
* Holds if `sink` is a relevant taint sink.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSink(DataFlow::Node sink);
/** Holds if the node `node` is a taint sanitizer. */
predicate isSanitizer(DataFlow::Node node) { none() }
final override predicate isBarrier(DataFlow::Node node) {
isSanitizer(node) or
defaultTaintSanitizer(node)
}
/** Holds if taint propagation into `node` is prohibited. */
predicate isSanitizerIn(DataFlow::Node node) { none() }
final override predicate isBarrierIn(DataFlow::Node node) { isSanitizerIn(node) }
/** Holds if taint propagation out of `node` is prohibited. */
predicate isSanitizerOut(DataFlow::Node node) { none() }
final override predicate isBarrierOut(DataFlow::Node node) { isSanitizerOut(node) }
/** Holds if taint propagation through nodes guarded by `guard` is prohibited. */
predicate isSanitizerGuard(DataFlow::BarrierGuard guard) { none() }
final override predicate isBarrierGuard(DataFlow::BarrierGuard guard) { isSanitizerGuard(guard) }
/**
* Holds if the additional taint propagation step from `node1` to `node2`
* must be taken into account in the analysis.
*/
predicate isAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) { none() }
final override predicate isAdditionalFlowStep(DataFlow::Node node1, DataFlow::Node node2) {
isAdditionalTaintStep(node1, node2) or
defaultAdditionalTaintStep(node1, node2)
}
override predicate allowImplicitRead(DataFlow::Node node, DataFlow::Content c) {
(this.isSink(node) or this.isAdditionalTaintStep(node, _)) and
defaultImplicitTaintRead(node, c)
}
/**
* Holds if taint may flow from `source` to `sink` for this configuration.
*/
// overridden to provide taint-tracking specific qldoc
override predicate hasFlow(DataFlow::Node source, DataFlow::Node sink) {
super.hasFlow(source, sink)
}
}

View File

@@ -1,120 +0,0 @@
/**
* Provides an implementation of global (interprocedural) taint tracking.
* This file re-exports the local (intraprocedural) taint-tracking analysis
* from `TaintTrackingParameter::Public` and adds a global analysis, mainly
* exposed through the `Configuration` class. For some languages, this file
* exists in several identical copies, allowing queries to use multiple
* `Configuration` classes that depend on each other without introducing
* mutual recursion among those configurations.
*/
import TaintTrackingParameter::Public
private import TaintTrackingParameter::Private
/**
* A configuration of interprocedural taint tracking analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the taint tracking library must define its own unique extension of
* this abstract class.
*
* A taint-tracking configuration is a special data flow configuration
* (`DataFlow::Configuration`) that allows for flow through nodes that do not
* necessarily preserve values but are still relevant from a taint tracking
* perspective. (For example, string concatenation, where one of the operands
* is tainted.)
*
* To create a configuration, extend this class with a subclass whose
* characteristic predicate is a unique singleton string. For example, write
*
* ```ql
* class MyAnalysisConfiguration extends TaintTracking::Configuration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isSanitizer`.
* // Optionally override `isSanitizerIn`.
* // Optionally override `isSanitizerOut`.
* // Optionally override `isSanitizerGuard`.
* // Optionally override `isAdditionalTaintStep`.
* }
* ```
*
* Then, to query whether there is flow between some `source` and `sink`,
* write
*
* ```ql
* exists(MyAnalysisConfiguration cfg | cfg.hasFlow(source, sink))
* ```
*
* Multiple configurations can coexist, but it is unsupported to depend on
* another `TaintTracking::Configuration` or a `DataFlow::Configuration` in the
* overridden predicates that define sources, sinks, or additional steps.
* Instead, the dependency should go to a `TaintTracking2::Configuration` or a
* `DataFlow2::Configuration`, `DataFlow3::Configuration`, etc.
*/
abstract class Configuration extends DataFlow::Configuration {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant taint source.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSource(DataFlow::Node source);
/**
* Holds if `sink` is a relevant taint sink.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSink(DataFlow::Node sink);
/** Holds if the node `node` is a taint sanitizer. */
predicate isSanitizer(DataFlow::Node node) { none() }
final override predicate isBarrier(DataFlow::Node node) {
isSanitizer(node) or
defaultTaintSanitizer(node)
}
/** Holds if taint propagation into `node` is prohibited. */
predicate isSanitizerIn(DataFlow::Node node) { none() }
final override predicate isBarrierIn(DataFlow::Node node) { isSanitizerIn(node) }
/** Holds if taint propagation out of `node` is prohibited. */
predicate isSanitizerOut(DataFlow::Node node) { none() }
final override predicate isBarrierOut(DataFlow::Node node) { isSanitizerOut(node) }
/** Holds if taint propagation through nodes guarded by `guard` is prohibited. */
predicate isSanitizerGuard(DataFlow::BarrierGuard guard) { none() }
final override predicate isBarrierGuard(DataFlow::BarrierGuard guard) { isSanitizerGuard(guard) }
/**
* Holds if the additional taint propagation step from `node1` to `node2`
* must be taken into account in the analysis.
*/
predicate isAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) { none() }
final override predicate isAdditionalFlowStep(DataFlow::Node node1, DataFlow::Node node2) {
isAdditionalTaintStep(node1, node2) or
defaultAdditionalTaintStep(node1, node2)
}
override predicate allowImplicitRead(DataFlow::Node node, DataFlow::Content c) {
(this.isSink(node) or this.isAdditionalTaintStep(node, _)) and
defaultImplicitTaintRead(node, c)
}
/**
* Holds if taint may flow from `source` to `sink` for this configuration.
*/
// overridden to provide taint-tracking specific qldoc
override predicate hasFlow(DataFlow::Node source, DataFlow::Node sink) {
super.hasFlow(source, sink)
}
}

View File

@@ -1,120 +0,0 @@
/**
* Provides an implementation of global (interprocedural) taint tracking.
* This file re-exports the local (intraprocedural) taint-tracking analysis
* from `TaintTrackingParameter::Public` and adds a global analysis, mainly
* exposed through the `Configuration` class. For some languages, this file
* exists in several identical copies, allowing queries to use multiple
* `Configuration` classes that depend on each other without introducing
* mutual recursion among those configurations.
*/
import TaintTrackingParameter::Public
private import TaintTrackingParameter::Private
/**
* A configuration of interprocedural taint tracking analysis. This defines
* sources, sinks, and any other configurable aspect of the analysis. Each
* use of the taint tracking library must define its own unique extension of
* this abstract class.
*
* A taint-tracking configuration is a special data flow configuration
* (`DataFlow::Configuration`) that allows for flow through nodes that do not
* necessarily preserve values but are still relevant from a taint tracking
* perspective. (For example, string concatenation, where one of the operands
* is tainted.)
*
* To create a configuration, extend this class with a subclass whose
* characteristic predicate is a unique singleton string. For example, write
*
* ```ql
* class MyAnalysisConfiguration extends TaintTracking::Configuration {
* MyAnalysisConfiguration() { this = "MyAnalysisConfiguration" }
* // Override `isSource` and `isSink`.
* // Optionally override `isSanitizer`.
* // Optionally override `isSanitizerIn`.
* // Optionally override `isSanitizerOut`.
* // Optionally override `isSanitizerGuard`.
* // Optionally override `isAdditionalTaintStep`.
* }
* ```
*
* Then, to query whether there is flow between some `source` and `sink`,
* write
*
* ```ql
* exists(MyAnalysisConfiguration cfg | cfg.hasFlow(source, sink))
* ```
*
* Multiple configurations can coexist, but it is unsupported to depend on
* another `TaintTracking::Configuration` or a `DataFlow::Configuration` in the
* overridden predicates that define sources, sinks, or additional steps.
* Instead, the dependency should go to a `TaintTracking2::Configuration` or a
* `DataFlow2::Configuration`, `DataFlow3::Configuration`, etc.
*/
abstract class Configuration extends DataFlow::Configuration {
bindingset[this]
Configuration() { any() }
/**
* Holds if `source` is a relevant taint source.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSource(DataFlow::Node source);
/**
* Holds if `sink` is a relevant taint sink.
*
* The smaller this predicate is, the faster `hasFlow()` will converge.
*/
// overridden to provide taint-tracking specific qldoc
abstract override predicate isSink(DataFlow::Node sink);
/** Holds if the node `node` is a taint sanitizer. */
predicate isSanitizer(DataFlow::Node node) { none() }
final override predicate isBarrier(DataFlow::Node node) {
isSanitizer(node) or
defaultTaintSanitizer(node)
}
/** Holds if taint propagation into `node` is prohibited. */
predicate isSanitizerIn(DataFlow::Node node) { none() }
final override predicate isBarrierIn(DataFlow::Node node) { isSanitizerIn(node) }
/** Holds if taint propagation out of `node` is prohibited. */
predicate isSanitizerOut(DataFlow::Node node) { none() }
final override predicate isBarrierOut(DataFlow::Node node) { isSanitizerOut(node) }
/** Holds if taint propagation through nodes guarded by `guard` is prohibited. */
predicate isSanitizerGuard(DataFlow::BarrierGuard guard) { none() }
final override predicate isBarrierGuard(DataFlow::BarrierGuard guard) { isSanitizerGuard(guard) }
/**
* Holds if the additional taint propagation step from `node1` to `node2`
* must be taken into account in the analysis.
*/
predicate isAdditionalTaintStep(DataFlow::Node node1, DataFlow::Node node2) { none() }
final override predicate isAdditionalFlowStep(DataFlow::Node node1, DataFlow::Node node2) {
isAdditionalTaintStep(node1, node2) or
defaultAdditionalTaintStep(node1, node2)
}
override predicate allowImplicitRead(DataFlow::Node node, DataFlow::Content c) {
(this.isSink(node) or this.isAdditionalTaintStep(node, _)) and
defaultImplicitTaintRead(node, c)
}
/**
* Holds if taint may flow from `source` to `sink` for this configuration.
*/
// overridden to provide taint-tracking specific qldoc
override predicate hasFlow(DataFlow::Node source, DataFlow::Node sink) {
super.hasFlow(source, sink)
}
}

View File

@@ -1,527 +0,0 @@
private import IR
import InstructionConsistency // module is below
import IRTypeConsistency // module is in IRType.qll
module InstructionConsistency {
private import internal.InstructionImports as Imports
private import Imports::OperandTag
private import Imports::Overlap
private import internal.IRInternal
private newtype TOptionalIRFunction =
TPresentIRFunction(IRFunction irFunc) or
TMissingIRFunction()
/**
* An `IRFunction` that might not exist. This is used so that we can produce consistency failures
* for IR that also incorrectly lacks a `getEnclosingIRFunction()`.
*/
abstract private class OptionalIRFunction extends TOptionalIRFunction {
abstract string toString();
abstract Language::Location getLocation();
}
private class PresentIRFunction extends OptionalIRFunction, TPresentIRFunction {
private IRFunction irFunc;
PresentIRFunction() { this = TPresentIRFunction(irFunc) }
override string toString() {
result = concat(Language::getIdentityString(irFunc.getFunction()), "; ")
}
override Language::Location getLocation() {
// To avoid an overwhelming number of results when the extractor merges functions with the
// same name, just pick a single location.
result =
min(Language::Location loc | loc = irFunc.getLocation() | loc order by loc.toString())
}
}
private class MissingIRFunction extends OptionalIRFunction, TMissingIRFunction {
override string toString() { result = "<Missing IRFunction>" }
override Language::Location getLocation() { result instanceof Language::UnknownDefaultLocation }
}
private OptionalIRFunction getInstructionIRFunction(Instruction instr) {
result = TPresentIRFunction(instr.getEnclosingIRFunction())
or
not exists(instr.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getInstructionIRFunction(Instruction instr, string irFuncText) {
result = getInstructionIRFunction(instr) and
irFuncText = result.toString()
}
private OptionalIRFunction getOperandIRFunction(Operand operand) {
result = TPresentIRFunction(operand.getEnclosingIRFunction())
or
not exists(operand.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getOperandIRFunction(Operand operand, string irFuncText) {
result = getOperandIRFunction(operand) and
irFuncText = result.toString()
}
private OptionalIRFunction getBlockIRFunction(IRBlock block) {
result = TPresentIRFunction(block.getEnclosingIRFunction())
or
not exists(block.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
/**
* Holds if instruction `instr` is missing an expected operand with tag `tag`.
*/
query predicate missingOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
instr.getOpcode().hasOperand(tag) and
not exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
message =
"Instruction '" + instr.getOpcode().toString() +
"' is missing an expected operand with tag '" + tag.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has an unexpected operand with tag `tag`.
*/
query predicate unexpectedOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
not instr.getOpcode().hasOperand(tag) and
not (instr instanceof CallInstruction and tag instanceof ArgumentOperandTag) and
not (
instr instanceof BuiltInOperationInstruction and tag instanceof PositionalArgumentOperandTag
) and
not (instr instanceof InlineAsmInstruction and tag instanceof AsmOperandTag) and
message =
"Instruction '" + instr.toString() + "' has unexpected operand '" + tag.toString() +
"' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has multiple operands with tag `tag`.
*/
query predicate duplicateOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag, int operandCount |
operandCount =
strictcount(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
operandCount > 1 and
message =
"Instruction has " + operandCount + " operands with tag '" + tag.toString() + "'" +
" in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if `Phi` instruction `instr` is missing an operand corresponding to
* the predecessor block `pred`.
*/
query predicate missingPhiOperand(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock pred |
pred = instr.getBlock().getAPredecessor() and
not exists(PhiInputOperand operand |
operand = instr.getAnOperand() and
operand.getPredecessorBlock() = pred
) and
message =
"Instruction '" + instr.toString() + "' is missing an operand for predecessor block '" +
pred.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
query predicate missingOperandType(
Operand operand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Instruction use |
not exists(operand.getType()) and
use = operand.getUse() and
message =
"Operand '" + operand.toString() + "' of instruction '" + use.getOpcode().toString() +
"' is missing a type in function '$@'." and
irFunc = getOperandIRFunction(operand, irFuncText)
)
}
query predicate duplicateChiOperand(
ChiInstruction chi, string message, OptionalIRFunction irFunc, string irFuncText
) {
chi.getTotal() = chi.getPartial() and
message =
"Chi instruction for " + chi.getPartial().toString() +
" has duplicate operands in function '$@'." and
irFunc = getInstructionIRFunction(chi, irFuncText)
}
query predicate sideEffectWithoutPrimary(
SideEffectInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getPrimaryInstruction()) and
message =
"Side effect instruction '" + instr + "' is missing a primary instruction in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if an instruction, other than `ExitFunction`, has no successors.
*/
query predicate instructionWithoutSuccessor(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getASuccessor()) and
not instr instanceof ExitFunctionInstruction and
// Phi instructions aren't linked into the instruction-level flow graph.
not instr instanceof PhiInstruction and
not instr instanceof UnreachedInstruction and
message = "Instruction '" + instr.toString() + "' has no successors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if there are multiple edges of the same kind from `source`.
*/
query predicate ambiguousSuccessors(
Instruction source, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(EdgeKind kind, int n |
n = strictcount(Instruction t | source.getSuccessor(kind) = t) and
n > 1 and
message =
"Instruction '" + source.toString() + "' has " + n.toString() + " successors of kind '" +
kind.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(source, irFuncText)
)
}
/**
* Holds if `instr` is part of a loop even though the AST of `instr`'s enclosing function
* contains no element that can cause loops.
*/
query predicate unexplainedLoop(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Language::Function f |
exists(IRBlock block |
instr.getBlock() = block and
block.getEnclosingFunction() = f and
block.getASuccessor+() = block
) and
not Language::hasPotentialLoop(f) and
message =
"Instruction '" + instr.toString() + "' is part of an unexplained loop in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a `Phi` instruction is present in a block with fewer than two
* predecessors.
*/
query predicate unnecessaryPhiInstruction(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int n |
n = count(instr.getBlock().getAPredecessor()) and
n < 2 and
message =
"Instruction '" + instr.toString() + "' is in a block with only " + n.toString() +
" predecessors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a memory operand is connected to a definition with an unmodeled result.
*/
query predicate memoryOperandDefinitionIsUnmodeled(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(MemoryOperand operand, Instruction def |
operand = instr.getAnOperand() and
def = operand.getAnyDef() and
not def.isResultModeled() and
message =
"Memory operand definition on instruction '" + instr.toString() +
"' has unmodeled result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if operand `operand` consumes a value that was defined in
* a different function.
*/
query predicate operandAcrossFunctions(
Operand operand, string message, OptionalIRFunction useIRFunc, string useIRFuncText,
OptionalIRFunction defIRFunc, string defIRFuncText
) {
exists(Instruction useInstr, Instruction defInstr |
operand.getUse() = useInstr and
operand.getAnyDef() = defInstr and
useIRFunc = getInstructionIRFunction(useInstr, useIRFuncText) and
defIRFunc = getInstructionIRFunction(defInstr, defIRFuncText) and
useIRFunc != defIRFunc and
message =
"Operand '" + operand.toString() + "' is used on instruction '" + useInstr.toString() +
"' in function '$@', but is defined on instruction '" + defInstr.toString() +
"' in function '$@'."
)
}
/**
* Holds if instruction `instr` is not in exactly one block.
*/
query predicate instructionWithoutUniqueBlock(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int blockCount |
blockCount = count(instr.getBlock()) and
blockCount != 1 and
message =
"Instruction '" + instr.toString() + "' is a member of " + blockCount.toString() +
" blocks in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
private predicate forwardEdge(IRBlock b1, IRBlock b2) {
b1.getASuccessor() = b2 and
not b1.getBackEdgeSuccessor(_) = b2
}
/**
* Holds if `f` contains a loop in which no edge is a back edge.
*
* This check ensures we don't have too _few_ back edges.
*/
query predicate containsLoopOfForwardEdges(IRFunction f, string message) {
exists(IRBlock block |
forwardEdge+(block, block) and
block.getEnclosingIRFunction() = f and
message = "Function contains a loop consisting of only forward edges."
)
}
/**
* Holds if `block` is reachable from its function entry point but would not
* be reachable by traversing only forward edges. This check is skipped for
* functions containing `goto` statements as the property does not generally
* hold there.
*
* This check ensures we don't have too _many_ back edges.
*/
query predicate lostReachability(
IRBlock block, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRFunction f, IRBlock entry |
entry = f.getEntryBlock() and
entry.getASuccessor+() = block and
not forwardEdge+(entry, block) and
not Language::hasGoto(f.getFunction()) and
message =
"Block '" + block.toString() +
"' is not reachable by traversing only forward edges in function '$@'." and
irFunc = TPresentIRFunction(f) and
irFuncText = irFunc.toString()
)
}
/**
* Holds if the number of back edges differs between the `Instruction` graph
* and the `IRBlock` graph.
*/
query predicate backEdgeCountMismatch(OptionalIRFunction irFunc, string message) {
exists(int fromInstr, int fromBlock |
fromInstr =
count(Instruction i1, Instruction i2 |
getInstructionIRFunction(i1) = irFunc and i1.getBackEdgeSuccessor(_) = i2
) and
fromBlock =
count(IRBlock b1, IRBlock b2 |
getBlockIRFunction(b1) = irFunc and b1.getBackEdgeSuccessor(_) = b2
) and
fromInstr != fromBlock and
message =
"The instruction graph for function '" + irFunc.toString() + "' contains " +
fromInstr.toString() + " back edges, but the block graph contains " + fromBlock.toString()
+ " back edges."
)
}
/**
* Gets the point in the function at which the specified operand is evaluated. For most operands,
* this is at the instruction that consumes the use. For a `PhiInputOperand`, the effective point
* of evaluation is at the end of the corresponding predecessor block.
*/
private predicate pointOfEvaluation(Operand operand, IRBlock block, int index) {
block = operand.(PhiInputOperand).getPredecessorBlock() and
index = block.getInstructionCount()
or
exists(Instruction use |
use = operand.(NonPhiOperand).getUse() and
block.getInstruction(index) = use
)
}
/**
* Holds if `useOperand` has a definition that does not dominate the use.
*/
query predicate useNotDominatedByDefinition(
Operand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock useBlock, int useIndex, Instruction defInstr, IRBlock defBlock, int defIndex |
pointOfEvaluation(useOperand, useBlock, useIndex) and
defInstr = useOperand.getAnyDef() and
(
defInstr instanceof PhiInstruction and
defBlock = defInstr.getBlock() and
defIndex = -1
or
defBlock.getInstruction(defIndex) = defInstr
) and
not (
defBlock.strictlyDominates(useBlock)
or
defBlock = useBlock and
defIndex < useIndex
) and
message =
"Operand '" + useOperand.toString() +
"' is not dominated by its definition in function '$@'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate switchInstructionWithoutDefaultEdge(
SwitchInstruction switchInstr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(switchInstr.getDefaultSuccessor()) and
message =
"SwitchInstruction " + switchInstr.toString() + " without a DefaultEdge in function '$@'." and
irFunc = getInstructionIRFunction(switchInstr, irFuncText)
}
/**
* Holds if `instr` is on the chain of chi/phi instructions for all aliased
* memory.
*/
private predicate isOnAliasedDefinitionChain(Instruction instr) {
instr instanceof AliasedDefinitionInstruction
or
isOnAliasedDefinitionChain(instr.(ChiInstruction).getTotal())
or
isOnAliasedDefinitionChain(instr.(PhiInstruction).getAnInputOperand().getAnyDef())
}
private predicate shouldBeConflated(Instruction instr) {
isOnAliasedDefinitionChain(instr)
or
instr.getOpcode() instanceof Opcode::InitializeNonLocal
}
query predicate notMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
shouldBeConflated(instr) and
not instr.isResultConflated() and
message =
"Instruction '" + instr.toString() +
"' should be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate wronglyMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
instr.isResultConflated() and
not shouldBeConflated(instr) and
message =
"Instruction '" + instr.toString() +
"' should not be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate invalidOverlap(
MemoryOperand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Overlap overlap |
overlap = useOperand.getDefinitionOverlap() and
overlap instanceof MayPartiallyOverlap and
message =
"MemoryOperand '" + useOperand.toString() + "' has a `getDefinitionOverlap()` of '" +
overlap.toString() + "'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate nonUniqueEnclosingIRFunction(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int irFuncCount |
irFuncCount = count(instr.getEnclosingIRFunction()) and
irFuncCount != 1 and
message =
"Instruction '" + instr.toString() + "' has " + irFuncCount.toString() +
" results for `getEnclosingIRFunction()` in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if the object address operand for the given `FieldAddress` instruction does not have an
* address type.
*/
query predicate fieldAddressOnNonPointer(
FieldAddressInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not instr.getObjectAddressOperand().getIRType() instanceof IRAddressType and
message =
"FieldAddress instruction '" + instr.toString() +
"' has an object address operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if the `this` argument operand for the given `Call` instruction does not have an address
* type.
*/
query predicate thisArgumentIsNonPointer(
CallInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(ThisArgumentOperand thisOperand | thisOperand = instr.getThisArgumentOperand() |
not thisOperand.getIRType() instanceof IRAddressType
) and
message =
"Call instruction '" + instr.toString() +
"' has a `this` argument operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
}

View File

@@ -1,527 +0,0 @@
private import IR
import InstructionConsistency // module is below
import IRTypeConsistency // module is in IRType.qll
module InstructionConsistency {
private import internal.InstructionImports as Imports
private import Imports::OperandTag
private import Imports::Overlap
private import internal.IRInternal
private newtype TOptionalIRFunction =
TPresentIRFunction(IRFunction irFunc) or
TMissingIRFunction()
/**
* An `IRFunction` that might not exist. This is used so that we can produce consistency failures
* for IR that also incorrectly lacks a `getEnclosingIRFunction()`.
*/
abstract private class OptionalIRFunction extends TOptionalIRFunction {
abstract string toString();
abstract Language::Location getLocation();
}
private class PresentIRFunction extends OptionalIRFunction, TPresentIRFunction {
private IRFunction irFunc;
PresentIRFunction() { this = TPresentIRFunction(irFunc) }
override string toString() {
result = concat(Language::getIdentityString(irFunc.getFunction()), "; ")
}
override Language::Location getLocation() {
// To avoid an overwhelming number of results when the extractor merges functions with the
// same name, just pick a single location.
result =
min(Language::Location loc | loc = irFunc.getLocation() | loc order by loc.toString())
}
}
private class MissingIRFunction extends OptionalIRFunction, TMissingIRFunction {
override string toString() { result = "<Missing IRFunction>" }
override Language::Location getLocation() { result instanceof Language::UnknownDefaultLocation }
}
private OptionalIRFunction getInstructionIRFunction(Instruction instr) {
result = TPresentIRFunction(instr.getEnclosingIRFunction())
or
not exists(instr.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getInstructionIRFunction(Instruction instr, string irFuncText) {
result = getInstructionIRFunction(instr) and
irFuncText = result.toString()
}
private OptionalIRFunction getOperandIRFunction(Operand operand) {
result = TPresentIRFunction(operand.getEnclosingIRFunction())
or
not exists(operand.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getOperandIRFunction(Operand operand, string irFuncText) {
result = getOperandIRFunction(operand) and
irFuncText = result.toString()
}
private OptionalIRFunction getBlockIRFunction(IRBlock block) {
result = TPresentIRFunction(block.getEnclosingIRFunction())
or
not exists(block.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
/**
* Holds if instruction `instr` is missing an expected operand with tag `tag`.
*/
query predicate missingOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
instr.getOpcode().hasOperand(tag) and
not exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
message =
"Instruction '" + instr.getOpcode().toString() +
"' is missing an expected operand with tag '" + tag.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has an unexpected operand with tag `tag`.
*/
query predicate unexpectedOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
not instr.getOpcode().hasOperand(tag) and
not (instr instanceof CallInstruction and tag instanceof ArgumentOperandTag) and
not (
instr instanceof BuiltInOperationInstruction and tag instanceof PositionalArgumentOperandTag
) and
not (instr instanceof InlineAsmInstruction and tag instanceof AsmOperandTag) and
message =
"Instruction '" + instr.toString() + "' has unexpected operand '" + tag.toString() +
"' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has multiple operands with tag `tag`.
*/
query predicate duplicateOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag, int operandCount |
operandCount =
strictcount(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
operandCount > 1 and
message =
"Instruction has " + operandCount + " operands with tag '" + tag.toString() + "'" +
" in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if `Phi` instruction `instr` is missing an operand corresponding to
* the predecessor block `pred`.
*/
query predicate missingPhiOperand(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock pred |
pred = instr.getBlock().getAPredecessor() and
not exists(PhiInputOperand operand |
operand = instr.getAnOperand() and
operand.getPredecessorBlock() = pred
) and
message =
"Instruction '" + instr.toString() + "' is missing an operand for predecessor block '" +
pred.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
query predicate missingOperandType(
Operand operand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Instruction use |
not exists(operand.getType()) and
use = operand.getUse() and
message =
"Operand '" + operand.toString() + "' of instruction '" + use.getOpcode().toString() +
"' is missing a type in function '$@'." and
irFunc = getOperandIRFunction(operand, irFuncText)
)
}
query predicate duplicateChiOperand(
ChiInstruction chi, string message, OptionalIRFunction irFunc, string irFuncText
) {
chi.getTotal() = chi.getPartial() and
message =
"Chi instruction for " + chi.getPartial().toString() +
" has duplicate operands in function '$@'." and
irFunc = getInstructionIRFunction(chi, irFuncText)
}
query predicate sideEffectWithoutPrimary(
SideEffectInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getPrimaryInstruction()) and
message =
"Side effect instruction '" + instr + "' is missing a primary instruction in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if an instruction, other than `ExitFunction`, has no successors.
*/
query predicate instructionWithoutSuccessor(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getASuccessor()) and
not instr instanceof ExitFunctionInstruction and
// Phi instructions aren't linked into the instruction-level flow graph.
not instr instanceof PhiInstruction and
not instr instanceof UnreachedInstruction and
message = "Instruction '" + instr.toString() + "' has no successors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if there are multiple edges of the same kind from `source`.
*/
query predicate ambiguousSuccessors(
Instruction source, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(EdgeKind kind, int n |
n = strictcount(Instruction t | source.getSuccessor(kind) = t) and
n > 1 and
message =
"Instruction '" + source.toString() + "' has " + n.toString() + " successors of kind '" +
kind.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(source, irFuncText)
)
}
/**
* Holds if `instr` is part of a loop even though the AST of `instr`'s enclosing function
* contains no element that can cause loops.
*/
query predicate unexplainedLoop(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Language::Function f |
exists(IRBlock block |
instr.getBlock() = block and
block.getEnclosingFunction() = f and
block.getASuccessor+() = block
) and
not Language::hasPotentialLoop(f) and
message =
"Instruction '" + instr.toString() + "' is part of an unexplained loop in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a `Phi` instruction is present in a block with fewer than two
* predecessors.
*/
query predicate unnecessaryPhiInstruction(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int n |
n = count(instr.getBlock().getAPredecessor()) and
n < 2 and
message =
"Instruction '" + instr.toString() + "' is in a block with only " + n.toString() +
" predecessors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a memory operand is connected to a definition with an unmodeled result.
*/
query predicate memoryOperandDefinitionIsUnmodeled(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(MemoryOperand operand, Instruction def |
operand = instr.getAnOperand() and
def = operand.getAnyDef() and
not def.isResultModeled() and
message =
"Memory operand definition on instruction '" + instr.toString() +
"' has unmodeled result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if operand `operand` consumes a value that was defined in
* a different function.
*/
query predicate operandAcrossFunctions(
Operand operand, string message, OptionalIRFunction useIRFunc, string useIRFuncText,
OptionalIRFunction defIRFunc, string defIRFuncText
) {
exists(Instruction useInstr, Instruction defInstr |
operand.getUse() = useInstr and
operand.getAnyDef() = defInstr and
useIRFunc = getInstructionIRFunction(useInstr, useIRFuncText) and
defIRFunc = getInstructionIRFunction(defInstr, defIRFuncText) and
useIRFunc != defIRFunc and
message =
"Operand '" + operand.toString() + "' is used on instruction '" + useInstr.toString() +
"' in function '$@', but is defined on instruction '" + defInstr.toString() +
"' in function '$@'."
)
}
/**
* Holds if instruction `instr` is not in exactly one block.
*/
query predicate instructionWithoutUniqueBlock(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int blockCount |
blockCount = count(instr.getBlock()) and
blockCount != 1 and
message =
"Instruction '" + instr.toString() + "' is a member of " + blockCount.toString() +
" blocks in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
private predicate forwardEdge(IRBlock b1, IRBlock b2) {
b1.getASuccessor() = b2 and
not b1.getBackEdgeSuccessor(_) = b2
}
/**
* Holds if `f` contains a loop in which no edge is a back edge.
*
* This check ensures we don't have too _few_ back edges.
*/
query predicate containsLoopOfForwardEdges(IRFunction f, string message) {
exists(IRBlock block |
forwardEdge+(block, block) and
block.getEnclosingIRFunction() = f and
message = "Function contains a loop consisting of only forward edges."
)
}
/**
* Holds if `block` is reachable from its function entry point but would not
* be reachable by traversing only forward edges. This check is skipped for
* functions containing `goto` statements as the property does not generally
* hold there.
*
* This check ensures we don't have too _many_ back edges.
*/
query predicate lostReachability(
IRBlock block, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRFunction f, IRBlock entry |
entry = f.getEntryBlock() and
entry.getASuccessor+() = block and
not forwardEdge+(entry, block) and
not Language::hasGoto(f.getFunction()) and
message =
"Block '" + block.toString() +
"' is not reachable by traversing only forward edges in function '$@'." and
irFunc = TPresentIRFunction(f) and
irFuncText = irFunc.toString()
)
}
/**
* Holds if the number of back edges differs between the `Instruction` graph
* and the `IRBlock` graph.
*/
query predicate backEdgeCountMismatch(OptionalIRFunction irFunc, string message) {
exists(int fromInstr, int fromBlock |
fromInstr =
count(Instruction i1, Instruction i2 |
getInstructionIRFunction(i1) = irFunc and i1.getBackEdgeSuccessor(_) = i2
) and
fromBlock =
count(IRBlock b1, IRBlock b2 |
getBlockIRFunction(b1) = irFunc and b1.getBackEdgeSuccessor(_) = b2
) and
fromInstr != fromBlock and
message =
"The instruction graph for function '" + irFunc.toString() + "' contains " +
fromInstr.toString() + " back edges, but the block graph contains " + fromBlock.toString()
+ " back edges."
)
}
/**
* Gets the point in the function at which the specified operand is evaluated. For most operands,
* this is at the instruction that consumes the use. For a `PhiInputOperand`, the effective point
* of evaluation is at the end of the corresponding predecessor block.
*/
private predicate pointOfEvaluation(Operand operand, IRBlock block, int index) {
block = operand.(PhiInputOperand).getPredecessorBlock() and
index = block.getInstructionCount()
or
exists(Instruction use |
use = operand.(NonPhiOperand).getUse() and
block.getInstruction(index) = use
)
}
/**
* Holds if `useOperand` has a definition that does not dominate the use.
*/
query predicate useNotDominatedByDefinition(
Operand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock useBlock, int useIndex, Instruction defInstr, IRBlock defBlock, int defIndex |
pointOfEvaluation(useOperand, useBlock, useIndex) and
defInstr = useOperand.getAnyDef() and
(
defInstr instanceof PhiInstruction and
defBlock = defInstr.getBlock() and
defIndex = -1
or
defBlock.getInstruction(defIndex) = defInstr
) and
not (
defBlock.strictlyDominates(useBlock)
or
defBlock = useBlock and
defIndex < useIndex
) and
message =
"Operand '" + useOperand.toString() +
"' is not dominated by its definition in function '$@'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate switchInstructionWithoutDefaultEdge(
SwitchInstruction switchInstr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(switchInstr.getDefaultSuccessor()) and
message =
"SwitchInstruction " + switchInstr.toString() + " without a DefaultEdge in function '$@'." and
irFunc = getInstructionIRFunction(switchInstr, irFuncText)
}
/**
* Holds if `instr` is on the chain of chi/phi instructions for all aliased
* memory.
*/
private predicate isOnAliasedDefinitionChain(Instruction instr) {
instr instanceof AliasedDefinitionInstruction
or
isOnAliasedDefinitionChain(instr.(ChiInstruction).getTotal())
or
isOnAliasedDefinitionChain(instr.(PhiInstruction).getAnInputOperand().getAnyDef())
}
private predicate shouldBeConflated(Instruction instr) {
isOnAliasedDefinitionChain(instr)
or
instr.getOpcode() instanceof Opcode::InitializeNonLocal
}
query predicate notMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
shouldBeConflated(instr) and
not instr.isResultConflated() and
message =
"Instruction '" + instr.toString() +
"' should be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate wronglyMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
instr.isResultConflated() and
not shouldBeConflated(instr) and
message =
"Instruction '" + instr.toString() +
"' should not be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate invalidOverlap(
MemoryOperand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Overlap overlap |
overlap = useOperand.getDefinitionOverlap() and
overlap instanceof MayPartiallyOverlap and
message =
"MemoryOperand '" + useOperand.toString() + "' has a `getDefinitionOverlap()` of '" +
overlap.toString() + "'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate nonUniqueEnclosingIRFunction(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int irFuncCount |
irFuncCount = count(instr.getEnclosingIRFunction()) and
irFuncCount != 1 and
message =
"Instruction '" + instr.toString() + "' has " + irFuncCount.toString() +
" results for `getEnclosingIRFunction()` in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if the object address operand for the given `FieldAddress` instruction does not have an
* address type.
*/
query predicate fieldAddressOnNonPointer(
FieldAddressInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not instr.getObjectAddressOperand().getIRType() instanceof IRAddressType and
message =
"FieldAddress instruction '" + instr.toString() +
"' has an object address operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if the `this` argument operand for the given `Call` instruction does not have an address
* type.
*/
query predicate thisArgumentIsNonPointer(
CallInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(ThisArgumentOperand thisOperand | thisOperand = instr.getThisArgumentOperand() |
not thisOperand.getIRType() instanceof IRAddressType
) and
message =
"Call instruction '" + instr.toString() +
"' has a `this` argument operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
}

View File

@@ -1,527 +0,0 @@
private import IR
import InstructionConsistency // module is below
import IRTypeConsistency // module is in IRType.qll
module InstructionConsistency {
private import internal.InstructionImports as Imports
private import Imports::OperandTag
private import Imports::Overlap
private import internal.IRInternal
private newtype TOptionalIRFunction =
TPresentIRFunction(IRFunction irFunc) or
TMissingIRFunction()
/**
* An `IRFunction` that might not exist. This is used so that we can produce consistency failures
* for IR that also incorrectly lacks a `getEnclosingIRFunction()`.
*/
abstract private class OptionalIRFunction extends TOptionalIRFunction {
abstract string toString();
abstract Language::Location getLocation();
}
private class PresentIRFunction extends OptionalIRFunction, TPresentIRFunction {
private IRFunction irFunc;
PresentIRFunction() { this = TPresentIRFunction(irFunc) }
override string toString() {
result = concat(Language::getIdentityString(irFunc.getFunction()), "; ")
}
override Language::Location getLocation() {
// To avoid an overwhelming number of results when the extractor merges functions with the
// same name, just pick a single location.
result =
min(Language::Location loc | loc = irFunc.getLocation() | loc order by loc.toString())
}
}
private class MissingIRFunction extends OptionalIRFunction, TMissingIRFunction {
override string toString() { result = "<Missing IRFunction>" }
override Language::Location getLocation() { result instanceof Language::UnknownDefaultLocation }
}
private OptionalIRFunction getInstructionIRFunction(Instruction instr) {
result = TPresentIRFunction(instr.getEnclosingIRFunction())
or
not exists(instr.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getInstructionIRFunction(Instruction instr, string irFuncText) {
result = getInstructionIRFunction(instr) and
irFuncText = result.toString()
}
private OptionalIRFunction getOperandIRFunction(Operand operand) {
result = TPresentIRFunction(operand.getEnclosingIRFunction())
or
not exists(operand.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
pragma[inline]
private OptionalIRFunction getOperandIRFunction(Operand operand, string irFuncText) {
result = getOperandIRFunction(operand) and
irFuncText = result.toString()
}
private OptionalIRFunction getBlockIRFunction(IRBlock block) {
result = TPresentIRFunction(block.getEnclosingIRFunction())
or
not exists(block.getEnclosingIRFunction()) and result = TMissingIRFunction()
}
/**
* Holds if instruction `instr` is missing an expected operand with tag `tag`.
*/
query predicate missingOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
instr.getOpcode().hasOperand(tag) and
not exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
message =
"Instruction '" + instr.getOpcode().toString() +
"' is missing an expected operand with tag '" + tag.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has an unexpected operand with tag `tag`.
*/
query predicate unexpectedOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag |
exists(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
not instr.getOpcode().hasOperand(tag) and
not (instr instanceof CallInstruction and tag instanceof ArgumentOperandTag) and
not (
instr instanceof BuiltInOperationInstruction and tag instanceof PositionalArgumentOperandTag
) and
not (instr instanceof InlineAsmInstruction and tag instanceof AsmOperandTag) and
message =
"Instruction '" + instr.toString() + "' has unexpected operand '" + tag.toString() +
"' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if instruction `instr` has multiple operands with tag `tag`.
*/
query predicate duplicateOperand(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(OperandTag tag, int operandCount |
operandCount =
strictcount(NonPhiOperand operand |
operand = instr.getAnOperand() and
operand.getOperandTag() = tag
) and
operandCount > 1 and
message =
"Instruction has " + operandCount + " operands with tag '" + tag.toString() + "'" +
" in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if `Phi` instruction `instr` is missing an operand corresponding to
* the predecessor block `pred`.
*/
query predicate missingPhiOperand(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock pred |
pred = instr.getBlock().getAPredecessor() and
not exists(PhiInputOperand operand |
operand = instr.getAnOperand() and
operand.getPredecessorBlock() = pred
) and
message =
"Instruction '" + instr.toString() + "' is missing an operand for predecessor block '" +
pred.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
query predicate missingOperandType(
Operand operand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Instruction use |
not exists(operand.getType()) and
use = operand.getUse() and
message =
"Operand '" + operand.toString() + "' of instruction '" + use.getOpcode().toString() +
"' is missing a type in function '$@'." and
irFunc = getOperandIRFunction(operand, irFuncText)
)
}
query predicate duplicateChiOperand(
ChiInstruction chi, string message, OptionalIRFunction irFunc, string irFuncText
) {
chi.getTotal() = chi.getPartial() and
message =
"Chi instruction for " + chi.getPartial().toString() +
" has duplicate operands in function '$@'." and
irFunc = getInstructionIRFunction(chi, irFuncText)
}
query predicate sideEffectWithoutPrimary(
SideEffectInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getPrimaryInstruction()) and
message =
"Side effect instruction '" + instr + "' is missing a primary instruction in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if an instruction, other than `ExitFunction`, has no successors.
*/
query predicate instructionWithoutSuccessor(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(instr.getASuccessor()) and
not instr instanceof ExitFunctionInstruction and
// Phi instructions aren't linked into the instruction-level flow graph.
not instr instanceof PhiInstruction and
not instr instanceof UnreachedInstruction and
message = "Instruction '" + instr.toString() + "' has no successors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if there are multiple edges of the same kind from `source`.
*/
query predicate ambiguousSuccessors(
Instruction source, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(EdgeKind kind, int n |
n = strictcount(Instruction t | source.getSuccessor(kind) = t) and
n > 1 and
message =
"Instruction '" + source.toString() + "' has " + n.toString() + " successors of kind '" +
kind.toString() + "' in function '$@'." and
irFunc = getInstructionIRFunction(source, irFuncText)
)
}
/**
* Holds if `instr` is part of a loop even though the AST of `instr`'s enclosing function
* contains no element that can cause loops.
*/
query predicate unexplainedLoop(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Language::Function f |
exists(IRBlock block |
instr.getBlock() = block and
block.getEnclosingFunction() = f and
block.getASuccessor+() = block
) and
not Language::hasPotentialLoop(f) and
message =
"Instruction '" + instr.toString() + "' is part of an unexplained loop in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a `Phi` instruction is present in a block with fewer than two
* predecessors.
*/
query predicate unnecessaryPhiInstruction(
PhiInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int n |
n = count(instr.getBlock().getAPredecessor()) and
n < 2 and
message =
"Instruction '" + instr.toString() + "' is in a block with only " + n.toString() +
" predecessors in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if a memory operand is connected to a definition with an unmodeled result.
*/
query predicate memoryOperandDefinitionIsUnmodeled(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(MemoryOperand operand, Instruction def |
operand = instr.getAnOperand() and
def = operand.getAnyDef() and
not def.isResultModeled() and
message =
"Memory operand definition on instruction '" + instr.toString() +
"' has unmodeled result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if operand `operand` consumes a value that was defined in
* a different function.
*/
query predicate operandAcrossFunctions(
Operand operand, string message, OptionalIRFunction useIRFunc, string useIRFuncText,
OptionalIRFunction defIRFunc, string defIRFuncText
) {
exists(Instruction useInstr, Instruction defInstr |
operand.getUse() = useInstr and
operand.getAnyDef() = defInstr and
useIRFunc = getInstructionIRFunction(useInstr, useIRFuncText) and
defIRFunc = getInstructionIRFunction(defInstr, defIRFuncText) and
useIRFunc != defIRFunc and
message =
"Operand '" + operand.toString() + "' is used on instruction '" + useInstr.toString() +
"' in function '$@', but is defined on instruction '" + defInstr.toString() +
"' in function '$@'."
)
}
/**
* Holds if instruction `instr` is not in exactly one block.
*/
query predicate instructionWithoutUniqueBlock(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int blockCount |
blockCount = count(instr.getBlock()) and
blockCount != 1 and
message =
"Instruction '" + instr.toString() + "' is a member of " + blockCount.toString() +
" blocks in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
private predicate forwardEdge(IRBlock b1, IRBlock b2) {
b1.getASuccessor() = b2 and
not b1.getBackEdgeSuccessor(_) = b2
}
/**
* Holds if `f` contains a loop in which no edge is a back edge.
*
* This check ensures we don't have too _few_ back edges.
*/
query predicate containsLoopOfForwardEdges(IRFunction f, string message) {
exists(IRBlock block |
forwardEdge+(block, block) and
block.getEnclosingIRFunction() = f and
message = "Function contains a loop consisting of only forward edges."
)
}
/**
* Holds if `block` is reachable from its function entry point but would not
* be reachable by traversing only forward edges. This check is skipped for
* functions containing `goto` statements as the property does not generally
* hold there.
*
* This check ensures we don't have too _many_ back edges.
*/
query predicate lostReachability(
IRBlock block, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRFunction f, IRBlock entry |
entry = f.getEntryBlock() and
entry.getASuccessor+() = block and
not forwardEdge+(entry, block) and
not Language::hasGoto(f.getFunction()) and
message =
"Block '" + block.toString() +
"' is not reachable by traversing only forward edges in function '$@'." and
irFunc = TPresentIRFunction(f) and
irFuncText = irFunc.toString()
)
}
/**
* Holds if the number of back edges differs between the `Instruction` graph
* and the `IRBlock` graph.
*/
query predicate backEdgeCountMismatch(OptionalIRFunction irFunc, string message) {
exists(int fromInstr, int fromBlock |
fromInstr =
count(Instruction i1, Instruction i2 |
getInstructionIRFunction(i1) = irFunc and i1.getBackEdgeSuccessor(_) = i2
) and
fromBlock =
count(IRBlock b1, IRBlock b2 |
getBlockIRFunction(b1) = irFunc and b1.getBackEdgeSuccessor(_) = b2
) and
fromInstr != fromBlock and
message =
"The instruction graph for function '" + irFunc.toString() + "' contains " +
fromInstr.toString() + " back edges, but the block graph contains " + fromBlock.toString()
+ " back edges."
)
}
/**
* Gets the point in the function at which the specified operand is evaluated. For most operands,
* this is at the instruction that consumes the use. For a `PhiInputOperand`, the effective point
* of evaluation is at the end of the corresponding predecessor block.
*/
private predicate pointOfEvaluation(Operand operand, IRBlock block, int index) {
block = operand.(PhiInputOperand).getPredecessorBlock() and
index = block.getInstructionCount()
or
exists(Instruction use |
use = operand.(NonPhiOperand).getUse() and
block.getInstruction(index) = use
)
}
/**
* Holds if `useOperand` has a definition that does not dominate the use.
*/
query predicate useNotDominatedByDefinition(
Operand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(IRBlock useBlock, int useIndex, Instruction defInstr, IRBlock defBlock, int defIndex |
pointOfEvaluation(useOperand, useBlock, useIndex) and
defInstr = useOperand.getAnyDef() and
(
defInstr instanceof PhiInstruction and
defBlock = defInstr.getBlock() and
defIndex = -1
or
defBlock.getInstruction(defIndex) = defInstr
) and
not (
defBlock.strictlyDominates(useBlock)
or
defBlock = useBlock and
defIndex < useIndex
) and
message =
"Operand '" + useOperand.toString() +
"' is not dominated by its definition in function '$@'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate switchInstructionWithoutDefaultEdge(
SwitchInstruction switchInstr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not exists(switchInstr.getDefaultSuccessor()) and
message =
"SwitchInstruction " + switchInstr.toString() + " without a DefaultEdge in function '$@'." and
irFunc = getInstructionIRFunction(switchInstr, irFuncText)
}
/**
* Holds if `instr` is on the chain of chi/phi instructions for all aliased
* memory.
*/
private predicate isOnAliasedDefinitionChain(Instruction instr) {
instr instanceof AliasedDefinitionInstruction
or
isOnAliasedDefinitionChain(instr.(ChiInstruction).getTotal())
or
isOnAliasedDefinitionChain(instr.(PhiInstruction).getAnInputOperand().getAnyDef())
}
private predicate shouldBeConflated(Instruction instr) {
isOnAliasedDefinitionChain(instr)
or
instr.getOpcode() instanceof Opcode::InitializeNonLocal
}
query predicate notMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
shouldBeConflated(instr) and
not instr.isResultConflated() and
message =
"Instruction '" + instr.toString() +
"' should be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate wronglyMarkedAsConflated(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
instr.isResultConflated() and
not shouldBeConflated(instr) and
message =
"Instruction '" + instr.toString() +
"' should not be marked as having a conflated result in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
query predicate invalidOverlap(
MemoryOperand useOperand, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(Overlap overlap |
overlap = useOperand.getDefinitionOverlap() and
overlap instanceof MayPartiallyOverlap and
message =
"MemoryOperand '" + useOperand.toString() + "' has a `getDefinitionOverlap()` of '" +
overlap.toString() + "'." and
irFunc = getOperandIRFunction(useOperand, irFuncText)
)
}
query predicate nonUniqueEnclosingIRFunction(
Instruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(int irFuncCount |
irFuncCount = count(instr.getEnclosingIRFunction()) and
irFuncCount != 1 and
message =
"Instruction '" + instr.toString() + "' has " + irFuncCount.toString() +
" results for `getEnclosingIRFunction()` in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
)
}
/**
* Holds if the object address operand for the given `FieldAddress` instruction does not have an
* address type.
*/
query predicate fieldAddressOnNonPointer(
FieldAddressInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
not instr.getObjectAddressOperand().getIRType() instanceof IRAddressType and
message =
"FieldAddress instruction '" + instr.toString() +
"' has an object address operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
/**
* Holds if the `this` argument operand for the given `Call` instruction does not have an address
* type.
*/
query predicate thisArgumentIsNonPointer(
CallInstruction instr, string message, OptionalIRFunction irFunc, string irFuncText
) {
exists(ThisArgumentOperand thisOperand | thisOperand = instr.getThisArgumentOperand() |
not thisOperand.getIRType() instanceof IRAddressType
) and
message =
"Call instruction '" + instr.toString() +
"' has a `this` argument operand that is not an address, in function '$@'." and
irFunc = getInstructionIRFunction(instr, irFuncText)
}
}

View File

@@ -1,32 +0,0 @@
/**
* Provides implementation classes modeling the MySql C API.
* See `semmle.code.cpp.models.Models` for usage information.
*/
private import semmle.code.cpp.models.interfaces.Sql
private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs
/**
* The `mysql_query` family of functions from the MySQL C API.
*/
private class MySqlExecutionFunction extends SqlExecutionFunction {
MySqlExecutionFunction() {
this.hasName(["mysql_query", "mysql_real_query", "mysql_real_query_nonblocking"])
}
override predicate hasSqlArgument(FunctionInput input) { input.isParameterDeref(1) }
}
/**
* The `mysql_real_escape_string` family of functions from the MySQL C API.
*/
private class MySqlBarrierFunction extends SqlBarrierFunction {
MySqlBarrierFunction() {
this.hasName(["mysql_real_escape_string", "mysql_real_escape_string_quote"])
}
override predicate barrierSqlArgument(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(2) and
output.isParameterDeref(1)
}
}

View File

@@ -1,94 +0,0 @@
private import semmle.code.cpp.models.interfaces.Sql
private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs
private predicate pqxxTransactionSqlArgument(string function, int arg) {
function = "exec" and arg = 0
or
function = "exec0" and arg = 0
or
function = "exec1" and arg = 0
or
function = "exec_n" and arg = 1
or
function = "exec_params" and arg = 0
or
function = "exec_params0" and arg = 0
or
function = "exec_params1" and arg = 0
or
function = "exec_params_n" and arg = 1
or
function = "query_value" and arg = 0
or
function = "stream" and arg = 0
}
private predicate pqxxConnectionSqlArgument(string function, int arg) {
function = "prepare" and arg = 1
}
private predicate pqxxTransationClassNames(string className, string namespace) {
namespace = "pqxx" and
className in [
"dbtransaction", "nontransaction", "basic_robusttransaction", "robusttransaction",
"subtransaction", "transaction", "basic_transaction", "transaction_base", "work"
]
}
private predicate pqxxConnectionClassNames(string className, string namespace) {
namespace = "pqxx" and
className in ["connection_base", "basic_connection", "connection"]
}
private predicate pqxxEscapeArgument(string function, int arg) {
arg = 0 and
function in ["esc", "esc_raw", "quote", "quote_raw", "quote_name", "quote_table", "esc_like"]
}
private class PostgreSqlExecutionFunction extends SqlExecutionFunction {
PostgreSqlExecutionFunction() {
exists(Class c |
this.getDeclaringType() = c and
// transaction exec and connection prepare variations
(
pqxxTransationClassNames(c.getName(), c.getNamespace().getName()) and
pqxxTransactionSqlArgument(this.getName(), _)
or
pqxxConnectionSqlArgument(this.getName(), _) and
pqxxConnectionClassNames(c.getName(), c.getNamespace().getName())
)
)
}
override predicate hasSqlArgument(FunctionInput input) {
exists(int argIndex |
pqxxTransactionSqlArgument(this.getName(), argIndex)
or
pqxxConnectionSqlArgument(this.getName(), argIndex)
|
input.isParameterDeref(argIndex)
)
}
}
private class PostgreSqlBarrierFunction extends SqlBarrierFunction {
PostgreSqlBarrierFunction() {
exists(Class c |
this.getDeclaringType() = c and
// transaction and connection escape functions
(
pqxxTransationClassNames(c.getName(), c.getNamespace().getName()) or
pqxxConnectionClassNames(c.getName(), c.getNamespace().getName())
) and
pqxxEscapeArgument(this.getName(), _)
)
}
override predicate barrierSqlArgument(FunctionInput input, FunctionOutput output) {
exists(int argIndex |
input.isParameterDeref(argIndex) and
output.isReturnValueDeref() and
pqxxEscapeArgument(this.getName(), argIndex)
)
}
}

View File

@@ -1,21 +0,0 @@
/**
* Provides implementation classes modeling the SQLite C API.
* See `semmle.code.cpp.models.Models` for usage information.
*/
private import semmle.code.cpp.models.interfaces.Sql
private import semmle.code.cpp.models.interfaces.FunctionInputsAndOutputs
/**
* The `sqlite3_exec` and `sqlite3_prepare` families of functions from the SQLite C API.
*/
private class SqLite3ExecutionFunction extends SqlExecutionFunction {
SqLite3ExecutionFunction() {
this.hasName([
"sqlite3_exec", "sqlite3_prepare", "sqlite3_prepare_v2", "sqlite3_prepare_v3",
"sqlite3_prepare16", "sqlite3_prepare16_v2", "sqlite3_prepare16_v3"
])
}
override predicate hasSqlArgument(FunctionInput input) { input.isParameterDeref(1) }
}

View File

@@ -1,30 +0,0 @@
/**
* Provides abstract classes for modeling functions that execute and escape SQL query strings.
* To extend this QL library, create a QL class extending `SqlExecutionFunction` or `SqlEscapeFunction`
* with a characteristic predicate that selects the function or set of functions you are modeling.
* Within that class, override the predicates provided by the class to match the way a
* parameter flows into the function and, in the case of `SqlEscapeFunction`, out of the function.
*/
private import cpp
/**
* An abstract class that represents a function that executes an SQL query.
*/
abstract class SqlExecutionFunction extends Function {
/**
* Holds if `input` to this function represents SQL code to be executed.
*/
abstract predicate hasSqlArgument(FunctionInput input);
}
/**
* An abstract class that represents a function that is a barrier to an SQL query string.
*/
abstract class SqlBarrierFunction extends Function {
/**
* Holds if the `output` is a barrier to the SQL input `input` such that is it safe to pass to
* an `SqlExecutionFunction`.
*/
abstract predicate barrierSqlArgument(FunctionInput input, FunctionOutput output);
}

View File

@@ -1,55 +0,0 @@
/**
* Provides classes for heuristically identifying variables and functions that
* might contain or return a password or other sensitive information.
*/
import cpp
/**
* Holds if the name `s` suggests something might contain or return a password
* or other sensitive information.
*/
bindingset[s]
private predicate suspicious(string s) {
(
s.matches("%password%") or
s.matches("%passwd%") or
s.matches("%trusted%")
) and
not (
s.matches("%hash%") or
s.matches("%crypt%") or
s.matches("%file%") or
s.matches("%path%")
)
}
/**
* A variable that might contain a password or other sensitive information.
*/
class SensitiveVariable extends Variable {
SensitiveVariable() {
suspicious(getName().toLowerCase()) and
not this.getUnspecifiedType() instanceof IntegralType
}
}
/**
* A function that might return a password or other sensitive information.
*/
class SensitiveFunction extends Function {
SensitiveFunction() {
suspicious(getName().toLowerCase()) and
not this.getUnspecifiedType() instanceof IntegralType
}
}
/**
* An expression whose value might be a password or other sensitive information.
*/
class SensitiveExpr extends Expr {
SensitiveExpr() {
this.(VariableAccess).getTarget() instanceof SensitiveVariable or
this.(FunctionCall).getTarget() instanceof SensitiveFunction
}
}

File diff suppressed because it is too large Load Diff

View File

@@ -7,6 +7,10 @@
import cpp
class AnonymousCompilation extends Compilation {
override string toString() { result = "<compilation>" }
}
string describe(Compilation c) {
if c.getArgument(1) = "--mimic"
then result = "compiler invocation " + concat(int i | i > 1 | c.getArgument(i), " " order by i)
@@ -15,4 +19,4 @@ string describe(Compilation c) {
from Compilation c
where not c.normalTermination()
select "Extraction aborted for " + describe(c)
select c, "Extraction aborted for " + describe(c), 2

View File

@@ -13,15 +13,10 @@
import cpp
from BitField fi, VariableAccess va, Type fct
from BitField fi, VariableAccess va
where
(
if va.getFullyConverted().getType() instanceof ReferenceType
then fct = va.getFullyConverted().getType().(ReferenceType).getBaseType()
else fct = va.getFullyConverted().getType()
) and
fi.getNumBits() > fct.getSize() * 8 and
va.getExplicitlyConverted().getType().getSize() > fct.getSize() and
fi.getNumBits() > va.getFullyConverted().getType().getSize() * 8 and
va.getExplicitlyConverted().getType().getSize() > va.getFullyConverted().getType().getSize() and
va.getTarget() = fi and
not fct.getUnspecifiedType() instanceof BoolType
not va.getActualType() instanceof BoolType
select va, "Implicit downcast of bitfield $@", fi, fi.toString()

View File

@@ -19,32 +19,28 @@ import cpp
* Holds if the argument corresponding to the `pos` conversion specifier
* of `ffc` is expected to have type `expected`.
*/
pragma[noopt]
private predicate formattingFunctionCallExpectedType(
FormattingFunctionCall ffc, int pos, Type expected
) {
ffc.getFormat().(FormatLiteral).getConversionType(pos) = expected
exists(FormattingFunction f, int i, FormatLiteral fl |
ffc instanceof FormattingFunctionCall and
ffc.getTarget() = f and
f.getFormatParameterIndex() = i and
ffc.getArgument(i) = fl and
fl.getConversionType(pos) = expected
)
}
/**
* Holds if the argument corresponding to the `pos` conversion specifier
* of `ffc` could alternatively have type `expected`, for example on a different
* platform.
*/
private predicate formattingFunctionCallAlternateType(
FormattingFunctionCall ffc, int pos, Type expected
) {
ffc.getFormat().(FormatLiteral).getConversionTypeAlternate(pos) = expected
}
/**
* Holds if the argument corresponding to the `pos` conversion specifier
* of `ffc` is `arg` and has type `actual`.
* of `ffc` is expected to have type `expected` and the corresponding
* argument `arg` has type `actual`.
*/
pragma[noopt]
predicate formattingFunctionCallActualType(
FormattingFunctionCall ffc, int pos, Expr arg, Type actual
) {
predicate formatArgType(FormattingFunctionCall ffc, int pos, Type expected, Expr arg, Type actual) {
exists(Expr argConverted |
formattingFunctionCallExpectedType(ffc, pos, expected) and
ffc.getConversionArgument(pos) = arg and
argConverted = arg.getFullyConverted() and
actual = argConverted.getType()
@@ -76,8 +72,7 @@ class ExpectedType extends Type {
ExpectedType() {
exists(Type t |
(
formattingFunctionCallExpectedType(_, _, t) or
formattingFunctionCallAlternateType(_, _, t) or
formatArgType(_, _, t, _, _) or
formatOtherArgType(_, _, t, _, _)
) and
this = t.getUnspecifiedType()
@@ -96,11 +91,7 @@ class ExpectedType extends Type {
*/
predicate trivialConversion(ExpectedType expected, Type actual) {
exists(Type exp, Type act |
(
formattingFunctionCallExpectedType(_, _, exp) or
formattingFunctionCallAlternateType(_, _, exp)
) and
formattingFunctionCallActualType(_, _, _, act) and
formatArgType(_, _, exp, _, act) and
expected = exp.getUnspecifiedType() and
actual = act.getUnspecifiedType()
) and
@@ -155,13 +146,9 @@ int sizeof_IntType() { exists(IntType it | result = it.getSize()) }
from FormattingFunctionCall ffc, int n, Expr arg, Type expected, Type actual
where
(
formattingFunctionCallExpectedType(ffc, n, expected) and
formattingFunctionCallActualType(ffc, n, arg, actual) and
formatArgType(ffc, n, expected, arg, actual) and
not exists(Type anyExpected |
(
formattingFunctionCallExpectedType(ffc, n, anyExpected) or
formattingFunctionCallAlternateType(ffc, n, anyExpected)
) and
formatArgType(ffc, n, anyExpected, arg, actual) and
trivialConversion(anyExpected.getUnspecifiedType(), actual.getUnspecifiedType())
)
or

View File

@@ -29,19 +29,11 @@ class ImproperNullTerminationReachability extends StackVariableReachabilityWithR
override predicate isSourceActual(ControlFlowNode node, StackVariable v) {
node = declWithNoInit(v)
or
exists(Call c, int bufferArg, int sizeArg |
exists(Call c, VariableAccess va |
c = node and
(
c.getTarget().hasName("readlink") and bufferArg = 1 and sizeArg = 2
or
c.getTarget().hasName("readlinkat") and bufferArg = 2 and sizeArg = 3
) and
c.getArgument(bufferArg).(VariableAccess).getTarget() = v and
(
// buffer size parameter likely matches the full buffer size
c.getArgument(sizeArg) instanceof SizeofOperator or
c.getArgument(sizeArg).getValue().toInt() = v.getType().getSize()
)
c.getTarget().hasName("readlink") and
c.getArgument(1) = va and
va.getTarget() = v
)
}

View File

@@ -41,7 +41,7 @@ DeclStmt declWithNoInit(LocalVariable v) {
result.getADeclaration() = v and
not exists(v.getInitializer()) and
/* The type of the variable is not stack-allocated. */
exists(Type t | t = v.getType() | not allocatedType(t))
not allocatedType(v.getType())
}
class UninitialisedLocalReachability extends StackVariableReachability {

View File

@@ -1,12 +0,0 @@
/**
* @name Sum of frontend and extractor time
* @description The sum of elapsed frontend time, and the sum of elapsed extractor time.
* This query is for internal use only and may change without notice.
* @kind table
* @id cpp/frontend-and-extractor-time
*/
import cpp
select sum(Compilation c, float seconds | compilation_time(c, _, 2, seconds) | seconds) as sum_frontend_elapsed_seconds,
sum(Compilation c, float seconds | compilation_time(c, _, 4, seconds) | seconds) as sum_extractor_elapsed_seconds

View File

@@ -9,6 +9,7 @@
* @tags reliability
* external/cwe/cwe-573
* external/cwe/cwe-252
* @opaque-id SM02344
* @microsoft.severity Important
*/

View File

@@ -30,15 +30,7 @@ class Configuration extends TaintTrackingConfiguration {
}
override predicate isBarrier(Expr e) {
super.isBarrier(e)
or
e.getUnspecifiedType() instanceof IntegralType
or
exists(SqlBarrierFunction sql, int arg, FunctionInput input |
e = sql.getACallToThisFunction().getArgument(arg) and
input.isParameterDeref(arg) and
sql.barrierSqlArgument(input, _)
)
super.isBarrier(e) or e.getUnspecifiedType() instanceof IntegralType
}
}

View File

@@ -15,122 +15,58 @@
import cpp
import semmle.code.cpp.security.Overflow
import semmle.code.cpp.security.Security
import semmle.code.cpp.security.FlowSources
import semmle.code.cpp.ir.dataflow.TaintTracking
import DataFlow::PathGraph
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
import Bounded
/**
* A function that outputs random data such as `std::rand`.
*/
abstract class RandomFunction extends Function {
/**
* Gets the `FunctionOutput` that describes how this function returns the random data.
*/
FunctionOutput getFunctionOutput() { result.isReturnValue() }
predicate isUnboundedRandCall(FunctionCall fc) {
exists(Function func | func = fc.getTarget() |
func.hasGlobalOrStdOrBslName("rand") and
not bounded(fc) and
func.getNumberOfParameters() = 0
)
}
/**
* The standard function `std::rand`.
*/
private class StdRand extends RandomFunction {
StdRand() {
this.hasGlobalOrStdOrBslName("rand") and
this.getNumberOfParameters() = 0
}
predicate isUnboundedRandCallOrParent(Expr e) {
isUnboundedRandCall(e)
or
isUnboundedRandCallOrParent(e.getAChild())
}
/**
* The Unix function `rand_r`.
*/
private class RandR extends RandomFunction {
RandR() {
this.hasGlobalName("rand_r") and
this.getNumberOfParameters() = 1
}
predicate isUnboundedRandValue(Expr e) {
isUnboundedRandCall(e)
or
exists(MacroInvocation mi |
e = mi.getExpr() and
isUnboundedRandCallOrParent(e)
)
}
/**
* The Unix function `random`.
*/
private class Random extends RandomFunction {
Random() {
this.hasGlobalName("random") and
this.getNumberOfParameters() = 1
class SecurityOptionsArith extends SecurityOptions {
override predicate isUserInput(Expr expr, string cause) {
isUnboundedRandValue(expr) and
cause = "rand"
}
}
/**
* The Windows `rand_s` function.
*/
private class RandS extends RandomFunction {
RandS() {
this.hasGlobalName("rand_s") and
this.getNumberOfParameters() = 1
}
override FunctionOutput getFunctionOutput() { result.isParameterDeref(0) }
}
predicate missingGuard(VariableAccess va, string effect) {
exists(Operation op | op.getAnOperand() = va |
// underflow - random numbers are usually non-negative, so underflow is
// only likely if the type is unsigned. Multiplication is also unlikely to
// cause underflow of a non-negative number.
missingGuardAgainstUnderflow(op, va) and
effect = "underflow" and
op.getUnspecifiedType().(IntegralType).isUnsigned() and
not op instanceof MulExpr
missingGuardAgainstUnderflow(op, va) and effect = "underflow"
or
// overflow
missingGuardAgainstOverflow(op, va) and effect = "overflow"
)
}
class UncontrolledArithConfiguration extends TaintTracking::Configuration {
UncontrolledArithConfiguration() { this = "UncontrolledArithConfiguration" }
class Configuration extends TaintTrackingConfiguration {
override predicate isSink(Element e) { missingGuard(e, _) }
override predicate isSource(DataFlow::Node source) {
exists(RandomFunction rand, Call call | call.getTarget() = rand |
rand.getFunctionOutput().isReturnValue() and
source.asExpr() = call
or
exists(int n |
source.asDefiningArgument() = call.getArgument(n) and
rand.getFunctionOutput().isParameterDeref(n)
)
)
}
override predicate isSink(DataFlow::Node sink) { missingGuard(sink.asExpr(), _) }
override predicate isSanitizer(DataFlow::Node node) {
bounded(node.asExpr())
or
// If this expression is part of bitwise 'and' or 'or' operation it's likely that the value is
// only used as a bit pattern.
node.asExpr() =
any(Operation op |
op instanceof BitwiseOrExpr or
op instanceof BitwiseAndExpr or
op instanceof ComplementExpr
).getAnOperand*()
or
// block unintended flow to pointers
node.asExpr().getUnspecifiedType() instanceof PointerType
}
override predicate isBarrier(Expr e) { super.isBarrier(e) or bounded(e) }
}
/** Gets the expression that corresponds to `node`, if any. */
Expr getExpr(DataFlow::Node node) { result = [node.asExpr(), node.asDefiningArgument()] }
from
UncontrolledArithConfiguration config, DataFlow::PathNode source, DataFlow::PathNode sink,
VariableAccess va, string effect
from Expr origin, VariableAccess va, string effect, PathNode sourceNode, PathNode sinkNode
where
config.hasFlowPath(source, sink) and
sink.getNode().asExpr() = va and
taintedWithPath(origin, va, sourceNode, sinkNode) and
missingGuard(va, effect)
select sink.getNode(), source, sink,
"$@ flows to here and is used in arithmetic, potentially causing an " + effect + ".",
getExpr(source.getNode()), "Uncontrolled value"
select va, sourceNode, sinkNode,
"$@ flows to here and is used in arithmetic, potentially causing an " + effect + ".", origin,
"Uncontrolled value"

View File

@@ -7,6 +7,25 @@ private import cpp
private import semmle.code.cpp.rangeanalysis.SimpleRangeAnalysis
private import semmle.code.cpp.rangeanalysis.RangeAnalysisUtils
/**
* An operand `e` of a division expression (i.e., `e` is an operand of either a `DivExpr` or
* a `AssignDivExpr`) is bounded when `e` is the left-hand side of the division.
*/
pragma[inline]
private predicate boundedDiv(Expr e, Expr left) { e = left }
/**
* An operand `e` of a remainder expression `rem` (i.e., `rem` is either a `RemExpr` or
* an `AssignRemExpr`) with left-hand side `left` and right-ahnd side `right` is bounded
* when `e` is `left` and `right` is upper bounded by some number that is less than the maximum integer
* allowed by the result type of `rem`.
*/
pragma[inline]
private predicate boundedRem(Expr e, Expr rem, Expr left, Expr right) {
e = left and
upperBound(right.getFullyConverted()) < exprMaxVal(rem.getFullyConverted())
}
/**
* An operand `e` of a bitwise and expression `andExpr` (i.e., `andExpr` is either an `BitwiseAndExpr`
* or an `AssignAndExpr`) with operands `operand1` and `operand2` is the operand that is not `e` is upper
@@ -31,10 +50,19 @@ predicate bounded(Expr e) {
) and
not convertedExprMightOverflow(e)
or
// Optimitically assume that a remainder expression always yields a much smaller value.
e = any(RemExpr rem).getLeftOperand()
// For `%` and `&` we require that `e` is bounded by a value that is strictly smaller than the
// maximum possible value of the result type of the operation.
// For example, the function call `rand()` is considered bounded in the following program:
// ```
// int i = rand() % (UINT8_MAX + 1);
// ```
// but not in:
// ```
// unsigned char uc = rand() % (UINT8_MAX + 1);
// ```
exists(RemExpr rem | boundedRem(e, rem, rem.getLeftOperand(), rem.getRightOperand()))
or
e = any(AssignRemExpr rem).getLValue()
exists(AssignRemExpr rem | boundedRem(e, rem, rem.getLValue(), rem.getRValue()))
or
exists(BitwiseAndExpr andExpr |
boundedBitwiseAnd(e, andExpr, andExpr.getAnOperand(), andExpr.getAnOperand())
@@ -45,11 +73,11 @@ predicate bounded(Expr e) {
)
or
// Optimitically assume that a division always yields a much smaller value.
e = any(DivExpr div).getLeftOperand()
boundedDiv(e, any(DivExpr div).getLeftOperand())
or
e = any(AssignDivExpr div).getLValue()
boundedDiv(e, any(AssignDivExpr div).getLValue())
or
e = any(RShiftExpr shift).getLeftOperand()
boundedDiv(e, any(RShiftExpr shift).getLeftOperand())
or
e = any(AssignRShiftExpr div).getLValue()
boundedDiv(e, any(AssignRShiftExpr div).getLValue())
}

View File

@@ -5,7 +5,7 @@
* @kind problem
* @problem.severity warning
* @security-severity 7.5
* @precision high
* @precision medium
* @id cpp/cleartext-storage-file
* @tags security
* external/cwe/cwe-313
@@ -14,40 +14,10 @@
import cpp
import semmle.code.cpp.security.SensitiveExprs
import semmle.code.cpp.security.FileWrite
import semmle.code.cpp.dataflow.DataFlow
import semmle.code.cpp.valuenumbering.GlobalValueNumbering
/**
* An operation on a filename.
*/
predicate filenameOperation(FunctionCall op, Expr path) {
exists(string name | name = op.getTarget().getName() |
name =
[
"remove", "unlink", "rmdir", "rename", "fopen", "open", "freopen", "_open", "_wopen",
"_wfopen", "_fsopen", "_wfsopen", "chmod", "chown", "stat", "lstat", "fstat", "access",
"_access", "_waccess", "_access_s", "_waccess_s"
] and
path = op.getArgument(0)
or
name = ["fopen_s", "wfopen_s", "rename"] and
path = op.getArgument(1)
)
}
predicate isFileName(GVN gvn) {
exists(FunctionCall op, Expr path |
filenameOperation(op, path) and
gvn = globalValueNumber(path)
)
}
from FileWrite w, SensitiveExpr source, Expr mid, Expr dest
from FileWrite w, SensitiveExpr source, Expr dest
where
DataFlow::localFlow(DataFlow::exprNode(source), DataFlow::exprNode(mid)) and
mid = w.getASource() and
dest = w.getDest() and
not isFileName(globalValueNumber(source)) and // file names are not passwords
not exists(string convChar | convChar = w.getSourceConvChar(mid) | not convChar = ["s", "S"]) // ignore things written with other conversion characters
source = w.getASource() and
dest = w.getDest()
select w, "This write into file '" + dest.toString() + "' may contain unencrypted data from $@",
source, "this source."

View File

@@ -6,7 +6,7 @@
* @kind problem
* @problem.severity warning
* @security-severity 7.7
* @precision high
* @precision medium
* @id cpp/toctou-race-condition
* @tags security
* external/cwe/cwe-367
@@ -16,60 +16,59 @@ import cpp
import semmle.code.cpp.controlflow.Guards
/**
* An operation on a filename that is likely to modify the corresponding file
* and may return an indication of success.
* An operation on a filename.
*
* Note: we're not interested in operations where the file is specified by a
* descriptor, rather than a filename, as they are better behaved. We are
* interested in functions that take a filename and return a file descriptor,
* however.
* Note: we're not interested in operations on file descriptors, as they
* are better behaved.
*/
FunctionCall filenameOperation(Expr path) {
exists(string name | name = result.getTarget().getName() |
name =
[
"remove", "unlink", "rmdir", "rename", "fopen", "open", "freopen", "_open", "_wopen",
"_wfopen", "_fsopen", "_wfsopen"
] and
(
name = "remove" or
name = "unlink" or
name = "rmdir" or
name = "rename" or
name = "chmod" or
name = "chown" or
name = "fopen" or
name = "open" or
name = "freopen" or
name = "_open" or
name = "_wopen" or
name = "_wfopen"
) and
result.getArgument(0) = path
or
name = ["fopen_s", "wfopen_s", "rename"] and
(
name = "fopen_s" or
name = "wfopen_s"
) and
result.getArgument(1) = path
)
or
result = sensitiveFilenameOperation(path)
}
/**
* An operation on a filename that is likely to modify the security properties
* of the corresponding file and may return an indication of success.
*/
FunctionCall sensitiveFilenameOperation(Expr path) {
exists(string name | name = result.getTarget().getName() |
name = ["chmod", "chown"] and
result.getArgument(0) = path
)
}
/**
* An operation on a filename that returns information in the return value but
* does not modify the corresponding file. For example, `access`.
* A use of `access` (or similar) on a filename.
*/
FunctionCall accessCheck(Expr path) {
exists(string name | name = result.getTarget().getName() |
name = ["access", "_access", "_waccess", "_access_s", "_waccess_s"]
name = "access" or
name = "_access" or
name = "_waccess" or
name = "_access_s" or
name = "_waccess_s"
) and
path = result.getArgument(0)
}
/**
* An operation on a filename that returns information via a pointer argument
* and any return value, but does not modify the corresponding file. For
* example, `stat`.
* A use of `stat` (or similar) on a filename.
*/
FunctionCall stat(Expr path, Expr buf) {
exists(string name | name = result.getTarget().getName() |
name = ["stat", "lstat", "fstat"] or
name = "stat" or
name = "lstat" or
name = "fstat" or
name.matches("\\_stat%") or
name.matches("\\_wstat%")
) and
@@ -78,7 +77,7 @@ FunctionCall stat(Expr path, Expr buf) {
}
/**
* Holds if `use` refers to `source`, either by being the same or by
* Holds if `use` points to `source`, either by being the same or by
* one step of variable indirection.
*/
predicate referenceTo(Expr source, Expr use) {
@@ -89,45 +88,36 @@ predicate referenceTo(Expr source, Expr use) {
)
}
from Expr check, Expr checkPath, FunctionCall use, Expr usePath
from FunctionCall fc, Expr check, Expr checkUse, Expr opUse
where
// `check` looks like a check on a filename
// checkUse looks like a check on a filename
(
(
// either:
// an access check
check = accessCheck(checkPath)
or
// a stat
check = stat(checkPath, _)
or
// access to a member variable on the stat buf
// (morally, this should be a use-use pair, but it seems unlikely
// that this variable will get reused in practice)
exists(Expr call, Expr e, Variable v |
call = stat(checkPath, e) and
e.getAChild*().(VariableAccess).getTarget() = v and
check.(VariableAccess).getTarget() = v and
not e.getAChild*() = check // the call that writes to the pointer is not where the pointer is checked.
)
) and
// `op` looks like an operation on a filename
use = filenameOperation(usePath)
// either:
// an access check
check = accessCheck(checkUse)
or
// a stat
check = stat(checkUse, _)
or
// another filename operation (null pointers can indicate errors)
check = filenameOperation(checkPath) and
// `op` looks like a sensitive operation on a filename
use = sensitiveFilenameOperation(usePath)
check = filenameOperation(checkUse)
or
// access to a member variable on the stat buf
// (morally, this should be a use-use pair, but it seems unlikely
// that this variable will get reused in practice)
exists(Variable buf | exists(stat(checkUse, buf.getAnAccess())) |
check.(VariableAccess).getQualifier() = buf.getAnAccess()
)
) and
// `checkPath` and `usePath` refer to the same SSA variable
exists(SsaDefinition def, StackVariable v |
def.getAUse(v) = checkPath and def.getAUse(v) = usePath
) and
// the return value of `check` is used (possibly with one step of
// variable indirection) in a guard which controls `use`
// checkUse and opUse refer to the same SSA variable
exists(SsaDefinition def, StackVariable v | def.getAUse(v) = checkUse and def.getAUse(v) = opUse) and
// opUse looks like an operation on a filename
fc = filenameOperation(opUse) and
// the return value of check is used (possibly with one step of
// variable indirection) in a guard which controls fc
exists(GuardCondition guard | referenceTo(check, guard.getAChild*()) |
guard.controls(use.(ControlFlowNode).getBasicBlock(), _)
guard.controls(fc.(ControlFlowNode).getBasicBlock(), _)
)
select use,
select fc,
"The $@ being operated upon was previously $@, but the underlying file may have been changed since then.",
usePath, "filename", check, "checked"
opUse, "filename", check, "checked"

View File

@@ -6,6 +6,7 @@
* @kind problem
* @problem.severity warning
* @security-severity 7.8
* @opaque-id SM02313
* @id cpp/conditionally-uninitialized-variable
* @tags security
* external/cwe/cwe-457

View File

@@ -182,7 +182,7 @@ class ThrowingAllocator extends Function {
// 3. the allocator isn't marked with `throw()` or `noexcept`.
not exists(this.getBlock()) and
not exists(Parameter p | p = this.getAParameter() |
p.getUnspecifiedType().stripType() instanceof NoThrowType
p.getUnspecifiedType() instanceof NoThrowType
) and
not this.isNoExcept() and
not this.isNoThrow()

View File

@@ -26,7 +26,7 @@ can use their own storage.</p>
<p>Similarly replace calls to <code>localtime</code> with
<code>localtime_r</code>, calls to <code>ctime</code> with
<code>ctime_r</code> and calls to <code>asctime</code> with
<code>asctime_r</code> (if those functions exist on your platform).</p>
<code>asctime_r</code>.</p>
</recommendation>
<example>

View File

@@ -4,7 +4,7 @@
* @kind problem
* @problem.severity warning
* @security-severity 10.0
* @precision medium
* @precision high
* @id cpp/potentially-dangerous-function
* @tags reliability
* security

View File

@@ -4,6 +4,7 @@
* @description The total number of lines of C/C++ code across all files, including system headers, libraries, and auto-generated files. This is a useful metric of the size of a database. For all files that were seen during the build, this query counts the lines of code, excluding whitespace or comments.
* @kind metric
* @tags summary
* lines-of-code
*/
import cpp

View File

@@ -1,6 +1,6 @@
- description: Standard Code Scanning queries for C and C++
- queries: .
- qlpack: codeql-cpp
- apply: code-scanning-selectors.yml
from: codeql/suite-helpers
from: codeql-suite-helpers
- apply: codeql-suites/exclude-slow-queries.yml
from: codeql/cpp-queries
from: codeql-cpp

View File

@@ -1,9 +1,9 @@
- description: Standard LGTM queries for C/C++, including ones not displayed by default
- queries: .
- qlpack: codeql-cpp
- apply: lgtm-selectors.yml
from: codeql/suite-helpers
from: codeql-suite-helpers
- apply: codeql-suites/exclude-slow-queries.yml
from: codeql/cpp-queries
from: codeql-cpp
# These are only for IDE use.
- exclude:
tags contain:

View File

@@ -1,4 +1,4 @@
- description: Standard LGTM queries for C/C++
- apply: codeql-suites/cpp-lgtm-full.qls
- apply: lgtm-displayed-only.yml
from: codeql/suite-helpers
from: codeql-suite-helpers

View File

@@ -1,6 +1,6 @@
- description: Security-and-quality queries for C and C++
- queries: .
- qlpack: codeql-cpp
- apply: security-and-quality-selectors.yml
from: codeql/suite-helpers
from: codeql-suite-helpers
- apply: codeql-suites/exclude-slow-queries.yml
from: codeql/cpp-queries
from: codeql-cpp

View File

@@ -1,6 +1,6 @@
- description: Security-extended queries for C and C++
- queries: .
- qlpack: codeql-cpp
- apply: security-extended-selectors.yml
from: codeql/suite-helpers
from: codeql-suite-helpers
- apply: codeql-suites/exclude-slow-queries.yml
from: codeql/cpp-queries
from: codeql-cpp

View File

@@ -0,0 +1,28 @@
#include <iostream>
#include <stdexcept>
#include <pqxx/pqxx>
int main(int argc, char ** argv) {
if (argc != 2) {
throw std::runtime_error("Give me a string!");
}
pqxx::connection c;
pqxx::work w(c);
// BAD
char *userName = argv[1];
char query1[1000] = {0};
sprintf(query1, "SELECT UID FROM USERS where name = \"%s\"", userName);
pqxx::row r = w.exec1(query1);
w.commit();
std::cout << r[0].as<int>() << std::endl;
// GOOD
pqxx::result r2 = w.exec("SELECT " + w.quote(argv[1]));
w.commit();
std::cout << r2[0][0].c_str() << std::endl;
return 0;
}

View File

@@ -0,0 +1,31 @@
<!DOCTYPE qhelp PUBLIC
"-//Semmle//qhelp//EN"
"qhelp.dtd">
<qhelp>
<overview>
<p>The code passes user input as part of a SQL query without escaping special elements.
It generates a SQL query to Postgres using <code>sprintf</code>,
with the user-supplied data directly passed as an argument
to <code>sprintf</code>. This leaves the code vulnerable to attack by SQL Injection.</p>
</overview>
<recommendation>
<p>Use a library routine to escape characters in the user-supplied
string before converting it to SQL. Use <code>esc</code> and <code>quote</code> pqxx library functions.</p>
</recommendation>
<example>
<sample src="SqlPqxxTainted.cpp" />
</example>
<references>
<li>MSDN Library: <a href="https://docs.microsoft.com/en-us/sql/relational-databases/security/sql-injection">SQL Injection</a>.</li>
<!-- LocalWords: SQL CWE
-->
</references>
</qhelp>

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