Author: DheerajK_81 Version: **
Translation - Japanese: ModusToolboxでのビルド時間- KBA226847- Community Translated (JA)
ModusToolbox™ uses a makefile to build the application. The first build might take some time, but subsequent builds will be faster and it only builds files that have changed which is tracked by the makefile. Eclipse provides the option of parallel builds which allows the makefile to spawn a set of parallel builds instead of compiling one file after the other.
You can find this option by right-clicking the project and selecting Properties. In the properties dialog, select C/C++ Build and then select the Behaviour tab.
What do the three options mean?
- Use optimal jobs – “Optimal” will set the number of jobs equal to the number of processors on the system
- Use parallel jobs – will set the number of jobs equal to the value specified by the user
- Use unlimited jobs – will make the builder run as many jobs as possible
Each computer configuration is different and the values for optimal and unlimited may not necessarily make them the best setting. Hence assigning the jobs manually might lead to better performance. Choosing a suitable number of parallel jobs depends on the configuration of the computer and hence will be a matter of trial and error.
Start by setting the number of parallel jobs equal to the number of CPU cores in the box show below:
Then try incrementing the values to find a point where there is not much improvement.
Please note: Increasing the value of parallel jobs can load the CPU and cause the computer to become slow. Therefore, finding the optimum setting is key. This is why using unlimited jobs may not be recommended as it might cause the computer to hang.
For testing the build time, follow these steps:
- (1) Set an appropriate value for parallel jobs
- (2) Right-click project and select Clean ModusToolbox Application
- (3) Right click-project and select Build ModusToolbox Application
- (4) Check the console for build time (for eg: 19:33:30 Build Finished (took 11s.695ms))
Repeat this process by increasing the value of parallel jobs to find the optimal setting.