| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| | |
|
| |
|
|
|
|
| |
Now that it doesn't call CircuitAccount::new() it has no error paths,
and clippy demands we remove the Result, so it must once again become
infallible.
|
| |
|
|
|
|
|
|
| |
We foolishly made *two* CircuitAccounts, one of which gets immediately
dropped. But we need to hold onto the account somewhere, because an
mq_queue doesn't keep the account alive.
Otherwise everything breaks when mq tracking is enabled.
|
| | |
|
| |
|
|
|
|
| |
Fixes #1682.
(This involves some noise in the tests.)
|
| |
|
|
|
| |
See
https://gitlab.torproject.org/tpo/core/arti/-/merge_requests/2509#note_3089913
|
| | |
|
| |
|
|
| |
This will save circuits etc. keeping their own clone of the time provider.
|
| |
|
|
|
|
|
|
| |
We use the *channel*'s memquota account. This is arguably wrong, but
it's hard to get right now. See #1652.
Change the type of the queue, and the places it's constructed.
The use sites can all stay the same.
|
| |
|
|
|
| |
This provides a nice place to change in the tests, when we want to
change the way the queue is made.
|
| |
|
|
|
| |
This gets it as far as the outbound circuit->channel mpsc queue creation.
Also, we provide an accessor for it.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
| |
Making a channel is going to involve making a memquota Participant,
which can fail.
|
| |
|
|
|
|
|
| |
The memquota arranagements are going to use this.
We *don't* apply this to Channel (or to other types that aren't
already generic over SleepProvider).
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Having this in the `tor-async-utils` crate prevents us from doing both
of the following without introducing a circular dependency:
* using it in `tor-rtmock` (which we currently do, particularly in
tests).
* using `tor-rtmock` to test things in `tor-async-utils`. We don't do
this yet, but it is generally sensible to do so. In particular we
want to move the `stream_peak` module there, which is currently tested
with `tor-rtmock`.
Moving this into its own crate avoids this circular dependency.
|
| | |
|
| | |
|
| | |
|
| |
|
|
| |
This is where it belongs.
|
| |
|
|
|
|
|
| |
This makes this available for any Sink + Unpin. Which we want because
we're about to wrap our ChannelSender in a Sink wrapper.
It's in the wrong place now; we'll move it in a moment.
|
| |
|
|
|
|
|
|
|
|
| |
This would otherwise shadow the poll_ready method, which is
confusing.
Also this paves the way for making it available for any
Sink + Unpin.
Improve the docs somewhat to explain what this thing actually is.
|
| |
|
|
|
|
|
|
| |
I think this error was in fact always Error::CircuitClosed because it
came from ChannelClosed.into(). Anyway, we shouldn't squash it.
Now this function has semantics identical to Sink::poll_ready, just a
slightly different signature.
|
| |
|
|
| |
Make it clear we're discarding `()`, not an actual value.
|
| |
|
|
|
|
|
|
|
| |
We're going to change this function, but first we are going to make
its behaviour identical to Sink::poll_ready.
This avoids open-coding the call to poll_read on cell_tx.
The error handling is still strange. We'll fix that in a moment.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously ChannelDetails had a double duty: It held elements shared
among the clones of a Channel, and it also held elements shared
between the Channel and the Reactor. But now that Channel doesn't
have to implement Clone, we can more the non-Reactor elements into
Channel itself.
This change may improve cache locality a bit, and should make it a
little easier to follow the channel code.
I've also moved unique_id out of ChannelDetails into Channel _and_
Reactor: it is small, immutable, and used all the time in logging.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously, Channel was a type that you could Clone that implicitly
its state. Now, Channel always appears as an Arc<Channel>.
This change has several benefits:
* It makes the relationship between Channel struct and the
underlying channel more clear.
* It enables Channel to participate in the RPC system,
where everything has to be an Arc<.>
* It enables us to have a Weak<Channel>, if we ever want to.
* It will let us move various members out of ChannelDetails.
We did this change a while ago with ClientCirc.
|
| |
|
|
|
|
|
|
|
|
|
|
| |
This serves three purposes:
* It removes the 'send a cell' method from the channel's public
API. Nothing outside of tor-proto should have to use this.
* It paves the way for giving each circuit a separate handle onto
the channel's send functionality. This will eventually let
the channel multiplex among circuits more intelligently.
* It prepares for the next commit, which will make Channel itself
universally Arc<.>ed.
|
| | |
|
| |
|
|
|
| |
For all mutable shared state, we ought to know which part of the
program sets it, which part of the program reads it, and why.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
This changes the internal representation to be `NonZeroU32` instead of
just `u32`.
Various places where a circuit ID is optional now use `Option<CircId>`.
Fixes a bug in `CircIdRange::sample` that would previously return a
circuit ID of 0, when the rng returned 0x8000_0000 for a low range.
|
| |
|
|
| |
cargo fmt, precisely.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The implementation here is perhaps excessively simple: we put
a `oneshot::Sender` in the `Reactor` object, and a
`Shared<oneshot::Receiver>` in the circuit or channel. When
the reactor is dropped, any copy of the `Shared<Receiver>` will
yield `Err(Cancelled)`.
I'm marking these methods as experimental because I'm not sure I've
thought of all the implications here, and we might want to change
things around.
Down the road, these methods might want to yield a `Result<>`
indicating why the reactor was shut down.
This feature was inspired by a request from Saksham Mittal, and a
felt need while working on !1472.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
| |
Unlike C tor, we treat unrecognized commands as reason to kill off
the connection entirely. That's fine; if we need to add an
unrecognized command in the future, we can use VERSIONS to negotiate
it.
Also, if someday we want this code to support relay channels as
well, we can use some type trickery to have that work too.
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
