Run sanitizers on CI

[taiki-e]

This adds AddressSanitizer/MemorySanitizer/ThreadSanitizer tests to CI.

See #589 (tsan: data race in deque), #644 (tsan: data race in atomic), and #614 (asan: memory leak in skiplist) for issues reported by sanitizers.

Closes #154
Closes #589
Closes #644

Refs:

• https://doc.rust-lang.org/nightly/unstable-book/compiler-flags/sanitizer.html
• https://github.com/google/sanitizers/wiki

[taiki-e]

I don't know why this test fails when we reduce internal::MAX_OBJECTS, but it also fails when I run a normal test with --cfg crossbeam_sanitize. (And, except for cfg(test) code, cfg(crossbeam_sanitize) only changes internal::MAX_OBJECTS.)

$ RUSTFLAGS='--cfg crossbeam_sanitize' cargo test --lib
...

---- collector::tests::incremental stdout ----
thread 'collector::tests::incremental' panicked at 'assertion failed: curr - last <= 1024', crossbeam-epoch/src/collector.rs:227:13
stack backtrace:
   0: rust_begin_unwind
             at /rustc/80184183ba0a53aa4f491753de9502acd3d6920c/library/std/src/panicking.rs:493:5
   1: core::panicking::panic_fmt
             at /rustc/80184183ba0a53aa4f491753de9502acd3d6920c/library/core/src/panicking.rs:92:14
   2: core::panicking::panic
             at /rustc/80184183ba0a53aa4f491753de9502acd3d6920c/library/core/src/panicking.rs:50:5
   3: crossbeam_epoch::collector::tests::incremental
             at ./src/collector.rs:227:13
   4: crossbeam_epoch::collector::tests::incremental::{{closure}}
             at ./src/collector.rs:204:5
   5: core::ops::function::FnOnce::call_once
             at /Users/taiki/.rustup/toolchains/nightly-x86_64-apple-darwin/lib/rustlib/src/rust/library/core/src/ops/function.rs:227:5
   6: core::ops::function::FnOnce::call_once
             at /rustc/80184183ba0a53aa4f491753de9502acd3d6920c/library/core/src/ops/function.rs:227:5
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.


failures:
    collector::tests::incremental

test result: FAILED. 35 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out; finished in 2.01s

error: test failed, to rerun pass '--lib'

[jeehoonkang]

I think it's worth reporting in a separate issue. @taiki-e would you please?

[taiki-e]

That makes sense. I've filed #662.

[taiki-e]

This should be ready for review now.

See #589 (tsan: data race in deque), #644 (tsan: data race in atomic), and #614 (asan: memory leak in skiplist) for issues reported by sanitizers.

[taiki-e]

@jeehoonkang Are there any changes that you'd like me to make?

[jeehoonkang]

@taiki-e LGTM. Please feel free to let bors merge it.

[jeehoonkang]

I think it's worth reporting in a separate issue. @taiki-e would you please?

[taiki-e]

bors r=jeehoonkang

[bors]

Build succeeded:

• ci