Add lifetime package

[knewbury01]

Description

add rules for lifetime package, two shared rules, one with improvements upon previous, one as simple shared

Change request type

• Release or process automation (GitHub workflows, internal scripts)
• Internal documentation
• External documentation
• Query files (.ql, .qll, .qls or unit tests)
• External scripts (analysis report or other code shipped as part of a release)

Rules with added or modified queries

• No rules added
• Queries have been added for the following rules:
  • RULE-11-6-2
  • RULE-6-8-3
• Queries have been modified for the following rules:
  • rule number here

Release change checklist

A change note (development_handbook.md#change-notes) is required for any pull request which modifies:

• The structure or layout of the release artifacts.
• The evaluation performance (memory, execution time) of an existing query.
• The results of an existing query in any circumstance.

If you are only adding new rule queries, a change note is not required.

Author: Is a change note required?

• Yes
• No

🚨🚨🚨
Reviewer: Confirm that format of shared queries (not the .qll file, the
.ql file that imports it) is valid by running them within VS Code.

• Confirmed

Reviewer: Confirm that either a change note is not required or the change note is required and has been added.

• Confirmed

Query development review checklist

For PRs that add new queries or modify existing queries, the following checklist should be completed by both the author and reviewer:

Author

• Have all the relevant rule package description files been checked in?
• Have you verified that the metadata properties of each new query is set appropriately?
• Do all the unit tests contain both "COMPLIANT" and "NON_COMPLIANT" cases?
• Are the alert messages properly formatted and consistent with the style guide?
• Have you run the queries on OpenPilot and verified that the performance and results are acceptable?
  As a rule of thumb, predicates specific to the query should take no more than 1 minute, and for simple queries be under 10 seconds. If this is not the case, this should be highlighted and agreed in the code review process.
• Does the query have an appropriate level of in-query comments/documentation?
• Have you considered/identified possible edge cases?
• Does the query not reinvent features in the standard library?
• Can the query be simplified further (not golfed!)

Reviewer

• Have all the relevant rule package description files been checked in?
• Have you verified that the metadata properties of each new query is set appropriately?
• Do all the unit tests contain both "COMPLIANT" and "NON_COMPLIANT" cases?
• Are the alert messages properly formatted and consistent with the style guide?
• Have you run the queries on OpenPilot and verified that the performance and results are acceptable?
  As a rule of thumb, predicates specific to the query should take no more than 1 minute, and for simple queries be under 10 seconds. If this is not the case, this should be highlighted and agreed in the code review process.
• Does the query have an appropriate level of in-query comments/documentation?
• Have you considered/identified possible edge cases?
• Does the query not reinvent features in the standard library?
• Can the query be simplified further (not golfed!)

[MichaelRFairhurst]

The test cases look great Kristen, thank you!

[knewbury01]

borrowed this from here

[Copilot]

Pull request overview

This PR adds a new "Lifetime" rule package for MISRA C++ 2023, implementing two new rules: RULE-11-6-2 (reading uninitialized objects) and RULE-6-8-3 (automatic storage assigned to wider lifetime). It also enhances the shared ReadOfUninitializedMemory query to detect uninitialized reads through new expressions without value initialization, and extends the AllocatedObjectIdentity class to recognize new/new[] expressions as allocated storage.

Changes:

• Adds the Lifetime rule package with RULE-11-6-2 and RULE-6-8-3, including exclusion metadata, query files, test references, and supporting infrastructure.
• Enhances the ReadOfUninitializedMemory shared library to detect uninitialized memory accessed via new expressions (without value initialization), incorporating a copied InitializationFunctions.qll from the CodeQL standard library.
• Extends AllocatedObjectIdentity in CppObjects.qll to include new/new[] expressions (excluding placement new) for storage duration tracking.

Reviewed changes

Copilot reviewed 17 out of 17 changed files in this pull request and generated 6 comments.

Show a summary per file

File	Description
rules.csv	Updates RULE-11-6-2 to reference the Lifetime package
rule_packages/cpp/Lifetime.json	New rule package definition for RULE-11-6-2 and RULE-6-8-3
cpp/common/src/codingstandards/cpp/exclusions/cpp/Lifetime.qll	Autogenerated exclusion module for Lifetime package
cpp/common/src/codingstandards/cpp/exclusions/cpp/RuleMetadata.qll	Integrates Lifetime package into rule metadata
cpp/misra/src/rules/RULE-11-6-2/ValueOfAnObjectMustNotBeReadBeforeItHasBeenSet.ql	New query for RULE-11-6-2 using shared ReadOfUninitializedMemory
cpp/misra/src/rules/RULE-6-8-3/AutomaticStorageAssignedToObjectGreaterLifetime.ql	New query for RULE-6-8-3 using shared DoNotCopyAddressOfAutoStorage
cpp/common/src/codingstandards/cpp/rules/readofuninitializedmemory/ReadOfUninitializedMemory.qll	Enhanced to handle new without value init
cpp/common/src/codingstandards/cpp/rules/readofuninitializedmemory/InitializationFunctions.qll	Copied from CodeQL standard library for init function tracking
cpp/common/src/codingstandards/cpp/lifetimes/CppObjects.qll	Extended AllocatedObjectIdentity for new/new[] expressions
cpp/common/test/rules/readofuninitializedmemory/test.cpp	Added test cases for new uninitialized memory scenarios
cpp/common/test/rules/readofuninitializedmemory/ReadOfUninitializedMemory.expected	Updated expected results for new test cases
cpp/common/test/rules/donotcopyaddressofautostorageobjecttootherobject/test.cpp	Added test cases for auto-storage address escape
cpp/common/test/rules/donotcopyaddressofautostorageobjecttootherobject/DoNotCopyAddressOfAutoStorageObjectToOtherObject.expected	Updated expected results for new test cases
cpp/misra/test/rules/RULE-6-8-3/AutomaticStorageAssignedToObjectGreaterLifetime.testref	Test reference for RULE-6-8-3
cpp/misra/test/rules/RULE-11-6-2/ValueOfAnObjectMustNotBeReadBeforeItHasBeenSet.testref	Test reference for RULE-11-6-2
change_notes/2026-02-03-uninitialized-mem-improve.md	Change note for shared query modifications
.vscode/tasks.json	Added Lifetime to VS Code task configuration

Comments suppressed due to low confidence (1)

cpp/common/src/codingstandards/cpp/rules/readofuninitializedmemory/ReadOfUninitializedMemory.qll:207

• Typo in docstring: "Get a access" should be "Gets an access".

   * Get a access of the variable that is assumed to initialize the variable.

[Copilot]

The RULE-11-6-2 row is missing a trailing comma after Import. All other rows in the CSV consistently end with a comma after the difficulty field (e.g., line 929: ...,Import,). This could cause parsing issues with downstream tools that process the CSV.

Suggested change

	cpp,MISRA-C++-2023,RULE-11-6-2,Yes,Mandatory,Undecidable,System,The value of an object must not be read before it has been set,A8-5-0,Lifetime,Import
	cpp,MISRA-C++-2023,RULE-11-6-2,Yes,Mandatory,Undecidable,System,The value of an object must not be read before it has been set,A8-5-0,Lifetime,Import,

[MichaelRFairhurst]

OK, there's a lot of text here, but I think it boils down to actually pretty simple changes, its just complicated to think through and fully explain the implications of X vs Y 😕 I would have written less if I olny had more time! :)

I'm 100% glad that you pursued the approach. I think with a small few tweaks this implementation will be very precise, and handling new int without a major overhaul was absolutely the correct choice. 🎉

Let me know if anything here is unclear or if I missed anything!

[MichaelRFairhurst]

Consider adding or result = c.getArgument(j), that would handle:

void f(int *p) {
  *p = 0;
}

void g() {
  int *p = new int;
  f(p); // compliant
  use(*p); // compliant
}

In fact we might want to rely on .isRValueCategory() and not exists(Call ... | ... result = c.getArgument(j)).

In general, if x appears in &x, that should mean imply not .isRValueCategory() for x. So technically, int *p = &x would previously have not been considered a use and now will be.

Overall, isRValueCategory() should find only cases where x could not be initialized (only reads + copies). So we were already very conservative in what we consider a use.

The catch is that copying or reading x may be how someone initializes *x, such as in the above example with f(p). So with our new x+*x analysis, we can't guarantee isRValueCategory() for x means we don't initialize *x. Really any rvalue use could lead to initialization of *x.

We also don't want to lose the ability to analyze uninitialized pointers more generally (as in, int *p; f(p); which is never valid and can segfault, and the current query should catch).

Ultimately, for *x vs x:

• if *x is used as an rvalue, that's problematic if x had a non-init new
• otherwise, all uses of x as an rvalue require that x had any initializer
• If x is used as an lvalue in *x, that's problematic if x had no initializer
• All other lvalue uses of x are valid
• It's always valid if *x is used as an lvalue if x had any initializer

Ultimately, I think it would probably be fine to say:

• track x = new int as an "uninitialized variable"
• count rvalue x as a use if not new int
• count lvalue x as a use if used in *x and not new int
• count rvalue *x as a use if has new int
• don't count lvalue *x as a use in and of itself