This adds experimental support for timeouts when accepting sockets through
`TcpAcceptor::accept`. This does not add a separate `accept_timeout` function,
but rather it adds a `set_timeout` function instead. This second function is
intended to be used as a hard deadline after which all accepts will never block
and fail immediately.
This idea was derived from Go's SetDeadline() methods. We do not currently have
a robust time abstraction in the standard library, so I opted to have the
argument be a relative time in millseconds into the future. I believe a more
appropriate argument type is an absolute time, but this concept does not exist
yet (this is also why the function is marked #[experimental]).
The native support is built on select(), similarly to connect_timeout(), and the
green support is based on channel select and a timer.
cc #13523
When a uv_tcp_t is closed in libuv, it will still invoke the pending connect_cb,
and I thought that it would always call it with ECANCELED, but it turns out that
sometimes we'll get a different error code instead. Handle this case by checking
to see if the request's data is NULL and bail out if so (the timeout expired).
This adds a `TcpStream::connect_timeout` function in order to assist opening
connections with a timeout (cc #13523). There isn't really much design space for
this specific operation (unlike timing out normal blocking reads/writes), so I
am fairly confident that this is the correct interface for this function.
The function is marked #[experimental] because it takes a u64 timeout argument,
and the u64 type is likely to change in the future.
Exposing ctpop, ctlz, cttz and bswap as taking signed i8/i16/... is just
exposing the internal LLVM names pointlessly (LLVM doesn't have "signed
integers" or "unsigned integers", it just has sized integer types
with (un)signed *operations*).
These operations are semantically working with raw bytes, which the
unsigned types model better.
There are currently a number of return values from the std::comm methods, not
all of which are necessarily completely expressive:
Sender::try_send(t: T) -> bool
This method currently doesn't transmit back the data `t` if the send fails
due to the other end having disconnected. Additionally, this shares the name
of the synchronous try_send method, but it differs in semantics in that it
only has one failure case, not two (the buffer can never be full).
SyncSender::try_send(t: T) -> TrySendResult<T>
This method accurately conveys all possible information, but it uses a
custom type to the std::comm module with no convenience methods on it.
Additionally, if you want to inspect the result you're forced to import
something from `std::comm`.
SyncSender::send_opt(t: T) -> Option<T>
This method uses Some(T) as an "error value" and None as a "success value",
but almost all other uses of Option<T> have Some/None the other way
Receiver::try_recv(t: T) -> TryRecvResult<T>
Similarly to the synchronous try_send, this custom return type is lacking in
terms of usability (no convenience methods).
With this number of drawbacks in mind, I believed it was time to re-work the
return types of these methods. The new API for the comm module is:
Sender::send(t: T) -> ()
Sender::send_opt(t: T) -> Result<(), T>
SyncSender::send(t: T) -> ()
SyncSender::send_opt(t: T) -> Result<(), T>
SyncSender::try_send(t: T) -> Result<(), TrySendError<T>>
Receiver::recv() -> T
Receiver::recv_opt() -> Result<T, ()>
Receiver::try_recv() -> Result<T, TryRecvError>
The notable changes made are:
* Sender::try_send => Sender::send_opt. This renaming brings the semantics in
line with the SyncSender::send_opt method. An asychronous send only has one
failure case, unlike the synchronous try_send method which has two failure
cases (full/disconnected).
* Sender::send_opt returns the data back to the caller if the send is guaranteed
to fail. This method previously returned `bool`, but then it was unable to
retrieve the data if the data was guaranteed to fail to send. There is still a
race such that when `Ok(())` is returned the data could still fail to be
received, but that's inherent to an asynchronous channel.
* Result is now the basis of all return values. This not only adds lots of
convenience methods to all return values for free, but it also means that you
can inspect the return values with no extra imports (Ok/Err are in the
prelude). Additionally, it's now self documenting when something failed or not
because the return value has "Err" in the name.
Things I'm a little uneasy about:
* The methods send_opt and recv_opt are not returning options, but rather
results. I felt more strongly that Option was the wrong return type than the
_opt prefix was wrong, and I coudn't think of a much better name for these
methods. One possible way to think about them is to read the _opt suffix as
"optionally".
* Result<T, ()> is often better expressed as Option<T>. This is only applicable
to the recv_opt() method, but I thought it would be more consistent for
everything to return Result rather than one method returning an Option.
Despite my two reasons to feel uneasy, I feel much better about the consistency
in return values at this point, and I think the only real open question is if
there's a better suffix for {send,recv}_opt.
Closes#11527
This is something that is plausibly useful, and is provided by libuv. This is
not currently surfaced as part of the `TcpStream` type, but it may possibly
appear in the future. For now only the raw functionality is provided through the
Rtio objects.
This is something that is plausibly useful, and is provided by libuv. This is
not currently surfaced as part of the `TcpStream` type, but it may possibly
appear in the future. For now only the raw functionality is provided through the
Rtio objects.
The green scheduler can optimize its runtime based on this by deciding to not go
to sleep in epoll() if there is no active I/O and there is a task to be stolen.
This is implemented for librustuv by keeping a count of the number of tasks
which are currently homed. If a task is homed, and then performs a blocking I/O
operation, the count will be nonzero while the task is blocked. The homing count
is intentionally 0 when there are I/O handles, but no handles currently blocked.
The reason for this is that epoll() would only be used to wake up the scheduler
anyway.
The crux of this change was to have a `HomingMissile` contain a mutable borrowed
reference back to the `HomeHandle`. The rest of the change was just dealing with
this fallout. This reference is used to decrement the homed handle count in a
HomingMissile's destructor.
Also note that the count maintained is not atomic because all of its
increments/decrements/reads are all on the same I/O thread.
This is part of the overall strategy I would like to take when approaching
issue #11165. The only two I/O objects that reasonably want to be "split" are
the network stream objects. Everything else can be "split" by just creating
another version.
The initial idea I had was the literally split the object into a reader and a
writer half, but that would just introduce lots of clutter with extra interfaces
that were a little unnnecssary, or it would return a ~Reader and a ~Writer which
means you couldn't access things like the remote peer name or local socket name.
The solution I found to be nicer was to just clone the stream itself. The clone
is just a clone of the handle, nothing fancy going on at the kernel level.
Conceptually I found this very easy to wrap my head around (everything else
supports clone()), and it solved the "split" problem at the same time.
The cloning support is pretty specific per platform/lib combination:
* native/win32 - uses some specific WSA apis to clone the SOCKET handle
* native/unix - uses dup() to get another file descriptor
* green/all - This is where things get interesting. When we support full clones
of a handle, this implies that we're allowing simultaneous writes
and reads to happen. It turns out that libuv doesn't support two
simultaneous reads or writes of the same object. It does support
*one* read and *one* write at the same time, however. Some extra
infrastructure was added to just block concurrent writers/readers
until the previous read/write operation was completed.
I've added tests to the tcp/unix modules to make sure that this functionality is
supported everywhere.
The `malloc` family of functions may return a null pointer for a
zero-size allocation, which should not be interpreted as an
out-of-memory error.
If the implementation does not return a null pointer, then handling
this will result in memory savings for zero-size types.
This also switches some code to `malloc_raw` in order to maintain a
centralized point for handling out-of-memory in `rt::global_heap`.
Closes#11634
All tests except for the homing tests are now working again with the
librustuv/libgreen refactoring. The homing-related tests are currently commented
out and now placed in the rustuv::homing module.
I plan on refactoring scheduler pool spawning in order to enable more homing
tests in a future commit.
This reimplements librustuv without using the interfaces provided by the
scheduler in libstd. This solely uses the new Runtime trait in order to
interface with the local task and perform the necessary scheduling operations.
The largest snag in this refactoring is reimplementing homing. The new runtime
trait exposes no concept of "homing" a task or forcibly sending a task to a
remote scheduler (there is no concept of a scheduler). In order to reimplement
homing, the transferrence of tasks is now done at the librustuv level instead of
the scheduler level. This means that all I/O loops now have a concurrent queue
which receives homing messages and requests.
This allows the entire implementation of librustuv to be only dependent on the
runtime trait, severing all dependence of librustuv on the scheduler and related
green-thread functions.
This is all in preparation of the introduction of libgreen and libnative.
At the same time, I also took the liberty of removing all glob imports from
librustuv.
This adds an implementation of the Chase-Lev work-stealing deque to libstd
under std::rt::deque. I've been unable to break the implementation of the deque
itself, and it's not super highly optimized just yet (everything uses a SeqCst
memory ordering).
The major snag in implementing the chase-lev deque is that the buffers used to
store data internally cannot get deallocated back to the OS. In the meantime, a
shared buffer pool (synchronized by a normal mutex) is used to
deallocate/allocate buffers from. This is done in hope of not overcommitting too
much memory. It is in theory possible to eventually free the buffers, but one
must be very careful in doing so.
I was unable to get some good numbers from src/test/bench tests (I don't think
many of them are slamming the work queue that much), but I was able to get some
good numbers from one of my own tests. In a recent rewrite of select::select(),
I found that my implementation was incredibly slow due to contention on the
shared work queue. Upon switching to the parallel deque, I saw the contention
drop to 0 and the runtime go from 1.6s to 0.9s with the most amount of time
spent in libuv awakening the schedulers (plus allocations).
Closes#4877
It turns out that libuv was returning ENOSPC to us in our usage of the
uv_ipX_name functions. It also turns out that there may be an off-by-one in
libuv. For now just add one to the buffer size and handle the return value
correctly.
Closes#10663
The reasons for doing this are:
* The model on which linked failure is based is inherently complex
* The implementation is also very complex, and there are few remaining who
fully understand the implementation
* There are existing race conditions in the core context switching function of
the scheduler, and possibly others.
* It's unclear whether this model of linked failure maps well to a 1:1 threading
model
Linked failure is often a desired aspect of tasks, but we would like to take a
much more conservative approach in re-implementing linked failure if at all.
Closes#8674Closes#8318Closes#8863
These two attributes are no longer useful now that Rust has decided to leave
segmented stacks behind. It is assumed that the rust task's stack is always
large enough to make an FFI call (due to the stack being very large).
There's always the case of stack overflow, however, to consider. This does not
change the behavior of stack overflow in Rust. This is still normally triggered
by the __morestack function and aborts the whole process.
C stack overflow will continue to corrupt the stack, however (as it did before
this commit as well). The future improvement of a guard page at the end of every
rust stack is still unimplemented and is intended to be the mechanism through
which we attempt to detect C stack overflow.
Closes#8822Closes#10155
