 | | Re: legOS-0.2.2
|
|
Hi Markus, sorry for the delayed response, but finally i'm back: (...) i found a way that may be somewhat cleaner, by separating the lnp related header files needed by the utilities from the libc stuff: created a directory in legOS/include named (...) (25 years ago, 12-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
Hi, if I am not wrong, I have a problem similar to Martin's: When I load the legOS-0.2.2 kernel to the RCX, the kernel dies leaving "on -" on the LCD. It doesn't accept any key input, only removing the batteries helps. When I nevertheless try to (...) (25 years ago, 14-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
(...) Hello Heiko & all other brave legOS'sers, i think i got some ideas related to my (and maybe someone else's) troubles. gcc 2.95.2 with Markus's patch applied looks like having general problems... I downloaded the legOS kernel contained in the (...) (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
Don't know much about the compiler differences (except that I thought most of the 1.1.2->2.9.5 differences would mainly cause compile-time failures, not run-time) but I have compiled every legOS thing I've done (including 0.2.1) with 1.1.2. I have (...) (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
(...) does this mean legOS will or at least should compile and run without the patch ? if that's true -- what is the actual effect of the patch ? (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
As I understand it, the patch provides support for interrupts, which, at least from my glance over the code, are necessary only for LNP stuff. (Markus, feel free to correct me on this.) So, it will compile with your "old" cross compiler and provide (...) (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
(...) In theory, everything should work with an unpatched cross compiler, too. legOS code checks a preprocessor define, CONF_RCX_COMPILER, I think, to determine whether patched or unpatched compilers are used. The problem is, I don't have an (...) (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Kernel dies after loading (was: Re: legOS-0.2.2)
|
|
(...) A quick grep of the sources show that only a few things (other than lnp, which depends very heavily on interrupts) need the interrupts- task visualization and sound appear to be the two victims. If you don't need these and don't want to worry (...) (25 years ago, 15-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Yippie, heLLo world (was: Re: Kernel dies after loading)
|
|
Hi Luis, i got thes egcs-1.1.2 sources today and rebuilt the cross-compiler and legOS and -- it works now. Thanks a lot for your help !! Could you perhaps give me a starting point to get some understanding of how legOS works internally ? (25 years ago, 16-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | dkey_multi (was: Kernel dies after loading)
|
|
(...) Now I know what makes the RCX freeze: On my RCX , the variable dkey_multi, which stores the status of the keys, always has the first bit (0x1) set. This causes the debounce loop at boot up to hang, because it thinks that the On-Off key is (...) (25 years ago, 17-Nov-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Yippie, heLLo world (was: Re: Kernel dies after loading)
|
|
Well, Markus is probably better qualified, but I'm going to try to get to it eventually. Don't bet on it for anytime soon- I seem to be getting less and less freetime, instead of more :( -Luis (...) ###...### Profanity is the one language that all (...) (25 years ago, 18-Nov-99, to lugnet.robotics.rcx.legos)
|
