You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
If sourcekitd or clangd don’t respond to a request for 5 minutes, terminate them and use crash recovery to restore behavior
This should be a last stop-gap measure in case sourcekitd or clangd get stuck, don’t respond to any requests anymore and don’t honor cancellation either. In that case we can restore SourceKit-LSP behavior by killing them and using the crash recovery logic to restore functionality.
rdar://149492159
Copy file name to clipboardExpand all lines: Documentation/Configuration File.md
+1Lines changed: 1 addition & 0 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -58,3 +58,4 @@ The structure of the file is currently not guaranteed to be stable. Options may
58
58
-`swiftPublishDiagnosticsDebounceDuration: number`: The time that `SwiftLanguageService` should wait after an edit before starting to compute diagnostics and sending a `PublishDiagnosticsNotification`.
59
59
-`workDoneProgressDebounceDuration: number`: When a task is started that should be displayed to the client as a work done progress, how many milliseconds to wait before actually starting the work done progress. This prevents flickering of the work done progress in the client for short-lived index tasks which end within this duration.
60
60
-`sourcekitdRequestTimeout: number`: The maximum duration that a sourcekitd request should be allowed to execute before being declared as timed out. In general, editors should cancel requests that they are no longer interested in, but in case editors don't cancel requests, this ensures that a long-running non-cancelled request is not blocking sourcekitd and thus most semantic functionality. In particular, VS Code does not cancel the semantic tokens request, which can cause a long-running AST build that blocks sourcekitd.
61
+
-`semanticServiceRestartTimeout: number`: If a request to sourcekitd or clangd exceeds this timeout, we assume that the semantic service provider is hanging for some reason and won't recover. To restore semantic functionality, we terminate and restart it.
"Already created SourceKitD with plugin paths \(resurrected.pluginPaths?.forLogging), now requesting incompatible plugin paths \(pluginPaths.forLogging)"
@@ -74,8 +74,8 @@ package actor SourceKitDRegistry<SourceKitDType: AnyObject> {
0 commit comments