I'm sure this doesn't intend to be misleading but I think it can mislead people anyway.
TSan should detect races but just like "repeatedly run the code in a loop" it's not actually checking what you wrote doesn't have races, it's just reporting any which happened during testing. This is valuable - if you eliminate a race you've almost certainly fixed a real bug, maybe a very subtle bug you'd have cursed for years otherwise, but you can't use TSan to prove there are no further bugs of this kind.
Tools to prove this stuff are often much harder to bring into a project, but you should put that work in if the difference between "It's probably fine" and "I proved it's correct" sounds necessary to you or for your work. I don't want my pacemaker to be "probably fine" but it's OK if the Youtube music player on my phone wasn't proved correct.
The real issue here isn't that TSAN is low-confidence about absence of data races in C++. The issue is that even if it statically proved the absence of data races in the C++ sense, that still wouldn't imply that your algorithm is race-free. Because it's trivial to patch every single instance of a data race by using an atomic to get TSAN to shut up, but that in no way implies the higher-level concurrent algorithm works correctly. (e.g., if two threads perform an atomic load of some variable, then add one, then perform an atomic store back in, they would obviously stomp on each other, but TSAN wouldn't care.) It's more like a UB checker than a correctness verifier.
Not to talk my own book, but there is a well-known alternative to C++ that can actually guarantee the absence of data races.
TSAN does not check for race conditions in general, and doesn't claim to do so at all as the documentation doesn't include the term race condition anywhere. TSAN is strictly for checking data races and deadlocks.
Consequently this claim is false:
>The issue is that even if it statically proved the absence of data races in the C++ sense, that still wouldn't imply that your algorithm is race-free.
Race-free code means absence of data races, it does not mean absence of the more general race condition. If you search Google Scholar for race free programming you'll find no one uses the term race-free to refer to complete absence of race conditions but rather to the absence of data races.
>Feel free to mentally rewrite my comment with whatever preferred terminology you feel would get my points across.
If I rewrite your comment to use data race, then your comment is plainly incorrect since the supporting example you give is not a data race but a race condition.
If I rewrite your comment to use race condition, then your comment is also incorrect since TSAN doesn't detect race conditions in general and doesn't claim to, it detects data races.
So what exactly am I supposed to do in order to make sense of your post?
The idea that you'd talk about the pros and cons of a tool like TSAN without knowing the difference between a race condition and a data race is kind of absurd. That you'd claim my clarification of these terms for the sake of better understanding your point is a form of pedantry is sheer hubris.
class Main {
private static String s = "uninitialized";
public static void main(String[] args) {
Thread t = new Thread() {
public void run() { s = args[0]; }
};
t.start();
System.out.println(s);
}
}
Yes, your program contains a data race, by the definition used in the JLS. The set of outcomes you may observe from a data race are specified. I'm not sure if this choice was intentional or not, but there is a guarantee that you will either print the argument or "uninitialized" and no other behavior, because String relies on final field semantics. This is would not be true in c/c++ where the equivalent code is undefined behavior and you could see any result.
In Java you can have a data race and use it productively for certain niche cases, like String.hashcode - I've also contributed some to the Guava concurrency library. This is not true in c/c++ where data races (by their definition) are undefined behavior. If you want to do the tricks you can in racy Java without UB, you have to declare your variables atomic and use relaxed memory order and possibly fences.
You see Java has a specific memory ordering model (many languages just give you a big shrug, including C before it adopted the C++ 11 model but Java spells out what happens) and that model is very sophisticated so it has an answer to what happens here.
Because we raced s, we lose Sequential Consistency. This means in general (this example is so trivial it won't matter) humans struggle to understand what's going on in their program, which makes debugging and other software engineering impractical. But, unlike C++ loss of Sequential Consistency isn't fatal in Java, instead we're promised that when s is observed in the main thread it will either be that initial "uninitialized" string or it will have the args[0] value, ie the name of the program because these are the only two values it could have and Java does not specify which of them observed in this case.
You could think of this as "atomic access" and that's likely the actual implementation in this case, but the Java specification only promises what I wrote.
In C++ this is game over, the language standard specifically says it is Undefined Behaviour to have any data race and so the behaviour of your program is outside the standard - anything at all might happen.
[Edited: I neglected originally to observe that s is set to "uninitialized", and instead I assumed it begins as null]
I have no idea what you mean here. Loss of sequential consistency is in no way fatal in C++. There are several access modes that are specifically designed to avoid sequential consistency.
Regarding the rest of your comment:
You're making exactly my point though. These are guaranteed atomic accesses -- and like you said, we are guaranteed to see either the old or new value, and nothing else -- and yet they are still data races. Anyone who agrees this is a data race despite the atomicity must necessarily understand that atomics don't imply lack of data races -- not in general CS terminology.
The only way it's correct to say they are mutually exclusive is when you define "data race" as they did in the C++ standard, to imply a non-atomic access. Which you're welcome to do, but it's an incredibly pedantic thing to do because, for probably >95% of the users of C++ (and probably even of TSAN itself), when they read "data race", they assume it to mean the concept they understand from CS. They don't know that the ISO standard defines it in its own peculiar way. My point here was to convey something to normal people rather than C++ committee language lawyers, hence the use of the general term.
Sure, if you work really hard you can write a C++ program which doesn't meet the 6.9.2.2 intro.races definition of a data race but does nevertheless lose sequential consistency and so it has in some sense well-defined meaning in C++ but humans can't usefully reason about it. You'll almost certainly trip and write UB when attempting this, but assuming you're inhumanly careful or the program is otherwise very simple so that you don't do that it's an exception.
My guess is that your program will be miscompiled by all extant C++ compilers and you'll be unhappy with the results, and further that if you can get committee focus on this at all they will prioritize making your program Undefined in C++ rather than "fixing" compilers.
Just don't do this. The reason for the exclusions in 6.9.2.2 is that what we want people to do is write correct SC code but using primitives which themselves can't guarantee that, so the person writing the code must do so correctly. The reason is not that somehow C++ programmers are magicians and the loss of SC won't negatively impact the correctness of code they attempt to write, quite the contrary.
A data-race does not imply a non-atomic operation, it implies an unsynchronized operation. Different languages have different requirements for what constitutes a synchronized operation, for example in Python all reads and writes are synchronized, in Java synchronization is generally accomplished through the use of a monitor or a volatile operation, and in C++ synchronization is generally accomplished through the use of std::atomic operations.
The fact that in C++ atomic operations result in synchronization, while in Java atomic operations are not sufficient for synchronization is not some kind of language lawyering or pedantry, it's because Java makes stronger guarantees about the consequences of a data race versus C++ where a data race can result in any arbitrary behavior whatsoever. As such it should not come as a surprise that the cost of those stronger guarantees is that Java has stronger requirements for data race free programs.
But of course this is mostly a deflection, since the discussion was about the use of TSAN, which is a data race detector for C and C++, not for Java. Hence to the extent that TSAN detects data races, it detects them with respect to C and C++'s memory model, not Java's memory model or Python's memory model, or any other memory model.
The objection I originally laid out was your example of a race condition, an example which can happen even in the absence of parallelism (ie. a single-core CPU) and even in the absence of multithreaded code altogether (your example can happen in a single threaded application with the use of coroutines). TSAN makes no claim with regards to detecting race conditions in general, it only seeks to detect data races and data races as they pertain the C and C++ memory models.
Let me lay this out very explicitly. This comment will likely be my final one on the matter as this back-and-forth is getting quite tiresome, and I'm not enjoying it, especially with the swipes directed at me.
Please take a look at the following two C++ and Java programs: https://godbolt.org/z/EjWWac1bG
For the sake of this discussion, assume the command-line arguments behave the same in both languages. (i.e. ignore the C++ argv[0] vs. Java args[0] distinction and whatnot.)
Somehow, you simultaneously believe (a) that Java program contains data races, while believing (b) the C++ program does not.
This is a self-contradictory position. The programs are well-defined, direct translations of each other. They are equivalent in everything but syntax. If one contains a data race, so must the other. If one does not, then neither can the other.
This implies that TSAN does not detect "data races" as it is usually understood in the CS field -- it does not detect anything in the C++ program. What TSAN detects is only a particular, distinct situation that the C++ standard also happens to call a "data race". So if you're talking to a C++ language lawyer, you can say TSAN detects all data races within its window/buffer limits. But if you're talking to someone who doesn't sleep with the C++ standard under their pillow, they're not going to realize C++ is using a language-specific definition, and they're going to assume it flags programs like the equivalent of the Java program above, which has a data race but whose equivalent TSAN would absolutely not flag.
That your position hinges on thinking all languages share the same memory model suggests a much deeper failure to understand some of the basic principles of writing correct parallel software and while numerous people have tried to correct you on this, you still seem adament on doubling down on your position so I don't think there is much point in continuing this.
I never suggested "all languages share the same memory model". You're severely mischaracterizing what I've said and putting words in my mouth.
What I said was (a) data races are generals properties of programs that people can and do discuss in language-agnostic contexts, and (b) it makes no sense to say two well-defined, equivalent programs differ in whether they have data races. Reducing these statements down to "all programs share the same memory model" as if they're somehow equivalent makes for an incredibly unfaithful caricature of everything I've said. Yes, I can see there's no point in continuing.
"Data race" is a specific property defined by a memory model, which is normally part of a language spec; it's not usually understood as an abstract property defined in terms of outcome, at least not usefully. If you talk about data races as abstract and language-spec-agnostic properties, then yes, you're assuming a memory model that's shared across all programs and their languages.
Really? To me [1] sure doesn't look useless:
> We use the standard definition of a data race: two memory accesses to the same address can be scheduled on different threads to happen concurrently, and at least one of the accesses is a write [16].
You're welcome to look at the [16] it cites, and observe that it is from 1991, entirely in pseudocode, with no mention of a "memory model". It so happens to use the phrase "access anomaly", but evidently that is used synonymously here, per [1].
> If you talk about data races as abstract and language-spec-agnostic properties, then yes, you're assuming a memory model that's shared across all programs and their languages.
No, nobody is assuming such a thing. Different memory models can still exhibit similar properties when analyzing file accesses. Just like how different network models can exhibit similar properties (like queue size bounds, latency, etc.) when discussing network communication. Just because two things are different that doesn't mean they can't exhibit common features you can talk about in a generic fashion.
TSAN defines its own specific definition of "data race" which is invariant to the languages that it evaluates. In particular the C++ definition of "data race" is more lenient than the TSAN definition of "data race"; C++ doesn't consider two atomic accesses of the same memory (at least one being a write) under memory_order_relaxed to be a data race, but TSAN does.
TSAN _could_ detect the C++ program as having a data race, for sure. And if it could evaluate Java programs, it _could_ also detect the Java program as having a data race, too.
I'm confused what you're saying here. If you take the program I linked, which uses relaxed ordering, and add -fsanitize=thread, TSAN doesn't flag anything. That seems inconsistent with what you're saying?
P.S. note that even using memory_order_seq_cst wouldn't change anything in that particular program.
You've now received many comments from many different commenters that all kind of say the same thing, which is basically that your understanding of a "data race" is not really accurate. Those comments have included pretty detailed information as to exactly how and when and why your definition isn't totally correct. If I were you I'd take the L and maybe read up a bit.
I stated as much in my own first comment, with more details on when/why this does/doesn't occur.
> at least as TSAN defines data race, which is stricter than how C++ defines data race. [...] If I were you I'd take the L and maybe read up a bit.
Before assuming I haven't: I have. And the reading does not agree [1] [2] with your idea that TSAN sometimes considers relaxed atomic writes to be data races. Hence my replies. Remember, you wrote:
>> C++ doesn't consider two atomic accesses of the same memory (at least one being a write) under memory_order_relaxed to be a data race, but TSAN does.
I have not seen a single word anywhere -- not in the docs, not in example code, and (I think) not even from anyone here other than you -- suggesting that TSAN considers memory_order_relaxed writes to constitute a data race. It certainly does not flag them in the most trivial test I could think of, which I had already linked here [3].
If this is just my ignorance or misunderstanding, then instead of telling me to go do my own reading, please enlighten me and provide one link or example that demonstrates that TSAN considers atomic writes to the same memory to be a data race? I would be very glad to learn this, as I wasn't aware of this before, and am not seeing anything suggesting such.
> Those comments have included pretty detailed information as to exactly how and when and why your definition isn't totally correct.
I have linked to papers going back decades showing that "data race" has a general definition that don't entirely match up with what people have said. I myself have also explained in great detail how the general definition differs from the C++ one. I don't know what else to possibly provide here, but I'm done.
[1] https://clang.llvm.org/docs/ThreadSanitizer.html
[2] https://github.com/google/sanitizers/wiki/threadsanitizercpp...
For something like TSan, which allows programs to execute normally with additional instrumentation, this timing matters, and so it's not a great example. An equivalent program being simulated in something like Loom would be much more convincing.
I'm a little confused, as you agree with your parent commenter that TSan not raising a flag is not conclusive. But you also appear to be using TSan not flagging the program as some kind of evidence in the same comment.
The answer to your questions is that timing is not the issue in my example. You can notice this easily if you strip std::atomic<> from the type. TSAN can and does catch it just fine. The atomicity itself is what tells TSAN to not consider this a data race.
What probably threw you off was that I (sloppily) used "timing" as a way to say "proximity in the history buffer". [1] It's not wall clock time that matters, it's the number of memory accesses that fit in TSAN's history buffer. (This should also explain your confusion w.r.t. "tight" timing.)
Hence, the conclusivity depends entirely on the reason it wasn't flagged. (This is why I explained the failure modes quite precisely in my very first comment: not all of them have randomness involved.) If it wasn't flagged because the history buffer wasn't large enough, then obviously it's not conclusive. But if it wasn't flagged because TSAN noticed it and deliberately exempted it, then obviously it doesn't consider it a data race.
[1] https://github.com/google/sanitizers/wiki/ThreadSanitizerFla...
If you strip the std::atomic from your example, then you obviously lose read/write atomicity on the value, which should be trivial for something like TSAN to detect and flag.
> The atomicity itself is what tells TSAN to not consider this a data race ... the conclusivity depends entirely on the reason it wasn't flagged. (This is why I explained the failure modes quite precisely in my very first comment: not all of them have randomness involved.)
"The conclusivity" can only ever be "inconclusive" or "has a data race", it can never be "does not have a data race", because that's just not how TSAN (or any similar tools) work. See [1]. In your original C++ program there is no happens-before relationship between the thread you spawn and the main thread, so there is a data race by the TSAN definition, even though the read and the write are atomic, and even if a given execution of that code by TSAN doesn't yield an error. It's not about timing, at least not exactly -- it's about the guarantees of the scheduler and its execution of threads, which is non-deterministic without explicit synchronization in the application (or something along those lines)..!
[1] https://github.com/google/sanitizers/wiki/ThreadSanitizerAbo...
But the write to the static variable from the second thread is entirely unordered with respect to the read, despite the atomicity. If lack of ordering is your criterion for data races, doesn't that imply there is a data race between that write and that read?
It's not my criterion, it's defined in the language standard. You can't have a data race on an atomic except for std::atomic_init. The reads on the string contents themselves are ordered because the args string is initialized before the thread is launched and the other one is static. If the launched thread allocated a new string and then populated the atomic, that would be a data race, unless stronger memory order was used by both the writer and the reader.
You can go the other way, porting that c++ to Java using varhandle and opaque memory order.
1. Per 17.4.5 your example can lead to a data race.
"When a program contains two conflicting accesses (§17.4.1) that are not ordered by a happens-before relationship, it is said to contain a data race."
2. Per 17.7 the variable s is accessed atomically.
"Writes to and reads of references are always atomic, regardless of whether they are implemented as 32-bit or 64-bit values."
As per 17.4.3, a data-race free program will not have this kind of behavior where two threads are in a perpetually inconsistent state, as the spec says "If a program has no data races, then all executions of the program will appear to be sequentially consistent."
So while atomicity protects against certain types of data corruption, it does not protect against data races.
[1] https://docs.oracle.com/javase/specs/jls/se24/html/jls-17.ht...
You seem to conflate the concepts of "data race" and "race condition", which are not the same thing.
Two threads writing to the same memory location without synchronization (without using atomic operations, without going thru a synchronization point like a mutex, etc.) is a data race, and almost certainly also a race condition. If access to that memory location is synchronized, whether thru atomics or otherwise, then there's no data race, but there can still be a race condition.
This isn't a pedantic distinction, it's actually pretty important.
Not only are you incorrect, it’s even worse than you might think. Unsynchronized access to data in c++ is not only a data race, it’s explicitly undefined behavior and the compiler can choose to do whatever in response of an observed data race (which you are promising it isn’t possible by using the language).
You are also misinformed about the efficacy of TSAN. Even in TSAN you have to run it in a loop - if TSAN never observes the specific execution order in a race it’ll remain silent. This isn’t a theoretical problem but a very real one you must deeply understand if you rely on these tools. I recall a bug in libc++ with condition_variable and reproducing it required running the repro case in a tight loop like a hundred times to get even one report. And when you fixed it, how long would you run to have confidence it was actually fixed?
And yes, race conditions are an even broader class of problems that no tool other than formal verification or DST can help with. Hypothesis testing can help mildly but really you want at least DST to probabilistically search the space to find the race conditions (and DST’s main weakness aside from the challenge of searching a combinatorial explosion of states is that it still relies on you to provide test coverage and expectations in the first place that the race condition might violate)
So yeah, just fix 'em. Most of them you dodged a bullet, and even when you didn't it will now be easier for the next person to reason about the software.
It's been a hobby of mine to collect concurrency examples from books and blog posts and simulating them in Temper, my Rust memory model simulator. As far as I know, it's the largest Rust/C++11 memory model test suite on the internet (but I'm happy to be corrected).
This is the file for Rust Atomics and Locks:
https://github.com/reitzensteinm/temper/blob/main/memlog/tes...
I didn't find any bugs in the examples, but with how good the book was, I didn't expect to :)
The Williams book for C++ contains many of the same ideas (Rust's memory model is a copy/paste from C++11 without the now deprecated Consume) and I can highly recommend that too.
Highly recommend!
Nitpick, but they absolutely can be split into several instructions, and this is the most common way it’s implemented on RISClike processors, and also single instructions aren’t necessarily atomic.
The actual guarantee is that the entire operation (load, store, RMW, whatever) occurs in one “go” and no other thread can perform an operation on that variable during this atomic operation (it can’t write into the low byte of your variable as you read it).
It’s probably best euphemismed by imagining that every atomic operation is just the normal operation wrapped in a mutex, but implemented in a much more efficient manner. Of course, with large enough types, Atomic variables may well be implemented via a mutex
Sort-of, but that's not quite right: if you imagine it as "taking a mutex on the memory", there's a possibility of starvation. Imagine two threads repeatedly "locking" the memory location to update it. With a mutex, it's possible that one of them get starved, never getting to update the location, stalling indefinitely.
At least x86 (and I'm sure ARM and RISC-V as well) make a much stronger progress guarantee than a mutex would: the operation is effectively wait-free. All threads are guaranteed to make progress in some finite amount of time, no one will be starved. At least, that's my understanding from reading much smarter people talking about the cache synchronization protocols of modern ISAs.
Given that, I think a better mental model is roughly something like the article describes: the operation might be slower under high contention, but not "blocking" slow, it is guaranteed to finish in a finite amount of time and atomically ("in one combined operation").
Definitely a good clarification though yeah, important
Executions of atomic functions that are either defined to be lock-free
(33.5.10) or indicated as lock-free (33.5.5) are lock-free executions.
When one or more lock-free executions run concurrently, at least one should
complete.
[Note 3 : It is difficult for some implementations to provide absolute
guarantees to this effect, since repeated and particularly inopportune
interference from other threads could prevent forward progress, e.g., by
repeatedly stealing a cache line for unrelated purposes between load-locked
and store-conditional instructions. For implementations that follow this
recommendation and ensure that such effects cannot indefinitely delay progress
under expected operating conditions, such anomalies can therefore safely be
ignored by programmers. Outside this document, this property is sometimes
termed lock-free. — end note]
[1] https://open-std.org/jtc1/sc22/wg21/docs/papers/2023/n4950.p...
That's to enable very minimal hardware implementations that can only track one line at a time.
Atomics and Concurrency https://news.ycombinator.com/item?id=38964675 (January 11, 2024 — 106 points, 48 comments)
Actually, nothing ever sets head so I'm not sure how anything is ever dequeued. Probably the implementation is incomplete and the queue needs a sentinel node somewhere.
Binary search provided. Pair abstraction provided. Lower bound for binary search yup.
Auto for type inference and fast as hell on top of it. Crazy powerful lang and also multiplayform threads.
In languages with type inference when it could be A or B and it's left to infer the type the compiler just always says it could be A or B, so the programmer needs to disambiguate.
But with deduction in some cases C++ will deduce that you meant A even though you could instead have explicitly chosen B here instead, and too bad if you did mean B because that's not what it deduced.
Delayed reclamation means that "when we don't need this Goat" will sometimes be arbitrarily delayed after we apparently didn't need it, and it's no longer possible to say for sure by examining the program. This will almost always be a trade you're willing to make, but you need to know it exists, this is not a fun thing to discover while debugging.
Hazard pointers are a bit fiddlier AIUI, since the reclamation step must be triggered explicitly and verify that no hazard pointers exist to the object being reclaimed, so it is quite possible that you won't promptly reclaim the object.
Maybe I'm wrong for some reason and there is a practical limit in play for RCU but not Hazard Pointers, I don't think so though.
What is delayed is the actual destruction/deallocation of the RCU-managed objects. Nobody cares about that delay unless the objects control limited resources (in which case, RCU is likely a bad fit) or there are so many objects and/or they are so large that the delay in deallocating them could cause memory pressure of some type.
Not to put too fine a point on things, but rust (and c++) very explicitly don’t guarantee this. They are both quite explicit about being allowed to leak memory and never free it (due to reference cycles), something a GC is not typically allowed to do. So yes it usually happens, it just is not guaranteed.
Implementing a garbage collector that is guaranteed to free memory when it's actually no longer needed is equivalent to solving the halting problem (via Rice's theorem), and so any garbage collection algorithm is going to have to leak some memory, it's simply unavoidable.
The "finalizer problem" in the Garbage Collected languages isn't about a Goat which never becomes unused, it's about a situation where the Goat is unused but the clean-up never happens.
If the finalizer was for an object that is just memory, it's not a problem. The GC has decided it doesn't need to recover the memory used by that object just yet. I've seen this with heaps defined to be large enough to handle a memory spike. When the spikes stop happening, a lot of stuff in the old gen pool just stays there. At one point I was looking at 50GB heap dumps looking for memory leaks.
It is a problem when the object has non memory resources like a socket descriptor. You can run out of descriptors. So not relying on GC and using try with resource is a good idea.