This month I want to highlight a blog, It Is Never a Compiler Bug Until It Is, by Russell O’Connor. The author finds a bug in the GCC compiler, and after fixing his code, demonstrates senior engineer behavior by checking for the bug in the rest of his code base. This behavior demonstrates terrific Ownership and Bias for Action by using the opportunity to harden his codebase as problems emerge. While inspecting the codebase takes extra time, and is not guaranteed fruitful, consistently demonstrating this behavior leads to more reliable codebases over time.
http://r6.ca/blog/20200929T023701Z.html
“Under some circumstances GCC 9 and 10 will cause memcmp to return an incorrect value when one of the inputs is statically known array that contains NULL bytes. As I rushed to recompile my computer system using GCC 8, I contemplated the vast consequences of such a bug could be, and pondered how it was possible that computers could function at all.”
Leave a comment