Java 17: Gradle daemons not shown in "Show Gradle daemons" window (Java 8 and 11 are shown) [361213686]

Fixed

Bug

Status Update

No update yet.

Description

dm...@gmail.com

created issue #1

Aug 21, 2024 02:13PM

####################################################

Please provide all of the following information, otherwise we may not be able to route your bug report.

####################################################


1. Gradle daemons started with Java 17, either JetBrains' own JVM or Temurin / Corretto, are never shown in the "Show Gradle Daemons" window. I have to stop them by killing the process in `htop`.
   On the other hand, Gradle daemons started with Java 8 and 11 are correctly shown, and can be managed by Android Studio.



2. Attach log files from Android Studio
  2A. In the IDE, select the Help..Collect Logs and Diagnostic Data menu option.
  2B. Create a diagnostic report and save it to your local computer.
  2C. Attach the report to this bug using the Add attachments button.

3. If you know what they are, write the steps to reproduce:

   3A. Create a project
   3B. Switch the Gradle JDK to Java 17 from you provider of choise (even JetBrains' internal)
   3C. Sync / Build the project
   3D. Open the "Show Gradle Daemons" window, and see that there are none.

In addition to logs, please attach a screenshot or recording that illustrates the problem.

For more information on how to get your bug routed quickly, see https://developer.android.com/studio/report-bugs.html

Build: AI-241.18034.62.2411.12169540, 202408010325
AS: Koala | 2024.1.1 Patch 2
AI-241.18034.62.2411.12169540, JRE 17.0.11+0-17.0.11b1207.24-11852314x64 JetBrains s.r.o., OS Mac OS X(x86_64) v13.6.9, screens 1440.0x900.0, 1920.0x1080.0
Android Gradle Plugin: 8.4.2
Gradle: 8.8
Gradle JDK: JetBrains Runtime 17.0.11
NDK: from local.properties: (not specified), latest from SDK: 27.0.12077973
CMake: from local.properties: (not specified), latest from SDK: 3.22.1-g37088a8, from PATH: 3.30.2

Captura de pantalla 2024-08-21 a la(s) 10.11.41.png

110 KB

View

Download

DiagnosticsReport20240821-100826.zip

4.4 MB

Download

Comments

sp...@google.com <sp...@google.com> Aug 21, 2024 02:17PM

Assigned to an...@google.com.

je...@google.com <je...@google.com> Aug 27, 2024 06:12PM

Reassigned to cm...@google.com.

cm...@google.com <cm...@google.com> Aug 28, 2024 09:07PM

Reassigned to ml...@google.com.

ub...@gmail.com <ub...@gmail.com> #2Jan 31, 2025 01:50PM

download_sdk.py requires a python module called colorama which I don't have on my mac. Is there a recommended way to get it for naclports?

ml...@google.com <ml...@google.com> #3Feb 17, 2025 02:49PM

gclient should install it during runhooks.

ml...@google.com <ml...@google.com> #4Feb 17, 2025 03:32PM

Oh.. and run it with build_tools/python_wrapper to get those modules.

ub...@gmail.com <ub...@gmail.com> #5Feb 17, 2025 04:14PM

[Empty comment from Monorail migration]

ml...@google.com <ml...@google.com> #6Feb 28, 2025 01:33PM

Marked as fixed.

This bug seems to be a problem with the inliner. In this trace seems to be inlining ~basic_string into ~basic_stringstream; in my repro it's a the same verifier failure but with different functions. It looks like the inliner is failing to update (or create) the inlinedAt field in the MDLocation metadata. The verifier starts with the original MDSubprogram (first output line), finds the function it describes (second output, the function prototye) checks each instruction in that function (3rd output, the load instruction) and follows the chain from its !dbg attachment (4th-6th outputs) lead back to an MDSubprogram describing the same function. Here they don't match.

an...@google.com <an...@google.com> #7Mar 18, 2025 07:54PM

Here's a bugpoint-reduced version of the testcase. bugpoint made the code pretty small but apparently it doesn't know how to remove debug info.