Timeline for Is C++ suitable for embedded systems?
Current License: CC BY-SA 2.5
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 21, 2010 at 0:48 | history | edited | Toby Jaffey | CC BY-SA 2.5 |
c
|
| Jun 17, 2010 at 22:46 | history | edited | Toby Jaffey | CC BY-SA 2.5 |
deleted 11 characters in body
|
| Jun 16, 2010 at 14:34 | comment | added | Kortuk | I really do not use the dynamic memory in C either though. There is nowhere that I have to have it. In long term I have read that it can get very very segmented and start causing problems. I need to have very clearly designed cases for running out of memory, and I need to be able to monitor exactly how much is left. | |
| Jun 16, 2010 at 14:30 | comment | added | Kortuk | I try to argue over how powerful a language is for getting things done. I think C++ gives great leaps and bounds in power for getting things done. With proper design and implementation I see great improvements in how we develop systems, which is what I want, and if it can help others, but just requires a little more training, then I am interested. | |
| Jun 16, 2010 at 14:21 | history | edited | Toby Jaffey | CC BY-SA 2.5 |
deleted 1 characters in body
|
| Jun 16, 2010 at 4:00 | comment | added | Mark | so what your saying is you don't want to use C++, you want to use something between C and C++ (lets just call it C+?). In that case i agree, theres a lot of crap in C++ people use just because its available, not because its optimal. Almost any language is capable of producing good, fast code, its a matter of how its used. Most holy wars over languages are not a result of the languages capabilities but an argument over how easy it is for an idiot to do idiotic things, which is an idiotic argument really :p | |
| Jun 15, 2010 at 18:43 | comment | added | Kortuk | I have done a very large amount of development on computers with C++, I have done almost none on embedded platforms. I have done a very very large amount of development on embedded platforms, and I mean in the range of MSP430s and PICs. This has all been done in C. No libraries or anything. I would like to use C++ on an embedded systems project, there are many good reasons, so far I have not heard reasons that cannot be resolved by using strict rules on what constructs of C++ that can be used. | |
| Jun 15, 2010 at 17:57 | comment | added | J. Polfer | I've done mainly larger-scale x86 SBC development in C++, so I find it useful to hear experiences (positive or negative) in using C++ on platforms with less resources. | |
| Jun 15, 2010 at 16:27 | comment | added | Kortuk | I am not arguing that C cannot have a smaller footprint, but you could have used C++ still and obtained a very similar result for designing for what was just discussed. I think the issue is most OOP programmers are used to systems with dynamic memory and using very inefficient constructs, resulting in completely useless code for lower power systems. | |
| Jun 15, 2010 at 16:22 | history | edited | Toby Jaffey | CC BY-SA 2.5 |
added 351 characters in body; added 21 characters in body
|
| Jun 15, 2010 at 16:20 | comment | added | Kortuk | I agree in reference to small systems, but I think we have different definitions of small systems. When you have 1kB of ROM and well written C code takes all but 1 byte of ROM, that is a small system. | |
| Jun 15, 2010 at 16:12 | history | answered | Toby Jaffey | CC BY-SA 2.5 |