Consider running unit tests (or a subset of them) under Miri

[wesleywiser]

This would help find issues like #530 and #544. Miri is a dynamic analysis (like ASAN or valgrind) so it likely won't find every instance of UB in your code but will help you with any code that is covered by your unit tests and does not run into the platform limitations the tool has.

When investigating #544, I was able to run the tool locally by finding a nightly toolchain that corresponds to Rust 1.85 (such as nightly-2025-01-03), installing it and then running whichever unit tests, you want. For example:

> rustup toolchain install nightly-2025-01-03 --component miri
> cargo +nightly-2025-01-03 miri test -- create_fill_check_test
...
error: Undefined Behavior: attempting a write access using <326320> at alloc132202[0x50], but that tag does not exist in the borrow stack for this location
   --> components\patina_adv_logger\src\memory_log.rs:294:9
    |
294 | /         ptr::write_volatile::<Self>(
295 | |             address as *mut Self,
296 | |             Self::new(log_entry.phase, log_entry.level, log_entry.timestamp, log_entry.data.len() as u16),      
297 | |         );
    | |         ^
    | |         |
    | |_________attempting a write access using <326320> at alloc132202[0x50], but that tag does not exist in the borrow stack for this location
    |           this error occurs as part of an access at alloc132202[0x50..0x68]
    |
    = help: this indicates a potential bug in the program: it performed an invalid operation, but the Stacked Borrows rules it violated are still experimental
    = help: see https://github.com/rust-lang/unsafe-code-guidelines/blob/master/wip/stacked-borrows.md for further information
help: <326320> was created by a SharedReadOnly retag at offsets [0x18..0x7c]
   --> components\patina_adv_logger\src\memory_log.rs:181:33
    |
181 |         let address = unsafe { (self as *const AdvLoggerInfo).byte_offset(current_offset as isize) };
    |                                 ^^^^
    = note: BACKTRACE (of the first span) on thread `memory_log::tests::create_fill_check_test`:
    = note: inside `memory_log::AdvLoggerMessageEntry::init_from_memory` at components\patina_adv_logger\src\memory_log.rs:294:9: 297:10
note: inside `memory_log::AdvLoggerInfo::add_log_entry`
   --> components\patina_adv_logger\src\memory_log.rs:182:18
    |
182 |         unsafe { AdvLoggerMessageEntry::init_from_memory(address as *mut c_void, message_size, log_entry) }       
    |                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: inside `memory_log::tests::create_fill_check_test`
   --> components\patina_adv_logger\src\memory_log.rs:407:29
    |
407 |             let log_entry = log.unwrap().add_log_entry(entry);
    |                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
note: inside closure
   --> components\patina_adv_logger\src\memory_log.rs:391:32
    |
390 |     #[test]
    |     ------- in this procedural macro expansion
391 |     fn create_fill_check_test() {
    |                                ^
    = note: this error originates in the attribute macro `test` (in Nightly builds, run with -Z macro-backtrace for more info)

note: some details are omitted, run with `MIRIFLAGS=-Zmiri-backtrace=full` for a verbose backtrace

error: aborting due to 1 previous error; 2 warnings emitted

test memory_log::tests::create_fill_check_test ... error: test failed, to rerun pass `-p patina_adv_logger --lib`

You'll probably also discover that some of your tests have UB (at least this one did #545) so I think this would be a great driver to improve quality!

[cfernald]

@Javagedes @makubacki for awareness

[makubacki]

Closing this discussion as GitHub issues are being used to track further miri work.