Index targets top-down instead of bottom up#2684
Open
ahoppen wants to merge 1 commit into
Open
Conversation
We originally started preparing and indexing the low-level targets in the hope that we can unblock the files of those targets as early as possible for indexing and thus get best indexing performance. Revisiting this design decision, we can actually get a ~2x speed up (SourceKit-LSP gets indexed in 2:45 instead of 5:49 minutes on my machine) in initial indexing performance by preparing indexing top-level targets first. This speed-up is for two reasons: - When preparing a top-level target, all of its dependencies are implicitly also prepared. That means that we don’t need to send any prepare requests to the build server for those targets, which is very advantages for SwiftPM projects, in which the constant cost in the order of 500ms (based on my memory) even for a null build. - The underlying build server is able to prepare dependencies of the top-level target in parallel instead of implicitly serializing them by preparing all targets in the package in dependency order.
rintaro
reviewed
Jun 3, 2026
| var sortedTargets: [BuildTargetIdentifier] = | ||
| await orLog("Sorting targets") { try await buildServerManager.topologicalSort(of: Array(filesByTarget.keys)) } | ||
| await orLog("Sorting targets") { | ||
| try await buildServerManager.targetsSortedForIndexing(Array(filesByTarget.keys)) |
Member
There was a problem hiding this comment.
If batchSize is greater than 1, the targets end up being partitioned in a rather odd way. I don't believe preparationBatchingStrategy is widely used, but this change puts it in a somewhat awkward position.
Member
Author
There was a problem hiding this comment.
Why do you think it sorts them in an odd way? It shouldn’t be any more odd than it is today, right?
Member
Author
|
@swift-ci Please test |
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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.
We originally started preparing and indexing the low-level targets in the hope that we can unblock the files of those targets as early as possible for indexing and thus get best indexing performance. Revisiting this design decision, we can actually get a ~2x speed up (SourceKit-LSP gets indexed in 2:45 instead of 5:49 minutes on my machine) in initial indexing performance by preparing indexing top-level targets first. This speed-up is for two reasons: