Mix-testing: revealing a new class of compiler bugs
A new "mix testing" approach uncovers compiler bugs by compiling test fragments with different compilers. Examples show issues in x86 and Arm architectures, emphasizing the importance of maintaining instruction ordering. Luke Geeson developed a tool to explore compiler combinations, identifying bugs and highlighting the need for clearer guidelines.
Read original article
The blog post discusses a new approach called "mix testing" to uncover a different class of compiler bugs. Unlike traditional "litmus testing," mix testing involves breaking down tests into fragments compiled by different compilers to detect bugs arising from interactions between compilers. The post provides examples of mixing bugs, where different compilation schemes lead to incorrect code generation. One example involves x86 architecture, highlighting the importance of maintaining ordering between instructions. Another example focuses on Arm architecture, showing how mixing compilation schemes can result in reordering issues. The author, Luke Geeson, has developed a tool to explore various compiler combinations and has identified bugs in the process. The post emphasizes the need for clearer guidelines on mixing compilation schemes to prevent such bugs in the future. The work was supported by a research grant and is not endorsed by any mentioned companies.
Related
Optimizing the Roc parser/compiler with data-oriented design
The blog post explores optimizing a parser/compiler with data-oriented design (DoD), comparing Array of Structs and Struct of Arrays for improved performance through memory efficiency and cache utilization. Restructuring data in the Roc compiler showcases enhanced efficiency and performance gains.
Getting 100% code coverage doesn't eliminate bugs
Achieving 100% code coverage doesn't ensure bug-free software. A blog post illustrates this with a critical bug missed despite full coverage, leading to a rocket explosion. It suggests alternative approaches and a 20% coverage minimum.
How GCC and Clang handle statically known undefined behaviour
Discussion on compilers handling statically known undefined behavior (UB) in C code reveals insights into optimizations. Compilers like gcc and clang optimize based on undefined language semantics, potentially crashing programs or ignoring problematic code. UB avoidance is crucial for program predictability and security. Compilers differ in handling UB, with gcc and clang showing variations in crash behavior and warnings. LLVM's 'poison' values allow optimizations despite UB, reflecting diverse compiler approaches. Compiler responses to UB are subjective, influenced by developers and user requirements.
The good, the bad, and the weird (2018)
Trail of Bits delves into "weird machines" in software exploitation, complex code snippets evading security measures. Techniques like Hoare triples and dynamic_casts aid in identifying and preventing exploitation, crucial in evolving security landscapes.
The weirdest QNX bug I've ever encountered
The author encountered a CPU usage bug in a QNX system's 'ps' utility due to a 15-year-old bug. Debugging revealed a race condition, leading to code modifications and a shift towards open-source solutions.
0 commentsRelated
Optimizing the Roc parser/compiler with data-oriented design
The blog post explores optimizing a parser/compiler with data-oriented design (DoD), comparing Array of Structs and Struct of Arrays for improved performance through memory efficiency and cache utilization. Restructuring data in the Roc compiler showcases enhanced efficiency and performance gains.
Getting 100% code coverage doesn't eliminate bugs
Achieving 100% code coverage doesn't ensure bug-free software. A blog post illustrates this with a critical bug missed despite full coverage, leading to a rocket explosion. It suggests alternative approaches and a 20% coverage minimum.
How GCC and Clang handle statically known undefined behaviour
Discussion on compilers handling statically known undefined behavior (UB) in C code reveals insights into optimizations. Compilers like gcc and clang optimize based on undefined language semantics, potentially crashing programs or ignoring problematic code. UB avoidance is crucial for program predictability and security. Compilers differ in handling UB, with gcc and clang showing variations in crash behavior and warnings. LLVM's 'poison' values allow optimizations despite UB, reflecting diverse compiler approaches. Compiler responses to UB are subjective, influenced by developers and user requirements.
The good, the bad, and the weird (2018)
Trail of Bits delves into "weird machines" in software exploitation, complex code snippets evading security measures. Techniques like Hoare triples and dynamic_casts aid in identifying and preventing exploitation, crucial in evolving security landscapes.
The weirdest QNX bug I've ever encountered
The author encountered a CPU usage bug in a QNX system's 'ps' utility due to a 15-year-old bug. Debugging revealed a race condition, leading to code modifications and a shift towards open-source solutions.