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
3642 changed files with 139050 additions and 251696 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

@@ -1,32 +0,0 @@
name: OS Version
description: Get OS version.
outputs:
version:
description: "OS version"
value: ${{ steps.version.outputs.version }}
runs:
using: composite
steps:
- if: runner.os == 'Linux'
shell: bash
run: |
. /etc/os-release
echo "VERSION=${NAME} ${VERSION}" >> $GITHUB_ENV
- if: runner.os == 'Windows'
shell: powershell
run: |
$objects = systeminfo.exe /FO CSV | ConvertFrom-Csv
"VERSION=$($objects.'OS Name') $($objects.'OS Version')" >> $env:GITHUB_ENV
- if: runner.os == 'macOS'
shell: bash
run: |
echo "VERSION=$(sw_vers -productName) $(sw_vers -productVersion)" >> $GITHUB_ENV
- name: Emit OS version
id: version
shell: bash
run: |
echo "$VERSION"
echo "version=${VERSION}" >> $GITHUB_OUTPUT

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

@@ -26,9 +26,3 @@ jobs:
run: |
gh api 'repos/${{github.repository}}/pulls/${{github.event.number}}/files' --paginate --jq 'any(.[].filename ; test("/change-notes/.*[.]md$"))' |
grep true -c
- name: Fail if the change note filename doesn't match the expected format. The file name must be of the form 'YYYY-MM-DD.md' or 'YYYY-MM-DD-{title}.md', where '{title}' is arbitrary text.
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
gh api 'repos/${{github.repository}}/pulls/${{github.event.number}}/files' --paginate --jq '[.[].filename | select(test("/change-notes/.*[.]md$"))] | all(test("/change-notes/[0-9]{4}-[0-9]{2}-[0-9]{2}.*[.]md$"))' |
grep true -c

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

@@ -11,7 +11,7 @@ on:
branches:
- main
paths:
- "java/ql/src/utils/modelgenerator/**/*.*"
- "java/ql/src/utils/model-generator/**/*.*"
- ".github/workflows/mad_modelDiff.yml"
permissions:
@@ -40,12 +40,12 @@ jobs:
- name: Download database
env:
SLUG: ${{ matrix.slug }}
GH_TOKEN: ${{ github.token }}
run: |
set -x
mkdir lib-dbs
SHORTNAME=${SLUG//[^a-zA-Z0-9_]/}
gh api -H "Accept: application/zip" "/repos/${SLUG}/code-scanning/codeql/databases/java" > "$SHORTNAME.zip"
projectId=`curl -s https://lgtm.com/api/v1.0/projects/g/${SLUG} | jq .id`
curl -L "https://lgtm.com/api/v1.0/snapshots/$projectId/java" -o "$SHORTNAME.zip"
unzip -q -d "${SHORTNAME}-db" "${SHORTNAME}.zip"
mkdir "lib-dbs/$SHORTNAME/"
mv "${SHORTNAME}-db/"$(ls -1 "${SHORTNAME}"-db)/* "lib-dbs/${SHORTNAME}/"
@@ -61,8 +61,8 @@ jobs:
DATABASE=$2
cd codeql-$QL_VARIANT
SHORTNAME=`basename $DATABASE`
python java/ql/src/utils/modelgenerator/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 ..
}
@@ -85,21 +85,19 @@ 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:
name: diffs
path: tmp-models/*.html
# An html file is only produced if the generated models differ.
if-no-files-found: ignore
retention-days: 20

View File

@@ -50,10 +50,10 @@ jobs:
SLUG: ${{ matrix.slug }}
run: |
SHORTNAME=${SLUG//[^a-zA-Z0-9_]/}
java/ql/src/utils/modelgenerator/RegenerateModels.py "${SLUG}" dbs/${SHORTNAME}
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@45955cb1830b640e2c1603ad72ad542a49d47b96
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: |
@@ -38,14 +34,12 @@ jobs:
shell: bash
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- uses: ./.github/actions/os-version
id: os_version
- name: Cache entire pack
id: cache-pack
uses: actions/cache@v3
with:
path: ${{ runner.temp }}/pack
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-pack-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}-${{ hashFiles('ql/**/*.ql*') }}-${{ hashFiles('ql/**/qlpack.yml') }}-${{ hashFiles('ql/ql/src/ql.dbscheme*') }}-${{ steps.get-codeql-version.outputs.version }}--${{ hashFiles('.github/workflows/ql-for-ql-build.yml') }}
key: ${{ runner.os }}-pack-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}-${{ hashFiles('ql/**/*.ql*') }}-${{ hashFiles('ql/**/qlpack.yml') }}-${{ hashFiles('ql/ql/src/ql.dbscheme*') }}-${{ steps.get-codeql-version.outputs.version }}--${{ hashFiles('.github/workflows/ql-for-ql-build.yml') }}
- name: Cache queries
if: steps.cache-pack.outputs.cache-hit != 'true'
id: cache-queries
@@ -79,7 +73,7 @@ jobs:
ql/target/release/ql-autobuilder.exe
ql/target/release/ql-extractor
ql/target/release/ql-extractor.exe
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-extractor-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}
key: ${{ runner.os }}-extractor-${{ hashFiles('ql/**/Cargo.lock') }}-${{ hashFiles('ql/**/*.rs') }}
- name: Cache cargo
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
uses: actions/cache@v3
@@ -88,7 +82,7 @@ jobs:
~/.cargo/registry
~/.cargo/git
ql/target
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-rust-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
key: ${{ runner.os }}-rust-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
- name: Check formatting
if: steps.cache-extractor.outputs.cache-hit != 'true' && steps.cache-pack.outputs.cache-hit != 'true'
run: cd ql; cargo fmt --all -- --check
@@ -139,20 +133,19 @@ jobs:
env:
CONF: ./ql-for-ql-config.yml
- name: Initialize CodeQL
uses: github/codeql-action/init@45955cb1830b640e2c1603ad72ad542a49d47b96
uses: github/codeql-action/init@77a8d2d10c0b403a8b4aadbd223dc489ecd22683
with:
languages: ql
db-location: ${{ runner.temp }}/db
config-file: ./ql-for-ql-config.yml
tools: ${{ steps.find-latest-bundle.outputs.url }}
- name: Move pack queries
- name: Move pack cache
run: |
cp -r ${PACK}/queries ql/ql/src
cp -r ${PACK}/.cache ql/ql/src/.cache
env:
PACK: ${{ runner.temp }}/pack
- name: Perform CodeQL Analysis
uses: github/codeql-action/analyze@45955cb1830b640e2c1603ad72ad542a49d47b96
uses: github/codeql-action/analyze@77a8d2d10c0b403a8b4aadbd223dc489ecd22683
with:
category: "ql-for-ql"
- name: Copy sarif file to CWD
@@ -174,4 +167,4 @@ jobs:
with:
name: ql-for-ql-langs
path: split-sarif
retention-days: 1
retention-days: 1

View File

@@ -25,18 +25,16 @@ jobs:
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@45955cb1830b640e2c1603ad72ad542a49d47b96
uses: github/codeql-action/init@77a8d2d10c0b403a8b4aadbd223dc489ecd22683
with:
languages: javascript # does not matter
- uses: ./.github/actions/os-version
id: os_version
- uses: actions/cache@v3
with:
path: |
~/.cargo/registry
~/.cargo/git
ql/target
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-qltest-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
key: ${{ runner.os }}-qltest-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
- name: Build Extractor
run: cd ql; env "PATH=$PATH:`dirname ${CODEQL}`" ./scripts/create-extractor-pack.sh
env:

View File

@@ -20,86 +20,30 @@ jobs:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v3
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@45955cb1830b640e2c1603ad72ad542a49d47b96
with:
languages: javascript # does not matter
- uses: ./.github/actions/os-version
id: os_version
- uses: actions/cache@v3
with:
path: |
~/.cargo/registry
~/.cargo/git
ql/target
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-qltest-cargo-${{ hashFiles('ql/rust-toolchain.toml', 'ql/**/Cargo.lock') }}
- name: Build extractor
run: |
cd ql;
codeqlpath=$(dirname ${{ steps.find-codeql.outputs.codeql-path }});
env "PATH=$PATH:$codeqlpath" ./scripts/create-extractor-pack.sh
- name: Cache compilation cache
id: query-cache
uses: ./.github/actions/cache-query-compilation
with:
key: ql-for-ql-tests
- name: Run QL tests
run: |
"${CODEQL}" test run --check-databases --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --search-path "${{ github.workspace }}/ql/extractor-pack" --consistency-queries ql/ql/consistency-queries --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}" ql/ql/test
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
other-os:
strategy:
matrix:
os: [macos-latest, windows-latest]
needs: [qltest]
runs-on: ${{ matrix.os }}
steps:
- uses: actions/checkout@v3
- name: Install GNU tar
if: runner.os == 'macOS'
run: |
brew install gnu-tar
echo "/usr/local/opt/gnu-tar/libexec/gnubin" >> $GITHUB_PATH
- name: Find codeql
id: find-codeql
uses: github/codeql-action/init@77a8d2d10c0b403a8b4aadbd223dc489ecd22683
with:
languages: javascript # does not matter
- uses: ./.github/actions/os-version
id: os_version
- uses: actions/cache@v3
with:
path: |
~/.cargo/registry
~/.cargo/git
ql/target
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-qltest-cargo-${{ hashFiles('ql/rust-toolchain.toml', 'ql/**/Cargo.lock') }}
key: ${{ runner.os }}-qltest-cargo-${{ hashFiles('ql/**/Cargo.lock') }}
- name: Build extractor
if: runner.os != 'Windows'
run: |
cd ql;
codeqlpath=$(dirname ${{ steps.find-codeql.outputs.codeql-path }});
env "PATH=$PATH:$codeqlpath" ./scripts/create-extractor-pack.sh
- name: Build extractor (Windows)
if: runner.os == 'Windows'
shell: pwsh
- name: Run QL tests
run: |
cd ql;
$Env:PATH += ";$(dirname ${{ steps.find-codeql.outputs.codeql-path }})"
pwsh ./scripts/create-extractor-pack.ps1
- name: Run a single QL tests - Unix
if: runner.os != 'Windows'
run: |
"${CODEQL}" test run --check-databases --search-path "${{ github.workspace }}/ql/extractor-pack" ql/ql/test/queries/style/DeadCode/DeadCode.qlref
"${CODEQL}" test run --check-databases --check-unused-labels --check-repeated-labels --check-redefined-labels --check-use-before-definition --search-path "${{ github.workspace }}/ql/extractor-pack" --consistency-queries ql/ql/consistency-queries ql/ql/test
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}
- name: Run a single QL tests - Windows
if: runner.os == 'Windows'
shell: pwsh
- name: Check QL formatting
run: |
$Env:PATH += ";$(dirname ${{ steps.find-codeql.outputs.codeql-path }})"
codeql test run --check-databases --search-path "${{ github.workspace }}/ql/extractor-pack" ql/ql/test/queries/style/DeadCode/DeadCode.qlref
find ql/ql/src "(" -name "*.ql" -or -name "*.qll" ")" -print0 | xargs -0 "${CODEQL}" query format --check-only
env:
CODEQL: ${{ steps.find-codeql.outputs.codeql-path }}

View File

@@ -48,41 +48,23 @@ jobs:
run: |
brew install gnu-tar
echo "/usr/local/opt/gnu-tar/libexec/gnubin" >> $GITHUB_PATH
- uses: ./.github/actions/os-version
id: os_version
- 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: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-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
~/.cargo/git
ruby/target
key: ${{ runner.os }}-${{ steps.os_version.outputs.version }}-ruby-rust-cargo-${{ hashFiles('ruby/rust-toolchain.toml', 'ruby/**/Cargo.lock') }}
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: ${{ matrix.os == 'ubuntu-latest' }}
@@ -117,10 +99,9 @@ jobs:
- name: Build Query Pack
run: |
rm -rf target/packs
codeql pack create ../shared/ssa --output target/packs
codeql pack create ../misc/suite-helpers --output target/packs
codeql pack create ../shared/regex --output target/packs
codeql pack create ../shared/ssa --output target/packs
codeql pack create ../shared/tutorial --output target/packs
codeql pack create ql/lib --output target/packs
codeql pack create -j0 ql/src --output target/packs --compilation-cache "${{ steps.query-cache.outputs.cache-dir }}"
PACK_FOLDER=$(readlink -f target/packs/codeql/ruby-queries/*)
@@ -205,6 +186,11 @@ jobs:
- name: Fetch CodeQL
uses: ./.github/actions/fetch-codeql
- uses: actions/checkout@v3
with:
repository: Shopify/example-ruby-app
ref: 67a0decc5eb550f3a9228eda53925c3afd40dfe9
- name: Download Ruby bundle
uses: actions/download-artifact@v3
with:
@@ -213,15 +199,26 @@ jobs:
- name: Unzip Ruby bundle
shell: bash
run: unzip -q -d "${{ runner.temp }}/ruby-bundle" "${{ runner.temp }}/codeql-ruby-bundle.zip"
- name: Prepare test files
shell: bash
run: |
echo "import codeql.ruby.AST select count(File f)" > "test.ql"
echo "| 4 |" > "test.expected"
echo 'name: sample-tests
version: 0.0.0
dependencies:
codeql/ruby-all: "*"
extractor: ruby
tests: .
' > qlpack.yml
- name: Run QL test
shell: bash
run: |
codeql test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" ruby/ql/test/library-tests/ast/constants/
codeql test run --search-path "${{ runner.temp }}/ruby-bundle" --additional-packs "${{ runner.temp }}/ruby-bundle" .
- name: Create database
shell: bash
run: |
codeql database create --search-path "${{ runner.temp }}/ruby-bundle" --language ruby --source-root ruby/ql/test/library-tests/ast/constants/ ../database
codeql database create --search-path "${{ runner.temp }}/ruby-bundle" --language ruby --source-root . ../database
- name: Analyze database
shell: bash
run: |

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 @@ If you have an idea for a query that you would like to share with other CodeQL u
Each language-specific directory contains further subdirectories that group queries based on their `@tags` or purpose.
- Experimental queries and libraries are stored in the `experimental` subdirectory within each language-specific directory in the [CodeQL repository](https://github.com/github/codeql). For example, experimental Java queries and libraries are stored in `java/ql/src/experimental` and any corresponding tests in `java/ql/test/experimental`.
- Experimental queries need to include `experimental` in their `@tags`
- The structure of an `experimental` subdirectory mirrors the structure of its parent directory.
- Select or create an appropriate directory in `experimental` based on the existing directory structure of `experimental` or its parent directory.

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"
@@ -402,6 +396,16 @@
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/ControlFlowReachability.qll",
"csharp/ql/lib/semmle/code/csharp/dataflow/internal/rangeanalysis/ControlFlowReachability.qll"
],
"Inline Test Expectations": [
"cpp/ql/test/TestUtilities/InlineExpectationsTest.qll",
"csharp/ql/test/TestUtilities/InlineExpectationsTest.qll",
"java/ql/test/TestUtilities/InlineExpectationsTest.qll",
"python/ql/test/TestUtilities/InlineExpectationsTest.qll",
"ruby/ql/test/TestUtilities/InlineExpectationsTest.qll",
"ql/ql/test/TestUtilities/InlineExpectationsTest.qll",
"go/ql/test/TestUtilities/InlineExpectationsTest.qll",
"swift/ql/test/TestUtilities/InlineExpectationsTest.qll"
],
"C++ ExternalAPIs": [
"cpp/ql/src/Security/CWE/CWE-020/ExternalAPIs.qll",
"cpp/ql/src/Security/CWE/CWE-020/ir/ExternalAPIs.qll"
@@ -460,10 +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",
@@ -495,9 +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",
@@ -513,16 +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"
@@ -566,9 +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

@@ -13,5 +13,5 @@ predicate isExprWithNewBuiltin(Expr expr) {
from Expr expr, int kind, int kind_new, Location location
where
exprs(expr, kind, location) and
if isExprWithNewBuiltin(expr) then kind_new = 1 else kind_new = kind
if isExprWithNewBuiltin(expr) then kind_new = 0 else kind_new = kind
select expr, kind_new, location

View File

@@ -9,5 +9,5 @@ class Location extends @location_expr {
from Expr expr, int kind, int kind_new, Location location
where
exprs(expr, kind, location) and
if expr instanceof @blockassignexpr then kind_new = 1 else kind_new = kind
if expr instanceof @blockassignexpr then kind_new = 0 else kind_new = kind
select expr, kind_new, location

View File

@@ -1,11 +0,0 @@
class BuiltinType extends @builtintype {
string toString() { none() }
}
from BuiltinType type, string name, int kind, int kind_new, int size, int sign, int alignment
where
builtintypes(type, name, kind, size, sign, alignment) and
if type instanceof @float16 or type instanceof @complex_float16
then kind_new = 2
else kind_new = kind
select type, name, kind_new, size, sign, alignment

View File

@@ -1,3 +0,0 @@
description: Introduce (_Complex) _Float16 type
compatibility: backwards
builtintypes.rel: run builtintypes.qlo

View File

@@ -1,2 +0,0 @@
description: Uncomment case splits in dbscheme
compatibility: full

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()`.
@@ -123,13 +123,6 @@ private predicate constructorCallTypeMention(ConstructorCall cc, TypeMention tm)
)
}
/** Holds if `loc` has the container `container` and is on the line starting at `startLine`. */
pragma[nomagic]
private predicate hasContainerAndStartLine(Location loc, Container container, int startLine) {
loc.getStartLine() = startLine and
loc.getContainer() = container
}
/**
* Gets an element, of kind `kind`, that element `e` uses, if any.
* Attention: This predicate yields multiple definitions for a single location.
@@ -166,9 +159,9 @@ Top definitionOf(Top e, string kind) {
// Multiple type mentions can be generated when a typedef is used, and
// in such cases we want to exclude all but the originating typedef.
not exists(Type secondary |
exists(File f, int startline, int startcol |
exists(TypeMention tm, File f, int startline, int startcol |
typeMentionStartLoc(e, result, f, startline, startcol) and
typeMentionStartLoc(_, secondary, f, startline, startcol) and
typeMentionStartLoc(tm, secondary, f, startline, startcol) and
(
result = secondary.(TypedefType).getBaseType() or
result = secondary.(TypedefType).getBaseType().(SpecifiedType).getBaseType()
@@ -191,9 +184,11 @@ Top definitionOf(Top e, string kind) {
kind = "I" and
result = e.(Include).getIncludedFile() and
// exclude `#include` directives containing macros
not exists(MacroInvocation mi, Container container, int startLine |
hasContainerAndStartLine(e.(Include).getLocation(), container, startLine) and
hasContainerAndStartLine(mi.getLocation(), container, startLine)
not exists(MacroInvocation mi, Location l1, Location l2 |
l1 = e.(Include).getLocation() and
l2 = mi.getLocation() and
l1.getContainer() = l2.getContainer() and
l1.getStartLine() = l2.getStartLine()
// (an #include directive must be always on it's own line)
)
) and

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

@@ -45,16 +45,6 @@ module Consistency {
) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodeAtPosition`. */
predicate uniqueParameterNodeAtPositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodePosition`. */
predicate uniqueParameterNodePositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
}
private class RelevantNode extends Node {
@@ -111,7 +101,9 @@ module Consistency {
exists(int c |
c =
strictcount(Node n |
not n.hasLocationInfo(_, _, _, _, _) and
not exists(string filepath, int startline, int startcolumn, int endline, int endcolumn |
n.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
) and
not any(ConsistencyConfiguration conf).missingLocationExclude(n)
) and
msg = "Nodes without location: " + c
@@ -252,27 +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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodeAtPositionExclude(c, pos, p) and
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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodePositionExclude(c, pos, p) and
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

@@ -218,7 +218,7 @@ private predicate allocation(Instruction array, Length length, int delta) {
length.(VNLength).getInstruction().getConvertedResultExpression() = lengthExpr
)
or
not deconstructMallocSizeExpr(alloc.getSizeExpr(), _, _) and
not exists(int d | deconstructMallocSizeExpr(alloc.getSizeExpr(), _, d)) and
length.(VNLength).getInstruction().getConvertedResultExpression() = alloc.getSizeExpr() and
delta = 0
)

View File

@@ -543,7 +543,9 @@ private predicate boundedPhiCand(
PhiInstruction phi, boolean upper, Bound b, int delta, boolean fromBackEdge, int origdelta,
Reason reason
) {
boundedPhiInp(phi, _, b, delta, upper, fromBackEdge, origdelta, reason)
exists(PhiInputOperand op |
boundedPhiInp(phi, op, b, delta, upper, fromBackEdge, origdelta, reason)
)
}
/**

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

@@ -664,7 +664,9 @@ private predicate boundedPhiCand(
SemSsaPhiNode phi, boolean upper, SemBound b, int delta, boolean fromBackEdge, int origdelta,
SemReason reason
) {
boundedPhiInp(phi, _, _, b, delta, upper, fromBackEdge, origdelta, reason)
exists(SemSsaVariable inp, SemSsaReadPositionPhiInputEdge edge |
boundedPhiInp(phi, inp, edge, b, delta, upper, fromBackEdge, origdelta, reason)
)
}
/**

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: 0.5.1-dev
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

@@ -318,7 +318,7 @@ class Function extends Declaration, ControlFlowNode, AccessHolder, @function {
MetricFunction getMetrics() { result = this }
/** Holds if this function calls the function `f`. */
predicate calls(Function f) { this.calls(f, _) }
predicate calls(Function f) { exists(Locatable l | this.calls(f, l)) }
/**
* Holds if this function calls the function `f` in the `FunctionCall`
@@ -335,7 +335,7 @@ class Function extends Declaration, ControlFlowNode, AccessHolder, @function {
}
/** Holds if this function accesses a function or variable or enumerator `a`. */
predicate accesses(Declaration a) { this.accesses(a, _) }
predicate accesses(Declaration a) { exists(Locatable l | this.accesses(a, l)) }
/**
* Holds if this function accesses a function or variable or enumerator `a`

View File

@@ -10,14 +10,12 @@ import semmle.code.cpp.File
*/
class Location extends @location {
/** Gets the container corresponding to this location. */
pragma[nomagic]
Container getContainer() { this.fullLocationInfo(result, _, _, _, _) }
/** Gets the file corresponding to this location, if any. */
File getFile() { result = this.getContainer() }
/** Gets the 1-based line number (inclusive) where this location starts. */
pragma[nomagic]
int getStartLine() { this.fullLocationInfo(_, result, _, _, _) }
/** Gets the 1-based column number (inclusive) where this location starts. */

View File

@@ -816,12 +816,6 @@ private predicate floatingPointTypeMapping(
or
// _Float128x
kind = 50 and base = 2 and domain = TRealDomain() and realKind = 50 and extended = true
or
// _Float16
kind = 52 and base = 2 and domain = TRealDomain() and realKind = 52 and extended = false
or
// _Complex _Float16
kind = 53 and base = 2 and domain = TComplexDomain() and realKind = 52 and extended = false
}
/**

View File

@@ -33,7 +33,7 @@ DependencyOptions getDependencyOptions() { any() }
class DependsSource extends Element {
DependsSource() {
// not inside a template instantiation
not this.isFromTemplateInstantiation(_) or
not exists(Element other | this.isFromTemplateInstantiation(other)) or
// allow DeclarationEntrys of template specializations
this.(DeclarationEntry).getDeclaration().(Function).isConstructedFrom(_) or
this.(DeclarationEntry).getDeclaration().(Class).isConstructedFrom(_)

View File

@@ -69,9 +69,12 @@ predicate functionContainsDisabledCode(Function f) {
*/
predicate functionContainsPreprocCode(Function f) {
// `f` contains a preprocessor branch
exists(string file, int pbdStartLine, int fBlockStartLine, int fBlockEndLine |
exists(
PreprocessorBranchDirective pbd, string file, int pbdStartLine, int fBlockStartLine,
int fBlockEndLine
|
functionLocation(f, file, fBlockStartLine, fBlockEndLine) and
pbdLocation(_, file, pbdStartLine) and
pbdLocation(pbd, file, pbdStartLine) and
pbdStartLine <= fBlockEndLine and
pbdStartLine >= fBlockStartLine
)

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 }
@@ -244,7 +258,9 @@ class ScanfFormatLiteral extends Expr {
/**
* Gets the maximum width option of the nth input (empty string if none is given).
*/
string getMaxWidthOpt(int n) { this.parseConvSpec(n, _, result, _, _) }
string getMaxWidthOpt(int n) {
exists(string spec, string len, string conv | this.parseConvSpec(n, spec, result, len, conv))
}
/**
* Gets the maximum width of the nth input.
@@ -254,12 +270,18 @@ class ScanfFormatLiteral extends Expr {
/**
* Gets the length flag of the nth conversion specifier.
*/
string getLength(int n) { this.parseConvSpec(n, _, _, result, _) }
string getLength(int n) {
exists(string spec, string width, string conv |
this.parseConvSpec(n, spec, width, result, conv)
)
}
/**
* Gets the conversion character of the nth conversion specifier.
*/
string getConversionChar(int n) { this.parseConvSpec(n, _, _, _, result) }
string getConversionChar(int n) {
exists(string spec, string width, string len | this.parseConvSpec(n, spec, width, len, result))
}
/**
* Gets the maximum length of the string that can be produced by the nth

View File

@@ -54,7 +54,7 @@ class SubBasicBlock extends ControlFlowNodeBase {
* only condition under which a `SubBasicBlock` may have multiple
* predecessors.
*/
predicate firstInBB() { this.getRankInBasicBlock(_) = 1 }
predicate firstInBB() { exists(BasicBlock bb | this.getRankInBasicBlock(bb) = 1) }
/**
* Holds if this `SubBasicBlock` comes last in its basic block. This is the

View File

@@ -441,8 +441,8 @@ library class ExprEvaluator extends int {
req = mid.(AssignExpr).getRValue()
)
or
exists(Variable v, boolean sub1 |
this.interestingVariableAccess(e, _, v, sub1) and
exists(VariableAccess va, Variable v, boolean sub1 |
this.interestingVariableAccess(e, va, v, sub1) and
req = v.getAnAssignedValue() and
(sub1 = true implies not this.ignoreVariableAssignment(e, v, req)) and
sub = false
@@ -876,7 +876,7 @@ private predicate nonAnalyzableVariableDefinition(Variable v, StmtParent def) {
* empirically to have effect only on a few rare and pathological examples.
*/
private predicate tractableVariable(Variable v) {
not nonAnalyzableVariableDefinition(v, _) or
not exists(StmtParent def | nonAnalyzableVariableDefinition(v, def)) or
strictcount(StmtParent def | nonAnalyzableVariableDefinition(v, def)) < 1000
}

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

@@ -45,16 +45,6 @@ module Consistency {
) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodeAtPosition`. */
predicate uniqueParameterNodeAtPositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodePosition`. */
predicate uniqueParameterNodePositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
}
private class RelevantNode extends Node {
@@ -111,7 +101,9 @@ module Consistency {
exists(int c |
c =
strictcount(Node n |
not n.hasLocationInfo(_, _, _, _, _) and
not exists(string filepath, int startline, int startcolumn, int endline, int endcolumn |
n.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
) and
not any(ConsistencyConfiguration conf).missingLocationExclude(n)
) and
msg = "Nodes without location: " + c
@@ -252,27 +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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodeAtPositionExclude(c, pos, p) and
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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodePositionExclude(c, pos, p) and
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

@@ -450,8 +450,10 @@ module FlowVar_internal {
}
override string toString() {
this.definedByExpr(_, _) and
result = "assignment to " + v
exists(Expr e |
this.definedByExpr(e, _) and
result = "assignment to " + v
)
or
this.definedByInitialValue(_) and
result = "initial value of " + v

View File

@@ -54,7 +54,7 @@ class SubBasicBlock extends ControlFlowNodeBase {
* only condition under which a `SubBasicBlock` may have multiple
* predecessors.
*/
predicate firstInBB() { this.getRankInBasicBlock(_) = 1 }
predicate firstInBB() { exists(BasicBlock bb | this.getRankInBasicBlock(bb) = 1) }
/**
* Holds if this `SubBasicBlock` comes last in its basic block. This is the

View File

@@ -70,8 +70,8 @@ module VirtualDispatch {
* that is, `c` or one of its supertypes overrides `f`.
*/
private predicate cannotInherit(Class c, MemberFunction f) {
exists(MemberFunction override |
cannotInheritHelper(c, f, _, override) and
exists(Class overridingType, MemberFunction override |
cannotInheritHelper(c, f, overridingType, override) and
override.overrides+(f)
)
}

View File

@@ -53,7 +53,7 @@ class Expr extends StmtParent, @expr {
Declaration getEnclosingDeclaration() { result = exprEnclosingElement(this) }
/** Gets a child of this expression. */
Expr getAChild() { result = this.getChild(_) }
Expr getAChild() { exists(int n | result = this.getChild(n)) }
/** Gets the parent of this expression, if any. */
Element getParent() { exprparents(underlyingElement(this), _, unresolveElement(result)) }

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

@@ -45,16 +45,6 @@ module Consistency {
) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodeAtPosition`. */
predicate uniqueParameterNodeAtPositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
/** Holds if `(c, pos, p)` should be excluded from the consistency test `uniqueParameterNodePosition`. */
predicate uniqueParameterNodePositionExclude(DataFlowCallable c, ParameterPosition pos, Node p) {
none()
}
}
private class RelevantNode extends Node {
@@ -111,7 +101,9 @@ module Consistency {
exists(int c |
c =
strictcount(Node n |
not n.hasLocationInfo(_, _, _, _, _) and
not exists(string filepath, int startline, int startcolumn, int endline, int endcolumn |
n.hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
) and
not any(ConsistencyConfiguration conf).missingLocationExclude(n)
) and
msg = "Nodes without location: " + c
@@ -252,27 +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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodeAtPositionExclude(c, pos, p) and
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
) {
not any(ConsistencyConfiguration conf).uniqueParameterNodePositionExclude(c, pos, p) and
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

@@ -30,7 +30,6 @@ private newtype TOpcode =
TNegate() or
TShiftLeft() or
TShiftRight() or
TUnsignedShiftRight() or
TBitAnd() or
TBitOr() or
TBitXor() or
@@ -653,15 +652,6 @@ module Opcode {
final override string toString() { result = "ShiftRight" }
}
/**
* The `Opcode` for a `UnsignedShiftRightInstruction`.
*
* See the `UnsignedShiftRightInstruction` documentation for more details.
*/
class UnsignedShiftRight extends BinaryBitwiseOpcode, TUnsignedShiftRight {
final override string toString() { result = "UnsignedShiftRight" }
}
/**
* The `Opcode` for a `BitAndInstruction`.
*

View File

@@ -1204,17 +1204,6 @@ class ShiftRightInstruction extends BinaryBitwiseInstruction {
ShiftRightInstruction() { this.getOpcode() instanceof Opcode::ShiftRight }
}
/**
* An instruction that shifts its left operand to the right by the number of bits specified by its
* right operand.
*
* Both operands must have an integer type. The result has the same type as the left operand.
* The leftmost bits are zero-filled.
*/
class UnsignedShiftRightInstruction extends BinaryBitwiseInstruction {
UnsignedShiftRightInstruction() { this.getOpcode() instanceof Opcode::UnsignedShiftRight }
}
/**
* An instruction that performs a binary arithmetic operation involving at least one pointer
* operand.

View File

@@ -45,7 +45,7 @@ class Operand extends TStageOperand {
this = reusedPhiOperand(use, def, predecessorBlock, _)
)
or
this = chiOperand(_, _)
exists(Instruction use | this = chiOperand(use, _))
}
/** Gets a textual representation of this element. */

View File

@@ -329,12 +329,12 @@ private module Cached {
cached
Instruction getChiInstructionTotalOperand(ChiInstruction chiInstr) {
exists(
Alias::VirtualVariable vvar, OldInstruction oldInstr, OldBlock defBlock, int defRank,
int defOffset, OldBlock useBlock, int useRank
Alias::VirtualVariable vvar, OldInstruction oldInstr, Alias::MemoryLocation defLocation,
OldBlock defBlock, int defRank, int defOffset, OldBlock useBlock, int useRank
|
chiInstr = getChi(oldInstr) and
vvar = Alias::getResultMemoryLocation(oldInstr).getVirtualVariable() and
hasDefinitionAtRank(vvar, _, defBlock, defRank, defOffset) and
hasDefinitionAtRank(vvar, defLocation, defBlock, defRank, defOffset) and
hasUseAtRank(vvar, useBlock, useRank, oldInstr) and
definitionReachesUse(vvar, defBlock, defRank, useBlock, useRank) and
result = getDefinitionOrChiInstruction(defBlock, defOffset, vvar, _)

View File

@@ -1204,17 +1204,6 @@ class ShiftRightInstruction extends BinaryBitwiseInstruction {
ShiftRightInstruction() { this.getOpcode() instanceof Opcode::ShiftRight }
}
/**
* An instruction that shifts its left operand to the right by the number of bits specified by its
* right operand.
*
* Both operands must have an integer type. The result has the same type as the left operand.
* The leftmost bits are zero-filled.
*/
class UnsignedShiftRightInstruction extends BinaryBitwiseInstruction {
UnsignedShiftRightInstruction() { this.getOpcode() instanceof Opcode::UnsignedShiftRight }
}
/**
* An instruction that performs a binary arithmetic operation involving at least one pointer
* operand.

View File

@@ -45,7 +45,7 @@ class Operand extends TStageOperand {
this = reusedPhiOperand(use, def, predecessorBlock, _)
)
or
this = chiOperand(_, _)
exists(Instruction use | this = chiOperand(use, _))
}
/** Gets a textual representation of this element. */

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

@@ -298,11 +298,11 @@ class TranslatedStringLiteralInitialization extends TranslatedDirectInitializati
opcode instanceof Opcode::Store and
resultType = getTypeForPRValue(expr.getType())
or
exists(int elementCount |
exists(int startIndex, int elementCount |
// If the initializer string isn't large enough to fill the target, then
// we have to generate another instruction sequence to store a constant
// zero into the remainder of the array.
zeroInitRange(_, elementCount) and
zeroInitRange(startIndex, elementCount) and
(
// Create a constant zero whose size is the size of the remaining
// space in the target array.

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

@@ -1204,17 +1204,6 @@ class ShiftRightInstruction extends BinaryBitwiseInstruction {
ShiftRightInstruction() { this.getOpcode() instanceof Opcode::ShiftRight }
}
/**
* An instruction that shifts its left operand to the right by the number of bits specified by its
* right operand.
*
* Both operands must have an integer type. The result has the same type as the left operand.
* The leftmost bits are zero-filled.
*/
class UnsignedShiftRightInstruction extends BinaryBitwiseInstruction {
UnsignedShiftRightInstruction() { this.getOpcode() instanceof Opcode::UnsignedShiftRight }
}
/**
* An instruction that performs a binary arithmetic operation involving at least one pointer
* operand.

View File

@@ -45,7 +45,7 @@ class Operand extends TStageOperand {
this = reusedPhiOperand(use, def, predecessorBlock, _)
)
or
this = chiOperand(_, _)
exists(Instruction use | this = chiOperand(use, _))
}
/** Gets a textual representation of this element. */

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