IssueTracker

From the log:

2024-08-23 08:24:16,009 [70180461]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 2000 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.getAccessibleRole(Unknown Source)
2024-08-23 08:24:16,423 [70180875]   WARN - #c.i.d.PerformanceWatcherImpl - UI was frozen for 7314ms, details saved to /Users/bhergert/Library/Logs/Google/AndroidStudio2024.1/threadDumps-freeze-20240823-082414-AI-241.18034.62.2412.12169539-ComponentManagerImpl$getComponentAdapter$1.invoke-7sec
2024-08-23 08:24:21,005 [70185457]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 1023 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.accessibilityHitTest(Unknown Source)
2024-08-23 08:24:23,000 [70187452]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 1995 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.getAccessibleRole(Unknown Source)
2024-08-23 08:24:25,005 [70189457]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 2002 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.accessibilityHitTest(Unknown Source)
2024-08-23 08:24:25,256 [70189708]   WARN - #c.i.d.PerformanceWatcherImpl - UI was frozen for 8741ms, details saved to /Users/bhergert/Library/Logs/Google/AndroidStudio2024.1/threadDumps-freeze-20240823-082421-AI-241.18034.62.2412.12169539-AndroidPsiTreeChangeListener.computeModulesToInvalidateAttributeDefinitions-8sec
2024-08-23 08:24:28,006 [70192458]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 1245 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.accessibilityHitTest(Unknown Source)
2024-08-23 08:24:30,003 [70194455]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 1997 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.getAccessibleRole(Unknown Source)
2024-08-23 08:24:32,004 [70196456]   INFO - sun.awt.AWTThreading - InvocationEvent has timed out (awaiting 1999 ms). Originated at java.desktop/sun.lwawt.macosx.CAccessibility.accessibilityHitTest(Unknown Source)
2024-08-23 08:24:33,614 [70198066]   WARN - #c.i.d.PerformanceWatcherImpl - UI was frozen for 8220ms, details saved to /Users/bhergert/Library/Logs/Google/AndroidStudio2024.1/threadDumps-freeze-20240823-082430-AI-241.18034.62.2412.12169539-HashMap.putVal-8sec

Do you still have the thread dumps in ~/Library/Logs/Google/AndroidStudio2024.1/ ? If yes, can you please attach them?

Also, do you have any accessibility features enabled? Because I see various calls to java.desktop/sun.lwawt.macosx.CAccessibility flagged for timeouts.

I'm not sure, I just can't get to reach the CAccessibility.accessibilityHitTest method (running under a debugger). But in any case, it doesn't seem like it is causing the freeze, quite the opposite: it's giving up posting the AWT event because the UI is frozen.

All three freeze reports point to similar stacks in the AWT thread:

        at java.base@17.0.11/java.util.HashSet.add(Unknown Source)
	at com.intellij.util.graph.impl.GraphAlgorithmsImpl.collectOutsRecursively(GraphAlgorithmsImpl.java:141)
	at org.jetbrains.android.util.AndroidUtils.getSetWithBackwardDependencies(AndroidUtils.java:603)
	at com.android.tools.idea.res.AndroidPsiTreeChangeListener.computeModulesToInvalidateAttributeDefinitions(AndroidPsiTreeChangeListener.kt:228)
	at com.android.tools.idea.res.AndroidPsiTreeChangeListener.childrenChanged(AndroidPsiTreeChangeListener.kt:151)
	at com.intellij.psi.impl.PsiManagerImpl.notifyPsiTreeChangeListener(PsiManagerImpl.java:414)
	at com.intellij.psi.impl.PsiManagerImpl.fireEvent(PsiManagerImpl.java:357)
	at com.intellij.psi.impl.PsiManagerImpl.childrenChanged(PsiManagerImpl.java:306)
	at com.intellij.pom.tree.events.impl.ChangeInfoImpl.childrenChanged(ChangeInfoImpl.java:122)
	at com.intellij.pom.tree.events.impl.ChangeInfoImpl.fireEvent(ChangeInfoImpl.java:72)
	at com.intellij.pom.tree.events.impl.TreeChangeImpl.fireEvents(TreeChangeImpl.java:143)
	at com.intellij.pom.tree.events.impl.TreeChangeEventImpl.fireEvents(TreeChangeEventImpl.java:130)
	at com.intellij.pom.wrappers.PsiEventWrapperAspect.update(PsiEventWrapperAspect.java:32)
	at com.intellij.pom.core.impl.PomModelImpl.updateDependentAspects(PomModelImpl.java:167)
[...]

(corresponding internal crash reports.)

There were two more freeze reports in the attached archive (thanks!) which point to https://youtrack.jetbrains.com/issue/KTIJ-27932/GradleBuildRootsManager.loadLinkedRoot-Slow-Operations-freeze, fixed in 2024.2 (Android Studio Ladybug). You may want to try a canary, and see if things improve.

I've submitted a fix for the stack mentioned in #comment6 that will be available in the next canary build.

Thank you for your patience while our engineering team worked to resolve this issue. A fix for this issue is now available in:

• Android Studio Ladybug Feature Drop | 2024.2.2 Canary 3
• Android Gradle Plugin 8.8.0-alpha03

We encourage you to try the latest update.

If you notice further issues or have questions, please file a new bug report.

Thank you for taking the time to submit feedback — we really appreciate it!

Thank you for your patience while our engineering team worked to resolve this issue. A fix for this issue is now available in:

• Android Studio Ladybug Feature Drop | 2024.2.2 Canary 2
• Android Gradle Plugin 8.8.0-alpha02

We encourage you to try the latest update.

If you notice further issues or have questions, please file a new bug report.

Thank you for taking the time to submit feedback — we really appreciate it!

Was a resolution ever found to the MissingValueException? I am seeing the same issue

Not that I know of.

Are you hitting that after trying the workaround from #4?

If so, does using AGP 8.8.0-alpha02 or newer fix the original problem for you?

Are you hitting that after trying the workaround from #4?

No, I am seeing the issue from #8 without having hit the original issue. I've been attempting to debug myself, but as the original reporter said, it can be frustrating to debug configuration internals in gradle.

However I have managed to work around it. As the original reporter said, I believe there is somewhere where AGP is setting a provider value on kapt stub gen KotlinCompile tasks with this that resolves to an empty provider.

I acknowledge that you were unable to find a location where this occurs, so possible this is being performed in kapt library. But the project I am working on uses freeCompilerArgs minimally, and I don't believe we are causing this ourselves.

Avoid the e via the following setup in the applicable areas where freeCompilerArgs are being set (simplified for sharing purposes):

android {
        kotlin {
            compilerOptions {
                jvmTarget.set(JvmTarget.JVM_17)
                try {
                    freeCompilerArgs.appendAll(
                        "-opt-in=kotlin.RequiresOptIn",
                        "-opt-in=kotlinx.coroutines.ExperimentalCoroutinesApi",
                    )
                } catch (e: MissingValueException) {
                    freeCompilerArgs.set(mutableListOf())
                    freeCompilerArgs.appendAll(
                        "-opt-in=kotlin.RequiresOptIn",
                        "-opt-in=kotlinx.coroutines.ExperimentalCoroutinesApi",
                    )
                }
            }
        }
    }

Thanks for the info!

If you use AGP 8.8.0-alpha02 or newer, do you still need the workaround in #13?

Thanks for the info. Any chance you can create a repro project hitting the MissingValueException? I wasn't able to repro it on my side previously.

Any chance you can create a repro project hitting the MissingValueException? I wasn't able to repro it on my side previously.

Apologies, this would not be something I would have the chance to do with my current workload.

I did note though, that when swapping freeCompilerArgs.append/appendAll for freeCompilerArgs.add/addAll the failure re-emerged. This is of interest to me since append* gets removed in Gradle 8.8: https://docs.gradle.org/8.8/userguide/upgrading_version_8.html#changes_to_collection_properties

Update:

The issue did lie in our usage of freeCompilerArgs. Traced the issue back to the following format in root build.gradle.kts. This setup had worked prior to upgrading to 2.0.0. Was able to work around the issue once I was aware of what caused it.

subprojects {
    tasks.withType<org.jetbrains.kotlin.gradle.tasks.KotlinCompile>().configureEach {
        compilerOptions {
            freeCompilerArgs.add(*any arg here*)
        }
    }
}

You should be able to reproduce now with this setup in any project.

Thanks for the update! I filed Issue 376481867 to track.