Feature: Embedded
processors. Some offer a compiler qualification kit with a proven methodology and resources such as test cases, tools for generating documentation and a framework for running validation tests. Solid Sands has SuperTest, a compiler test and validation suite
for C and C++. It can be used to verify the compiler or library for qualification if there is no qualification kit. For those that are unwilling or unable to do this themselves, Solid Sands’s compiler and library qualification services can save resources as well as time, and provide independent quality control. Qualifying a compiler in-house for safety-critical
applications is now easier; however, a company called Bugseng has introduced a new tool for automated use-case analysis called CerTran. This stage is a critical part of compiler qualification and is certainly complex. Great care is needed to ensure that the compiler conforms to the required safety standard and will perform as expected. Various parameters define the compiler configuration. Basic
settings include the compiler as well as programming language versions. Tese are extremely important, since even a small change can have a large effect on compiler behaviour. Tere are hundreds of option settings changing behaviour, analysis, optimisations, diagnostics, and other effects. Diagnostics and other settings govern linking, treatment of headers, and handling of output files. Once these options are selected and the compiler is qualified, soſtware developers can use only the qualified use case. Compilers are notoriously sensitive to configuration. Te
effects of a small change in one setting can be amplified as the compiler works through the multiple stages involved in generating object code from the input source code. Even a small configuration change can have a huge effect on the generated code and invalidate the qualification. If a change is necessary, the compiler must be re-tested and re-qualified. Selecting the right options at the beginning is critical for the
project. However, manually analysing the compiler use case and determining option settings is difficult, time consuming and carries many risks. Automating this part of the qualification process with CerTran now relieves these risks and accelerates qualification. Additional benefits include greater repeatability and a reduction of manual steps. Once correctly set up, CerTran’s automated analysis is easy
to use and allows developers to perform their own analyses independently. Moreover, automation now makes it feasible any necessary updating of compiler options, re-testing and re-qualifying with the new settings to be included as part of the routine nightly build process. Although automating error-prone manual processes can help
to enhance the robustness and reliability of soſtware for any purpose, it is extremely valuable in safety-critical applications. Automated use-case analysis could revolutionise functional- safety soſtware development and remove many risks of errors, and the high costs and delays incurred to fix them. It is a valuable step forward that significantly eases the challenge of ensuring compliance with functional-safety standards in the automotive and industrial markets.
Figure 1: The transformation of a loop in source code under different compiler optimisation levels. SuperTest’s loop optimisation test suite ensures that both code efficiency and correct behaviour are maintained
Figure 2: SuperTest supports all versions of C and C++, ranging from C90 all the way to C++20. The CerTran integration enhances safety- critical C/C++ development by ensuring compiler and library reliability, whilst enforcing MISRA guidelines
Figure 3: The traceability overview report, shown in the V-model, demonstrates the comprehensive testing coverage achieved by SuperTest, which operates at the level of the compiler specification
www.electronicsworld.co.uk May 2024 17
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48
