 | | Re: LegOS 0.2.0a multi-threading problem
|
|
Hi, yes, currently, you'll have to make main() your first function. version 0.2.1 will support arbitrary starting offsets. could you report about the stability of repeated program runs? if this works fine, I'll change the ON/OFF behaviour to (...) (25 years ago, 26-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: LegOS 0.2.0a multi-threading problem
|
|
(...) That would be nice, actually. I have only one problem with stability - every once in a while, the "Run" key stops working, and all the other keys as well (don't remember exactly, but I think that's how it is). The program keeps running (...) (25 years ago, 27-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: LegOS 0.2.0a multi-threading problem
|
|
(...) Following up to myself, but here's the error (kernel/program/.c) lines 255-260: case KEY_ONOFF: // works only if no programs are running. if(nb_tasks<=3) { killall(PRIO_HIGHEST); } return 0; ^^^^ This should be 'break'. The way the code is (...) (25 years ago, 27-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: LegOS 0.2.0a multi-threading problem
|
|
(...) If ON/OFF is desired while programs run, just drop the if. If ON/OFF isn't to work while programs run, it has to be this: if(nb_tasks<=3) { killall(PRIO_HIGHEST); return 0; // killall doesn't kill self } break; What does standard firmware do? (...) (25 years ago, 27-Oct-99, to lugnet.robotics.rcx.legos)
|
| |
| | Re: LegOS 0.2.0a multi-threading problem
|
|
(...) I'd say this is the way to leave it, since it seems to be how you meant it in the first place. It's just very annoying to take out the batteries if ON/OFF is pressed by mistake while something is running. (...) Don't know, I've never used (...) (25 years ago, 27-Oct-99, to lugnet.robotics.rcx.legos)
|
