The trace-cmd command interacts only with the Ftrace tracer that is built inside the Linux kernel. It is a user-space front-end command-line tool for Ftrace. However Ftrace is not enough to analysis performance problem. We need to use systrace and perf tools with the trace-cmd to optimize the performance in android systems. We can store a dataframe object in pandas with the raw data in Ftrace using the trace-cmd, but it is disable in the case of the systrace. So this systrace parser llbrary support a way to store a dataframe with the raw data in systrace, likes the trace-cmd.
The main goals of Systrace Parser are:
- Support analysis of systrace to be taken from android devices.
- Support some dataframes for various analysis.
- Get insights on what's working on the linux scheduler.
- Enables kernel developers to analysis the time delayed points.
- Monitor some tracking points, lock contention and processing core number.
This systrace parser use some pyhton libraries, numpy and pandas. So those should be included in your local computer.
pip install numpy
pip install pandas