| |
| ▲ | cpgxiii a day ago | parent | next [-] | | In the case of the Toyota/Denso mess, the code in question had both auto-generated and hand-written elements, including places where the autogenerated code had been modified by hand later. That is the worst place to be, where you no longer have whatever structure and/or guarantees the code gen might provide, but you also don't have the structure and choices that a good SWE team would have to develop that level of complexity by hand. | | |
| ▲ | superxpro12 10 hours ago | parent [-] | | The toyota code was a case of truly abysmal software development methodology. The resultant code they released was so bad that neither NASA, nor Barr, nor Koopman could successfully decipher. (Although Barr posited that the issue was VERY LIKELY in one of a few places with complex multithreaded interactions). Which therein lies the clue. They wrote software that was simply unmaintainable. Autogenerated code isnt any better. |
| |
| ▲ | creato a day ago | parent | prev | next [-] | | This isn't necessarily a problem if you don't consider the output to be "source" code. Assembly is also garbage spaghetti code but that doesn't stop you from using a compiler does it? | |
| ▲ | vodou a day ago | parent | prev | next [-] | | Modern autogenerated C code from Simulink is rather effective. It is neither garbage nor spaghetti, it is just... peculiar. | | |
| ▲ | addaon a day ago | parent [-] | | It’s also much, much more resource intensive (both compute and memory) than what a human would right for the same requirements. | | |
| ▲ | stackghost a day ago | parent [-] | | For control systems like avionics it either passes the suite of tests for certification, or it doesn't. Whether a human could write code that uses less memory is simply not important. In the event the autocode isn't performant enough to run on the box you just spec a faster chip or more memory. | | |
| ▲ | addaon a day ago | parent [-] | | I’m sorry, but I disagree. Building these real-time safety-critical systems is what I do for a living. Once the system is designed and hardware is selected, I agree that if the required tasks fit in the hardware, it’s good to go — there’s no bonus points for leaving memory empty. But the sizing of the system, and even the decomposition of the system to multiple ECUs and the level of integration, depends on how efficient the code is. And there are step functions here — even a decade ago it wasn’t possible to get safety processors with sufficient performance for eVTOL control loops (there’s no “just spec a faster chip”), so the system design needed to deal with lower-ASIL capable hardware and achieve reliability, at the cost of system complexity, at a higher level. Today doing that in a safety processors is possible for hand-written code, but still marginal for autogen code, meaning that if you want to allow for the bloat of code gen you’ll pay for it at the system level. | | |
| ▲ | stackghost a day ago | parent [-] | | >And there are step functions here — even a decade ago it wasn’t possible to get safety processors with sufficient performance for eVTOL control loops (there’s no “just spec a faster chip”) The idea that processors from the last decade were slower than those available today isn't a novel or interesting revelation. All that means is that 10 years ago you had to rely on humans to write the code that today can be done more safely with auto generation. 50+ years of off by ones and use after frees should have disabused us of the hubristic notion that humans can write safe code. We demonstrably can't. In any other problem domain, if our bodies can't do something we use a tool. This is why we invented axes, screwdrivers, and forklifts. But for some reason in software there are people who, despite all evidence to the contrary, cling to the absurd notion that people can write safe code. | | |
| ▲ | addaon a day ago | parent [-] | | > All that means is that 10 years ago you had to rely on humans to write the code that today can be done more safely with auto generation. No. It means more than that. There's a cross-product here. On one axis, you have "resources needed", higher for code gen. On another axis, you have "available hardware safety features." If the higher resources needed for code gen pushes you to fewer hardware safety features available at that performance bucket, then you're stuck with a more complex safety concept, pushing the overall system complexity up. The choice isn't "code gen, with corresponding hopefully better tool safety, and more hardware cost" vs. "hand written code, with human-written bugs that need to be mitigated by test processes, and less hardware cost." It's "code gen, better tool safety, more system complexity, much much larger test matrix for fault injection" vs "human-written code, human-written bugs, but an overall much simpler system." And while it is possible to discuss systems that are so simple that safety processors can be used either way, or systems so complex that non-safety processors must be used either way... in my experience, there are real, interesting, and relevant systems over the past decade that are right on the edge. It's also worth saying that for high-criticality avionics built to DAL B or DAL A via DO-178, the incidence of bugs found in the wild is very, very low. That's accomplished by spending outrageous time (money) on testing, but it's achievable -- defects in real-world avionics systems overwhelming are defects in the requirement specifications, not in the implementation, hand-written or not. | | |
| ▲ | stackghost a day ago | parent [-] | | HN is a very poor platform for good conversations so we'll have to agree to disagree, as I'm not willing to go further in this format | | |
| ▲ | menaerus 18 hours ago | parent [-] | | Codegen from Matlab/Simulink/whatever is good for proof of concept design. It largely helps engineers who are not very good with coding to hypothesize about different algorithmic approaches. Engineers who actually implement that algorithm in a system that will be deployed are coming from a different group with different domain expertise. |
|
|
|
|
|
|
| |
| ▲ | fluorinerocket a day ago | parent | prev | next [-] | | Rockets have flown to orbit on auto coded simulink, seen it myself | |
| ▲ | AnimalMuppet a day ago | parent | prev | next [-] | | > It was probably the reason for Toyotas unintended acceleration glitch. Do you have any evidence for "probably"? | | |
| ▲ | garyfirestorm a day ago | parent [-] | | I know for the fact simulink generates spaghetti and spaghetti code was partially blamed for Toyotas problems. Hence the inference See https://www.safetyresearch.net/toyota-unintended-acceleratio... | | |
| ▲ | CamouflagedKiwi a day ago | parent | next [-] | | That's a nonsensical connection. "Spaghetti code" is a very general term, that's nowhere near specific enough for the two to be related. "I know for a fact that Italian cooks generate spaghetti, and the deceased's last meal contained spaghetti, therefore an Italian chef must have poisoned him" | |
| ▲ | stackghost a day ago | parent | prev [-] | | SRS is a for-profit corporation whose income comes from lawsuits, so their reports/investigations are tainted by their financial incentive to overstate the significance of their findings. |
|
| |
| ▲ | stackghost a day ago | parent | prev [-] | | No I'm not joking at all. The Autocode feature generates code that has high fidelity to the model in simulink, and is immensely more reliable than a human. It is impossible for a simulink model to accidentally type `i > 0` when they meant `i >= 0`, for example. Any human who tells you they have not made this mistake is a liar. Unless there was a second uncommanded acceleration problem with Toyotas, my understanding is that it was caused by poor mechanical design of the accelerator pedal that caused it to get stuck on floor mats. In any case, when we're talking about safety critical control systems like avionics, it's better to abstract away the actual act of typing code into an editor, because it eliminates a potential source of errors. You verify the model at a higher level, and the code is produced in a deterministic manner. | | |
| ▲ | fl7305 a day ago | parent | next [-] | | > It is impossible for a simulink model to accidentally type `i > 0` when they meant `i >= 0` The Simulink Coder tool is a piece of software. It is designed and implemented by humans. It will have bugs. Autogenerated code is different from human written code. It hits soft spots in the C/C++ compilers. For example, autogenerated code can have really huge switch statements. You know, larger than the 15-bit branch offset the compiler implementer thought was big enough to handle any switch-statement any sane human would ever write? So now the switch jumps backwards instead when trying to get the the correct case-statement. I'm not saying that Simulink Coder + a C/C++ compiler is bad. It might be better than the "manual coding" options available. But it's not 100% bug free either. | | |
| ▲ | stackghost a day ago | parent [-] | | >But it's not 100% bug free either. Nobody said it was bug free, and this is a straw man argument of your own construction. Using Autocode completely eliminates certain types of errors that human C programmers have continued to make for more than half a century. |
| |
| ▲ | mmooss a day ago | parent | prev [-] | | > It is impossible for a simulink model to accidentally type `i > 0` when they meant `i >= 0` That's a classic bias: Comparing A and B, show that B doesn't have some A flaws. If they are different systems, of course that's true. But it's also true that A doesn't have some B flaws. That is, what flaws does Autocode have that humans don't? The fantasy that machines are infallible - another (implicit) argument in this thread - is just ignorance for any professional in technology. | | |
| ▲ | coderenegade 21 hours ago | parent [-] | | What's the difference between autogenerated C code and compiling to assembly or machine code? Seems academic to me. The main flaw of autocode is that a human can't easily read and validate it, so you can't really use it as source code. In my experience, this is one of the biggest flaws of these types of systems. You have to version control the file for whatever proprietary graphical programming software generated the code in the first place, and as much as we like to complain about git, it looks like a miracle by comparison. | | |
| ▲ | mmooss 21 hours ago | parent [-] | | > What's the difference between autogenerated C code and compiling to assembly or machine code? Seems academic to me. It's an interesting question and point, but those are two different things and there is no reason to think you'll get the same results. Why not compile from natural language, if that theory is true? | | |
| ▲ | _flux 19 hours ago | parent [-] | | Natural language does not have a specification, while both C and assembly do. | | |
| ▲ | mmooss 18 hours ago | parent [-] | | The C specification is orders of magnitude more complex and is much less defined than assembly. Arguably, the same could be said comparing natural language with C. I admit that's mostly philosphical. But I think saying 'C can autogenerate reliable assembly, therefore a specification can autogenerate reliable C' is also about two different problems. |
|
|
|
|
|
|
