-
Notifications
You must be signed in to change notification settings - Fork 20
Add changes used in Devoxx Demo #54
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Pull Request Overview
This PR adds configuration changes and scripts used for a Devoxx Demo presentation. The changes primarily involve adjusting TornadoVM runtime settings, modifying model loader configurations, and adding demo-specific scripts and configuration files.
- Removes Timer logging from multiple model loaders to improve performance for demo
- Configures TornadoVM with demo-specific paths and increased GPU memory allocation
- Adds comprehensive argument files and shell scripts for running multiple AI models
Reviewed Changes
Copilot reviewed 7 out of 7 changed files in this pull request and generated 5 comments.
Show a summary per file
File | Description |
---|---|
src/main/java/org/beehive/gpullama3/model/loader/Qwen3ModelLoader.java | Removes Timer.log wrapper from model loading |
src/main/java/org/beehive/gpullama3/model/loader/Phi3ModelLoader.java | Removes Timer.log wrapper and adds TornadoVM debug output |
src/main/java/org/beehive/gpullama3/model/loader/MistralModelLoader.java | Removes Timer.log wrapper from model loading |
set_paths | Comments out TornadoVM path exports |
llama-tornado | Updates TornadoVM configuration with demo-specific settings and increased GPU memory |
argfile | Adds comprehensive JVM argument configuration file for TornadoVM |
all.sh | Adds shell script for running multiple AI models in sequence |
Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.
@Override | ||
public Qwen3 loadModel() { | ||
try (var ignored = Timer.log("Load Qwen3 model")) { | ||
try { |
Copilot
AI
Oct 9, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The try-with-resources statement has been changed to a regular try block, but there's no corresponding resource management. This could lead to resource leaks if exceptions occur during model loading.
Copilot uses AI. Check for mistakes.
@Override | ||
public Phi3 loadModel() { | ||
try (var ignored = Timer.log("Load Phi3 model")) { | ||
try { |
Copilot
AI
Oct 9, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The try-with-resources statement has been changed to a regular try block, but there's no corresponding resource management. This could lead to resource leaks if exceptions occur during model loading.
Copilot uses AI. Check for mistakes.
@Override | ||
public Mistral loadModel() { | ||
try (var ignored = Timer.log("Load Mistral model")) { | ||
try { |
Copilot
AI
Oct 9, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The try-with-resources statement has been changed to a regular try block, but there's no corresponding resource management. This could lead to resource leaks if exceptions occur during model loading.
Copilot uses AI. Check for mistakes.
llama-tornado
Outdated
"-Dtornado.enable.mathOptimizations=false", | ||
"-Dtornado.enable.nativeFunctions=true", | ||
"-Dtornado.loop.interchange=true", | ||
"-Dtornado.loop.interchange=true -Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java", |
Copilot
AI
Oct 9, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This line contains two JVM arguments concatenated in a single string. They should be separated into individual array elements for proper argument parsing.
"-Dtornado.loop.interchange=true -Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java", | |
"-Dtornado.loop.interchange=true", | |
"-Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java", |
Copilot uses AI. Check for mistakes.
"-Dtornado.loop.interchange=true -Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java" | ||
-Dtornado.eventpool.maxwaitevents=32000 | ||
"-Dtornado.opencl.compiler.flags=-cl-denorms-are-zero -cl-no-signed-zeros -cl-finite-math-only" |
Copilot
AI
Oct 9, 2025
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Multiple JVM arguments are concatenated in a single quoted string. Each argument should be on its own line for proper parsing by the JVM.
"-Dtornado.loop.interchange=true -Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java" | |
-Dtornado.eventpool.maxwaitevents=32000 | |
"-Dtornado.opencl.compiler.flags=-cl-denorms-are-zero -cl-no-signed-zeros -cl-finite-math-only" | |
-Dtornado.loop.interchange=true | |
-Dtornado.dump.bytecodes.dir=/home/devoxx2025-demo/java-ai-demos/GPULlama3.java | |
-Dtornado.eventpool.maxwaitevents=32000 | |
-Dtornado.opencl.compiler.flags=-cl-denorms-are-zero | |
-cl-no-signed-zeros | |
-cl-finite-math-only |
Copilot uses AI. Check for mistakes.
No description provided.