Compare commits

..

5 Commits

Author SHA1 Message Date
Michael B. Gale
059d33f033 Fix typo in documentation
Co-authored-by: hubwriter <hubwriter@github.com>
2022-11-28 12:07:15 +00:00
Michael B. Gale
b5b23f12a5 Add change note 2022-11-25 11:07:32 +00:00
Michael B. Gale
5620d247fd Fix query not working for anonymous functions 2022-11-24 13:51:34 +00:00
Michael B. Gale
c907a9e707 Remove unnecessary cast 2022-11-24 13:21:36 +00:00
Michael B. Gale
a71245abc2 Add query to check for deferred errors 2022-11-24 11:01:48 +00:00
2683 changed files with 117243 additions and 165685 deletions

View File

@@ -10,5 +10,5 @@ assignees: ''
**Description of the issue**
<!-- Please explain briefly what is the problem.
If it is about a GitHub project, please include its URL. -->
If it is about an LGTM project, please include its URL.-->

View File

@@ -1,5 +1,5 @@
---
name: CodeQL false positive
name: CodeQL False positive
about: Report CodeQL alerts that you think should not have been detected (not applicable, not exploitable, etc.)
title: False positive
labels: false-positive

View File

@@ -14,7 +14,8 @@ outputs:
runs:
using: composite
steps:
# calculate the merge-base with main, in a way that works both on PRs and pushes to main.
# Cache the query compilation caches.
# calculate the merge-base with main, in a way that works both on PRs and pushes to main.
- name: Calculate merge-base
shell: bash
if: ${{ github.event_name == 'pull_request' }}
@@ -23,32 +24,38 @@ runs:
run: |
MERGE_BASE=$(git cat-file commit $GITHUB_SHA | grep '^parent ' | head -1 | cut -f 2 -d " ")
echo "merge_base=$MERGE_BASE" >> $GITHUB_ENV
- name: Restore cache (PR)
- name: Read CodeQL query compilation - PR
if: ${{ github.event_name == 'pull_request' }}
uses: actions/cache/restore@v3
uses: erik-krogh/actions-cache@a88d0603fe5fb5606db9f002dfcadeb32b5f84c6
with:
path: '**/.cache'
key: codeql-compile-${{ inputs.key }}-pr-${{ github.sha }}
read-only: true
key: codeql-compile-${{ inputs.key }}-pr-${{ github.sha }} # deliberately not using the `compile-compile-main` keys here.
restore-keys: |
codeql-compile-${{ inputs.key }}-${{ github.base_ref }}-${{ env.merge_base }}
codeql-compile-${{ inputs.key }}-${{ github.base_ref }}-
codeql-compile-${{ inputs.key }}-main-
- name: Fill cache (only branch push)
- name: Fill CodeQL query compilation cache - main
if: ${{ github.event_name != 'pull_request' }}
uses: actions/cache@v3
uses: erik-krogh/actions-cache@a88d0603fe5fb5606db9f002dfcadeb32b5f84c6
with:
path: '**/.cache'
key: codeql-compile-${{ inputs.key }}-${{ github.ref_name }}-${{ github.sha }} # just fill on main
restore-keys: | # restore the latest cache if the exact cache is unavailable, to speed up compilation.
codeql-compile-${{ inputs.key }}-${{ github.ref_name }}-
restore-keys: | # restore from another random commit, to speed up compilation.
codeql-compile-${{ inputs.key }}-${{ github.ref_name }}-
codeql-compile-${{ inputs.key }}-main-
- name: Fill compilation cache directory
id: fill-compilation-dir
shell: bash
shell: bash
run: |
# Move all the existing cache into another folder, so we only preserve the cache for the current queries.
node $GITHUB_WORKSPACE/.github/actions/cache-query-compilation/move-caches.js ${COMBINED_CACHE_DIR}
mkdir -p ${COMBINED_CACHE_DIR}
rm -f **/.cache/{lock,size} # -f to avoid errors if the cache is empty.
# copy the contents of the .cache folders into the combined cache folder.
cp -r **/.cache/* ${COMBINED_CACHE_DIR}/ || : # ignore missing files
# clean up the .cache folders
rm -rf **/.cache/*
echo "compdir=${COMBINED_CACHE_DIR}" >> $GITHUB_OUTPUT
env:
COMBINED_CACHE_DIR: ${{ runner.temp }}/compilation-dir
env:
COMBINED_CACHE_DIR: ${{ github.workspace }}/compilation-dir

View File

@@ -1,75 +0,0 @@
// # Move all the existing cache into another folder, so we only preserve the cache for the current queries.
// mkdir -p ${COMBINED_CACHE_DIR}
// rm -f **/.cache/{lock,size} # -f to avoid errors if the cache is empty.
// # copy the contents of the .cache folders into the combined cache folder.
// cp -r **/.cache/* ${COMBINED_CACHE_DIR}/ || : # ignore missing files
// # clean up the .cache folders
// rm -rf **/.cache/*
const fs = require("fs");
const path = require("path");
// the first argv is the cache folder to create.
const COMBINED_CACHE_DIR = process.argv[2];
function* walkCaches(dir) {
const files = fs.readdirSync(dir, { withFileTypes: true });
for (const file of files) {
if (file.isDirectory()) {
const filePath = path.join(dir, file.name);
yield* walkCaches(filePath);
if (file.name === ".cache") {
yield filePath;
}
}
}
}
async function copyDir(src, dest) {
for await (const file of await fs.promises.readdir(src, { withFileTypes: true })) {
const srcPath = path.join(src, file.name);
const destPath = path.join(dest, file.name);
if (file.isDirectory()) {
if (!fs.existsSync(destPath)) {
fs.mkdirSync(destPath);
}
await copyDir(srcPath, destPath);
} else {
await fs.promises.copyFile(srcPath, destPath);
}
}
}
async function main() {
const cacheDirs = [...walkCaches(".")];
for (const dir of cacheDirs) {
console.log(`Found .cache dir at ${dir}`);
}
// mkdir -p ${COMBINED_CACHE_DIR}
fs.mkdirSync(COMBINED_CACHE_DIR, { recursive: true });
// rm -f **/.cache/{lock,size} # -f to avoid errors if the cache is empty.
await Promise.all(
cacheDirs.map((cacheDir) =>
(async function () {
await fs.promises.rm(path.join(cacheDir, "lock"), { force: true });
await fs.promises.rm(path.join(cacheDir, "size"), { force: true });
})()
)
);
// # copy the contents of the .cache folders into the combined cache folder.
// cp -r **/.cache/* ${COMBINED_CACHE_DIR}/ || : # ignore missing files
await Promise.all(
cacheDirs.map((cacheDir) => copyDir(cacheDir, COMBINED_CACHE_DIR))
);
// # clean up the .cache folders
// rm -rf **/.cache/*
await Promise.all(
cacheDirs.map((cacheDir) => fs.promises.rm(cacheDir, { recursive: true }))
);
}
main();

View File

@@ -1,26 +0,0 @@
name: Find Latest CodeQL Bundle
description: Finds the URL of the latest released version of the CodeQL bundle.
outputs:
url:
description: The download URL of the latest CodeQL bundle release
value: ${{ steps.find-latest.outputs.url }}
runs:
using: composite
steps:
- name: Find Latest Release
id: find-latest
shell: pwsh
run: |
$Latest = gh release list --repo github/codeql-action --exclude-drafts --limit 1000 |
ForEach-Object { $C = $_ -split "`t"; return @{ type = $C[1]; tag = $C[2]; } } |
Where-Object { $_.type -eq 'Latest' }
$Tag = $Latest.tag
if ($Tag -eq '') {
throw 'Failed to find latest bundle release.'
}
Write-Output "Latest bundle tag is '${Tag}'."
"url=https://github.com/github/codeql-action/releases/download/${Tag}/codeql-bundle-linux64.tar.gz" >> $env:GITHUB_OUTPUT
env:
GITHUB_TOKEN: ${{ github.token }}

View File

@@ -13,7 +13,7 @@ on:
jobs:
atm-check-query-suite:
runs-on: ubuntu-latest-xl
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
@@ -23,12 +23,6 @@ jobs:
with:
channel: release
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
with:
key: atm-suite
- name: Install ATM model
run: |
set -exu
@@ -56,13 +50,10 @@ jobs:
echo "SARIF_PATH=${SARIF_PATH}" >> "${GITHUB_ENV}"
codeql database analyze \
--threads=0 \
--ram 50000 \
--format sarif-latest \
--output "${SARIF_PATH}" \
--sarif-group-rules-by-pack \
-vv \
--compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" \
-- \
"${DB_PATH}" \
"${QUERY_PACK}/${QUERY_SUITE}"

View File

@@ -1,21 +0,0 @@
name: Check query IDs
on:
pull_request:
paths:
- "**/src/**/*.ql"
- misc/scripts/check-query-ids.py
- .github/workflows/check-query-ids.yml
branches:
- main
- "rc/*"
workflow_dispatch:
jobs:
check:
name: Check query IDs
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Check for duplicate query IDs
run: python3 misc/scripts/check-query-ids.py

View File

@@ -12,7 +12,7 @@ jobs:
runs-on: ubuntu-latest
steps:
- uses: actions/stale@v7
- uses: actions/stale@v6
with:
repo-token: ${{ secrets.GITHUB_TOKEN }}
stale-issue-message: 'This issue is stale because it has been open 14 days with no activity. Comment or remove the `Stale` label in order to avoid having this issue closed in 7 days.'

View File

@@ -35,3 +35,5 @@ jobs:
if : ${{ github.event_name != 'pull_request' }}
shell: bash
run: codeql query compile -j0 */ql/{src,examples} --keep-going --warnings=error --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
env:
COMBINED_CACHE_DIR: ${{ github.workspace }}/compilation-dir

View File

@@ -67,7 +67,7 @@ jobs:
mv "$CODEQL_PATH/csharp/tools/extractor-asp.jar" "${{ github.workspace }}/csharp/extractor-pack/tools"
# Safe guard against using the bundled extractor
rm -rf "$CODEQL_PATH/csharp"
codeql test run --threads=0 --ram 50000 --slice ${{ matrix.slice }} --search-path "${{ github.workspace }}/csharp/extractor-pack" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
codeql test run --threads=0 --ram 52000 --slice ${{ matrix.slice }} --search-path "${{ github.workspace }}/csharp/extractor-pack" --check-databases --check-undefined-labels --check-repeated-labels --check-redefined-labels --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
env:
GITHUB_TOKEN: ${{ github.token }}
unit-tests:

View File

@@ -1,80 +0,0 @@
name: "Go: Run Tests - Other OS"
on:
pull_request:
paths:
- "go/**"
- "!go/ql/**" # don't run other-os if only ql/ files changed
- .github/workflows/go-tests-other-os.yml
- .github/actions/**
- codeql-workspace.yml
jobs:
test-mac:
name: Test MacOS
runs-on: macos-latest
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
make
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
with:
key: go-qltest
- name: Test
run: |
cd go
make test cache="${{ steps.query-cache.outputs.cache-dir }}"
test-win:
name: Test Windows
runs-on: windows-latest-xl
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
make
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
with:
key: go-qltest
- name: Test
run: |
cd go
make test cache="${{ steps.query-cache.outputs.cache-dir }}"

View File

@@ -1,24 +1,15 @@
name: "Go: Run Tests"
on:
push:
paths:
- "go/**"
- .github/workflows/go-tests.yml
- .github/actions/**
- codeql-workspace.yml
branches:
- main
- "rc/*"
pull_request:
paths:
- "go/**"
- .github/workflows/go-tests.yml
- .github/actions/**
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
jobs:
test-linux:
name: Test Linux (Ubuntu)
runs-on: ubuntu-latest-xl
runs-on: ubuntu-latest
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
@@ -41,7 +32,7 @@ jobs:
cd go
make
- name: Check that all Go code is autoformatted
- name: Check that all QL and Go code is autoformatted
run: |
cd go
make check-formatting
@@ -57,13 +48,67 @@ jobs:
name: qhelp-markdown
path: go/qhelp-out/**/*.md
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
- name: Test
run: |
cd go
make test
test-mac:
name: Test MacOS
runs-on: macos-latest
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
key: go-qltest
go-version: 1.19
id: go
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
make
- name: Test
run: |
cd go
make test cache="${{ steps.query-cache.outputs.cache-dir }}"
make test
test-win:
name: Test Windows
runs-on: windows-2019
steps:
- name: Set up Go 1.19
uses: actions/setup-go@v3
with:
go-version: 1.19
id: go
- name: Check out code
uses: actions/checkout@v2
- name: Set up CodeQL CLI
uses: ./.github/actions/fetch-codeql
- name: Enable problem matchers in repository
shell: bash
run: 'find .github/problem-matchers -name \*.json -exec echo "::add-matcher::{}" \;'
- name: Build
run: |
cd go
make
- name: Test
run: |
cd go
make test

View File

@@ -23,9 +23,9 @@ defaults:
working-directory: javascript/ql/experimental/adaptivethreatmodeling
jobs:
qltest:
name: Test QL
runs-on: ubuntu-latest-xl
qlcompile:
name: Check QL compilation
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
@@ -33,33 +33,36 @@ jobs:
- name: Install pack dependencies
run: |
for pack in modelbuilding src test; do
for pack in modelbuilding src; do
codeql pack install --mode verify -- "${pack}"
done
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
with:
key: js-ml-test
- name: Check QL compilation
run: |
codeql query compile \
--check-only \
--ram 50000 \
--ram 5120 \
--additional-packs "${{ github.workspace }}" \
--threads=0 \
--compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" \
-- \
lib modelbuilding src
qltest:
name: Run QL tests
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: ./.github/actions/fetch-codeql
- name: Install pack dependencies
run: codeql pack install -- test
- name: Run QL tests
run: |
codeql test run \
--threads=0 \
--ram 50000 \
--ram 5120 \
--additional-packs "${{ github.workspace }}" \
--compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" \
-- \
test
test

View File

@@ -61,8 +61,8 @@ jobs:
DATABASE=$2
cd codeql-$QL_VARIANT
SHORTNAME=`basename $DATABASE`
python java/ql/src/utils/model-generator/GenerateFlowModel.py --with-summaries --with-sinks $DATABASE ${SHORTNAME}.temp.model.yml
mv java/ql/lib/ext/generated/${SHORTNAME}.temp.model.yml $MODELS/${SHORTNAME}Generated_${QL_VARIANT}.model.yml
python java/ql/src/utils/model-generator/GenerateFlowModel.py --with-summaries --with-sinks $DATABASE $MODELS/${SHORTNAME}.qll
mv $MODELS/${SHORTNAME}.qll $MODELS/${SHORTNAME}Generated_${QL_VARIANT}.qll
cd ..
}
@@ -85,16 +85,16 @@ jobs:
set -x
MODELS=`pwd`/tmp-models
ls -1 tmp-models/
for m in $MODELS/*_main.model.yml ; do
for m in $MODELS/*_main.qll ; do
t="${m/main/"pr"}"
basename=`basename $m`
name="diff_${basename/_main.model.yml/""}"
name="diff_${basename/_main.qll/""}"
(diff -w -u $m $t | diff2html -i stdin -F $MODELS/$name.html) || true
done
- uses: actions/upload-artifact@v3
with:
name: models
path: tmp-models/*.model.yml
path: tmp-models/*.qll
retention-days: 20
- uses: actions/upload-artifact@v3
with:

View File

@@ -53,7 +53,7 @@ jobs:
java/ql/src/utils/model-generator/RegenerateModels.py "${SLUG}" dbs/${SHORTNAME}
- name: Stage changes
run: |
find java -name "*.model.yml" -print0 | xargs -0 git add
find java -name "*.qll" -print0 | xargs -0 git add
git status
git diff --cached > models.patch
- uses: actions/upload-artifact@v3

View File

@@ -22,15 +22,11 @@ jobs:
steps:
### Build the queries ###
- uses: actions/checkout@v3
- name: Find latest bundle
id: find-latest-bundle
uses: ./.github/actions/find-latest-bundle
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@77a8d2d10c0b403a8b4aadbd223dc489ecd22683
with:
languages: javascript # does not matter
tools: ${{ steps.find-latest-bundle.outputs.url }}
- name: Get CodeQL version
id: get-codeql-version
run: |
@@ -142,7 +138,6 @@ jobs:
languages: ql
db-location: ${{ runner.temp }}/db
config-file: ./ql-for-ql-config.yml
tools: ${{ steps.find-latest-bundle.outputs.url }}
- name: Move pack cache
run: |
cp -r ${PACK}/.cache ql/ql/src/.cache

View File

@@ -48,19 +48,7 @@ jobs:
run: |
brew install gnu-tar
echo "/usr/local/opt/gnu-tar/libexec/gnubin" >> $GITHUB_PATH
- name: Cache entire extractor
uses: actions/cache@v3
id: cache-extractor
with:
path: |
ruby/target/release/ruby-autobuilder
ruby/target/release/ruby-autobuilder.exe
ruby/target/release/ruby-extractor
ruby/target/release/ruby-extractor.exe
ruby/ql/lib/codeql/ruby/ast/internal/TreeSitter.qll
key: ${{ runner.os }}-ruby-extractor-${{ hashFiles('ruby/rust-toolchain.toml', 'ruby/**/Cargo.lock') }}--${{ hashFiles('ruby/**/*.rs') }}
- uses: actions/cache@v3
if: steps.cache-extractor.outputs.cache-hit != 'true'
with:
path: |
~/.cargo/registry
@@ -68,19 +56,15 @@ jobs:
ruby/target
key: ${{ runner.os }}-ruby-rust-cargo-${{ hashFiles('ruby/rust-toolchain.toml', 'ruby/**/Cargo.lock') }}
- name: Check formatting
if: steps.cache-extractor.outputs.cache-hit != 'true'
run: cargo fmt --all -- --check
- name: Build
if: steps.cache-extractor.outputs.cache-hit != 'true'
run: cargo build --verbose
- name: Run tests
if: steps.cache-extractor.outputs.cache-hit != 'true'
run: cargo test --verbose
- name: Release build
if: steps.cache-extractor.outputs.cache-hit != 'true'
run: cargo build --release
- name: Generate dbscheme
if: ${{ matrix.os == 'ubuntu-latest' && steps.cache-extractor.outputs.cache-hit != 'true'}}
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@v3
if: ${{ matrix.os == 'ubuntu-latest' }}
@@ -115,10 +99,9 @@ jobs:
- name: Build Query Pack
run: |
rm -rf target/packs
codeql pack create ../shared/ssa --output target/packs
codeql pack create ../misc/suite-helpers --output target/packs
codeql pack create ../shared/regex --output target/packs
codeql pack create ../shared/ssa --output target/packs
codeql pack create ../shared/tutorial --output target/packs
codeql pack create ql/lib --output target/packs
codeql pack create -j0 ql/src --output target/packs --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
PACK_FOLDER=$(readlink -f target/packs/codeql/ruby-queries/*)

View File

@@ -62,6 +62,6 @@ jobs:
key: ruby-qltest
- name: Run QL tests
run: |
codeql test run --threads=0 --ram 50000 --search-path "${{ github.workspace }}/ruby/extractor-pack" --check-databases --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
codeql test run --threads=0 --ram 52000 --search-path "${{ github.workspace }}/ruby/extractor-pack" --check-databases --check-undefined-labels --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --consistency-queries ql/consistency-queries ql/test --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
env:
GITHUB_TOKEN: ${{ github.token }}

View File

@@ -7,77 +7,96 @@ on:
- "misc/bazel/**"
- "*.bazel*"
- .github/workflows/swift.yml
- .github/actions/**
- .github/actions/fetch-codeql/action.yml
- codeql-workspace.yml
- .pre-commit-config.yaml
- "!**/*.md"
- "!**/*.qhelp"
branches:
- main
- rc/*
push:
paths:
- "swift/**"
- "misc/bazel/**"
- "*.bazel*"
- .github/workflows/swift.yml
- .github/actions/**
- codeql-workspace.yml
- "!**/*.md"
- "!**/*.qhelp"
branches:
- main
- rc/*
jobs:
changes:
runs-on: ubuntu-latest
outputs:
codegen: ${{ steps.filter.outputs.codegen }}
ql: ${{ steps.filter.outputs.ql }}
steps:
- uses: dorny/paths-filter@4512585405083f25c027a35db413c2b3b9006d50
id: filter
with:
filters: |
codegen:
- '.github/workflows/swift.yml'
- "misc/bazel/**"
- "*.bazel*"
- 'swift/actions/setup-env/**'
- '.pre-commit-config.yaml'
- 'swift/codegen/**'
- 'swift/schema.py'
- 'swift/**/*.dbscheme'
- 'swift/ql/lib/codeql/swift/elements.qll'
- 'swift/ql/lib/codeql/swift/elements/**'
- 'swift/ql/lib/codeql/swift/generated/**'
- 'swift/ql/test/extractor-tests/generated/**'
- 'swift/ql/.generated.list'
ql:
- 'github/workflows/swift.yml'
- 'swift/**/*.ql'
- 'swift/**/*.qll'
# not using a matrix as you cannot depend on a specific job in a matrix, and we want to start linux checks
# without waiting for the macOS build
build-and-test-macos:
runs-on: macos-12-xl
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/build-and-test
- uses: ./swift/actions/create-extractor-pack
- uses: ./swift/actions/run-quick-tests
- uses: ./swift/actions/print-unextracted
build-and-test-linux:
runs-on: ubuntu-latest-xl
runs-on: ubuntu-20.04
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/build-and-test
- uses: ./swift/actions/create-extractor-pack
- uses: ./swift/actions/run-quick-tests
- uses: ./swift/actions/print-unextracted
qltests-linux:
needs: build-and-test-linux
runs-on: ubuntu-latest-xl
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/run-ql-tests
qltests-macos:
if : ${{ github.event_name == 'pull_request' }}
needs: build-and-test-macos
runs-on: macos-12-xl
strategy:
fail-fast: false
matrix:
slice: ["1/2", "2/2"]
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/run-ql-tests
with:
flags: --slice ${{ matrix.slice }}
integration-tests-linux:
needs: build-and-test-linux
runs-on: ubuntu-latest-xl
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/run-integration-tests
integration-tests-macos:
if : ${{ github.event_name == 'pull_request' }}
needs: build-and-test-macos
runs-on: macos-12-xl
timeout-minutes: 60
steps:
- uses: actions/checkout@v3
- uses: ./swift/actions/run-integration-tests
codegen:
if : ${{ github.event_name == 'pull_request' }}
runs-on: ubuntu-latest
needs: changes
if: ${{ needs.changes.outputs.codegen == 'true' }}
steps:
- uses: actions/checkout@v3
- uses: bazelbuild/setup-bazelisk@v2
- uses: actions/setup-python@v4
with:
python-version-file: 'swift/.python-version'
- uses: ./swift/actions/setup-env
- uses: pre-commit/action@v3.0.0
name: Check that python code is properly formatted
with:
@@ -95,7 +114,6 @@ jobs:
name: swift-generated-cpp-files
path: generated-cpp-files/**
database-upgrade-scripts:
if : ${{ github.event_name == 'pull_request' }}
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3

View File

@@ -12,6 +12,13 @@
# ML-powered queries
/javascript/ql/experimental/adaptivethreatmodeling/ @github/codeql-ml-powered-queries-reviewers
# 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
/java/ql/src/semmle/code/java/dataflow/internal/DataFlowImpl2.qll @github/codeql-java @github/codeql-go
/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/codeql-cli/ @github/codeql-cli-reviewers
/docs/codeql/codeql-for-visual-studio-code/ @github/codeql-vscode-reviewers

View File

@@ -25,7 +25,6 @@ provide:
- "misc/suite-helpers/qlpack.yml"
- "ruby/extractor-pack/codeql-extractor.yml"
- "swift/extractor-pack/codeql-extractor.yml"
- "swift/integration-tests/qlpack.yml"
- "ql/extractor-pack/codeql-extractor.yml"
versionPolicies:

View File

@@ -1,5 +1,5 @@
{
"DataFlow Java/C++/C#/Go/Python/Ruby/Swift": [
"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",
@@ -27,8 +27,6 @@
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl4.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImpl5.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplForContentDataFlow.qll",
"go/ql/lib/semmle/go/dataflow/internal/DataFlowImpl.qll",
"go/ql/lib/semmle/go/dataflow/internal/DataFlowImpl2.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",
@@ -40,18 +38,17 @@
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplForPathname.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImpl.qll"
],
"DataFlow Java/C++/C#/Go/Python/Ruby/Swift Common": [
"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",
"cpp/ql/lib/experimental/semmle/code/cpp/ir/dataflow/internal/DataFlowImplCommon.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/DataFlowImplCommon.qll",
"go/ql/lib/semmle/go/dataflow/internal/DataFlowImplCommon.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/DataFlowImplCommon.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplCommon.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImplCommon.qll"
],
"TaintTracking::Configuration Java/C++/C#/Go/Python/Ruby/Swift": [
"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",
@@ -65,8 +62,6 @@
"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",
"go/ql/lib/semmle/go/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"go/ql/lib/semmle/go/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/tainttracking2/TaintTrackingImpl.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/tainttracking3/TaintTrackingImpl.qll",
@@ -77,7 +72,7 @@
"ruby/ql/lib/codeql/ruby/dataflow/internal/tainttracking1/TaintTrackingImpl.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/tainttracking1/TaintTrackingImpl.qll"
],
"DataFlow Java/C++/C#/Python/Ruby/Swift Consistency checks": [
"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",
@@ -87,10 +82,9 @@
"ruby/ql/lib/codeql/ruby/dataflow/internal/DataFlowImplConsistency.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/DataFlowImplConsistency.qll"
],
"DataFlow Java/C#/Go/Ruby/Python/Swift Flow Summaries": [
"DataFlow Java/C#/Ruby/Python/Swift Flow Summaries": [
"java/ql/lib/semmle/code/java/dataflow/internal/FlowSummaryImpl.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/FlowSummaryImpl.qll",
"go/ql/lib/semmle/go/dataflow/internal/FlowSummaryImpl.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/FlowSummaryImpl.qll",
"python/ql/lib/semmle/python/dataflow/new/internal/FlowSummaryImpl.qll",
"swift/ql/lib/codeql/swift/dataflow/internal/FlowSummaryImpl.qll"
@@ -470,10 +464,6 @@
"javascript/ql/src/Comments/CommentedOutCodeReferences.inc.qhelp",
"python/ql/src/Lexical/CommentedOutCodeReferences.inc.qhelp"
],
"ThreadResourceAbuse qhelp": [
"java/ql/src/experimental/Security/CWE/CWE-400/LocalThreadResourceAbuse.qhelp",
"java/ql/src/experimental/Security/CWE/CWE-400/ThreadResourceAbuse.qhelp"
],
"IDE Contextual Queries": [
"cpp/ql/lib/IDEContextual.qll",
"csharp/ql/lib/IDEContextual.qll",
@@ -505,9 +495,16 @@
"python/ql/lib/semmle/python/dataflow/new/internal/TypeTracker.qll",
"ruby/ql/lib/codeql/ruby/typetracking/TypeTracker.qll"
],
"CodeQL Tutorial": [
"cpp/ql/lib/tutorial.qll",
"csharp/ql/lib/tutorial.qll",
"java/ql/lib/tutorial.qll",
"javascript/ql/lib/tutorial.qll",
"python/ql/lib/tutorial.qll",
"ruby/ql/lib/tutorial.qll"
],
"AccessPathSyntax": [
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/AccessPathSyntax.qll",
"go/ql/lib/semmle/go/dataflow/internal/AccessPathSyntax.qll",
"java/ql/lib/semmle/code/java/dataflow/internal/AccessPathSyntax.qll",
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/AccessPathSyntax.qll",
"ruby/ql/lib/codeql/ruby/dataflow/internal/AccessPathSyntax.qll",
@@ -523,16 +520,16 @@
"ruby/ql/lib/codeql/ruby/internal/ConceptsShared.qll",
"javascript/ql/lib/semmle/javascript/internal/ConceptsShared.qll"
],
"Hostname Regexp queries": [
"javascript/ql/src/Security/CWE-020/HostnameRegexpShared.qll",
"python/ql/src/Security/CWE-020/HostnameRegexpShared.qll",
"ruby/ql/src/queries/security/cwe-020/HostnameRegexpShared.qll"
],
"ApiGraphModels": [
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/ApiGraphModels.qll",
"ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModels.qll",
"python/ql/lib/semmle/python/frameworks/data/internal/ApiGraphModels.qll"
],
"ApiGraphModelsExtensions": [
"javascript/ql/lib/semmle/javascript/frameworks/data/internal/ApiGraphModelsExtensions.qll",
"ruby/ql/lib/codeql/ruby/frameworks/data/internal/ApiGraphModelsExtensions.qll",
"python/ql/lib/semmle/python/frameworks/data/internal/ApiGraphModelsExtensions.qll"
],
"TaintedFormatStringQuery Ruby/JS": [
"javascript/ql/lib/semmle/javascript/security/dataflow/TaintedFormatStringQuery.qll",
"ruby/ql/lib/codeql/ruby/security/TaintedFormatStringQuery.qll"
@@ -576,9 +573,5 @@
"IncompleteMultiCharacterSanitization JS/Ruby": [
"javascript/ql/lib/semmle/javascript/security/IncompleteMultiCharacterSanitizationQuery.qll",
"ruby/ql/lib/codeql/ruby/security/IncompleteMultiCharacterSanitizationQuery.qll"
],
"EncryptionKeySizes Python/Java": [
"python/ql/lib/semmle/python/security/internal/EncryptionKeySizes.qll",
"java/ql/lib/semmle/code/java/security/internal/EncryptionKeySizes.qll"
]
}

View File

@@ -1,32 +1,3 @@
## 0.5.0
### Breaking Changes
The predicates in the `MustFlow::Configuration` class used by the `MustFlow` library (`semmle.code.cpp.ir.dataflow.MustFlow`) have changed to be defined directly in terms of the C++ IR instead of IR dataflow nodes.
### Deprecated APIs
* Deprecated `semmle.code.cpp.ir.dataflow.DefaultTaintTracking`. Use `semmle.code.cpp.ir.dataflow.TaintTracking`.
* Deprecated `semmle.code.cpp.security.TaintTrackingImpl`. Use `semmle.code.cpp.ir.dataflow.TaintTracking`.
* Deprecated `semmle.code.cpp.valuenumbering.GlobalValueNumberingImpl`. Use `semmle.code.cpp.valuenumbering.GlobalValueNumbering`, which exposes the same API.
### Minor Analysis Improvements
* The `ArgvSource` flow source now uses the second parameter of `main` as its source instead of the uses of this parameter.
* The `ArgvSource` flow source has been generalized to handle cases where the argument vector of `main` is not named `argv`.
* The `getaddrinfo` function is now recognized as a flow source.
* The `secure_getenv` and `_wgetenv` functions are now recognized as local flow sources.
* The `scanf` and `fscanf` functions and their variants are now recognized as flow sources.
* Deleted the deprecated `getName` and `getShortName` predicates from the `Folder` class.
## 0.4.6
No user-facing changes.
## 0.4.5
No user-facing changes.
## 0.4.4
No user-facing changes.

View File

@@ -0,0 +1,6 @@
---
category: deprecated
---
* Deprecated `semmle.code.cpp.valuenumbering.GlobalValueNumberingImpl`. Use `semmle.code.cpp.valuenumbering.GlobalValueNumbering`, which exposes the same API.

View File

@@ -0,0 +1,4 @@
---
category: breaking
---
The predicates in the `MustFlow::Configuration` class used by the `MustFlow` library (`semmle.code.cpp.ir.dataflow.MustFlow`) have changed to be defined directly in terms of the C++ IR instead of IR dataflow nodes.

View File

@@ -0,0 +1,4 @@
---
category: minorAnalysis
---
* Deleted the deprecated `getName` and `getShortName` predicates from the `Folder` class.

View File

@@ -1,3 +0,0 @@
## 0.4.5
No user-facing changes.

View File

@@ -1,3 +0,0 @@
## 0.4.6
No user-facing changes.

View File

@@ -1,20 +0,0 @@
## 0.5.0
### Breaking Changes
The predicates in the `MustFlow::Configuration` class used by the `MustFlow` library (`semmle.code.cpp.ir.dataflow.MustFlow`) have changed to be defined directly in terms of the C++ IR instead of IR dataflow nodes.
### Deprecated APIs
* Deprecated `semmle.code.cpp.ir.dataflow.DefaultTaintTracking`. Use `semmle.code.cpp.ir.dataflow.TaintTracking`.
* Deprecated `semmle.code.cpp.security.TaintTrackingImpl`. Use `semmle.code.cpp.ir.dataflow.TaintTracking`.
* Deprecated `semmle.code.cpp.valuenumbering.GlobalValueNumberingImpl`. Use `semmle.code.cpp.valuenumbering.GlobalValueNumbering`, which exposes the same API.
### Minor Analysis Improvements
* The `ArgvSource` flow source now uses the second parameter of `main` as its source instead of the uses of this parameter.
* The `ArgvSource` flow source has been generalized to handle cases where the argument vector of `main` is not named `argv`.
* The `getaddrinfo` function is now recognized as a flow source.
* The `secure_getenv` and `_wgetenv` functions are now recognized as local flow sources.
* The `scanf` and `fscanf` functions and their variants are now recognized as flow sources.
* Deleted the deprecated `getName` and `getShortName` predicates from the `Folder` class.

View File

@@ -1,2 +1,2 @@
---
lastReleaseVersion: 0.5.0
lastReleaseVersion: 0.4.4

View File

@@ -12,8 +12,8 @@ import IDEContextual
*
* In some cases it is preferable to modify locations (the
* `hasLocationInfo()` predicate) so that they are short, and
* non-overlapping with other locations that might be reported as
* code scanning alerts on GitHub.
* non-overlapping with other locations that might be highlighted in
* the LGTM interface.
*
* We need to give locations that may not be in the database, so
* we use `hasLocationInfo()` rather than `getLocation()`.

View File

@@ -915,57 +915,18 @@ private module Cached {
TDataFlowCallNone() or
TDataFlowCallSome(DataFlowCall call)
cached
newtype TParamNodeOption =
TParamNodeNone() or
TParamNodeSome(ParamNode p)
cached
newtype TReturnCtx =
TReturnCtxNone() or
TReturnCtxNoFlowThrough() or
TReturnCtxMaybeFlowThrough(ReturnPosition pos)
cached
newtype TTypedContentApprox =
MkTypedContentApprox(ContentApprox c, DataFlowType t) {
exists(Content cont |
c = getContentApprox(cont) and
store(_, cont, _, _, t)
)
}
cached
newtype TTypedContent = MkTypedContent(Content c, DataFlowType t) { store(_, c, _, _, t) }
cached
TypedContent getATypedContent(TypedContentApprox c) {
exists(ContentApprox cls, DataFlowType t, Content cont |
c = MkTypedContentApprox(cls, pragma[only_bind_into](t)) and
result = MkTypedContent(cont, pragma[only_bind_into](t)) and
cls = getContentApprox(cont)
)
}
cached
newtype TAccessPathFront =
TFrontNil(DataFlowType t) or
TFrontHead(TypedContent tc)
cached
newtype TApproxAccessPathFront =
TApproxFrontNil(DataFlowType t) or
TApproxFrontHead(TypedContentApprox tc)
cached
newtype TAccessPathFrontOption =
TAccessPathFrontNone() or
TAccessPathFrontSome(AccessPathFront apf)
cached
newtype TApproxAccessPathFrontOption =
TApproxAccessPathFrontNone() or
TApproxAccessPathFrontSome(ApproxAccessPathFront apf)
}
/**
@@ -1343,113 +1304,6 @@ class DataFlowCallOption extends TDataFlowCallOption {
}
}
/** An optional `ParamNode`. */
class ParamNodeOption extends TParamNodeOption {
string toString() {
this = TParamNodeNone() and
result = "(none)"
or
exists(ParamNode p |
this = TParamNodeSome(p) and
result = p.toString()
)
}
}
/**
* A return context used to calculate flow summaries in reverse flow.
*
* The possible values are:
*
* - `TReturnCtxNone()`: no return flow.
* - `TReturnCtxNoFlowThrough()`: return flow, but flow through is not possible.
* - `TReturnCtxMaybeFlowThrough(ReturnPosition pos)`: return flow, of kind `pos`, and
* flow through may be possible.
*/
class ReturnCtx extends TReturnCtx {
string toString() {
this = TReturnCtxNone() and
result = "(none)"
or
this = TReturnCtxNoFlowThrough() and
result = "(no flow through)"
or
exists(ReturnPosition pos |
this = TReturnCtxMaybeFlowThrough(pos) and
result = pos.toString()
)
}
}
/** An approximated `Content` tagged with the type of a containing object. */
class TypedContentApprox extends MkTypedContentApprox {
private ContentApprox c;
private DataFlowType t;
TypedContentApprox() { this = MkTypedContentApprox(c, t) }
/** Gets a typed content approximated by this value. */
TypedContent getATypedContent() { result = getATypedContent(this) }
/** Gets the container type. */
DataFlowType getContainerType() { result = t }
/** Gets a textual representation of this approximated content. */
string toString() { result = c.toString() }
}
/**
* The front of an approximated access path. This is either a head or a nil.
*/
abstract class ApproxAccessPathFront extends TApproxAccessPathFront {
abstract string toString();
abstract DataFlowType getType();
abstract boolean toBoolNonEmpty();
pragma[nomagic]
TypedContent getAHead() {
exists(TypedContentApprox cont |
this = TApproxFrontHead(cont) and
result = cont.getATypedContent()
)
}
}
class ApproxAccessPathFrontNil extends ApproxAccessPathFront, TApproxFrontNil {
private DataFlowType t;
ApproxAccessPathFrontNil() { this = TApproxFrontNil(t) }
override string toString() { result = ppReprType(t) }
override DataFlowType getType() { result = t }
override boolean toBoolNonEmpty() { result = false }
}
class ApproxAccessPathFrontHead extends ApproxAccessPathFront, TApproxFrontHead {
private TypedContentApprox tc;
ApproxAccessPathFrontHead() { this = TApproxFrontHead(tc) }
override string toString() { result = tc.toString() }
override DataFlowType getType() { result = tc.getContainerType() }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional approximated access path front. */
class ApproxAccessPathFrontOption extends TApproxAccessPathFrontOption {
string toString() {
this = TApproxAccessPathFrontNone() and result = "<none>"
or
this = TApproxAccessPathFrontSome(any(ApproxAccessPathFront apf | result = apf.toString()))
}
}
/** A `Content` tagged with the type of a containing object. */
class TypedContent extends MkTypedContent {
private Content c;
@@ -1482,7 +1336,7 @@ abstract class AccessPathFront extends TAccessPathFront {
abstract DataFlowType getType();
abstract ApproxAccessPathFront toApprox();
abstract boolean toBoolNonEmpty();
TypedContent getHead() { this = TFrontHead(result) }
}
@@ -1496,7 +1350,7 @@ class AccessPathFrontNil extends AccessPathFront, TFrontNil {
override DataFlowType getType() { result = t }
override ApproxAccessPathFront toApprox() { result = TApproxFrontNil(t) }
override boolean toBoolNonEmpty() { result = false }
}
class AccessPathFrontHead extends AccessPathFront, TFrontHead {
@@ -1508,7 +1362,7 @@ class AccessPathFrontHead extends AccessPathFront, TFrontHead {
override DataFlowType getType() { result = tc.getContainerType() }
override ApproxAccessPathFront toApprox() { result.getAHead() = tc }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional access path front. */

View File

@@ -244,25 +244,4 @@ module Consistency {
not callable = viableCallable(call) and
not any(ConsistencyConfiguration c).viableImplInCallContextTooLargeExclude(call, ctx, callable)
}
query predicate uniqueParameterNodeAtPosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(Node p0 | isParameterNode(p0, c, pos))) and
msg = "Parameters with overlapping positions."
}
query predicate uniqueParameterNodePosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(ParameterPosition pos0 | isParameterNode(p, c, pos0))) and
msg = "Parameter node with multiple positions."
}
query predicate uniqueContentApprox(Content c, string msg) {
not exists(unique(ContentApprox approx | approx = getContentApprox(c))) and
msg = "Non-unique content approximation."
}
}

View File

@@ -551,13 +551,6 @@ predicate additionalLambdaFlowStep(Node nodeFrom, Node nodeTo, boolean preserves
*/
predicate allowParameterReturnInSelf(ParameterNode p) { none() }
/** An approximated `Content`. */
class ContentApprox = Unit;
/** Gets an approximated value for content `c`. */
pragma[inline]
ContentApprox getContentApprox(Content c) { any() }
private class MyConsistencyConfiguration extends Consistency::ConsistencyConfiguration {
override predicate argHasPostUpdateExclude(ArgumentNode n) {
// The rules for whether an IR argument gets a post-update node are too

View File

@@ -292,8 +292,12 @@ module SemanticExprConfig {
final Location getLocation() { result = super.getLocation() }
}
private class ValueNumberBound extends Bound instanceof IRBound::ValueNumberBound {
override string toString() { result = IRBound::ValueNumberBound.super.toString() }
private class ValueNumberBound extends Bound {
IRBound::ValueNumberBound bound;
ValueNumberBound() { bound = this }
override string toString() { result = bound.toString() }
}
predicate zeroBound(Bound bound) { bound instanceof IRBound::ZeroBound }

View File

@@ -33,15 +33,23 @@ abstract private class FlowSignDef extends SignDef {
}
/** An SSA definition whose sign is determined by the sign of that definitions source expression. */
private class ExplicitSignDef extends FlowSignDef instanceof SemSsaExplicitUpdate {
final override Sign getSign() { result = semExprSign(super.getSourceExpr()) }
private class ExplicitSignDef extends FlowSignDef {
SemSsaExplicitUpdate update;
ExplicitSignDef() { update = this }
final override Sign getSign() { result = semExprSign(update.getSourceExpr()) }
}
/** An SSA Phi definition, whose sign is the union of the signs of its inputs. */
private class PhiSignDef extends FlowSignDef instanceof SemSsaPhiNode {
private class PhiSignDef extends FlowSignDef {
SemSsaPhiNode phi;
PhiSignDef() { phi = this }
final override Sign getSign() {
exists(SemSsaVariable inp, SemSsaReadPositionPhiInputEdge edge |
edge.phiInput(this, inp) and
edge.phiInput(phi, inp) and
result = semSsaSign(inp, edge)
)
}

View File

@@ -1,5 +1,5 @@
name: codeql/cpp-all
version: 0.5.0
version: 0.4.5-dev
groups: cpp
dbscheme: semmlecode.cpp.dbscheme
extractor: cpp
@@ -7,4 +7,3 @@ library: true
upgrades: upgrades
dependencies:
codeql/ssa: ${workspace}
codeql/tutorial: ${workspace}

View File

@@ -397,8 +397,11 @@ private int lengthInBase16(float f) {
/**
* A class to represent format strings that occur as arguments to invocations of formatting functions.
*/
class FormatLiteral extends Literal instanceof StringLiteral {
FormatLiteral() { exists(FormattingFunctionCall ffc | ffc.getFormat() = this) }
class FormatLiteral extends Literal {
FormatLiteral() {
exists(FormattingFunctionCall ffc | ffc.getFormat() = this) and
this instanceof StringLiteral
}
/**
* Gets the function call where this format string is used.

View File

@@ -30,12 +30,15 @@ abstract class ScanfFunction extends Function {
/**
* The standard function `scanf` (and variations).
*/
class Scanf extends ScanfFunction instanceof TopLevelFunction {
class Scanf extends ScanfFunction {
Scanf() {
this.hasGlobalOrStdOrBslName("scanf") or // scanf(format, args...)
this.hasGlobalOrStdOrBslName("wscanf") or // wscanf(format, args...)
this.hasGlobalName("_scanf_l") or // _scanf_l(format, locale, args...)
this.hasGlobalName("_wscanf_l")
this instanceof TopLevelFunction and
(
this.hasGlobalOrStdOrBslName("scanf") or // scanf(format, args...)
this.hasGlobalOrStdOrBslName("wscanf") or // wscanf(format, args...)
this.hasGlobalName("_scanf_l") or // _scanf_l(format, locale, args...)
this.hasGlobalName("_wscanf_l") // _wscanf_l(format, locale, args...)
)
}
override int getInputParameterIndex() { none() }
@@ -46,12 +49,15 @@ class Scanf extends ScanfFunction instanceof TopLevelFunction {
/**
* The standard function `fscanf` (and variations).
*/
class Fscanf extends ScanfFunction instanceof TopLevelFunction {
class Fscanf extends ScanfFunction {
Fscanf() {
this.hasGlobalOrStdOrBslName("fscanf") or // fscanf(src_stream, format, args...)
this.hasGlobalOrStdOrBslName("fwscanf") or // fwscanf(src_stream, format, args...)
this.hasGlobalName("_fscanf_l") or // _fscanf_l(src_stream, format, locale, args...)
this.hasGlobalName("_fwscanf_l")
this instanceof TopLevelFunction and
(
this.hasGlobalOrStdOrBslName("fscanf") or // fscanf(src_stream, format, args...)
this.hasGlobalOrStdOrBslName("fwscanf") or // fwscanf(src_stream, format, args...)
this.hasGlobalName("_fscanf_l") or // _fscanf_l(src_stream, format, locale, args...)
this.hasGlobalName("_fwscanf_l") // _fwscanf_l(src_stream, format, locale, args...)
)
}
override int getInputParameterIndex() { result = 0 }
@@ -62,12 +68,15 @@ class Fscanf extends ScanfFunction instanceof TopLevelFunction {
/**
* The standard function `sscanf` (and variations).
*/
class Sscanf extends ScanfFunction instanceof TopLevelFunction {
class Sscanf extends ScanfFunction {
Sscanf() {
this.hasGlobalOrStdOrBslName("sscanf") or // sscanf(src_stream, format, args...)
this.hasGlobalOrStdOrBslName("swscanf") or // swscanf(src, format, args...)
this.hasGlobalName("_sscanf_l") or // _sscanf_l(src, format, locale, args...)
this.hasGlobalName("_swscanf_l")
this instanceof TopLevelFunction and
(
this.hasGlobalOrStdOrBslName("sscanf") or // sscanf(src_stream, format, args...)
this.hasGlobalOrStdOrBslName("swscanf") or // swscanf(src, format, args...)
this.hasGlobalName("_sscanf_l") or // _sscanf_l(src, format, locale, args...)
this.hasGlobalName("_swscanf_l") // _swscanf_l(src, format, locale, args...)
)
}
override int getInputParameterIndex() { result = 0 }
@@ -78,12 +87,17 @@ class Sscanf extends ScanfFunction instanceof TopLevelFunction {
/**
* The standard(ish) function `snscanf` (and variations).
*/
class Snscanf extends ScanfFunction instanceof TopLevelFunction {
class Snscanf extends ScanfFunction {
Snscanf() {
this.hasGlobalName("_snscanf") or // _snscanf(src, max_amount, format, args...)
this.hasGlobalName("_snwscanf") or // _snwscanf(src, max_amount, format, args...)
this.hasGlobalName("_snscanf_l") or // _snscanf_l(src, max_amount, format, locale, args...)
this.hasGlobalName("_snwscanf_l")
this instanceof TopLevelFunction and
(
this.hasGlobalName("_snscanf") or // _snscanf(src, max_amount, format, args...)
this.hasGlobalName("_snwscanf") or // _snwscanf(src, max_amount, format, args...)
this.hasGlobalName("_snscanf_l") or // _snscanf_l(src, max_amount, format, locale, args...)
this.hasGlobalName("_snwscanf_l") // _snwscanf_l(src, max_amount, format, locale, args...)
// note that the max_amount is not a limit on the output length, it's an input length
// limit used with non null-terminated strings.
)
}
override int getInputParameterIndex() { result = 0 }

File diff suppressed because it is too large Load Diff

View File

@@ -915,57 +915,18 @@ private module Cached {
TDataFlowCallNone() or
TDataFlowCallSome(DataFlowCall call)
cached
newtype TParamNodeOption =
TParamNodeNone() or
TParamNodeSome(ParamNode p)
cached
newtype TReturnCtx =
TReturnCtxNone() or
TReturnCtxNoFlowThrough() or
TReturnCtxMaybeFlowThrough(ReturnPosition pos)
cached
newtype TTypedContentApprox =
MkTypedContentApprox(ContentApprox c, DataFlowType t) {
exists(Content cont |
c = getContentApprox(cont) and
store(_, cont, _, _, t)
)
}
cached
newtype TTypedContent = MkTypedContent(Content c, DataFlowType t) { store(_, c, _, _, t) }
cached
TypedContent getATypedContent(TypedContentApprox c) {
exists(ContentApprox cls, DataFlowType t, Content cont |
c = MkTypedContentApprox(cls, pragma[only_bind_into](t)) and
result = MkTypedContent(cont, pragma[only_bind_into](t)) and
cls = getContentApprox(cont)
)
}
cached
newtype TAccessPathFront =
TFrontNil(DataFlowType t) or
TFrontHead(TypedContent tc)
cached
newtype TApproxAccessPathFront =
TApproxFrontNil(DataFlowType t) or
TApproxFrontHead(TypedContentApprox tc)
cached
newtype TAccessPathFrontOption =
TAccessPathFrontNone() or
TAccessPathFrontSome(AccessPathFront apf)
cached
newtype TApproxAccessPathFrontOption =
TApproxAccessPathFrontNone() or
TApproxAccessPathFrontSome(ApproxAccessPathFront apf)
}
/**
@@ -1343,113 +1304,6 @@ class DataFlowCallOption extends TDataFlowCallOption {
}
}
/** An optional `ParamNode`. */
class ParamNodeOption extends TParamNodeOption {
string toString() {
this = TParamNodeNone() and
result = "(none)"
or
exists(ParamNode p |
this = TParamNodeSome(p) and
result = p.toString()
)
}
}
/**
* A return context used to calculate flow summaries in reverse flow.
*
* The possible values are:
*
* - `TReturnCtxNone()`: no return flow.
* - `TReturnCtxNoFlowThrough()`: return flow, but flow through is not possible.
* - `TReturnCtxMaybeFlowThrough(ReturnPosition pos)`: return flow, of kind `pos`, and
* flow through may be possible.
*/
class ReturnCtx extends TReturnCtx {
string toString() {
this = TReturnCtxNone() and
result = "(none)"
or
this = TReturnCtxNoFlowThrough() and
result = "(no flow through)"
or
exists(ReturnPosition pos |
this = TReturnCtxMaybeFlowThrough(pos) and
result = pos.toString()
)
}
}
/** An approximated `Content` tagged with the type of a containing object. */
class TypedContentApprox extends MkTypedContentApprox {
private ContentApprox c;
private DataFlowType t;
TypedContentApprox() { this = MkTypedContentApprox(c, t) }
/** Gets a typed content approximated by this value. */
TypedContent getATypedContent() { result = getATypedContent(this) }
/** Gets the container type. */
DataFlowType getContainerType() { result = t }
/** Gets a textual representation of this approximated content. */
string toString() { result = c.toString() }
}
/**
* The front of an approximated access path. This is either a head or a nil.
*/
abstract class ApproxAccessPathFront extends TApproxAccessPathFront {
abstract string toString();
abstract DataFlowType getType();
abstract boolean toBoolNonEmpty();
pragma[nomagic]
TypedContent getAHead() {
exists(TypedContentApprox cont |
this = TApproxFrontHead(cont) and
result = cont.getATypedContent()
)
}
}
class ApproxAccessPathFrontNil extends ApproxAccessPathFront, TApproxFrontNil {
private DataFlowType t;
ApproxAccessPathFrontNil() { this = TApproxFrontNil(t) }
override string toString() { result = ppReprType(t) }
override DataFlowType getType() { result = t }
override boolean toBoolNonEmpty() { result = false }
}
class ApproxAccessPathFrontHead extends ApproxAccessPathFront, TApproxFrontHead {
private TypedContentApprox tc;
ApproxAccessPathFrontHead() { this = TApproxFrontHead(tc) }
override string toString() { result = tc.toString() }
override DataFlowType getType() { result = tc.getContainerType() }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional approximated access path front. */
class ApproxAccessPathFrontOption extends TApproxAccessPathFrontOption {
string toString() {
this = TApproxAccessPathFrontNone() and result = "<none>"
or
this = TApproxAccessPathFrontSome(any(ApproxAccessPathFront apf | result = apf.toString()))
}
}
/** A `Content` tagged with the type of a containing object. */
class TypedContent extends MkTypedContent {
private Content c;
@@ -1482,7 +1336,7 @@ abstract class AccessPathFront extends TAccessPathFront {
abstract DataFlowType getType();
abstract ApproxAccessPathFront toApprox();
abstract boolean toBoolNonEmpty();
TypedContent getHead() { this = TFrontHead(result) }
}
@@ -1496,7 +1350,7 @@ class AccessPathFrontNil extends AccessPathFront, TFrontNil {
override DataFlowType getType() { result = t }
override ApproxAccessPathFront toApprox() { result = TApproxFrontNil(t) }
override boolean toBoolNonEmpty() { result = false }
}
class AccessPathFrontHead extends AccessPathFront, TFrontHead {
@@ -1508,7 +1362,7 @@ class AccessPathFrontHead extends AccessPathFront, TFrontHead {
override DataFlowType getType() { result = tc.getContainerType() }
override ApproxAccessPathFront toApprox() { result.getAHead() = tc }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional access path front. */

View File

@@ -244,25 +244,4 @@ module Consistency {
not callable = viableCallable(call) and
not any(ConsistencyConfiguration c).viableImplInCallContextTooLargeExclude(call, ctx, callable)
}
query predicate uniqueParameterNodeAtPosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(Node p0 | isParameterNode(p0, c, pos))) and
msg = "Parameters with overlapping positions."
}
query predicate uniqueParameterNodePosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(ParameterPosition pos0 | isParameterNode(p, c, pos0))) and
msg = "Parameter node with multiple positions."
}
query predicate uniqueContentApprox(Content c, string msg) {
not exists(unique(ContentApprox approx | approx = getContentApprox(c))) and
msg = "Non-unique content approximation."
}
}

View File

@@ -296,13 +296,6 @@ predicate additionalLambdaFlowStep(Node nodeFrom, Node nodeTo, boolean preserves
*/
predicate allowParameterReturnInSelf(ParameterNode p) { none() }
/** An approximated `Content`. */
class ContentApprox = Unit;
/** Gets an approximated value for content `c`. */
pragma[inline]
ContentApprox getContentApprox(Content c) { any() }
private class MyConsistencyConfiguration extends Consistency::ConsistencyConfiguration {
override predicate argHasPostUpdateExclude(ArgumentNode n) {
// Is the null pointer (or something that's not really a pointer)

View File

@@ -1,21 +1,642 @@
/**
* DEPRECATED: Use `semmle.code.cpp.ir.dataflow.TaintTracking` as a replacement.
*
* An IR taint tracking library that uses an IR DataFlow configuration to track
* taint from user inputs as defined by `semmle.code.cpp.security.Security`.
*/
import cpp
import semmle.code.cpp.security.Security
private import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl as DefaultTaintTrackingImpl
private import semmle.code.cpp.ir.dataflow.DataFlow
private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import semmle.code.cpp.ir.dataflow.DataFlow3
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.ResolveCall
private import semmle.code.cpp.controlflow.IRGuards
private import semmle.code.cpp.models.interfaces.Taint
private import semmle.code.cpp.models.interfaces.DataFlow
private import semmle.code.cpp.ir.dataflow.TaintTracking
private import semmle.code.cpp.ir.dataflow.TaintTracking2
private import semmle.code.cpp.ir.dataflow.TaintTracking3
private import semmle.code.cpp.ir.dataflow.internal.ModelUtil
deprecated predicate predictableOnlyFlow = DefaultTaintTrackingImpl::predictableOnlyFlow/1;
/**
* A predictable instruction is one where an external user can predict
* the value. For example, a literal in the source code is considered
* predictable.
*/
private predicate predictableInstruction(Instruction instr) {
instr instanceof ConstantInstruction
or
instr instanceof StringConstantInstruction
or
// This could be a conversion on a string literal
predictableInstruction(instr.(UnaryInstruction).getUnary())
}
deprecated predicate tainted = DefaultTaintTrackingImpl::tainted/2;
/**
* Functions that we should only allow taint to flow through (to the return
* value) if all but the source argument are 'predictable'. This is done to
* emulate the old security library's implementation rather than due to any
* strong belief that this is the right approach.
*
* Note that the list itself is not very principled; it consists of all the
* functions listed in the old security library's [default] `isPureFunction`
* that have more than one argument, but are not in the old taint tracking
* library's `returnArgument` predicate.
*/
predicate predictableOnlyFlow(string name) {
name =
[
"strcasestr", "strchnul", "strchr", "strchrnul", "strcmp", "strcspn", "strncmp", "strndup",
"strnlen", "strrchr", "strspn", "strstr", "strtod", "strtof", "strtol", "strtoll", "strtoq",
"strtoul"
]
}
deprecated predicate taintedIncludingGlobalVars =
DefaultTaintTrackingImpl::taintedIncludingGlobalVars/3;
private DataFlow::Node getNodeForSource(Expr source) {
isUserInput(source, _) and
result = getNodeForExpr(source)
}
deprecated predicate globalVarFromId = DefaultTaintTrackingImpl::globalVarFromId/1;
private DataFlow::Node getNodeForExpr(Expr node) {
result = DataFlow::exprNode(node)
or
// Some of the sources in `isUserInput` are intended to match the value of
// an expression, while others (those modeled below) are intended to match
// the taint that propagates out of an argument, like the `char *` argument
// to `gets`. It's impossible here to tell which is which, but the "access
// to argv" source is definitely not intended to match an output argument,
// and it causes false positives if we let it.
//
// This case goes together with the similar (but not identical) rule in
// `nodeIsBarrierIn`.
result = DataFlow::definitionByReferenceNodeFromArgument(node) and
not argv(node.(VariableAccess).getTarget())
}
deprecated module TaintedWithPath = DefaultTaintTrackingImpl::TaintedWithPath;
private class DefaultTaintTrackingCfg extends TaintTracking::Configuration {
DefaultTaintTrackingCfg() { this = "DefaultTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) { source = getNodeForSource(_) }
override predicate isSink(DataFlow::Node sink) { exists(adjustedSink(sink)) }
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private class ToGlobalVarTaintTrackingCfg extends TaintTracking::Configuration {
ToGlobalVarTaintTrackingCfg() { this = "GlobalVarTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) { source = getNodeForSource(_) }
override predicate isSink(DataFlow::Node sink) {
sink.asVariable() instanceof GlobalOrNamespaceVariable
}
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
writesVariable(n1.asInstruction(), n2.asVariable().(GlobalOrNamespaceVariable))
or
readsVariable(n2.asInstruction(), n1.asVariable().(GlobalOrNamespaceVariable))
}
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private class FromGlobalVarTaintTrackingCfg extends TaintTracking2::Configuration {
FromGlobalVarTaintTrackingCfg() { this = "FromGlobalVarTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) {
// This set of sources should be reasonably small, which is good for
// performance since the set of sinks is very large.
exists(ToGlobalVarTaintTrackingCfg otherCfg | otherCfg.hasFlowTo(source))
}
override predicate isSink(DataFlow::Node sink) { exists(adjustedSink(sink)) }
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
// Additional step for flow out of variables. There is no flow _into_
// variables in this configuration, so this step only serves to take flow
// out of a variable that's a source.
readsVariable(n2.asInstruction(), n1.asVariable())
}
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private predicate readsVariable(LoadInstruction load, Variable var) {
load.getSourceAddress().(VariableAddressInstruction).getAstVariable() = var
}
private predicate writesVariable(StoreInstruction store, Variable var) {
store.getDestinationAddress().(VariableAddressInstruction).getAstVariable() = var
}
/**
* A variable that has any kind of upper-bound check anywhere in the program. This is
* biased towards being inclusive because there are a lot of valid ways of doing an
* upper bounds checks if we don't consider where it occurs, for example:
* ```
* if (x < 10) { sink(x); }
*
* if (10 > y) { sink(y); }
*
* if (z > 10) { z = 10; }
* sink(z);
* ```
*/
// TODO: This coarse overapproximation, ported from the old taint tracking
// library, could be replaced with an actual semantic check that a particular
// variable _access_ is guarded by an upper-bound check. We probably don't want
// to do this right away since it could expose a lot of FPs that were
// previously suppressed by this predicate by coincidence.
private predicate hasUpperBoundsCheck(Variable var) {
exists(RelationalOperation oper, VariableAccess access |
oper.getAnOperand() = access and
access.getTarget() = var and
// Comparing to 0 is not an upper bound check
not oper.getAnOperand().getValue() = "0"
)
}
private predicate nodeIsBarrierEqualityCandidate(
DataFlow::Node node, Operand access, Variable checkedVar
) {
readsVariable(node.asInstruction(), checkedVar) and
any(IRGuardCondition guard).ensuresEq(access, _, _, node.asInstruction().getBlock(), true)
}
cached
private module Cached {
cached
predicate nodeIsBarrier(DataFlow::Node node) {
exists(Variable checkedVar |
readsVariable(node.asInstruction(), checkedVar) and
hasUpperBoundsCheck(checkedVar)
)
or
exists(Variable checkedVar, Operand access |
/*
* This node is guarded by a condition that forces the accessed variable
* to equal something else. For example:
* ```
* x = taintsource()
* if (x == 10) {
* taintsink(x); // not considered tainted
* }
* ```
*/
nodeIsBarrierEqualityCandidate(node, access, checkedVar) and
readsVariable(access.getDef(), checkedVar)
)
}
cached
predicate nodeIsBarrierIn(DataFlow::Node node) {
// don't use dataflow into taint sources, as this leads to duplicate results.
exists(Expr source | isUserInput(source, _) |
node = DataFlow::exprNode(source)
or
// This case goes together with the similar (but not identical) rule in
// `getNodeForSource`.
node = DataFlow::definitionByReferenceNodeFromArgument(source)
)
or
// don't use dataflow into binary instructions if both operands are unpredictable
exists(BinaryInstruction iTo |
iTo = node.asInstruction() and
not predictableInstruction(iTo.getLeft()) and
not predictableInstruction(iTo.getRight()) and
// propagate taint from either the pointer or the offset, regardless of predictability
not iTo instanceof PointerArithmeticInstruction
)
or
// don't use dataflow through calls to pure functions if two or more operands
// are unpredictable
exists(Instruction iFrom1, Instruction iFrom2, CallInstruction iTo |
iTo = node.asInstruction() and
isPureFunction(iTo.getStaticCallTarget().getName()) and
iFrom1 = iTo.getAnArgument() and
iFrom2 = iTo.getAnArgument() and
not predictableInstruction(iFrom1) and
not predictableInstruction(iFrom2) and
iFrom1 != iFrom2
)
}
cached
Element adjustedSink(DataFlow::Node sink) {
// TODO: is it more appropriate to use asConvertedExpr here and avoid
// `getConversion*`? Or will that cause us to miss some cases where there's
// flow to a conversion (like a `ReferenceDereferenceExpr`) and we want to
// pretend there was flow to the converted `Expr` for the sake of
// compatibility.
sink.asExpr().getConversion*() = result
or
// For compatibility, send flow from arguments to parameters, even for
// functions with no body.
exists(FunctionCall call, int i |
sink.asExpr() = call.getArgument(pragma[only_bind_into](i)) and
result = resolveCall(call).getParameter(pragma[only_bind_into](i))
)
or
// For compatibility, send flow into a `Variable` if there is flow to any
// Load or Store of that variable.
exists(CopyInstruction copy |
copy.getSourceValue() = sink.asInstruction() and
(
readsVariable(copy, result) or
writesVariable(copy, result)
) and
not hasUpperBoundsCheck(result)
)
or
// For compatibility, send flow into a `NotExpr` even if it's part of a
// short-circuiting condition and thus might get skipped.
result.(NotExpr).getOperand() = sink.asExpr()
or
// Taint postfix and prefix crement operations when their operand is tainted.
result.(CrementOperation).getAnOperand() = sink.asExpr()
or
// Taint `e1 += e2`, `e &= e2` and friends when `e1` or `e2` is tainted.
result.(AssignOperation).getAnOperand() = sink.asExpr()
or
result =
sink.asOperand()
.(SideEffectOperand)
.getUse()
.(ReadSideEffectInstruction)
.getArgumentDef()
.getUnconvertedResultExpression()
}
/**
* Step to return value of a modeled function when an input taints the
* dereference of the return value.
*/
cached
predicate additionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
exists(CallInstruction call, Function func, FunctionInput modelIn, FunctionOutput modelOut |
n1.asOperand() = callInput(call, modelIn) and
(
func.(TaintFunction).hasTaintFlow(modelIn, modelOut)
or
func.(DataFlowFunction).hasDataFlow(modelIn, modelOut)
) and
call.getStaticCallTarget() = func and
modelOut.isReturnValueDeref() and
call = n2.asInstruction()
)
}
}
private import Cached
/**
* Holds if `tainted` may contain taint from `source`.
*
* A tainted expression is either directly user input, or is
* computed from user input in a way that users can probably
* control the exact output of the computation.
*
* This doesn't include data flow through global variables.
* If you need that you must call `taintedIncludingGlobalVars`.
*/
cached
predicate tainted(Expr source, Element tainted) {
exists(DefaultTaintTrackingCfg cfg, DataFlow::Node sink |
cfg.hasFlow(getNodeForSource(source), sink) and
tainted = adjustedSink(sink)
)
}
/**
* Holds if `tainted` may contain taint from `source`, where the taint passed
* through a global variable named `globalVar`.
*
* A tainted expression is either directly user input, or is
* computed from user input in a way that users can probably
* control the exact output of the computation.
*
* This version gives the same results as tainted but also includes
* data flow through global variables.
*
* The parameter `globalVar` is the qualified name of the last global variable
* used to move the value from source to tainted. If the taint did not pass
* through a global variable, then `globalVar = ""`.
*/
cached
predicate taintedIncludingGlobalVars(Expr source, Element tainted, string globalVar) {
tainted(source, tainted) and
globalVar = ""
or
exists(
ToGlobalVarTaintTrackingCfg toCfg, FromGlobalVarTaintTrackingCfg fromCfg,
DataFlow::VariableNode variableNode, GlobalOrNamespaceVariable global, DataFlow::Node sink
|
global = variableNode.getVariable() and
toCfg.hasFlow(getNodeForSource(source), variableNode) and
fromCfg.hasFlow(variableNode, sink) and
tainted = adjustedSink(sink) and
global = globalVarFromId(globalVar)
)
}
/**
* Gets the global variable whose qualified name is `id`. Use this predicate
* together with `taintedIncludingGlobalVars`. Example:
*
* ```
* exists(string varName |
* taintedIncludingGlobalVars(source, tainted, varName) and
* var = globalVarFromId(varName)
* )
* ```
*/
GlobalOrNamespaceVariable globalVarFromId(string id) { id = result.getQualifiedName() }
/**
* Provides definitions for augmenting source/sink pairs with data-flow paths
* between them. From a `@kind path-problem` query, import this module in the
* global scope, extend `TaintTrackingConfiguration`, and use `taintedWithPath`
* in place of `tainted`.
*
* Importing this module will also import the query predicates that contain the
* taint paths.
*/
module TaintedWithPath {
private newtype TSingleton = MkSingleton()
/**
* A taint-tracking configuration that matches sources and sinks in the same
* way as the `tainted` predicate.
*
* Override `isSink` and `taintThroughGlobals` as needed, but do not provide
* a characteristic predicate.
*/
class TaintTrackingConfiguration extends TSingleton {
/** Override this to specify which elements are sources in this configuration. */
predicate isSource(Expr source) { exists(getNodeForSource(source)) }
/** Override this to specify which elements are sinks in this configuration. */
abstract predicate isSink(Element e);
/** Override this to specify which expressions are barriers in this configuration. */
predicate isBarrier(Expr e) { nodeIsBarrier(getNodeForExpr(e)) }
/**
* Override this predicate to `any()` to allow taint to flow through global
* variables.
*/
predicate taintThroughGlobals() { none() }
/** Gets a textual representation of this element. */
string toString() { result = "TaintTrackingConfiguration" }
}
private class AdjustedConfiguration extends TaintTracking3::Configuration {
AdjustedConfiguration() { this = "AdjustedConfiguration" }
override predicate isSource(DataFlow::Node source) {
exists(TaintTrackingConfiguration cfg, Expr e |
cfg.isSource(e) and source = getNodeForExpr(e)
)
}
override predicate isSink(DataFlow::Node sink) {
exists(TaintTrackingConfiguration cfg | cfg.isSink(adjustedSink(sink)))
}
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
// Steps into and out of global variables
exists(TaintTrackingConfiguration cfg | cfg.taintThroughGlobals() |
writesVariable(n1.asInstruction(), n2.asVariable().(GlobalOrNamespaceVariable))
or
readsVariable(n2.asInstruction(), n1.asVariable().(GlobalOrNamespaceVariable))
)
or
additionalTaintStep(n1, n2)
}
override predicate isSanitizer(DataFlow::Node node) {
exists(TaintTrackingConfiguration cfg, Expr e | cfg.isBarrier(e) and node = getNodeForExpr(e))
}
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
/*
* A sink `Element` may map to multiple `DataFlowX::PathNode`s via (the
* inverse of) `adjustedSink`. For example, an `Expr` maps to all its
* conversions, and a `Variable` maps to all loads and stores from it. Because
* the path node is part of the tuple that constitutes the alert, this leads
* to duplicate alerts.
*
* To avoid showing duplicates, we edit the graph to replace the final node
* coming from the data-flow library with a node that matches exactly the
* `Element` sink that's requested.
*
* The same is done for sources.
*/
private newtype TPathNode =
TWrapPathNode(DataFlow3::PathNode n) or
// There's a single newtype constructor for both sources and sinks since
// that makes it easiest to deal with the case where source = sink.
TEndpointPathNode(Element e) {
exists(AdjustedConfiguration cfg, DataFlow3::Node sourceNode, DataFlow3::Node sinkNode |
cfg.hasFlow(sourceNode, sinkNode)
|
sourceNode = getNodeForExpr(e) and
exists(TaintTrackingConfiguration ttCfg | ttCfg.isSource(e))
or
e = adjustedSink(sinkNode) and
exists(TaintTrackingConfiguration ttCfg | ttCfg.isSink(e))
)
}
/** An opaque type used for the nodes of a data-flow path. */
class PathNode extends TPathNode {
/** Gets a textual representation of this element. */
string toString() { none() }
/**
* 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://codeql.github.com/docs/writing-codeql-queries/providing-locations-in-codeql-queries/).
*/
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
none()
}
}
/**
* INTERNAL: Do not use.
*/
module Private {
/** Gets a predecessor `PathNode` of `pathNode`, if any. */
PathNode getAPredecessor(PathNode pathNode) { edges(result, pathNode) }
/** Gets the element that `pathNode` wraps, if any. */
Element getElementFromPathNode(PathNode pathNode) {
exists(DataFlow::Node node | node = pathNode.(WrapPathNode).inner().getNode() |
result = node.asInstruction().getAst()
or
result = node.asOperand().getDef().getAst()
)
or
result = pathNode.(EndpointPathNode).inner()
}
}
private class WrapPathNode extends PathNode, TWrapPathNode {
DataFlow3::PathNode inner() { this = TWrapPathNode(result) }
override string toString() { result = this.inner().toString() }
override predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.inner().hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
}
private class EndpointPathNode extends PathNode, TEndpointPathNode {
Expr inner() { this = TEndpointPathNode(result) }
override string toString() { result = this.inner().toString() }
override predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.inner()
.getLocation()
.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
}
/** A PathNode whose `Element` is a source. It may also be a sink. */
private class InitialPathNode extends EndpointPathNode {
InitialPathNode() { exists(TaintTrackingConfiguration cfg | cfg.isSource(this.inner())) }
}
/** A PathNode whose `Element` is a sink. It may also be a source. */
private class FinalPathNode extends EndpointPathNode {
FinalPathNode() { exists(TaintTrackingConfiguration cfg | cfg.isSink(this.inner())) }
}
/** Holds if `(a,b)` is an edge in the graph of data flow path explanations. */
query predicate edges(PathNode a, PathNode b) {
DataFlow3::PathGraph::edges(a.(WrapPathNode).inner(), b.(WrapPathNode).inner())
or
// To avoid showing trivial-looking steps, we _replace_ the last node instead
// of adding an edge out of it.
exists(WrapPathNode sinkNode |
DataFlow3::PathGraph::edges(a.(WrapPathNode).inner(), sinkNode.inner()) and
b.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
or
// Same for the first node
exists(WrapPathNode sourceNode |
DataFlow3::PathGraph::edges(sourceNode.inner(), b.(WrapPathNode).inner()) and
sourceNode.inner().getNode() = getNodeForExpr(a.(InitialPathNode).inner())
)
or
// Finally, handle the case where the path goes directly from a source to a
// sink, meaning that they both need to be translated.
exists(WrapPathNode sinkNode, WrapPathNode sourceNode |
DataFlow3::PathGraph::edges(sourceNode.inner(), sinkNode.inner()) and
sourceNode.inner().getNode() = getNodeForExpr(a.(InitialPathNode).inner()) and
b.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
}
/**
* Holds if there is flow from `arg` to `out` across a call that can by summarized by the flow
* from `par` to `ret` within it, in the graph of data flow path explanations.
*/
query predicate subpaths(PathNode arg, PathNode par, PathNode ret, PathNode out) {
DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), out.(WrapPathNode).inner())
or
// To avoid showing trivial-looking steps, we _replace_ the last node instead
// of adding an edge out of it.
exists(WrapPathNode sinkNode |
DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), sinkNode.inner()) and
out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
or
// Same for the first node
exists(WrapPathNode sourceNode |
DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), out.(WrapPathNode).inner()) and
sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner())
)
or
// Finally, handle the case where the path goes directly from a source to a
// sink, meaning that they both need to be translated.
exists(WrapPathNode sinkNode, WrapPathNode sourceNode |
DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), sinkNode.inner()) and
sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner()) and
out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
}
/** Holds if `n` is a node in the graph of data flow path explanations. */
query predicate nodes(PathNode n, string key, string val) {
key = "semmle.label" and val = n.toString()
}
/**
* Holds if `tainted` may contain taint from `source`, where `sourceNode` and
* `sinkNode` are the corresponding `PathNode`s that can be used in a query
* to provide path explanations. Extend `TaintTrackingConfiguration` to use
* this predicate.
*
* A tainted expression is either directly user input, or is computed from
* user input in a way that users can probably control the exact output of
* the computation.
*/
predicate taintedWithPath(Expr source, Element tainted, PathNode sourceNode, PathNode sinkNode) {
exists(AdjustedConfiguration cfg, DataFlow3::Node flowSource, DataFlow3::Node flowSink |
source = sourceNode.(InitialPathNode).inner() and
flowSource = getNodeForExpr(source) and
cfg.hasFlow(flowSource, flowSink) and
tainted = adjustedSink(flowSink) and
tainted = sinkNode.(FinalPathNode).inner()
)
}
private predicate isGlobalVariablePathNode(WrapPathNode n) {
n.inner().getNode().asVariable() instanceof GlobalOrNamespaceVariable
}
private predicate edgesWithoutGlobals(PathNode a, PathNode b) {
edges(a, b) and
not isGlobalVariablePathNode(a) and
not isGlobalVariablePathNode(b)
}
/**
* Holds if `tainted` can be reached from a taint source without passing
* through a global variable.
*/
predicate taintedWithoutGlobals(Element tainted) {
exists(AdjustedConfiguration cfg, PathNode sourceNode, FinalPathNode sinkNode |
cfg.isSource(sourceNode.(WrapPathNode).inner().getNode()) and
edgesWithoutGlobals+(sourceNode, sinkNode) and
tainted = sinkNode.inner()
)
}
}

View File

@@ -915,57 +915,18 @@ private module Cached {
TDataFlowCallNone() or
TDataFlowCallSome(DataFlowCall call)
cached
newtype TParamNodeOption =
TParamNodeNone() or
TParamNodeSome(ParamNode p)
cached
newtype TReturnCtx =
TReturnCtxNone() or
TReturnCtxNoFlowThrough() or
TReturnCtxMaybeFlowThrough(ReturnPosition pos)
cached
newtype TTypedContentApprox =
MkTypedContentApprox(ContentApprox c, DataFlowType t) {
exists(Content cont |
c = getContentApprox(cont) and
store(_, cont, _, _, t)
)
}
cached
newtype TTypedContent = MkTypedContent(Content c, DataFlowType t) { store(_, c, _, _, t) }
cached
TypedContent getATypedContent(TypedContentApprox c) {
exists(ContentApprox cls, DataFlowType t, Content cont |
c = MkTypedContentApprox(cls, pragma[only_bind_into](t)) and
result = MkTypedContent(cont, pragma[only_bind_into](t)) and
cls = getContentApprox(cont)
)
}
cached
newtype TAccessPathFront =
TFrontNil(DataFlowType t) or
TFrontHead(TypedContent tc)
cached
newtype TApproxAccessPathFront =
TApproxFrontNil(DataFlowType t) or
TApproxFrontHead(TypedContentApprox tc)
cached
newtype TAccessPathFrontOption =
TAccessPathFrontNone() or
TAccessPathFrontSome(AccessPathFront apf)
cached
newtype TApproxAccessPathFrontOption =
TApproxAccessPathFrontNone() or
TApproxAccessPathFrontSome(ApproxAccessPathFront apf)
}
/**
@@ -1343,113 +1304,6 @@ class DataFlowCallOption extends TDataFlowCallOption {
}
}
/** An optional `ParamNode`. */
class ParamNodeOption extends TParamNodeOption {
string toString() {
this = TParamNodeNone() and
result = "(none)"
or
exists(ParamNode p |
this = TParamNodeSome(p) and
result = p.toString()
)
}
}
/**
* A return context used to calculate flow summaries in reverse flow.
*
* The possible values are:
*
* - `TReturnCtxNone()`: no return flow.
* - `TReturnCtxNoFlowThrough()`: return flow, but flow through is not possible.
* - `TReturnCtxMaybeFlowThrough(ReturnPosition pos)`: return flow, of kind `pos`, and
* flow through may be possible.
*/
class ReturnCtx extends TReturnCtx {
string toString() {
this = TReturnCtxNone() and
result = "(none)"
or
this = TReturnCtxNoFlowThrough() and
result = "(no flow through)"
or
exists(ReturnPosition pos |
this = TReturnCtxMaybeFlowThrough(pos) and
result = pos.toString()
)
}
}
/** An approximated `Content` tagged with the type of a containing object. */
class TypedContentApprox extends MkTypedContentApprox {
private ContentApprox c;
private DataFlowType t;
TypedContentApprox() { this = MkTypedContentApprox(c, t) }
/** Gets a typed content approximated by this value. */
TypedContent getATypedContent() { result = getATypedContent(this) }
/** Gets the container type. */
DataFlowType getContainerType() { result = t }
/** Gets a textual representation of this approximated content. */
string toString() { result = c.toString() }
}
/**
* The front of an approximated access path. This is either a head or a nil.
*/
abstract class ApproxAccessPathFront extends TApproxAccessPathFront {
abstract string toString();
abstract DataFlowType getType();
abstract boolean toBoolNonEmpty();
pragma[nomagic]
TypedContent getAHead() {
exists(TypedContentApprox cont |
this = TApproxFrontHead(cont) and
result = cont.getATypedContent()
)
}
}
class ApproxAccessPathFrontNil extends ApproxAccessPathFront, TApproxFrontNil {
private DataFlowType t;
ApproxAccessPathFrontNil() { this = TApproxFrontNil(t) }
override string toString() { result = ppReprType(t) }
override DataFlowType getType() { result = t }
override boolean toBoolNonEmpty() { result = false }
}
class ApproxAccessPathFrontHead extends ApproxAccessPathFront, TApproxFrontHead {
private TypedContentApprox tc;
ApproxAccessPathFrontHead() { this = TApproxFrontHead(tc) }
override string toString() { result = tc.toString() }
override DataFlowType getType() { result = tc.getContainerType() }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional approximated access path front. */
class ApproxAccessPathFrontOption extends TApproxAccessPathFrontOption {
string toString() {
this = TApproxAccessPathFrontNone() and result = "<none>"
or
this = TApproxAccessPathFrontSome(any(ApproxAccessPathFront apf | result = apf.toString()))
}
}
/** A `Content` tagged with the type of a containing object. */
class TypedContent extends MkTypedContent {
private Content c;
@@ -1482,7 +1336,7 @@ abstract class AccessPathFront extends TAccessPathFront {
abstract DataFlowType getType();
abstract ApproxAccessPathFront toApprox();
abstract boolean toBoolNonEmpty();
TypedContent getHead() { this = TFrontHead(result) }
}
@@ -1496,7 +1350,7 @@ class AccessPathFrontNil extends AccessPathFront, TFrontNil {
override DataFlowType getType() { result = t }
override ApproxAccessPathFront toApprox() { result = TApproxFrontNil(t) }
override boolean toBoolNonEmpty() { result = false }
}
class AccessPathFrontHead extends AccessPathFront, TFrontHead {
@@ -1508,7 +1362,7 @@ class AccessPathFrontHead extends AccessPathFront, TFrontHead {
override DataFlowType getType() { result = tc.getContainerType() }
override ApproxAccessPathFront toApprox() { result.getAHead() = tc }
override boolean toBoolNonEmpty() { result = true }
}
/** An optional access path front. */

View File

@@ -244,25 +244,4 @@ module Consistency {
not callable = viableCallable(call) and
not any(ConsistencyConfiguration c).viableImplInCallContextTooLargeExclude(call, ctx, callable)
}
query predicate uniqueParameterNodeAtPosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(Node p0 | isParameterNode(p0, c, pos))) and
msg = "Parameters with overlapping positions."
}
query predicate uniqueParameterNodePosition(
DataFlowCallable c, ParameterPosition pos, Node p, string msg
) {
isParameterNode(p, c, pos) and
not exists(unique(ParameterPosition pos0 | isParameterNode(p, c, pos0))) and
msg = "Parameter node with multiple positions."
}
query predicate uniqueContentApprox(Content c, string msg) {
not exists(unique(ContentApprox approx | approx = getContentApprox(c))) and
msg = "Non-unique content approximation."
}
}

View File

@@ -400,13 +400,6 @@ predicate additionalLambdaFlowStep(Node nodeFrom, Node nodeTo, boolean preserves
*/
predicate allowParameterReturnInSelf(ParameterNode p) { none() }
/** An approximated `Content`. */
class ContentApprox = Unit;
/** Gets an approximated value for content `c`. */
pragma[inline]
ContentApprox getContentApprox(Content c) { any() }
private class MyConsistencyConfiguration extends Consistency::ConsistencyConfiguration {
override predicate argHasPostUpdateExclude(ArgumentNode n) {
// The rules for whether an IR argument gets a post-update node are too

View File

@@ -1,644 +0,0 @@
/**
* INTERNAL: Do not use.
*
* An IR taint tracking library that uses an IR DataFlow configuration to track
* taint from user inputs as defined by `semmle.code.cpp.security.Security`.
*/
import cpp
import semmle.code.cpp.security.Security
private import semmle.code.cpp.ir.dataflow.DataFlow
private import semmle.code.cpp.ir.dataflow.internal.DataFlowUtil
private import semmle.code.cpp.ir.dataflow.DataFlow3
private import semmle.code.cpp.ir.IR
private import semmle.code.cpp.ir.dataflow.ResolveCall
private import semmle.code.cpp.controlflow.IRGuards
private import semmle.code.cpp.models.interfaces.Taint
private import semmle.code.cpp.models.interfaces.DataFlow
private import semmle.code.cpp.ir.dataflow.TaintTracking
private import semmle.code.cpp.ir.dataflow.TaintTracking2
private import semmle.code.cpp.ir.dataflow.TaintTracking3
private import semmle.code.cpp.ir.dataflow.internal.ModelUtil
/**
* A predictable instruction is one where an external user can predict
* the value. For example, a literal in the source code is considered
* predictable.
*/
private predicate predictableInstruction(Instruction instr) {
instr instanceof ConstantInstruction
or
instr instanceof StringConstantInstruction
or
// This could be a conversion on a string literal
predictableInstruction(instr.(UnaryInstruction).getUnary())
}
/**
* Functions that we should only allow taint to flow through (to the return
* value) if all but the source argument are 'predictable'. This is done to
* emulate the old security library's implementation rather than due to any
* strong belief that this is the right approach.
*
* Note that the list itself is not very principled; it consists of all the
* functions listed in the old security library's [default] `isPureFunction`
* that have more than one argument, but are not in the old taint tracking
* library's `returnArgument` predicate.
*/
predicate predictableOnlyFlow(string name) {
name =
[
"strcasestr", "strchnul", "strchr", "strchrnul", "strcmp", "strcspn", "strncmp", "strndup",
"strnlen", "strrchr", "strspn", "strstr", "strtod", "strtof", "strtol", "strtoll", "strtoq",
"strtoul"
]
}
private DataFlow::Node getNodeForSource(Expr source) {
isUserInput(source, _) and
result = getNodeForExpr(source)
}
private DataFlow::Node getNodeForExpr(Expr node) {
result = DataFlow::exprNode(node)
or
// Some of the sources in `isUserInput` are intended to match the value of
// an expression, while others (those modeled below) are intended to match
// the taint that propagates out of an argument, like the `char *` argument
// to `gets`. It's impossible here to tell which is which, but the "access
// to argv" source is definitely not intended to match an output argument,
// and it causes false positives if we let it.
//
// This case goes together with the similar (but not identical) rule in
// `nodeIsBarrierIn`.
result = DataFlow::definitionByReferenceNodeFromArgument(node) and
not argv(node.(VariableAccess).getTarget())
}
private class DefaultTaintTrackingCfg extends TaintTracking::Configuration {
DefaultTaintTrackingCfg() { this = "DefaultTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) { source = getNodeForSource(_) }
override predicate isSink(DataFlow::Node sink) { exists(adjustedSink(sink)) }
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private class ToGlobalVarTaintTrackingCfg extends TaintTracking::Configuration {
ToGlobalVarTaintTrackingCfg() { this = "GlobalVarTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) { source = getNodeForSource(_) }
override predicate isSink(DataFlow::Node sink) {
sink.asVariable() instanceof GlobalOrNamespaceVariable
}
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
writesVariable(n1.asInstruction(), n2.asVariable().(GlobalOrNamespaceVariable))
or
readsVariable(n2.asInstruction(), n1.asVariable().(GlobalOrNamespaceVariable))
}
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private class FromGlobalVarTaintTrackingCfg extends TaintTracking2::Configuration {
FromGlobalVarTaintTrackingCfg() { this = "FromGlobalVarTaintTrackingCfg" }
override predicate isSource(DataFlow::Node source) {
// This set of sources should be reasonably small, which is good for
// performance since the set of sinks is very large.
exists(ToGlobalVarTaintTrackingCfg otherCfg | otherCfg.hasFlowTo(source))
}
override predicate isSink(DataFlow::Node sink) { exists(adjustedSink(sink)) }
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
// Additional step for flow out of variables. There is no flow _into_
// variables in this configuration, so this step only serves to take flow
// out of a variable that's a source.
readsVariable(n2.asInstruction(), n1.asVariable())
}
override predicate isSanitizer(DataFlow::Node node) { nodeIsBarrier(node) }
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
private predicate readsVariable(LoadInstruction load, Variable var) {
load.getSourceAddress().(VariableAddressInstruction).getAstVariable() = var
}
private predicate writesVariable(StoreInstruction store, Variable var) {
store.getDestinationAddress().(VariableAddressInstruction).getAstVariable() = var
}
/**
* A variable that has any kind of upper-bound check anywhere in the program. This is
* biased towards being inclusive because there are a lot of valid ways of doing an
* upper bounds checks if we don't consider where it occurs, for example:
* ```
* if (x < 10) { sink(x); }
*
* if (10 > y) { sink(y); }
*
* if (z > 10) { z = 10; }
* sink(z);
* ```
*/
// TODO: This coarse overapproximation, ported from the old taint tracking
// library, could be replaced with an actual semantic check that a particular
// variable _access_ is guarded by an upper-bound check. We probably don't want
// to do this right away since it could expose a lot of FPs that were
// previously suppressed by this predicate by coincidence.
private predicate hasUpperBoundsCheck(Variable var) {
exists(RelationalOperation oper, VariableAccess access |
oper.getAnOperand() = access and
access.getTarget() = var and
// Comparing to 0 is not an upper bound check
not oper.getAnOperand().getValue() = "0"
)
}
private predicate nodeIsBarrierEqualityCandidate(
DataFlow::Node node, Operand access, Variable checkedVar
) {
readsVariable(node.asInstruction(), checkedVar) and
any(IRGuardCondition guard).ensuresEq(access, _, _, node.asInstruction().getBlock(), true)
}
cached
private module Cached {
cached
predicate nodeIsBarrier(DataFlow::Node node) {
exists(Variable checkedVar |
readsVariable(node.asInstruction(), checkedVar) and
hasUpperBoundsCheck(checkedVar)
)
or
exists(Variable checkedVar, Operand access |
/*
* This node is guarded by a condition that forces the accessed variable
* to equal something else. For example:
* ```
* x = taintsource()
* if (x == 10) {
* taintsink(x); // not considered tainted
* }
* ```
*/
nodeIsBarrierEqualityCandidate(node, access, checkedVar) and
readsVariable(access.getDef(), checkedVar)
)
}
cached
predicate nodeIsBarrierIn(DataFlow::Node node) {
// don't use dataflow into taint sources, as this leads to duplicate results.
exists(Expr source | isUserInput(source, _) |
node = DataFlow::exprNode(source)
or
// This case goes together with the similar (but not identical) rule in
// `getNodeForSource`.
node = DataFlow::definitionByReferenceNodeFromArgument(source)
)
or
// don't use dataflow into binary instructions if both operands are unpredictable
exists(BinaryInstruction iTo |
iTo = node.asInstruction() and
not predictableInstruction(iTo.getLeft()) and
not predictableInstruction(iTo.getRight()) and
// propagate taint from either the pointer or the offset, regardless of predictability
not iTo instanceof PointerArithmeticInstruction
)
or
// don't use dataflow through calls to pure functions if two or more operands
// are unpredictable
exists(Instruction iFrom1, Instruction iFrom2, CallInstruction iTo |
iTo = node.asInstruction() and
isPureFunction(iTo.getStaticCallTarget().getName()) and
iFrom1 = iTo.getAnArgument() and
iFrom2 = iTo.getAnArgument() and
not predictableInstruction(iFrom1) and
not predictableInstruction(iFrom2) and
iFrom1 != iFrom2
)
}
cached
Element adjustedSink(DataFlow::Node sink) {
// TODO: is it more appropriate to use asConvertedExpr here and avoid
// `getConversion*`? Or will that cause us to miss some cases where there's
// flow to a conversion (like a `ReferenceDereferenceExpr`) and we want to
// pretend there was flow to the converted `Expr` for the sake of
// compatibility.
sink.asExpr().getConversion*() = result
or
// For compatibility, send flow from arguments to parameters, even for
// functions with no body.
exists(FunctionCall call, int i |
sink.asExpr() = call.getArgument(pragma[only_bind_into](i)) and
result = resolveCall(call).getParameter(pragma[only_bind_into](i))
)
or
// For compatibility, send flow into a `Variable` if there is flow to any
// Load or Store of that variable.
exists(CopyInstruction copy |
copy.getSourceValue() = sink.asInstruction() and
(
readsVariable(copy, result) or
writesVariable(copy, result)
) and
not hasUpperBoundsCheck(result)
)
or
// For compatibility, send flow into a `NotExpr` even if it's part of a
// short-circuiting condition and thus might get skipped.
result.(NotExpr).getOperand() = sink.asExpr()
or
// Taint postfix and prefix crement operations when their operand is tainted.
result.(CrementOperation).getAnOperand() = sink.asExpr()
or
// Taint `e1 += e2`, `e &= e2` and friends when `e1` or `e2` is tainted.
result.(AssignOperation).getAnOperand() = sink.asExpr()
or
result =
sink.asOperand()
.(SideEffectOperand)
.getUse()
.(ReadSideEffectInstruction)
.getArgumentDef()
.getUnconvertedResultExpression()
}
/**
* Step to return value of a modeled function when an input taints the
* dereference of the return value.
*/
cached
predicate additionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
exists(CallInstruction call, Function func, FunctionInput modelIn, FunctionOutput modelOut |
n1.asOperand() = callInput(call, modelIn) and
(
func.(TaintFunction).hasTaintFlow(modelIn, modelOut)
or
func.(DataFlowFunction).hasDataFlow(modelIn, modelOut)
) and
call.getStaticCallTarget() = func and
modelOut.isReturnValueDeref() and
call = n2.asInstruction()
)
}
}
private import Cached
/**
* Holds if `tainted` may contain taint from `source`.
*
* A tainted expression is either directly user input, or is
* computed from user input in a way that users can probably
* control the exact output of the computation.
*
* This doesn't include data flow through global variables.
* If you need that you must call `taintedIncludingGlobalVars`.
*/
cached
predicate tainted(Expr source, Element tainted) {
exists(DefaultTaintTrackingCfg cfg, DataFlow::Node sink |
cfg.hasFlow(getNodeForSource(source), sink) and
tainted = adjustedSink(sink)
)
}
/**
* Holds if `tainted` may contain taint from `source`, where the taint passed
* through a global variable named `globalVar`.
*
* A tainted expression is either directly user input, or is
* computed from user input in a way that users can probably
* control the exact output of the computation.
*
* This version gives the same results as tainted but also includes
* data flow through global variables.
*
* The parameter `globalVar` is the qualified name of the last global variable
* used to move the value from source to tainted. If the taint did not pass
* through a global variable, then `globalVar = ""`.
*/
cached
predicate taintedIncludingGlobalVars(Expr source, Element tainted, string globalVar) {
tainted(source, tainted) and
globalVar = ""
or
exists(
ToGlobalVarTaintTrackingCfg toCfg, FromGlobalVarTaintTrackingCfg fromCfg,
DataFlow::VariableNode variableNode, GlobalOrNamespaceVariable global, DataFlow::Node sink
|
global = variableNode.getVariable() and
toCfg.hasFlow(getNodeForSource(source), variableNode) and
fromCfg.hasFlow(variableNode, sink) and
tainted = adjustedSink(sink) and
global = globalVarFromId(globalVar)
)
}
/**
* Gets the global variable whose qualified name is `id`. Use this predicate
* together with `taintedIncludingGlobalVars`. Example:
*
* ```
* exists(string varName |
* taintedIncludingGlobalVars(source, tainted, varName) and
* var = globalVarFromId(varName)
* )
* ```
*/
GlobalOrNamespaceVariable globalVarFromId(string id) { id = result.getQualifiedName() }
/**
* Provides definitions for augmenting source/sink pairs with data-flow paths
* between them. From a `@kind path-problem` query, import this module in the
* global scope, extend `TaintTrackingConfiguration`, and use `taintedWithPath`
* in place of `tainted`.
*
* Importing this module will also import the query predicates that contain the
* taint paths.
*/
module TaintedWithPath {
private newtype TSingleton = MkSingleton()
/**
* A taint-tracking configuration that matches sources and sinks in the same
* way as the `tainted` predicate.
*
* Override `isSink` and `taintThroughGlobals` as needed, but do not provide
* a characteristic predicate.
*/
class TaintTrackingConfiguration extends TSingleton {
/** Override this to specify which elements are sources in this configuration. */
predicate isSource(Expr source) { exists(getNodeForSource(source)) }
/** Override this to specify which elements are sinks in this configuration. */
abstract predicate isSink(Element e);
/** Override this to specify which expressions are barriers in this configuration. */
predicate isBarrier(Expr e) { nodeIsBarrier(getNodeForExpr(e)) }
/**
* Override this predicate to `any()` to allow taint to flow through global
* variables.
*/
predicate taintThroughGlobals() { none() }
/** Gets a textual representation of this element. */
string toString() { result = "TaintTrackingConfiguration" }
}
private class AdjustedConfiguration extends TaintTracking3::Configuration {
AdjustedConfiguration() { this = "AdjustedConfiguration" }
override predicate isSource(DataFlow::Node source) {
exists(TaintTrackingConfiguration cfg, Expr e |
cfg.isSource(e) and source = getNodeForExpr(e)
)
}
override predicate isSink(DataFlow::Node sink) {
exists(TaintTrackingConfiguration cfg | cfg.isSink(adjustedSink(sink)))
}
override predicate isAdditionalTaintStep(DataFlow::Node n1, DataFlow::Node n2) {
// Steps into and out of global variables
exists(TaintTrackingConfiguration cfg | cfg.taintThroughGlobals() |
writesVariable(n1.asInstruction(), n2.asVariable().(GlobalOrNamespaceVariable))
or
readsVariable(n2.asInstruction(), n1.asVariable().(GlobalOrNamespaceVariable))
)
or
additionalTaintStep(n1, n2)
}
override predicate isSanitizer(DataFlow::Node node) {
exists(TaintTrackingConfiguration cfg, Expr e | cfg.isBarrier(e) and node = getNodeForExpr(e))
}
override predicate isSanitizerIn(DataFlow::Node node) { nodeIsBarrierIn(node) }
}
/*
* A sink `Element` may map to multiple `DataFlowX::PathNode`s via (the
* inverse of) `adjustedSink`. For example, an `Expr` maps to all its
* conversions, and a `Variable` maps to all loads and stores from it. Because
* the path node is part of the tuple that constitutes the alert, this leads
* to duplicate alerts.
*
* To avoid showing duplicates, we edit the graph to replace the final node
* coming from the data-flow library with a node that matches exactly the
* `Element` sink that's requested.
*
* The same is done for sources.
*/
private newtype TPathNode =
TWrapPathNode(DataFlow3::PathNode n) or
// There's a single newtype constructor for both sources and sinks since
// that makes it easiest to deal with the case where source = sink.
TEndpointPathNode(Element e) {
exists(AdjustedConfiguration cfg, DataFlow3::Node sourceNode, DataFlow3::Node sinkNode |
cfg.hasFlow(sourceNode, sinkNode)
|
sourceNode = getNodeForExpr(e) and
exists(TaintTrackingConfiguration ttCfg | ttCfg.isSource(e))
or
e = adjustedSink(sinkNode) and
exists(TaintTrackingConfiguration ttCfg | ttCfg.isSink(e))
)
}
/** An opaque type used for the nodes of a data-flow path. */
class PathNode extends TPathNode {
/** Gets a textual representation of this element. */
string toString() { none() }
/**
* 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://codeql.github.com/docs/writing-codeql-queries/providing-locations-in-codeql-queries/).
*/
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
none()
}
}
/**
* INTERNAL: Do not use.
*/
module Private {
/** Gets a predecessor `PathNode` of `pathNode`, if any. */
PathNode getAPredecessor(PathNode pathNode) { edges(result, pathNode) }
/** Gets the element that `pathNode` wraps, if any. */
Element getElementFromPathNode(PathNode pathNode) {
exists(DataFlow::Node node | node = pathNode.(WrapPathNode).inner().getNode() |
result = node.asInstruction().getAst()
or
result = node.asOperand().getDef().getAst()
)
or
result = pathNode.(EndpointPathNode).inner()
}
}
private class WrapPathNode extends PathNode, TWrapPathNode {
DataFlow3::PathNode inner() { this = TWrapPathNode(result) }
override string toString() { result = this.inner().toString() }
override predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.inner().hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
}
private class EndpointPathNode extends PathNode, TEndpointPathNode {
Expr inner() { this = TEndpointPathNode(result) }
override string toString() { result = this.inner().toString() }
override predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.inner()
.getLocation()
.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
}
/** A PathNode whose `Element` is a source. It may also be a sink. */
private class InitialPathNode extends EndpointPathNode {
InitialPathNode() { exists(TaintTrackingConfiguration cfg | cfg.isSource(this.inner())) }
}
/** A PathNode whose `Element` is a sink. It may also be a source. */
private class FinalPathNode extends EndpointPathNode {
FinalPathNode() { exists(TaintTrackingConfiguration cfg | cfg.isSink(this.inner())) }
}
/** Holds if `(a,b)` is an edge in the graph of data flow path explanations. */
query predicate edges(PathNode a, PathNode b) {
DataFlow3::PathGraph::edges(a.(WrapPathNode).inner(), b.(WrapPathNode).inner())
or
// To avoid showing trivial-looking steps, we _replace_ the last node instead
// of adding an edge out of it.
exists(WrapPathNode sinkNode |
DataFlow3::PathGraph::edges(a.(WrapPathNode).inner(), sinkNode.inner()) and
b.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
or
// Same for the first node
exists(WrapPathNode sourceNode |
DataFlow3::PathGraph::edges(sourceNode.inner(), b.(WrapPathNode).inner()) and
sourceNode.inner().getNode() = getNodeForExpr(a.(InitialPathNode).inner())
)
or
// Finally, handle the case where the path goes directly from a source to a
// sink, meaning that they both need to be translated.
exists(WrapPathNode sinkNode, WrapPathNode sourceNode |
DataFlow3::PathGraph::edges(sourceNode.inner(), sinkNode.inner()) and
sourceNode.inner().getNode() = getNodeForExpr(a.(InitialPathNode).inner()) and
b.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
}
/**
* Holds if there is flow from `arg` to `out` across a call that can by summarized by the flow
* from `par` to `ret` within it, in the graph of data flow path explanations.
*/
query predicate subpaths(PathNode arg, PathNode par, PathNode ret, PathNode out) {
DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), out.(WrapPathNode).inner())
or
// To avoid showing trivial-looking steps, we _replace_ the last node instead
// of adding an edge out of it.
exists(WrapPathNode sinkNode |
DataFlow3::PathGraph::subpaths(arg.(WrapPathNode).inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), sinkNode.inner()) and
out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
or
// Same for the first node
exists(WrapPathNode sourceNode |
DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), out.(WrapPathNode).inner()) and
sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner())
)
or
// Finally, handle the case where the path goes directly from a source to a
// sink, meaning that they both need to be translated.
exists(WrapPathNode sinkNode, WrapPathNode sourceNode |
DataFlow3::PathGraph::subpaths(sourceNode.inner(), par.(WrapPathNode).inner(),
ret.(WrapPathNode).inner(), sinkNode.inner()) and
sourceNode.inner().getNode() = getNodeForExpr(arg.(InitialPathNode).inner()) and
out.(FinalPathNode).inner() = adjustedSink(sinkNode.inner().getNode())
)
}
/** Holds if `n` is a node in the graph of data flow path explanations. */
query predicate nodes(PathNode n, string key, string val) {
key = "semmle.label" and val = n.toString()
}
/**
* Holds if `tainted` may contain taint from `source`, where `sourceNode` and
* `sinkNode` are the corresponding `PathNode`s that can be used in a query
* to provide path explanations. Extend `TaintTrackingConfiguration` to use
* this predicate.
*
* A tainted expression is either directly user input, or is computed from
* user input in a way that users can probably control the exact output of
* the computation.
*/
predicate taintedWithPath(Expr source, Element tainted, PathNode sourceNode, PathNode sinkNode) {
exists(AdjustedConfiguration cfg, DataFlow3::Node flowSource, DataFlow3::Node flowSink |
source = sourceNode.(InitialPathNode).inner() and
flowSource = getNodeForExpr(source) and
cfg.hasFlow(flowSource, flowSink) and
tainted = adjustedSink(flowSink) and
tainted = sinkNode.(FinalPathNode).inner()
)
}
private predicate isGlobalVariablePathNode(WrapPathNode n) {
n.inner().getNode().asVariable() instanceof GlobalOrNamespaceVariable
}
private predicate edgesWithoutGlobals(PathNode a, PathNode b) {
edges(a, b) and
not isGlobalVariablePathNode(a) and
not isGlobalVariablePathNode(b)
}
/**
* Holds if `tainted` can be reached from a taint source without passing
* through a global variable.
*/
predicate taintedWithoutGlobals(Element tainted) {
exists(AdjustedConfiguration cfg, PathNode sourceNode, FinalPathNode sinkNode |
cfg.isSource(sourceNode.(WrapPathNode).inner().getNode()) and
edgesWithoutGlobals+(sourceNode, sinkNode) and
tainted = sinkNode.inner()
)
}
}

View File

@@ -72,19 +72,7 @@ newtype TInstructionTag =
AsmInputTag(int elementIndex) { exists(AsmStmt asm | exists(asm.getChild(elementIndex))) } or
ThisAddressTag() or
ThisLoadTag() or
StructuredBindingAccessTag() or
// The next three cases handle generation of the constants -1, 0 and 1 for __except handling.
TryExceptGenerateNegativeOne() or
TryExceptGenerateZero() or
TryExceptGenerateOne() or
// The next three cases handle generation of comparisons for __except handling.
TryExceptCompareNegativeOne() or
TryExceptCompareZero() or
TryExceptCompareOne() or
// The next three cases handle generation of branching for __except handling.
TryExceptCompareNegativeOneBranch() or
TryExceptCompareZeroBranch() or
TryExceptCompareOneBranch()
StructuredBindingAccessTag()
class InstructionTag extends TInstructionTag {
final string toString() { result = "Tag" }
@@ -236,22 +224,4 @@ string getInstructionTagId(TInstructionTag tag) {
tag = ThisLoadTag() and result = "ThisLoad"
or
tag = StructuredBindingAccessTag() and result = "StructuredBindingAccess"
or
tag = TryExceptCompareNegativeOne() and result = "TryExceptCompareNegativeOne"
or
tag = TryExceptCompareZero() and result = "TryExceptCompareZero"
or
tag = TryExceptCompareOne() and result = "TryExceptCompareOne"
or
tag = TryExceptGenerateNegativeOne() and result = "TryExceptGenerateNegativeOne"
or
tag = TryExceptGenerateZero() and result = "TryExceptGenerateNegativeOne"
or
tag = TryExceptGenerateOne() and result = "TryExceptGenerateOne"
or
tag = TryExceptCompareNegativeOneBranch() and result = "TryExceptCompareNegativeOneBranch"
or
tag = TryExceptCompareZeroBranch() and result = "TryExceptCompareZeroBranch"
or
tag = TryExceptCompareOneBranch() and result = "TryExceptCompareOneBranch"
}

View File

@@ -675,7 +675,6 @@ newtype TTranslatedElement =
} or
// A statement
TTranslatedStmt(Stmt stmt) { translateStmt(stmt) } or
TTranslatedMicrosoftTryExceptHandler(MicrosoftTryExceptStmt stmt) or
// A function
TTranslatedFunction(Function func) { translateFunction(func) } or
// A constructor init list

View File

@@ -13,222 +13,6 @@ private import TranslatedInitialization
TranslatedStmt getTranslatedStmt(Stmt stmt) { result.getAst() = stmt }
TranslatedMicrosoftTryExceptHandler getTranslatedMicrosoftTryExceptHandler(
MicrosoftTryExceptStmt tryExcept
) {
result.getAst() = tryExcept.getExcept()
}
class TranslatedMicrosoftTryExceptHandler extends TranslatedElement,
TTranslatedMicrosoftTryExceptHandler {
MicrosoftTryExceptStmt tryExcept;
TranslatedMicrosoftTryExceptHandler() { this = TTranslatedMicrosoftTryExceptHandler(tryExcept) }
final override string toString() { result = tryExcept.toString() }
final override Locatable getAst() { result = tryExcept.getExcept() }
override Instruction getFirstInstruction() { result = this.getChild(0).getFirstInstruction() }
override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
// t1 = -1
tag = TryExceptGenerateNegativeOne() and
opcode instanceof Opcode::Constant and
resultType = getIntType()
or
// t2 = cmp t1, condition
tag = TryExceptCompareNegativeOne() and
opcode instanceof Opcode::CompareEQ and
resultType = getBoolType()
or
// if t2 goto ... else goto ...
tag = TryExceptCompareNegativeOneBranch() and
opcode instanceof Opcode::ConditionalBranch and
resultType = getVoidType()
or
// t1 = 0
tag = TryExceptGenerateZero() and
opcode instanceof Opcode::Constant and
resultType = getIntType()
or
// t2 = cmp t1, condition
tag = TryExceptCompareZero() and
opcode instanceof Opcode::CompareEQ and
resultType = getBoolType()
or
// if t2 goto ... else goto ...
tag = TryExceptCompareZeroBranch() and
opcode instanceof Opcode::ConditionalBranch and
resultType = getVoidType()
or
// t1 = 1
tag = TryExceptGenerateOne() and
opcode instanceof Opcode::Constant and
resultType = getIntType()
or
// t2 = cmp t1, condition
tag = TryExceptCompareOne() and
opcode instanceof Opcode::CompareEQ and
resultType = getBoolType()
or
// if t2 goto ... else goto ...
tag = TryExceptCompareOneBranch() and
opcode instanceof Opcode::ConditionalBranch and
resultType = getVoidType()
or
// unwind stack
tag = UnwindTag() and
opcode instanceof Opcode::Unwind and
resultType = getVoidType()
}
final override Instruction getInstructionRegisterOperand(InstructionTag tag, OperandTag operandTag) {
tag = TryExceptCompareNegativeOne() and
(
operandTag instanceof LeftOperandTag and
result = this.getTranslatedCondition().getResult()
or
operandTag instanceof RightOperandTag and
result = this.getInstruction(TryExceptGenerateNegativeOne())
)
or
tag = TryExceptCompareNegativeOneBranch() and
operandTag instanceof ConditionOperandTag and
result = this.getInstruction(TryExceptCompareNegativeOne())
or
tag = TryExceptCompareZero() and
(
operandTag instanceof LeftOperandTag and
result = this.getTranslatedCondition().getResult()
or
operandTag instanceof RightOperandTag and
result = this.getInstruction(TryExceptGenerateZero())
)
or
tag = TryExceptCompareZeroBranch() and
operandTag instanceof ConditionOperandTag and
result = this.getInstruction(TryExceptCompareZero())
or
tag = TryExceptCompareOne() and
(
operandTag instanceof LeftOperandTag and
result = this.getTranslatedCondition().getResult()
or
operandTag instanceof RightOperandTag and
result = this.getInstruction(TryExceptGenerateOne())
)
or
tag = TryExceptCompareOneBranch() and
operandTag instanceof ConditionOperandTag and
result = this.getInstruction(TryExceptCompareOne())
}
override string getInstructionConstantValue(InstructionTag tag) {
tag = TryExceptGenerateNegativeOne() and
result = "-1"
or
tag = TryExceptGenerateZero() and
result = "0"
or
tag = TryExceptGenerateOne() and
result = "1"
}
override Instruction getInstructionSuccessor(InstructionTag tag, EdgeKind kind) {
// Generate -1 -> Compare condition
tag = TryExceptGenerateNegativeOne() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareNegativeOne())
or
// Compare condition -> Branch
tag = TryExceptCompareNegativeOne() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareNegativeOneBranch())
or
// Branch -> Unwind or Generate 0
tag = TryExceptCompareNegativeOneBranch() and
(
kind instanceof TrueEdge and
// TODO: This is not really correct. The semantics of `EXCEPTION_CONTINUE_EXECUTION` is that
// we should continue execution at the point where the exception occurred. But we don't have
// any instruction to model this behavior.
result = this.getInstruction(UnwindTag())
or
kind instanceof FalseEdge and
result = this.getInstruction(TryExceptGenerateZero())
)
or
// Generate 0 -> Compare condition
tag = TryExceptGenerateZero() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareZero())
or
// Compare condition -> Branch
tag = TryExceptCompareZero() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareZeroBranch())
or
// Branch -> Unwind or Generate 1
tag = TryExceptCompareZeroBranch() and
(
kind instanceof TrueEdge and
result = this.getInstruction(UnwindTag())
or
kind instanceof FalseEdge and
result = this.getInstruction(TryExceptGenerateOne())
)
or
// Generate 1 -> Compare condition
tag = TryExceptGenerateOne() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareOne())
or
// Compare condition -> Branch
tag = TryExceptCompareOne() and
kind instanceof GotoEdge and
result = this.getInstruction(TryExceptCompareOneBranch())
or
// Branch -> Handler (the condition value is always 0, -1 or 1, and we've checked for 0 or -1 already.)
tag = TryExceptCompareOneBranch() and
(
kind instanceof TrueEdge and
result = this.getTranslatedHandler().getFirstInstruction()
)
or
// Unwind -> Parent
tag = UnwindTag() and
kind instanceof GotoEdge and
result = this.getParent().getChildSuccessor(this)
}
override Instruction getChildSuccessor(TranslatedElement child) {
child = this.getTranslatedCondition() and
result = this.getInstruction(TryExceptGenerateNegativeOne())
or
child = this.getTranslatedHandler() and
result = this.getParent().getChildSuccessor(this)
}
private TranslatedExpr getTranslatedCondition() {
result = getTranslatedExpr(tryExcept.getCondition())
}
private TranslatedStmt getTranslatedHandler() {
result = getTranslatedStmt(tryExcept.getExcept())
}
override TranslatedElement getChild(int id) {
id = 0 and
result = this.getTranslatedCondition()
or
id = 1 and
result = this.getTranslatedHandler()
}
final override Function getFunction() { result = tryExcept.getEnclosingFunction() }
}
abstract class TranslatedStmt extends TranslatedElement, TTranslatedStmt {
Stmt stmt;
@@ -465,57 +249,15 @@ class TranslatedUnreachableReturnStmt extends TranslatedReturnStmt {
}
/**
* A C/C++ `try` statement, or a `__try __except` or `__try __finally` statement.
*/
private class TryOrMicrosoftTryStmt extends Stmt {
TryOrMicrosoftTryStmt() {
this instanceof TryStmt or
this instanceof MicrosoftTryStmt
}
/** Gets the number of `catch block`s of this statement. */
int getNumberOfCatchClauses() {
result = this.(TryStmt).getNumberOfCatchClauses()
or
this instanceof MicrosoftTryExceptStmt and
result = 1
or
this instanceof MicrosoftTryFinallyStmt and
result = 0
}
/** Gets the `body` statement of this statement. */
Stmt getStmt() {
result = this.(TryStmt).getStmt()
or
result = this.(MicrosoftTryStmt).getStmt()
}
/** Gets the `i`th translated handler of this statement. */
TranslatedElement getTranslatedHandler(int index) {
result = getTranslatedStmt(this.(TryStmt).getChild(index + 1))
or
index = 0 and
result = getTranslatedMicrosoftTryExceptHandler(this)
}
/** Gets the `finally` statement (usually a BlockStmt), if any. */
Stmt getFinally() { result = this.(MicrosoftTryFinallyStmt).getFinally() }
}
/**
* The IR translation of a C++ `try` (or a `__try __except` or `__try __finally`) statement.
* The IR translation of a C++ `try` statement.
*/
class TranslatedTryStmt extends TranslatedStmt {
override TryOrMicrosoftTryStmt stmt;
override TryStmt stmt;
override TranslatedElement getChild(int id) {
id = 0 and result = getBody()
or
result = getHandler(id - 1)
or
id = stmt.getNumberOfCatchClauses() + 1 and
result = this.getFinally()
}
override predicate hasInstruction(Opcode opcode, InstructionTag tag, CppType resultType) {
@@ -527,20 +269,8 @@ class TranslatedTryStmt extends TranslatedStmt {
override Instruction getFirstInstruction() { result = getBody().getFirstInstruction() }
override Instruction getChildSuccessor(TranslatedElement child) {
// All non-finally children go to the successor of the `try` if
// there is no finally block, but if there is a finally block
// then we go to that one.
child = [this.getBody(), this.getHandler(_)] and
(
not exists(this.getFinally()) and
result = this.getParent().getChildSuccessor(this)
or
result = this.getFinally().getFirstInstruction()
)
or
// And after the finally block we go to the successor of the `try`.
child = this.getFinally() and
result = this.getParent().getChildSuccessor(this)
// All children go to the successor of the `try`.
child = getAChild() and result = getParent().getChildSuccessor(this)
}
final Instruction getNextHandler(TranslatedHandler handler) {
@@ -560,9 +290,9 @@ class TranslatedTryStmt extends TranslatedStmt {
result = getHandler(0).getFirstInstruction()
}
private TranslatedElement getHandler(int index) { result = stmt.getTranslatedHandler(index) }
private TranslatedStmt getFinally() { result = getTranslatedStmt(stmt.getFinally()) }
private TranslatedHandler getHandler(int index) {
result = getTranslatedStmt(stmt.getChild(index + 1))
}
private TranslatedStmt getBody() { result = getTranslatedStmt(stmt.getStmt()) }
}

View File

@@ -27,7 +27,7 @@ private import implementations.StdString
private import implementations.Swap
private import implementations.GetDelim
private import implementations.SmartPointer
private import implementations.Scanf
private import implementations.Sscanf
private import implementations.Send
private import implementations.Recv
private import implementations.Accept

View File

@@ -15,6 +15,6 @@ private class Fread extends AliasFunction, RemoteFlowSourceFunction {
override predicate hasRemoteFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(0) and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
}
}

View File

@@ -36,6 +36,6 @@ private class GetDelimFunction extends TaintFunction, AliasFunction, SideEffectF
override predicate hasRemoteFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(0) and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
}
}

View File

@@ -1,19 +1,15 @@
/**
* Provides an implementation class modeling the POSIX function `getenv` and
* various similar functions.
* Provides an implementation class modeling the POSIX function `getenv`.
*/
import cpp
import semmle.code.cpp.models.interfaces.FlowSource
/**
* The POSIX function `getenv`, the GNU function `secure_getenv`, and the
* Windows function `_wgetenv`.
* The POSIX function `getenv`.
*/
class Getenv extends LocalFlowSourceFunction {
Getenv() {
this.hasGlobalOrStdOrBslName("getenv") or this.hasGlobalName(["secure_getenv", "_wgetenv"])
}
Getenv() { this.hasGlobalOrStdOrBslName("getenv") }
override predicate hasLocalFlowSource(FunctionOutput output, string description) {
(

View File

@@ -49,10 +49,10 @@ private class FgetsFunction extends DataFlowFunction, TaintFunction, ArrayFuncti
override predicate hasRemoteFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(0) and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
or
output.isReturnValue() and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
}
override predicate hasArrayWithVariableSize(int bufParam, int countParam) {
@@ -98,10 +98,10 @@ private class GetsFunction extends DataFlowFunction, ArrayFunction, AliasFunctio
override predicate hasLocalFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(0) and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
or
output.isReturnValue() and
description = "string read by " + this.getName()
description = "String read by " + this.getName()
}
override predicate hasArrayWithUnknownSize(int bufParam) { bufParam = 0 }

View File

@@ -1,10 +1,9 @@
import semmle.code.cpp.models.interfaces.Taint
import semmle.code.cpp.models.interfaces.Alias
import semmle.code.cpp.models.interfaces.ArrayFunction
import semmle.code.cpp.models.interfaces.FlowSource
private class InetNtoa extends TaintFunction {
InetNtoa() { this.hasGlobalName("inet_ntoa") }
InetNtoa() { hasGlobalName("inet_ntoa") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameter(0) and
@@ -13,7 +12,7 @@ private class InetNtoa extends TaintFunction {
}
private class InetAton extends TaintFunction, ArrayFunction {
InetAton() { this.hasGlobalName("inet_aton") }
InetAton() { hasGlobalName("inet_aton") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(0) and
@@ -33,7 +32,7 @@ private class InetAton extends TaintFunction, ArrayFunction {
}
private class InetAddr extends TaintFunction, ArrayFunction, AliasFunction {
InetAddr() { this.hasGlobalName("inet_addr") }
InetAddr() { hasGlobalName("inet_addr") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(0) and
@@ -52,7 +51,7 @@ private class InetAddr extends TaintFunction, ArrayFunction, AliasFunction {
}
private class InetNetwork extends TaintFunction, ArrayFunction {
InetNetwork() { this.hasGlobalName("inet_network") }
InetNetwork() { hasGlobalName("inet_network") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(0) and
@@ -65,7 +64,7 @@ private class InetNetwork extends TaintFunction, ArrayFunction {
}
private class InetMakeaddr extends TaintFunction {
InetMakeaddr() { this.hasGlobalName("inet_makeaddr") }
InetMakeaddr() { hasGlobalName("inet_makeaddr") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
(
@@ -77,7 +76,7 @@ private class InetMakeaddr extends TaintFunction {
}
private class InetLnaof extends TaintFunction {
InetLnaof() { this.hasGlobalName("inet_lnaof") }
InetLnaof() { hasGlobalName("inet_lnaof") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameter(0) and
@@ -86,7 +85,7 @@ private class InetLnaof extends TaintFunction {
}
private class InetNetof extends TaintFunction {
InetNetof() { this.hasGlobalName("inet_netof") }
InetNetof() { hasGlobalName("inet_netof") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameter(0) and
@@ -95,7 +94,7 @@ private class InetNetof extends TaintFunction {
}
private class InetPton extends TaintFunction, ArrayFunction {
InetPton() { this.hasGlobalName("inet_pton") }
InetPton() { hasGlobalName("inet_pton") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
(
@@ -115,7 +114,7 @@ private class InetPton extends TaintFunction, ArrayFunction {
}
private class Gethostbyname extends TaintFunction, ArrayFunction {
Gethostbyname() { this.hasGlobalName("gethostbyname") }
Gethostbyname() { hasGlobalName("gethostbyname") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(0) and
@@ -128,7 +127,7 @@ private class Gethostbyname extends TaintFunction, ArrayFunction {
}
private class Gethostbyaddr extends TaintFunction, ArrayFunction {
Gethostbyaddr() { this.hasGlobalName("gethostbyaddr") }
Gethostbyaddr() { hasGlobalName("gethostbyaddr") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
(
@@ -143,21 +142,3 @@ private class Gethostbyaddr extends TaintFunction, ArrayFunction {
override predicate hasArrayWithNullTerminator(int bufParam) { bufParam = 0 }
}
private class Getaddrinfo extends TaintFunction, ArrayFunction, RemoteFlowSourceFunction {
Getaddrinfo() { this.hasGlobalName("getaddrinfo") }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref([0 .. 2]) and
output.isParameterDeref(3)
}
override predicate hasArrayInput(int bufParam) { bufParam in [0, 1] }
override predicate hasArrayWithNullTerminator(int bufParam) { bufParam in [0, 1] }
override predicate hasRemoteFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(3) and
description = "address returned by " + this.getName()
}
}

View File

@@ -31,17 +31,7 @@ private class IteratorTraits extends Class {
* `std::iterator_traits` instantiation for it.
*/
private class IteratorByTraits extends Iterator {
IteratorTraits trait;
IteratorByTraits() { trait.getIteratorType() = this }
override Type getValueType() {
exists(TypedefType t |
trait.getAMember() = t and
t.getName() = "value_type" and
result = t.getUnderlyingType()
)
}
IteratorByTraits() { exists(IteratorTraits it | it.getIteratorType() = this) }
}
/**
@@ -52,27 +42,20 @@ private class IteratorByTraits extends Iterator {
*/
private class IteratorByPointer extends Iterator instanceof PointerType {
IteratorByPointer() { not this instanceof IteratorByTraits }
override Type getValueType() { result = super.getBaseType() }
}
/**
* A type which has the typedefs expected for an iterator.
*/
private class IteratorByTypedefs extends Iterator, Class {
TypedefType valueType;
IteratorByTypedefs() {
this.getAMember().(TypedefType).hasName("difference_type") and
valueType = this.getAMember() and
valueType.hasName("value_type") and
this.getAMember().(TypedefType).hasName("value_type") and
this.getAMember().(TypedefType).hasName("pointer") and
this.getAMember().(TypedefType).hasName("reference") and
this.getAMember().(TypedefType).hasName("iterator_category") and
not this.hasQualifiedName(["std", "bsl"], "iterator_traits")
}
override Type getValueType() { result = valueType.getUnderlyingType() }
}
/**
@@ -80,8 +63,6 @@ private class IteratorByTypedefs extends Iterator, Class {
*/
private class StdIterator extends Iterator, Class {
StdIterator() { this.hasQualifiedName(["std", "bsl"], "iterator") }
override Type getValueType() { result = this.getTemplateArgument(1).(Type).getUnderlyingType() }
}
/**
@@ -185,15 +166,12 @@ private class IteratorSubOperator extends Operator, TaintFunction {
/**
* A non-member `operator+=` or `operator-=` function for an iterator type.
*/
class IteratorAssignArithmeticOperator extends Operator {
private class IteratorAssignArithmeticOperator extends Operator, DataFlowFunction, TaintFunction {
IteratorAssignArithmeticOperator() {
this.hasName(["operator+=", "operator-="]) and
exists(getIteratorArgumentInput(this, 0))
}
}
private class IteratorAssignArithmeticOperatorModel extends IteratorAssignArithmeticOperator,
DataFlowFunction, TaintFunction {
override predicate hasDataFlow(FunctionInput input, FunctionOutput output) {
input.isParameter(0) and
output.isReturnValue()
@@ -232,14 +210,11 @@ class IteratorPointerDereferenceMemberOperator extends MemberFunction, TaintFunc
/**
* An `operator++` or `operator--` member function for an iterator type.
*/
class IteratorCrementMemberOperator extends MemberFunction {
private class IteratorCrementMemberOperator extends MemberFunction, DataFlowFunction, TaintFunction {
IteratorCrementMemberOperator() {
this.getClassAndName(["operator++", "operator--"]) instanceof Iterator
}
}
private class IteratorCrementMemberOperatorModel extends IteratorCrementMemberOperator,
DataFlowFunction, TaintFunction {
override predicate hasDataFlow(FunctionInput input, FunctionOutput output) {
input.isQualifierAddress() and
output.isReturnValue()

View File

@@ -83,7 +83,7 @@ private class Recv extends AliasFunction, ArrayFunction, SideEffectFunction,
or
this.hasGlobalName("recvfrom") and output.isParameterDeref([4, 5])
) and
description = "buffer read by " + this.getName()
description = "Buffer read by " + this.getName()
}
override predicate hasSocketInput(FunctionInput input) { input.isParameter(0) }

View File

@@ -58,7 +58,7 @@ private class Send extends AliasFunction, ArrayFunction, SideEffectFunction, Rem
override ParameterIndex getParameterSizeIndex(ParameterIndex i) { i = 1 and result = 2 }
override predicate hasRemoteFlowSink(FunctionInput input, string description) {
input.isParameterDeref(1) and description = "buffer sent by " + this.getName()
input.isParameterDeref(1) and description = "Buffer sent by " + this.getName()
}
override predicate hasSocketInput(FunctionInput input) { input.isParameter(0) }

View File

@@ -1,6 +1,6 @@
/**
* Provides implementation classes modeling the `scanf` family of functions.
* See `semmle.code.cpp.models.Models` for usage information.
* Provides implementation classes modeling `sscanf`, `fscanf` and various similar
* functions. See `semmle.code.cpp.models.Models` for usage information.
*/
import semmle.code.cpp.Function
@@ -9,15 +9,18 @@ import semmle.code.cpp.models.interfaces.ArrayFunction
import semmle.code.cpp.models.interfaces.Taint
import semmle.code.cpp.models.interfaces.Alias
import semmle.code.cpp.models.interfaces.SideEffect
import semmle.code.cpp.models.interfaces.FlowSource
/**
* The `scanf` family of functions.
* The standard function `sscanf`, `fscanf` and its assorted variants
*/
abstract private class ScanfFunctionModel extends ArrayFunction, TaintFunction, AliasFunction,
SideEffectFunction {
private class SscanfModel extends ArrayFunction, TaintFunction, AliasFunction, SideEffectFunction {
SscanfModel() { this instanceof Sscanf or this instanceof Fscanf or this instanceof Snscanf }
override predicate hasArrayWithNullTerminator(int bufParam) {
bufParam = this.(ScanfFunction).getFormatParameterIndex()
or
not this instanceof Fscanf and
bufParam = this.(ScanfFunction).getInputParameterIndex()
}
override predicate hasArrayInput(int bufParam) { this.hasArrayWithNullTerminator(bufParam) }
@@ -33,7 +36,7 @@ abstract private class ScanfFunctionModel extends ArrayFunction, TaintFunction,
)
}
int getArgsStartPosition() { result = this.getNumberOfParameters() }
private int getArgsStartPosition() { result = this.getNumberOfParameters() }
override predicate hasTaintFlow(FunctionInput input, FunctionOutput output) {
input.isParameterDeref(this.(ScanfFunction).getInputParameterIndex()) and
@@ -67,36 +70,3 @@ abstract private class ScanfFunctionModel extends ArrayFunction, TaintFunction,
]
}
}
/**
* The standard function `scanf` and its assorted variants
*/
private class ScanfModel extends ScanfFunctionModel, LocalFlowSourceFunction instanceof Scanf {
override predicate hasLocalFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(any(int i | i >= this.getArgsStartPosition())) and
description = "value read by " + this.getName()
}
}
/**
* The standard function `fscanf` and its assorted variants
*/
private class FscanfModel extends ScanfFunctionModel, RemoteFlowSourceFunction instanceof Fscanf {
override predicate hasRemoteFlowSource(FunctionOutput output, string description) {
output.isParameterDeref(any(int i | i >= this.getArgsStartPosition())) and
description = "value read by " + this.getName()
}
}
/**
* The standard function `sscanf` and its assorted variants
*/
private class SscanfModel extends ScanfFunctionModel {
SscanfModel() { this instanceof Sscanf or this instanceof Snscanf }
override predicate hasArrayWithNullTerminator(int bufParam) {
super.hasArrayWithNullTerminator(bufParam)
or
bufParam = this.(ScanfFunction).getInputParameterIndex()
}
}

View File

@@ -5,53 +5,38 @@
import semmle.code.cpp.models.interfaces.Taint
import semmle.code.cpp.models.interfaces.Iterator
/**
* A sequence container template class (for example, `std::vector`) from the
* standard library.
*/
abstract class StdSequenceContainer extends Class {
Type getElementType() { result = this.getTemplateArgument(0) }
}
/**
* The `std::array` template class.
*/
private class Array extends StdSequenceContainer {
private class Array extends Class {
Array() { this.hasQualifiedName(["std", "bsl"], "array") }
}
/**
* The `std::string` template class.
*/
private class String extends StdSequenceContainer {
String() { this.hasQualifiedName(["std", "bsl"], "basic_string") }
}
/**
* The `std::deque` template class.
*/
private class Deque extends StdSequenceContainer {
private class Deque extends Class {
Deque() { this.hasQualifiedName(["std", "bsl"], "deque") }
}
/**
* The `std::forward_list` template class.
*/
private class ForwardList extends StdSequenceContainer {
private class ForwardList extends Class {
ForwardList() { this.hasQualifiedName(["std", "bsl"], "forward_list") }
}
/**
* The `std::list` template class.
*/
private class List extends StdSequenceContainer {
private class List extends Class {
List() { this.hasQualifiedName(["std", "bsl"], "list") }
}
/**
* The `std::vector` template class.
*/
private class Vector extends StdSequenceContainer {
private class Vector extends Class {
Vector() { this.hasQualifiedName(["std", "bsl"], "vector") }
}

View File

@@ -17,15 +17,6 @@ private class StdBasicString extends ClassTemplateInstantiation {
/**
* The `std::basic_string::iterator` declaration.
*
* Intuitively, this class shouldn't be necessary as it's already captured
* by the `StdIterator` class. However, this class ensures that the typedef inside the
* body of the `std::string` class is also seen as an iterator.
*
* Eventually, we should be consistent about which of the following should be recognized as iterators:
* 1. The typedef type.
* 2. The template class of the resolved type.
* 3. Any instantiation of the resolved type.
*/
private class StdBasicStringIterator extends Iterator, Type {
StdBasicStringIterator() {

View File

@@ -11,6 +11,38 @@
import semmle.code.cpp.Function
import semmle.code.cpp.models.Models
/**
* An allocation function such as `malloc`.
*/
abstract class AllocationFunction extends Function {
/**
* Gets the index of the argument for the allocation size, if any. The actual
* allocation size is the value of this argument multiplied by the result of
* `getSizeMult()`, in bytes.
*/
int getSizeArg() { none() }
/**
* Gets the index of an argument that multiplies the allocation size given by
* `getSizeArg`, if any.
*/
int getSizeMult() { none() }
/**
* Gets the index of the input pointer argument to be reallocated, if this
* is a `realloc` function.
*/
int getReallocPtrArg() { none() }
/**
* Whether or not this allocation requires a corresponding deallocation of
* some sort (most do, but `alloca` for example does not). If it is unclear,
* we default to no (for example a placement `new` allocation may or may not
* require a corresponding `delete`).
*/
predicate requiresDealloc() { any() }
}
/**
* An allocation expression such as call to `malloc` or a `new` expression.
*/
@@ -54,41 +86,6 @@ abstract class AllocationExpr extends Expr {
predicate requiresDealloc() { any() }
}
/**
* An allocation function such as `malloc`.
*
* Note: `AllocationExpr` includes calls to allocation functions, so prefer
* to use that class unless you specifically need to reason about functions.
*/
abstract class AllocationFunction extends Function {
/**
* Gets the index of the argument for the allocation size, if any. The actual
* allocation size is the value of this argument multiplied by the result of
* `getSizeMult()`, in bytes.
*/
int getSizeArg() { none() }
/**
* Gets the index of an argument that multiplies the allocation size given by
* `getSizeArg`, if any.
*/
int getSizeMult() { none() }
/**
* Gets the index of the input pointer argument to be reallocated, if this
* is a `realloc` function.
*/
int getReallocPtrArg() { none() }
/**
* Whether or not this allocation requires a corresponding deallocation of
* some sort (most do, but `alloca` for example does not). If it is unclear,
* we default to no (for example a placement `new` allocation may or may not
* require a corresponding `delete`).
*/
predicate requiresDealloc() { any() }
}
/**
* An `operator new` or `operator new[]` function that may be associated with
* `new` or `new[]` expressions. Note that `new` and `new[]` are not function

View File

@@ -11,6 +11,16 @@
import semmle.code.cpp.Function
import semmle.code.cpp.models.Models
/**
* A deallocation function such as `free`.
*/
abstract class DeallocationFunction extends Function {
/**
* Gets the index of the argument that is freed by this function.
*/
int getFreedArg() { none() }
}
/**
* An deallocation expression such as call to `free` or a `delete` expression.
*/
@@ -21,19 +31,6 @@ abstract class DeallocationExpr extends Expr {
Expr getFreedExpr() { none() }
}
/**
* A deallocation function such as `free`.
*
* Note: `DeallocationExpr` includes calls to deallocation functions, so prefer
* to use that class unless you specifically need to reason about functions.
*/
abstract class DeallocationFunction extends Function {
/**
* Gets the index of the argument that is freed by this function.
*/
int getFreedArg() { none() }
}
/**
* An `operator delete` or `operator delete[]` function that may be associated
* with `delete` or `delete[]` expressions. Note that `delete` and `delete[]`

View File

@@ -29,17 +29,5 @@ abstract class GetIteratorFunction extends Function {
/**
* A type which can be used as an iterator.
*
* Note: Do _not_ `extend` when inheriting from this class in queries. Always use `instanceof`:
* ```
* class MyIterator instanceof Iterator { ... }
* ```
*/
abstract class Iterator extends Type {
/**
* Gets the value type of this iterator, if any.
*
* For example, the value type of a `std::vector<int>::iterator` is `int`.
*/
Type getValueType() { none() }
}
abstract class Iterator extends Type { }

View File

@@ -89,10 +89,10 @@ private class LocalParameterSource extends LocalFlowSource {
private class ArgvSource extends LocalFlowSource {
ArgvSource() {
exists(Function main, Parameter argv |
main.hasGlobalName("main") and
main.getParameter(1) = argv and
this.asParameter() = argv
exists(Parameter argv |
argv.hasName("argv") and
argv.getFunction().hasGlobalName("main") and
this.asExpr() = argv.getAnAccess()
)
}

View File

@@ -121,9 +121,7 @@ private predicate moveToDependingOnSide(Expr src, Expr dest) {
* (this is done to avoid false positives). Because of this we need to track if the tainted element came from an argument
* or not, and for that we use destFromArg
*/
deprecated private predicate betweenFunctionsValueMoveTo(
Element src, Element dest, boolean destFromArg
) {
private predicate betweenFunctionsValueMoveTo(Element src, Element dest, boolean destFromArg) {
not unreachable(src) and
not unreachable(dest) and
(
@@ -164,13 +162,13 @@ deprecated private predicate betweenFunctionsValueMoveTo(
// predicate folding for proper join-order
// bad magic: pushes down predicate that ruins join-order
pragma[nomagic]
deprecated private predicate resolveCallWithParam(Call call, Function called, int i, Parameter p) {
private predicate resolveCallWithParam(Call call, Function called, int i, Parameter p) {
called = resolveCall(call) and
p = called.getParameter(i)
}
/** A variable for which flow through is allowed. */
deprecated library class FlowVariable extends Variable {
library class FlowVariable extends Variable {
FlowVariable() {
(
this instanceof LocalScopeVariable or
@@ -181,11 +179,11 @@ deprecated library class FlowVariable extends Variable {
}
/** A local scope variable for which flow through is allowed. */
deprecated library class FlowLocalScopeVariable extends Variable {
library class FlowLocalScopeVariable extends Variable {
FlowLocalScopeVariable() { this instanceof LocalScopeVariable }
}
deprecated private predicate insideFunctionValueMoveTo(Element src, Element dest) {
private predicate insideFunctionValueMoveTo(Element src, Element dest) {
not unreachable(src) and
not unreachable(dest) and
(
@@ -326,7 +324,7 @@ private predicate unionAccess(Variable v, Field f, FieldAccess a) {
a.getQualifier() = v.getAnAccess()
}
deprecated GlobalOrNamespaceVariable globalVarFromId(string id) {
GlobalOrNamespaceVariable globalVarFromId(string id) {
if result instanceof NamespaceVariable
then id = result.getNamespace() + "::" + result.getName()
else id = result.getName()
@@ -355,7 +353,7 @@ private predicate hasUpperBoundsCheck(Variable var) {
}
cached
deprecated private predicate taintedWithArgsAndGlobalVars(
private predicate taintedWithArgsAndGlobalVars(
Element src, Element dest, boolean destFromArg, string globalVar
) {
isUserInput(src, _) and
@@ -397,7 +395,7 @@ deprecated private predicate taintedWithArgsAndGlobalVars(
* This doesn't include data flow through global variables.
* If you need that you must call taintedIncludingGlobalVars.
*/
deprecated predicate tainted(Expr source, Element tainted) {
predicate tainted(Expr source, Element tainted) {
taintedWithArgsAndGlobalVars(source, tainted, _, "")
}
@@ -412,7 +410,7 @@ deprecated predicate tainted(Expr source, Element tainted) {
* The parameter `globalVar` is the name of the last global variable used to move the
* value from source to tainted.
*/
deprecated predicate taintedIncludingGlobalVars(Expr source, Element tainted, string globalVar) {
predicate taintedIncludingGlobalVars(Expr source, Element tainted, string globalVar) {
taintedWithArgsAndGlobalVars(source, tainted, _, globalVar)
}
@@ -543,14 +541,14 @@ private predicate returnArgument(Function f, int sourceArg) {
* targets a virtual method, simple data flow analysis is performed
* in order to identify target(s).
*/
deprecated Function resolveCall(Call call) {
Function resolveCall(Call call) {
result = call.getTarget()
or
result = call.(DataSensitiveCallExpr).resolve()
}
/** A data sensitive call expression. */
abstract deprecated library class DataSensitiveCallExpr extends Expr {
abstract library class DataSensitiveCallExpr extends Expr {
DataSensitiveCallExpr() { not unreachable(this) }
abstract Expr getSrc();
@@ -581,7 +579,7 @@ abstract deprecated library class DataSensitiveCallExpr extends Expr {
}
/** Call through a function pointer. */
deprecated library class DataSensitiveExprCall extends DataSensitiveCallExpr, ExprCall {
library class DataSensitiveExprCall extends DataSensitiveCallExpr, ExprCall {
override Expr getSrc() { result = getExpr() }
override Function resolve() {
@@ -590,8 +588,7 @@ deprecated library class DataSensitiveExprCall extends DataSensitiveCallExpr, Ex
}
/** Call to a virtual function. */
deprecated library class DataSensitiveOverriddenFunctionCall extends DataSensitiveCallExpr,
FunctionCall {
library class DataSensitiveOverriddenFunctionCall extends DataSensitiveCallExpr, FunctionCall {
DataSensitiveOverriddenFunctionCall() {
exists(getTarget().(VirtualFunction).getAnOverridingFunction())
}

View File

@@ -183,7 +183,7 @@ private newtype GvnBase =
// global variable will only get the same value number if they are
// guaranteed to have the same value.
GVN_OtherVariable(Variable x, ControlFlowNode dominator) { mk_OtherVariable(x, dominator, _) } or
deprecated GVN_FieldAccess(GVN s, Field f) {
GVN_FieldAccess(GVN s, Field f) {
mk_DotFieldAccess(s, f, _) or
mk_PointerFieldAccess_with_deref(s, f, _) or
mk_ImplicitThisFieldAccess_with_deref(s, f, _)
@@ -192,7 +192,7 @@ private newtype GvnBase =
// time the pointer was dereferenced, so we need to include a definition
// location. As a crude (but safe) approximation, we use
// `mostRecentSideEffect` to compute a definition location.
deprecated GVN_Deref(GVN p, ControlFlowNode dominator) {
GVN_Deref(GVN p, ControlFlowNode dominator) {
mk_Deref(p, dominator, _) or
mk_PointerFieldAccess(p, _, dominator, _) or
mk_ImplicitThisFieldAccess_with_qualifier(p, _, dominator, _)
@@ -201,12 +201,10 @@ private newtype GvnBase =
mk_ThisExpr(fcn, _) or
mk_ImplicitThisFieldAccess(fcn, _, _, _)
} or
deprecated GVN_Conversion(Type t, GVN child) { mk_Conversion(t, child, _) } or
deprecated GVN_BinaryOp(GVN lhs, GVN rhs, string opname) { mk_BinaryOp(lhs, rhs, opname, _) } or
deprecated GVN_UnaryOp(GVN child, string opname) { mk_UnaryOp(child, opname, _) } or
deprecated GVN_ArrayAccess(GVN x, GVN i, ControlFlowNode dominator) {
mk_ArrayAccess(x, i, dominator, _)
} or
GVN_Conversion(Type t, GVN child) { mk_Conversion(t, child, _) } or
GVN_BinaryOp(GVN lhs, GVN rhs, string opname) { mk_BinaryOp(lhs, rhs, opname, _) } or
GVN_UnaryOp(GVN child, string opname) { mk_UnaryOp(child, opname, _) } or
GVN_ArrayAccess(GVN x, GVN i, ControlFlowNode dominator) { mk_ArrayAccess(x, i, dominator, _) } or
// Any expression that is not handled by the cases above is
// given a unique number based on the expression itself.
GVN_Unanalyzable(Expr e) { not analyzableExpr(e) }
@@ -342,7 +340,7 @@ private predicate analyzableDotFieldAccess(DotFieldAccess access) {
not analyzableConst(access)
}
deprecated private predicate mk_DotFieldAccess(GVN qualifier, Field target, DotFieldAccess access) {
private predicate mk_DotFieldAccess(GVN qualifier, Field target, DotFieldAccess access) {
analyzableDotFieldAccess(access) and
target = access.getTarget() and
qualifier = globalValueNumber(access.getQualifier().getFullyConverted())
@@ -355,7 +353,7 @@ private predicate analyzablePointerFieldAccess(PointerFieldAccess access) {
not analyzableConst(access)
}
deprecated private predicate mk_PointerFieldAccess(
private predicate mk_PointerFieldAccess(
GVN qualifier, Field target, ControlFlowNode dominator, PointerFieldAccess access
) {
analyzablePointerFieldAccess(access) and
@@ -368,7 +366,7 @@ deprecated private predicate mk_PointerFieldAccess(
* `obj->field` is equivalent to `(*obj).field`, so we need to wrap an
* extra `GVN_Deref` around the qualifier.
*/
deprecated private predicate mk_PointerFieldAccess_with_deref(
private predicate mk_PointerFieldAccess_with_deref(
GVN new_qualifier, Field target, PointerFieldAccess access
) {
exists(GVN qualifier, ControlFlowNode dominator |
@@ -393,7 +391,7 @@ private predicate mk_ImplicitThisFieldAccess(
fcn = access.getEnclosingFunction()
}
deprecated private predicate mk_ImplicitThisFieldAccess_with_qualifier(
private predicate mk_ImplicitThisFieldAccess_with_qualifier(
GVN qualifier, Field target, ControlFlowNode dominator, ImplicitThisFieldAccess access
) {
exists(Function fcn |
@@ -402,7 +400,7 @@ deprecated private predicate mk_ImplicitThisFieldAccess_with_qualifier(
)
}
deprecated private predicate mk_ImplicitThisFieldAccess_with_deref(
private predicate mk_ImplicitThisFieldAccess_with_deref(
GVN new_qualifier, Field target, ImplicitThisFieldAccess access
) {
exists(GVN qualifier, ControlFlowNode dominator |
@@ -436,7 +434,7 @@ private predicate analyzableConversion(Conversion conv) {
not analyzableConst(conv)
}
deprecated private predicate mk_Conversion(Type t, GVN child, Conversion conv) {
private predicate mk_Conversion(Type t, GVN child, Conversion conv) {
analyzableConversion(conv) and
t = conv.getUnspecifiedType() and
child = globalValueNumber(conv.getExpr())
@@ -450,7 +448,7 @@ private predicate analyzableBinaryOp(BinaryOperation op) {
not analyzableConst(op)
}
deprecated private predicate mk_BinaryOp(GVN lhs, GVN rhs, string opname, BinaryOperation op) {
private predicate mk_BinaryOp(GVN lhs, GVN rhs, string opname, BinaryOperation op) {
analyzableBinaryOp(op) and
lhs = globalValueNumber(op.getLeftOperand().getFullyConverted()) and
rhs = globalValueNumber(op.getRightOperand().getFullyConverted()) and
@@ -465,7 +463,7 @@ private predicate analyzableUnaryOp(UnaryOperation op) {
not analyzableConst(op)
}
deprecated private predicate mk_UnaryOp(GVN child, string opname, UnaryOperation op) {
private predicate mk_UnaryOp(GVN child, string opname, UnaryOperation op) {
analyzableUnaryOp(op) and
child = globalValueNumber(op.getOperand().getFullyConverted()) and
opname = op.getOperator()
@@ -488,9 +486,7 @@ private predicate analyzableArrayAccess(ArrayExpr ae) {
not analyzableConst(ae)
}
deprecated private predicate mk_ArrayAccess(
GVN base, GVN offset, ControlFlowNode dominator, ArrayExpr ae
) {
private predicate mk_ArrayAccess(GVN base, GVN offset, ControlFlowNode dominator, ArrayExpr ae) {
analyzableArrayAccess(ae) and
base = globalValueNumber(ae.getArrayBase().getFullyConverted()) and
offset = globalValueNumber(ae.getArrayOffset().getFullyConverted()) and
@@ -503,7 +499,7 @@ private predicate analyzablePointerDereferenceExpr(PointerDereferenceExpr deref)
not analyzableConst(deref)
}
deprecated private predicate mk_Deref(GVN p, ControlFlowNode dominator, PointerDereferenceExpr deref) {
private predicate mk_Deref(GVN p, ControlFlowNode dominator, PointerDereferenceExpr deref) {
analyzablePointerDereferenceExpr(deref) and
p = globalValueNumber(deref.getOperand().getFullyConverted()) and
dominator = mostRecentSideEffect(deref)

View File

@@ -4,11 +4,6 @@
* Note: Data is usually stored in a separate database and the QL libraries only contain predicates,
* but for this tutorial both the data and the predicates are stored in the library.
*/
/**
* A person known to the QL detective tutorials, represented by a string
* (their first name).
*/
class Person extends string {
Person() {
this =

View File

@@ -5,37 +5,78 @@
* @id cpp/alert-suppression
*/
private import codeql.util.suppression.AlertSuppression as AS
private import semmle.code.cpp.Element
import cpp
class AstNode extends Locatable {
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
this.getLocation().hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
}
}
/**
* An alert suppression comment.
*/
class SuppressionComment extends Comment {
string annotation;
string text;
class SingleLineComment extends Comment, AstNode {
private string text;
SingleLineComment() {
this instanceof CppStyleComment and
// strip the beginning slashes
text = this.getContents().suffix(2)
or
this instanceof CStyleComment and
// strip both the beginning /* and the end */ the comment
exists(string text0 |
text0 = this.getContents().suffix(2) and
text = text0.prefix(text0.length() - 2)
SuppressionComment() {
(
this instanceof CppStyleComment and
// strip the beginning slashes
text = this.getContents().suffix(2)
or
this instanceof CStyleComment and
// strip both the beginning /* and the end */ the comment
exists(string text0 |
text0 = this.getContents().suffix(2) and
text = text0.prefix(text0.length() - 2)
) and
// The /* */ comment must be a single-line comment
not text.matches("%\n%")
) and
// The /* */ comment must be a single-line comment
not text.matches("%\n%")
(
// match `lgtm[...]` anywhere in the comment
annotation = text.regexpFind("(?i)\\blgtm\\s*\\[[^\\]]*\\]", _, _)
or
// match `lgtm` at the start of the comment and after semicolon
annotation = text.regexpFind("(?i)(?<=^|;)\\s*lgtm(?!\\B|\\s*\\[)", _, _).trim()
)
}
/** Gets the text in this comment, excluding the leading //. */
string getText() { result = text }
/** Gets the suppression annotation in this comment. */
string getAnnotation() { result = annotation }
/**
* Holds if this comment applies to the range from column `startcolumn` of line `startline`
* to column `endcolumn` of line `endline` in file `filepath`.
*/
predicate covers(string filepath, int startline, int startcolumn, int endline, int endcolumn) {
this.getLocation().hasLocationInfo(filepath, startline, _, endline, endcolumn) and
startcolumn = 1
}
/** Gets the scope of this suppression. */
SuppressionScope getScope() { result = this }
}
import AS::Make<AstNode, SingleLineComment>
/**
* The scope of an alert suppression comment.
*/
class SuppressionScope extends ElementBase instanceof SuppressionComment {
/**
* 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://codeql.github.com/docs/writing-codeql-queries/providing-locations-in-codeql-queries/).
*/
predicate hasLocationInfo(
string filepath, int startline, int startcolumn, int endline, int endcolumn
) {
super.covers(filepath, startline, startcolumn, endline, endcolumn)
}
}
from SuppressionComment c
select c, // suppression comment
c.getText(), // text of suppression comment (excluding delimiters)
c.getAnnotation(), // text of suppression annotation
c.getScope() // scope of suppression

View File

@@ -15,7 +15,6 @@
import cpp
pragma[nomagic]
predicate beforeArrayAccess(Variable v, ArrayExpr access, Expr before) {
exists(LogicalAndExpr andexpr |
access.getArrayOffset() = v.getAnAccess() and
@@ -24,7 +23,6 @@ predicate beforeArrayAccess(Variable v, ArrayExpr access, Expr before) {
)
}
pragma[nomagic]
predicate afterArrayAccess(Variable v, ArrayExpr access, Expr after) {
exists(LogicalAndExpr andexpr |
access.getArrayOffset() = v.getAnAccess() and

View File

@@ -1,18 +1,3 @@
## 0.5.0
### Minor Analysis Improvements
* The `AlertSuppression.ql` query has been updated to support the new `// codeql[query-id]` supression comments. These comments can be used to suppress an alert and must be placed on a blank line before the alert. In addition the legacy `// lgtm` and `// lgtm[query-id]` comments can now also be placed on the line before an alert.
* The `cpp/missing-check-scanf` query no longer reports the free'ing of `scanf` output variables as potential reads.
## 0.4.6
No user-facing changes.
## 0.4.5
No user-facing changes.
## 0.4.4
No user-facing changes.

View File

@@ -115,8 +115,7 @@ BasicBlock blockGuardedBy(int value, string op, ScanfFunctionCall call) {
from ScanfOutput output, ScanfFunctionCall call, Access access
where
output.getCall() = call and
output.hasGuardedAccess(access, false) and
not exists(DeallocationExpr dealloc | dealloc.getFreedExpr() = access)
output.hasGuardedAccess(access, false)
select access,
"This variable is read, but may not have been written. " +
"It should be guarded by a check that the $@ returns at least " +

View File

@@ -66,7 +66,9 @@ class ElseDirective extends Directive {
override predicate mismatched() { depth() < 1 }
}
class EndifDirective extends Directive instanceof PreprocessorEndif {
class EndifDirective extends Directive {
EndifDirective() { this instanceof PreprocessorEndif }
override int depthChange() { result = -1 }
override predicate mismatched() { depth() < 0 }

View File

@@ -12,8 +12,8 @@
*/
/*
* Note: this query is not assigned a precision yet because we don't want it
* to be included in query suites until its performance is well understood.
* Note: this query is not assigned a precision yet because we don't want it on
* LGTM until its performance is well understood.
*/
import cpp

View File

@@ -52,7 +52,7 @@ class Library extends LibraryT {
// The versions reported for C/C++ dependencies are just the versions that
// happen to be installed on the system where the build takes place.
// Reporting those versions is likely to cause misunderstandings, both for
// people reading them and for vulnerability checkers.
// people reading them and for the vulnerability checker of lgtm.
result = "unknown"
}

View File

@@ -16,10 +16,9 @@
import cpp
import semmle.code.cpp.security.FunctionWithWrappers
import semmle.code.cpp.security.FlowSources
import semmle.code.cpp.ir.IR
import semmle.code.cpp.ir.dataflow.TaintTracking
import DataFlow::PathGraph
import semmle.code.cpp.security.Security
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
/**
* A function for opening a file.
@@ -47,69 +46,19 @@ class FileFunction extends FunctionWithWrappers {
override predicate interestingArg(int arg) { arg = 0 }
}
Expr asSinkExpr(DataFlow::Node node) {
result =
node.asOperand()
.(SideEffectOperand)
.getUse()
.(ReadSideEffectInstruction)
.getArgumentDef()
.getUnconvertedResultExpression()
}
/**
* Holds for a variable that has any kind of upper-bound check anywhere in the program.
* This is biased towards being inclusive and being a coarse overapproximation because
* there are a lot of valid ways of doing an upper bounds checks if we don't consider
* where it occurs, for example:
* ```cpp
* if (x < 10) { sink(x); }
*
* if (10 > y) { sink(y); }
*
* if (z > 10) { z = 10; }
* sink(z);
* ```
*/
predicate hasUpperBoundsCheck(Variable var) {
exists(RelationalOperation oper, VariableAccess access |
oper.getAnOperand() = access and
access.getTarget() = var and
// Comparing to 0 is not an upper bound check
not oper.getAnOperand().getValue() = "0"
)
}
class TaintedPathConfiguration extends TaintTracking::Configuration {
TaintedPathConfiguration() { this = "TaintedPathConfiguration" }
override predicate isSource(DataFlow::Node node) { node instanceof FlowSource }
override predicate isSink(DataFlow::Node node) {
exists(FileFunction fileFunction |
fileFunction.outermostWrapperFunctionCall(asSinkExpr(node), _)
)
}
override predicate isSanitizer(DataFlow::Node node) {
node.asExpr().(Call).getTarget().getUnspecifiedType() instanceof ArithmeticType
or
exists(LoadInstruction load, Variable checkedVar |
load = node.asInstruction() and
checkedVar = load.getSourceAddress().(VariableAddressInstruction).getAstVariable() and
hasUpperBoundsCheck(checkedVar)
)
class TaintedPathConfiguration extends TaintTrackingConfiguration {
override predicate isSink(Element tainted) {
exists(FileFunction fileFunction | fileFunction.outermostWrapperFunctionCall(tainted, _))
}
}
from
FileFunction fileFunction, Expr taintedArg, FlowSource taintSource, TaintedPathConfiguration cfg,
DataFlow::PathNode sourceNode, DataFlow::PathNode sinkNode, string callChain
FileFunction fileFunction, Expr taintedArg, Expr taintSource, PathNode sourceNode,
PathNode sinkNode, string taintCause, string callChain
where
taintedArg = asSinkExpr(sinkNode.getNode()) and
fileFunction.outermostWrapperFunctionCall(taintedArg, callChain) and
cfg.hasFlowPath(sourceNode, sinkNode) and
taintSource = sourceNode.getNode()
taintedWithPath(taintSource, taintedArg, sourceNode, sinkNode) and
isUserInput(taintSource, taintCause)
select taintedArg, sourceNode, sinkNode,
"This argument to a file access function is derived from $@ and then passed to " + callChain + ".",
taintSource, "user input (" + taintSource.getSourceType() + ")"
taintSource, "user input (" + taintCause + ")"

View File

@@ -13,7 +13,7 @@
import cpp
import semmle.code.cpp.commons.Environment
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
/** A call that prints its arguments to `stdout`. */

View File

@@ -15,7 +15,7 @@
import cpp
import semmle.code.cpp.security.Security
import semmle.code.cpp.security.FunctionWithWrappers
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
class SqlLikeFunction extends FunctionWithWrappers {

View File

@@ -14,7 +14,7 @@
import cpp
import semmle.code.cpp.security.Security
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
predicate isProcessOperationExplanation(Expr arg, string processOperation) {

View File

@@ -16,7 +16,7 @@
import semmle.code.cpp.security.BufferWrite
import semmle.code.cpp.security.Security
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
/*

View File

@@ -116,6 +116,10 @@ class ImproperArrayIndexValidationConfig extends TaintTracking::Configuration {
}
}
/** Gets `str` where the first letter has been lowercased. */
bindingset[str]
string lowerFirst(string str) { result = str.prefix(1).toLowerCase() + str.suffix(1) }
from
ImproperArrayIndexValidationConfig conf, DataFlow::PathNode source, DataFlow::PathNode sink,
string sourceType
@@ -124,4 +128,4 @@ where
isFlowSource(source.getNode(), sourceType)
select sink.getNode(), source, sink,
"An array indexing expression depends on $@ that might be outside the bounds of the array.",
source.getNode(), sourceType
source.getNode(), lowerFirst(sourceType)

View File

@@ -16,7 +16,7 @@
import cpp
import semmle.code.cpp.security.Security
import semmle.code.cpp.security.FunctionWithWrappers
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
class Configuration extends TaintTrackingConfiguration {

View File

@@ -16,7 +16,7 @@
import cpp
import semmle.code.cpp.security.FunctionWithWrappers
import semmle.code.cpp.security.Security
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
class Configuration extends TaintTrackingConfiguration {

View File

@@ -12,7 +12,7 @@
import cpp
import semmle.code.cpp.commons.NullTermination
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
/** A user-controlled expression that may not be null terminated. */
class TaintSource extends VariableAccess {

View File

@@ -15,7 +15,7 @@
import cpp
import semmle.code.cpp.security.Overflow
import semmle.code.cpp.security.Security
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
import TaintedWithPath
import Bounded

View File

@@ -17,7 +17,7 @@
import cpp
import semmle.code.cpp.security.Overflow
import semmle.code.cpp.security.Security
import semmle.code.cpp.ir.dataflow.internal.DefaultTaintTrackingImpl
import semmle.code.cpp.security.TaintTracking
predicate isMaxValue(Expr mie) {
exists(MacroInvocation mi |

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