Previously, if AFD failed to initialize `polling` would return a custom
I/O error with a string error, containing the formatted version of the
underlying system error. However, this means that information about the
underlying system error is lost to the user.
This commit makes it so the returned `io::Error` wraps a user
inaccessible type: `AfdError`. This `AfdError`, when stringified,
returns a similar error message as what was previously returned. In
addition when `.source()` is used it returns the underlying system
error.
Closes#174
Signed-off-by: John Nunley <dev@notgull.net>
This lint does not take into account destructors.
```
error: field `0` is never read
--> src\iocp\mod.rs:1155:13
|
1155 | Waiting(WaitHandle),
| ------- ^^^^^^^^^^
| |
| field in this variant
|
= note: `-D dead-code` implied by `-D warnings`
= help: to override `-D warnings` add `#[allow(dead_code)]`
help: consider changing the field to be of unit type to suppress this warning while preserving the field numbering, or remove the field
|
1155 | Waiting(()),
| ~~
```
Previous, `Poller::wait` would bubble signal interruption error to the user.
However, this may be unexpected for simple use cases. Thus, this commit makes
it so, if `ErrorKind::Interrupted` is received by the underlying `wait()` call,
it clears the events and tries to wait again.
This also adds a test for this interruption written by @psychon.
Co-Authored-By: Uli Schlachter <psychon@users.noreply.github.com>
Signed-off-by: John Nunley <dev@notgull.net>
Once upon a time, this got a Vec as an argument, but that was replaced
with the Events struct. Thus, this should link to Events and not Vec.
Signed-off-by: Uli Schlachter <psychon@znc.in>
Thus far, our kqueue implementation has been a relatively thin layer on
top of the OS kqueue. However, kqueue doesn't keep track of when the
same source is inserted twice, or when a source that doesn't exist is
removed. In the interest of keeping consistent behavior between backends
this commit adds a system for tracking when sources are inserted.
Closes#151
Signed-off-by: John Nunley <dev@notgull.net>
Added in 272bb11eaf for reasons that are
unclear to me. It serves no purpose aside from disabling the API of the
entire crate, so it's best to remove it.
Signed-off-by: John Nunley <dev@notgull.net>
This adds documentation to add() describing what happens when a source
is registered in multiple pollers. A test is also added to ensure this
behavior.
Signed-off-by: John Nunley <dev@notgull.net>
In some containers, eventfd is not available as it cannot be implemented
securely in some hosts. This commit adds a fallback notifier that uses
a pipe instead of eventfd.
Closes#122
Signed-off-by: John Nunley <dev@notgull.net>
* Add better documentation for the IOCP module
* Extract Events from Poller
This prevents the need to have an intermediate buffer to read events
from, reducing the need for an allocation and a copy. This is a breaking
change.
* Add event extra information
Foundation for more details later on.
* Add PRI and HUP events
* Fix various failing tests
- Make sure that waitable handles interact properly with the new
infrastructure
- Fix failing doctests
* Review comments
- Make set_* take a boolean for the value of the flag
- Make Events !Sync
- Fix visibility modifiers
- Inline more methods
- Use a better strategy for testing
* Move completion packets into the Events buffer
This removes one of the mutexes that we have to lock.
* Review comments
Signed-off-by: John Nunley <dev@notgull.net>
This means that we only depend on rustix, so all of the system calls
are direct. This should make mustang integration trivial.
Signed-off-by: John Nunley <dev@notgull.net>
* Add support for polling waitable handles
* Add a smoke test
* Fix failing tests
* Rebase on latest master
Signed-off-by: John Nunley <dev@notgull.net>
* Update semantics for new system
Signed-off-by: John Nunley <dev@notgull.net>
* Forgot about doctests
Signed-off-by: John Nunley <dev@notgull.net>
---------
Signed-off-by: John Nunley <dev@notgull.net>
* Support for the ESP-IDF framework
* Restore the spans to work with the raw notify fd
* On Linux eventfd needs PollFlags::IN
* Add cargo check for ESP IDF to the CI
---------
Co-authored-by: imarkov <imarkov@vmware.com>
Reimplements the C-based wepoll backend in Rust, using some handwritten code. This PR also implements bindings to the I/O Completion Ports and \Device\Afd APIs. For more information on the latter, see my blog post on the subject: https://notgull.github.io/device-afd/
Note that the IOCP API is wrapped using a `Pin`-oriented "CompletionHandle" system that is relatively brittle. This should be replaced with a better model when one becomes available.
* feat: Expose other kqueue filters
* Fix netbsd/openbsd compilation
* Build MSRV for FreeBsd/OpenBsd in CI
* Only run MSRV BSD builds on Linux
* Change API a little + fix netbsd timer
* Add inlines + move PollerSealed
* rustfmt
* Make filter fields public
* Fix examples
In FreeBSD 12, kevent grew some extra fields. libc currently implements
a FreeBSD 11 ABI, but that will change some day. Tweak the kevent
initialization so it will compile with either version.
In a future release of the `libc` crate, `libc::timespec` will contain
private padding fields on `*-linux-musl` targets and so the struct will
no longer be able to be created using the literal initialization syntax.
Update `TS_ZERO` to create a value by initializing an array of the
correct size to `0` and then transmuting to `libc::timespec`. Update
struct literal use of `libc::timespec` to initialize to `TS_ZERO` and
then manually update the appropriate fields. Also updates a raw syscall
to use the libc function instead as on musl 1.2, it correctly handles
`libc::timespec` values which, in musl 1.2, are always 16 bytes in
length regardless of platform.
John Nunley
Taiki Endo
dependabot[bot]
Uli Schlachter
tison
Al Hoang
David Hotham
John Nunley
ivmarkov
Alan Somers
Wesley Wiser