DataDog/java-profiler
Dec 2024 – Apr 2025
2 Months active
Languages Used
C++Java
Technical Skills
Low-level programmingPerformance profilingSystem programmingError HandlingJNILow-Level Programming
Long Yang
PROFILE
Long Yang
Worked on the DataDog/java-profiler repository to enhance profiling accuracy and system stability through targeted bug fixes in C++ and Java. Addressed stack trace memory pollution by refining C++ header metadata, introducing a macro to retrieve the caller’s program counter, and improving the ASGCT_CallFrame structure for better memory layout and data handling. Further stabilized context initialization by implementing robust error handling for memory allocation failures, adding a boolean success indicator, and introducing a Java-side sentinel to prevent double-free scenarios. Focused on low-level programming, memory management, and JNI integration, these changes improved reliability and reduced the crash surface under memory pressure.
Overall Statistics
Feature vs Bugs
0%Features
Repository Contributions
2Total
Bugs
2
Commits
2
Features
0
Lines of code
136
Activity Months2
Your Network
384 peopleSame Organization
@alibaba-inc.com369
Shared Repositories
15
Brice DutheilMember
Edouard SchweisguthMember
Felix GeisendörferMember
Ivo AnjoMember
Jaroslav BachorikMember
Jaroslav BachorikMember
Datadog Java ProfilerMember
krkMember
MattMember
Work History
April 2025
1 Commits
Apr 1, 20252025-04 Monthly Summary: Focused on stabilizing the DataDog/java-profiler context initialization by fixing a crash and improving memory allocation failure handling. The changes introduce a boolean success indicator and a Java-side sentinel to gracefully handle allocation failures, preventing double-free and ensuring robustness under memory pressure. Delivered measurable improvements in stability and reliability for production profiling.
1 Commits
Apr 1, 20252025-04 Monthly Summary: Focused on stabilizing the DataDog/java-profiler context initialization by fixing a crash and improving memory allocation failure handling. The changes introduce a boolean success indicator and a Java-side sentinel to gracefully handle allocation failures, preventing double-free and ensuring robustness under memory pressure. Delivered measurable improvements in stability and reliability for production profiling.
April 2025
December 2024
1 Commits
Dec 1, 2024December 2024 — DataDog/java-profiler: Targeted maintenance of the Profiling Library focusing on stack-trace integrity and API enhancements. Fixed memory pollution in stack traces by updating copyright and license information across C++ header files, added callerPC() macro to retrieve the caller's program counter, and refined the ASGCT_CallFrame structure to improve memory layout and data handling. The change is committed in 34ff5f09ad34dfbb2b7832a3e144448e10a7b57d (#1089: Fix stack trace memory pollution). Business impact: improved profiling accuracy, reduced memory footprint, and more robust API for downstream integrations.
December 2024
1 Commits
Dec 1, 2024December 2024 — DataDog/java-profiler: Targeted maintenance of the Profiling Library focusing on stack-trace integrity and API enhancements. Fixed memory pollution in stack traces by updating copyright and license information across C++ header files, added callerPC() macro to retrieve the caller's program counter, and refined the ASGCT_CallFrame structure to improve memory layout and data handling. The change is committed in 34ff5f09ad34dfbb2b7832a3e144448e10a7b57d (#1089: Fix stack trace memory pollution). Business impact: improved profiling accuracy, reduced memory footprint, and more robust API for downstream integrations.
Activity
Loading activity data...
Quality Metrics
Correctness85.0%
Maintainability80.0%
Architecture70.0%
Performance70.0%
AI Usage20.0%
Skills & Technologies
Programming Languages
C++Java
Technical Skills
Error HandlingJNILow-Level ProgrammingLow-level programmingMemory ManagementPerformance profilingSystem programming
Repositories Contributed To
Overview of all repositories you've contributed to across your timeline