 | | Re: Rotation Sensor Values
|
|
Oops. When I copied it, that statement got left out, and I put it bck in in the incoorect place. The original version of the code had the return in the outside loop. I know it is this way because I could turn the rotation sensor slowly, and see the (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Persistent globals
|
|
(...) Wonderful! (Oh, and good morning, Eddie ;) I'm putting some stuff into the HOWTO as we speak- I'll mention that. Luis ---...--- "Summertime... and the living is easy... fish are jumping and the cotton is high... So hush, little baby, baby (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Persistent globals
|
|
(...) I just committed a header file persistent.h with comment on how to use __persistent in your files. I also updated genlds.c to move the persistent data to the right place. Eddie C. Dost ecd@skynet.be (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
(...) Could you please post the compile command the makefile generates to the list, including the error? Like in: ecd@reset:~/lego/min...egOS/demo> make /usr/local/bin/h8300...hi-hms-gcc -O2 -fno-builtin -fomit-frame-pointer -Wall (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
(...) Yes. It will only display two counter readings, then stop! See the "return 0;" statement. It is inside the while (1) loop. Take it out of the loop, and I am shure this one will also work. At least with 0.2.4 it does. (...) Here the return (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Persistent globals
|
|
(...) No. You can not easily see the difference between static and non static data in the linker scripts to decide into which section to put the data. I would recomment a header file "persistent.h" and some documentation about that. We could modify (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
I am using Linux. It is a minimalistic setup, but I have downloaded all of the extra stff that needs to be used for legOS. Before this error happens, legOS.srec is built, which I did not know before. Now I get errors involving the include files in (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
(...) Hmm. DJGPP or Cygwin? I thought it was cygwin, but that sounds like a DJGPP problem. (...) Again, DJGPP or Cygwin? Thanks, Andy- I'm sorry to bug you, but I'd really hate to be responsible for putting out something that I know is broken. Luis (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
I downloaded the tarball, and tried to compile it. when I an in /boot, it tries to do a genlds, and exits with an error. When I try to run genlds, it says Segmentation Error(or fault, I can't remember). I think the only reason that I can use 0.2.3 (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
Ack! You are still having problems with 0.2.4? I thought your problem was with the compiler, and not legOS? <flips through old Andy Gombos posts on lugnet> Is this still the merge-map problem? I figured out the situation, and fixed it in CVS. Sorry (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
I figured you would be glad that I was not running 0.2.4. I cannot get it to compile correctly, so I use 0.2.3 instead. The two full programs are below, in case you wanted to see them in full. Andy ----This one does not work---- #include <dsensor.h> (...) (24 years ago, 25-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
(...) phew :) Had me scared there for a while. I'll probably still take a look at it later- I just didn't want to have caused a bug so soon after starting to apply patches. Luis (...) ---...--- "Summertime... and the living is easy... fish are (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
Hee, Hee. It was 0.2.3, not 0.2.4. Here is the code snippet that setup and read the values. The current(working) code takes out the refresh, and changes the delay to msleep. ds_active(&SENSOR_2); ds_rotation_on(&SENSOR_2); (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Persistent globals
|
|
(...) Do you think that is possible to change the behavier of legOS loading to obtain the expected behavier for static data without the trick? Bye, Paolo. --- "Do not follow where the path my lead Go instead where there is no path and leave a (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: light sensor problem?
|
|
(...) That should be great! :-) (...) use (...) Absolutely True!!!! :-) (...) I'm quite sure that you can do it!!!! :-) And this should be a real step forward!!!! What a beautiful thing to have some time to spend in this section of code... :-) I'm (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: light sensor problem?
|
|
(...) Let me think about how to uncouple this from sys_time, so the code can get higher resolutions. (...) Yes, but even if all active sensors are turned of we are running with a lot of CPU power due to the high sampling speed. If you lower the (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
(...) That's a good question. Is this the latest tarball? If so, there have been some modifications to the refresh loop, and this may indicate a bug in them. In other words, it could be the double sets of lcd_refresh() and not the actual reading (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: Rotation Sensor Values
|
|
I added back in the display code, and left out the lcd_refreshes. Then I got a proper counter. I wonder why the refresh was throwing the reading off. Thanks for all your help, Andy (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | RE: light sensor problem?
|
|
(...) I've used the OC1B interrupt for my servo driver code in pbForth. You could use the OC1B to set the settle time for each reading, and then actually read the results when the interrupt expires. Having a structure to tell you which A/D to read (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
| |
| | Re: light sensor problem?
|
|
(...) :-) It is not the best implementation I've seen... :-) (...) Yes. I hate the delay loop but we I can't imagine how to do it better if we still need maximum speed in reading values from sensors. The routine also change his sampling speed if we (...) (24 years ago, 24-Jul-00, to lugnet.robotics.rcx.legos)
|
