Correct, but, to bring it back to the original point, there's a difference between "sloppy code" and "sloppy code that cascades into unintended acceleration". The fact that it didn't actually cascade isn't a reason to keep writing sloppy code, of course. But such sloppiness also remains a red herring until they can actually find a concrete way that code could have contributed.
An analysis by expert witnesses in the trial found that a small amount of memory corruption could trigger task death and unintended acceleration. The report did not find the cause of the memory corruption, but many software errors can corrupt memory. https://www.safetyresearch.net/Library/BarrSlides_FINAL_SCRU...
Last I heard there was speculation that the CPU was vulnerable to an alpha particle flipped bit error and fixing that was the eventual patch. A one in a million shot happens ten times a year (sic) when you ship ten million cars a year covering many billions of miles.
