refactor: Remove slot mutex and simplify per blob state #104
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
will probably inline this once it is proven to work well
futures-buffered is reexported by n0-future, as I learned yesterday
...by reverting a change that I did because of some older version of clippy
it is present in global, so now we don't need to impl drop, we have it in on_shutdown
|
Documentation for this PR has been generated and is available at: https://n0-computer.github.io/iroh-blobs/pr/104/docs/iroh_blobs/ Last updated: 2025-08-06T12:21:41Z |
rklaehn
commented
Jul 10, 2025
rklaehn
changed the title
We are just destructuring a Command::Spawn that we have created a few lines before.
matheus23
reviewed
Aug 5, 2025
rklaehn
force-pushed
the
remove-slot-mutex
branch
from
701cb08 to
5b3953d
Compare
matheus23
approved these changes
Aug 6, 2025
matheus23
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Before, the state we kept per hash was a bit convoluted. The entry point was a Slot, which contained a tokio mutex to cover the async loading of an entry state from the metadata db. Then inside that, there was the actual entry state, wrapped in an option to cover the case that we don't have anything about the hash.
This was working, but it was an arc in an arc in an arc, and also led to bugs in the case of trying to export a blob that doesn't exist.
Now all these states are flattened into a single enum, and we can easily define what should happen when we e.g. do an export of an entry that does not exist - return an appropriate io error.
Breaking Changes
None
Notes & open questions
Note: you could remove the Initial and Loading state by using a tokio::sync::OnceCell. But that is a true sync primitive, and we want to use an AtomicRefCell, and also using a OnceLock would come with its own sync primitive (a semaphore) just for init, and then we have our own one for the non-load state changes. I don't think it is that bad...
Note: a lot of changes are to add the wait_idle fn in the rpc protocol, which isn't really related to the changes in the PR but just a way to get rid of some flaky tests.
Change checklist