P2300R9
std::execution

Published Proposal,

Authors:
Source:
GitHub
Issue Tracking:
GitHub
Project:
ISO/IEC 14882 Programming Languages — C++, ISO/IEC JTC1/SC22/WG21
Audience:
SG1, LEWG

1. Introduction

This paper proposes a self-contained design for a Standard C++ framework for managing asynchronous execution on generic execution resources. It is based on the ideas in A Unified Executors Proposal for C++ and its companion papers.

1.1. Motivation

Today, C++ software is increasingly asynchronous and parallel, a trend that is likely to only continue going forward. Asynchrony and parallelism appears everywhere, from processor hardware interfaces, to networking, to file I/O, to GUIs, to accelerators. Every C++ domain and every platform needs to deal with asynchrony and parallelism, from scientific computing to video games to financial services, from the smallest mobile devices to your laptop to GPUs in the world’s fastest supercomputer.

While the C++ Standard Library has a rich set of concurrency primitives (std::atomic, std::mutex, std::counting_semaphore, etc) and lower level building blocks (std::thread, etc), we lack a Standard vocabulary and framework for asynchrony and parallelism that C++ programmers desperately need. std::async/std::future/std::promise, C++11’s intended exposure for asynchrony, is inefficient, hard to use correctly, and severely lacking in genericity, making it unusable in many contexts. We introduced parallel algorithms to the C++ Standard Library in C++17, and while they are an excellent start, they are all inherently synchronous and not composable.

This paper proposes a Standard C++ model for asynchrony based around three key abstractions: schedulers, senders, and receivers, and a set of customizable asynchronous algorithms.

1.2. Priorities

1.3. Examples: End User

In this section we demonstrate the end-user experience of asynchronous programming directly with the sender algorithms presented in this paper. See § 4.19 User-facing sender factories, § 4.20 User-facing sender adaptors, and § 4.21 User-facing sender consumers for short explanations of the algorithms used in these code examples.

1.3.1. Hello world

using namespace std::execution;

scheduler auto sch = thread_pool.scheduler();                                 // 1

sender auto begin = schedule(sch);                                            // 2
sender auto hi = then(begin, []{                                              // 3
    std::cout << "Hello world! Have an int.";                                 // 3
    return 13;                                                                // 3
});                                                                           // 3
sender auto add_42 = then(hi, [](int arg) { return arg + 42; });              // 4

auto [i] = this_thread::sync_wait(add_42).value();                            // 5

This example demonstrates the basics of schedulers, senders, and receivers:

  1. First we need to get a scheduler from somewhere, such as a thread pool. A scheduler is a lightweight handle to an execution resource.

  2. To start a chain of work on a scheduler, we call § 4.19.1 execution::schedule, which returns a sender that completes on the scheduler. A sender describes asynchronous work and sends a signal (value, error, or stopped) to some recipient(s) when that work completes.

  3. We use sender algorithms to produce senders and compose asynchronous work. § 4.20.2 execution::then is a sender adaptor that takes an input sender and a std::invocable, and calls the std::invocable on the signal sent by the input sender. The sender returned by then sends the result of that invocation. In this case, the input sender came from schedule, so its void, meaning it won’t send us a value, so our std::invocable takes no parameters. But we return an int, which will be sent to the next recipient.

  4. Now, we add another operation to the chain, again using § 4.20.2 execution::then. This time, we get sent a value - the int from the previous step. We add 42 to it, and then return the result.

  5. Finally, we’re ready to submit the entire asynchronous pipeline and wait for its completion. Everything up until this point has been completely asynchronous; the work may not have even started yet. To ensure the work has started and then block pending its completion, we use § 4.21.2 this_thread::sync_wait, which will either return a std::optional<std::tuple<...>> with the value sent by the last sender, or an empty std::optional if the last sender sent a stopped signal, or it throws an exception if the last sender sent an error.

1.3.2. Asynchronous inclusive scan

using namespace std::execution;

sender auto async_inclusive_scan(scheduler auto sch,                          // 2
                                 std::span<const double> input,               // 1
                                 std::span<double> output,                    // 1
                                 double init,                                 // 1
                                 std::size_t tile_count)                      // 3
{
  std::size_t const tile_size = (input.size() + tile_count - 1) / tile_count;

  std::vector<double> partials(tile_count + 1);                               // 4
  partials[0] = init;                                                         // 4

  return just(std::move(partials))                                            // 5
       | transfer(sch)
       | bulk(tile_count,                                                     // 6
           [ = ](std::size_t i, std::vector<double>& partials) {              // 7
             auto start = i * tile_size;                                      // 8
             auto end   = std::min(input.size(), (i + 1) * tile_size);        // 8
             partials[i + 1] = *--std::inclusive_scan(begin(input) + start,   // 9
                                                      begin(input) + end,     // 9
                                                      begin(output) + start); // 9
           })                                                                 // 10
       | then(                                                                // 11
           [](std::vector<double>&& partials) {
             std::inclusive_scan(begin(partials), end(partials),              // 12
                                 begin(partials));                            // 12
             return std::move(partials);                                      // 13
           })
       | bulk(tile_count,                                                     // 14
           [ = ](std::size_t i, std::vector<double>& partials) {              // 14
             auto start = i * tile_size;                                      // 14
             auto end   = std::min(input.size(), (i + 1) * tile_size);        // 14
             std::for_each(begin(output) + start, begin(output) + end,        // 14
               [&] (double& e) { e = partials[i] + e; }                       // 14
             );
           })
       | then(                                                                // 15
           [ = ](std::vector<double>&& partials) {                            // 15
             return output;                                                   // 15
           });                                                                // 15
}

This example builds an asynchronous computation of an inclusive scan:

  1. It scans a sequence of doubles (represented as the std::span<const double> input) and stores the result in another sequence of doubles (represented as std::span<double> output).

  2. It takes a scheduler, which specifies what execution resource the scan should be launched on.

  3. It also takes a tile_count parameter that controls the number of execution agents that will be spawned.

  4. First we need to allocate temporary storage needed for the algorithm, which we’ll do with a std::vector, partials. We need one double of temporary storage for each execution agent we create.

  5. Next we’ll create our initial sender with § 4.19.2 execution::just and § 4.20.1 execution::transfer. These senders will send the temporary storage, which we’ve moved into the sender. The sender has a completion scheduler of sch, which means the next item in the chain will use sch.

  6. Senders and sender adaptors support composition via operator|, similar to C++ ranges. We’ll use operator| to attach the next piece of work, which will spawn tile_count execution agents using § 4.20.9 execution::bulk (see § 4.12 Most sender adaptors are pipeable for details).

  7. Each agent will call a std::invocable, passing it two arguments. The first is the agent’s index (i) in the § 4.20.9 execution::bulk operation, in this case a unique integer in [0, tile_count). The second argument is what the input sender sent - the temporary storage.

  8. We start by computing the start and end of the range of input and output elements that this agent is responsible for, based on our agent index.

  9. Then we do a sequential std::inclusive_scan over our elements. We store the scan result for our last element, which is the sum of all of our elements, in our temporary storage partials.

  10. After all computation in that initial § 4.20.9 execution::bulk pass has completed, every one of the spawned execution agents will have written the sum of its elements into its slot in partials.

  11. Now we need to scan all of the values in partials. We’ll do that with a single execution agent which will execute after the § 4.20.9 execution::bulk completes. We create that execution agent with § 4.20.2 execution::then.

  12. § 4.20.2 execution::then takes an input sender and an std::invocable and calls the std::invocable with the value sent by the input sender. Inside our std::invocable, we call std::inclusive_scan on partials, which the input senders will send to us.

  13. Then we return partials, which the next phase will need.

  14. Finally we do another § 4.20.9 execution::bulk of the same shape as before. In this § 4.20.9 execution::bulk, we will use the scanned values in partials to integrate the sums from other tiles into our elements, completing the inclusive scan.

  15. async_inclusive_scan returns a sender that sends the output std::span<double>. A consumer of the algorithm can chain additional work that uses the scan result. At the point at which async_inclusive_scan returns, the computation may not have completed. In fact, it may not have even started.

1.3.3. Asynchronous dynamically-sized read

using namespace std::execution;

sender_of<std::size_t> auto async_read(                                       // 1
    sender_of<std::span<std::byte>> auto buffer,                              // 1
    auto handle);                                                             // 1

struct dynamic_buffer {                                                       // 3
  std::unique_ptr<std::byte[]> data;                                          // 3
  std::size_t size;                                                           // 3
};                                                                            // 3

sender_of<dynamic_buffer> auto async_read_array(auto handle) {                // 2
  return just(dynamic_buffer{})                                               // 4
       | let_value([handle] (dynamic_buffer& buf) {                           // 5
           return just(std::as_writeable_bytes(std::span(&buf.size, 1))       // 6
                | async_read(handle)                                          // 7
                | then(                                                       // 8
                    [&buf] (std::size_t bytes_read) {                         // 9
                      assert(bytes_read == sizeof(buf.size));                 // 10
                      buf.data = std::make_unique<std::byte[]>(buf.size);     // 11
                      return std::span(buf.data.get(), buf.size);             // 12
                    })
                | async_read(handle)                                          // 13
                | then(
                    [&buf] (std::size_t bytes_read) {
                      assert(bytes_read == buf.size);                         // 14
                      return std::move(buf);                                  // 15
                    });
       });
}

This example demonstrates a common asynchronous I/O pattern - reading a payload of a dynamic size by first reading the size, then reading the number of bytes specified by the size:

  1. async_read is a pipeable sender adaptor. It’s a customization point object, but this is what it’s call signature looks like. It takes a sender parameter which must send an input buffer in the form of a std::span<std::byte>, and a handle to an I/O context. It will asynchronously read into the input buffer, up to the size of the std::span. It returns a sender which will send the number of bytes read once the read completes.

  2. async_read_array takes an I/O handle and reads a size from it, and then a buffer of that many bytes. It returns a sender that sends a dynamic_buffer object that owns the data that was sent.

  3. dynamic_buffer is an aggregate struct that contains a std::unique_ptr<std::byte[]> and a size.

  4. The first thing we do inside of async_read_array is create a sender that will send a new, empty dynamic_array object using § 4.19.2 execution::just. We can attach more work to the pipeline using operator| composition (see § 4.12 Most sender adaptors are pipeable for details).

  5. We need the lifetime of this dynamic_array object to last for the entire pipeline. So, we use let_value, which takes an input sender and a std::invocable that must return a sender itself (see § 4.20.4 execution::let_* for details). let_value sends the value from the input sender to the std::invocable. Critically, the lifetime of the sent object will last until the sender returned by the std::invocable completes.

  6. Inside of the let_value std::invocable, we have the rest of our logic. First, we want to initiate an async_read of the buffer size. To do that, we need to send a std::span pointing to buf.size. We can do that with § 4.19.2 execution::just.

  7. We chain the async_read onto the § 4.19.2 execution::just sender with operator|.

  8. Next, we pipe a std::invocable that will be invoked after the async_read completes using § 4.20.2 execution::then.

  9. That std::invocable gets sent the number of bytes read.

  10. We need to check that the number of bytes read is what we expected.

  11. Now that we have read the size of the data, we can allocate storage for it.

  12. We return a std::span<std::byte> to the storage for the data from the std::invocable. This will be sent to the next recipient in the pipeline.

  13. And that recipient will be another async_read, which will read the data.

  14. Once the data has been read, in another § 4.20.2 execution::then, we confirm that we read the right number of bytes.

  15. Finally, we move out of and return our dynamic_buffer object. It will get sent by the sender returned by async_read_array. We can attach more things to that sender to use the data in the buffer.

1.4. Asynchronous Windows socket recv

To get a better feel for how this interface might be used by low-level operations see this example implementation of a cancellable async_recv() operation for a Windows Socket.

struct operation_base : WSAOVERALAPPED {
    using completion_fn = void(operation_base* op, DWORD bytesTransferred, int errorCode) noexcept;

    // Assume IOCP event loop will call this when this OVERLAPPED structure is dequeued.
    completion_fn* completed;
};

template<class Receiver>
struct recv_op : operation_base {
    using operation_state_concept = std::execution::operation_state_t;

    recv_op(SOCKET s, void* data, size_t len, Receiver r)
    : receiver(std::move(r))
    , sock(s) {
        this->Internal = 0;
        this->InternalHigh = 0;
        this->Offset = 0;
        this->OffsetHigh = 0;
        this->hEvent = NULL;
        this->completed = &recv_op::on_complete;
        buffer.len = len;
        buffer.buf = static_cast<CHAR*>(data);
    }

    void start() & noexcept {
        // Avoid even calling WSARecv() if operation already cancelled
        auto st = std::execution::get_stop_token(
          std::execution::get_env(receiver));
        if (st.stop_requested()) {
            std::execution::set_stopped(std::move(receiver));
            return;
        }

        // Store and cache result here in case it changes during execution
        const bool stopPossible = st.stop_possible();
        if (!stopPossible) {
            ready.store(true, std::memory_order_relaxed);
        }

        // Launch the operation
        DWORD bytesTransferred = 0;
        DWORD flags = 0;
        int result = WSARecv(sock, &buffer, 1, &bytesTransferred, &flags,
                             static_cast<WSAOVERLAPPED*>(this), NULL);
        if (result == SOCKET_ERROR) {
            int errorCode = WSAGetLastError();
            if (errorCode != WSA_IO_PENDING) {
                if (errorCode == WSA_OPERATION_ABORTED) {
                    std::execution::set_stopped(std::move(receiver));
                } else {
                    std::execution::set_error(std::move(receiver),
                                              std::error_code(errorCode, std::system_category()));
                }
                return;
            }
        } else {
            // Completed synchronously (assuming FILE_SKIP_COMPLETION_PORT_ON_SUCCESS has been set)
            execution::set_value(std::move(receiver), bytesTransferred);
            return;
        }

        // If we get here then operation has launched successfully and will complete asynchronously.
        // May be completing concurrently on another thread already.
        if (stopPossible) {
            // Register the stop callback
            stopCallback.emplace(std::move(st), cancel_cb{*this});

            // Mark as 'completed'
            if (ready.load(std::memory_order_acquire) ||
                ready.exchange(true, std::memory_order_acq_rel)) {
                // Already completed on another thread
                stopCallback.reset();

                BOOL ok = WSAGetOverlappedResult(sock, (WSAOVERLAPPED*)this, &bytesTransferred, FALSE, &flags);
                if (ok) {
                    std::execution::set_value(std::move(receiver), bytesTransferred);
                } else {
                    int errorCode = WSAGetLastError();
                    std::execution::set_error(std::move(receiver),
                                              std::error_code(errorCode, std::system_category()));
                }
            }
        }
    }

    struct cancel_cb {
        recv_op& op;

        void operator()() noexcept {
            CancelIoEx((HANDLE)op.sock, (OVERLAPPED*)(WSAOVERLAPPED*)&op);
        }
    };

    static void on_complete(operation_base* op, DWORD bytesTransferred, int errorCode) noexcept {
        recv_op& self = *static_cast<recv_op*>(op);

        if (self.ready.load(std::memory_order_acquire) ||
            self.ready.exchange(true, std::memory_order_acq_rel)) {
            // Unsubscribe any stop-callback so we know that CancelIoEx() is not accessing 'op'
            // any more
            self.stopCallback.reset();

            if (errorCode == 0) {
                std::execution::set_value(std::move(self.receiver), bytesTransferred);
            } else {
                std::execution::set_error(std::move(self.receiver),
                                          std::error_code(errorCode, std::system_category()));
            }
        }
    }

    using stop_callback_t = stop_callback_of_t<stop_token_of_t<env_of_t<Receiver>>, cancel_cb>;

    Receiver receiver;
    SOCKET sock;
    WSABUF buffer;
    std::optional<stop_callback_t> stopCallback;
    std::atomic<bool> ready{false};
};

struct recv_sender {
    using sender_concept = std::execution::sender_t;
    SOCKET sock;
    void* data;
    size_t len;

    template<class Receiver>
    recv_op<Receiver> connect(Receiver r) const {
        return recv_op<Receiver>{sock, data, len, std::move(r)};
    }
};

recv_sender async_recv(SOCKET s, void* data, size_t len) {
    return recv_sender{s, data, len};
}

1.4.1. More end-user examples

1.4.1.1. Sudoku solver

This example comes from Kirk Shoop, who ported an example from TBB’s documentation to sender/receiver in his fork of the libunifex repo. It is a Sudoku solver that uses a configurable number of threads to explore the search space for solutions.

The sender/receiver-based Sudoku solver can be found here. Some things that are worth noting about Kirk’s solution:

  1. Although it schedules asychronous work onto a thread pool, and each unit of work will schedule more work, its use of structured concurrency patterns make reference counting unnecessary. The solution does not make use of shared_ptr.

  2. In addition to eliminating the need for reference counting, the use of structured concurrency makes it easy to ensure that resources are cleaned up on all code paths. In contrast, the TBB example that inspired this one leaks memory.

For comparison, the TBB-based Sudoku solver can be found here.

1.4.1.2. File copy

This example also comes from Kirk Shoop which uses sender/receiver to recursively copy the files a directory tree. It demonstrates how sender/receiver can be used to do IO, using a scheduler that schedules work on Linux’s io_uring.

As with the Sudoku example, this example obviates the need for reference counting by employing structured concurrency. It uses iteration with an upper limit to avoid having too many open file handles.

You can find the example here.

1.4.1.3. Echo server

Dietmar Kuehl has proposed networking APIs that use the sender/receiver abstraction (see P2762). He has implemented an echo server as a demo. His echo server code can be found here.

Below, I show the part of the echo server code. This code is executed for each client that connects to the echo server. In a loop, it reads input from a socket and echos the input back to the same socket. All of this, including the loop, is implemented with generic async algorithms.

outstanding.start(
    EX::repeat_effect_until(
          EX::let_value(
              NN::async_read_some(ptr->d_socket,
                                  context.scheduler(),
                                  NN::buffer(ptr->d_buffer))
        | EX::then([ptr](::std::size_t n){
            ::std::cout << "read='" << ::std::string_view(ptr->d_buffer, n) << "'\n";
            ptr->d_done = n == 0;
            return n;
        }),
          [&context, ptr](::std::size_t n){
            return NN::async_write_some(ptr->d_socket,
                                        context.scheduler(),
                                        NN::buffer(ptr->d_buffer, n));
          })
        | EX::then([](auto&&...){})
        , [owner = ::std::move(owner)]{ return owner->d_done; }
    )
);

In this code, NN::async_read_some and NN::async_write_some are asynchronous socket-based networking APIs that return senders. EX::repeat_effect_until, EX::let_value, and EX::then are fully generic sender adaptor algorithms that accept and return senders.

This is a good example of seamless composition of async IO functions with non-IO operations. And by composing the senders in this structured way, all the state for the composite operation -- the repeat_effect_until expression and all its child operations -- is stored altogether in a single object.

1.5. Examples: Algorithms

In this section we show a few simple sender/receiver-based algorithm implementations.

1.5.1. then

namespace stdexec = std::execution;

template <class R, class F>
class _then_receiver : public R {
  F f_;

 public:
  _then_receiver(R r, F f) : R(std::move(r)), f_(std::move(f)) {}

  // Customize set_value by invoking the callable and passing the result to
  // the inner receiver
  template <class... As>
    requires std::invocable<F, As...>
  void set_value(As&&... as) && noexcept {
    try {
      stdexec::set_value(std::move(*this).base(), std::invoke((F&&) f_, (As&&) as...));
    } catch(...) {
      stdexec::set_error(std::move(*this).base(), std::current_exception());
    }
  }
};

template <stdexec::sender S, class F>
struct _then_sender {
  using sender_concept = stdexec::sender_t;
  S s_;
  F f_;

  template <class... Args>
    using _set_value_t = stdexec::completion_signatures<
      stdexec::set_value_t(std::invoke_result_t<F, Args...>)>;

  using _except_ptr_sig =
    stdexec::completion_signatures<stdexec::set_error_t(std::exception_ptr)>;

  // Compute the completion signatures
  template <class Env>
  auto get_completion_signatures(Env&& env) && noexcept
    -> stdexec::transform_completion_signatures_of<
        S, Env, _except_ptr_sig, _set_value_t> {
    return {};
  }

  // Connect:
  template <stdexec::receiver R>
  auto connect(R r) && -> stdexec::connect_result_t<S, _then_receiver<R, F>> {
    return stdexec::connect(
      (S&&) s_, _then_receiver{(R&&) r, (F&&) f_});
  }

  decltype(auto) get_env() const noexcept {
    return get_env(s_);
  }
};

template <stdexec::sender S, class F>
stdexec::sender auto then(S s, F f) {
  return _then_sender<S, F>{(S&&) s, (F&&) f};
}

This code builds a then algorithm that transforms the value(s) from the input sender with a transformation function. The result of the transformation becomes the new value. The other receiver functions (set_error and set_stopped), as well as all receiver queries, are passed through unchanged.

In detail, it does the following:

  1. Defines a receiver in terms of receiver and an invocable that:

    • Defines a constrained set_value member function for transforming the value channel.

    • Delegates set_error and set_stopped to the inner receiver.

  2. Defines a sender that aggregates another sender and the invocable, which defines a connect member function that wraps the incoming receiver in the receiver from (1) and passes it and the incoming sender to std::execution::connect, returning the result. It also defines a get_completion_signatures member function that declares the sender’s completion signatures when executed within a particular environment.

1.5.2. retry

using namespace std;
namespace stdexec = execution;

template<class From, class To>
concept _decays_to = same_as<decay_t<From>, To>;

// _conv needed so we can emplace construct non-movable types into
// a std::optional.
template<invocable F>
struct _conv {
  F f_;

  static_assert(is_nothrow_move_constructible_v<F>);
  explicit _conv(F f) noexcept : f_((F&&) f) {}

  operator invoke_result_t<F>() && {
    return ((F&&) f_)();
  }
};

template<class S, class R>
struct _retry_op;

// pass through all customizations except set_error, which retries
// the operation.
template<class S, class R>
struct _retry_receiver {
  _retry_op<S, R>* o_;

  void set_value(auto&&... as) && noexcept {
    stdexec::set_value(std::move(o_->r_), (decltype(as)&&) as...);
  }

  void set_error(auto&&) && noexcept {
    o_->_retry(); // This causes the op to be retried
  }

  void set_stopped() && noexcept {
    stdexec::set_stopped(std::move(o_->r_));
  }

  decltype(auto) get_env() const noexcept {
    return get_env(o_->r_);
  }
};

// Hold the nested operation state in an optional so we can
// re-construct and re-start it if the operation fails.
template<class S, class R>
struct _retry_op {
  using operation_state_concept = stdexec::operation_state_t;
  using _child_op_t =
    stdexec::connect_result_t<S&, _retry_receiver<S, R>>;

  S s_;
  R r_;
  optional<_child_op_t> o_;

  _op(_op&&) = delete;
  _op(S s, R r)
    : s_(std::move(s)), r_(std::move(r)), o_{_connect()} {}

  auto _connect() noexcept {
    return _conv{[this] {
      return stdexec::connect(s_, _retry_receiver<S, R>{this});
    }};
  }

  void _retry() noexcept {
    try {
      o_.emplace(_connect()); // potentially-throwing
      stdexec::start(*o_);
    } catch(...) {
      stdexec::set_error(std::move(r_), std::current_exception());
    }
  }

  void start() & noexcept {
    stdexec::start(*o_);
  }
};

// Helpers for computing the <code data-opaque bs-autolink-syntax='`then`'>then</code> sender’s completion signatures: 
template <class... Ts>
  using _value_t =
    stdexec::completion_signatures<stdexec::set_value_t(Ts...)>;

template <class>
  using _error_t = stdexec::completion_signatures<>;

using _except_sig =
  stdexec::completion_signatures<stdexec::set_error_t(std::exception_ptr)>;

template<class S>
struct _retry_sender {
  using sender_concept = stdexec::sender_t;
  S s_;
  explicit _retry_sender(S s) : s_(std::move(s)) {}

  // Declare the signatures with which this sender can complete
  template <class Env>
    using _compl_sigs =
      stdexec::transform_completion_signatures_of<
        S&, Env, _except_sig, _value_t, _error_t>;

  template <class Env>
  auto get_completion_signatures(Env&&) const noexcept -> _compl_sigs<Env> {
    return {};
  }

  template <stdexec::receiver R>
    requires stdexec::sender_to<S&, _retry_receiver<S, R>>
  _retry_op<S, R> connect(R r) && {
    return {std::move(s_), std::move(r)};
  }

  decltype(auto) get_env() const noexcept {
    return get_env(s_);
  }
};

template <stdexec::sender S>
stdexec::sender auto retry(S s) {
  return _retry_sender{std::move(s)};
}

The retry algorithm takes a multi-shot sender and causes it to repeat on error, passing through values and stopped signals. Each time the input sender is restarted, a new receiver is connected and the resulting operation state is stored in an optional, which allows us to reinitialize it multiple times.

This example does the following:

  1. Defines a _conv utility that takes advantage of C++17’s guaranteed copy elision to emplace a non-movable type in a std::optional.

  2. Defines a _retry_receiver that holds a pointer back to the operation state. It passes all customizations through unmodified to the inner receiver owned by the operation state except for set_error, which causes a _retry() function to be called instead.

  3. Defines an operation state that aggregates the input sender and receiver, and declares storage for the nested operation state in an optional. Constructing the operation state constructs a _retry_receiver with a pointer to the (under construction) operation state and uses it to connect to the input sender.

  4. Starting the operation state dispatches to start on the inner operation state.

  5. The _retry() function reinitializes the inner operation state by connecting the sender to a new receiver, holding a pointer back to the outer operation state as before.

  6. After reinitializing the inner operation state, _retry() calls start on it, causing the failed operation to be rescheduled.

  7. Defines a _retry_sender that implements a connect member function to return an operation state constructed from the passed-in sender and receiver.

  8. _retry_sender also implements a get_completion_signatures member function to describe the ways this sender may complete when executed in a particular execution resource.

1.6. Examples: Schedulers

In this section we look at some schedulers of varying complexity.

1.6.1. Inline scheduler

namespace stdexec = std::execution;

class inline_scheduler {
  template <class R>
  struct _op {
    using operation_state_concept = operation_state_t;
    R rec_;

    void start() & noexcept {
      stdexec::set_value(std::move(rec_));
    }
  };

  struct _env {
    template <class Tag>
    inline_scheduler query(stdexec::get_completion_scheduler_t<Tag>) const noexcept {
      return {};
    }
  };

  struct _sender {
    using sender_concept = stdexec::sender_t;
    using _compl_sigs = stdexec::completion_signatures<stdexec::set_value_t()>;
    using completion_signatures = _compl_sigs;

    template <stdexec::receiver_of<_compl_sigs> R>
    _op<R> connect(R rec) noexcept(std::is_nothrow_move_constructible_v<R>) {
      return {std::move(rec)};
    }

    _env get_env() const noexcept {
      return {};
    }
  };

 public:
  inline_scheduler() = default;

  _sender schedule() const noexcept {
    return {};
  }

  bool operator==(const inline_scheduler&) const noexcept = default;
};

The inline scheduler is a trivial scheduler that completes immediately and synchronously on the thread that calls std::execution::start on the operation state produced by its sender. In other words, start(connect(schedule(inline_scheduler()), receiver)) is just a fancy way of saying set_value(receiver), with the exception of the fact that start wants to be passed an lvalue.

Although not a particularly useful scheduler, it serves to illustrate the basics of implementing one. The inline_scheduler:

  1. Customizes execution::schedule to return an instance of the sender type _sender.

  2. The _sender type models the sender concept and provides the metadata needed to describe it as a sender of no values and that never calls set_error or set_stopped. This metadata is provided with the help of the execution::completion_signatures utility.

  3. The _sender type customizes execution::connect to accept a receiver of no values. It returns an instance of type _op that holds the receiver by value.

  4. The operation state customizes std::execution::start to call std::execution::set_value on the receiver.

1.6.2. Single thread scheduler

This example shows how to create a scheduler for an execution resource that consists of a single thread. It is implemented in terms of a lower-level execution resource called std::execution::run_loop.

class single_thread_context {
  std::execution::run_loop loop_;
  std::thread thread_;

public:
  single_thread_context()
    : loop_()
    , thread_([this] { loop_.run(); })
  {}
  single_thread_context(single_thread_context&&) = delete;

  ~single_thread_context() {
    loop_.finish();
    thread_.join();
  }

  auto get_scheduler() noexcept {
    return loop_.get_scheduler();
  }

  std::thread::id get_thread_id() const noexcept {
    return thread_.get_id();
  }
};

The single_thread_context owns an event loop and a thread to drive it. In the destructor, it tells the event loop to finish up what it’s doing and then joins the thread, blocking for the event loop to drain.

The interesting bits are in the execution::run_loop context implementation. It is slightly too long to include here, so we only provide a reference to it, but there is one noteworthy detail about its implementation: It uses space in its operation states to build an intrusive linked list of work items. In structured concurrency patterns, the operation states of nested operations compose statically, and in an algorithm like this_thread::sync_wait, the composite operation state lives on the stack for the duration of the operation. The end result is that work can be scheduled onto this thread with zero allocations.

1.7. Examples: Server theme

In this section we look at some examples of how one would use senders to implement an HTTP server. The examples ignore the low-level details of the HTTP server and looks at how senders can be combined to achieve the goals of the project.

General application context:

1.7.1. Composability with execution::let_*

Example context:

Goals:

namespace stdexec = std::execution;

// Returns a sender that yields an http_request object for an incoming request
stdexec::sender auto schedule_request_start(read_requests_ctx ctx) {...}

// Sends a response back to the client; yields a void signal on success
stdexec::sender auto send_response(const http_response& resp) {...}

// Validate that the HTTP request is well-formed; forwards the request on success
stdexec::sender auto validate_request(const http_request& req) {...}

// Handle the request; main application logic
stdexec::sender auto handle_request(const http_request& req) {
  //...
  return stdexec::just(http_response{200, result_body});
}

// Transforms server errors into responses to be sent to the client
stdexec::sender auto error_to_response(std::exception_ptr err) {
  try {
    std::rethrow_exception(err);
  } catch (const std::invalid_argument& e) {
    return stdexec::just(http_response{404, e.what()});
  } catch (const std::exception& e) {
    return stdexec::just(http_response{500, e.what()});
  } catch (...) {
    return stdexec::just(http_response{500, "Unknown server error"});
  }
}

// Transforms cancellation of the server into responses to be sent to the client
stdexec::sender auto stopped_to_response() {
  return stdexec::just(http_response{503, "Service temporarily unavailable"});
}

//...

// The whole flow for transforming incoming requests into responses
stdexec::sender auto snd =
    // get a sender when a new request comes
    schedule_request_start(the_read_requests_ctx)
    // make sure the request is valid; throw if not
    | stdexec::let_value(validate_request)
    // process the request in a function that may be using a different execution resource
    | stdexec::let_value(handle_request)
    // If there are errors transform them into proper responses
    | stdexec::let_error(error_to_response)
    // If the flow is cancelled, send back a proper response
    | stdexec::let_stopped(stopped_to_response)
    // write the result back to the client
    | stdexec::let_value(send_response)
    // done
    ;

// execute the whole flow asynchronously
stdexec::start_detached(std::move(snd));

The example shows how one can separate out the concerns for interpreting requests, validating requests, running the main logic for handling the request, generating error responses, handling cancellation and sending the response back to the client. They are all different phases in the application, and can be joined together with the let_* functions.

All our functions return execution::sender objects, so that they can all generate success, failure and cancellation paths. For example, regardless where an error is generated (reading request, validating request or handling the response), we would have one common block to handle the error, and following error flows is easy.

Also, because of using execution::sender objects at any step, we might expect any of these steps to be completely asynchronous; the overall flow doesn’t care. Regardless of the execution resource in which the steps, or part of the steps are executed in, the flow is still the same.

1.7.2. Moving between execution resources with execution::on and execution::transfer

Example context:

Goals:

namespace stdexec = std::execution;

size_t legacy_read_from_socket(int sock, char* buffer, size_t buffer_len);
void process_read_data(const char* read_data, size_t read_len);
//...

// A sender that just calls the legacy read function
auto snd_read = stdexec::just(sock, buf, buf_len)
              | stdexec::then(legacy_read_from_socket);

// The entire flow
auto snd =
    // start by reading data on the I/O thread
    stdexec::on(io_sched, std::move(snd_read))
    // do the processing on the worker threads pool
    | stdexec::transfer(work_sched)
    // process the incoming data (on worker threads)
    | stdexec::then([buf](int read_len) { process_read_data(buf, read_len); })
    // done
    ;

// execute the whole flow asynchronously
stdexec::start_detached(std::move(snd));

The example assume that we need to wrap some legacy code of reading sockets, and handle execution resource switching. (This style of reading from socket may not be the most efficient one, but it’s working for our purposes.) For performance reasons, the reading from the socket needs to be done on the I/O thread, and all the processing needs to happen on a work-specific execution resource (i.e., thread pool).

Calling execution::on will ensure that the given sender will be started on the given scheduler. In our example, snd_read is going to be started on the I/O scheduler. This sender will just call the legacy code.

The completion-signal will be issued in the I/O execution resource, so we have to move it to the work thread pool. This is achieved with the help of the execution::transfer algorithm. The rest of the processing (in our case, the last call to then) will happen in the work thread pool.

The reader should notice the difference between execution::on and execution::transfer. The execution::on algorithm will ensure that the given sender will start in the specified context, and doesn’t care where the completion-signal for that sender is sent. The execution::transfer algorithm will not care where the given sender is going to be started, but will ensure that the completion-signal of will be transferred to the given context.

1.8. Design changes from P0443

  1. The executor concept has been removed and all of its proposed functionality is now based on schedulers and senders, as per SG1 direction.

  2. Properties are not included in this paper. We see them as a possible future extension, if the committee gets more comfortable with them.

  3. Senders now advertise what scheduler, if any, their evaluation will complete on.

  4. The places of execution of user code in P0443 weren’t precisely defined, whereas they are in this paper. See § 4.5 Senders can propagate completion schedulers.

  5. P0443 did not propose a suite of sender algorithms necessary for writing sender code; this paper does. See § 4.19 User-facing sender factories, § 4.20 User-facing sender adaptors, and § 4.21 User-facing sender consumers.

  6. P0443 did not specify the semantics of variously qualified connect overloads; this paper does. See § 4.7 Senders can be either multi-shot or single-shot.

  7. This paper extends the sender traits/typed sender design to support typed senders whose value/error types depend on type information provided late via the receiver.

  8. Support for untyped senders is dropped; the typed_sender concept is renamed sender; sender_traits is replaced with completion_signatures_of_t.

  9. Specific type erasure facilities are omitted, as per LEWG direction. Type erasure facilities can be built on top of this proposal, as discussed in § 5.9 Customization points.

  10. A specific thread pool implementation is omitted, as per LEWG direction.

  11. Some additional utilities are added:

    • run_loop: An execution resource that provides a multi-producer, single-consumer, first-in-first-out work queue.

    • completion_signatures and transform_completion_signatures: Utilities for describing the ways in which a sender can complete in a declarative syntax.

1.9. Prior art

This proposal builds upon and learns from years of prior art with asynchronous and parallel programming frameworks in C++. In this section, we discuss async abstractions that have previously been suggested as a possible basis for asynchronous algorithms and why they fall short.

1.9.1. Futures

A future is a handle to work that has already been scheduled for execution. It is one end of a communication channel; the other end is a promise, used to receive the result from the concurrent operation and to communicate it to the future.

Futures, as traditionally realized, require the dynamic allocation and management of a shared state, synchronization, and typically type-erasure of work and continuation. Many of these costs are inherent in the nature of "future" as a handle to work that is already scheduled for execution. These expenses rule out the future abstraction for many uses and makes it a poor choice for a basis of a generic mechanism.

1.9.2. Coroutines

C++20 coroutines are frequently suggested as a basis for asynchronous algorithms. It’s fair to ask why, if we added coroutines to C++, are we suggesting the addition of a library-based abstraction for asynchrony. Certainly, coroutines come with huge syntactic and semantic advantages over the alternatives.

Although coroutines are lighter weight than futures, coroutines suffer many of the same problems. Since they typically start suspended, they can avoid synchronizing the chaining of dependent work. However in many cases, coroutine frames require an unavoidable dynamic allocation and indirect function calls. This is done to hide the layout of the coroutine frame from the C++ type system, which in turn makes possible the separate compilation of coroutines and certain compiler optimizations, such as optimization of the coroutine frame size.

Those advantages come at a cost, though. Because of the dynamic allocation of coroutine frames, coroutines in embedded or heterogeneous environments, which often lack support for dynamic allocation, require great attention to detail. And the allocations and indirections tend to complicate the job of the inliner, often resulting in sub-optimal codegen.

The coroutine language feature mitigates these shortcomings somewhat with the HALO optimization Halo: coroutine Heap Allocation eLision Optimization: the joint response, which leverages existing compiler optimizations such as allocation elision and devirtualization to inline the coroutine, completely eliminating the runtime overhead. However, HALO requires a sophisiticated compiler, and a fair number of stars need to align for the optimization to kick in. In our experience, more often than not in real-world code today’s compilers are not able to inline the coroutine, resulting in allocations and indirections in the generated code.

In a suite of generic async algorithms that are expected to be callable from hot code paths, the extra allocations and indirections are a deal-breaker. It is for these reasons that we consider coroutines a poor choise for a basis of all standard async.

1.9.3. Callbacks

Callbacks are the oldest, simplest, most powerful, and most efficient mechanism for creating chains of work, but suffer problems of their own. Callbacks must propagate either errors or values. This simple requirement yields many different interface possibilities. The lack of a standard callback shape obstructs generic design.

Additionally, few of these possibilities accommodate cancellation signals when the user requests upstream work to stop and clean up.

1.10. Field experience

1.10.1. libunifex

This proposal draws heavily from our field experience with libunifex. Libunifex implements all of the concepts and customization points defined in this paper (with slight variations -- the design of P2300 has evolved due to LEWG feedback), many of this paper’s algorithms (some under different names), and much more besides.

Libunifex has several concrete schedulers in addition to the run_loop suggested here (where it is called manual_event_loop). It has schedulers that dispatch efficiently to epoll and io_uring on Linux and the Windows Thread Pool on Windows.

In addition to the proposed interfaces and the additional schedulers, it has several important extensions to the facilities described in this paper, which demonstrate directions in which these abstractions may be evolved over time, including:

Libunifex has seen heavy production use at Meta. An employee summarizes it as follows:

As of June, 2023, Unifex is still used in production at Meta. It’s used to express the asynchrony in rsys, and is therefore serving video calling to billions of people every month on Meta’s social networking apps on iOS, Android, Windows, and macOS. It’s also serving the Virtual Desktop experience on Oculus Quest devices, and some internal uses that run on Linux.

One team at Meta has migrated from folly::Future to unifex::task and seen significant developer efficiency improvements. Coroutines are easier to understand than chained futures so the team was able to meet requirements for certain constrained environments that would have been too complicated to maintain with futures.

In all the cases mentioned above, developers mix-and-match between the sender algorithms in Unifex and Unifex’s coroutine type, unifex::task. We also rely on unifex::task's scheduler affinity to minimize surprise when programming with coroutines.

1.10.2. stdexec

stdexec is the reference implementation of this proposal. It is a complete implementation, written from the specification in this paper, and is current with \R8.

The original purpose of stdexec was to help find specification bugs and to harden the wording of the proposal, but it has since become one of NVIDIA’s core C++ libraries for high-performance computing. In addition to the facilities proposed in this paper, stdexec has schedulers for CUDA, Intel TBB, and MacOS. Like libunifex, its scope has also expanded to include a streaming abstraction and stream algorithms, and time-based schedulers and algorithms.

The stdexec project has seen lots of community interest and contributions. At the time of writing (March, 2024), the GitHub repository has 1.2k stars, 130 forks, and 50 contributors.

stdexec is fit for broad use and for ultimate contribution to libc++.

1.10.3. Other implementations

The authors are aware of a number of other implementations of sender/receiver from this paper. These are presented here in perceived order of maturity and field experience.

1.10.4. Inspirations

This proposal also draws heavily from our experience with Thrust and Agency. It is also inspired by the needs of countless other C++ frameworks for asynchrony, parallelism, and concurrency, including:

2. Revision history

2.1. R9

The changes since R8 are as follows:

Fixes:

Enhancements:

2.2. R8

The changes since R7 are as follows:

Fixes:

Enhancements:

2.3. R7

The changes since R6 are as follows:

Fixes:

Enhancements:

2.4. R6

The changes since R5 are as follows:

Fixes:

Enhancements:

2.4.1. Environments and attributes

In earlier revisions, receivers, senders, and schedulers all were directly queryable. In R4, receiver queries were moved into a separate "environment" object, obtainable from a receiver with a get_env accessor. In R6, the sender queries are given similar treatment, relocating to a "attributes" object obtainable from a sender with a get_attrs accessor. This was done to solve a number of design problems with the split and ensure_started algorithms; _e.g._, see NVIDIA/stdexec#466.

Schedulers, however, remain directly queryable. As lightweight handles that are required to be movable and copyable, there is little reason to want to dispose of a scheduler and yet persist the scheduler’s queries.

This revision also makes operation states directly queryable, even though there isn’t yet a use for such. Some early prototypes of cooperative bulk parallel sender algorithms done at NVIDIA suggest the utility of forwardable operation state queries. The authors chose to make opstates directly queryable since the opstate object is itself required to be kept alive for the duration of asynchronous operation.

2.5. R5

The changes since R4 are as follows:

Fixes:

Enhancements:

2.6. R4

The changes since R3 are as follows:

Fixes:

Enhancements:

2.6.1. Dependently-typed senders

Background:

In the sender/receiver model, as with coroutines, contextual information about the current execution is most naturally propagated from the consumer to the producer. In coroutines, that means information like stop tokens, allocators and schedulers are propagated from the calling coroutine to the callee. In sender/receiver, that means that that contextual information is associated with the receiver and is queried by the sender and/or operation state after the sender and the receiver are connect-ed.

Problem:

The implication of the above is that the sender alone does not have all the information about the async computation it will ultimately initiate; some of that information is provided late via the receiver. However, the sender_traits mechanism, by which an algorithm can introspect the value and error types the sender will propagate, only accepts a sender parameter. It does not take into consideration the type information that will come in late via the receiver. The effect of this is that some senders cannot be typed senders when they otherwise could be.

Example:

To get concrete, consider the case of the "get_scheduler()" sender: when connect-ed and start-ed, it queries the receiver for its associated scheduler and passes it back to the receiver through the value channel. That sender’s "value type" is the type of the receiver’s scheduler. What then should sender_traits<get_scheduler_sender>::value_types report for the get_scheduler()'s value type? It can’t answer because it doesn’t know.

This causes knock-on problems since some important algorithms require a typed sender, such as sync_wait. To illustrate the problem, consider the following code:

namespace ex = std::execution;

ex::sender auto task =
  ex::let_value(
    ex::get_scheduler(), // Fetches scheduler from receiver.
    [](auto current_sched) {
      // Lauch some nested work on the current scheduler:
      return ex::on(current_sched, nested work...);
    });

std::this_thread::sync_wait(std::move(task));

The code above is attempting to schedule some work onto the sync_wait's run_loop execution resource. But let_value only returns a typed sender when the input sender is typed. As we explained above, get_scheduler() is not typed, so task is likewise not typed. Since task isn’t typed, it cannot be passed to sync_wait which is expecting a typed sender. The above code would fail to compile.

Solution:

The solution is conceptually quite simple: extend the sender_traits mechanism to optionally accept a receiver in addition to the sender. The algorithms can use sender_traits<Sender, Receiver> to inspect the async operation’s completion-signals. The typed_sender concept would also need to take an optional receiver parameter. This is the simplest change, and it would solve the immediate problem.

Design:

Using the receiver type to compute the sender traits turns out to have pitfalls in practice. Many receivers make use of that type information in their implementation. It is very easy to create cycles in the type system, leading to inscrutible errors. The design pursued in R4 is to give receivers an associated environment object -- a bag of key/value pairs -- and to move the contextual information (schedulers, etc) out of the receiver and into the environment. The sender_traits template and the typed_sender concept, rather than taking a receiver, take an environment. This is a much more robust design.

A further refinement of this design would be to separate the receiver and the environment entirely, passing then as separate arguments along with the sender to connect. This paper does not propose that change.

Impact:

This change, apart from increasing the expressive power of the sender/receiver abstraction, has the following impact:

"Has it been implemented?"

Yes, the reference implementation, which can be found at https://github.com/NVIDIA/stdexec, has implemented this design as well as some dependently-typed senders to confirm that it works.

Implementation experience

Although this change has not yet been made in libunifex, the most widely adopted sender/receiver implementation, a similar design can be found in Folly’s coroutine support library. In Folly.Coro, it is possible to await a special awaitable to obtain the current coroutine’s associated scheduler (called an executor in Folly).

For instance, the following Folly code grabs the current executor, schedules a task for execution on that executor, and starts the resulting (scheduled) task by enqueueing it for execution.

// From Facebook’s Folly open source library:
template <class T>
folly::coro::Task<void> CancellableAsyncScope::co_schedule(folly::coro::Task<T>&& task) {
  this->add(std::move(task).scheduleOn(co_await co_current_executor));
  co_return;
}

Facebook relies heavily on this pattern in its coroutine code. But as described above, this pattern doesn’t work with R3 of std::execution because of the lack of dependently-typed schedulers. The change to sender_traits in R4 rectifies that.

Why now?

The authors are loathe to make any changes to the design, however small, at this stage of the C++23 release cycle. But we feel that, for a relatively minor design change -- adding an extra template parameter to sender_traits and typed_sender -- the returns are large enough to justify the change. And there is no better time to make this change than as early as possible.

One might wonder why this missing feature not been added to sender/receiver before now. The designers of sender/receiver have long been aware of the need. What was missing was a clean, robust, and simple design for the change, which we now have.

Drive-by:

We took the opportunity to make an additional drive-by change: Rather than providing the sender traits via a class template for users to specialize, we changed it into a sender query: get_completion_signatures(sender, env). That function’s return type is used as the sender’s traits. The authors feel this leads to a more uniform design and gives sender authors a straightforward way to make the value/error types dependent on the cv- and ref-qualification of the sender if need be.

Details:

Below are the salient parts of the new support for dependently-typed senders in R4:

2.7. R3

The changes since R2 are as follows:

Fixes:

Enhancements:

2.8. R2

The changes since R1 are as follows:

2.9. R1

The changes since R0 are as follows:

2.10. R0

Initial revision.

3. Design - introduction

The following three sections describe the entirety of the proposed design.

3.1. Conventions

The following conventions are used throughout the design section:

  1. The namespace proposed in this paper is the same as in A Unified Executors Proposal for C++: std::execution; however, for brevity, the std:: part of this name is omitted. When you see execution::foo, treat that as std::execution::foo.

  2. Universal references and explicit calls to std::move/std::forward are omitted in code samples and signatures for simplicity; assume universal references and perfect forwarding unless stated otherwise.

  3. None of the names proposed here are names that we are particularly attached to; consider the names to be reasonable placeholders that can freely be changed, should the committee want to do so.

3.2. Queries and algorithms

A query is a callable that takes some set of objects (usually one) as parameters and returns facts about those objects without modifying them. Queries are usually customization point objects, but in some cases may be functions.

An algorithm is a callable that takes some set of objects as parameters and causes those objects to do something. Algorithms are usually customization point objects, but in some cases may be functions.

4. Design - user side

4.1. Execution resources describe the place of execution

An execution resource is a resource that represents the place where execution will happen. This could be a concrete resource - like a specific thread pool object, or a GPU - or a more abstract one, like the current thread of execution. Execution contexts don’t need to have a representation in code; they are simply a term describing certain properties of execution of a function.

4.2. Schedulers represent execution resources

A scheduler is a lightweight handle that represents a strategy for scheduling work onto an execution resource. Since execution resources don’t necessarily manifest in C++ code, it’s not possible to program directly against their API. A scheduler is a solution to that problem: the scheduler concept is defined by a single sender algorithm, schedule, which returns a sender that will complete on an execution resource determined by the scheduler. Logic that you want to run on that context can be placed in the receiver’s completion-signalling method.

execution::scheduler auto sch = thread_pool.scheduler();
execution::sender auto snd = execution::schedule(sch);
// snd is a sender (see below) describing the creation of a new execution resource
// on the execution resource associated with sch

Note that a particular scheduler type may provide other kinds of scheduling operations which are supported by its associated execution resource. It is not limited to scheduling purely using the execution::schedule API.

Future papers will propose additional scheduler concepts that extend scheduler to add other capabilities. For example:

4.3. Senders describe work

A sender is an object that describes work. Senders are similar to futures in existing asynchrony designs, but unlike futures, the work that is being done to arrive at the values they will send is also directly described by the sender object itself. A sender is said to send some values if a receiver connected (see § 5.3 execution::connect) to that sender will eventually receive said values.

The primary defining sender algorithm is § 5.3 execution::connect; this function, however, is not a user-facing API; it is used to facilitate communication between senders and various sender algorithms, but end user code is not expected to invoke it directly.

The way user code is expected to interact with senders is by using sender algorithms. This paper proposes an initial set of such sender algorithms, which are described in § 4.4 Senders are composable through sender algorithms, § 4.19 User-facing sender factories, § 4.20 User-facing sender adaptors, and § 4.21 User-facing sender consumers. For example, here is how a user can create a new sender on a scheduler, attach a continuation to it, and then wait for execution of the continuation to complete:

execution::scheduler auto sch = thread_pool.scheduler();
execution::sender auto snd = execution::schedule(sch);
execution::sender auto cont = execution::then(snd, []{
    std::fstream file{ "result.txt" };
    file << compute_result;
});

this_thread::sync_wait(cont);
// at this point, cont has completed execution

4.4. Senders are composable through sender algorithms

Asynchronous programming often departs from traditional code structure and control flow that we are familiar with. A successful asynchronous framework must provide an intuitive story for composition of asynchronous work: expressing dependencies, passing objects, managing object lifetimes, etc.

The true power and utility of senders is in their composability. With senders, users can describe generic execution pipelines and graphs, and then run them on and across a variety of different schedulers. Senders are composed using sender algorithms:

4.5. Senders can propagate completion schedulers

One of the goals of executors is to support a diverse set of execution resources, including traditional thread pools, task and fiber frameworks (like HPX Legion), and GPUs and other accelerators (managed by runtimes such as CUDA or SYCL). On many of these systems, not all execution agents are created equal and not all functions can be run on all execution agents. Having precise control over the execution resource used for any given function call being submitted is important on such systems, and the users of standard execution facilities will expect to be able to express such requirements.

A Unified Executors Proposal for C++ was not always clear about the place of execution of any given piece of code. Precise control was present in the two-way execution API present in earlier executor designs, but it has so far been missing from the senders design. There has been a proposal (Towards C++23 executors: A proposal for an initial set of algorithms) to provide a number of sender algorithms that would enforce certain rules on the places of execution of the work described by a sender, but we have found those sender algorithms to be insufficient for achieving the best performance on all platforms that are of interest to us. The implementation strategies that we are aware of result in one of the following situations:

  1. trying to submit work to one execution resource (such as a CPU thread pool) from another execution resource (such as a GPU or a task framework), which assumes that all execution agents are as capable as a std::thread (which they aren’t).

  2. forcibly interleaving two adjacent execution graph nodes that are both executing on one execution resource (such as a GPU) with glue code that runs on another execution resource (such as a CPU), which is prohibitively expensive for some execution resources (such as CUDA or SYCL).

  3. having to customise most or all sender algorithms to support an execution resource, so that you can avoid problems described in 1. and 2, which we believe is impractical and brittle based on months of field experience attempting this in Agency.

None of these implementation strategies are acceptable for many classes of parallel runtimes, such as task frameworks (like HPX) or accelerator runtimes (like CUDA or SYCL).

Therefore, in addition to the on sender algorithm from Towards C++23 executors: A proposal for an initial set of algorithms, we are proposing a way for senders to advertise what scheduler (and by extension what execution resource) they will complete on. Any given sender may have completion schedulers for some or all of the signals (value, error, or stopped) it completes with (for more detail on the completion-signals, see § 5.1 Receivers serve as glue between senders). When further work is attached to that sender by invoking sender algorithms, that work will also complete on an appropriate completion scheduler.

4.5.1. execution::get_completion_scheduler

get_completion_scheduler is a query that retrieves the completion scheduler for a specific completion-signal from a sender’s environment. For a sender that lacks a completion scheduler query for a given signal, calling get_completion_scheduler is ill-formed. If a sender advertises a completion scheduler for a signal in this way, that sender must ensure that it sends that signal on an execution agent belonging to an execution resource represented by a scheduler returned from this function. See § 4.5 Senders can propagate completion schedulers for more details.

execution::scheduler auto cpu_sched = new_thread_scheduler{};
execution::scheduler auto gpu_sched = cuda::scheduler();

execution::sender auto snd0 = execution::schedule(cpu_sched);
execution::scheduler auto completion_sch0 =
  execution::get_completion_scheduler<execution::set_value_t>(get_env(snd0));
// completion_sch0 is equivalent to cpu_sched

execution::sender auto snd1 = execution::then(snd0, []{
    std::cout << "I am running on cpu_sched!\n";
});
execution::scheduler auto completion_sch1 =
  execution::get_completion_scheduler<execution::set_value_t>(get_env(snd1));
// completion_sch1 is equivalent to cpu_sched

execution::sender auto snd2 = execution::transfer(snd1, gpu_sched);
execution::sender auto snd3 = execution::then(snd2, []{
    std::cout << "I am running on gpu_sched!\n";
});
execution::scheduler auto completion_sch3 =
  execution::get_completion_scheduler<execution::set_value_t>(get_env(snd3));
// completion_sch3 is equivalent to gpu_sched

4.6. Execution resource transitions are explicit

A Unified Executors Proposal for C++ does not contain any mechanisms for performing an execution resource transition. The only sender algorithm that can create a sender that will move execution to a specific execution resource is execution::schedule, which does not take an input sender. That means that there’s no way to construct sender chains that traverse different execution resources. This is necessary to fulfill the promise of senders being able to replace two-way executors, which had this capability.

We propose that, for senders advertising their completion scheduler, all execution resource transitions must be explicit; running user code anywhere but where they defined it to run must be considered a bug.

The execution::transfer sender adaptor performs a transition from one execution resource to another:

execution::scheduler auto sch1 = ...;
execution::scheduler auto sch2 = ...;

execution::sender auto snd1 = execution::schedule(sch1);
execution::sender auto then1 = execution::then(snd1, []{
    std::cout << "I am running on sch1!\n";
});

execution::sender auto snd2 = execution::transfer(then1, sch2);
execution::sender auto then2 = execution::then(snd2, []{
    std::cout << "I am running on sch2!\n";
});

this_thread::sync_wait(then2);

4.7. Senders can be either multi-shot or single-shot

Some senders may only support launching their operation a single time, while others may be repeatable and support being launched multiple times. Executing the operation may consume resources owned by the sender.

For example, a sender may contain a std::unique_ptr that it will be transferring ownership of to the operation-state returned by a call to execution::connect so that the operation has access to this resource. In such a sender, calling execution::connect consumes the sender such that after the call the input sender is no longer valid. Such a sender will also typically be move-only so that it can maintain unique ownership of that resource.

A single-shot sender can only be connected to a receiver at most once. Its implementation of execution::connect only has overloads for an rvalue-qualified sender. Callers must pass the sender as an rvalue to the call to execution::connect, indicating that the call consumes the sender.

A multi-shot sender can be connected to multiple receivers and can be launched multiple times. Multi-shot senders customise execution::connect to accept an lvalue reference to the sender. Callers can indicate that they want the sender to remain valid after the call to execution::connect by passing an lvalue reference to the sender to call these overloads. Multi-shot senders should also define overloads of execution::connect that accept rvalue-qualified senders to allow the sender to be also used in places where only a single-shot sender is required.

If the user of a sender does not require the sender to remain valid after connecting it to a receiver then it can pass an rvalue-reference to the sender to the call to execution::connect. Such usages should be able to accept either single-shot or multi-shot senders.

If the caller does wish for the sender to remain valid after the call then it can pass an lvalue-qualified sender to the call to execution::connect. Such usages will only accept multi-shot senders.

Algorithms that accept senders will typically either decay-copy an input sender and store it somewhere for later usage (for example as a data-member of the returned sender) or will immediately call execution::connect on the input sender, such as in this_thread::sync_wait or execution::start_detached.

Some multi-use sender algorithms may require that an input sender be copy-constructible but will only call execution::connect on an rvalue of each copy, which still results in effectively executing the operation multiple times. Other multi-use sender algorithms may require that the sender is move-constructible but will invoke execution::connect on an lvalue reference to the sender.

For a sender to be usable in both multi-use scenarios, it will generally be required to be both copy-constructible and lvalue-connectable.

4.8. Senders are forkable

Any non-trivial program will eventually want to fork a chain of senders into independent streams of work, regardless of whether they are single-shot or multi-shot. For instance, an incoming event to a middleware system may be required to trigger events on more than one downstream system. This requires that we provide well defined mechanisms for making sure that connecting a sender multiple times is possible and correct.

The split sender adaptor facilitates connecting to a sender multiple times, regardless of whether it is single-shot or multi-shot:

auto some_algorithm(execution::sender auto&& input) {
    execution::sender auto multi_shot = split(input);
    // "multi_shot" is guaranteed to be multi-shot,
    // regardless of whether "input" was multi-shot or not

    return when_all(
      then(multi_shot, [] { std::cout << "First continuation\n"; }),
      then(multi_shot, [] { std::cout << "Second continuation\n"; })
    );
}

4.9. Senders support cancellation

Senders are often used in scenarios where the application may be concurrently executing multiple strategies for achieving some program goal. When one of these strategies succeeds (or fails) it may not make sense to continue pursuing the other strategies as their results are no longer useful.

For example, we may want to try to simultaneously connect to multiple network servers and use whichever server responds first. Once the first server responds we no longer need to continue trying to connect to the other servers.

Ideally, in these scenarios, we would somehow be able to request that those other strategies stop executing promptly so that their resources (e.g. cpu, memory, I/O bandwidth) can be released and used for other work.

While the design of senders has support for cancelling an operation before it starts by simply destroying the sender or the operation-state returned from execution::connect() before calling execution::start(), there also needs to be a standard, generic mechanism to ask for an already-started operation to complete early.

The ability to be able to cancel in-flight operations is fundamental to supporting some kinds of generic concurrency algorithms.

For example:

The mechanism used for communcating cancellation-requests, or stop-requests, needs to have a uniform interface so that generic algorithms that compose sender-based operations, such as the ones listed above, are able to communicate these cancellation requests to senders that they don’t know anything about.

The design is intended to be composable so that cancellation of higher-level operations can propagate those cancellation requests through intermediate layers to lower-level operations that need to actually respond to the cancellation requests.

For example, we can compose the algorithms mentioned above so that child operations are cancelled when any one of the multiple cancellation conditions occurs:

sender auto composed_cancellation_example(auto query) {
  return stop_when(
    timeout(
      when_all(
        first_successful(
          query_server_a(query),
          query_server_b(query)),
        load_file("some_file.jpg")),
      5s),
    cancelButton.on_click());
}

In this example, if we take the operation returned by query_server_b(query), this operation will receive a stop-request when any of the following happens:

Note that within this code there is no explicit mention of cancellation, stop-tokens, callbacks, etc. yet the example fully supports and responds to the various cancellation sources.

The intent of the design is that the common usage of cancellation in sender/receiver-based code is primarily through use of concurrency algorithms that manage the detailed plumbing of cancellation for you. Much like algorithms that compose senders relieve the user from having to write their own receiver types, algorithms that introduce concurrency and provide higher-level cancellation semantics relieve the user from having to deal with low-level details of cancellation.

4.9.1. Cancellation design summary

The design of cancellation described in this paper is built on top of and extends the std::stop_token-based cancellation facilities added in C++20, first proposed in Composable cancellation for sender-based async operations.

At a high-level, the facilities proposed by this paper for supporting cancellation include:

In addition, there are requirements added to some of the algorithms to specify what their cancellation behaviour is and what the requirements of customisations of those algorithms are with respect to cancellation.

The key component that enables generic cancellation within sender-based operations is the execution::get_stop_token() CPO. This CPO takes a single parameter, which is the execution environment of the receiver passed to execution::connect, and returns a std::stoppable_token that the operation can use to check for stop-requests for that operation.

As the caller of execution::connect typically has control over the receiver type it passes, it is able to customise the std::execution::get_env() CPO for that receiver to return an execution environment that hooks the execution::get_stop_token() CPO to return a stop-token that the receiver has control over and that it can use to communicate a stop-request to the operation once it has started.

4.9.2. Support for cancellation is optional

Support for cancellation is optional, both on part of the author of the receiver and on part of the author of the sender.

If the receiver’s execution environment does not customise the execution::get_stop_token() CPO then invoking the CPO on that receiver’s environment will invoke the default implementation which returns std::never_stop_token. This is a special stoppable_token type that is statically known to always return false from the stop_possible() method.

Sender code that tries to use this stop-token will in general result in code that handles stop-requests being compiled out and having little to no run-time overhead.

If the sender doesn’t call execution::get_stop_token(), for example because the operation does not support cancellation, then it will simply not respond to stop-requests from the caller.

Note that stop-requests are generally racy in nature as there is often a race betwen an operation completing naturally and the stop-request being made. If the operation has already completed or past the point at which it can be cancelled when the stop-request is sent then the stop-request may just be ignored. An application will typically need to be able to cope with senders that might ignore a stop-request anyway.

4.9.3. Cancellation is inherently racy

Usually, an operation will attach a stop-callback at some point inside the call to execution::start() so that a subsequent stop-request will interrupt the logic.

A stop-request can be issued concurrently from another thread. This means the implementation of execution::start() needs to be careful to ensure that, once a stop-callback has been registered, that there are no data-races between a potentially concurrently-executing stop-callback and the rest of the execution::start() implementation.

An implementation of execution::start() that supports cancellation will generally need to perform (at least) two separate steps: launch the operation, subscribe a stop-callback to the receiver’s stop-token. Care needs to be taken depending on the order in which these two steps are performed.

If the stop-callback is subscribed first and then the operation is launched, care needs to be taken to ensure that a stop-request that invokes the stop-callback on another thread after the stop-callback is registered but before the operation finishes launching does not either result in a missed cancellation request or a data-race. e.g. by performing an atomic write after the launch has finished executing

If the operation is launched first and then the stop-callback is subscribed, care needs to be taken to ensure that if the launched operation completes concurrently on another thread that it does not destroy the operation-state until after the stop-callback has been registered. e.g. by having the execution::start implementation write to an atomic variable once it has finished registering the stop-callback and having the concurrent completion handler check that variable and either call the completion-signalling operation or store the result and defer calling the receiver’s completion-signalling operation to the execution::start() call (which is still executing).

For an example of an implementation strategy for solving these data-races see § 1.4 Asynchronous Windows socket recv.

4.9.4. Cancellation design status

This paper currently includes the design for cancellation as proposed in Composable cancellation for sender-based async operations - "Composable cancellation for sender-based async operations". P2175R0 contains more details on the background motivation and prior-art and design rationale of this design.

It is important to note, however, that initial review of this design in the SG1 concurrency subgroup raised some concerns related to runtime overhead of the design in single-threaded scenarios and these concerns are still being investigated.

The design of P2175R0 has been included in this paper for now, despite its potential to change, as we believe that support for cancellation is a fundamental requirement for an async model and is required in some form to be able to talk about the semantics of some of the algorithms proposed in this paper.

This paper will be updated in the future with any changes that arise from the investigations into P2175R0.

4.10. Sender factories and adaptors are lazy

In an earlier revision of this paper, some of the proposed algorithms supported executing their logic eagerly; i.e., before the returned sender has been connected to a receiver and started. These algorithms were removed because eager execution has a number of negative semantic and performance implications.

We have originally included this functionality in the paper because of a long-standing belief that eager execution is a mandatory feature to be included in the standard Executors facility for that facility to be acceptable for accelerator vendors. A particular concern was that we must be able to write generic algorithms that can run either eagerly or lazily, depending on the kind of an input sender or scheduler that have been passed into them as arguments. We considered this a requirement, because the _latency_ of launching work on an accelerator can sometimes be considerable.

However, in the process of working on this paper and implementations of the features proposed within, our set of requirements has shifted, as we understood the different implementation strategies that are available for the feature set of this paper better, and, after weighting the earlier concerns against the points presented below, we have arrived at the conclusion that a purely lazy model is enough for most algorithms, and users who intend to launch work earlier may use an algorithm such as ensure_started to achieve that goal. We have also come to deeply appreciate the fact that a purely lazy model allows both the implementation and the compiler to have a much better understanding of what the complete graph of tasks looks like, allowing them to better optimize the code - also when targetting accelerators.

4.10.1. Eager execution leads to detached work or worse

One of the questions that arises with APIs that can potentially return eagerly-executing senders is "What happens when those senders are destructed without a call to execution::connect?" or similarly, "What happens if a call to execution::connect is made, but the returned operation state is destroyed before execution::start is called on that operation state"?

In these cases, the operation represented by the sender is potentially executing concurrently in another thread at the time that the destructor of the sender and/or operation-state is running. In the case that the operation has not completed executing by the time that the destructor is run we need to decide what the semantics of the destructor is.

There are three main strategies that can be adopted here, none of which is particularly satisfactory:

  1. Make this undefined-behaviour - the caller must ensure that any eagerly-executing sender is always joined by connecting and starting that sender. This approach is generally pretty hostile to programmers, particularly in the presence of exceptions, since it complicates the ability to compose these operations.

    Eager operations typically need to acquire resources when they are first called in order to start the operation early. This makes eager algorithms prone to failure. Consider, then, what might happen in an expression such as when_all(eager_op_1(), eager_op_2()). Imagine eager_op_1() starts an asynchronous operation successfully, but then eager_op_2() throws. For lazy senders, that failure happens in the context of the when_all algorithm, which handles the failure and ensures that async work joins on all code paths. In this case though -- the eager case -- the child operation has failed even before when_all has been called.

    It then becomes the responsibility, not of the algorithm, but of the end user to handle the exception and ensure that eager_op_1() is joined before allowing the exception to propagate. If they fail to do that, they incur undefined behavior.

  2. Detach from the computation - let the operation continue in the background - like an implicit call to std::thread::detach(). While this approach can work in some circumstances for some kinds of applications, in general it is also pretty user-hostile; it makes it difficult to reason about the safe destruction of resources used by these eager operations. In general, detached work necessitates some kind of garbage collection; e.g., std::shared_ptr, to ensure resources are kept alive until the operations complete, and can make clean shutdown nigh impossible.

  3. Block in the destructor until the operation completes. This approach is probably the safest to use as it preserves the structured nature of the concurrent operations, but also introduces the potential for deadlocking the application if the completion of the operation depends on the current thread making forward progress.

    The risk of deadlock might occur, for example, if a thread-pool with a small number of threads is executing code that creates a sender representing an eagerly-executing operation and then calls the destructor of that sender without joining it (e.g. because an exception was thrown). If the current thread blocks waiting for that eager operation to complete and that eager operation cannot complete until some entry enqueued to the thread-pool’s queue of work is run then the thread may wait for an indefinite amount of time. If all threads of the thread-pool are simultaneously performing such blocking operations then deadlock can result.

There are also minor variations on each of these choices. For example:

  1. A variation of (1): Call std::terminate if an eager sender is destructed without joining it. This is the approach that std::thread destructor takes.

  2. A variation of (2): Request cancellation of the operation before detaching. This reduces the chances of operations continuing to run indefinitely in the background once they have been detached but does not solve the lifetime- or shutdown-related challenges.

  3. A variation of (3): Request cancellation of the operation before blocking on its completion. This is the strategy that std::jthread uses for its destructor. It reduces the risk of deadlock but does not eliminate it.

4.10.2. Eager senders complicate algorithm implementations

Algorithms that can assume they are operating on senders with strictly lazy semantics are able to make certain optimizations that are not available if senders can be potentially eager. With lazy senders, an algorithm can safely assume that a call to execution::start on an operation state strictly happens before the execution of that async operation. This frees the algorithm from needing to resolve potential race conditions. For example, consider an algorithm sequence that puts async operations in sequence by starting an operation only after the preceding one has completed. In an expression like sequence(a(), then(src, [] { b(); }), c()), one may reasonably assume that a(), b() and c() are sequenced and therefore do not need synchronisation. Eager algorithms break that assumption.

When an algorithm needs to deal with potentially eager senders, the potential race conditions can be resolved one of two ways, neither of which is desirable:

  1. Assume the worst and implement the algorithm defensively, assuming all senders are eager. This obviously has overheads both at runtime and in algorithm complexity. Resolving race conditions is hard.

  2. Require senders to declare whether they are eager or not with a query. Algorithms can then implement two different implementation strategies, one for strictly lazy senders and one for potentially eager senders. This addresses the performance problem of (1) while compounding the complexity problem.

4.10.3. Eager senders incur cancellation-related overhead

Another implication of the use of eager operations is with regards to cancellation. The eagerly executing operation will not have access to the caller’s stop token until the sender is connected to a receiver. If we still want to be able to cancel the eager operation then it will need to create a new stop source and pass its associated stop token down to child operations. Then when the returned sender is eventually connected it will register a stop callback with the receiver’s stop token that will request stop on the eager sender’s stop source.

As the eager operation does not know at the time that it is launched what the type of the receiver is going to be, and thus whether or not the stop token returned from execution::get_stop_token is an std::unstoppable_token or not, the eager operation is going to need to assume it might be later connected to a receiver with a stop token that might actually issue a stop request. Thus it needs to declare space in the operation state for a type-erased stop callback and incur the runtime overhead of supporting cancellation, even if cancellation will never be requested by the caller.

The eager operation will also need to do this to support sending a stop request to the eager operation in the case that the sender representing the eager work is destroyed before it has been joined (assuming strategy (5) or (6) listed above is chosen).

4.10.4. Eager senders cannot access execution resource from the receiver

In sender/receiver, contextual information is passed from parent operations to their children by way of receivers. Information like stop tokens, allocators, current scheduler, priority, and deadline are propagated to child operations with custom receivers at the time the operation is connected. That way, each operation has the contextual information it needs before it is started.

But if the operation is started before it is connected to a receiver, then there isn’t a way for a parent operation to communicate contextual information to its child operations, which may complete before a receiver is ever attached.

4.11. Schedulers advertise their forward progress guarantees

To decide whether a scheduler (and its associated execution resource) is sufficient for a specific task, it may be necessary to know what kind of forward progress guarantees it provides for the execution agents it creates. The C++ Standard defines the following forward progress guarantees:

This paper introduces a scheduler query function, get_forward_progress_guarantee, which returns one of the enumerators of a new enum type, forward_progress_guarantee. Each enumerator of forward_progress_guarantee corresponds to one of the aforementioned guarantees.

4.12. Most sender adaptors are pipeable

To facilitate an intuitive syntax for composition, most sender adaptors are pipeable; they can be composed (piped) together with operator|. This mechanism is similar to the operator| composition that C++ range adaptors support and draws inspiration from piping in *nix shells. Pipeable sender adaptors take a sender as their first parameter and have no other sender parameters.

a | b will pass the sender a as the first argument to the pipeable sender adaptor b. Pipeable sender adaptors support partial application of the parameters after the first. For example, all of the following are equivalent:

execution::bulk(snd, N, [] (std::size_t i, auto d) {});
execution::bulk(N, [] (std::size_t i, auto d) {})(snd);
snd | execution::bulk(N, [] (std::size_t i, auto d) {});

Piping enables you to compose together senders with a linear syntax. Without it, you’d have to use either nested function call syntax, which would cause a syntactic inversion of the direction of control flow, or you’d have to introduce a temporary variable for each stage of the pipeline. Consider the following example where we want to execute first on a CPU thread pool, then on a CUDA GPU, then back on the CPU thread pool:

Syntax Style Example
Function call
(nested)
auto snd = execution::then(
             execution::transfer(
               execution::then(
                 execution::transfer(
                   execution::then(
                     execution::schedule(thread_pool.scheduler())
                     []{ return 123; }),
                   cuda::new_stream_scheduler()),
                 [](int i){ return 123 * 5; }),
               thread_pool.scheduler()),
             [](int i){ return i - 5; });
auto [result] = this_thread::sync_wait(snd).value();
// result == 610
Function call
(named temporaries)
auto snd0 = execution::schedule(thread_pool.scheduler());
auto snd1 = execution::then(snd0, []{ return 123; });
auto snd2 = execution::transfer(snd1, cuda::new_stream_scheduler());
auto snd3 = execution::then(snd2, [](int i){ return 123 * 5; })
auto snd4 = execution::transfer(snd3, thread_pool.scheduler())
auto snd5 = execution::then(snd4, [](int i){ return i - 5; });
auto [result] = *this_thread::sync_wait(snd4);
// result == 610
Pipe
auto snd = execution::schedule(thread_pool.scheduler())
         | execution::then([]{ return 123; })
         | execution::transfer(cuda::new_stream_scheduler())
         | execution::then([](int i){ return 123 * 5; })
         | execution::transfer(thread_pool.scheduler())
         | execution::then([](int i){ return i - 5; });
auto [result] = this_thread::sync_wait(snd).value();
// result == 610

Certain sender adaptors are not pipeable, because using the pipeline syntax can result in confusion of the semantics of the adaptors involved. Specifically, the following sender adaptors are not pipeable.

Sender consumers could be made pipeable, but we have chosen to not do so. However, since these are terminal nodes in a pipeline and nothing can be piped after them, we believe a pipe syntax may be confusing as well as unnecessary, as consumers cannot be chained. We believe sender consumers read better with function call syntax.

4.13. A range of senders represents an async sequence of data

Senders represent a single unit of asynchronous work. In many cases though, what is being modeled is a sequence of data arriving asynchronously, and you want computation to happen on demand, when each element arrives. This requires nothing more than what is in this paper and the range support in C++20. A range of senders would allow you to model such input as keystrikes, mouse movements, sensor readings, or network requests.

Given some expression R that is a range of senders, consider the following in a coroutine that returns an async generator type:

for (auto snd : R) {
  if (auto opt = co_await execution::stopped_as_optional(std::move(snd)))
    co_yield fn(*std::move(opt));
  else
    break;
}

This transforms each element of the asynchronous sequence R with the function fn on demand, as the data arrives. The result is a new asynchronous sequence of the transformed values.

Now imagine that R is the simple expression views::iota(0) | views::transform(execution::just). This creates a lazy range of senders, each of which completes immediately with monotonically increasing integers. The above code churns through the range, generating a new infine asynchronous range of values [fn(0), fn(1), fn(2), ...].

Far more interesting would be if R were a range of senders representing, say, user actions in a UI. The above code gives a simple way to respond to user actions on demand.

4.14. Senders can represent partial success

Receivers have three ways they can complete: with success, failure, or cancellation. This begs the question of how they can be used to represent async operations that partially succeed. For example, consider an API that reads from a socket. The connection could drop after the API has filled in some of the buffer. In cases like that, it makes sense to want to report both that the connection dropped and that some data has been successfully read.

Often in the case of partial success, the error condition is not fatal nor does it mean the API has failed to satisfy its post-conditions. It is merely an extra piece of information about the nature of the completion. In those cases, "partial success" is another way of saying "success". As a result, it is sensible to pass both the error code and the result (if any) through the value channel, as shown below:

// Capture a buffer for read_socket_async to fill in
execution::just(array<byte, 1024>{})
  | execution::let_value([socket](array<byte, 1024>& buff) {
      // read_socket_async completes with two values: an error_code and
      // a count of bytes:
      return read_socket_async(socket, span{buff})
          // For success (partial and full), specify the next action:
        | execution::let_value([](error_code err, size_t bytes_read) {
            if (err != 0) {
              // OK, partial success. Decide how to deal with the partial results
            } else {
              // OK, full success here.
            }
          });
    })

In other cases, the partial success is more of a partial failure. That happens when the error condition indicates that in some way the function failed to satisfy its post-conditions. In those cases, sending the error through the value channel loses valuable contextual information. It’s possible that bundling the error and the incomplete results into an object and passing it through the error channel makes more sense. In that way, generic algorithms will not miss the fact that a post-condition has not been met and react inappropriately.

Another possibility is for an async API to return a range of senders: if the API completes with full success, full error, or cancellation, the returned range contains just one sender with the result. Otherwise, if the API partially fails (doesn’t satisfy its post-conditions, but some incomplete result is available), the returned range would have two senders: the first containing the partial result, and the second containing the error. Such an API might be used in a coroutine as follows:

// Declare a buffer for read_socket_async to fill in
array<byte, 1024> buff;

for (auto snd : read_socket_async(socket, span{buff})) {
  try {
    if (optional<size_t> bytes_read =
          co_await execution::stopped_as_optional(std::move(snd))) {
      // OK, we read some bytes into buff. Process them here....
    } else {
      // The socket read was cancelled and returned no data. React
      // appropriately.
    }
  } catch (...) {
    // read_socket_async failed to meet its post-conditions.
    // Do some cleanup and propagate the error...
  }
}

Finally, it’s possible to combine these two approaches when the API can both partially succeed (meeting its post-conditions) and partially fail (not meeting its post-conditions).

4.15. All awaitables are senders

Since C++20 added coroutines to the standard, we expect that coroutines and awaitables will be how a great many will choose to express their asynchronous code. However, in this paper, we are proposing to add a suite of asynchronous algorithms that accept senders, not awaitables. One might wonder whether and how these algorithms will be accessible to those who choose coroutines instead of senders.

In truth there will be no problem because all generally awaitable types automatically model the sender concept. The adaptation is transparent and happens in the sender customization points, which are aware of awaitables. (By "generally awaitable" we mean types that don’t require custom await_transform trickery from a promise type to make them awaitable.)

For an example, imagine a coroutine type called task<T> that knows nothing about senders. It doesn’t implement any of the sender customization points. Despite that fact, and despite the fact that the this_thread::sync_wait algorithm is constrained with the sender concept, the following would compile and do what the user wants:

task<int> doSomeAsyncWork();

int main() {
  // OK, awaitable types satisfy the requirements for senders:
  auto o = this_thread::sync_wait(doSomeAsyncWork());
}

Since awaitables are senders, writing a sender-based asynchronous algorithm is trivial if you have a coroutine task type: implement the algorithm as a coroutine. If you are not bothered by the possibility of allocations and indirections as a result of using coroutines, then there is no need to ever write a sender, a receiver, or an operation state.

4.16. Many senders can be trivially made awaitable

If you choose to implement your sender-based algorithms as coroutines, you’ll run into the issue of how to retrieve results from a passed-in sender. This is not a problem. If the coroutine type opts in to sender support -- trivial with the execution::with_awaitable_senders utility -- then a large class of senders are transparently awaitable from within the coroutine.

For example, consider the following trivial implementation of the sender-based retry algorithm:

template<class S>
  requires single-sender<S&> // see [exec.as.awaitable]
task<single-sender-value-type<S>> retry(S s) {
  for (;;) {
    try {
      co_return co_await s;
    } catch(...) {
    }
  }
}

Only some senders can be made awaitable directly because of the fact that callbacks are more expressive than coroutines. An awaitable expression has a single type: the result value of the async operation. In contrast, a callback can accept multiple arguments as the result of an operation. What’s more, the callback can have overloaded function call signatures that take different sets of arguments. There is no way to automatically map such senders into awaitables. The with_awaitable_senders utility recognizes as awaitables those senders that send a single value of a single type. To await another kind of sender, a user would have to first map its value channel into a single value of a single type -- say, with the into_variant sender algorithm -- before co_await-ing that sender.

4.17. Cancellation of a sender can unwind a stack of coroutines

When looking at the sender-based retry algorithm in the previous section, we can see that the value and error cases are correctly handled. But what about cancellation? What happens to a coroutine that is suspended awaiting a sender that completes by calling execution::set_stopped?

When your task type’s promise inherits from with_awaitable_senders, what happens is this: the coroutine behaves as if an uncatchable exception had been thrown from the co_await expression. (It is not really an exception, but it’s helpful to think of it that way.) Provided that the promise types of the calling coroutines also inherit from with_awaitable_senders, or more generally implement a member function called unhandled_stopped, the exception unwinds the chain of coroutines as if an exception were thrown except that it bypasses catch(...) clauses.

In order to "catch" this uncatchable stopped exception, one of the calling coroutines in the stack would have to await a sender that maps the stopped channel into either a value or an error. That is achievable with the execution::let_stopped, execution::upon_stopped, execution::stopped_as_optional, or execution::stopped_as_error sender adaptors. For instance, we can use execution::stopped_as_optional to "catch" the stopped signal and map it into an empty optional as shown below:

if (auto opt = co_await execution::stopped_as_optional(some_sender)) {
  // OK, some_sender completed successfully, and opt contains the result.
} else {
  // some_sender completed with a cancellation signal.
}

As described in the section "All awaitables are senders", the sender customization points recognize awaitables and adapt them transparently to model the sender concept. When connect-ing an awaitable and a receiver, the adaptation layer awaits the awaitable within a coroutine that implements unhandled_stopped in its promise type. The effect of this is that an "uncatchable" stopped exception propagates seamlessly out of awaitables, causing execution::set_stopped to be called on the receiver.

Obviously, unhandled_stopped is a library extension of the coroutine promise interface. Many promise types will not implement unhandled_stopped. When an uncatchable stopped exception tries to propagate through such a coroutine, it is treated as an unhandled exception and terminate is called. The solution, as described above, is to use a sender adaptor to handle the stopped exception before awaiting it. It goes without saying that any future Standard Library coroutine types ought to implement unhandled_stopped. The author of Add lazy coroutine (coroutine task) type, which proposes a standard coroutine task type, is in agreement.

4.18. Composition with parallel algorithms

The C++ Standard Library provides a large number of algorithms that offer the potential for non-sequential execution via the use of execution policies. The set of algorithms with execution policy overloads are often referred to as "parallel algorithms", although additional policies are available.

Existing policies, such as execution::par, give the implementation permission to execute the algorithm in parallel. However, the choice of execution resources used to perform the work is left to the implementation.

We will propose a customization point for combining schedulers with policies in order to provide control over where work will execute.

template<class ExecutionPolicy>
unspecified executing_on(
    execution::scheduler auto scheduler,
    ExecutionPolicy && policy
);

This function would return an object of an unspecified type which can be used in place of an execution policy as the first argument to one of the parallel algorithms. The overload selected by that object should execute its computation as requested by policy while using scheduler to create any work to be run. The expression may be ill-formed if scheduler is not able to support the given policy.

The existing parallel algorithms are synchronous; all of the effects performed by the computation are complete before the algorithm returns to its caller. This remains unchanged with the executing_on customization point.

In the future, we expect additional papers will propose asynchronous forms of the parallel algorithms which (1) return senders rather than values or void and (2) where a customization point pairing a sender with an execution policy would similarly be used to obtain an object of unspecified type to be provided as the first argument to the algorithm.

4.19. User-facing sender factories

A sender factory is an algorithm that takes no senders as parameters and returns a sender.

4.19.1. execution::schedule

execution::sender auto schedule(
    execution::scheduler auto scheduler
);

Returns a sender describing the start of a task graph on the provided scheduler. See § 4.2 Schedulers represent execution resources.

execution::scheduler auto sch1 = get_system_thread_pool().scheduler();

execution::sender auto snd1 = execution::schedule(sch1);
// snd1 describes the creation of a new task on the system thread pool

4.19.2. execution::just

execution::sender auto just(
    auto ...&& values
);

Returns a sender with no completion schedulers, which sends the provided values. The input values are decay-copied into the returned sender. When the returned sender is connected to a receiver, the values are moved into the operation state if the sender is an rvalue; otherwise, they are copied. Then xvalues referencing the values in the operation state are passed to the receiver’s set_value.

execution::sender auto snd1 = execution::just(3.14);
execution::sender auto then1 = execution::then(snd1, [] (double d) {
  std::cout << d << "\n";
});

execution::sender auto snd2 = execution::just(3.14, 42);
execution::sender auto then2 = execution::then(snd2, [] (double d, int i) {
  std::cout << d << ", " << i << "\n";
});

std::vector v3{1, 2, 3, 4, 5};
execution::sender auto snd3 = execution::just(v3);
execution::sender auto then3 = execution::then(snd3, [] (std::vector<int>&& v3copy) {
  for (auto&& e : v3copy) { e *= 2; }
  return std::move(v3copy);
}
auto&& [v3copy] = this_thread::sync_wait(then3).value();
// v3 contains {1, 2, 3, 4, 5}; v3copy will contain {2, 4, 6, 8, 10}.

execution::sender auto snd4 = execution::just(std::vector{1, 2, 3, 4, 5});
execution::sender auto then4 = execution::then(std::move(snd4), [] (std::vector<int>&& v4) {
  for (auto&& e : v4) { e *= 2; }
  return std::move(v4);
});
auto&& [v4] = this_thread::sync_wait(std::move(then4)).value();
// v4 contains {2, 4, 6, 8, 10}. No vectors were copied in this example.

4.19.3. execution::just_error

execution::sender auto just_error(
    auto && error
);

Returns a sender with no completion schedulers, which completes with the specified error. If the provided error is an lvalue reference, a copy is made inside the returned sender and a non-const lvalue reference to the copy is sent to the receiver’s set_error. If the provided value is an rvalue reference, it is moved into the returned sender and an rvalue reference to it is sent to the receiver’s set_error.

4.19.4. execution::just_stopped

execution::sender auto just_stopped();

Returns a sender with no completion schedulers, which completes immediately by calling the receiver’s set_stopped.

4.19.5. execution::read

execution::sender auto read(auto tag);

execution::sender auto get_scheduler() {
  return read(execution::get_scheduler);
}
execution::sender auto get_delegatee_scheduler() {
  return read(execution::get_delegatee_scheduler);
}
execution::sender auto get_allocator() {
  return read(execution::get_allocator);
}
execution::sender auto get_stop_token() {
  return read(execution::get_stop_token);
}

Returns a sender that reaches into a receiver’s environment and pulls out the current value associated with the customization point denoted by Tag. It then sends the value read back to the receiver through the value channel. For instance, get_scheduler() (with no arguments) is a sender that asks the receiver for the currently suggested scheduler and passes it to the receiver’s set_value completion-signal.

This can be useful when scheduling nested dependent work. The following sender pulls the current schduler into the value channel and then schedules more work onto it.

execution::sender auto task =
  execution::get_scheduler()
    | execution::let_value([](auto sched) {
        return execution::on(sched, some nested work here);
    });

this_thread::sync_wait( std::move(task) ); // wait for it to finish

This code uses the fact that sync_wait associates a scheduler with the receiver that it connects with task. get_scheduler() reads that scheduler out of the receiver, and passes it to let_value's receiver’s set_value function, which in turn passes it to the lambda. That lambda returns a new sender that uses the scheduler to schedule some nested work onto sync_wait's scheduler.

4.20. User-facing sender adaptors

A sender adaptor is an algorithm that takes one or more senders, which it may execution::connect, as parameters, and returns a sender, whose completion is related to the sender arguments it has received.

Sender adaptors are lazy, that is, they are never allowed to submit any work for execution prior to the returned sender being started later on, and are also guaranteed to not start any input senders passed into them. Sender consumers such as § 4.21.1 execution::start_detached and § 4.21.2 this_thread::sync_wait start senders.

For more implementer-centric description of starting senders, see § 5.5 Sender adaptors are lazy.

4.20.1. execution::transfer

execution::sender auto transfer(
    execution::sender auto input,
    execution::scheduler auto scheduler
);

Returns a sender describing the transition from the execution agent of the input sender to the execution agent of the target scheduler. See § 4.6 Execution resource transitions are explicit.

execution::scheduler auto cpu_sched = get_system_thread_pool().scheduler();
execution::scheduler auto gpu_sched = cuda::scheduler();

execution::sender auto cpu_task = execution::schedule(cpu_sched);
// cpu_task describes the creation of a new task on the system thread pool

execution::sender auto gpu_task = execution::transfer(cpu_task, gpu_sched);
// gpu_task describes the transition of the task graph described by cpu_task to the gpu

4.20.2. execution::then

execution::sender auto then(
    execution::sender auto input,
    std::invocable<values-sent-by(input)...> function
);

then returns a sender describing the task graph described by the input sender, with an added node of invoking the provided function with the values sent by the input sender as arguments.

then is guaranteed to not begin executing function until the returned sender is started.

execution::sender auto input = get_input();
execution::sender auto snd = execution::then(input, [](auto... args) {
    std::print(args...);
});
// snd describes the work described by pred
// followed by printing all of the values sent by pred

This adaptor is included as it is necessary for writing any sender code that actually performs a useful function.

4.20.3. execution::upon_*

execution::sender auto upon_error(
    execution::sender auto input,
    std::invocable<errors-sent-by(input)...> function
);

execution::sender auto upon_stopped(
    execution::sender auto input,
    std::invocable auto function
);

upon_error and upon_stopped are similar to then, but where then works with values sent by the input sender, upon_error works with errors, and upon_stopped is invoked when the "stopped" signal is sent.

4.20.4. execution::let_*

execution::sender auto let_value(
    execution::sender auto input,
    std::invocable<values-sent-by(input)...> function
);

execution::sender auto let_error(
    execution::sender auto input,
    std::invocable<errors-sent-by(input)...> function
);

execution::sender auto let_stopped(
    execution::sender auto input,
    std::invocable auto function
);

let_value is very similar to then: when it is started, it invokes the provided function with the values sent by the input sender as arguments. However, where the sender returned from then sends exactly what that function ends up returning - let_value requires that the function return a sender, and the sender returned by let_value sends the values sent by the sender returned from the callback. This is similar to the notion of "future unwrapping" in future/promise-based frameworks.

let_value is guaranteed to not begin executing function until the returned sender is started.

let_error and let_stopped are similar to let_value, but where let_value works with values sent by the input sender, let_error works with errors, and let_stopped is invoked when the "stopped" signal is sent.

4.20.5. execution::on

execution::sender auto on(
    execution::scheduler auto sched,
    execution::sender auto snd
);

Returns a sender which, when started, will start the provided sender on an execution agent belonging to the execution resource associated with the provided scheduler. This returned sender has no completion schedulers.

4.20.6. execution::into_variant

execution::sender auto into_variant(
    execution::sender auto snd
);

Returns a sender which sends a variant of tuples of all the possible sets of types sent by the input sender. Senders can send multiple sets of values depending on runtime conditions; this is a helper function that turns them into a single variant value.

4.20.7. execution::stopped_as_optional

execution::sender auto stopped_as_optional(
    single-sender auto snd
);

Returns a sender that maps the value channel from a T to an optional<decay_t<T>>, and maps the stopped channel to a value of an empty optional<decay_t<T>>.

4.20.8. execution::stopped_as_error

template<move_constructible Error>
execution::sender auto stopped_as_error(
    execution::sender auto snd,
    Error err
);

Returns a sender that maps the stopped channel to an error of err.

4.20.9. execution::bulk

execution::sender auto bulk(
    execution::sender auto input,
    std::integral auto shape,
    invocable<decltype(size), values-sent-by(input)...> function
);

Returns a sender describing the task of invoking the provided function with every index in the provided shape along with the values sent by the input sender. The returned sender completes once all invocations have completed, or an error has occurred. If it completes by sending values, they are equivalent to those sent by the input sender.

No instance of function will begin executing until the returned sender is started. Each invocation of function runs in an execution agent whose forward progress guarantees are determined by the scheduler on which they are run. All agents created by a single use of bulk execute with the same guarantee. The number of execution agents used by bulk is not specified. This allows a scheduler to execute some invocations of the function in parallel.

In this proposal, only integral types are used to specify the shape of the bulk section. We expect that future papers may wish to explore extensions of the interface to explore additional kinds of shapes, such as multi-dimensional grids, that are commonly used for parallel computing tasks.

4.20.10. execution::split

execution::sender auto split(execution::sender auto sender);

If the provided sender is a multi-shot sender, returns that sender. Otherwise, returns a multi-shot sender which sends values equivalent to the values sent by the provided sender. See § 4.7 Senders can be either multi-shot or single-shot.

4.20.11. execution::when_all

execution::sender auto when_all(
    execution::sender auto ...inputs
);

execution::sender auto when_all_with_variant(
    execution::sender auto ...inputs
);

when_all returns a sender that completes once all of the input senders have completed. It is constrained to only accept senders that can complete with a single set of values (_i.e._, it only calls one overload of set_value on its receiver). The values sent by this sender are the values sent by each of the input senders, in order of the arguments passed to when_all. It completes inline on the execution resource on which the last input sender completes, unless stop is requested before when_all is started, in which case it completes inline within the call to start.

when_all_with_variant does the same, but it adapts all the input senders using into_variant, and so it does not constrain the input arguments as when_all does.

The returned sender has no completion schedulers.

execution::scheduler auto sched = thread_pool.scheduler();

execution::sender auto sends_1 = ...;
execution::sender auto sends_abc = ...;

execution::sender auto both = execution::when_all(sched,
    sends_1,
    sends_abc
);

execution::sender auto final = execution::then(both, [](auto... args){
    std::cout << std::format("the two args: {}, {}", args...);
});
// when final executes, it will print "the two args: 1, abc"

4.20.12. execution::ensure_started

execution::sender auto ensure_started(
    execution::sender auto sender
);

Once ensure_started returns, it is known that the provided sender has been connected and start has been called on the resulting operation state (see § 5.2 Operation states represent work); in other words, the work described by the provided sender has been submitted for execution on the appropriate execution resources. Returns a sender which completes when the provided sender completes and sends values equivalent to those of the provided sender.

If the returned sender is destroyed before execution::connect() is called, or if execution::connect() is called but the returned operation-state is destroyed before execution::start() is called, then a stop-request is sent to the eagerly launched operation and the operation is detached and will run to completion in the background. Its result will be discarded when it eventually completes.

Note that the application will need to make sure that resources are kept alive in the case that the operation detaches. e.g. by holding a std::shared_ptr to those resources or otherwise having some out-of-band way to signal completion of the operation so that resource release can be sequenced after the completion.

4.21. User-facing sender consumers

A sender consumer is an algorithm that takes one or more senders, which it may execution::connect, as parameters, and does not return a sender.

4.21.1. execution::start_detached

void start_detached(
    execution::sender auto sender
);

Like ensure_started, but does not return a value; if the provided sender sends an error instead of a value, std::terminate is called.

4.21.2. this_thread::sync_wait

auto sync_wait(
    execution::sender auto sender
) requires (always-sends-same-values(sender))
    -> std::optional<std::tuple<values-sent-by(sender)>>;

this_thread::sync_wait is a sender consumer that submits the work described by the provided sender for execution, similarly to ensure_started, except that it blocks the current std::thread or thread of main until the work is completed, and returns an optional tuple of values that were sent by the provided sender on its completion of work. Where § 4.19.1 execution::schedule and § 4.19.2 execution::just are meant to enter the domain of senders, sync_wait is meant to exit the domain of senders, retrieving the result of the task graph.

If the provided sender sends an error instead of values, sync_wait throws that error as an exception, or rethrows the original exception if the error is of type std::exception_ptr.

If the provided sender sends the "stopped" signal instead of values, sync_wait returns an empty optional.

For an explanation of the requires clause, see § 5.8 All senders are typed. That clause also explains another sender consumer, built on top of sync_wait: sync_wait_with_variant.

Note: This function is specified inside std::this_thread, and not inside execution. This is because sync_wait has to block the current execution agent, but determining what the current execution agent is is not reliable. Since the standard does not specify any functions on the current execution agent other than those in std::this_thread, this is the flavor of this function that is being proposed. If C++ ever obtains fibers, for instance, we expect that a variant of this function called std::this_fiber::sync_wait would be provided. We also expect that runtimes with execution agents that use different synchronization mechanisms than std::thread's will provide their own flavors of sync_wait as well (assuming their execution agents have the means to block in a non-deadlock manner).

4.22. execution::execute

In addition to the three categories of functions presented above, we also propose to include a convenience function for fire-and-forget eager one-way submission of an invocable to a scheduler, to fulfil the role of one-way executors from P0443.

void execution::execute(
    execution::schedule auto sched,
    std::invocable auto fn
);

Submits the provided function for execution on the provided scheduler, as-if by:

auto snd = execution::schedule(sched);
auto work = execution::then(snd, fn);
execution::start_detached(work);

5. Design - implementer side

5.1. Receivers serve as glue between senders

A receiver is a callback that supports more than one channel. In fact, it supports three of them:

Once an async operation has been started exactly one of these functions must be invoked on a receiver before it is destroyed.

While the receiver interface may look novel, it is in fact very similar to the interface of std::promise, which provides the first two signals as set_value and set_exception, and it’s possible to emulate the third channel with lifetime management of the promise.

Receivers are not a part of the end-user-facing API of this proposal; they are necessary to allow unrelated senders communicate with each other, but the only users who will interact with receivers directly are authors of senders.

Receivers are what is passed as the second argument to § 5.3 execution::connect.

5.2. Operation states represent work

An operation state is an object that represents work. Unlike senders, it is not a chaining mechanism; instead, it is a concrete object that packages the work described by a full sender chain, ready to be executed. An operation state is neither movable nor copyable, and its interface consists of a single algorithm: start, which serves as the submission point of the work represented by a given operation state.

Operation states are not a part of the user-facing API of this proposal; they are necessary for implementing sender consumers like execution::ensure_started and this_thread::sync_wait, and the knowledge of them is necessary to implement senders, so the only users who will interact with operation states directly are authors of senders and authors of sender algorithms.

The return value of § 5.3 execution::connect must satisfy the operation state concept.

5.3. execution::connect

execution::connect is a customization point which connects senders with receivers, resulting in an operation state that will ensure that if start is called that one of the completion operations will be called on the receiver passed to connect.

execution::sender auto snd = some input sender;
execution::receiver auto rcv = some receiver;
execution::operation_state auto state = execution::connect(snd, rcv);

execution::start(state);
// at this point, it is guaranteed that the work represented by state has been submitted
// to an execution resource, and that execution resource will eventually call one of the
// completion operations on rcv

// operation states are not movable, and therefore this operation state object must be
// kept alive until the operation finishes

5.4. Sender algorithms are customizable

Senders being able to advertise what their completion schedulers are fulfills one of the promises of senders: that of being able to customize an implementation of a sender algorithm based on what scheduler any work it depends on will complete on.

The simple way to provide customizations for functions like then, that is for sender adaptors and sender consumers, is to follow the customization scheme that has been adopted for C++20 ranges library; to do that, we would define the expression execution::then(sender, invocable) to be equivalent to:

  1. sender.then(invocable), if that expression is well-formed; otherwise

  2. then(sender, invocable), performed in a context where this call always performs ADL, if that expression is well-formed; otherwise

  3. a default implementation of then, which returns a sender adaptor, and then define the exact semantics of said adaptor.

However, this definition is problematic. Imagine another sender adaptor, bulk, which is a structured abstraction for a loop over an index space. Its default implementation is just a for loop. However, for accelerator runtimes like CUDA, we would like sender algorithms like bulk to have specialized behavior, which invokes a kernel of more than one thread (with its size defined by the call to bulk); therefore, we would like to customize bulk for CUDA senders to achieve this. However, there’s no reason for CUDA kernels to necessarily customize the then sender adaptor, as the generic implementation is perfectly sufficient. This creates a problem, though; consider the following snippet:

execution::scheduler auto cuda_sch = cuda_scheduler{};

execution::sender auto initial = execution::schedule(cuda_sch);
// the type of initial is a type defined by the cuda_scheduler
// let’s call it cuda::schedule_sender<>

execution::sender auto next = execution::then(cuda_sch, []{ return 1; });
// the type of next is a standard-library unspecified sender adaptor
// that wraps the cuda sender
// let’s call it execution::then_sender_adaptor<cuda::schedule_sender<>>

execution::sender auto kernel_sender = execution::bulk(next, shape, [](int i){ ... });

How can we specialize the bulk sender adaptor for our wrapped schedule_sender? Well, here’s one possible approach, taking advantage of ADL (and the fact that the definition of "associated namespace" also recursively enumerates the associated namespaces of all template parameters of a type):

namespace cuda::for_adl_purposes {
template<typename... SentValues>
class schedule_sender {
    execution::operation_state auto connect(execution::receiver auto rcv);
    execution::scheduler auto get_completion_scheduler() const;
};

execution::sender auto bulk(
    execution::sender auto && input,
    execution::shape auto && shape,
    invocable%lt;sender-values(input)> auto && fn)
{
    // return a cuda sender representing a bulk kernel launch
}
} // namespace cuda::for_adl_purposes

However, if the input sender is not just a then_sender_adaptor like in the example above, but another sender that overrides bulk by itself, as a member function, because its author believes they know an optimization for bulk - the specialization above will no longer be selected, because a member function of the first argument is a better match than the ADL-found overload.

This means that well-meant specialization of sender algorithms that are entirely scheduler-agnostic can have negative consequences. The scheduler-specific specialization - which is essential for good performance on platforms providing specialized ways to launch certain sender algorithms - would not be selected in such cases. But it’s really the scheduler that should control the behavior of sender algorithms when a non-default implementation exists, not the sender. Senders merely describe work; schedulers, however, are the handle to the runtime that will eventually execute said work, and should thus have the final say in how the work is going to be executed.

Therefore, we are proposing the following customization scheme: the expression execution::<sender-algorithm>(sender, args...), for any given sender algorithm that accepts a sender as its first argument, should do the following:

  1. Create a sender that implements the default implementation of the sender algorithm. That sender is tuple-like; it can be destructured into its constituent parts: algorithm tag, data, and child sender(s).

  2. We query the child sender for its domain. A domain is a tag type associated with the scheduler that the child sender will complete on. If there are multiple child senders, we query all of them for their domains and require that they all be the same.

  3. We use the domain to dispatch to a transform_sender customization, which accepts the sender and optionally performs a domain-specific transformation on it. This customization is expected to return a new sender, which will be returned from <sender-algorithm> in place of the original sender.

5.5. Sender adaptors are lazy

Contrary to early revisions of this paper, we propose to make all sender adaptors perform strictly lazy submission, unless specified otherwise (the one notable exception in this paper is § 4.20.12 execution::ensure_started, whose sole purpose is to start an input sender).

Strictly lazy submission means that there is a guarantee that no work is submitted to an execution resource before a receiver is connected to a sender, and execution::start is called on the resulting operation state.

5.6. Lazy senders provide optimization opportunities

Because lazy senders fundamentally describe work, instead of describing or representing the submission of said work to an execution resource, and thanks to the flexibility of the customization of most sender algorithms, they provide an opportunity for fusing multiple algorithms in a sender chain together, into a single function that can later be submitted for execution by an execution resource. There are two ways this can happen.

The first (and most common) way for such optimizations to happen is thanks to the structure of the implementation: because all the work is done within callbacks invoked on the completion of an earlier sender, recursively up to the original source of computation, the compiler is able to see a chain of work described using senders as a tree of tail calls, allowing for inlining and removal of most of the sender machinery. In fact, when work is not submitted to execution resources outside of the current thread of execution, compilers are capable of removing the senders abstraction entirely, while still allowing for composition of functions across different parts of a program.

The second way for this to occur is when a sender algorithm is specialized for a specific set of arguments. For instance, we expect that, for senders which are known to have been started already, § 4.20.12 execution::ensure_started will be an identity transformation, because the sender algorithm will be specialized for such senders. Similarly, an implementation could recognize two subsequent § 4.20.9 execution::bulks of compatible shapes, and merge them together into a single submission of a GPU kernel.

5.7. Execution resource transitions are two-step

Because execution::transfer takes a sender as its first argument, it is not actually directly customizable by the target scheduler. This is by design: the target scheduler may not know how to transition from a scheduler such as a CUDA scheduler; transitioning away from a GPU in an efficient manner requires making runtime calls that are specific to the GPU in question, and the same is usually true for other kinds of accelerators too (or for scheduler running on remote systems). To avoid this problem, specialized schedulers like the ones mentioned here can still hook into the transition mechanism, and inject a sender which will perform a transition to the regular CPU execution resource, so that any sender can be attached to it.

This, however, is a problem: because customization of sender algorithms must be controlled by the scheduler they will run on (see § 5.4 Sender algorithms are customizable), the type of the sender returned from transfer must be controllable by the target scheduler. Besides, the target scheduler may itself represent a specialized execution resource, which requires additional work to be performed to transition to it. GPUs and remote node schedulers are once again good examples of such schedulers: executing code on their execution resources requires making runtime API calls for work submission, and quite possibly for the data movement of the values being sent by the input sender passed into transfer.

To allow for such customization from both ends, we propose the inclusion of a secondary transitioning sender adaptor, called schedule_from. This adaptor is a form of schedule, but takes an additional, second argument: the input sender. This adaptor is not meant to be invoked manually by the end users; they are always supposed to invoke transfer, to ensure that both schedulers have a say in how the transitions are made. Any scheduler that specializes transfer(snd, sch) shall ensure that the return value of their customization is equivalent to schedule_from(sch, snd2), where snd2 is a successor of snd that sends values equivalent to those sent by snd.

The default implementation of transfer(snd, sched) is schedule_from(sched, snd).

5.8. All senders are typed

All senders must advertise the types they will send when they complete. There are many sender adaptors that need this information. Even just transitioning from one execution context to another requires temporarily storing the async result data so it can be propagated in the new execution context. Doing that efficiently requires knowing the type of the data.

The mechanism a sender uses to advertise its completions is the get_completion_signatures customization point, which takes an environment and must return a specialization of the execution::completion_signatures class template. The template parameters of execution::completion_signatures is a list of function types that represent the completion operations of the sender. for example, the type execution::set_value_t(size_t, const char*) indicates that the sender can complete successfully by passing a size_t and a const char* to the receiver’s set_value function.

This proposal includes utilities for parsing and manipulating the list of a sender’s completion signatures. For instance, values_of_t is a template alias for accessing a sender’s value completions. It takes a sender, an environment, and two variadic template template parameters: a tuple-like template and a variant-like template. You can get the value completions of S and Env with value_types_of_t<S, Env, tuple-like, variant-like>. For example, for a sender that can complete successfully with either Ts... or Us..., value_types_of_t<S, Env, std::tuple, std::variant> would name the type std::variant<std::tuple<Ts...>, std::tuple<Us...>>.

5.9. Customization points

Earlier versions of this paper used a dispatching technique known as tag_invoke (see tag_invoke: A general pattern for supporting customisable functions) to allow for customization of basis operations and sender algorithms. This technique used private friend functions named "tag_invoke" that are found by argument-dependent look-up. The tag_invoke overloads are distinguished from each other by their first argument, which is the type of the customization point object being customized. For instance, to customize the execution::set_value operation, a receiver type might do the following:

struct my_receiver {
  friend void tag_invoke(execution::set_value_t, my_receiver&& self, int value) noexcept {
    std::cout << "received value: " << value;
  }
  //...
};

The tag_invoke technique, although it had its strengths, has been replaced with a new (or rather, a very old) technique that uses explicit concept opt-ins and named member functions. For instance, the execution::set_value operation is now customized by defining a member function named set_value in the receiver type. This technique is more explicit and easier to understand than tag_invoke. This is what a receiver author would do to customize execution::set_value now:

struct my_receiver {
  using receiver_concept = execution::receiver_t;

  void set_value(int value) && noexcept {
    std::cout << "received value: " << value;
  }
  //...
};

The only exception to this is the customization of queries. There is a need to build queryable adaptors that can forward and open and unknowable set of queries to some wrapped object. This is done by defining a member function named query in the adaptor type that takes the query CPO object as its first (and usually only) argument. A queryable adaptor might look like this:

template <class Query, class Queryable, class... Args>
concept query_for =
  execution::queryable<Queryable> &&
  requires (const Queryable& o, Args&&... args) {
    o.query(Query(), (Args&&) args...);
  };

template<class Allocator = std::allocator<>,
         execution::queryable Base = execution::empty_env>
struct with_allocator {
  Allocator alloc{};
  Base base{};

  // Forward unknown queries to the wrapped object:
  template<query_for<Base> Query>
  decltype(auto) query(Query q) const {
    return base.query(q);
  }

  // Specialize the query for the allocator:
  Allocator query(execution::get_allocator_t) const {
    return alloc;
  }
};

Customization of sender algorithms such as execution::then and execution::bulk are handled differently because they must dispatch based on where the sender is executing. See the section on § 5.4 Sender algorithms are customizable for more information.

6. Specification

Much of this wording follows the wording of A Unified Executors Proposal for C++.

§ 22 General utilities library [utilities] is meant to be a diff relative to the wording of the [utilities] clause of Working Draft, Standard for Programming Language C++.

§ 33 Concurrency support library [thread] is meant to be a diff relative to the wording of the [thread] clause of Working Draft, Standard for Programming Language C++. This diff applies changes from Composable cancellation for sender-based async operations.

§ 34 Execution control library [exec] is meant to be added as a new library clause to the working draft of C++.

7.

8.

9.

10.

11.

12.

13.

14. Exception handling [except]

14.6. Special functions [except.special]

14.6.2. The std::terminate function [except.terminate]

At the end of the bulleted list in the Note in paragraph 1, add a new bullet as follows:

  • when a call to a wait(), wait_until(), or wait_for() function on a condition variable (33.7.4, 33.7.5) fails to meet a postcondition.

  • when a callback invocation exits via an exception when requesting stop on a std::stop_source or a std::inplace_stop_source ([stopsource.mem], [stopsource.inplace.mem]), or in the constructor of std::stop_callback or std::inplace_stop_callback ([stopcallback.cons], [stopcallback.inplace.cons]) when a callback invocation exits via an exception.

16. Library introduction [library]

At the end of [expos.only.entity], add the following:

  1. The following are defined for exposition only to aid in the specification of the library:

    namespace std {
      // ...as before...
    }
    
  1. An object dst is said to be decay-copied from a subexpression src if the type of dst is decay_t<decltype((src))>, and dst is copy-initialized from src.

22. General utilities library [utilities]

22.10. Function objects [function.objects]

22.10.2. Header <functional> synopsis [functional.syn]

At the end of this subclause, insert the following declarations into the synopsis within namespace std:

namespace std {
  // ...as before...

  namespace ranges {
    // 22.10.9, concept-constrained comparisons
    struct equal_to;                                    // freestanding
    struct not_equal_to;                                // freestanding
    struct greater;                                     // freestanding
    struct less;                                        // freestanding
    struct greater_equal;                               // freestanding
    struct less_equal;                                  // freestanding
  }

template<class Fn, class... Args> concept callable = // exposition only requires (Fn&& fn, Args&&... args) { std::forward<Fn>(fn)(std::forward<Args>(args)...); }; template<class Fn, class... Args> concept nothrow-callable = // exposition only callable<Fn, Args...> && requires (Fn&& fn, Args&&... args) { { std::forward<Fn>(fn)(std::forward<Args>(args)...) } noexcept; }; // exposition only: template<class Fn, class... Args> using call-result-t = decltype(declval<Fn>()(declval<Args>()...)); template<const auto& Tag> using decayed-typeof = decltype(auto(Tag)); // exposition only
}

33. Concurrency support library [thread]

33.3. Stop tokens [thread.stoptoken]

33.3.1. Introduction [thread.stoptoken.intro]

  1. Subclause [thread.stoptoken] describes components that can be used to asynchronously request that an operation stops execution in a timely manner, typically because the result is no longer required. Such a request is called a stop request.

  2. stop_source, stop_token, and stop_callback implement stoppable-source, stoppable_token, and stoppable-callback-for are concepts that specify the required syntax and semantics of shared ownership access of a stop state. Any stop_source, stop_token, or stop_callback object that shares ownership of the same stop state is an associated stop_source, stop_token, or stop_callback, respectively. Any object modeling stoppable-source, stoppable_token, or stoppable-callback-for that refers to the same stop state is an associated stoppable-source, stoppable_token, or stoppable-callback-for, respectively. The last remaining owner of the stop state automatically releases the resources associated with the stop state.
  3. A stoppable_token can be passed to an operation which that can either

    • actively poll the token to check if there has been a stop request, or

    • register a callback using the stop_callback class template which that will be called in the event that a stop request is made.

    A stop request made via a stop_source an object that models stoppable-source will be visible to all associated stoppable_token and stop_source stoppable-source objects. Once a stop request has been made it cannot be withdrawn (a subsequent stop request has no effect).

  4. Callbacks registered via a stop_callback object an object that models stoppable-callback-for are called when a stop request is first made by any associated stop_source stoppable-source object.

The following paragraph is moved to the specification of the new stoppable-source concept.

  1. Calls to the functions request_stop, stop_requested, and stop_possible do not introduce data races. A call to request_stop that returns true synchronizes with a call to stop_requested on an associated stop_token or stop_source object that returns true. Registration of a callback synchronizes with the invocation of that callback.

  1. The types stop_source and stop_token and the class template stop_callback implement the semantics of shared ownership of a stop state. The last remaining owner of the stop state automatically releases the resources associated with the stop state.

  2. The types inplace_stop_source and inplace_stop_token and the class template inplace_stop_callback do no dynamic memory allocation or reference counting of the stop state. They are for use when the lifetimes of the tokens and the callbacks are known to nest within the lifetime of the source.

33.3.2. Header <stop_token> synopsis [thread.stoptoken.syn]

In this subclause, insert the following declarations into the <stop_token> synopsis:

namespace std {
// [stoptoken.concepts], stop token concepts template<class Callback, class Token, class Initializer = Callback> concept stoppable-callback-for = see below; // exposition only template<class Token> concept stoppable_token = see below; template<class Token> concept unstoppable_token = see below; template<class Source> concept stoppable-source = see below; // exposition only
// 33.3.3, class stop_token class stop_token; // 33.3.4, class stop_source class stop_source; // no-shared-stop-state indicator struct nostopstate_t { explicit nostopstate_t() = default; }; inline constexpr nostopstate_t nostopstate{}; // 33.3.5, class template stop_callback template<class Callback> class stop_callback;
// [stoptoken.never], class never_stop_token class never_stop_token; // [stoptoken.inplace], class inplace_stop_token class inplace_stop_token; // [stopsource.inplace], class inplace_stop_source class inplace_stop_source; // [stopcallback.inplace], class template inplace_stop_callback template<class Callback> class inplace_stop_callback; template<class T, class Callback> using stop_callback_for_t = T::template callback_type<Callback>;
}

Insert the following subclause as a new subclause between Header <stop_token> synopsis [thread.stoptoken.syn] and Class stop_token [stoptoken].

33.3.3. Stop token concepts [stoptoken.concepts]

  1. For a stop token type Token and a type Callback that is callable with no arguments, if the type stop_callback_for_t<Token, Callback> is valid, it denotes the type of a stop callback to use to register a callback to be executed if a stop request is ever made on the stoppable_token's associated stop source. The exposition-only stoppable-callback-for concept checks for a callback compatible with a given stop token type.

    template<class Callback, class Token, class Initializer = Callback>
      concept stoppable-callback-for = // exposition only
        invocable<Callback> &&
        constructible_from<Callback, Initializer> &&
        requires { typename stop_callback_for_t<Token, Callback>; } &&
        constructible_from<stop_callback_for_t<Token, Callback>, const Token&, Initializer>;
    
  2. Let t and u be distinct, valid objects of type Token that reference the same logical stop state; let init be an object of type Initializer; and let CB denote the type stop_callback_for_t<Token, Callback>.

  3. The concept stoppable-callback-for<Callback, Token, Initializer> is modeled only if:

    1. The following concepts are modeled:

      • constructible_from<CB, Token, Initializer>

      • constructible_from<CB, Token&, Initializer>

      • constructible_from<CB, const Token, Initializer>

    2. An object of type CB has an associated callback function of type Callback. Let cb be an object of type CB and let callback denote cb's associated callback function. Direct-non-list-initializing cb from arguments t and init shall execute a stoppable callback registration as follows:

      1. If t.stop_possible() is true:

        1. callback shall be direct-initialized with init.

        2. Construction of cb shall only throw exceptions thrown by the initialization of callback from init.

        3. The callback invocation std::forward<Callback>(callback)() shall be registered with t's associated stop state as follows:

          1. If t.stop_requested() evaluates to false at the time of registration, the callback invocation is added to the stop state’s list of callbacks such that std::forward<Callback>(callback)() is evaluated if a stop request is made on the stop state.

          2. Otherwise, std::forward<Callback>(callback)() shall be immediately evaluated on the thread executing cb's constructor, and the callback invocation shall not be added to the list of callback invocations.

        4. If the callback invocation was added to stop state’s list of callbacks, cb shall be associated with the stop state.

      2. If t.stop_possible() is false, there is no requirement that the initialization of cb causes the initialization of callback.

    3. An evaluation of u.stop_requested() that strongly happens after ([intro.races]/10) the beginning of the invocation of callback shall return true.

    4. Destruction of cb shall execute a stoppable callback deregistration as follows:

      1. If the constructor of cb did not register a callback invocation with t's stop state, then the stoppable callback deregistration shall have no effect.

      2. Otherwise, the invocation of callback shall be removed from the associated stop state.

      3. If callback is currently being invoked on another thread then the stoppable callback deregistration shall block ([defns.block]) until the invocation of callback returns such that the return from the invocation of callback strongly happens before ([intro.races]) the destruction of callback.

      4. If callback is executing on the current thread, then the destructor shall not block waiting for the return from the invocation of callback.

      5. A stoppable callback deregistration shall not block on the completion of the invocation of some other callback registered with the same logical stop state.

      6. As a final step, the stoppable callback deregistration shall destroy the callback function.

  4. The stoppable_token concept checks for the basic interface of a stop token that is copyable and allows polling to see if stop has been requested and also whether a stop request is possible. The unstoppable_token concept checks for a stop token type that does not allow stopping.

    template<template<class> class>
      struct check-type-alias-exists; // exposition-only
    
    template<class Token>
      concept stoppable_token =
        requires (const Token tok) {
          typename check-type-alias-exists<Token::template callback_type>;
          { tok.stop_requested() } noexcept -> same_as<bool>;
          { tok.stop_possible() } noexcept -> same_as<bool>;
          { Token(tok) } noexcept; // see implicit expression variations
                                   // ([concepts.equality])
        } &&
        copyable<Token> &&
        equality_comparable<Token> &&
        swappable;
    
    template<class Token>
      concept unstoppable_token =
        stoppable_token<Token> &&
        requires (const Token tok) {
          requires bool_constant<(!tok.stop_possible())>::value;
        };
    
  5. The type Token models stoppable_token only if:

    1. If an evaluation, E, of t.stop_possible() evaluates to false, then evaluations of u.stop_possible() and u.stop_requested() that happen after E shall evaluate to false.

    2. If an evaluation, E, of t.stop_requested() evaluates to true, then evaluations of u.stop_possible() and u.stop_requested() that happens after E shall evaluate to true.

    3. For any types Callback and Initializer, if stoppable-callback-for<Callback, Token, Initializer> is satisfied, then stoppable-callback-for<Callback, Token, Initializer> shall be modeled.

    4. An object that models stoppable_token has at most one associated logical stop state. A stoppable_token object with no associated stop state is said to be disengaged. For a disengaged stoppable_token object, stop_possible and stop_requested shall return false. If t and u reference the same stop state, or if both t and u are disengaged, t == u shall be true; otherwise, it shall be false.

  6. A model of the exposition-only stoppable-source concept can be queried whether stop has been requested (stop_requested) and whether stop is possible (stop_possible). It is a factory for associated stop tokens (get_token) and can be explicitly placed into the "stop requested" state (request_stop). It maintains a list of registered stop callback invocations that it executes when a stop request is first made.

    template<class Source>
      concept stoppable-source = // exposition only
        requires (Source& src, const Source csrc) { // see implicit expression variations
                                                    // ([concepts.equality])
          { csrc.get_token() } -> stoppable_token;
          { csrc.stop_possible() } noexcept -> same_as<bool>;
          { csrc.stop_requested() } noexcept -> same_as<bool>;
          { src.request_stop() } -> same_as<bool>;
        };
    
    1. A stoppable-source object has at most one associated logical stop state. A stoppable-source object with no associated stop state is said to be disengaged. For a disengaged stoppable-source object, stop_possible and stop_requested shall return false.

    2. A disengaged stoppable-source object shall return a disengaged stop token from get_token(). Otherwise, get_token() shall return a stop token that is associated with the stop state of the source.

    The following paragraph is moved from the introduction, with minor modifications (underlined in green).

    1. Calls to the member functions request_stop, stop_requested, and stop_possible and similarly named member functions on associated stoppable_token objects do not introduce data races. A call to request_stop that returns true synchronizes with a call to stop_requested on an associated stoppable_token or stop_source stoppable-source object that returns true. Registration of a callback synchronizes with the invocation of that callback.

    The following paragraph is taken from § 33.3.5.3 Member functions [stopsource.mem] and modified.

    1. If the stoppable-source is disengaged, request_stop shall have no effect and return false. Otherwise, it shall execute a stop request operation on the associated stop state. A stop request operation determines whether the stop state has received a stop request, and if not, makes a stop request. The determination and making of the stop request shall happen atomically, as-if by a read-modify-write operation ([intro.races]). If the request was made, the stop state’s registered callback invocations shall be synchronously executed. If an invocation of a callback exits via an exception then terminate shall be invoked ([except.terminate]). No constraint is placed on the order in which the callback invocations are executed. request_stop shall return true if a stop request was made, and false otherwise. After a call to request_stop either stop_possible() shall be false or stop_requested() shall be true.

      A stop request includes notifying all condition variables of type condition_variable_any temporarily registered during an interruptible wait ([thread.condvarany.intwait]).

Modify subclause [stoptoken] as follows:

33.3.4. Class stop_token [stoptoken]

33.3.4.1. General [stoptoken.general]
  1. The class stop_token provides an interface for querying whether a stop request has been made (stop_requested) or can ever be made (stop_possible) using an associated stop_source object ([stopsource]). A stop_token can also be passed to a stop_callback ([stopcallback]) constructor to register a callback to be called when a stop request has been made from an associated stop_source. The class stop_token models the concept stoppable_token. It shares ownership of its stop state, if any, with its associated stop_source object ([stopsource]) and any stop_token objects to which it compares equal.
namespace std {
  class stop_token {
  public:
template<class T> using callback_type = stop_callback<T>;
// [stoptoken.cons], constructors, copy, and assignment stop_token() noexcept = default;
stop_token(const stop_token&) noexcept; stop_token(stop_token&&) noexcept; stop_token& operator=(const stop_token&) noexcept; stop_token& operator=(stop_token&&) noexcept; ~stop_token();
// [stoptoken.mem], Member functions void swap(stop_token&) noexcept; // [stoptoken.mem], stop handling [[nodiscard]] bool stop_requested() const noexcept; [[nodiscard]] bool stop_possible() const noexcept; bool operator==(const stop_token& rhs) const noexcept = default; [[nodiscard]] friend bool operator==(const stop_token& lhs, const stop_token& rhs) noexcept; friend void swap(stop_token& lhs, stop_token& rhs) noexcept; private: shared_ptr<unspecified> stop-state{}; // exposition only }; }
  1. stop-state refers to the stop_token's associated stop state. A stop_token object is disengaged when stop-state is null.

33.3.4.2. Constructors, copy, and assignment [stoptoken.cons]
stop_token() noexcept;
  1. Postconditions: stop_possible() is false and stop_requested() is false. Because the created stop_token object can never receive a stop request, no resources are allocated for a stop state.

stop_token(const stop_token& rhs) noexcept;
  1. Postconditions: *this == rhs is true. *this and rhs share the ownership of the same stop state, if any.

stop_token(stop_token&& rhs) noexcept;
  1. Postconditions: *this contains the value of rhs prior to the start of construction and rhs.stop_possible() is false.

~stop_token();
  1. Effects: Releases ownership of the stop state, if any.

stop_token& operator=(const stop_token& rhs) noexcept;
  1. Effects: Equivalent to: stop_token(rhs).swap(*this).

  2. Returns: *this.

stop_token& operator=(stop_token&& rhs) noexcept;
  1. Effects: Equivalent to: stop_token(std::move(rhs)).swap(*this).

  2. Returns: *this.

Move swap into [stoptoken.mem]:

33.3.4.3. Member functions [stoptoken.mem]
void swap(stop_token& rhs) noexcept;
  1. Effects: Exchanges the values of *this and rhs. Equivalent to: std::swap(stop-state, rhs.stop-state).

[[nodiscard]] bool stop_requested() const noexcept;
  1. Returns: true if *this has ownership of stop-state refers to a stop state that has received a stop request; otherwise, false.

[[nodiscard]] bool stop_possible() const noexcept;
  1. Returns: false if:

    • *this does not have ownership of a stop state is disengaged , or

    • a stop request was not made and there are no associated stop_source objects; otherwise, true.

The following are covered by the equality_comparable and swappable concepts.

33.3.4.4. Non-member functions [stoptoken.nonmembers]
[[nodiscard]] bool operator==(const stop_token& lhs, const stop_token& rhs) noexcept;
  1. Returns: true if lhs and rhs have ownership of the same stop state or if both lhs and rhs do not have ownership of a stop state; otherwise false.

friend void swap(stop_token& x, stop_token& y) noexcept;
  1. Effects: Equivalent to: x.swap(y).

33.3.5. Class stop_source [stopsource]

33.3.5.1. General [stopsource.general]
  1. The class stop_source implements the semantics of making a stop request. A stop request made on a stop_source object is visible to all associated stop_source and stop_token ([thread.stoptoken]) objects. Once a stop request has been made it cannot be withdrawn (a subsequent stop request has no effect) models stoppable-source .

namespace std {
  The following definitions are already specified in the <stop_token> synopsis:
// no-shared-stop-state indicator struct nostopstate_t { explicit nostopstate_t() = default; }; inline constexpr nostopstate_t nostopstate{};
class stop_source { public: // 33.3.4.2, constructors, copy, and assignment stop_source(); explicit stop_source(nostopstate_t) noexcept; {}
stop_source(const stop_source&) noexcept; stop_source(stop_source&&) noexcept; stop_source& operator=(const stop_source&) noexcept; stop_source& operator=(stop_source&&) noexcept; ~stop_source();
// [stopsource.mem], Member functions void swap(stop_source&) noexcept; // 33.3.4.3, stop handling [[nodiscard]] stop_token get_token() const noexcept; [[nodiscard]] bool stop_possible() const noexcept; [[nodiscard]] bool stop_requested() const noexcept; bool request_stop() noexcept; bool operator==(const stop_source& rhs) const noexcept = default;
[[nodiscard]] friend bool operator==(const stop_source& lhs, const stop_source& rhs) noexcept; friend void swap(stop_source& lhs, stop_source& rhs) noexcept;
private: shared_ptr<unspecified> stop-state{}; // exposition only }; }
  1. stop-state refers to the stop_source's associated stop state. A stop_source object is disengaged when stop-state is null.

  2. stop_source models copyable, equality_comparable, and swappable.

33.3.5.2. Constructors, copy, and assignment [stopsource.cons]
stop_source();
  1. Effects: Initialises *this to have ownership of stop-state with a pointer to a new stop state.

  2. Postconditions: stop_possible() is true and stop_requested() is false.

  3. Throws: bad_alloc if memory cannot be allocated for the stop state.

explicit stop_source(nostopstate_t) noexcept;
  1. Postconditions: stop_possible() is false and stop_requested() is false. No resources are allocated for the state.

stop_source(const stop_source& rhs) noexcept;
  1. Postconditions: *this == rhs is true. *this and rhs share the ownership of the same stop state, if any.

stop_source(stop_source&& rhs) noexcept;
  1. Postconditions: *this contains the value of rhs prior to the start of construction and rhs.stop_possible() is false.

~stop_source();
  1. Effects: Releases ownership of the stop state, if any.

stop_source& operator=(const stop_source& rhs) noexcept;
  1. Effects: Equivalent to: stop_source(rhs).swap(*this).

  2. Returns: *this.

stop_source& operator=(stop_source&& rhs) noexcept;
  1. Effects: Equivalent to: stop_source(std::move(rhs)).swap(*this).

  2. Returns: *this.

Move swap into [stopsource.mem]:

33.3.5.3. Member functions [stopsource.mem]
void swap(stop_source& rhs) noexcept;
  1. Effects: Exchanges the values of *this and rhs Equivalent to: std::swap(stop-state, rhs.stop-state) .

[[nodiscard]] stop_token get_token() const noexcept;
  1. Returns: stop_token() if stop_possible() is false; otherwise a new associated stop_token object ; i.e., its stop-state pointer is equal to the stop-state pointer of *this .

[[nodiscard]] bool stop_possible() const noexcept;
  1. Returns: true if *this has ownership of a stop state; otherwise, false stop-state != nullptr .

[[nodiscard]] bool stop_requested() const noexcept;
  1. Returns: true if *this has ownership of stop-state refers to a stop state that has received a stop request; otherwise, false.

bool request_stop() noexcept;
  1. Effects: Executes a stop request operation ([stoptoken.concepts]) on the associated stop state, if any.

  1. Effects: If *this does not have ownership of a stop state, returns false. Otherwise, atomically determines whether the owned stop state has received a stop request, and if not, makes a stop request. The determination and making of the stop request are an atomic read-modify-write operation ([intro.races]). If the request was made, the callbacks registered by associated stop_callback objects are synchronously called. If an invocation of a callback exits via an exception then terminate is invoked ([except.terminate]).

    A stop request includes notifying all condition variables of type condition_variable_any temporarily registered during an interruptible wait ([thread.condvarany.intwait]).

  2. Postconditions: stop_possible() is false or stop_requested() is true.

  3. Returns: true if this call made a stop request; otherwise false.

33.3.5.4. Non-member functions [stopsource.nonmembers]
[[nodiscard]] friend bool
  operator==(const stop_source& lhs, const stop_source& rhs) noexcept;
  1. Returns: true if lhs and rhs have ownership of the same stop state or if both lhs and rhs do not have ownership of a stop state; otherwise false.

friend void swap(stop_source& x, stop_source& y) noexcept;
  1. Effects: Equivalent to: x.swap(y).

33.3.6. Class template stop_callback [stopcallback]

33.3.6.1. General [stopcallback.general]
namespace std {
  template<class Callback>
  class stop_callback {
  public:
    using callback_type = Callback;

    // 33.3.5.2, constructors and destructor
    template<class C>
      explicit stop_callback(const stop_token& st, C&& cb)
        noexcept(is_nothrow_constructible_v<Callback, C>);
    template<class C>
      explicit stop_callback(stop_token&& st, C&& cb)
        noexcept(is_nothrow_constructible_v<Callback, C>);
    ~stop_callback();

    stop_callback(const stop_callback&) = delete;
    stop_callback(stop_callback&&) = delete;
    stop_callback& operator=(const stop_callback&) = delete;
    stop_callback& operator=(stop_callback&&) = delete;

  private:
    Callback callback; // exposition only
  };

  template<class Callback>
    stop_callback(stop_token, Callback) -> stop_callback<Callback>;
}
  1. Mandates: stop_callback is instantiated with an argument for the template parameter Callback that satisfies both invocable and destructible.

  1. Preconditions: stop_callback is instantiated with an argument for the template parameter Callback that models both invocable and destructible.

  1. Remarks: For a type C, if stoppable-callback-for<Callback, stop_token, C> is satisfied, then stoppable-callback-for<Callback, stop_token, C> is modeled. The exposition-only callback member is the associated callback function ([stoptoken.concepts]) of stop_callback<Callback> objects.

33.3.6.2. Constructors and destructor [stopcallback.cons]
template<class C>
explicit stop_callback(const stop_token& st, C&& cb)
  noexcept(is_nothrow_constructible_v<Callback, C>);
template<class C>
explicit stop_callback(stop_token&& st, C&& cb)
  noexcept(is_nothrow_constructible_v<Callback, C>);
  1. Constraints: Callback and C satisfy constructible_from<Callback, C>.

  1. Preconditions: Callback and C model constructible_from<Callback, C>.

  1. Effects: Initializes callback with std::forward<C>(cb) and executes a stoppable callback registration ([stoptoken.concepts]) . If st.stop_requested() is true, then std::forward<Callback>(callback)() is evaluated in the current thread before the constructor returns. Otherwise, if st has ownership of a stop state, acquires shared ownership of that stop state and registers the callback with that stop state such that std::forward<Callback>(callback)() is evaluated by the first call to request_stop() on an associated stop_source. If a callback is registered with st's shared stop state, then *this acquires shared ownership of that stop state.

  1. Throws: Any exception thrown by the initialization of callback.

  2. Remarks: If evaluating std::forward<Callback>(callback)() exits via an exception, then terminate is invoked ([except.terminate]).

~stop_callback();
  1. Effects: Unregisters the callback from the owned stop state, if any. The destructor does not block waiting for the execution of another callback registered by an associated stop_callback. If callback is concurrently executing on another thread, then the return from the invocation of callback strongly happens before ([intro.races]) callback is destroyed. If callback is executing on the current thread, then the destructor does not block ([defns.block]) waiting for the return from the invocation of callback. Releases Executes a stoppable callback deregistration ([stoptoken.concepts]), and releases ownership of the stop state, if any.

Insert a new subclause, Class never_stop_token [stoptoken.never], after subclause Class template stop_callback [stopcallback], as a new subclause of Stop tokens [thread.stoptoken].

33.3.7. Class never_stop_token [stoptoken.never]

33.3.7.1. General [stoptoken.never.general]
  1. The class never_stop_token models the unstoppable_token concept. It provides a stop token interface, but also provides static information that a stop is never possible nor requested.

    namespace std {
      class never_stop_token {
        struct callback { // exposition only
          explicit callback(never_stop_token, auto&&) noexcept {}
        };
      public:
        template<class>
          using callback_type = callback;
    
        static constexpr bool stop_requested() noexcept { return false; }
        static constexpr bool stop_possible() noexcept { return false; }
    
        bool operator==(const never_stop_token&) const noexcept = default;
      };
    }
    

Insert a new subclause, Class inplace_stop_token [stoptoken.inplace], after the subclause added above, as a new subclause of Stop tokens [thread.stoptoken].

33.3.8. Class inplace_stop_token [stoptoken.inplace]

33.3.8.1. General [stoptoken.inplace.general]
  1. The class inplace_stop_token models the concept stoppable_token. It references the stop state of its associated inplace_stop_source object ([stopsource.inplace]), if any.

    namespace std {
      class inplace_stop_token {
      public:
        template<class CB>
          using callback_type = inplace_stop_callback<CB>;
    
        inplace_stop_token() noexcept = default;
        bool operator==(const inplace_stop_token&) const noexcept = default;
    
        // [stoptoken.inplace.mem], member functions
        bool stop_requested() const noexcept;
        bool stop_possible() const noexcept;
        void swap(inplace_stop_token&) noexcept;
    
      private:
        const inplace_stop_source* stop-source = nullptr; // exposition only
      };
    }
    
33.3.8.2. Member functions [stoptoken.inplace.members]
void swap(inplace_stop_token& rhs) noexcept;
  1. Effects: Exchanges the values of stop-source and rhs.stop-source.

bool stop_requested() const noexcept;
  1. Effects: Equivalent to: return stop-source != nullptr && stop-source->stop_requested();

  2. As specified in [basic.life], the behavior of stop_requested() is undefined unless the call strongly happens before the start of the destructor of the associated inplace_stop_source, if any.

bool stop_possible() const noexcept;
  1. Returns: stop-source != nullptr.

  2. As specified in [basic.stc.general], the behavior of stop_possible() is implementation-defined unless the call strongly happens before the end of the storage duration of the associated inplace_stop_source object, if any.

Insert a new subclause, Class inplace_stop_source [stopsource.inplace], after the subclause added above, as a new subclause of Stop tokens [thread.stoptoken].

33.3.9. Class inplace_stop_source [stopsource.inplace]

33.3.9.1. General [stopsource.inplace.general]
  1. The class inplace_stop_source models stoppable-source. Unlike stop_source, inplace_stop_source does not require dynamic allocation or reference counting of a shared stop state. Instead, it requires that all uses of associated inplace_stop_token and inplace_stop_callback objects happen before the inplace_stop_source is destroyed.

    namespace std {
      class inplace_stop_source {
      public:
        // [stopsource.inplace.cons], constructors, copy, and assignment
        inplace_stop_source() noexcept;
    
        inplace_stop_source(inplace_stop_source&&) = delete;
        inplace_stop_source(const inplace_stop_source&) = delete;
        inplace_stop_source& operator=(inplace_stop_source&&) = delete;
        inplace_stop_source& operator=(const inplace_stop_source&) = delete;
        ~inplace_stop_source();
    
        //[stopsource.inplace.mem], stop handling
        inplace_stop_token get_token() const noexcept;
        static constexpr bool stop_possible() noexcept { return true; }
        bool stop_requested() const noexcept;
        bool request_stop() noexcept;
      };
    }
    
33.3.9.2. Constructors, copy, and assignment [stopsource.inplace.cons]
inplace_stop_source() noexcept;
  1. Effects: Initializes a new stop state inside *this.

  2. Postconditions: stop_requested() is false.

33.3.9.3. Members [stopsource.inplace.mem]
inplace_stop_token get_token() const noexcept;
  1. Returns: A new associated inplace_stop_token object.

bool stop_requested() const noexcept;
  1. Returns: true if the stop state inside *this has received a stop request; otherwise, false.

bool request_stop() noexcept;
  1. Effects: Executes a stop request operation ([stoptoken.concepts]).

  2. Postconditions: stop_requested() is true.

Insert a new subclause, Class template inplace_stop_callback [stopcallback.inplace], after the subclause added above, as a new subclause of Stop tokens [thread.stoptoken].

33.3.10. Class template inplace_stop_callback [stopcallback.inplace]

33.3.10.1. General [stopcallback.inplace.general]
  1. namespace std {
      template<class Callback>
      class inplace_stop_callback {
      public:
        using callback_type = Callback;
    
        // [stopcallback.inplace.cons], constructors and destructor
        template<class C>
          explicit inplace_stop_callback(inplace_stop_token st, C&& cb)
            noexcept(is_nothrow_constructible_v<Callback, C>);
        ~inplace_stop_callback();
    
        inplace_stop_callback(inplace_stop_callback&&) = delete;
        inplace_stop_callback(const inplace_stop_callback&) = delete;
        inplace_stop_callback& operator=(inplace_stop_callback&&) = delete;
        inplace_stop_callback& operator=(const inplace_stop_callback&) = delete;
    
      private:
        Callback stop-callback;      // exposition only
      };
    
      template<class Callback>
        inplace_stop_callback(inplace_stop_token, Callback)
          -> inplace_stop_callback<Callback>;
    }
    
  2. Mandates: inplace_stop_callback is instantiated with an argument for the template parameter Callback that satisfies both invocable and destructible.

  3. Remarks: For a type C, if stoppable-callback-for<Callback, inplace_stop_token, C> is satisfied, then stoppable-callback-for<Callback, inplace_stop_token, C> is modeled. The exposition-only stop-callback member is the associated callback function ([stoptoken.concepts]) of inplace_stop_callback<Callback> objects.

    Implementations are not permitted to use additional storage, such as dynamic memory, to store the state necessary for an inplace_stop_callback's association with an inplace_stop_source object or to register the callback invocation with the associated inplace_stop_source object.

33.3.10.2. Constructors and destructor [stopcallback.inplace.cons]
template<class C>
  explicit inplace_stop_callback(inplace_stop_token st, C&& cb)
    noexcept(is_nothrow_constructible_v<Callback, C>);
  1. Constraints: constructible_from<Callback, C> is satisfied.

  2. Effects: Initializes stop-callback with std::forward<C>(cb) and executes a stoppable callback registration ([stoptoken.concepts]).

~inplace_stop_callback();
  1. Effects: Executes a stoppable callback deregistration ([stoptoken.concepts]).

Insert a new top-level clause

34. Execution control library [exec]

34.1. General [exec.general]

  1. This Clause describes components supporting execution of function objects [function.objects].

  2. The following subclauses describe the requirements, concepts, and components for execution control primitives as summarized in Table 1.

Table N: Execution control library summary [tab:execution.summary]
Subclause Header
[exec.sched] Schedulers <execution>
[exec.recv] Receivers
[exec.opstate] Operation states
[exec.snd] Senders
[exec.execute] One-way execution
  1. [Note: A large number of execution control primitives are customization point objects. For an object one might define multiple types of customization point objects, for which different rules apply. Table 2 shows the types of customization point objects used in the execution control library:

Table N+1: Types of customization point objects in the execution control library [tab:execution.cpos]
Customization point object type Purpose Examples
core provide core execution functionality, and connection between core components connect, start, execute
completion functions called by senders to announce the completion of the work (success, error, or cancellation) set_value, set_error, set_stopped
senders allow the specialization of the provided sender algorithms
  • sender factories (schedule, just, read, ...)
  • sender adaptors (transfer, then, let_value, ...)
  • sender consumers (start_detached, sync_wait)
queries allow querying different properties of objects
  • general queries (get_allocator, get_stop_token, ...)
  • environment queries (get_scheduler, get_delegatee_scheduler, ...)
  • scheduler queries (get_forward_progress_guarantee, execute_may_block_caller, ...)
  • sender attribute queries (get_completion_scheduler)

-- end note]

  1. This clause makes use of the following exposition-only entities:

    1. For a subexpression expr, let MANDATE-NOTHROW(expr) be expression-equivalent to expr.

      • Mandates: noexcept(expr) is true.

    2. namespace std {
        template<class T>
          concept movable-value =
            move_constructible<decay_t<T>> &&
            constructible_from<decay_t<T>, T> &&
            (!is_array_v<remove_cvref_t<T>>);
      }
      
    3. For function types F1 and F2 denoting R1(Args1...) and R2(Args2...) respectively, MATCHING-SIG(F1, F2) is true if and only if same_as<R1(Args&&...), R2(Args2&&...)> is true.

    4. For a subexpression err, let Err be decltype((err)) and let AS-EXCEPT-PTR(err) be:

      1. err if decay_t<Err> denotes the type exception_ptr.

        • Mandates: err != exception_ptr() is true

      2. Otherwise, make_exception_ptr(system_error(err)) if decay_t<Err> denotes the type error_code,

      3. Otherwise, make_exception_ptr(err).

34.2. Queries and queryables [exec.queryable]

34.2.1. General [exec.queryable.general]

  1. A queryable object is a read-only collection of key/value pairs where each key is a customization point object known as a query object. A query is an invocation of a query object with a queryable object as its first argument and a (possibly empty) set of additional arguments. The result of a query expression is valid as long as the queryable object is valid. A query imposes syntactic and semantic requirements on its invocations.

  2. Given a subexpression env that refers to a queryable object o, a query object q, and a (possibly empty) pack of subexpressions args, the expression q(env, args...) is equal to ([concepts.equality]) the expression q(c, args...) where c is a const lvalue reference to o.

  3. The type of a query expression can not be void.

  4. The expression q(env, args...) is equality-preserving ([concepts.equality]) and does not modify the function object or the arguments.

  5. If the expression env.query(q, args...) is well-formed, then it is expression-equivalent to q(env, args...).

  6. Unless otherwise specified, the value returned by the expression q(env, args...) is valid as long as env is valid.

34.2.2. queryable concept [exec.queryable.concept]

namespace std {
  template<class T>
    concept queryable = destructible<T>;
}
  1. The queryable concept specifies the constraints on the types of queryable objects.

  2. Let env be an object of type Env. The type Env models queryable if for each callable object q and a pack of subexpressions args, if requires { q(env, args...) } is true then q(env, args...) meets any semantic requirements imposed by q.

34.3. Asynchronous operations [async.ops]

  1. An execution resource is a program entity that manages a (possibly dynamic) set of execution agents ([thread.req.lockable.general]), which it uses to execute parallel work on behalf of callers. [Example 1: The currently active thread, a system-provided thread pool, and uses of an API associated with an external hardware accelerator are all examples of execution resources. -- end example] Execution resources execute asynchronous operations. An execution resource is either valid or invalid.

  2. An asynchronous operation is a distinct unit of program execution that:

    • is explicitly created;

    • can be explicitly started; an asynchronous operation can be started once at most;

    • if started, eventually completes with a (possibly empty) set of result datums, and in exactly one of three modes: success, failure, or cancellation, known as the operation’s disposition; an asychronous operation can only complete once; a successful completion, also known as a value completion, can have an arbitrary number of result datums; a failure completion, also known as an error completion, has a single result datum; a cancellation completion, also known as a stopped completion, has no result datum; an asynchronous operation’s async result is its disposition and its (possibly empty) set of result datums.

    • can complete on a different execution resource than that on which it started; and

    • can create and start other asychronous operations called child operations. A child operation is an asynchronous operation that is created by the parent operation and, if started, completes before the parent operation completes. A parent operation is the asynchronous operation that created a particular child operation.

    An asynchronous operation can in fact execute synchronously; that is, it can complete during the execution of its start operation on the thread of execution that started it.

  3. An asynchronous operation has associated state known as its operation state.

  4. An asynchronous operation has an associated environment. An environment is a queryable object ([exec.queryable]) representing the execution-time properties of the operation’s caller. The caller of an asynchronous operation is its parent operation or the function that created it. An asynchronous operation’s operation state owns the operation’s environment.

  5. An asynchronous operation has an associated receiver. A receiver is an aggregation of three handlers for the three asynchronous completion dispositions: a value completion handler for a value completion, an error completion handler for an error completion, and a stopped completion handler for a stopped completion. A receiver has an associated environment. An asynchronous operation’s operation state owns the operation’s receiver. The environment of an asynchronous operation is equal to its receiver’s environment.

  6. For each completion disposition, there is a completion function. A completion function is a customization point object ([customization.point.object]) that accepts an asynchronous operation’s receiver as the first argument and the result datums of the asynchronous operation as additional arguments. The value completion function invokes the receiver’s value completion handler with the value result datums; likewise for the error completion function and the stopped completion function. A completion function has an associated type known as its completion tag that names the unqualified type of the completion function. A valid invocation of a completion function is called a completion operation.

  7. The lifetime of an asynchronous operation, also known as the operation’s async lifetime, begins when its start operation begins executing and ends when its completion operation begins executing. If the lifetime of an asynchronous operation’s associated operation state ends before the lifetime of the asynchronous operation, the behavior is undefined. After an asynchronous operation executes a completion operation, its associated operation state is invalid. Accessing any part of an invalid operation state is undefined behavior.

  8. An asynchronous operation shall not execute a completion operation before its start operation has begun executing. After its start operation has begun executing, exactly one completion operation shall execute. The lifetime of an asynchronous operation’s operation state can end during the execution of the completion operation.

  9. A sender is a factory for one or more asynchronous operations. Connecting a sender and a receiver creates an asynchronous operation. The asynchronous operation’s associated receiver is equal to the receiver used to create it, and its associated environment is equal to the environment associated with the receiver used to create it. The lifetime of an asynchronous operation’s associated operation state does not depend on the lifetimes of either the sender or the receiver from which it was created. A sender sends its results by way of the asynchronous operation(s) it produces, and a receiver receives those results. A sender is either valid or invalid; it becomes invalid when its parent sender (see below) becomes invalid.

  10. A scheduler is an abstraction of an execution resource with a uniform, generic interface for scheduling work onto that resource. It is a factory for senders whose asynchronous operations execute value completion operations on an execution agent belonging to the scheduler’s associated execution resource. A schedule-expression obtains such a sender from a scheduler. A schedule sender is the result of a schedule expression. On success, an asynchronous operation produced by a schedule sender executes a value completion operation with an empty set of result datums. Multiple schedulers can refer to the same execution resource. A scheduler can be valid or invalid. A scheduler becomes invalid when the execution resource to which it refers becomes invalid, as do any schedule senders obtained from the scheduler, and any operation states obtained from those senders.

  11. An asynchronous operation has one or more associated completion schedulers for each of its possible dispositions. A completion scheduler is a scheduler whose associated execution resource is used to execute a completion operation for an asynchronous operation. A value completion scheduler is a scheduler on which an asynchronous operation’s value completion operation can execute. Likewise for error completion schedulers and stopped completion schedulers.

  12. A sender has an associated queryable object ([exec.queryable]) known as its attributes that describes various characteristics of the sender and of the asynchronous operation(s) it produces. For each disposition, there is a query object for reading the associated completion scheduler from a sender’s attributes; i.e., a value completion scheduler query object for reading a sender’s value completion scheduler, etc. If a completion scheduler query is well-formed, the returned completion scheduler is unique for that disposition for any asynchronous operation the sender creates. A schedule sender is required to have a value completion scheduler attribute whose value is equal to the scheduler that produced the schedule sender.

  13. A completion signature is a function type that describes a completion operation. An asychronous operation has a finite set of possible completion signatures corresponding to the completion operations that the asynchronous operation potentially evaluates ([basic.def.odr]). For a completion function set, receiver rcvr, and pack of arguments args, let c be the completion operation set(rcvr, args...), and let F be the function type decltype(auto(set))(decltype((args))...). A completion signature Sig is associated with c if and only if MATCHING-SIG(Sig, F) is true ([exec.general]). Together, a sender type and an environment type Env determine the set of completion signatures of an asynchronous operation that results from connecting the sender with a receiver that has an environment of type Env. The type of the receiver does not affect an asychronous operation’s completion signatures, only the type of the receiver’s environment.

  14. A sender algorithm is a function that takes and/or returns a sender. There are three categories of sender algorithms:

    • A sender factory is a function that takes non-senders as arguments and that returns a sender.

    • A sender adaptor is a function that constructs and returns a parent sender from a set of one or more child senders and a (possibly empty) set of additional arguments. An asynchronous operation created by a parent sender is a parent operation to the child operations created by the child senders.

    • A sender consumer is a function that takes one or more senders and a (possibly empty) set of additional arguments, and whose return type is not the type of a sender.

34.4. Header <execution> synopsis [exec.syn]

namespace std {
  // [exec.general], helper concepts
  template<class T>
    concept movable-value = see below; // exposition only

  template<class From, class To>
    concept decays-to = same_as<decay_t<From>, To>; // exposition only

  template<class T>
    concept class-type = decays-to<T, T> && is_class_v<T>;  // exposition only

  // [exec.queryable], queryable objects
  template<class T>
    concept queryable = destructible<T>;

  // [exec.queries], queries
  struct forwarding_query_t;
  struct get_allocator_t;
  struct get_stop_token_t;

  inline constexpr forwarding_query_t forwarding_query{};
  inline constexpr get_allocator_t get_allocator{};
  inline constexpr get_stop_token_t get_stop_token{};

  template<class T>
    using stop_token_of_t =
      remove_cvref_t<decltype(get_stop_token(declval<T>()))>;

  template<class T>
    concept forwarding-query = // exposition only
      forwarding_query(T{});
}

namespace std::execution {
  // [exec.queries], queries
  enum class forward_progress_guarantee;
  struct get_domain_t;
  struct get_scheduler_t;
  struct get_delegatee_scheduler_t;
  struct get_forward_progress_guarantee_t;
  template<class CPO>
    struct get_completion_scheduler_t;

  inline constexpr get_domain_t get_domain{};
  inline constexpr get_scheduler_t get_scheduler{};
  inline constexpr get_delegatee_scheduler_t get_delegatee_scheduler{};
  inline constexpr get_forward_progress_guarantee_t get_forward_progress_guarantee{};
  template<class CPO>
    inline constexpr get_completion_scheduler_t<CPO> get_completion_scheduler{};

  struct empty_env {};
  struct get_env_t;
  inline constexpr get_env_t get_env {};

  template<class T>
    using env_of_t = decltype(get_env(declval<T>()));

  // [exec.domain.default], execution domains
  struct default_domain;

  // [exec.sched], schedulers
  struct scheduler_t {};

  template<class Sch>
    concept scheduler = see below;

  // [exec.recv], receivers
  struct receiver_t {};

  template<class Rcvr>
    concept receiver = see below;

  template<class Rcvr, class Completions>
    concept receiver_of = see below;

  struct set_value_t;
  struct set_error_t;
  struct set_stopped_t;

  inline constexpr set_value_t set_value{};
  inline constexpr set_error_t set_error{};
  inline constexpr set_stopped_t set_stopped{};

  // [exec.opstate], operation states
  struct operation_state_t {};

  template<class O>
    concept operation_state = see below;

  struct start_t;
  inline constexpr start_t start{};

  // [exec.snd], senders
  struct sender_t {};

  template<class Sndr>
    concept sender = see below;

  template<class Sndr, class Env = empty_env>
    concept sender_in = see below;

  template<class Sndr, class Rcvr>
    concept sender_to = see below;

  template<class... Ts>
    struct type-list; // exposition only

  template<class Sndr, class Env = empty_env>
    using single-sender-value-type = see below; // exposition only

  template<class Sndr, class Env = empty_env>
    concept single-sender = see below; // exposition only

  // [exec.getcomplsigs], completion signatures
  struct get_completion_signatures_t;
  inline constexpr get_completion_signatures_t get_completion_signatures {};

  template<class Sndr, class Env = empty_env>
      requires sender_in<Sndr, Env>
    using completion_signatures_of_t = call-result-t<get_completion_signatures_t, Sndr, Env>;

  template<class... Ts>
    using decayed-tuple = tuple<decay_t<Ts>...>; // exposition only

  template<class... Ts>
    using variant-or-empty = see below; // exposition only

  template<class Sndr,
           class Env = empty_env,
           template<class...> class Tuple = decayed-tuple,
           template<class...> class Variant = variant-or-empty>
      requires sender_in<Sndr, Env>
    using value_types_of_t = see below;

  template<class Sndr,
           class Env = empty_env,
           template<class...> class Variant = variant-or-empty>
      requires sender_in<Sndr, Env>
    using error_types_of_t = see below;

  template<class Sndr, class Env = empty_env>
      requires sender_in<Sndr, Env>
    inline constexpr bool sends_stopped = see below;

  template <sender Sndr>
    using tag_of_t = see below;

  // [exec.snd.transform], sender transformations
  template<class Domain, sender Sndr, queryable... Env>
      requires (sizeof...(Env) <= 1)
    constexpr sender decltype(auto) transform_sender(
      Domain dom, Sndr&& sndr, const Env&... env) noexcept(see below);

  // [exec.snd.transform.env], environment transformations
  template<class Domain, sender Sndr, queryable Env>
    constexpr queryable decltype(auto) transform_env(
      Domain dom, Sndr&& sndr, Env&& env) noexcept;

  // [exec.snd.apply], sender algorithm application
  template<class Domain, class Tag, sender Sndr, class... Args>
    constexpr decltype(auto) apply_sender(
      Domain dom, Tag, Sndr&& sndr, Args&&... args) noexcept(see below);

  // [exec.connect], the connect sender algorithm
  struct connect_t;
  inline constexpr connect_t connect{};

  template<class Sndr, class Rcvr>
    using connect_result_t =
      decltype(connect(declval<Sndr>(), declval<Rcvr>()));

  // [exec.factories], sender factories
  struct just_t;
  struct just_error_t;
  struct just_stopped_t;
  struct schedule_t;

  inline constexpr just_t just{};
  inline constexpr just_error_t just_error{};
  inline constexpr just_stopped_t just_stopped{};
  inline constexpr schedule_t schedule{};
  inline constexpr unspecified read{};

  template<scheduler Sndr>
    using schedule_result_t = decltype(schedule(declval<Sndr>()));

  // [exec.adapt], sender adaptors
  template<class-type D>
    struct sender_adaptor_closure { };

  struct on_t;
  struct transfer_t;
  struct schedule_from_t;
  struct then_t;
  struct upon_error_t;
  struct upon_stopped_t;
  struct let_value_t;
  struct let_error_t;
  struct let_stopped_t;
  struct bulk_t;
  struct split_t;
  struct ensure_started_t;
  struct when_all_t;
  struct when_all_with_variant_t;
  struct into_variant_t;
  struct stopped_as_optional_t;
  struct stopped_as_error_t;

  inline constexpr on_t on{};
  inline constexpr transfer_t transfer{};
  inline constexpr schedule_from_t schedule_from{};
  inline constexpr then_t then{};
  inline constexpr upon_error_t upon_error{};
  inline constexpr upon_stopped_t upon_stopped{};
  inline constexpr let_value_t let_value{};
  inline constexpr let_error_t let_error{};
  inline constexpr let_stopped_t let_stopped{};
  inline constexpr bulk_t bulk{};
  inline constexpr split_t split{};
  inline constexpr ensure_started_t ensure_started{};
  inline constexpr when_all_t when_all{};
  inline constexpr when_all_with_variant_t when_all_with_variant{};
  inline constexpr into_variant_t into_variant{};
  inline constexpr stopped_as_optional_t stopped_as_optional;
  inline constexpr stopped_as_error_t stopped_as_error;

  // [exec.consumers], sender consumers
  struct start_detached_t;
  inline constexpr start_detached_t start_detached{};

  // [exec.utils], sender and receiver utilities
  // [exec.utils.cmplsigs]
  template<class Fn>
    concept completion-signature = // exposition only
      see below;

  template<completion-signature... Fns>
    struct completion_signatures {};

  template<class Sigs> // exposition only
    concept valid-completion-signatures = see below;

  // [exec.utils.tfxcmplsigs]
  template<
    valid-completion-signatures InputSignatures,
    valid-completion-signatures AdditionalSignatures = completion_signatures<>,
    template<class...> class SetValue = see below,
    template<class> class SetError = see below,
    valid-completion-signatures SetStopped = completion_signatures<set_stopped_t()>>
  using transform_completion_signatures = completion_signatures<see below>;

  template<
    sender Sndr,
    class Env = empty_env,
    valid-completion-signatures AdditionalSignatures = completion_signatures<>,
    template<class...> class SetValue = see below,
    template<class> class SetError = see below,
    valid-completion-signatures SetStopped = completion_signatures<set_stopped_t()>>
      requires sender_in<Sndr, Env>
  using transform_completion_signatures_of =
    transform_completion_signatures<
      completion_signatures_of_t<Sndr, Env>,
      AdditionalSignatures, SetValue, SetError, SetStopped>;

  // [exec.ctx], execution resources
  // [exec.run.loop], run_loop
  class run_loop;
}

namespace std::this_thread {
  // [exec.queries], queries
  struct execute_may_block_caller_t;
  inline constexpr execute_may_block_caller_t execute_may_block_caller{};

  struct sync_wait_t;
  struct sync_wait_with_variant_t;

  inline constexpr sync_wait_t sync_wait{};
  inline constexpr sync_wait_with_variant_t sync_wait_with_variant{};
}

namespace std::execution {
  // [exec.execute], one-way execution
  struct execute_t;
  inline constexpr execute_t execute{};

  // [exec.as.awaitable]
  struct as_awaitable_t;
  inline constexpr as_awaitable_t as_awaitable;

  // [exec.with.awaitable.senders]
  template<class-type Promise>
    struct with_awaitable_senders;
}
  1. The exposition-only type variant-or-empty<Ts...> is defined as follows:

    1. If sizeof...(Ts) is greater than zero, variant-or-empty<Ts...> names the type variant<Us...> where Us... is the pack decay_t<Ts>... with duplicate types removed.

    2. Otherwise, variant-or-empty<Ts...> names the exposition-only class type:

      namespace std::execution {
        struct empty-variant {
          empty-variant() = delete;
        };
      }
      

34.5. Queries [exec.queries]

34.5.1. std::forwarding_query [exec.fwd.env]

  1. forwarding_query asks a query object whether it should be forwarded through queryable adaptors.

  2. The name forwarding_query denotes a query object. For some query object q of type Q, forwarding_query(q) is expression-equivalent to:

    1. MANDATE-NOTHROW(q.query(forwarding_query)) if that expression is well-formed.

      • Mandates: The expression above has type bool and is a core constant expressions if q is a core constant expression.

    2. Otherwise, true if derived_from<Q, forwarding_query_t> is true.

    3. Otherwise, false.

34.5.2. std::get_allocator [exec.get.allocator]

  1. get_allocator asks an object for its associated allocator.

  2. The name get_allocator denotes a query object. For a subexpression env, get_allocator(env) is expression-equivalent to MANDATE-NOTHROW(as_const(env).query(get_allocator)).

    • Mandates: If the expression above is well-formed, its type satisfies Allocator.

  3. forwarding_query(get_allocator) is a core constant expression and has value true.

  4. get_allocator() (with no arguments) is expression-equivalent to execution::read(get_allocator) ([exec.read]).

34.5.3. std::get_stop_token [exec.get.stop.token]

  1. get_stop_token asks an object for an associated stop token.

  2. The name get_stop_token denotes a query object. For a subexpression env, get_stop_token(env) is expression-equivalent to:

    1. MANDATE-NOTHROW(as_const(env).query(get_stop_token)) if that expression is well-formed.

      • Mandates: The type of the expression above satisfies stoppable_token.

    2. Otherwise, never_stop_token{}.

  3. forwarding_query(get_stop_token) is a core constant expression and has value true.

  4. get_stop_token() (with no arguments) is expression-equivalent to execution::read(get_stop_token) ([exec.read]).

34.5.4. execution::get_env [exec.get.env]

  1. execution::get_env is a customization point object. For a subexpression o, execution::get_env(o) is expression-equivalent to:

    1. as_const(o).get_env() if that expression is well-formed.

      • Mandates: The expression above is not potentially throwing, and its type satisfies queryable ([exec.queryable]).

    2. Otherwise, empty_env{}.

  2. The value of get_env(o) shall be valid while o is valid.

  3. When passed a sender object, get_env returns the sender’s attributes. When passed a receiver, get_env returns the receiver’s environment.

34.5.5. execution::get_domain [exec.get.domain]

  1. get_domain asks an object for an associated execution domain tag.

  2. The name get_domain denotes a query object. For a subexpression env, get_domain(env) is expression-equivalent to MANDATE-NOTHROW(as_const(env).query(get_domain)).

  3. forwarding_query(execution::get_domain) is a core constant expression and has value true.

  4. get_domain() (with no arguments) is expression-equivalent to execution::read(get_domain) ([exec.read]).

34.5.6. execution::get_scheduler [exec.get.scheduler]

  1. get_scheduler asks an object for its associated scheduler.

  2. The name get_scheduler denotes a query object. For a subexpression env, get_scheduler(env) is expression-equivalent to MANDATE-NOTHROW(as_const(env).query(get_scheduler)).

    • Mandates: If the expression above is well-formed, its type satisfies scheduler.

  3. forwarding_query(execution::get_scheduler) is a core constant expression and has value true.

  4. get_scheduler() (with no arguments) is expression-equivalent to execution::read(get_scheduler) ([exec.read]).

34.5.7. execution::get_delegatee_scheduler [exec.get.delegatee.scheduler]

  1. get_delegatee_scheduler asks an object for a scheduler that can be used to delegate work to for the purpose of forward progress delegation.

  2. The name get_delegatee_scheduler denotes a query object. For a subexpression env, get_delegatee_scheduler(env) is expression-equivalent to MANDATE-NOTHROW(as_const(env).query(get_delegatee_scheduler)).

    • Mandates: If the expression above is well-formed, its type satisfies scheduler.

  3. forwarding_query(execution::get_delegatee_scheduler) is a core constant expression and has value true.

  4. get_delegatee_scheduler() (with no arguments) is expression-equivalent to execution::read(get_delegatee_scheduler) ([exec.read]).

34.5.8. execution::get_forward_progress_guarantee [exec.get.forward.progress.guarantee]

namespace std::execution {
  enum class forward_progress_guarantee {
    concurrent,
    parallel,
    weakly_parallel
  };
}
  1. get_forward_progress_guarantee asks a scheduler about the forward progress guarantee of execution agents created by that scheduler.

  2. The name get_forward_progress_guarantee denotes a query object. For a subexpression sch, let Sch be decltype((sch)). If Sch does not satisfy scheduler, get_forward_progress_guarantee is ill-formed. Otherwise, get_forward_progress_guarantee(sch) is expression-equivalent to:

    1. MANDATE-NOTHROW(as_const(sch).query(get_forward_progress_guarantee)), if this expression is well-formed.

      • Mandates: The type of the expression above is forward_progress_guarantee.

    2. Otherwise, forward_progress_guarantee::weakly_parallel.

  3. If get_forward_progress_guarantee(sch) for some scheduler sch returns forward_progress_guarantee::concurrent, all execution agents created by that scheduler shall provide the concurrent forward progress guarantee. If it returns forward_progress_guarantee::parallel, all execution agents created by that scheduler shall provide at least the parallel forward progress guarantee.

34.5.9. this_thread::execute_may_block_caller [exec.execute.may.block.caller]

  1. this_thread::execute_may_block_caller asks a scheduler sch whether a call execute(sch, f) with any invocable f may block the thread where such a call occurs.

  2. The name this_thread::execute_may_block_caller denotes a query object. For a subexpression sch, let Sch be decltype((sch)). If Sch does not satisfy scheduler, this_thread::execute_may_block_caller is ill-formed. Otherwise, this_thread::execute_may_block_caller(sch) is expression-equivalent to:

    1. MANDATE-NOTHROW(as_const(sch).query(this_thread::execute_may_block_caller)), if this expression is well-formed.

      • Mandates: The type of the expression above is bool.

    2. Otherwise, true.

  3. If this_thread::execute_may_block_caller(sch) for some scheduler sch returns false, no execute(sch, f) call with some invocable f shall block the calling thread.

34.5.10. execution::get_completion_scheduler [exec.completion.scheduler]

  1. get_completion_scheduler<completion-tag> obtains the completion scheduler associated with a completion tag from a sender’s attributes.

  2. The name get_completion_scheduler denotes a query object template. For a subexpression q, let Q be decltype((q)). If the template argument Tag in get_completion_scheduler<Tag>(q) is not one of set_value_t, set_error_t, or set_stopped_t, get_completion_scheduler<Tag>(q) is ill-formed. Otherwise, get_completion_scheduler<Tag>(q) is expression-equivalent to MANDATE-NOTHROW(as_const(q).query(get_completion_scheduler<Tag>)).

    • Mandates: If the expression above is well-formed, its type satisfies scheduler.

  3. If, for some sender sndr and completion function C that has an associated completion tag Tag, get_completion_scheduler<Tag>(get_env(sndr)) is well-formed and results in a scheduler sch, and the sender sndr invokes C(rcvr, args...), for some receiver rcvr that has been connected to sndr, with additional arguments args..., on an execution agent that does not belong to the associated execution resource of sch, the behavior is undefined.

  4. The expression forwarding_query(get_completion_scheduler<CPO>) is a core constant expression and has value true.

34.6. Schedulers [exec.sched]

  1. The scheduler concept defines the requirements of a scheduler type ([async.ops]). schedule is a customization point object that accepts a scheduler. A valid invocation of schedule is a schedule-expression.

    namespace std::execution {
      template<class Sch>
        concept enable-scheduler = // exposition only
          requires {
            requires derived_from<typename Sch::scheduler_concept, scheduler_t>;
          };
    
      template<class Sch>
        concept scheduler =
          enable-scheduler<remove_cvref_t<Sch>> &&
          queryable<Sch> &&
          requires(Sch&& sch) {
            { schedule(std::forward<Sch>(sch)) } -> sender;
            { get_completion_scheduler<set_value_t>(
                get_env(schedule(std::forward<Sch>(sch)))) }
                  -> same_as<remove_cvref_t<Sch>>;
          } &&
          equality_comparable<remove_cvref_t<Sch>> &&
          copy_constructible<remove_cvref_t<Sch>>;
    }
    
  2. Let Sch be the type of a scheduler and let Env be the type of an execution environment for which sender_in<schedule_result_t<Sch>, Env> is true. Then sender-of-in<schedule_result_t<Sch>, Env> shall be true.

  3. None of a scheduler’s copy constructor, destructor, equality comparison, or swap member functions shall exit via an exception.

  4. None of these member functions, nor a scheduler type’s schedule function, shall introduce data races as a result of concurrent invocations of those functions from different threads.

  5. For any two (possibly const) values sch1 and sch2 of some scheduler type Sch, sch1 == sch2 shall return true only if both sch1 and sch2 share the same associated execution resource.

  6. For a given scheduler expression sch, the expression get_completion_scheduler<set_value_t>(get_env(schedule(sch))) shall compare equal to sch.

  7. For a given scheduler expression sch, if the expression get_domain(sch) is well-formed, then the expression get_domain(get_env(schedule(sch))) is also well-formed and has the same type.

  8. A scheduler type’s destructor shall not block pending completion of any receivers connected to the sender objects returned from schedule. The ability to wait for completion of submitted function objects can be provided by the associated execution resource of the scheduler.

34.7. Receivers [exec.recv]

34.7.1. Receiver concepts [exec.recv.concepts]

  1. A receiver represents the continuation of an asynchronous operation. The receiver concept defines the requirements for a receiver type ([async.ops]). The receiver_of concept defines the requirements for a receiver type that is usable as the first argument of a set of completion operations corresponding to a set of completion signatures. The get_env customization point is used to access a receiver’s associated environment.

    namespace std::execution {
      template<class Rcvr>
        concept enable-receiver = // exposition only
          requires {
            requires derived_from<typename Rcvr::receiver_concept, receiver_t>;
          };
    
      template<class Rcvr>
        concept receiver =
          enable-receiver<remove_cvref_t<Rcvr>> &&
          requires(const remove_cvref_t<Rcvr>& rcvr) {
            { get_env(rcvr) } -> queryable;
          } &&
          move_constructible<remove_cvref_t<Rcvr>> &&  // rvalues are movable, and
          constructible_from<remove_cvref_t<Rcvr>, Rcvr>; // lvalues are copyable
    
      template<class Signature, class Rcvr>
        concept valid-completion-for = // exposition only
          requires (Signature* sig) {
            []<class Tag, class... Args>(Tag(*)(Args...))
                requires callable<Tag, remove_cvref_t<Rcvr>, Args...>
            {}(sig);
          };
    
      template<class Rcvr, class Completions>
        concept has-completions = // exposition only
          requires (Completions* completions) {
            []<valid-completion-for<Rcvr>...Sigs>(completion_signatures<Sigs...>*)
            {}(completions);
          };
    
      template<class Rcvr, class Completions>
        concept receiver_of =
          receiver<Rcvr> && has-completions<Rcvr, Completions>;
    }
    
  2. Class types that are final do not model the receiver concept.

  3. Let rcvr be a receiver and let op_state be an operation state associated with an asynchronous operation created by connecting rcvr with a sender. Let token be a stop token equal to get_stop_token(get_env(rcvr)). token shall remain valid for the duration of the asynchronous operation’s lifetime ([async.ops]). This means that, unless it knows about further guarantees provided by the type of receiver rcvr, the implementation of op_state can not use token after it executes a completion operation. This also implies that any stop callbacks registered on token must be destroyed before the invocation of the completion operation.

34.7.2. execution::set_value [exec.set.value]

  1. set_value is a value completion function ([async.ops]). Its associated completion tag is set_value_t. The expression set_value(rcvr, vs...) for a subexpression rcvr and pack of subexpressions vs is ill-formed if rcvr is an lvalue or a const rvalue. Otherwise, it is expression-equivalent to MANDATE-NOTHROW(rcvr.set_value(vs...)).

34.7.3. execution::set_error [exec.set.error]

  1. set_error is an error completion function. Its associated completion tag is set_error_t. The expression set_error(rcvr, err) for some subexpressions rcvr and err is ill-formed if rcvr is an lvalue or a const rvalue. Otherwise, it is expression-equivalent to MANDATE-NOTHROW(rcvr.set_error(err)).

34.7.4. execution::set_stopped [exec.set.stopped]

  1. set_stopped is a stopped completion function. Its associated completion tag is set_stopped_t. The expression set_stopped(rcvr) for a subexpression rcvr is ill-formed if rcvr is an lvalue or a const rvalue. Otherwise, it is expression-equivalent to MANDATE-NOTHROW(rcvr.set_stopped()).

34.8. Operation states [exec.opstate]

  1. The operation_state concept defines the requirements of an operation state type ([async.ops]).

    namespace std::execution {
      template<class Rcvr>
        concept enable-opstate = // exposition only
          requires {
            requires derived_from<typename Rcvr::operation_state_concept, operation_state_t>;
          };
    
      template<class O>
        concept operation_state =
          enable-opstate<remove_cvref_t<O>> &&
          queryable<O> &&
          is_object_v<O> &&
          requires (O& o) {
            { start(o) } noexcept;
          };
    }
    
  2. If an operation_state object is moved during the lifetime of its asynchronous operation ([async.ops]), the behavior is undefined.

  3. Library-provided operation state types are non-movable.

34.8.1. execution::start [exec.opstate.start]

  1. The name start denotes a customization point object that starts ([async.ops]) the asynchronous operation associated with the operation state object. For a subexpression op, the expression start(op) is ill-formed if op is an rvalue. Otherwise, it is expression-equivalent to:

    MANDATE-NOTHROW(op.start())
    
  2. If op.start() does not start the asynchronous operation associated with the operation state op, the behavior of calling start(op) is undefined.

34.9. Senders [exec.snd]

34.9.1. General [exec.snd.general]

  1. For the purposes of this subclause, a sender is an object that satisfies the sender concept ([async.ops]).

  2. Subclauses [exec.factories] and [exec.adapt] define customizable algorithms that return senders. Each algorithm has a default implementation. Let sndr be the result of an invocation of such an algorithm or an object equal to such ([concepts.equality]), and let Sndr be decltype((sndr)). Let rcvr be a receiver with associated environment env of type Env such that sender_in<Sndr, Env> is true. For the default implementation of the algorithm that produced sndr, connecting sndr to rcvr and starting the resulting operation state ([async.ops]) necessarily results in the potential evaluation ([basic.def.odr]) of a set of completion operations whose first argument is a subexpression equal to rcvr. Let Sigs be a pack of completion signatures corresponding to this set of completion operations. Then the type of the expression get_completion_signatures(sndr, env) is a specialization of the class template completion_signatures, ([exec.utils.cmplsigs]) the set of whose template arguments is Sigs. If a user-provided implementation of the algorithm that produced sndr is selected instead of the default, any completion signature that is in the set of types denoted by completion_signatures_of_t<Sndr, Env> and that is not part of Sigs shall correspond to error or stopped completion operations, unless otherwise specified.

  3. This subclause makes use of the following exposition-only entities.

    1. For a queryable object env, let FWD-ENV(env) be a queryable object such that for a query object q and a pack of subexpressions as, the expression FWD-ENV(env).query(q, as...) is ill-formed if forwarding_query(q) is false; otherwise, it is expression-equivalent to env.query(q, as...).

    2. For a query object q and a subexpression v, let MAKE-ENV(q, v) be a queryable object env such that the result of env.query(q) has a value equal to v ([concepts.equality]). Unless otherwise stated, the object to which env.query(q) refers remains valid while env remains valid.

    3. For two queryable objects env1 and env2, a query object q and a pack of subexpressions as, let JOIN-ENV(env1, env2) be a queryable object env3 such that env3.query(q, as...) is expression-equivalent to:

      • env1.query(q, as...) if that expression is well-formed,

      • otherwise, env2.query(q, as...) if that expression is well-formed,

      • otherwise, env3.query(q, as...) is ill-formed.

    4. The expansions of FWD-ENV, MAKE-ENV, and JOIN-ENV can be context-dependent; i.e., they can expand to expressions with different types and value categories in different contexts for the same arguments.

    5. For a scheduler sch, let SCHED-ATTRS(sch) be a queryable object o1 such that o1.query(get_completion_scheduler<Tag>) is a prvalue with the same type and value as sch where Tag is one of set_value_t or set_stopped_t; and let o1.query(get_domain) be expression-equivalent to sch.query(get_domain). Let SCHED-ENV(sch) be a queryable object o2 such that o1.query(get_scheduler) is a prvalue with the same type and value as sch, and let o2.query(get_domain) be expression-equivalent to sch.query(get_domain).

    6. For two subexpressions rcvr and expr, let SET-VALUE(rcvr, expr) be (expr, set_value(rcvr)) if the type of expr is void; otherwise, it is set_value(rcvr, expr). Let TRY-EVAL(rcvr, expr) be:

      try {
        expr;
      } catch(...) {
        set_error(rcvr, current_exception());
      }
      

      if expr is potentially-throwing; otherwise, expr. Let TRY-SET-VALUE(rcvr, expr) be TRY-EVAL(rcvr, SET-VALUE(rcvr, expr)) except that rcvr is evaluated only once.

    7. template<class Default = default_domain, class Sndr>
        constexpr auto completion-domain(const Sndr& sndr) noexcept;
      
      1. Effects: Let COMPL-DOMAIN(T) be the type of the expression get_domain(get_completion_scheduler<T>(get_env(sndr))). If COMPL-DOMAIN(set_value_t), COMPL-DOMAIN(set_error_t), and COMPL-DOMAIN(set_stopped_t) all share a common type [meta.trans.other] (ignoring those types that are ill-formed), then completion-domain<Default>(sndr) is a default-constructed prvalue of that type. Otherwise, if all of those types are ill-formed, completion-domain<Default>(sndr) is a default-constructed prvalue of type Default. Otherwise, completion-domain<Default>(sndr) is ill-formed.

    8. template<class Tag, class Env, class Default>
        constexpr decltype(auto) query-with-default(
          Tag, const Env& env, Default&& value) noexcept(see below);
      
      1. Let e be the expression Tag()(env) if that expression is well-formed; otherwise, it is static_cast<Default>(std::forward<Default>(value)).

      2. Returns: e.

      3. Remarks: The expression in the noexcept clause is noexcept(e).

    9. template<class Sndr>
        constexpr auto get-domain-early(const Sndr& sndr) noexcept;
      
      1. Effects: Equivalent to return Domain(); where Domain is the decayed type of the first of the following expressions that is well-formed:

        • get_domain(get_env(sndr))

        • completion-domain(sndr)

        • default_domain()

    10. template<class Sndr, class Env>
        constexpr auto get-domain-late(const Sndr& sndr, const Env& env) noexcept;
      
      1. Effects: Equivalent to:

        • If sender-for<Sndr, transfer_t> is true, then return Domain(); where Domain is the type of the following expression:

          [] {
            auto [ignore1, sch, ignore2] = sndr;
            return query-or-default(get_domain, sch, default_domain());
          }();
          
        • Otherwise, return Domain(); where Domain is the first of the following expressions that is well-formed and has class type:

          • get_domain(get_env(sndr))

          • completion-domain<void>(sndr)

          • get_domain(env)

          • get_domain(get_scheduler(env))

          • default_domain().

        The transfer algorithm is unique in that it ignores the execution domain of its predecessor, using only the domain of its destination scheduler to select a customization.

    11. template<callable Fun>
        requires is_nothrow_move_constructible_v<Fun>
      struct emplace-from { // exposition only
        Fun fun; // exposition only
        using type = call-result-t<Fun>;
      
        constexpr operator type() && noexcept(nothrow-callable<Fun>) {
          return std::move(fun)();
        }
      
        constexpr type operator()() && noexcept(nothrow-callable<Fun>) {
          return std::move(fun)();
        }
      };
      
      1. emplace-from is used to emplace non-movable types into containers like tuple, optional, and variant.

    12. struct on-stop-request { // exposition only
        inplace_stop_source& stop-src; // exposition only
        void operator()() noexcept { stop-src.request_stop(); }
      };
      
    13. template<class... T>
      struct product-type {  // exposition only
        using type0 = T0;      // exposition only
        using type1 = T1;      // exposition only
          ...
        using typen-1 = Tn-1;   // exposition only
      
        T0 t0;      // exposition only
        T1 t1;      // exposition only
          ...
        Tn-1 tn-1;   // exposition only
      };
      
      1. An expression of type product-type is usable as the initializer of a structured binding declaration [dcl.struct.bind].

    14. template <semiregular Tag, movable-value Data = see below, sender... Child>
        constexpr auto make-sender(Tag, Data&& data, Child&&... child);
      
      1. Returns: A prvalue of type basic-sender<Tag, decay_t<Data>, decay_t<Child>...> where the tag member has been default-initialized and the data and childn... members have been direct initialized from their respective forwarded arguments, where basic-sender is the following exposition-only class template except as noted below:

        namespace std::execution {
          template<class T, class... Us>
          concept one-of = (same_as<T, Us> ||...); // exposition only
        
          template<class Tag>
          concept completion-tag = // exposition only
            one-of<Tag, set_value_t, set_error_t, set_stopped_t>;
        
          template<template<class...> class T, class... Args>
          concept well-formed = requires { typename T<Args...>; }; // exposition only
        
          template<const auto& Fun, class... Args>
          concept cpo-callable = callable<decltype(Fun), Args...>; // exposition only
        
          template<const auto& Fun, class... Args>
          using cpo-result-t = call-result-t<decltype(Fun), Args...>; // exposition only
        
          struct default-impls {  // exposition only
            static constexpr auto get-attrs = see below;
            static constexpr auto get-env = see below;
            static constexpr auto get-state = see below;
            static constexpr auto start = see below;
            static constexpr auto complete = see below;
          };
        
          template<class Tag>
          struct impls-for : default-impls {}; // exposition only
        
          template<class Sndr, class Rcvr> // exposition only
          using state-type = decay_t<cpo-result-t<
            impls-for<tag_of_t<Sndr>>::get-state, Sndr, Rcvr&>>;
        
          template<class Index, class Sndr, class Rcvr> // exposition only
          using env-type = cpo-result-t<
            impls-for<tag_of_t<Sndr>>::get-env, Index,
            state-type<Sndr, Rcvr>&, const Rcvr&>>;
        
          template<class Sndr, class Rcvr, class Index> 
            requires well-formed<env-type, Index, Sndr, Rcvr>
          struct basic-receiver {  // exposition only
            using receiver_concept = receiver_t;
        
            using tag-type = tag_of_t<Sndr>; // exposition only
            using state-type = state-type<Sndr, Rcvr>; // exposition only
            static constexpr const auto& complete = impls-for<tag-type>::complete; // exposition only
        
            template<class... Args>
              requires cpo-callable<complete, Index, state-type&, Rcvr&, set_value_t, Args...>
            void set_value(Args&&... args) && noexcept {
              complete(Index(), op->state, op->rcvr, set_value_t(), std::forward<Args>(args)...);
            }
        
            template<class Error>
              requires cpo-callable<complete, Index, state-type&, Rcvr&, set_error_t, Error>
            void set_error(Error&& err) && noexcept {
              complete(Index(), op->state, op->rcvr, set_error_t(), std::forward<Error>(err));
            }
        
            void set_stopped() && noexcept
              requires cpo-callable<complete, Index, state-type&, Rcvr&, set_stopped_t> {
              complete(Index(), op->state, op->rcvr, set_stopped_t());
            }
        
            auto get_env() const noexcept -> env-type<Index, Sndr, Rcvr> {
              const auto& rcvr = op->rcvr;
              return impls-for<tag-type>::get-env(Index(), op->state, rcvr);
            }
        
            basic-operation<Sndr, Rcvr>* op; // exposition only
          };
        
          constexpr auto connect-all =   // exposition only
            []<class Sndr, class Rcvr, size_t... Is>(
              basic-operation<Sndr, Rcvr>* op, Sndr&& sndr, index_sequence<Is...>)
                noexcept( TODO ) requires ( TODO ) {
                auto&& [ign1, ign2, ...child] = std::forward<Sndr>(sndr);
                return product-type{connect(
                  std::forward_like<Sndr>(child),
                  basic-receiver<Sndr, Rcvr, integral_constant<size_t, Is>>{op})...};
              };
        
          template<class Sndr>
          using indices-for = make_index_sequence<tuple_size_v<Sndr>-2>; // exposition only
        
          template<class Sndr, class Rcvr>
          using inner-ops-tuple =   // exposition only
            cpo-result-t<connect-all, basic-operation<Sndr, Rcvr>*, Sndr,
              indices-for<Sndr>>;
        
          template<class Sndr, class Rcvr>
            requires well-formed<state-type, Sndr, Rcvr> &&
              well-formed<inner-ops-tuple, Sndr, Rcvr>
          struct basic-operation {  // exposition only
            using operation_state_concept = operation_state_t;
            using tag-type = tag_of_t<Sndr>; // exposition only
        
            Rcvr rcvr; // exposition only
            state-type<Sndr, Rcvr> state; // exposition only
            inner-ops-tuple<Sndr, Rcvr> inner-ops; // exposition only
        
            basic-operation(Sndr&& sndr, Rcvr rcvr_init)  // exposition only
              : rcvr(std::move(rcvr_init))
              , state(impls-for<tag-type>::get-state(std::forward<Sndr>(sndr), rcvr))
              , inner-ops(connect-all(this, std::forward<Sndr>(sndr), indices-for<Sndr>()))
            {}
        
            void start() & noexcept {
              auto& [...ops] = inner-ops;
              impls-for<tag-type>::start(state, rcvr, ops...);
            }
          };
        
          template<class Sndr, class Env>
          using completion-signatures-for =  see below; // exposition only
        
          template<class Tag, class Data, class... Child>
          struct basic-sender {  // exposition only
            using sender_concept = sender_t;
        
            decltype(auto) get_env() const noexcept {
              return impls-for<Tag>::get-attrs(data, child0, ... childn-1);
            }
        
            template<decays-to<basic-sender> Self, receiver Rcvr>
            auto connect(this Self&& self, Rcvr rcvr) -> basic-operation<Self, Rcvr> {
              return {std::forward<Self>(self), std::move(rcvr)};
            }
        
            template<decays-to<basic-sender> Self, class Env>
            auto get_completion_signatures(this Self&& self, Env&& env) noexcept
              -> completion-signatures-for<Self, Env> {
              return {};
            }
        
            Tag tag;            // exposition only
            Data data;          // exposition only
            Child0 child0;      // exposition only
            Child1 child1;      // exposition only
              ...
            Childn-1 childn-1;   // exposition only
          };
        
          template <class Sndr>
          using data-type = decltype((declval<Sndr>().data)); // exposition only
        
          template <class Sndr, size_t N = 0>
          using child-type = decltype((declval<Sndr>().childN)); // exposition only
        }
        
      2. Remarks: The default template argument for the Data template parameter denotes an unspecified empty trivial class type.

      3. It is unspecified whether instances of basic-sender can be aggregate initialized.

      4. An expression of type basic-sender is usable as the initializer of a structured binding declaration [dcl.struct.bind].

      5. The member default-impls::get-attrs is initialized with a callable object equivalent to the following lambda:

        [](const auto& data, const auto&... child) noexcept -> decltype(auto) {
          if constexpr (sizeof...(child) == 1)
            return FWD-ENV(get_env(child...));
          else
            return empty_env();
        }
        
      6. The member default-impls::get-env is initialized with a callable object equivalent to the following lambda:

        []<class Rcvr>(auto index, auto& state, const Rcvr& rcvr) noexcept
          -> decltype(FWD-ENV(get_env(rcvr))) {
          return FWD-ENV(get_env(rcvr));
        }
        
      7. The member default-impls::get-state is initialized with a callable object equivalent to the following lambda:

        []<class Sndr, class Rcvr>(Sndr&& sndr, Rcvr& rcvr) noexcept -> decltype(auto) {
          return get<1>(std::forward<Sndr>(sndr));
        }
        
      8. The member default-impls::start is initialized with a callable object equivalent to the following lambda:

        [](auto& state, auto& rcvr, auto&... ops) noexcept -> void {
          (execution::start(ops), ...);
        }
        
      9. The member default-impls::complete is initialized with a callable object equivalent to the following lambda:

        []<class Index, class Rcvr, class Tag, class... Args>(
          Index, auto& state, Rcvr& rcvr, Tag, Args&&... args) noexcept
            -> void requires callable<Tag, Rcvr, Args...> {
          // Mandates: Index::value == 0
          Tag()(std::move(rcvr), std::forward<Args>(args)...);
        }
        
      10. For a subexpression sndr let Sndr be decltype((sndr)). Let rcvr be a receiver that has an associated environment of type Env such that sender_in<Sndr, Env> is true. completion-signatures-for<Sndr, Env> denotes a specialization of completion_signatures, the set of whose template arguments correspond to the set of completion operations that are potentially evaluated as a result of calling start on the operation state that results from connecting sndr and rcvr. When sender_in<Sndr, Env> is false, the type denoted by completion-signatures-for<Sndr, Env>, if any, is not a specialization of completion_signatures.

        Recommended practice: When sender_in<Sndr, Env> is false, implementations are encouraged to use the type denoted by completion-signatures-for<Sndr, Env> to communicate to users why.

34.9.2. Sender concepts [exec.snd.concepts]

  1. The sender concept defines the requirements for a sender type ([async.ops]). The sender_in concept defines the requirements for a sender type that can create asynchronous operations given an associated environment type. The sender_to concept defines the requirements for a sender type that can connect with a specific receiver type. The get_env customization point object is used to access a sender’s associated attributes. The connect customization point object is used to connect ([async.ops]) a sender and a receiver to produce an operation state.

    namespace std::execution {
      template<class Sigs>
        concept valid-completion-signatures = see below; // exposition only
    
      template<class Sndr>
        concept is-sender = // exposition only
          requires {
            requires derived_from<typename Sndr::sender_concept, sender_t>;
          };
    
      template<class Sndr>
        concept enable-sender = // exposition only
          is-sender<Sndr> ||
          is-awaitable<Sndr, env-promise<empty_env>>;  // [exec.awaitables]
    
      template<class Sndr>
        concept sender =
          bool(enable-sender<remove_cvref_t<Sndr>>) && // atomic constraint
          requires (const remove_cvref_t<Sndr>& sndr) {
            { get_env(sndr) } -> queryable;
          } &&
          move_constructible<remove_cvref_t<Sndr>> &&  // rvalues are movable, and
          constructible_from<remove_cvref_t<Sndr>, Sndr>; // lvalues are copyable
    
      template<class Sndr, class Env = empty_env>
        concept sender_in =
          sender<Sndr> &&
          queryable<Env> &&
          requires (Sndr&& sndr, Env&& env) {
            { get_completion_signatures(std::forward<Sndr>(sndr), std::forward<Env>(env)) }
              -> valid-completion-signatures;
          };
    
      template<class Sndr, class Rcvr>
        concept sender_to =
          sender_in<Sndr, env_of_t<Rcvr>> &&
          receiver_of<Rcvr, completion_signatures_of_t<Sndr, env_of_t<Rcvr>>> &&
          requires (Sndr&& sndr, Rcvr&& rcvr) {
            connect(std::forward<Sndr>(sndr), std::forward<Rcvr>(rcvr));
          };
    }
    
  2. Given a subexpression sndr, let Sndr be decltype((sndr)), let Env be the type of an environment, and let rcvr be a receiver with an associated environment Env. A completion operation is a permissible completion for Sndr and Env if its completion signature appears in the argument list of the specialization of completion_signatures denoted by completion_signatures_of_t<Sndr, Env>. Sndr and Env model sender_in<Sndr, Env> if all the completion operations that are potentially evaluated by connecting sndr to rcvr and starting the resulting operation state are permissible completions for Sndr and Env.

  3. A type Sigs satisfies and models the exposition-only concept valid-completion-signatures if it denotes a specialization of the completion_signatures class template.

  4. The exposition-only concepts sender-of and sender-of-in define the requirements for a sender type that completes with a given unique set of value result types.

    namespace std::execution {
      template<class... As>
        using value-signature = set_value_t(As...); // exposition only
    
      template<class Sndr, class Env, class... Values>
        concept sender-of-in =
          sender_in<Sndr, Env> &&
          MATCHING-SIG( // see [exec.general]
            set_value_t(Values...),
            value_types_of_t<Sndr, Env, value-signature, type_identity_t>);
    
      template<class Sndr, class... Values>
        concept sender-of = sender-of-in<Sndr, empty_env, Values...>;
    }
    
  5. Let sndr be an expression such that decltype((sndr)) is Sndr. The type tag_of_t<Sndr> is as follows:

    • If the declaration auto&& [tag, data, ...children] = sndr; would be well-formed, tag_of_t<Sndr> is an alias for decltype(auto(tag)).

    • Otherwise, tag_of_t<Sndr> is ill-formed.

    There is no way in standard C++ to determine whether the above declaration is well-formed without causing a hard error, so this presumes compiler magic. However, the author anticipates the adoption of [@P2141R1], which makes it possible to implement this purely in the library. P2141 has already been approved by EWG for C++26.

  6. Let sender-for be an exposition-only concept defined as follows:

    namespace std::execution {
      template<class Sndr, class Tag>
      concept sender-for =
        sender<Sndr> &&
        same_as<tag_of_t<Sndr>, Tag>;
    }
    
  7. For a type T, SET-VALUE-SIG(T) denotes the type set_value_t() if T is cv void; otherwise, it denotes the type set_value_t(T).

  8. Library-provided sender types:

    • Always expose an overload of a member connect that accepts an rvalue sender.

    • Only expose an overload of a member connect that accepts an lvalue sender if they model copy_constructible.

    • Model copy_constructible if they satisfy copy_constructible.

34.9.3. Awaitable helpers [exec.awaitables]

  1. The sender concepts recognize awaitables as senders. For this clause ([exec]), an awaitable is an expression that would be well-formed as the operand of a co_await expression within a given context.

  2. For a subexpression c, let GET-AWAITER(c, p) be expression-equivalent to the series of transformations and conversions applied to c as the operand of an await-expression in a coroutine, resulting in lvalue e as described by [expr.await]/3.2-4, where p is an lvalue referring to the coroutine’s promise type, Promise. This includes the invocation of the promise type’s await_transform member if any, the invocation of the operator co_await picked by overload resolution if any, and any necessary implicit conversions and materializations.

    I have opened cwg#250 to give these transformations a term-of-art so we can more easily refer to it here.

  3. Let is-awaitable be the following exposition-only concept:

    namespace std {
      template<class T>
      concept await-suspend-result = see below;
    
      template<class A, class Promise>
      concept is-awaiter = // exposition only
        requires (A& a, coroutine_handle<Promise> h) {
          a.await_ready() ? 1 : 0;
          { a.await_suspend(h) } -> await-suspend-result;
          a.await_resume();
        };
    
      template<class C, class Promise>
      concept is-awaitable =
        requires (C (*fc)() noexcept, Promise& p) {
          { GET-AWAITER(fc(), p) } -> is-awaiter<Promise>;
        };
    }
    

    await-suspend-result<T> is true if and only if one of the following is true:

    • T is void, or

    • T is bool, or

    • T is a specialization of coroutine_handle.

  4. For a subexpression c such that decltype((c)) is type C, and an lvalue p of type Promise, await-result-type<C, Promise> denotes the type decltype(GET-AWAITER(c, p).await_resume()).

  5. Let with-await-transform be the exposition-only class template:

    namespace std::execution {
      template<class T, class Promise>
        concept has-as-awaitable = // exposition only
          requires (T&& t, Promise& p) {
            { std::forward<T>(t).as_awaitable(p) } -> is-awaitable<Promise&>;
          };
    
      template<class Derived>
        struct with-await-transform {
          template<class T>
            T&& await_transform(T&& value) noexcept {
              return std::forward<T>(value);
            }
    
          template<has-as-awaitable<Derived> T>
            auto await_transform(T&& value)
              noexcept(noexcept(std::forward<T>(value).as_awaitable(declval<Derived&>())))
              -> decltype(std::forward<T>(value).as_awaitable(declval<Derived&>())) {
              return std::forward<T>(value).as_awaitable(static_cast<Derived&>(*this));
            }
        };
    }
    
  6. Let env-promise be the exposition-only class template:

    namespace std::execution {
      template<class Env>
      struct env-promise : with-await-transform<env-promise<Env>> {
        unspecified get_return_object() noexcept;
        unspecified initial_suspend() noexcept;
        unspecified final_suspend() noexcept;
        void unhandled_exception() noexcept;
        void return_void() noexcept;
        coroutine_handle<> unhandled_stopped() noexcept;
    
        const Env& get_env() const noexcept;
      };
    }
    

    Specializations of env-promise are only used for the purpose of type computation; its members need not be defined.

34.9.4. execution::default_domain [exec.domain.default]

namespace std::execution {
  struct default_domain {
    template <sender Sndr, queryable... Env>
        requires (sizeof...(Env) <= 1)
      static constexpr sender decltype(auto) transform_sender(Sndr&& sndr, const Env&... env)
        noexcept(see below);

    template <sender Sndr, queryable Env>
      static constexpr queryable decltype(auto) transform_env(Sndr&& sndr, Env&& env) noexcept;

    template<class Tag, sender Sndr, class... Args>
      static constexpr decltype(auto) apply_sender(Tag, Sndr&& sndr, Args&&... args)
        noexcept(see below);
  };
}
34.9.4.1. Static members [exec.domain.default.statics]
template <sender Sndr, queryable... Env>
    requires (sizeof...(Env) <= 1)
  constexpr sender decltype(auto) transform_sender(Sndr&& sndr, const Env&... env)
    noexcept(see below);
  1. Let e be the expression tag_of_t<Sndr>().transform_sender(std::forward<Sndr>(sndr), env...) if that expression is well-formed; otherwise, std::forward<Sndr>(sndr).

  2. Returns: e.

  3. Remarks: The exception specification is equivalent to noexcept(e).

template <sender Sndr, queryable Env>
  constexpr queryable decltype(auto) transform_env(Sndr&& sndr, Env&& env) noexcept;
  1. Let e be the expression tag_of_t<Sndr>().transform_env(std::forward<Sndr>(sndr), std::forward<Env>(env)) if that expression is well-formed; otherwise, static_cast<Env>(std::forward<Env>(env)).

  2. Mandates: noexcept(e) is true.

  3. Returns: e.

template<class Tag, sender Sndr, class... Args>
  constexpr decltype(auto) apply_sender(Tag, Sndr&& sndr, Args&&... args)
    noexcept(see below);
  1. Let e be the expression Tag().apply_sender(std::forward<Sndr>(sndr), std::forward<Args>(args)...).

  2. Constraints: e is a well-formed expression.

  3. Returns: e.

  4. Remarks: The exception specification is equivalent to noexcept(e).

34.9.5. execution::transform_sender [exec.snd.transform]

namespace std::execution {
  template<class Domain, sender Sndr, queryable... Env>
      requires (sizeof...(Env) <= 1)
    constexpr sender decltype(auto) transform_sender(Domain dom, Sndr&& sndr, const Env&... env)
      noexcept(see below);
}
  1. Let transformed-sndr be the expression dom.transform_sender(std::forward<Sndr>(sndr), env...) if that expression is well-formed; otherwise, default_domain().transform_sender(std::forward<Sndr>(sndr), env...). Let final-sndr be the expression transformed-sndr if transformed-sndr and sndr have the same type ignoring cv qualifiers; otherwise, it is the expression transform_sender(dom, transformed-sndr, env...).

  2. Returns: final-sndr.

  3. Remarks: The exception specification is equivalent to noexcept(final-sndr).

34.9.6. execution::transform_env [exec.snd.transform.env]

namespace std::execution {
  template<class Domain, sender Sndr, queryable Env>
    constexpr queryable decltype(auto) transform_env(Domain dom, Sndr&& sndr, Env&& env) noexcept;
}
  1. Let e be the expression dom.transform_env(std::forward<Sndr>(sndr), std::forward<Env>(env)) if that expression is well-formed; otherwise, default_domain().transform_env(std::forward<Sndr>(sndr), std::forward<Env>(env)).

  2. Mandates: noexcept(e) is true.

  3. Returns: e.

34.9.7. execution::apply_sender [exec.snd.apply]

namespace std::execution {
  template<class Domain, class Tag, sender Sndr, class... Args>
    constexpr decltype(auto) apply_sender(Domain dom, Tag, Sndr&& sndr, Args&&... args)
      noexcept(see below);
}
  1. Let e be the expression dom.apply_sender(Tag(), std::forward<Sndr>(sndr), std::forward<Args>(args)...) if that expression is well-formed; otherwise, default_domain().apply_sender(Tag(), std::forward<Sndr>(sndr), std::forward<Args>(args)...).

  2. Constraints: The expression e is well-formed.

  3. Returns: e.

  4. Remarks: The exception specification is equivalent to noexcept(e).

34.9.8. execution::get_completion_signatures [exec.getcomplsigs]

  1. get_completion_signatures is a customization point object. Let sndr be an expression such that decltype((sndr)) is Sndr, and let env be an expression such that decltype((env)) is Env. Then get_completion_signatures(sndr, env) is expression-equivalent to:

    1. decltype(sndr.get_completion_signatures(env)){} if that expression is well-formed,

    2. Otherwise, remove_cvref_t<Sndr>::completion_signatures{} if that expression is well-formed,

    3. Otherwise, if is-awaitable<Sndr, env-promise<Env>> is true, then:

      completion_signatures<
        SET-VALUE-SIG(await-result-type<Sndr,
                      env-promise<Env>>), // see [exec.snd.concepts]
        set_error_t(exception_ptr),
        set_stopped_t()>{}
      
    4. Otherwise, get_completion_signatures(sndr, env) is ill-formed.

  2. Let rcvr be an rvalue receiver of type Rcvr, and let Sndr be the type of a sender such that sender_in<Sndr, env_of_t<Rcvr>> is true. Let Sigs... be the template arguments of the completion_signatures specialization named by completion_signatures_of_t<Sndr, env_of_t<Rcvr>>. Let CSO be a completion function. If sender Sndr or its operation state cause the expression CSO(rcvr, args...) to be potentially evaluated ([basic.def.odr]) then there shall be a signature Sig in Sigs... such that MATCHING-SIG(decayed-typeof<CSO>(decltype(args)...), Sig) is true ([exec.general]).

34.9.9. execution::connect [exec.connect]

  1. connect connects ([async.ops]) a sender with a receiver.

  2. The name connect denotes a customization point object. For subexpressions sndr and rcvr, let Sndr be decltype((sndr)) and Rcvr be decltype((rcvr)), and let DS and DR be the decayed types of Sndr and Rcvr, respectively.

  3. Let connect-awaitable-promise be the following class:

    namespace std::execution {
      struct connect-awaitable-promise
        : with-await-transform<connect-awaitable-promise> {
        DR& rcvr; // exposition only
    
        connect-awaitable-promise(DS&, DR& rcvr) noexcept : rcvr(rcvr) {}
    
        suspend_always initial_suspend() noexcept { return {}; }
        [[noreturn]] suspend_always final_suspend() noexcept { terminate(); }
        [[noreturn]] void unhandled_exception() noexcept { terminate(); }
        [[noreturn]] void return_void() noexcept { terminate(); }
    
        coroutine_handle<> unhandled_stopped() noexcept {
          set_stopped((DR&&) rcvr);
          return noop_coroutine();
        }
    
        operation-state-task get_return_object() noexcept {
          return operation-state-task{
            coroutine_handle<connect-awaitable-promise>::from_promise(*this)};
        }
    
        env_of_t<const DR&> get_env() const noexcept {
          return execution::get_env(rcvr);
        }
      };
    }
    
  4. Let operation-state-task be the following class:

    namespace std::execution {
      struct operation-state-task {
        using operation_state_concept = operation_state_t;
        using promise_type = connect-awaitable-promise;
        coroutine_handle<> coro; // exposition only
    
        explicit operation-state-task(coroutine_handle<> h) noexcept : coro(h) {}
        operation-state-task(operation-state-task&& o) noexcept
          : coro(exchange(o.coro, {})) {}
        ~operation-state-task() { if (coro) coro.destroy(); }
    
        void start() & noexcept {
          coro.resume();
        }
      };
    }
    
  5. Let V name the type await-result-type<DS, connect-awaitable-promise>, let Sigs name the type:

    completion_signatures<
      SET-VALUE-SIG(V), // see [exec.snd.concepts]
      set_error_t(exception_ptr),
      set_stopped_t()>
    

    and let connect-awaitable be an exposition-only coroutine defined as follows:

    namespace std::execution {
      template<class Fun, class... Ts>
      auto suspend-complete(Fun fun, Ts&&... as) noexcept { // exposition only
        auto fn = [&, fun]() noexcept { fun(std::forward<Ts>(as)...); };
    
        struct awaiter {
          decltype(fn) fn;
    
          static constexpr bool await_ready() noexcept { return false; }
          void await_suspend(coroutine_handle<>) noexcept { fn(); }
          [[noreturn]] void await_resume() noexcept { unreachable(); }
        };
        return awaiter{fn};
      };
    
      operation-state-task connect-awaitable(DS sndr, DR rcvr) requires receiver_of<DR, Sigs> {
        exception_ptr ep;
        try {
          if constexpr (same_as<V, void>) {
            co_await std::move(sndr);
            co_await suspend-complete(set_value, std::move(rcvr));
          } else {
            co_await suspend-complete(set_value, std::move(rcvr), co_await std::move(sndr));
          }
        } catch(...) {
          ep = current_exception();
        }
        co_await suspend-complete(set_error, std::move(rcvr), std::move(ep));
      }
    }
    
  6. If Sndr does not satisfy sender or if Rcvr does not satisfy receiver, connect(sndr, rcvr) is ill-formed. Otherwise, the expression connect(sndr, rcvr) is expression-equivalent to:

    1. sndr.connect(rcvr) if that expression is well-formed.

      • Mandates: The type of the expression above satisfies operation_state.

    2. Otherwise, connect-awaitable(sndr, rcvr) if that expression is well-formed.

    3. Otherwise, connect(sndr, rcvr) is ill-formed.

34.9.10. Sender factories [exec.factories]

34.9.10.1. execution::schedule [exec.schedule]
  1. schedule obtains a schedule-sender ([async.ops]) from a scheduler.

  2. The name schedule denotes a customization point object. For a subexpression sch, the expression schedule(sch) is expression-equivalent to:

    1. sch.schedule() if that expression is valid. If sch.schedule() does not return a sender whose set_value completion scheduler is equal to sch, the behavior of calling schedule(sch) is undefined.

      • Mandates: The type of sch.schedule() satisfies sender.

    2. Otherwise, schedule(sch) is ill-formed.

34.9.10.2. execution::just, execution::just_error, execution::just_stopped [exec.just]
  1. just, just_error, and just_stopped are sender factories whose asynchronous operations complete synchronously in their start operation with a value completion operation, an error completion operation, or a stopped completion operation respectively.

  2. The names just, just_error, and just_stopped denote customization point objects. Let just-cpo be one of just, just_error, or just_stopped. For a pack of subexpressions ts, let Ts be the template parameter pack decltype((ts)). The expression just-cpo(ts...) is ill-formed if:

    • (movable-value<Ts> &&...) is false, or

    • just-cpo is just_error and sizeof...(ts) == 1 is false, or

    • just-cpo is just_stopped and sizeof...(ts) == 0 is false;

    Otherwise, it is expression-equivalent to make-sender(just-cpo, product-type{vs...}).

  3. For just, just_error, and just_stopped, let set-cpo be set_value, set_error, and set_stopped respectively. The exposition-only class template impls-for ([exec.snd.general]) is specialized for just-cpo as follows:

    namespace std::execution {
      template<>
      struct impls-for<decayed-typeof<just-cpo>> : default-impls {
        static constexpr auto start =
          [](auto& state, auto& rcvr) noexcept -> void {
            auto& [...ts] = state;
            set-cpo(std::move(rcvr), std::move(ts)...);
          };
      };
    }
    
34.9.10.3. execution::read [exec.read]
  1. read is a sender factory for a sender whose asynchronous operation completes synchronously in its start operation with a value completion result equal to a value read from the receiver’s associated environment.

  2. read is a customization point object. For some query object q, the expression read(q) is expression-equivalent to make-sender(read, q).

  3. The exposition-only class template impls-for ([exec.snd.general]) is specialized for read as follows:

    namespace std::execution {
      template<>
      struct impls-for<decayed-typeof<read>> : default-impls {
        static constexpr auto start =
          [](auto query, auto& rcvr) noexcept -> void {
            TRY-SET-VALUE(std::move(rcvr), query(get_env(rcvr)));
          };
      };
    }
    

34.9.11. Sender adaptors [exec.adapt]

34.9.11.1. General [exec.adapt.general]
  1. Subclause [exec.adapt] specifies a set of sender adaptors.

  2. The bitwise OR operator is overloaded for the purpose of creating sender chains. The adaptors also support function call syntax with equivalent semantics.

  3. Unless otherwise specified, a sender adaptor is prohibited from causing observable effects, apart from moving and copying its arguments, before the returned sender is connected with a receiver using connect, and start is called on the resulting operation state. This requirement applies to any function that is selected by the implementation of the sender adaptor.

  4. Unless otherwise specified, a parent sender ([async.ops]) with a single child sender sndr has an associated attribute object equal to FWD-ENV(get_env(sndr)) ([exec.fwd.env]). Unless otherwise specified, a parent sender with more than one child senders has an associated attributes object equal to empty_env{}. These requirements apply to any function that is selected by the implementation of the sender adaptor.

  5. Unless otherwise specified, when a parent sender is connected to a receiver rcvr, any receiver used to connect a child sender has an associated environment equal to FWD-ENV(get_env(rcvr)). This requirement applies to any sender returned from a function that is selected by the implementation of such sender adaptor.

  6. If a sender returned from a sender adaptor specified in this subclause is specified to include set_error_t(Err) among its set of completion signatures where decay_t<Err> denotes the type exception_ptr, but the implementation does not potentially evaluate an error completion operation with an exception_ptr argument, the implementation is allowed to omit the exception_ptr error completion signature from the set.

34.9.11.2. Sender adaptor closure objects [exec.adapt.objects]
  1. A pipeable sender adaptor closure object is a function object that accepts one or more sender arguments and returns a sender. For a sender adaptor closure object c and an expression sndr such that decltype((sndr)) models sender, the following expressions are equivalent and yield a sender:

    c(sndr)
    sndr | c
    

    Given an additional pipeable sender adaptor closure object d, the expression c | d produces another pipeable sender adaptor closure object e:

    e is a perfect forwarding call wrapper ([func.require]) with the following properties:

    • Its target object is an object d2 of type decay_t<decltype((d))> direct-non-list-initialized with d.

    • It has one bound argument entity, an object c2 of type decay_t<decltype((c))> direct-non-list-initialized with C.

    • Its call pattern is d2(c2(arg)), where arg is the argument used in a function call expression of e.

The expression c | d is well-formed if and only if the initializations of the state entities of e are all well-formed.

  1. An object t of type T is a pipeable sender adaptor closure object if T models derived_from<sender_adaptor_closure<T>>, T has no other base classes of type sender_adaptor_closure<U> for any other type U, and T does not model sender.

  2. The template parameter D for sender_adaptor_closure can be an incomplete type. Before any expression of type cv D appears as an operand to the | operator, D shall be complete and model derived_from<sender_adaptor_closure<D>>. The behavior of an expression involving an object of type cv D as an operand to the | operator is undefined if overload resolution selects a program-defined operator| function.

  3. A pipeable sender adaptor object is a customization point object that accepts a sender as its first argument and returns a sender.

  4. If a pipeable sender adaptor object accepts only one argument, then it is a pipeable sender adaptor closure object.

  5. If a pipeable sender adaptor object adaptor accepts more than one argument, then let sndr be an expression such that decltype((sndr)) models sender, let args... be arguments such that adaptor(sndr, args...) is a well-formed expression as specified in the rest of this subclause ([exec.adapt.objects]), and let BoundArgs be a pack that denotes decay_t<decltype((args))>.... The expression adaptor(args...) produces a pipeable sender adaptor closure object f that is a perfect forwarding call wrapper with the following properties:

    • Its target object is a copy of adaptor.

    • Its bound argument entities bound_args consist of objects of types BoundArgs... direct-non-list-initialized with std::forward<decltype((args))>(args)..., respectively.

    • Its call pattern is adaptor(rcvr, bound_args...), where rcvr is the argument used in a function call expression of f.

    The expression adaptor(args...) is well-formed if and only if the initializations of the bound argument entities of the result, as specified above, are all well-formed.

34.9.11.3. execution::on [exec.on]
  1. on adapts an input sender into a sender that will start on an execution agent belonging to a particular scheduler’s associated execution resource.

  2. The name on denotes a customization point object. For some subexpressions sch and sndr, if decltype((sch)) does not satisfy scheduler, or decltype((sndr)) does not satisfy sender, on(sch, sndr) is ill-formed.

  3. Otherwise, the expression on(sch, sndr) is expression-equivalent to:

    transform_sender(
      query-or-default(get_domain, sch, default_domain()),
      make-sender(on, sch, sndr));
    
  4. Let out_sndr and env be subexpressions such that OutSndr is decltype((out_sndr)). If sender-for<OutSndr, on_t> is false, then the expressions on.transform_env(out_sndr, env) and on.transform_sender(out_sndr, env) are ill-formed; otherwise:

    • on.transform_env(out_sndr, env) is equivalent to:

      auto&& [ign1, sch, ign2] = out_sndr;
      return JOIN-ENV(SCHED-ENV(sch), FWD-ENV(env));
      
    • on.transform_sender(out_sndr, env) is equivalent to:

      auto&& [ign, sch, sndr] = out_sndr;
      return let_value(
        schedule(sch),
        [sndr = std::forward_like<OutSndr>(sndr)]() mutable {
          return std::move(sndr);
        });
      
  5. Let out_sndr be a subexpression denoting a sender returned from on(sch, sndr) or one equal to such, and let OutSndr be the type decltype((out_sndr)). Let out_rcvr be a subexpression denoting a receiver that has an environment of type Env such that sender_in<OutSndr, Env> is true. Let op be an lvalue referring to the operation state that results from connecting out_sndr with out_rcvr. Calling start(op) shall start sndr on an execution agent of the associated execution resource of sch, or failing that, shall execute an error completion on out_rcvr.

34.9.11.4. execution::transfer [exec.transfer]
  1. transfer adapts a sender into one with a different associated set_value completion scheduler. It results in a transition between different execution resources when executed.

  2. The name transfer denotes a customization point object. For some subexpressions sch and sndr, if decltype((sch)) does not satisfy scheduler, or decltype((sndr)) does not satisfy sender, transfer(sndr, sch) is ill-formed.

  3. Otherwise, the expression transfer(sndr, sch) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(transfer, sch, sndr));
    
  4. The exposition-only class template impls-for is specialized for transfer_t as follows:

    namespace std::execution {
      template<>
      struct impls-for<transfer_t> : default-impls {
        static constexpr auto get_attrs =
          [](const auto& data, const auto& child) noexcept -> decltype(auto) {
            return JOIN-ENV(SCHED-ATTRS(data), FWD-ENV(get_env(child)));
          };
      };
    }
    
  5. Let sndr and env be subexpressions such that Sndr is decltype((sndr)). If sender-for<Sndr, transfer_t> is false, then the expression transfer.transform_sender(sndr, env) is ill-formed; otherwise, it is equal to:

    auto [tag, data, child] = sndr;
    return schedule_from(std::move(data), std::move(child));
    

    This causes the transfer(sndr, sch) sender to become schedule_from(sch, sndr) when it is connected with a receiver with an execution domain that does not customize transfer.

  6. Let out_sndr be a subexpression denoting a sender returned from transfer(sndr, sch) or one equal to such, and let OutSndr be the type decltype((out_sndr)). Let out_rcvr be a subexpression denoting a receiver that has an environment of type Env such that sender_in<OutSndr, Env> is true. Let op be an lvalue referring to the operation state that results from connecting out_sndr with out_rcvr. Calling start(op) shall start sndr on the current execution agent and execute completion operations on out_rcvr on an execution agent of the execution resource associated with sch. If scheduling onto sch fails, execute an error completion on out_rcvr on an unspecified execution agent.

34.9.11.5. execution::schedule_from [exec.schedule.from]
  1. schedule_from schedules work dependent on the completion of a sender onto a scheduler’s associated execution resource. schedule_from is not meant to be used in user code; it is used in the implementation of transfer.

  2. The name schedule_from denotes a customization point object. For some subexpressions sch and sndr, let Sch be decltype((sch)) and Sndr be decltype((sndr)). If Sch does not satisfy scheduler, or Sndr does not satisfy sender, schedule_from is ill-formed.

  3. Otherwise, the expression schedule_from(sch, sndr) is expression-equivalent to:

    transform_sender(
      query-or-default(get_domain, sch, default_domain()),
      make-sender(schedule_from, sch, sndr));
    
  4. The exposition-only class template impls-for ([exec.snd.general]) is specialized for schedule_from_t as follows:

    namespace std::execution {
      template<>
      struct impls-for<schedule_from_t> : default-impls {
        static constexpr auto get-attrs = see below;
        static constexpr auto get-state = see below;
        static constexpr auto complete = see below;
      };
    }
    
    1. The member impls-for<schedule_from_t>::get-attrs is initialized with a callable object equivalent to the following lambda:

      [](const auto& data, const auto& child) noexcept -> decltype(auto) {
        return JOIN-ENV(SCHED-ATTRS(data), FWD-ENV(get_env(child)));
      }
      
    2. The member impls-for<schedule_from_t>::get-state is initialized with a callable object equivalent to the following lambda:

      []<class Sndr, class Rcvr>(Sndr&& sndr, Rcvr& rcvr)
          requires sender_in<child-type<Sndr>, env_of_t<Rcvr>> {
        return apply(
          [&]<class Sch, class Child>(auto, Sch sch, Child&& child) {
            using variant-type = see below;
            using receiver-type = see below;
            using operation-type = connect_result_t<schedule_result_t<Sch>, receiver-type>;
      
            struct state-type {
              Rcvr& rcvr;
              variant-type async-result;
              operation-type op-state;
      
              explicit state-type(Sch sch, Rcvr& rcvr)
                : rcvr(rcvr), op-state(connect(schedule(sch), receiver-type{{}, this})) {}
            };
      
            return state-type{sch, rcvr};
          },
          std::forward<Sndr>(sndr));
      }
      
      1. The local class state-type is a structural type.

      2. Let Sigs be a pack of the arguments to the completion_signatures specialization named by completion_signatures_of_t<Child, env_of_t<Rcvr>>. Let as-tuple be an alias template that transforms a completion signature Tag(Args...) into the tuple specialization decayed-tuple<Tag, Args...>. Then variant-type denotes the type variant<monostate, as-tuple<Sigs>...>, except with duplicate types removed.

      3. Let receiver-type denote the following exposition-only class:

        namespace std::execution {
          struct receiver-type {
            using receiver_concept = receiver_t;
            state-type* state; // exposition only
        
            Rcvr&& base() && noexcept { return std::move(state->rcvr); }
            const Rcvr& base() const & noexcept { return state->rcvr; }
        
            void set_value() && noexcept {
              visit(
                [this]<class Tuple>(Tuple& result) noexcept -> void {
                  if constexpr (!same_as<monostate, Tuple>) {
                    auto& [tag, ...args] = result;
                    tag(std::move(state->rcvr), std::move(args)...);
                  }
                },
                state->async-result);
            }
        
            template<class Error>
            void set_error(Error&& err) && noexcept {
              execution::set_error(std::move(state->rcvr), std::forward<Error>(err));
            }
        
            void set_stopped() && noexcept {
              execution::set_stopped(std::move(state->rcvr));
            }
        
            decltype(auto) get_env() const noexcept {
              return FWD-ENV(execution::get_env(state->rcvr));
            }
          };
        }
        
    3. The member impls-for<schedule_from_t>::complete is initialized with a callable object equivalent to the following lambda:

      []<class Tag, class... Args>(auto, auto& state, auto& rcvr, Tag, Args&&... args) noexcept -> void {
        using result_t = decayed-tuple<Tag, Args...>;
        constexpr bool nothrow = is_nothrow_constructible_v<result_t, Tag, Args...>;
      
        TRY-EVAL(std::move(rcvr), [&]() noexcept(nothrow) {
          state.async-result.template emplace<result_t>(Tag(), std::forward<Args>(args)...);
        }());
      
        if (state.async-result.valueless_by_exception())
          return;
        if (state.async-result.index() == 0)
          return;
      
        start(state.op-state);
      };
      
  5. Let the subexpression out_sndr denote the result of the invocation schedule_from(sch, sndr) or an object copied or moved from such, and let the subexpression rcvr denote a receiver such that the expression connect(out_sndr, rcvr) is well-formed. The expression connect(out_sndr, rcvr) has undefined behavior unless it creates an asynchronous operation ([async.ops]) that, when started:

    • eventually completes on an execution agent belonging to the associated execution resource of sch, and

    • completes with the same async result as sndr.

34.9.11.6. execution::then, execution::upon_error, execution::upon_stopped [exec.then]
  1. then attaches an invocable as a continuation for an input sender’s value completion operation. upon_error and upon_stopped do the same for the error and stopped completion operations respectively, sending the result of the invocable as a value completion.

  2. The names then, upon_error, and upon_stopped denote customization point objects. Let the expression then-cpo be one of then, upon_error, or upon_stopped. For subexpressions sndr and f, let Sndr be decltype((sndr)) and let F be the decayed type of f. If Sndr does not satisfy sender, or F does not satisfy movable-value, then-cpo(sndr, f) is ill-formed.

  3. Otherwise, the expression then-cpo(sndr, f) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(then-cpo, f, sndr));
    
  4. For then, upon_error, and upon_stopped, let set-cpo be set_value, set_error, and set_stopped respectively. The exposition-only class template impls-for ([exec.snd.general]) is specialized for then-cpo as follows:

    namespace std::execution {
      template<>
      struct impls-for<decayed-typeof<then-cpo>> : default-impls {
        static constexpr auto complete =
          []<class Tag, class... Args>
            (auto /*index*/, auto& fn, auto& rcvr, Tag, Args&&... args) noexcept -> void {
              if constexpr (same_as<Tag, decayed-typeof<set-cpo>>) {
                TRY-SET-VALUE(std::move(rcvr),
                              invoke(std::move(fn), std::forward<Args>(args)...));
              } else {
                Tag()(std::move(rcvr), std::forward<Args>(args)...);
              }
            };
      };
    }
    
  5. The expression then-cpo(sndr, f) has undefined behavior unless it returns a sender out_sndr that:

    1. Invokes f or a copy of such with the value, error, or stopped result datums of sndr (for then, upon_error, and upon_stopped respectively), using the result value of f as out_sndr's value completion, and

    2. Forwards all other completion operations unchanged.

34.9.11.7. execution::let_value, execution::let_error, execution::let_stopped, [exec.let]
  1. let_value, let_error, and let_stopped transform a sender’s value, error, and stopped completions respectively into a new child asynchronous operation by passing the sender’s result datums to a user-specified callable, which returns a new sender that is connected and started.

  2. Let the expression let-cpo be one of let_value, let_error, or let_stopped and let set-cpo be the completion function that corresponds to let-cpo (set_value for let_value, etc.). For a subexpression sndr, let let-env(sndr) be expression-equivalent to the first well-formed expression below:

    • SCHED-ENV(get_completion_scheduler<decayed-typeof<set-cpo>>(get_env(sndr)))

    • MAKE-ENV(get_domain, get_domain(get_env(sndr)))

    • empty_env{}

  3. The names let_value, let_error, and let_stopped denote customization point objects. For subexpressions sndr and f, let Sndr be decltype((sndr)), let F be the decayed type of f. If Sndr does not satisfy sender or if F does not satisfy movable-value, the expression let-cpo(sndr, f) is ill-formed. If F does not satisfy invocable, the expression let_stopped(sndr, f) is ill-formed.

  4. Otherwise, the expression let-cpo(sndr, f) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(let-cpo, f, sndr));
    
  5. The exposition-only class template impls-for ([exec.snd.general]) is specialized for let-cpo as follows:

    namespace std::execution {
      template<class State, class Rcvr, class... Args>
      void let-bind(State& state, Rcvr& rcvr, Args&&... args); // exposition only
    
      template<>
      struct impls-for<decayed-typeof<let-cpo>> : default-impls {
        static constexpr auto get-state = see below;
        static constexpr auto complete = see below;
      };
    }
    
    1. Let receiver2 denote the following exposition-only class template:

      namespace std::execution {
        template<class Rcvr, class Env>
        struct receiver2 : Rcvr {
          explicit receiver2(Rcvr rcvr, Env env)
            : Rcvr(std::move(rcvr)), env(std::move(env)) {}
      
          auto get_env() const noexcept {
            const Rcvr& rcvr = *this;
            return JOIN-ENV(env, FWD-ENV(execution::get_env(rcvr)));
          }
      
          Env env; // exposition only
        };
      }
      
    2. impls-for<decayed-typeof<let-cpo>>::get-state is is initialized with a callable object equivalent to the following:

      []<class Sndr, class Rcvr>(Sndr&& sndr, Rcvr& rcvr) requires see below {
        auto&& [tag, data, child] = std::forward<Sndr>(sndr);
        return [&]<class Fn, class Env>(Fn fn, Env env) {
          using args-variant-type = see below;
          using ops2-variant-type = see below;
      
          struct state-type {
            Fn fn;
            Env env;
            args-variant-type args;
            ops2-variant-type ops2;
          };
          return state-type{std::move(fn), std::move(env), {}, {}};
        }(std::forward_like<Sndr>(data), let-env(child));
      }
      
      1. Let Sigs be a pack of the arguments to the completion_signatures specialization named by completion_signatures_of_t<child-type<Sndr>, env_of_t<Rcvr>>. Let LetSigs be a pack of those types in Sigs with a return type of decayed-typeof<set-cpo>. Let as-tuple be an alias template such that as-tuple<Tag(Args...)> denotes the type decayed-tuple<Args...>. Then args-variant-type denotes the type variant<monostate, as-tuple<LetSigs>...>.

      2. Let as-sndr2 be an alias template such that as-sndr2<Tag(Args...)> denotes the type call-result-t<Fn, decay_t<Args>&...>. Then ops2-variant-type denotes the type variant<monostate, connect_result_t<as-sndr2<LetSigs>, receiver2<Rcvr, Env>>...>.

      3. The requires-clause constraining the above lambda is satisfied if and only if the types args-variant-type and ops2-variant-type are well-formed.

    3. The exposition-only function template let-bind is equal to:

      auto& args = state.args.emplace<decayed-tuple<Args...>>(std::forward<Args>(args)...);
      auto sndr2 = apply(std::move(state.fn), args);
      auto rcvr2 = receiver2{std::move(rcvr), std::move(state.env)};
      auto mkop2 = [&] { return connect(std::move(sndr2), std::move(rcvr2)); };
      auto& op2 = state.ops2.emplace<decltype(mkop2())>(emplace-from{mkop2});
      start(op2);
      
    4. impls-for<decayed-typeof<let-cpo>>::complete is is initialized with a callable object equivalent to the following:

      []<class Tag, class... Args>
        (auto, auto& state, auto& rcvr, Tag, Args&&... args) noexcept -> void {
          if constexpr (same_as<Tag, decayed-typeof<set-cpo>>) {
            TRY-EVAL(std::move(rcvr), let-bind(state, rcvr, std::forward<Args>(args)...));
          } else {
            Tag()(std::move(rcvr), std::forward<Args>(args)...);
          }
        }
      
  6. Let sndr and env be subexpressions, and let Sndr be decltype((sndr)). If sender-for<Sndr, decayed-typeof<let-cpo>> is false, then the expression let-cpo.transform_env(sndr, env) is ill-formed. Otherwise, it is equal to JOIN-ENV(let-env(sndr), FWD-ENV(env)).

  7. Let the subexpression out_sndr denote the result of the invocation let-cpo(sndr, f) or an object copied or moved from such, and let the subexpression rcvr denote a receiver such that the expression connect(out_sndr, rcvr) is well-formed. The expression connect(out_sndr, rcvr) has undefined behavior unless it creates an asynchronous operation ([async.ops]) that, when started:

    • invokes f when set-cpo is called with sndr's result datums,

    • makes its completion dependent on the completion of a sender returned by f, and

    • propagates the other completion operations sent by sndr.

34.9.11.8. execution::bulk [exec.bulk]
  1. bulk runs a task repeatedly for every index in an index space.

  2. The name bulk denotes a customization point object. For subexpressions sndr, shape, and f, let Sndr be decltype((sndr)), let Shape be the decayed type of shape, and let F be the decayed type of f. If Sndr does not satisfy sender, or if Shape does not satisfy integral, or if F does not satisfy movable-value, bulk(sndr, shape, f) is ill-formed.

  3. Otherwise, the expression bulk(sndr, shape, f) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(bulk, product-type{shape, f}, sndr));
    
  4. The exposition-only class template impls-for ([exec.snd.general]) is specialized for bulk_t as follows:

    namespace std::execution {
      template<>
      struct impls-for<bulk_t> : default-impls {
        static constexpr auto complete = see below;
      };
    }
    
    1. The member impls-for<bulk_t>::complete is initialized with a callable object equivalent to the following lambda:

      []<class Index, class State, class Rcvr, class Tag, class... Args>
        (Index, State& state, Rcvr& rcvr, Tag, Args&&... args) noexcept -> void requires see below {
          if constexpr (same_as<Tag, set_value_t>) {
            auto& [shape, f] = state;
            constexpr bool nothrow = noexcept(f(auto(shape), args...));
            TRY-EVAL(std::move(rcvr), [&]() noexcept(nothrow) {
              for (auto max = shape, i = 0; i < max; ++i) {
                f(auto(i), args...);
              }
              Tag()(std::move(rcvr), std::forward<Args>(args)...);
            }());
          } else {
            Tag()(std::move(rcvr), std::forward<Args>(args)...);
          }
        }
      
      1. The expression in the requires-clause of the lambda above is true if and only if Tag denotes a type other than set_value_t or if the expression f(auto(shape), args...) is well-formed.

  5. Let the subexpression out_sndr denote the result of the invocation bulk(sndr, shape, f) or an object copied or moved from such, and let the subexpression rcvr denote a receiver such that the expression connect(out_sndr, rcvr) is well-formed. The expression connect(out_sndr, rcvr) has undefined behavior unless it creates an asynchronous operation ([async.ops]) that, when started:

    • on a value completion operation, invokes f(i, args...) for every i of type Shape from 0 to shape, where args is a pack of lvalue subexpressions referring to the value completion result datums of the input sender, and

    • propagates all completion operations sent by sndr.

34.9.11.9. execution::split and execution::ensure_started [exec.split]
  1. split adapts an arbitrary sender into a sender that can be connected multiple times. ensure_started eagerly starts the execution of a sender, returning a sender that is usable as input to additional sender algorithms.

  2. Let shared-env be the type of an environment such that, given an instance env, the expression get_stop_token(env) is well-formed and has type inplace_stop_token.

  3. The names split and ensure_started denote customization point objects. Let the expression shared-cpo be one of split or ensure_started. For a subexpression sndr, let Sndr be decltype((sndr)). If sender_in<Sndr, shared-env> or constructible_from<decay_t<env_of_t<Sndr>>, env_of_t<Sndr>> is false, shared-cpo(sndr) is ill-formed.

    Although it has not yet been approved by LEWG, there is a bug in the current wording that makes it impossible to safely copy the attributes of a sender; it may have reference semantics, leading to a dangling reference. I am striking this part for now and will bring a fix to LEWG.

  4. Otherwise, the expression shared-cpo(sndr) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(shared-cpo, {}, sndr));
    
    • The default implementation of transform_sender will have the effect of connecting the sender to a receiver and, in the case of ensure_started, calling start on the resulting operation state. It will return a sender with a different tag type.

  5. Let local-state denote the following exposition-only class:

    namespace std::execution {
      struct local-state-base {
        virtual ~local-state-base() = default;
        virtual void notify() noexcept = 0;
        virtual void detach() noexcept = 0;
      };
    
      template<class Sndr, class Rcvr>
      struct local-state : local-state-base {
        using on-stop-request = see below;
        using on-stop-callback = stop_token_of_t<env_of_t<Rcvr>>::
                                    template callback_type<on-stop-request>;
    
        local-state(Sndr&& sndr, Rcvr& rcvr) noexcept;
        ~local-state();
    
        void notify() noexcept override;
        void detach() noexcept override;
    
        optional<on-stop-callback> on_stop;
        shared-state<Sndr>* sh_state;
        Rcvr* rcvr;
      };
    }
    
    1. local-state(Sndr&& sndr, Rcvr& rcvr) noexcept;
      1. Effects: Equivalent to:

        auto&& [tag, data, child] = std::forward<Sndr>(sndr);
        this->sh_state = data.sh_state.get();
        this->sh_state->inc-ref();
        this->rcvr = &rcvr;
        
    2. ~local-state();
      1. Effects: Equivalent to:

        detach();
        sh_state->dec-ref();
        
    3. void notify() noexcept override;
      1. Effects: Equivalent to:

        on_stop.reset();
        visit(
          [this]<class Tuple>(Tuple&& tupl) noexcept -> void {
            apply(
              [this](auto tag, auto&... args) noexcept -> void {
                tag(std::move(*rcvr), std::forward_like<Tuple>(args)...);
              },
              tupl);
          },
          QUAL(sh_state->result));
        );
        

        where QUAL is std::move if same_as<tag_of_t<Sndr>, ensure-started-impl-tag> is true, and as_const otherwise.

    4. void detach() noexcept override;
      1. Effects: Equivalent to sh_state->detach() if same_as<tag_of_t<Sndr>, ensure-started-impl-tag> is true; otherwise, nothing.

  6. Let shared-receiver denote the following exposition-only class template:

    namespace std::execution {
      template<class Sndr>
      struct shared-receiver {
        using receiver_concept = receiver_t;
    
        template<class Tag, class... Args>
        void complete(Tag, Args&&... args) noexcept { // exposition only
          try {
            using tuple_t = decayed-tuple<Tag, Args...>;
            sh_state->result.template emplace<tuple_t>(Tag(), std::forward<Args>(args)...);
          } catch (...) {
            using tuple_t = tuple<set_error_t, exception_ptr>;
            sh_state->result.template emplace<tuple_t>(set_error, current_exception());
          }
          sh_state->notify();
        }
    
        template<class... Args>
        void set_value(Args&&... args) && noexcept {
          complete(execution::set_value, std::forward<Args>(args)...);
        }
    
        template<class Error>
        void set_error(Error&& err) && noexcept {
          complete(execution::set_error, std::forward<Error>(err));
        }
    
        void set_stopped() && noexcept {
          complete(execution::set_stopped);
        }
    
        struct env { // exposition only
          shared-state<Sndr>* sh-state; // exposition only
    
          inplace_stop_source query(get_stop_token_t) const noexcept {
            return sh-state->stop_src.get_token();
          }
        };
    
        env get_env() const noexcept {
          return env{sh_state};
        }
    
        shared-state<Sndr>* sh_state;
      };
    }
    
  7. Let shared-state denote the following exposition-only class template:

    namespace std::execution {
      template<class Sndr>
      struct shared-state {
        using variant-type = see below;
        using state-list-type = see below;
        using state-flag-type = see below;
    
        explicit shared-state(Sndr&& sndr);
    
        void start-op() noexcept;
        void notify() noexcept;
        void detach() noexcept;
        void inc-ref() noexcept;
        void dec-ref() noexcept;
    
        inplace_stop_source stop_src{};
        variant-type result{};
        state-list-type waiting_states;
        state-flag-type completed;
        atomic<size_t> ref_count{1};
        connect_result_t<Sndr, shared-receiver<Sndr>> op_state;
      };
    }
    
    1. Let Sigs be a pack of the arguments to the completion_signatures specialization named by completion_signatures_of_t<Sndr>. Let as-tuple be an alias template such that as-tuple<Tag(Args...)> denotes the type decayed-tuple<Tag, Args...>. Then variant-type denotes the type variant<tuple<set_stopped_t>, tuple<set_error_t, exception_ptr>, as-tuple<Sigs>...>, but with duplicate types removed.

    2. Let state-list-type be a type that stores a list of pointers to local-state-base objects and that permits atomic insertion. Let state-flag-type be a type that can be atomically toggled between true and false.

    3.   explicit shared-state(Sndr&& sndr);
      1. Effects: Initializes op_state with the result of connect(std::forward<Sndr>(sndr), shared-receiver{this}).

      2. Postcondition: waiting_states is empty, and completed is false.

    4.   void start-op() noexcept;
      1. Effects: inc-ref(). If stop_src.stop_requested() is true, calls notify(); otherwise, calls start(op_state).

    5.   void notify() noexcept;
      1. Effects: Atomically does the following:

        • Sets completed to true, and

        • Exchanges waiting_states with an empty list, storing the old value in a local prior_states. For each pointer p in prior_states, calls p->notify(). Finally, calls dec-ref().

    6.   void detach() noexcept;
      1. Effects: If completed is false and waiting_states is empty, calls stop_src.request_stop(). This has the effect of requesting early termination of any asynchronous operation that was started as a result of a call to ensure_started, but only if the resulting sender was never connected and started.

    7.   void inc-ref() noexcept;
      1. Effects: Increments ref_count.

    8.   void dec-ref() noexcept;
      1. Effects: Decrements ref_count. If the new value of ref_count is 0, calls delete this.

      2. Synchronization: If dec_ref() does not decrement the ref_count to 0 then synchronizes with the call to dec_ref() that decrements ref_count to 0.

  8. For each type split_t and ensure_started_t, there is a different, associated exposition-only implementation tag type, split-impl-tag and ensure-started-impl-tag, respectively. Let shared-impl-tag be the associated implementation tag type of shared-cpo. Given an expression sndr, the expression shared-cpo.transform_sender(sndr) is equivalent to:

    auto&& [tag, data, child] = sndr;
    auto* sh_state = new shared-state{std::forward_like<decltype((sndr))>(child)};
    return make-sender(shared-impl-tag(), shared-wrapper{sh_state, tag});
    

    where shared-wrapper is an exposition-only class that manages the reference count of the shared-state object pointed to by sh_state. shared-wrapper models movable with move operations nulling out the moved-from object. If tag is split_t, shared-wrapper models copyable with copy operations incrementing the reference count by calling sh_state->inc-ref(). The constructor calls sh_state->start-op() if tag is ensure_started_t. The destructor has no effect if sh_state is null; otherwise, it calls sh_state->detach() if tag is ensure_started_t; and finally, it decrements the reference count by calling sh_state->dec-ref().

  9. The exposition-only class template impls-for ([exec.snd.general]) is specialized for shared-impl-tag as follows:

    namespace std::execution {
      template<>
      struct impls-for<shared-impl-tag> : default-impls {
        static constexpr auto get-state = see below;
        static constexpr auto start = see below;
      };
    }
    
    1. The member impls-for<shared-impl-tag>::get-state is initialized with a callable object equivalent to the following lambda expression:

      []<class Sndr>(Sndr&& sndr, auto& rcvr) noexcept {
        return local-state{std::forward<Sndr>(sndr), rcvr};
      }
      
    2. The member impls-for<shared-impl-tag>::start is initialized with a callable object that has a call operator equivalent to the following:

      template <class Sndr, class Rcvr>
      void operator()(local-state<Sndr, Rcvr>& state, Rcvr& rcvr) const noexcept;
      
      1. Effects:

        1. If state.sh_state->completed is true, calls state.notify() and returns.

        2. Otherwise, calls:

          state.on_stop.emplace(
            get_stop_token(get_env(rcvr)),
            on-stop-request{state.sh_state->stop_src})
          

          If shared-impl-tag is ensure-started-impl-tag, and if state.sh_state->stop_src.stop_requested() is true, calls set_stopped(std::move(rcvr)) and returns.

        3. Then atomically does the following:

          • Inserts &state into state.sh_state->waiting_states, and

          • Reads the value of state.sh_state->completed.

          If the value is true, calls state.notify() and returns.

        4. If shared-impl-tag is split-impl-tag, and if &state is the first item added to state.sh_state->waiting_states, calls state.sh_state->start-op().

  10. Under the following conditions, the results of the child operation are discarded:
    • When a sender returned from ensure_started is destroyed without being connected to a receiver, or

    • If the sender is connected to a receiver but the operation state is destroyed without having been started, or

    • If polling the receiver’s stop token indicates that stop has been requested when start is called, and the operation has not yet completed.

34.9.11.10. execution::when_all [exec.when.all]
  1. when_all and when_all_with_variant both adapt multiple input senders into a sender that completes when all input senders have completed. when_all only accepts senders with a single value completion signature and on success concatenates all the input senders' value result datums into its own value completion operation. when_all_with_variant(sndrs...) is semantically equivalent to when_all(into_variant(sndrs)...), where sndrs is a pack of subexpressions of sender types.

  2. The names when_all and when_all_with_variant denote customization point objects. For some subexpressions sndri..., let Sndri... be decltype((sndri)).... The expressions when_all(sndri...) and when_all_with_variant(sndri...) are ill-formed if any of the following is true:

    • If the number of subexpressions in sndri... is 0, or

    • If any type Sndri does not satisfy sender.

    • If the types of the expressions get-domain-early(sndri) do not share a common type ([meta.trans.other]) for all values of i.

    Otherwise, let CD be the common type of the input senders' domains.

  3. The expression when_all(sndri...) is expression-equivalent to:

    transform_sender(
      CD(),
      make-sender(when_all, {}, sndr0, ... sndrn-1));
    
  4. The exposition-only class template impls-for ([exec.snd.general]) is specialized for when_all_t as follows:

    namespace std::execution {
      template<>
      struct impls-for<when_all_t> : default-impls {
        static constexpr auto get-attrs = see below;
        static constexpr auto get-env = see below;
        static constexpr auto get-state = see below;
        static constexpr auto start = see below;
        static constexpr auto complete = see below;
      };
    }
    
    1. The member impls-for<when_all_t>::get-attrs is initialized with a callable object equivalent to the following lambda expression:

      [](auto&&, auto&&... child) noexcept {
        auto domain_fn = []<class... Ds>(Ds...) noexcept { return common_type_t<Ds...>(); };
        using domain_type = decltype(domain_fn(get-domain-early(child)...));
        if constexpr (same_as<domain_type, default_domain>) {
          return empty_env();
        } else {
          return MAKE-ENV(get_domain, domain_type());
        }
      }
      
    2. The member impls-for<when_all_t>::get-env is initialized with a callable object equivalent to the following lambda expression:

      []<class State, class Rcvr>(auto&&, State& state, const Receiver& rcvr) noexcept {
        return JOIN-ENV(
          MAKE-ENV(get_stop_token, state.stop_src.get_token()), get_env(rcvr));
      }
      
    3. The member impls-for<when_all_t>::get-state is initialized with a callable object equivalent to the following lambda expression:

      BUG: apply isnt constrained
      []<class Sndr, class Rcvr>(Sndr&& sndr, Rcvr& rcvr)
        -> decltype(apply(make-state<Rcvr>{get_env(rcvr)}, std::forward<Sndr>(sndr))) {
        return apply(make-state<Rcvr>{get_env(rcvr)}, std::forward<Sndr>(sndr));
      }
      

      where make-state is the following exposition-only class type:

      template<class Sndr, class Env>
      concept max-1-sender-in = sender_in<Sndr, Env> &&
        (tuple_size_v<value_types_of_t<Sndr, Env, tuple, tuple>> <= 1);
      
      enum class disposition { started, error, stopped };
      
      template <class Rcvr>
      struct make-state {
        const env_of_t<Rcvr>& env;
      
        template <max-1-sender-in<env_of_t<Rcvr>>... Sndrs>
        auto operator()(auto, auto, Sndrs&&... sndrs) const {
          using values_tuple = see below;
          using errors_variant = see below;
          using stop_token = stop_token_of_t<env_of_t<Rcvr>>;
          using stop_callback = stop_token::template callback_type<on-stop-request>;
      
          struct state {
            void arrive(Rcvr& rcvr) noexcept {
              if (0 == --count) {
                complete(rcvr);
              }
            }
      
            void complete(Rcvr& rcvr) noexcept; // see below
      
            atomic<size_t> count{sizeof...(sndrs)};
            inplace_stop_source stop_src{};
            atomic<disposition> disp{disposition::started};
            errors_variant errors{};
            values_tuple values{};
            optional<stop_callback> on_stop{nullopt};
          };
      
          return state{};
        }
      };
      
      1. Let copy-fail be exception_ptr if decay-copying any of the input senders' result datums can potentially throw; otherwise, none-such, where none-such is an unspecified empty class type.

      2. The alias values_tuple denotes the type tuple<value_types_of_t<Sndrs, env_of_t<Rcvr>, decayed-tuple, optional>...> if that type is well-formed; otherwise, tuple<>.

      3. The alias errors_variant denotes the type variant<none-such, copy-fail, Es...> with duplicate types removed, where Es is the pack of the decayed types of all the input senders' possible error result datums.

      4. The member void state::complete(Rcvr& rcvr) noexcept behaves as follows:

        1. If disp is equal to disposition::started, evaluates:

          auto tie = []<class... T>(tuple<T...>& t) noexcept { return tuple<T&...>(t); };
          auto set = [&](auto&... t) noexcept { set_value(std::move(rcvr), std::move(t)...); };
          
          on_stop.reset();
          apply(
            [&](auto&... opts) noexcept {
              apply(set, tuple_cat(tie(*opts)...));
            },
            values);
          
        2. Otherwise, if disp is equal to disposition::error, evaluates:

          on_stop.reset();
          visit(
            [&]<class Error>(Error& error) noexcept {
              if constexpr (!same_as<Error, none-such>) {
                set_error(std::move(rcvr), std::move(error));
              }
            },
            errors);
          
        3. Otherwise, evaluates:

          on_stop.reset();
          set_stopped(std::move(rcvr));
          
    4. The member impls-for<when_all_t>::start is initialized with a callable object equivalent to the following lambda expression:

      []<class State, class Rcvr, class... Ops>(
          State& state, Rcvr& rcvr, Ops&... ops) noexcept -> void {
        state.on_stop.emplace(
          get_stop_token(get_env(rcvr)),
          on-stop-request{state.stop_src});
        if (state.stop_src.stop_requested()) {
          state.on_stop.reset();
          set_stopped(std::move(rcvr));
        } else {
          (start(ops), ...);
        }
      }
      
    5. The member impls-for<when_all_t>::complete is initialized with a callable object equivalent to the following lambda expression:

      []<class Index, class State, class Rcvr, class Set, class... Args>(
          this auto& complete, Index, State& state, Rcvr& rcvr, Set, Args&&... args) noexcept -> void {
        if constexpr (same_as<Set, set_error_t>) {
          if (disposition::error != state.disp.exchange(disposition::error)) {
            state.stop_src.request_stop();
            TRY-EMPLACE-ERROR(state.errors, std::forward<Args>(args)...);
          }
        } else if constexpr (same_as<Set, set_stopped_t>) {
          auto expected = disposition::started;
          if (state.disp.compare_exchange_strong(expected, disposition::stopped)) {
            state.stop_src.request_stop();
          }
        } else if constexpr (!same_as<decltype(State::values), tuple<>>) {
          if (state.disp == disposition::started) {
            auto& opt = get<Index::value>(state.values);
            TRY-EMPLACE-VALUE(complete, opt, std::forward<Args>(args)...);
          }
        }
      
        state.arrive(rcvr);
      }
      

      where TRY-EMPLACE-ERROR(v, e), for subexpressions v and e, is equivalent to:

      try {
        v.template emplace<decltype(auto(e))>(e);
      } catch (...) {
        v.template emplace<exception_ptr>(current_exception());
      }
      

      if the expression decltype(auto(e))(e) is potentially throwing; otherwise, v.template emplace<decltype(auto(e))>(e); and where TRY-EMPLACE-VALUE(c, o, as...), for subexpressions c, o, and pack of subexpressions as, is equivalent to:

      try {
        o.emplace(as...);
      } catch (...) {
        c(Index(), state, rcvr, set_error, current_exception());
        return;
      }
      

      if the expression decayed-tuple<decltype(as)...>{as...} is potentially throwing; otherwise, o.emplace(as...).

  5. The expression when_all_with_variant(sndri...) is expression-equivalent to:

    transform_sender(
      CD(),
      make-sender(when_all_with_variant, {}, sndr0, ... sndrn-1));
    
  6. Given subexpressions sndr and env, if sender-for<decltype((sndr)), when_all_with_variant_t> is false, then the expression when_all_with_variant.transform_sender(sndr, env) is ill-formed; otherwise, the body of the transform_sender member-function is equivalent to:

    auto [tag, data, ...child] = sndr;
    return when_all(into_variant(std::move(child))...);
    

    This causes the when_all_with_variant(sndr...) sender to become when_all(into_variant(sndr)...) when it is connected with a receiver with an execution domain that does not customize when_all_with_variant.

34.9.11.11. execution::into_variant [exec.into.variant]
  1. into_variant adapts a sender with multiple value completion signatures into a sender with just one consisting of a variant of tuples.

  2. The template into-variant-type computes the type sent by a sender returned from into_variant.

    namespace std::execution {
      template<class Sndr, class Env>
          requires sender_in<Sndr, Env>
        using into-variant-type =
          value_types_of_t<Sndr, Env>;
    }
    
  3. The name into_variant denotes a customization point object. For a subexpression sndr, let Sndr be decltype((sndr)). If Sndr does not satisfy sender, into_variant(sndr) is ill-formed.

  4. Otherwise, the expression into_variant(sndr) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(into_variant, {}, sndr))
    
  5. The exposition-only class template impls-for ([exec.snd.general]) is specialized for into_variant as follows:

    namespace std::execution {
      template<>
      struct impls-for<into_variant_t> : default-impls {
        static constexpr auto get-state = see below;
        static constexpr auto complete = see below;
      };
    }
    
    1. The member impls-for<into_variant_t>::get-state is initialized with a callable object equivalent to the following lambda:

      []<class Sndr, class Rcvr>(Sndr&& sndr, Rcvr& rcvr) noexcept
        -> type_identity<into-variant-type<child-type<Sndr>, env_of_t<Rcvr>>> {
        return {};
      }
      
    2. The member impls-for<into_variant_t>::complete is initialized with a callable object equivalent to the following lambda:

      []<class State, class Rcvr, class Tag, class... Args>(
          auto, State, Rcvr& rcvr, Tag, Args&&... args) noexcept {
        if constexpr (same_as<Tag, set_value_t>) {
          using variant_type = typename State::type;
          using tuple_type = decayed-tuple<Args...>;
          try {
            set_value(std::move(rcvr),
                      variant_type(tuple_type(std::forward<Args>(args)...)));
          }
          catch (...) {
            set_error(std::move(rcvr), current_exception());
          }
        } else {
          Tag()(std::move(rcvr), std::forward<Args>(args)...);
        }
      }
      
34.9.11.12. execution::stopped_as_optional [exec.stopped.as.optional]
  1. stopped_as_optional maps an input sender’s stopped completion operation into the value completion operation as an empty optional. The input sender’s value completion operation is also converted into an optional. The result is a sender that never completes with stopped, reporting cancellation by completing with an empty optional.

  2. The name stopped_as_optional denotes a customization point object. For a subexpression sndr, let Sndr be decltype((sndr)). The expression stopped_as_optional(sndr) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(stopped_as_optional, {}, sndr))
    
  3. Let sndr and env be subexpressions such that Sndr is decltype((sndr)) and Env is decltype((env)). If either sender-for<Sndr, stopped_as_optional_t> or single-sender<Sndr, Env> is false then the expression stopped_as_optional.transform_sender(sndr, env) is ill-formed; otherwise, it is equal to:

    auto [tag, data, child] = sndr;
    using V = single-sender-value-type<Sndr, Env>;
    return let_stopped(
        then(std::move(child),
                  []<class T>(T&& t) { return optional(std::forward(t)); }),
        []() noexcept { return just(optional()); });
    
34.9.11.13. execution::stopped_as_error [exec.stopped.as.error]
  1. stopped_as_error maps an input sender’s stopped completion operation into an error completion operation as a custom error type. The result is a sender that never completes with stopped, reporting cancellation by completing with an error.

  2. The name stopped_as_error denotes a customization point object. For some subexpressions sndr and err, let Sndr be decltype((sndr)) and let Err be decltype((err)). If the type Sndr does not satisfy sender or if the type Err doesn’t satisfy movable-value, stopped_as_error(sndr, err) is ill-formed. Otherwise, the expression stopped_as_error(sndr, err) is expression-equivalent to:

    transform_sender(
      get-domain-early(sndr),
      make-sender(stopped_as_error, err, sndr))
    
  3. Let sndr and env be subexpressions such that Sndr is decltype((sndr)) and Env is decltype((env)). If sender-for<Sndr, stopped_as_error_t> is false, then the expression stopped_as_error.transform_sender(sndr, env) is ill-formed; otherwise, it is equal to:

    auto [tag, data, child] = sndr;
    return let_stopped(
        std::move(child),
        [err = std::move(data)]() mutable { return just_error(std::move(err)); });
    

34.9.12. Sender consumers [exec.consumers]

34.9.12.1. execution::start_detached [exec.start.detached]
  1. start_detached eagerly starts a sender without the caller needing to manage the lifetimes of any objects.

  2. The name start_detached denotes a customization point object. For a subexpression sndr, let Sndr be decltype((sndr)). If sender_in<Sndr, empty_env> is false, start_detached is ill-formed. Otherwise, the expression start_detached(sndr) is expression-equivalent to the following except that sndr is evaluated only once:

    apply_sender(get-domain-early(sndr), start_detached, sndr)
    
    • Mandates: same_as<decltype(e), void> is true where e is the expression above.

    If the expression above does not eagerly start the sender sndr after connecting it with a receiver that ignores value and stopped completion operations and calls terminate() on error completions, the behavior of calling start_detached(sndr) is undefined.

  3. Let sndr be a subexpression such that Sndr is decltype((sndr)), and let detached-receiver and detached-operation be the following exposition-only class templates:

    namespace std::execution {
      template<class Sndr>
      struct detached-receiver {
        using receiver_concept = receiver_t;
        detached-operation<Sndr>* op; // exposition only
    
        void set_value() && noexcept { delete op; }
        void set_error() && noexcept { terminate(); }
        void set_stopped() && noexcept { delete op; }
        empty_env get_env() const noexcept { return {}; }
      };
    
      template<class Sndr>
      struct detached-operation {
        connect_result_t<Sndr, detached-receiver<Sndr>> op; // exposition only
    
        explicit detached-operation(Sndr&& sndr)
          : op(connect(std::forward<Sndr>(sndr), detached-receiver<Sndr>{this}))
        {}
      };
    }
    
  4. If sender_to<Sndr, detached-receiver<Sndr>> is false, the expression start_detached.apply_sender(sndr) is ill-formed; otherwise, it is expression-equivalent to start((new detached-operation<Sndr>(sndr))->op).

34.9.12.2. this_thread::sync_wait [exec.sync.wait]
  1. this_thread::sync_wait and this_thread::sync_wait_with_variant are used to block the current thread of execution until the specified sender completes and to return its async result. sync_wait mandates that the input sender has exactly one value completion signature.

  2. Let sync-wait-env be the following exposition-only class type:

    namespace std::this_thread {
      struct sync-wait-env {
        execution::run_loop* loop; // exposition only
    
        auto query(execution::get_scheduler_t) const noexcept { loop->get_scheduler(); }
        auto query(execution::get_delegatee_scheduler_t) const noexcept { loop->get_scheduler(); }
      };
    }
    
  3. Let sync-wait-result-type and sync-wait-with-variant-result-type be exposition-only template aliases defined as follows:

    namespace std::this_thread {
      template<execution::sender_in<sync-wait-env> Sndr>
        using sync-wait-result-type =
          optional<execution::value_types_of_t<Sndr, sync-wait-env, decayed-tuple, type_identity_t>>;
    
      template<execution::sender_in<sync-wait-env> Sndr>
        using sync-wait-with-variant-result-type =
          optional<execution::into-variant-type<Sndr, sync-wait-env>>; // see [exec.into.variant]
    }
    
  4. The name this_thread::sync_wait denotes a customization point object. For a subexpression sndr, let Sndr be decltype((sndr)). If sender_in<Sndr, sync-wait-env> is false, the expression this_thread::sync_wait(sndr) is ill-formed. Otherwise, it is expression-equivalent to the following, except that sndr is evaluated only once:

    apply_sender(get-domain-early(sndr), sync_wait, sndr)
    

    Mandates:

    • The type sync-wait-result-type<Sndr> is well-formed.

    • same_as<decltype(e), sync-wait-result-type<Sndr>> is true, where e is apply_sender the expression above.

  5. Let sync-wait-state and sync-wait-receiver be the following exposition-only class templates:

    namespace std::this_thread {
      template<class Sndr>
      struct sync-wait-state { // exposition only
        execution::run_loop loop;
        exception_ptr error;
        sync-wait-result-type<Sndr> result;
      };
    
      template<class Sndr>
      struct sync-wait-receiver {
        using receiver_concept = execution::receiver_t;
        sync-wait-state<Sndr>* state; // exposition only
    
        template<class... Args>
        void set_value(Args&&... args) && noexcept;
    
        template<class Error>
        void set_error(Error&& err) && noexcept;
    
        void set_stopped() && noexcept;
    
        sync-wait-env get_env() const noexcept { return {&state->loop}; }
      };
    }
    
    1. template<class... Args>
      void set_value(Args&&... args) && noexcept;
      
      1. Effects: Equivalent to:

        try {
          state->result.emplace(std::forward<Args>(args)...);
        } catch (...) {
          state->error = current_exception();
        }
        state->loop.finish();
        
    2. template<class Error>
      void set_error(Error&& err) && noexcept;
      
      1. Effects: Equivalent to:

        state->error = AS-EXCEPT-PTR(std::forward<Error>(err)); // see [exec.general]
        state->loop.finish();
        
    3. template<class Error>
      void set_stopped() && noexcept;
      
      1. Effects: Equivalent to state->loop.finish().

  6. For a subexpression sndr, let Sndr be decltype((sndr)). If sender_to<Sndr, sync-wait-receiver<Sndr>> is false, the expression sync_wait.apply_sender(sndr) is ill-formed; otherwise, it is equivalent to:

    sync-wait-state<Sndr> state;
    auto op = connect(sndr, sync-wait-receiver<Sndr>{&state});
    start(op);
    
    state.loop.run();
    if (state.error) {
      rethrow_exception(std::move(state.error));
    }
    return std::move(state.result);
    
  7. The behavior of this_thread::sync_wait(sndr) is undefined unless:

    1. It blocks the current thread of execution ([defns.block]) with forward progress guarantee delegation ([intro.progress]) until the specified sender completes. The default implementation of sync_wait achieves forward progress guarantee delegation by providing a run_loop scheduler via the get_delegatee_scheduler query on the sync-wait-receiver’s environment. The run_loop is driven by the current thread of execution.

    2. It returns the specified sender’s async results as follows:

      1. For a value completion, the result datums are returned in a tuple in an engaged optional object.

      2. For an error completion, the result datum is rethrown.

      3. For a stopped completion, a disengaged optional object is returned.

  8. The name this_thread::sync_wait_with_variant denotes a customization point object. For a subexpression sndr, let Sndr be the type of into_variant(sndr). If sender_in<Sndr, sync-wait-env> is false, this_thread::sync_wait_with_variant(sndr) is ill-formed. Otherwise, it is expression-equivalent to the following, except sndr is evaluated only once:

    apply_sender(get-domain-early(sndr), sync_wait_with_variant, sndr)
    

    Mandates:

    • The type sync-wait-with-variant-result-type<Sndr> is well-formed.

    • same_as<decltype(e), sync-wait-with-variant-result-type<Sndr>> is true, where e is the expression above.

  9. If callable<sync_wait_t, Sndr> is false, the expression sync_wait_with_variant.apply_sender(sndr) is ill-formed. Otherwise, it is equivalent to:

    using result_type = sync-wait-with-variant-result-type<Sndr>;
    if (auto opt_value = sync_wait(into_variant(sndr))) {
      return result_type(std::move(get<0>(*opt_value)));
    }
    return result_type(nullopt);
    
  10. The behavior of this_thread::sync_wait_with_variant(sndr) is undefined unless:

    1. It blocks the current thread of execution ([defns.block]) with forward progress guarantee delegation ([intro.progress]) until the specified sender completes. The default implementation of sync_wait_with_variant achieves forward progress guarantee delegation by relying on the forward progress guarantee delegation provided by sync_wait.

    2. It returns the specified sender’s async results as follows:

      1. For a value completion, the result datums are returned in an engaged optional object that contains a variant of tuples.

      2. For an error completion, the result datum is rethrown.

      3. For a stopped completion, a disengaged optional object is returned.

34.10. execution::execute [exec.execute]

  1. execute executes a specified callable object on a specified scheduler.

  2. The name execute denotes a customization point object. For some subexpressions sch and f, let Sch be decltype((sch)) and F be the decayed type of f. If Sch does not satisfy scheduler or F does not satisfy invocable, execute(sch, f) is ill-formed. Otherwise, execute(sch, f) is expression-equivalent to:

    apply_sender(
      query-or-default(get_domain, sch, default_domain()),
      execute, schedule(sch), f)
    
    • Mandates: The type of the expression above is void.

  3. For some subexpressions sndr and f where F is the decayed type of f, if F does not satisfy invocable, the expression execute.apply_sender(sndr, f) is ill-formed; otherwise it is expression-equivalent to start_detached(then(sndr, f)).

34.11. Sender/receiver utilities [exec.utils]

34.11.1. execution::completion_signatures [exec.utils.cmplsigs]

  1. completion_signatures is a type that encodes a set of completion signatures ([async.ops]).

  2. [Example:

    class my_sender {
      using completion_signatures =
        completion_signatures<
          set_value_t(),
          set_value_t(int, float),
          set_error_t(exception_ptr),
          set_error_t(error_code),
          set_stopped_t()>;
    };
    
    // Declares my_sender to be a sender that can complete by calling
    // one of the following for a receiver expression rcvr:
    //    set_value(rcvr)
    //    set_value(rcvr, int{...}, float{...})
    //    set_error(rcvr, exception_ptr{...})
    //    set_error(rcvr, error_code{...})
    //    set_stopped(rcvr)
    

    -- end example]

  3. This subclause makes use of the following exposition-only entities:

    template<class Fn>
      concept completion-signature = see below;
    
    template<bool>
      struct indirect-meta-apply {
        template<template<class...> class T, class... As>
          using meta-apply = T<As...>; // exposition only
      };
    
    template<class...>
      concept always-true = true; // exposition only
    
    1. A type Fn satisfies completion-signature if and only if it is a function type with one of the following forms:

      • set_value_t(Vs...), where Vs is an arbitrary parameter pack.

      • set_error_t(Err), where Err is an arbitrary type.

      • set_stopped_t()

    template<class Tag,
              valid-completion-signatures Completions,
              template<class...> class Tuple,
              template<class...> class Variant>
      using gather-signatures = see below;
    
    1. Let Fns... be a template parameter pack of the arguments of the completion_signatures specialization named by Completions, let TagFns be a template parameter pack of the function types in Fns whose return types are Tag, and let Tsn be a template parameter pack of the function argument types in the n-th type in TagFns. Then, given two variadic templates Tuple and Variant, the type gather-signatures<Tag, Completions, Tuple, Variant> names the type META-APPLY(Variant, META-APPLY(Tuple, Ts0...), META-APPLY(Tuple, Ts1...), ... META-APPLY(Tuple, Tsm-1...)), where m is the size of the parameter pack TagFns and META-APPLY(T, As...) is equivalent to:

      typename indirect-meta-apply<always-true<As...>>::template meta-apply<T, As...>;
      
    2. The purpose of META-APPLY is to make it valid to use non-variadic templates as Variant and Tuple arguments to gather-signatures.

  4. namespace std::execution {
      template<completion-signature... Fns>
        struct completion_signatures {};
    
      template<class Sndr,
                class Env = empty_env,
                template<class...> class Tuple = decayed-tuple,
                template<class...> class Variant = variant-or-empty>
          requires sender_in<Sndr, Env>
        using value_types_of_t =
            gather-signatures<set_value_t, completion_signatures_of_t<Sndr, Env>, Tuple, Variant>;
    
      template<class Sndr,
                class Env = empty_env,
                template<class...> class Variant = variant-or-empty>
          requires sender_in<Sndr, Env>
        using error_types_of_t =
            gather-signatures<set_error_t, completion_signatures_of_t<Sndr, Env>, type_identity_t, Variant>;
    
      template<class Sndr, class Env = empty_env>
          requires sender_in<Sndr, Env>
        inline constexpr bool sends_stopped =
            !same_as<
              type-list<>,
              gather-signatures<set_stopped_t, completion_signatures_of_t<Sndr, Env>, type-list, type-list>>;
    }
    

34.11.2. execution::transform_completion_signatures [exec.utils.tfxcmplsigs]

  1. transform_completion_signatures is an alias template used to transform one set of completion signatures into another. It takes a set of completion signatures and several other template arguments that apply modifications to each completion signature in the set to generate a new specialization of completion_signatures.

  2. [Example:

    // Given a sender Sndr and an environment Env, adapt the completion
    // signatures of Sndr by lvalue-ref qualifying the values, adding an additional
    // exception_ptr error completion if its not already there, and leaving the
    // other completion signatures alone.
    template<class... Args>
      using my_set_value_t =
        completion_signatures<
          set_value_t(add_lvalue_reference_t<Args>...)>;
    
    using my_completion_signatures =
      transform_completion_signatures<
        completion_signatures_of_t<Sndr, Env>,
        completion_signatures<set_error_t(exception_ptr)>,
        my_set_value_t>;
    

    -- end example]

  3. This subclause makes use of the following exposition-only entities:

    template<class... As>
      using default-set-value =
        completion_signatures<set_value_t(As...)>;
    
    template<class Err>
      using default-set-error =
        completion_signatures<set_error_t(Err)>;
    
  4. namespace std::execution {
      template<valid-completion-signatures InputSignatures,
              valid-completion-signatures AdditionalSignatures =
                  completion_signatures<>,
              template<class...> class SetValue = default-set-value,
              template<class> class SetError = default-set-error,
              valid-completion-signatures SetStopped =
                  completion_signatures<set_stopped_t()>>
      using transform_completion_signatures =
        completion_signatures<see below>;
    }
    
    1. SetValue shall name an alias template such that for any template parameter pack As..., the type SetValue<As...> is either ill-formed or else valid-completion-signatures<SetValue<As...>> is satisfied.

    2. SetError shall name an alias template such that for any type Err, SetError<Err> is either ill-formed or else valid-completion-signatures<SetError<Err>> is satisfied.

    Then:

    1. Let Vs... be a pack of the types in the type-list named by gather-signatures<set_value_t, InputSignatures, SetValue, type-list>.

    2. Let Es... be a pack of the types in the type-list named by gather-signatures<set_error_t, InputSignatures, type_identity_t, error-list>, where error-list is an alias template such that error-list<Ts...> names type-list<SetError<Ts>...>.

    3. Let Ss name the type completion_signatures<> if gather-signatures<set_stopped_t, InputSignatures, type-list, type-list> is an alias for the type type-list<>; otherwise, SetStopped.

    Then:

    1. If any of the above types are ill-formed, then transform_completion_signatures<InputSignatures, AdditionalSignatures, SetValue, SetError, SetStopped> is ill-formed,

    2. Otherwise, transform_completion_signatures<InputSignatures, AdditionalSignatures, SetValue, SetError, SetStopped> names the type completion_signatures<Sigs...> where Sigs... is the unique set of types in all the template arguments of all the completion_signatures specializations in [AdditionalSignatures, Vs..., Es..., Ss].

34.12. Execution contexts [exec.ctx]

  1. This subclause specifies some execution resources on which work can be scheduled.

34.12.1. run_loop [exec.run.loop]

  1. A run_loop is an execution resource on which work can be scheduled. It maintains a simple, thread-safe first-in-first-out queue of work. Its run() member function removes elements from the queue and executes them in a loop on whatever thread of execution calls run().

  2. A run_loop instance has an associated count that corresponds to the number of work items that are in its queue. Additionally, a run_loop has an associated state that can be one of starting, running, or finishing.

  3. Concurrent invocations of the member functions of run_loop, other than run and its destructor, do not introduce data races. The member functions pop_front, push_back, and finish execute atomically.

  4. Implementations are encouraged to use an intrusive queue of operation states to hold the work units to make scheduling allocation-free.

    namespace std::execution {
      class run_loop {
        // [exec.run.loop.types] Associated types
        class run-loop-scheduler; // exposition only
        class run-loop-sender; // exposition only
        struct run-loop-opstate-base { // exposition only
          virtual void execute() = 0;
          run_loop* loop;
          run-loop-opstate-base* next;
        };
        template<receiver_of<completion_signatures<set_value_t()>> Rcvr>
          using run-loop-opstate = unspecified; // exposition only
    
        // [exec.run.loop.members] Member functions:
        run-loop-opstate-base* pop_front(); // exposition only
        void push_back(run-loop-opstate-base*); // exposition only
    
      public:
        // [exec.run.loop.ctor] construct/copy/destroy
        run_loop() noexcept;
        run_loop(run_loop&&) = delete;
        ~run_loop();
    
        // [exec.run.loop.members] Member functions:
        run-loop-scheduler get_scheduler();
        void run();
        void finish();
      };
    }
    
34.12.1.1. Associated types [exec.run.loop.types]
class run-loop-scheduler;
  1. run-loop-scheduler is an unspecified type that models the scheduler concept.

  2. Instances of run-loop-scheduler remain valid until the end of the lifetime of the run_loop instance from which they were obtained.

  3. Two instances of run-loop-scheduler compare equal if and only if they were obtained from the same run_loop instance.

  4. Let sch be an expression of type run-loop-scheduler. The expression schedule(sch) is not potentially-throwing and has type run-loop-sender.

class run-loop-sender;
  1. run-loop-sender is an unspecified type such that sender-of<run-loop-sender> is true. Additionally, the types reported by its error_types associated type is exception_ptr, and the value of its sends_stopped trait is true.

  2. An instance of run-loop-sender remains valid until the end of the lifetime of its associated run_loop instance.

  3. Let sndr be an expression of type run-loop-sender, let rcvr be an expression such that decltype(rcvr) models the receiver_of concept, and let C be either set_value_t or set_stopped_t. Then:

    • The expression connect(sndr, rcvr) has type run-loop-opstate<decay_t<decltype(rcvr)>> and is potentially-throwing if and only if the initialiation of decay_t<decltype(rcvr)> from rcvr is potentially-throwing.

    • The expression get_completion_scheduler<C>(get_env(sndr)) is not potentially-throwing, has type run-loop-scheduler, and compares equal to the run-loop-scheduler instance from which sndr was obtained.

template<receiver_of<completion_signatures<set_value_t()>> Rcvr>
  struct run-loop-opstate;
  1. run-loop-opstate<Rcvr> inherits unambiguously from run-loop-opstate-base.

  2. Let o be a non-const lvalue of type run-loop-opstate<Rcvr>, and let REC(o) be a non-const lvalue reference to an instance of type Rcvr that was initialized with the expression rcvr passed to the invocation of connect that returned o. Then:

    • The object to which REC(o) refers remains valid for the lifetime of the object to which o refers.

    • The type run-loop-opstate<Rcvr> overrides run-loop-opstate-base::execute() such that o.execute() is equivalent to the following:

      if (get_stop_token(REC(o)).stop_requested()) {
        set_stopped(std::move(REC(o)));
      } else {
        set_value(std::move(REC(o)));
      }
      
    • The expression start(o) is equivalent to the following:

      try {
        o.loop->push_back(&o);
      } catch(...) {
        set_error(std::move(REC(o)), current_exception());
      }
      
34.12.1.2. Constructor and destructor [exec.run.loop.ctor]
run_loop() noexcept;
  1. Postconditions: count is 0 and state is starting.

~run_loop();
  1. Effects: If count is not 0 or if state is running, invokes terminate(). Otherwise, has no effects.

34.12.1.3. Member functions [exec.run.loop.members]
run-loop-opstate-base* pop_front();
  1. Effects: Blocks ([defns.block]) until one of the following conditions is true:

    • count is 0 and state is finishing, in which case pop_front returns nullptr; or

    • count is greater than 0, in which case an item is removed from the front of the queue, count is decremented by 1, and the removed item is returned.

void push_back(run-loop-opstate-base* item);
  1. Effects: Adds item to the back of the queue and increments count by 1.

  2. Synchronization: This operation synchronizes with the pop_front operation that obtains item.

run-loop-scheduler run_loop::get_scheduler();
  1. Returns: an instance of run-loop-scheduler that can be used to schedule work onto this run_loop instance.

void run();
  1. Effects: Equivalent to:

    while (auto* op = pop_front()) {
      op->execute();
    }
    
  2. Precondition: state is starting.

  3. Postcondition: state is finishing.

  4. Remarks: While the loop is executing, state is running. When state changes, it does so without introducing data races.

void finish();
  1. Effects: Changes state to finishing.

  2. Synchronization: This operation synchronizes with all pop_front operations on this object.

34.13. Coroutine utilities [exec.coro.utils]

34.13.1. execution::as_awaitable [exec.as.awaitable]

  1. as_awaitable transforms an object into one that is awaitable within a particular coroutine. This subclause makes use of the following exposition-only entities:

    namespace std::execution {
      template<class Sndr, class Env>
        using single-sender-value-type = see below;
    
      template<class Sndr, class Env>
        concept single-sender =
          sender_in<Sndr, Env> &&
          requires { typename single-sender-value-type<Sndr, Env>; };
    
      template<class Sndr, class Promise>
        concept awaitable-sender =
          single-sender<Sndr, env_of_t> &&
          sender_to<Sndr, awaitable-receiver> && // see below
          requires (Promise& p) {
            { p.unhandled_stopped() } -> convertible_to<coroutine_handle<>>;
          };
    
      template<class Sndr, class Promise>
        class sender-awaitable;
    }
    
    1. Alias template single-sender-value-type is defined as follows:

      1. If value_types_of_t<Sndr, Env, Tuple, Variant> would have the form Variant<Tuple<T>>, then single-sender-value-type<Sndr, Env> is an alias for type decay_t<T>.

      2. Otherwise, if value_types_of_t<Sndr, Env, Tuple, Variant> would have the form Variant<Tuple<>> or Variant<>, then single-sender-value-type<Sndr, Env> is an alias for type void.

      3. Otherwise, if value_types_of_t<Sndr, Env, Tuple, Variant> would have the form Variant<Tuple<Ts...>> where Ts is a parameter pack, then single-sender-value-type<Sndr, Env> is an alias for type std::tuple<decay_t<Ts>...>.

      4. Otherwise, single-sender-value-type<Sndr, Env> is ill-formed.

    2. The type sender-awaitable<Sndr, Promise> is equivalent to the following:

      namespace std::execution {
        template<class Sndr, class Promise>
        class sender-awaitable {
          struct unit {};                                          // exposition only
          using value-type =                                       // exposition only
            single-sender-value-type<Sndr, env_of_t<Promise>>;
          using result-type =                                      // exposition only
            conditional_t<is_void_v<value-type>, unit, value-type>;
          struct awaitable-receiver;                               // exposition only
      
          variant<monostate, result-type, exception_ptr> result{}; // exposition only
          connect_result_t<Sndr, awaitable-receiver> state;        // exposition only
      
        public:
          sender-awaitable(Sndr&& sndr, Promise& p);
          static constexpr bool await_ready() noexcept { return false; }
          void await_suspend(coroutine_handle<Promise>) noexcept { start(state); }
          value-type await_resume();
        };
      }
      
      1. awaitable-receiver is equivalent to the following:

        struct awaitable-receiver {
          using receiver_concept = receiver_t;
          variant<monostate, result-type, exception_ptr>* result-ptr; // exposition only
          coroutine_handle<Promise> continuation;                     // exposition only
          // ... see below
        };
        

        Let rcvr be an rvalue expression of type awaitable-receiver, let crcvr be a const lvalue that refers to rcvr, let vs be a parameter pack of types Vs..., and let err be an arbitrary expression of type Err. Then:

        1. If constructible_from<result-type, Vs...> is satisfied, the expression set_value(rcvr, vs...) is equivalent to:

          try {
            rcvr.result-ptr->emplace<1>(vs...);
          } catch(...) {
            rcvr.result-ptr->emplace<2>(current_exception());
          }
          rcvr.continuation.resume();
          

          Otherwise, set_value(rcvr, vs...) is ill-formed.

        2. The expression set_error(rcvr, err) is equivalent to:

          rcvr.result-ptr->emplace<2>(AS-EXCEPT-PTR(err)); // see [exec.general]
          rcvr.continuation.resume();
          
        3. The expression set_stopped(rcvr) is equivalent to static_cast<coroutine_handle<>>(rcvr.continuation.promise().unhandled_stopped()).resume().

        4. For any expression tag whose type satisfies forwarding-query and for any pack of subexpressions as, get_env(crcvr).query(tag, as...) is expression-equivalent to tag(get_env(as_const(crcvr.continuation.promise())), as...) when that expression is well-formed.

      2. sender-awaitable(Sndr&& sndr, Promise& p)

        • Effects: initializes state with connect(std::forward<Sndr>(sndr), awaitable-receiver{&result, coroutine_handle<Promise>::from_promise(p)}).

      3. value-type await_resume()

        • Effects: equivalent to:

          if (result.index() == 2)
            rethrow_exception(get<2>(result));
          if constexpr (!is_void_v<value-type>)
            return std::forward<value-type>(get<1>(result));
          
  2. as_awaitable is a customization point object. For some subexpressions expr and p where p is an lvalue, Expr names the type decltype((expr)) and Promise names the type decltype((p)), as_awaitable(expr, p) is expression-equivalent to the following:

    1. expr.as_awaitable(p) if that expression is well-formed.

      • Mandates: is-awaitable<A, Promise> is true, where A is the type of the expression above.

    2. Otherwise, expr if is-awaitable<Expr, U> is true, where U is an unspecified class type that lacks a member named await_transform. The condition is not is-awaitable<Expr, Promise> as that creates the potential for constraint recursion.

      • Preconditions: is-awaitable<Expr, Promise> is true and the expression co_await expr in a coroutine with promise type U is expression-equivalent to the same expression in a coroutine with promise type Promise.

    3. Otherwise, sender-awaitable{expr, p} if awaitable-sender<Expr, Promise> is true.

    4. Otherwise, expr.

34.13.2. execution::with_awaitable_senders [exec.with.awaitable.senders]

  1. with_awaitable_senders, when used as the base class of a coroutine promise type, makes senders awaitable in that coroutine type.

    In addition, it provides a default implementation of unhandled_stopped() such that if a sender completes by calling set_stopped, it is treated as if an uncatchable "stopped" exception were thrown from the await-expression. In practice, the coroutine is never resumed, and the unhandled_stopped of the coroutine caller’s promise type is called.

    namespace std::execution {
      template<class-type Promise>
        struct with_awaitable_senders {
          template<OtherPromise>
            requires (!same_as<OtherPromise, void>)
          void set_continuation(coroutine_handle<OtherPromise> h) noexcept;
    
          coroutine_handle<> continuation() const noexcept { return continuation; }
    
          coroutine_handle<> unhandled_stopped() noexcept {
            return stopped-handler(continuation.address());
          }
    
          template<class Value>
          see below await_transform(Value&& value);
    
          private:
          // exposition only
          [[noreturn]] static coroutine_handle<> default_unhandled_stopped(void*) noexcept {
            terminate();
          }
          coroutine_handle<> continuation{}; // exposition only
          // exposition only
          coroutine_handle<> (*stopped-handler)(void*) noexcept = &default_unhandled_stopped;
        };
    }
    
  2. void set_continuation(coroutine_handle<OtherPromise> h) noexcept

    • Effects: equivalent to:

    continuation = h;
    if constexpr ( requires(OtherPromise& other) { other.unhandled_stopped(); } ) {
      stopped-handler = [](void* p) noexcept -> coroutine_handle<> {
        return coroutine_handle<OtherPromise>::from_address(p)
          .promise().unhandled_stopped();
      };
    } else {
      stopped-handler = default_unhandled_stopped;
    }
    
  3. call-result-t<as_awaitable_t, Value, Promise&> await_transform(Value&& value)

    • Effects: equivalent to:

    return as_awaitable(std::forward<Value>(value), static_cast<Promise&>(*this));
    

Index

Terms defined by this specification

References

Informative References

[HPX]
Hartmut Kaiser; et al. HPX - The C++ Standard Library for Parallelism and Concurrency. URL: https://doi.org/10.21105/joss.02352
[N4885]
Thomas Köppe. Working Draft, Standard for Programming Language C++. 17 March 2021. URL: https://wg21.link/n4885
[P0443R14]
Jared Hoberock, Michael Garland, Chris Kohlhoff, Chris Mysen, H. Carter Edwards, Gordon Brown, D. S. Hollman. A Unified Executors Proposal for C++. 15 September 2020. URL: https://wg21.link/p0443r14
[P0981R0]
Richard Smith, Gor Nishanov. Halo: coroutine Heap Allocation eLision Optimization: the joint response. 18 March 2018. URL: https://wg21.link/p0981r0
[P1056R1]
Lewis Baker, Gor Nishanov. Add lazy coroutine (coroutine task) type. 7 October 2018. URL: https://wg21.link/p1056r1
[P1895R0]
Lewis Baker, Eric Niebler, Kirk Shoop. tag_invoke: A general pattern for supporting customisable functions. 8 October 2019. URL: https://wg21.link/p1895r0
[P1897R3]
Lee Howes. Towards C++23 executors: A proposal for an initial set of algorithms. 16 May 2020. URL: https://wg21.link/p1897r3
[P2175R0]
Lewis Baker. Composable cancellation for sender-based async operations. 15 December 2020. URL: https://wg21.link/p2175r0