Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

wait for in-flight ops before cloning File #5803

Merged
merged 1 commit into from
Jun 27, 2023

Conversation

djmitche
Copy link
Contributor

If there is an ongoing operation on a file, wait for that to complete before performing the clone in File::try_clone. This avoids a race between the ongoing operation and any subsequent operations performed on the clone.

Fixes: #5759

Motivation

#5759 shows a situation where a file is written to with write_all, and once that completes (Poll::Ready), cloned with try_clone. The file contents are then read back from the clone. In unlucky situations, this read produces an empty result because the read occurs before the write has completed.

Solution

Like most methods on File, the try_clone function should wait until any previous asynchronously-completing operations on the file are complete before actually cloning the file.

No Tests?

The failure fixed by this PR appears very difficult to test:

  • It is a race condition, and one which almost always succeeds (that is, the write almost always completes before the read).
  • It's not well-defined what an operating system will do if a file is duplicated while a write(2) syscall is in progress. Reproducing the race requires that the read happen before the invocation of write(2). In practice, this bug reproduces best on a very busy system, such as in CI, where the worker thread that calls wait(2) is not scheduled immediately.
  • Tests run with a mocked spawn_blocking which does not exhibit this bug.
  • If I was able to produce a test case where the closure passed to spawn_blocking was blocked from executing, then when the fix was applied the test would block indefinitely.

I'm open to suggestions as to how to test this, but perhaps the fix is simple enough to accept without them?

If there is an ongoing operation on a file, wait for that to complete
before performing the clone in `File::try_clone`. This avoids a race
between the ongoing operation and any subsequent operations performed on
the clone.

Fixes: tokio-rs#5759
@Darksonn Darksonn added A-tokio Area: The main tokio crate M-fs Module: tokio/fs labels Jun 27, 2023
@Darksonn Darksonn merged commit 6b076a2 into tokio-rs:master Jun 27, 2023
crapStone pushed a commit to Calciumdibromid/CaBr2 that referenced this pull request Jul 6, 2023
This PR contains the following updates:

| Package | Type | Update | Change |
|---|---|---|---|
| [tokio](https://tokio.rs) ([source](https://github.com/tokio-rs/tokio)) | dependencies | minor | `1.28.2` -> `1.29.1` |
| [tokio](https://tokio.rs) ([source](https://github.com/tokio-rs/tokio)) | dev-dependencies | minor | `1.28.2` -> `1.29.1` |

---

### Release Notes

<details>
<summary>tokio-rs/tokio (tokio)</summary>

### [`v1.29.1`](https://github.com/tokio-rs/tokio/releases/tag/tokio-1.29.1): Tokio v1.29.1

[Compare Source](tokio-rs/tokio@tokio-1.29.0...tokio-1.29.1)

##### Fixed

-   rt: fix nesting two `block_in_place` with a `block_on` between (#&#8203;5837])

#&#8203;5837]: tokio-rs/tokio#5837

### [`v1.29.0`](https://github.com/tokio-rs/tokio/releases/tag/tokio-1.29.0): Tokio v1.29.0

[Compare Source](tokio-rs/tokio@tokio-1.28.2...tokio-1.29.0)

Technically a breaking change, the `Send` implementation is removed from
`runtime::EnterGuard`. This change fixes a bug and should not impact most users.

##### Breaking

-   rt: `EnterGuard` should not be `Send` (#&#8203;5766])

##### Fixed

-   fs: reduce blocking ops in `fs::read_dir` (#&#8203;5653])
-   rt: fix possible starvation (#&#8203;5686], #&#8203;5712])
-   rt: fix stacked borrows issue in `JoinSet` (#&#8203;5693])
-   rt: panic if `EnterGuard` dropped incorrect order (#&#8203;5772])
-   time: do not overflow to signal value (#&#8203;5710])
-   fs: wait for in-flight ops before cloning `File` (#&#8203;5803])

##### Changed

-   rt: reduce time to poll tasks scheduled from outside the runtime (#&#8203;5705], #&#8203;5720])

##### Added

-   net: add uds doc alias for unix sockets (#&#8203;5659])
-   rt: add metric for number of tasks (#&#8203;5628])
-   sync: implement more traits for channel errors (#&#8203;5666])
-   net: add nodelay methods on TcpSocket (#&#8203;5672])
-   sync: add `broadcast::Receiver::blocking_recv` (#&#8203;5690])
-   process: add `raw_arg` method to `Command` (#&#8203;5704])
-   io: support PRIORITY epoll events (#&#8203;5566])
-   task: add `JoinSet::poll_join_next` (#&#8203;5721])
-   net: add support for Redox OS (#&#8203;5790])

##### Unstable

-   rt: add the ability to dump task backtraces (#&#8203;5608], #&#8203;5676], #&#8203;5708], #&#8203;5717])
-   rt: instrument task poll times with a histogram (#&#8203;5685])

#&#8203;5766]: tokio-rs/tokio#5766

#&#8203;5653]: tokio-rs/tokio#5653

#&#8203;5686]: tokio-rs/tokio#5686

#&#8203;5712]: tokio-rs/tokio#5712

#&#8203;5693]: tokio-rs/tokio#5693

#&#8203;5772]: tokio-rs/tokio#5772

#&#8203;5710]: tokio-rs/tokio#5710

#&#8203;5803]: tokio-rs/tokio#5803

#&#8203;5705]: tokio-rs/tokio#5705

#&#8203;5720]: tokio-rs/tokio#5720

#&#8203;5659]: tokio-rs/tokio#5659

#&#8203;5628]: tokio-rs/tokio#5628

#&#8203;5666]: tokio-rs/tokio#5666

#&#8203;5672]: tokio-rs/tokio#5672

#&#8203;5690]: tokio-rs/tokio#5690

#&#8203;5704]: tokio-rs/tokio#5704

#&#8203;5566]: tokio-rs/tokio#5566

#&#8203;5721]: tokio-rs/tokio#5721

#&#8203;5790]: tokio-rs/tokio#5790

#&#8203;5608]: tokio-rs/tokio#5608

#&#8203;5676]: tokio-rs/tokio#5676

#&#8203;5708]: tokio-rs/tokio#5708

#&#8203;5717]: tokio-rs/tokio#5717

#&#8203;5685]: tokio-rs/tokio#5685

</details>

---

### Configuration

📅 **Schedule**: Branch creation - At any time (no schedule defined), Automerge - At any time (no schedule defined).

🚦 **Automerge**: Disabled by config. Please merge this manually once you are satisfied.

♻ **Rebasing**: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 **Ignore**: Close this PR and you won't be reminded about these updates again.

---

 - [ ] <!-- rebase-check -->If you want to rebase/retry this PR, check this box

---

This PR has been generated by [Renovate Bot](https://github.com/renovatebot/renovate).
<!--renovate-debug:eyJjcmVhdGVkSW5WZXIiOiIzNi4wLjAiLCJ1cGRhdGVkSW5WZXIiOiIzNi4wLjAiLCJ0YXJnZXRCcmFuY2giOiJkZXZlbG9wIn0=-->

Co-authored-by: cabr2-bot <cabr2.help@gmail.com>
Reviewed-on: https://codeberg.org/Calciumdibromid/CaBr2/pulls/1958
Reviewed-by: crapStone <crapstone01@gmail.com>
Co-authored-by: Calciumdibromid Bot <cabr2_bot@noreply.codeberg.org>
Co-committed-by: Calciumdibromid Bot <cabr2_bot@noreply.codeberg.org>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
A-tokio Area: The main tokio crate M-fs Module: tokio/fs
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Clone may "miss" ongoing operations
2 participants