 | | problems with legOS 0.2.5 and C++ compiling
|
|
due to makefile used in 0.2.5 version, when i try to make a c++ program it searches for h8300-hitachi-g++ but this executable cross compiler is not present in the file conataining gcc cross compiler and the other tools - i downloaded it from this (...) (23 years ago, 4-May-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Floating ponts ??
|
|
(...) Double check the actual program you are test to make sure that you don't have: p = 0.4 /* Only one = */ This is the number one "c" typo and is easy to overlook Lester been there, done that (23 years ago, 4-May-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Floating ponts ??
|
|
(...) I can't check this out now, but keep e-mailing me until I do. -Kekoa kekoa@graphics.stanford.edu (23 years ago, 4-May-01, to lugnet.robotics.rcx.legos)
|
| |
| | Floating ponts ??
|
|
Hello everybody I am writing a program that uses floats, but I have some problems. If I write /******/ float p; p = 0.3; if(p == 0.4) do something; /******/ The if statement evaluates to true. Is there anything special I have to do to use floats in (...) (23 years ago, 3-May-01, to lugnet.robotics.rcx.legos)
|
| |
| | RE: LegOS bug - firmware unlock string not below CC00h in some circumstances
|
|
(...) Argh! This is so typical of many embedded projects that I have seen go sour over the years. It's due to not using the linker to full advantage. In my firmware, I declare the string in its own .text section (you can do this in inline assembler (...) (23 years ago, 29-Apr-01, to lugnet.robotics.rcx.legos)
|
| |
| | LegOS bug - firmware unlock string not below CC00h in some circumstances
|
|
Just thought I'd better report this... Firmdl3 is unable to download very large statically linked programs (with a .text segment greater than 4c00h bytes). Unlike previous size related bugs, this is *not* caused by firmdl, but rather by kmain.c. (...) (23 years ago, 28-Apr-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: C standard functions with legOS?
|
|
(...) As other people have pointed out, its not possible to use them straight off. I've written some floating point code for my own use to do conversions between rectangular (x,y) and polar (dist, bearing) calculations. The codes not documented I'm (...) (23 years ago, 27-Apr-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: C standard functions with legOS?
|
|
(...) I've been meaning to butt in on this discussion sometime, but I got caught up in things, but better late than never. The source you refer to is probably mine. I wrote a suite of fixed-point math routines for multiplication, division, and (...) (23 years ago, 27-Apr-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: C standard functions with legOS?
|
|
(...) An alternative is using fixed point math. There was some posts time ago, with C source for trigonometric functions. If you are interested, I should have links to them, somewhere (or do a `search' :-). bye Bernardo New e-mail: (...) (23 years ago, 26-Apr-01, to lugnet.robotics.rcx.legos)
|
| |
| | Re: C standard functions with legOS?
|
|
(...) The cross-compiler is set up to not use standard libraries. This is to give more control over the code produced. Also, many standard functions just wouldn't work on the H8. There is a float library (lib/float), which allows use of floating (...) (23 years ago, 25-Apr-01, to lugnet.robotics.rcx.legos)
|
