Related
Hi I had a general question regarding usage of volatile and memory barriers in C while making memory changes in shared memory being concurrently accessed by multiple threads without locks.
As I understand volatile and memory barriers serve the following general purposes
memory barriers
A) make sure all pending memory accesses (read/writes(depending on the barrier)) have been properly completed before the barrier and only then the memory accesses following the barrier are executed.
B) Make sure that the compiler does not reorder load/store instructions(depending on the barrier) across the barriers.
Basically the purpose of point A is to handle out of order execution and write buffer flush delay scenarios where the processor itself ends up reordering instructions generated by the compiler OR memory accesses made by the said instructions.
The purpose of the point B is that when C code is translated to machine code the compiler does not itself move those accesses in assembly around.
Now for volatile
volatile is basically meant in a loose way so that so that the compiler does not perform its optimisations while optimising code written with volatile variables. The following broad purposes are served
A) memory accesses are not cached in cpu registers when translating C code to machine level code and every time a read in code happens it’s converted into a load instruction to be done through the memory in assembly.
B) relative order of memory accesses in assembly with other volatile variables are kept in the same order when the compiler transforms C code to machine code while the memory accesses in assembly with non volatile variables can be interleaved.
I have the following questions
is my understanding correct and complete ? Like are there cases I am missing or something I am saying incorrect.
so then whenever we are writing code making memory changes in shared memory being concurrently accessed by multiple threads we need to make sure we have barriers so that behaviour corresponding to point 1.A and 1.B doesn’t happen.
The behaviour corresponding to 2.B will be handled by 1.B and for 2.A we need to cast our pointer to a volatile pointer for the access.
Basically I am trying to understand should we always be casting the pointer to a volatile pointer and then making the memory access so that we are sure 2.A doesn’t happen or are there are cases where only using barriers suffice ?
is my understanding correct and complete ?
Yeah, it looks that way, except for not mentioning that C11 <stdatomic.h> made all this obsolete for almost all purposes.
There are more bad/weird things that can happen without volatile (or better, _Atomic) that you didn't list: the LWN article Who's afraid of a big bad optimizing compiler? goes into detail about things like inventing extra loads (and expecting them both to read the same value). It's aimed at Linux kernel code, where C11 _Atomic isn't how they do things.
Other than the Linux kernel, new code should pretty much always use <stdatomic.h> instead of rolling your own atomics with volatile and inline asm for RMWs and barriers. But that does continue to work because all real-world CPUs that we run threads across have coherent shared memory, so making a memory access happen in the asm is enough for inter-thread visibility, like memory_order_relaxed. See When to use volatile with multi threading? (basically never, except in the Linux kernel or maybe a handful of other codebases that already have good implementations of hand-rolled stuff).
In ISO C11, it's data-race undefined behaviour for two threads to do unsynchronized read+write on the same object, but mainstream compilers do define the behaviour, just compiling the way you'd expect so hardware guarantees or lack thereof come into play.
Other than that, yeah, looks accurate except for your final question 2: there are use-cases for memory_order_relaxed atomics, which is like volatile with no barriers, e.g. an exit_now flag.
or are there are cases where only using barriers suffice ?
No, unless you get lucky and the compiler happens to generate correct asm anyway.
Or unless other synchronization means this code only runs while no other threads are reading/writing the object. (C++20 has std::atomic_ref<T> to handle the case where some parts of the code need to do atomic accesses to data, but other parts of your program don't, and you want to let them auto-vectorize or whatever. C doesn't have any such thing yet, other than using plain variables with/without GNU C __atomic_load_n() and other builtins, which is how C++ headers implement std::atomic<T>, and which is the same underlying support that C11 _Atomic compiles to. Probably also the C11 functions like atomic_load_explicit defined in stdatomic.h, but unlike C++, _Atomic is a true keyword not defined in any header.)
https://www.gnu.org/software/libc/manual/html_node/Atomic-Types.html#Atomic-Types says - In practice, you can assume that int is atomic. You can also assume that pointer types are atomic; that is very convenient. Both of these assumptions are true on all of the machines that the GNU C Library supports and on all POSIX systems we know of.
My question is whether pointer assignment can be considered atomic on x86_64 architecture for a C program compiled with gcc m64 flag. OS is 64bit Linux and CPU is Intel(R) Xeon(R) CPU D-1548. One thread will be setting a pointer and another thread accessing the pointer. There is only one writer thread and one reader thread. Reader should either be getting the previous value of the pointer or the latest value and no garbage value in between.
If it is not considered atomic, please let me know how can I use the gcc atomic builtins or maybe memory barrier like __sync_synchronize to achieve the same without using locks. Interested only in C solution and not C++. Thanks!
Bear in mind that atomicity alone is not enough for communicating between threads. Nothing prevents the compiler and CPU from reordering previous/subsequent load and store instructions with that "atomic" store. In old days people used volatile to prevent that reordering but that was never intended for use with threads and doesn't provide means to specify less or more restrictive memory order (see "Relationship with volatile" in there).
You should use C11 atomics because they guarantee both atomicity and memory order.
For almost all architectures, pointer load and store are atomic. A once notable exception was 8086/80286 where pointers could be seg:offset; there was an l[des]s instruction which could make an atomic load; but no corresponding atomic store.
The integrity of the pointer is only a small concern; your bigger issue revolves around synchronization: the pointer was at value Y, you set it to X; how will you know when nobody is using the (old) Y value?
A somewhat related problem is that you may have stored things at X, which the other thread expects to find. Without synchronization, other might see the new pointer value, however what it points to might not be up to date yet.
A plain global char *ptr should not be considered atomic. It might work sometimes, especially with optimization disabled, but you can get the compiler to make safe and efficient optimized asm by using modern language features to tell it you want atomicity.
Use C11 stdatomic.h or GNU C __atomic builtins. And see Why is integer assignment on a naturally aligned variable atomic on x86? - yes the underlying asm operations are atomic "for free", but you need to control the compiler's code-gen to get sane behaviour for multithreading.
See also LWN: Who's afraid of a big bad optimizing compiler? - weird effects of using plain vars include several really bad well-known things, but also more obscure stuff like invented loads, reading a variable more than once if the compiler decides to optimize away a local tmp and load the shared var twice, instead of loading it into a register. Using asm("" ::: "memory") compiler barriers may not be sufficient to defeat that depending on where you put them.
So use proper atomic stores and loads that tell the compiler what you want: You should generally use atomic loads to read them, too.
#include <stdatomic.h> // C11 way
_Atomic char *c11_shared_var; // all access to this is atomic, functions needed only if you want weaker ordering
void foo(){
atomic_store_explicit(&c11_shared_var, newval, memory_order_relaxed);
}
char *plain_shared_var; // GNU C
// This is a plain C var. Only specific accesses to it are atomic; be careful!
void foo() {
__atomic_store_n(&plain_shared_var, newval, __ATOMIC_RELAXED);
}
Using __atomic_store_n on a plain var is the functionality that C++20 atomic_ref exposes. If multiple threads access a variable for the entire time that it needs to exist, you might as well just use C11 stdatomic because every access needs to be atomic (not optimized into a register or whatever). When you want to let the compiler load once and reuse that value, do char *tmp = c11_shared_var; (or atomic_load_explicit if you only want acquire instead of seq_cst; cheaper on a few non-x86 ISAs).
Besides lack of tearing (atomicity of asm load or store), the other key parts of _Atomic foo * are:
The compiler will assume that other threads may have changed memory contents (like volatile effectively implies), otherwise the assumption of no data-race UB will let the compiler hoist loads out of loops. Without this, dead-store elimination might only do one store at the end of a loop, not updating the value multiple times.
The read side of the problem is usually what bites people in practice, see Multithreading program stuck in optimized mode but runs normally in -O0 - e.g. while(!flag){} becomes if(!flag) infinite_loop; with optimization enabled.
Ordering wrt. other code. e.g. you can use memory_order_release to make sure that other threads that see the pointer update also see all changes to the pointed-to data. (On x86 that's as simple as compile-time ordering, no extra barriers needed for acquire/release, only for seq_cst. Avoid seq_cst if you can; mfence or locked operations are slow.)
Guarantee that the store will compile to a single asm instruction. You'd be depending on this. It does happen in practice with sane compilers, although it's conceivable that a compiler might decide to use rep movsb to copy a few contiguous pointers, and that some machine somewhere might have a microcoded implementation that does some stores narrower than 8 bytes.
(This failure mode is highly unlikely; the Linux kernel relies on volatile load/store compiling to a single instruction with GCC / clang for its hand-rolled intrinsics. But if you just used asm("" ::: "memory") to make sure a store happened on a non-volatile variable, there's a chance.)
Also, something like ptr++ will compile to an atomic RMW operation like lock add qword [mem], 4, rather than separate load and store like volatile would. (See Can num++ be atomic for 'int num'? for more about atomic RMWs). Avoid that if you don't need it, it's slower. e.g. atomic_store_explicit(&ptr, ptr + 1, mo_release); - seq_cst loads are cheap on x86-64 but seq_cst stores aren't.
Also note that memory barriers can't create atomicity (lack of tearing), they can only create ordering wrt other ops.
In practice x86-64 ABIs do have alignof(void*) = 8 so all pointer objects should be naturally aligned (except in a __attribute__((packed)) struct which violates the ABI, so you can use __atomic_store_n on them. It should compile to what you want (plain store, no overhead), and meet the asm requirements to be atomic.
See also When to use volatile with multi threading? - you can roll your own atomics with volatile and asm memory barriers, but don't. The Linux kernel does that, but it's a lot of effort for basically no gain, especially for a user-space program.
Side note: an often repeated misconception is that volatile or _Atomic are needed to avoid reading stale values from cache. This is not the case.
All machines that run C11 threads across multiple cores have coherent caches, not needing explicit flush instructions in the reader or writer. Just ordinary load or store instructions, like x86 mov. The key is to not let the compiler keep values of shared variable in CPU registers (which are thread-private). It normally can do this optimization because of the assumption of no data-race Undefined Behaviour. Registers are very much not the same thing as L1d CPU cache; managing what's in registers vs. memory is done by the compiler, while hardware keeps cache in sync. See When to use volatile with multi threading? for more details about why coherent caches is sufficient to make volatile work like memory_order_relaxed.
See Multithreading program stuck in optimized mode but runs normally in -O0 for an example.
"Atomic" is treated as this quantum state where something can be both atomic and not atomic at the same time because "it's possible" that "some machines" "somewhere" "might not" write "a certain value" atomically. Maybe.
That is not the case. Atomicity has a very specific meaning, and it solves a very specific problem: threads being pre-empted by the OS to schedule another thread in its place on that core. And you cannot stop a thread from executing mid-assembly instruction.
What that means is that any single assembly instruction is "atomic" by definition. And since you have registry moving instructions, any register-sized copy is atomic by definition. That means a 32-bit integer on a 32-bit CPU, and a 64-bit integer on a 64-bit CPU are all atomic -- and of course that includes pointers (ignore all the people who will tell you "some architectures" have pointers of "different size" than registers, that hasn't been the case since 386).
You should however be careful not to hit variable caching problems (ie one thread writing a pointer, and another trying to read it but getting an old value from the cache), use volatile as needed to prevent this.
If there are two threads accessing a global variable then many tutorials say make the variable volatile to prevent the compiler caching the variable in a register and it thus not getting updated correctly.
However two threads both accessing a shared variable is something which calls for protection via a mutex isn't it?
But in that case, between the thread locking and releasing the mutex the code is in a critical section where only that one thread can access the variable, in which case the variable doesn't need to be volatile?
So therefore what is the use/purpose of volatile in a multi-threaded program?
Short & quick answer: volatile is (nearly) useless for platform-agnostic, multithreaded application programming. It does not provide any synchronization, it does not create memory fences, nor does it ensure the order of execution of operations. It does not make operations atomic. It does not make your code magically thread safe. volatile may be the single-most misunderstood facility in all of C++. See this, this and this for more information about volatile
On the other hand, volatile does have some use that may not be so obvious. It can be used much in the same way one would use const to help the compiler show you where you might be making a mistake in accessing some shared resource in a non-protected way. This use is discussed by Alexandrescu in this article. However, this is basically using the C++ type system in a way that is often viewed as a contrivance and can evoke Undefined Behavior.
volatile was specifically intended to be used when interfacing with memory-mapped hardware, signal handlers, and the setjmp machine code instruction. This makes volatile directly applicable to systems-level programming rather than normal applications-level programming.
The 2003 C++ Standard does not say that volatile applies any kind of Acquire or Release semantics on variables. In fact, the Standard is completely silent on all matters of multithreading. However, specific platforms do apply Acquire and Release semantics on volatile variables.
[Update for C++11]
The C++11 Standard now does acknowledge multithreading directly in the memory model and the language, and it provides library facilities to deal with it in a platform-independent way. However the semantics of volatile still have not changed. volatile is still not a synchronization mechanism. Bjarne Stroustrup says as much in TCPPPL4E:
Do not use volatile except in low-level code that deals directly
with hardware.
Do not assume volatile has special meaning in the memory model. It
does not. It is not -- as in some later languages -- a
synchronization mechanism. To get synchronization, use atomic, a
mutex, or a condition_variable.
[/End update]
The above all applies to the C++ language itself, as defined by the 2003 Standard (and now the 2011 Standard). Some specific platforms however do add additional functionality or restrictions to what volatile does. For example, in MSVC 2010 (at least) Acquire and Release semantics do apply to certain operations on volatile variables. From the MSDN:
When optimizing, the compiler must maintain ordering among references
to volatile objects as well as references to other global objects. In
particular,
A write to a volatile object (volatile write) has Release semantics; a
reference to a global or static object that occurs before a write to a
volatile object in the instruction sequence will occur before that
volatile write in the compiled binary.
A read of a volatile object (volatile read) has Acquire semantics; a
reference to a global or static object that occurs after a read of
volatile memory in the instruction sequence will occur after that
volatile read in the compiled binary.
However, you might take note of the fact that if you follow the above link, there is some debate in the comments as to whether or not acquire/release semantics actually apply in this case.
In C++11, don't use volatile for threading, only for MMIO
But TL:DR, it does "work" sort of like atomic with mo_relaxed on hardware with coherent caches (i.e. everything); it is sufficient to stop compilers keeping vars in registers. atomic doesn't need memory barriers to create atomicity or inter-thread visibility, only to make the current thread wait before/after an operation to create ordering between this thread's accesses to different variables. mo_relaxed never needs any barriers, just load, store, or RMW.
For roll-your-own atomics with volatile (and inline-asm for barriers) in the bad old days before C++11 std::atomic, volatile was the only good way to get some things to work. But it depended on a lot of assumptions about how implementations worked and was never guaranteed by any standard.
For example the Linux kernel still uses its own hand-rolled atomics with volatile, but only supports a few specific C implementations (GNU C, clang, and maybe ICC). Partly that's because of GNU C extensions and inline asm syntax and semantics, but also because it depends on some assumptions about how compilers work.
It's almost always the wrong choice for new projects; you can use std::atomic (with std::memory_order_relaxed) to get a compiler to emit the same efficient machine code you could with volatile. std::atomic with mo_relaxed obsoletes volatile for threading purposes. (except maybe to work around missed-optimization bugs with atomic<double> on some compilers.)
The internal implementation of std::atomic on mainstream compilers (like gcc and clang) does not just use volatile internally; compilers directly expose atomic load, store and RMW builtin functions. (e.g. GNU C __atomic builtins which operate on "plain" objects.)
Volatile is usable in practice (but don't do it)
That said, volatile is usable in practice for things like an exit_now flag on all(?) existing C++ implementations on real CPUs, because of how CPUs work (coherent caches) and shared assumptions about how volatile should work. But not much else, and is not recommended. The purpose of this answer is to explain how existing CPUs and C++ implementations actually work. If you don't care about that, all you need to know is that std::atomic with mo_relaxed obsoletes volatile for threading.
(The ISO C++ standard is pretty vague on it, just saying that volatile accesses should be evaluated strictly according to the rules of the C++ abstract machine, not optimized away. Given that real implementations use the machine's memory address-space to model C++ address space, this means volatile reads and assignments have to compile to load/store instructions to access the object-representation in memory.)
As another answer points out, an exit_now flag is a simple case of inter-thread communication that doesn't need any synchronization: it's not publishing that array contents are ready or anything like that. Just a store that's noticed promptly by a not-optimized-away load in another thread.
// global
bool exit_now = false;
// in one thread
while (!exit_now) { do_stuff; }
// in another thread, or signal handler in this thread
exit_now = true;
Without volatile or atomic, the as-if rule and assumption of no data-race UB allows a compiler to optimize it into asm that only checks the flag once, before entering (or not) an infinite loop. This is exactly what happens in real life for real compilers. (And usually optimize away much of do_stuff because the loop never exits, so any later code that might have used the result is not reachable if we enter the loop).
// Optimizing compilers transform the loop into asm like this
if (!exit_now) { // check once before entering loop
while(1) do_stuff; // infinite loop
}
Multithreading program stuck in optimized mode but runs normally in -O0 is an example (with description of GCC's asm output) of how exactly this happens with GCC on x86-64. Also MCU programming - C++ O2 optimization breaks while loop on electronics.SE shows another example.
We normally want aggressive optimizations that CSE and hoist loads out of loops, including for global variables.
Before C++11, volatile bool exit_now was one way to make this work as intended (on normal C++ implementations). But in C++11, data-race UB still applies to volatile so it's not actually guaranteed by the ISO standard to work everywhere, even assuming HW coherent caches.
Note that for wider types, volatile gives no guarantee of lack of tearing. I ignored that distinction here for bool because it's a non-issue on normal implementations. But that's also part of why volatile is still subject to data-race UB instead of being equivalent to relaxed atomic.
Note that "as intended" doesn't mean the thread doing exit_now waits for the other thread to actually exit. Or even that it waits for the volatile exit_now=true store to even be globally visible before continuing to later operations in this thread. (atomic<bool> with the default mo_seq_cst would make it wait before any later seq_cst loads at least. On many ISAs you'd just get a full barrier after the store).
C++11 provides a non-UB way that compiles the same
A "keep running" or "exit now" flag should use std::atomic<bool> flag with mo_relaxed
Using
flag.store(true, std::memory_order_relaxed)
while( !flag.load(std::memory_order_relaxed) ) { ... }
will give you the exact same asm (with no expensive barrier instructions) that you'd get from volatile flag.
As well as no-tearing, atomic also gives you the ability to store in one thread and load in another without UB, so the compiler can't hoist the load out of a loop. (The assumption of no data-race UB is what allows the aggressive optimizations we want for non-atomic non-volatile objects.) This feature of atomic<T> is pretty much the same as what volatile does for pure loads and pure stores.
atomic<T> also make += and so on into atomic RMW operations (significantly more expensive than an atomic load into a temporary, operate, then a separate atomic store. If you don't want an atomic RMW, write your code with a local temporary).
With the default seq_cst ordering you'd get from while(!flag), it also adds ordering guarantees wrt. non-atomic accesses, and to other atomic accesses.
(In theory, the ISO C++ standard doesn't rule out compile-time optimization of atomics. But in practice compilers don't because there's no way to control when that wouldn't be ok. There are a few cases where even volatile atomic<T> might not be enough control over optimization of atomics if compilers did optimize, so for now compilers don't. See Why don't compilers merge redundant std::atomic writes? Note that wg21/p0062 recommends against using volatile atomic in current code to guard against optimization of atomics.)
volatile does actually work for this on real CPUs (but still don't use it)
even with weakly-ordered memory models (non-x86). But don't actually use it, use atomic<T> with mo_relaxed instead!! The point of this section is to address misconceptions about how real CPUs work, not to justify volatile. If you're writing lockless code, you probably care about performance. Understanding caches and the costs of inter-thread communication is usually important for good performance.
Real CPUs have coherent caches / shared memory: after a store from one core becomes globally visible, no other core can load a stale value. (See also Myths Programmers Believe about CPU Caches which talks some about Java volatiles, equivalent to C++ atomic<T> with seq_cst memory order.)
When I say load, I mean an asm instruction that accesses memory. That's what a volatile access ensures, and is not the same thing as lvalue-to-rvalue conversion of a non-atomic / non-volatile C++ variable. (e.g. local_tmp = flag or while(!flag)).
The only thing you need to defeat is compile-time optimizations that don't reload at all after the first check. Any load+check on each iteration is sufficient, without any ordering. Without synchronization between this thread and the main thread, it's not meaningful to talk about when exactly the store happened, or ordering of the load wrt. other operations in the loop. Only when it's visible to this thread is what matters. When you see the exit_now flag set, you exit. Inter-core latency on a typical x86 Xeon can be something like 40ns between separate physical cores.
In theory: C++ threads on hardware without coherent caches
I don't see any way this could be remotely efficient, with just pure ISO C++ without requiring the programmer to do explicit flushes in the source code.
In theory you could have a C++ implementation on a machine that wasn't like this, requiring compiler-generated explicit flushes to make things visible to other threads on other cores. (Or for reads to not use a maybe-stale copy). The C++ standard doesn't make this impossible, but C++'s memory model is designed around being efficient on coherent shared-memory machines. E.g. the C++ standard even talks about "read-read coherence", "write-read coherence", etc. One note in the standard even points the connection to hardware:
http://eel.is/c++draft/intro.races#19
[ Note: The four preceding coherence requirements effectively disallow compiler reordering of atomic operations to a single object, even if both operations are relaxed loads. This effectively makes the cache coherence guarantee provided by most hardware available to C++ atomic operations. — end note ]
There's no mechanism for a release store to only flush itself and a few select address-ranges: it would have to sync everything because it wouldn't know what other threads might want to read if their acquire-load saw this release-store (forming a release-sequence that establishes a happens-before relationship across threads, guaranteeing that earlier non-atomic operations done by the writing thread are now safe to read. Unless it did further writes to them after the release store...) Or compilers would have to be really smart to prove that only a few cache lines needed flushing.
Related: my answer on Is mov + mfence safe on NUMA? goes into detail about the non-existence of x86 systems without coherent shared memory. Also related: Loads and stores reordering on ARM for more about loads/stores to the same location.
There are I think clusters with non-coherent shared memory, but they're not single-system-image machines. Each coherency domain runs a separate kernel, so you can't run threads of a single C++ program across it. Instead you run separate instances of the program (each with their own address space: pointers in one instance aren't valid in the other).
To get them to communicate with each other via explicit flushes, you'd typically use MPI or other message-passing API to make the program specify which address ranges need flushing.
Real hardware doesn't run std::thread across cache coherency boundaries:
Some asymmetric ARM chips exist, with shared physical address space but not inner-shareable cache domains. So not coherent. (e.g. comment thread an A8 core and an Cortex-M3 like TI Sitara AM335x).
But different kernels would run on those cores, not a single system image that could run threads across both cores. I'm not aware of any C++ implementations that run std::thread threads across CPU cores without coherent caches.
For ARM specifically, GCC and clang generate code assuming all threads run in the same inner-shareable domain. In fact, the ARMv7 ISA manual says
This architecture (ARMv7) is written with an expectation that all processors using the same operating system or hypervisor are in the same Inner Shareable shareability domain
So non-coherent shared memory between separate domains is only a thing for explicit system-specific use of shared memory regions for communication between different processes under different kernels.
See also this CoreCLR discussion about code-gen using dmb ish (Inner Shareable barrier) vs. dmb sy (System) memory barriers in that compiler.
I make the assertion that no C++ implementation for other any other ISA runs std::thread across cores with non-coherent caches. I don't have proof that no such implementation exists, but it seems highly unlikely. Unless you're targeting a specific exotic piece of HW that works that way, your thinking about performance should assume MESI-like cache coherency between all threads. (Preferably use atomic<T> in ways that guarantees correctness, though!)
Coherent caches makes it simple
But on a multi-core system with coherent caches, implementing a release-store just means ordering commit into cache for this thread's stores, not doing any explicit flushing. (https://preshing.com/20120913/acquire-and-release-semantics/ and https://preshing.com/20120710/memory-barriers-are-like-source-control-operations/). (And an acquire-load means ordering access to cache in the other core).
A memory barrier instruction just blocks the current thread's loads and/or stores until the store buffer drains; that always happens as fast as possible on its own. (Or for LoadLoad / LoadStore barriers, block until previous loads have completed.) (Does a memory barrier ensure that the cache coherence has been completed? addresses this misconception). So if you don't need ordering, just prompt visibility in other threads, mo_relaxed is fine. (And so is volatile, but don't do that.)
See also C/C++11 mappings to processors
Fun fact: on x86, every asm store is a release-store because the x86 memory model is basically seq-cst plus a store buffer (with store forwarding).
Semi-related re: store buffer, global visibility, and coherency: C++11 guarantees very little. Most real ISAs (except PowerPC) do guarantee that all threads can agree on the order of a appearance of two stores by two other threads. (In formal computer-architecture memory model terminology, they're "multi-copy atomic").
Will two atomic writes to different locations in different threads always be seen in the same order by other threads?
Concurrent stores seen in a consistent order
Another misconception is that memory fence asm instructions are needed to flush the store buffer for other cores to see our stores at all. Actually the store buffer is always trying to drain itself (commit to L1d cache) as fast as possible, otherwise it would fill up and stall execution. What a full barrier / fence does is stall the current thread until the store buffer is drained, so our later loads appear in the global order after our earlier stores.
Are loads and stores the only instructions that gets reordered?
x86 mfence and C++ memory barrier
Globally Invisible load instructions
(x86's strongly ordered asm memory model means that volatile on x86 may end up giving you closer to mo_acq_rel, except that compile-time reordering with non-atomic variables can still happen. But most non-x86 have weakly-ordered memory models so volatile and relaxed are about as weak as mo_relaxed allows.)
(Editor's note: in C++11 volatile is not the right tool for this job and still has data-race UB. Use std::atomic<bool> with std::memory_order_relaxed loads/stores to do this without UB. On real implementations it will compile to the same asm as volatile. I added an answer with more detail, and also addressing the misconceptions in comments that weakly-ordered memory might be a problem for this use-case: all real-world CPUs have coherent shared memory so volatile will work for this on real C++ implementations. But still don't do it.
Some discussion in comments seems to be talking about other use-cases where you would need something stronger than relaxed atomics. This answer already points out that volatile gives you no ordering.)
Volatile is occasionally useful for the following reason: this code:
/* global */ bool flag = false;
while (!flag) {}
is optimized by gcc to:
if (!flag) { while (true) {} }
Which is obviously incorrect if the flag is written to by the other thread. Note that without this optimization the synchronization mechanism probably works (depending on the other code some memory barriers may be needed) - there is no need for a mutex in 1 producer - 1 consumer scenario.
Otherwise the volatile keyword is too weird to be useable - it does not provide any memory ordering guarantees wrt both volatile and non-volatile accesses and does not provide any atomic operations - i.e. you get no help from the compiler with volatile keyword except disabled register caching.
You need volatile and possibly locking.
volatile tells the optimiser that the value can change asynchronously, thus
volatile bool flag = false;
while (!flag) {
/*do something*/
}
will read flag every time around the loop.
If you turn optimisation off or make every variable volatile a program will behave the same but slower. volatile just means 'I know you may have just read it and know what it says, but if I say read it then read it.
Locking is a part of the program. So ,by the way, if you are implementing semaphores then among other things they must be volatile. (Don't try it, it is hard, will probably need a little assembler or the new atomic stuff, and it has already been done.)
#include <iostream>
#include <thread>
#include <unistd.h>
using namespace std;
bool checkValue = false;
int main()
{
std::thread writer([&](){
sleep(2);
checkValue = true;
std::cout << "Value of checkValue set to " << checkValue << std::endl;
});
std::thread reader([&](){
while(!checkValue);
});
writer.join();
reader.join();
}
Once an interviewer who also believed that volatile is useless argued with me that Optimisation wouldn't cause any issues and was referring to different cores having separate cache lines and all that (didn't really understand what he was exactly referring to). But this piece of code when compiled with -O3 on g++ (g++ -O3 thread.cpp -lpthread), it shows undefined behaviour. Basically if the value gets set before the while check it works fine and if not it goes into a loop without bothering to fetch the value (which was actually changed by the other thread). Basically i believe the value of checkValue only gets fetched once into the register and never gets checked again under the highest level of optimisation. If its set to true before the fetch, it works fine and if not it goes into a loop. Please correct me if am wrong.
As I know that in C we could use the keyword "register" to suggest to the compiler that the variable should be stored in the CPU register. Isn't it true that all variables that involved in CPU instructions will be eventually stored in CPU registers for execution?
The register keyword is a way of telling the compiler that the variable is heavily used. It's true that values must usually be loaded temporarily into registers to perform calculations on them. The name comes from the idea that a compiler might keep the variable in a register for the entire duration that it is in scope, rather than only temporarily when it is being used in a calculation.
The keyword is obsolete for the purpose of optimisation, since modern compilers can determine when a variable is heavily used (and when it does not have its address taken) without help from the programmer.
You should not use that register keyword. It is an antique relic, maintained for backward compatibility. Most compilers will ignore it (by default).
There could be exceptions but they are very rare, consult your compiler manual.
Isn't it true that all variables that involved in CPU instructions will be eventually stored in CPU registers for execution?
Yes, that is true. But CPU registers are limited so variables are usually LOAD/STOREd from 'normal' memory and live in a register only briefly. The register keyword is (was) a way of indicating high priority variables that should occupy a register longer. Like the i in for(i = 0; ...).
In old times, compilers weren't that smart as they are today. It was a hint from the programmer to the compiler that this variable should be stored in a register to allow fast access/modification. Today, almost any decent compiler implements clever registers allocation algorithms that beats the mind of humans.
Most variables will be loaded into registers for a short while...as long as necessary to do what needs to be done with them. The register keyword hints that they should be kept there.
Compilers' optimization has gotten so much better, though, that the register keyword isn't very helpful. In fact, if your compiler respects it at all (and many don't), it could even mess you up (by tying the compiler's hands, making certain optimizations impossible). So it's a pretty bad idea these days.
In the C programming language and Pthreads as the threading library; do variables/structures that are shared between threads need to be declared as volatile? Assuming that they might be protected by a lock or not (barriers perhaps).
Does the pthread POSIX standard have any say about this, is this compiler-dependent or neither?
Edit to add: Thanks for the great answers. But what if you're not using locks; what if you're using barriers for example? Or code that uses primitives such as compare-and-swap to directly and atomically modify a shared variable...
As long as you are using locks to control access to the variable, you do not need volatile on it. In fact, if you're putting volatile on any variable you're probably already wrong.
https://software.intel.com/en-us/blogs/2007/11/30/volatile-almost-useless-for-multi-threaded-programming/
The answer is absolutely, unequivocally, NO. You do not need to use 'volatile' in addition to proper synchronization primitives. Everything that needs to be done are done by these primitives.
The use of 'volatile' is neither necessary nor sufficient. It's not necessary because the proper synchronization primitives are sufficient. It's not sufficient because it only disables some optimizations, not all of the ones that might bite you. For example, it does not guarantee either atomicity or visibility on another CPU.
But unless you use volatile, the compiler is free to cache the shared data in a register for any length of time... if you want your data to be written to be predictably written to actual memory and not just cached in a register by the compiler at its discretion, you will need to mark it as volatile. Alternatively, if you only access the shared data after you have left a function modifying it, you might be fine. But I would suggest not relying on blind luck to make sure that values are written back from registers to memory.
Right, but even if you do use volatile, the CPU is free to cache the shared data in a write posting buffer for any length of time. The set of optimizations that can bite you is not precisely the same as the set of optimizations that 'volatile' disables. So if you use 'volatile', you are relying on blind luck.
On the other hand, if you use sychronization primitives with defined multi-threaded semantics, you are guaranteed that things will work. As a plus, you don't take the huge performance hit of 'volatile'. So why not do things that way?
I think one very important property of volatile is that it makes the variable be written to memory when modified, and reread from memory each time it accessed. The other answers here mix volatile and synchronization, and it is clear from some other answers than this that volatile is NOT a sync primitive (credit where credit is due).
But unless you use volatile, the compiler is free to cache the shared data in a register for any length of time... if you want your data to be written to be predictably written to actual memory and not just cached in a register by the compiler at its discretion, you will need to mark it as volatile. Alternatively, if you only access the shared data after you have left a function modifying it, you might be fine. But I would suggest not relying on blind luck to make sure that values are written back from registers to memory.
Especially on register-rich machines (i.e., not x86), variables can live for quite long periods in registers, and a good compiler can cache even parts of structures or entire structures in registers. So you should use volatile, but for performance, also copy values to local variables for computation and then do an explicit write-back. Essentially, using volatile efficiently means doing a bit of load-store thinking in your C code.
In any case, you positively have to use some kind of OS-level provided sync mechanism to create a correct program.
For an example of the weakness of volatile, see my Decker's algorithm example at http://jakob.engbloms.se/archives/65, which proves pretty well that volatile does not work to synchronize.
There is a widespread notion that the keyword volatile is good for multi-threaded programming.
Hans Boehm points out that there are only three portable uses for volatile:
volatile may be used to mark local variables in the same scope as a setjmp whose value should be preserved across a longjmp. It is unclear what fraction of such uses would be slowed down, since the atomicity and ordering constraints have no effect if there is no way to share the local variable in question. (It is even unclear what fraction of such uses would be slowed down by requiring all variables to be preserved across a longjmp, but that is a separate matter and is not considered here.)
volatile may be used when variables may be "externally modified", but the modification in fact is triggered synchronously by the thread itself, e.g. because the underlying memory is mapped at multiple locations.
A volatile sigatomic_t may be used to communicate with a signal handler in the same thread, in a restricted manner. One could consider weakening the requirements for the sigatomic_t case, but that seems rather counterintuitive.
If you are multi-threading for the sake of speed, slowing down code is definitely not what you want. For multi-threaded programming, there two key issues that volatile is often mistakenly thought to address:
atomicity
memory consistency, i.e. the order of a thread's operations as seen by another thread.
Let's deal with (1) first. Volatile does not guarantee atomic reads or writes. For example, a volatile read or write of a 129-bit structure is not going to be atomic on most modern hardware. A volatile read or write of a 32-bit int is atomic on most modern hardware, but volatile has nothing to do with it. It would likely be atomic without the volatile. The atomicity is at the whim of the compiler. There's nothing in the C or C++ standards that says it has to be atomic.
Now consider issue (2). Sometimes programmers think of volatile as turning off optimization of volatile accesses. That's largely true in practice. But that's only the volatile accesses, not the non-volatile ones. Consider this fragment:
volatile int Ready;
int Message[100];
void foo( int i ) {
Message[i/10] = 42;
Ready = 1;
}
It's trying to do something very reasonable in multi-threaded programming: write a message and then send it to another thread. The other thread will wait until Ready becomes non-zero and then read Message. Try compiling this with "gcc -O2 -S" using gcc 4.0, or icc. Both will do the store to Ready first, so it can be overlapped with the computation of i/10. The reordering is not a compiler bug. It's an aggressive optimizer doing its job.
You might think the solution is to mark all your memory references volatile. That's just plain silly. As the earlier quotes say, it will just slow down your code. Worst yet, it might not fix the problem. Even if the compiler does not reorder the references, the hardware might. In this example, x86 hardware will not reorder it. Neither will an Itanium(TM) processor, because Itanium compilers insert memory fences for volatile stores. That's a clever Itanium extension. But chips like Power(TM) will reorder. What you really need for ordering are memory fences, also called memory barriers. A memory fence prevents reordering of memory operations across the fence, or in some cases, prevents reordering in one direction.Volatile has nothing to do with memory fences.
So what's the solution for multi-threaded programming? Use a library or language extension that implements the atomic and fence semantics. When used as intended, the operations in the library will insert the right fences. Some examples:
POSIX threads
Windows(TM) threads
OpenMP
TBB
Based on article by Arch Robison (Intel)
In my experience, no; you just have to properly mutex yourself when you write to those values, or structure your program such that the threads will stop before they need to access data that depends on another thread's actions. My project, x264, uses this method; threads share an enormous amount of data but the vast majority of it doesn't need mutexes because its either read-only or a thread will wait for the data to become available and finalized before it needs to access it.
Now, if you have many threads that are all heavily interleaved in their operations (they depend on each others' output on a very fine-grained level), this may be a lot harder--in fact, in such a case I'd consider revisiting the threading model to see if it can possibly be done more cleanly with more separation between threads.
NO.
Volatile is only required when reading a memory location that can change independently of the CPU read/write commands. In the situation of threading, the CPU is in full control of read/writes to memory for each thread, therefore the compiler can assume the memory is coherent and optimizes the CPU instructions to reduce unnecessary memory access.
The primary usage for volatile is for accessing memory-mapped I/O. In this case, the underlying device can change the value of a memory location independently from CPU. If you do not use volatile under this condition, the CPU may use a previously cached memory value, instead of reading the newly updated value.
POSIX 7 guarantees that functions such as pthread_lock also synchronize memory
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_11 "4.12 Memory Synchronization" says:
The following functions synchronize memory with respect to other threads:
pthread_barrier_wait()
pthread_cond_broadcast()
pthread_cond_signal()
pthread_cond_timedwait()
pthread_cond_wait()
pthread_create()
pthread_join()
pthread_mutex_lock()
pthread_mutex_timedlock()
pthread_mutex_trylock()
pthread_mutex_unlock()
pthread_spin_lock()
pthread_spin_trylock()
pthread_spin_unlock()
pthread_rwlock_rdlock()
pthread_rwlock_timedrdlock()
pthread_rwlock_timedwrlock()
pthread_rwlock_tryrdlock()
pthread_rwlock_trywrlock()
pthread_rwlock_unlock()
pthread_rwlock_wrlock()
sem_post()
sem_timedwait()
sem_trywait()
sem_wait()
semctl()
semop()
wait()
waitpid()
Therefore if your variable is guarded between pthread_mutex_lock and pthread_mutex_unlock then it does not need further synchronization as you might attempt to provide with volatile.
Related questions:
Does guarding a variable with a pthread mutex guarantee it's also not cached?
Does pthread_mutex_lock contains memory fence instruction?
Volatile would only be useful if you need absolutely no delay between when one thread writes something and another thread reads it. Without some sort of lock, though, you have no idea of when the other thread wrote the data, only that it's the most recent possible value.
For simple values (int and float in their various sizes) a mutex might be overkill if you don't need an explicit synch point. If you don't use a mutex or lock of some sort, you should declare the variable volatile. If you use a mutex you're all set.
For complicated types, you must use a mutex. Operations on them are non-atomic, so you could read a half-changed version without a mutex.
Volatile means that we have to go to memory to get or set this value. If you don't set volatile, the compiled code might store the data in a register for a long time.
What this means is that you should mark variables that you share between threads as volatile so that you don't have situations where one thread starts modifying the value but doesn't write its result before a second thread comes along and tries to read the value.
Volatile is a compiler hint that disables certain optimizations. The output assembly of the compiler might have been safe without it but you should always use it for shared values.
This is especially important if you are NOT using the expensive thread sync objects provided by your system - you might for example have a data structure where you can keep it valid with a series of atomic changes. Many stacks that do not allocate memory are examples of such data structures, because you can add a value to the stack then move the end pointer or remove a value from the stack after moving the end pointer. When implementing such a structure, volatile becomes crucial to ensure that your atomic instructions are actually atomic.
The underlying reason is that the C language semantic is based upon a single-threaded abstract machine. And the compiler is within its own right to transform the program as long as the program's 'observable behaviors' on the abstract machine stay unchanged. It can merge adjacent or overlapping memory accesses, redo a memory access multiple times (upon register spilling for example), or simply discard a memory access, if it thinks the program's behaviors, when executed in a single thread, doesn't change. Therefore as you may suspect, the behaviors do change if the program is actually supposed to be executing in a multi-threaded way.
As Paul Mckenney pointed out in a famous Linux kernel document:
It _must_not_ be assumed that the compiler will do what you want
with memory references that are not protected by READ_ONCE() and
WRITE_ONCE(). Without them, the compiler is within its rights to
do all sorts of "creative" transformations, which are covered in
the COMPILER BARRIER section.
READ_ONCE() and WRITE_ONCE() are defined as volatile casts on referenced variables. Thus:
int y;
int x = READ_ONCE(y);
is equivalent to:
int y;
int x = *(volatile int *)&y;
So, unless you make a 'volatile' access, you are not assured that the access happens exactly once, no matter what synchronization mechanism you are using. Calling an external function (pthread_mutex_lock for example) may force the compiler do memory accesses to global variables. But this happens only when the compiler fails to figure out whether the external function changes these global variables or not. Modern compilers employing sophisticated inter-procedure analysis and link-time optimization make this trick simply useless.
In summary, you should mark variables shared by multiple threads volatile or access them using volatile casts.
As Paul McKenney has also pointed out:
I have seen the glint in their eyes when they discuss optimization techniques that you would not want your children to know about!
But see what happens to C11/C++11.
Some people obviously are assuming that the compiler treats the synchronization calls as memory barriers. "Casey" is assuming there is exactly one CPU.
If the sync primitives are external functions and the symbols in question are visible outside the compilation unit (global names, exported pointer, exported function that may modify them) then the compiler will treat them -- or any other external function call -- as a memory fence with respect to all externally visible objects.
Otherwise, you are on your own. And volatile may be the best tool available for making the compiler produce correct, fast code. It generally won't be portable though, when you need volatile and what it actually does for you depends a lot on the system and compiler.
No.
First, volatile is not necessary. There are numerous other operations that provide guaranteed multithreaded semantics that don't use volatile. These include atomic operations, mutexes, and so on.
Second, volatile is not sufficient. The C standard does not provide any guarantees about multithreaded behavior for variables declared volatile.
So being neither necessary nor sufficient, there's not much point in using it.
One exception would be particular platforms (such as Visual Studio) where it does have documented multithreaded semantics.
Variables that are shared among threads should be declared 'volatile'. This tells the
compiler that when one thread writes to such variables, the write should be to memory
(as opposed to a register).