Yahoo Groups archive
Thread
2004-11-04 by dasbento@aeiou.pt
Hello,
Somebody can say which boot.s or start.s to use for LPC2129 since I
have problems in putting the interruptions to work. But I don't know
if it is for that reason!!!
Thanks!!!
Regards
Domingos
_________________________________________________________
CEAC Cursos de forma\ufffd\ufffdo profissional - pe\ufffda informa\ufffd\ufffdes aqui.:
http://ceac.online.pt/2004-11-04 by Robert Adsett
At 04:44 PM 11/4/04 +0000, you wrote:
>Somebody can say which boot.s or start.s to use for LPC2129 since I
>have problems in putting the interruptions to work. But I don't know
>if it is for that reason!!!
Well, I know the one in the newlib_lpc support will support interrupts
(although not IAP).
( http://www.aeolusdevelopment.com )
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-04 by dasbento@aeiou.pt
Hello,
I'm using GCC 4.4.1 in linux.
I used de crt0.s and lpc2119.ld of the newlib_lpc but when compiling
doesn't find lnewlib-lpc. This line is in the file lpc2119.ld:
/* Libraries to link against. */
INPUT( -lc -lnewlib-lpc -lc -lgcc )
This is important for working the interrupt??? because I comment this
line and the interrupt working incorrectly!!!!!
Thanks...
Regards
Domingos
Well, I know the one in the newlib_lpc support will support interrupts
(although not IAP).
( http://www.aeolusdevelopment.com )
Robert
_________________________________________________________
CEAC Cursos de forma\ufffd\ufffdo profissional - pe\ufffda informa\ufffd\ufffdes aqui.:
http://ceac.online.pt/2004-11-04 by Robert Adsett
At 07:15 PM 11/4/04 +0000, you wrote:
>I'm using GCC 4.4.1 in linux.
>
>I used de crt0.s and lpc2119.ld of the newlib_lpc but when compiling
>doesn't find lnewlib-lpc. This line is in the file lpc2119.ld:
>
>
>/* Libraries to link against. */
>INPUT( -lc -lnewlib-lpc -lc -lgcc )
>
>This is important for working the interrupt??? because I comment this
>line and the interrupt working incorrectly!!!!!
It shouldn't have too much to do with the interrupts themselves. It might
cause other issues if you have libraries on your command line though. What
that line tells ld to do is (in order)
1 - search through libc to find any references not satisfied.
2 - search through libnewlib-lpc to find any references not already
staisfied (or new references created by libc)
3 - search through libc to find any references created when libnewlib-lpc
statisfied references from step 2
4 - finally attempt to satisfy any leftover references by searching through
libgcc
I hope that was clear.
If, when you comment out the INPUT line, the program works and doesn't
complain about missing newlib-lpc functions (IE you have the I/O and chip
support you asked for there is no problem. You are probably just linking
in another fashion. There are multiple ways to skin this particular cat.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-04 by dasbento@aeiou.pt
I'm not using any libnewlib-lpc.a. A long time I can't work with the
interruptions to move forward with my project. Will be possible to see
the code that is in the link below to know what is wrong??
http://www.lsa.isep.ipp.pt/~dbento/program/
Thanks!
Best Regards
Domingos
_________________________________________________________
CEAC Cursos de forma\ufffd\ufffdo profissional - pe\ufffda informa\ufffd\ufffdes aqui.:
http://ceac.online.pt/2004-11-04 by Robert Adsett
At 08:46 PM 11/4/04 +0000, you wrote:
>I'm not using any libnewlib-lpc.a. A long time I can't work with the
In that case definitely comment out the INPUT line. There is no sense
attempting to link to a library you are not using.
>interruptions to move forward with my project. Will be possible to see
>the code that is in the link below to know what is wrong??
>
> http://www.lsa.isep.ipp.pt/~dbento/program/
What do you mean by 'can't work'? What is or is not happening?
I'll take a look but it will probably be a while until I can get back to
you (others might get back more quickly). In the meantime there are
examples of working interrupt code both in the newlib-lpc code and in the
files forum.
The newlib chip (ie VIC setup, memory setup, pll support etc..) support
routines should be useable without needing the I/O routines so if you just
want to use them that should be quite doable. And both those and the
examples in the files section should provide models.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-04 by Robert Adsett
At 04:03 PM 11/4/04 -0500, you wrote:
>At 08:46 PM 11/4/04 +0000, you wrote:
> >interruptions to move forward with my project. Will be possible to see
> >the code that is in the link below to know what is wrong??
> >
> > http://www.lsa.isep.ipp.pt/~dbento/program/
>
>What do you mean by 'can't work'? What is or is not happening?
>
>I'll take a look but it will probably be a while until I can get back to
>you (others might get back more quickly). In the meantime there are
>examples of working interrupt code both in the newlib-lpc code and in the
>files forum.
I've only done a quick scan but I see an immediate potential trouble
spot. You have no actual interrupt function.
You placed a macro of inline assembly at the beginning of what you expect
to be your interrupt function.
That might work, although I certainly wouldn't place any great expectations
that it would without inspecting the actually output assembly code. Even
if it did work I wouldn't place any great faith in it continuing to work as
either the compiler or other code is changed.
My first advice build a proper interrupt wrapper (and yes do it in
assembly), It's not difficult and that way you will know that it is
correct. I'm sounding a bit like a broken record but there is an example
of such a wrapper in the newlib-lpc code.
The other option would be to upgrade the compiler to get a working
interrupt keyword (me I don't trust compilers to get this 'right', but
that's just me).
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III2004-11-05 by dasbento@aeiou.pt
Hello Robert,
Thanks for you dedication in my problem!!!
What wanted saying is that the program doesn't work well!!
What seems happen is the program to do for some time reset, i.e.,
sometimes makes following configuration as capture the ports including
enter in the interruption routine without existing and later it ends
for blocking.
Remember that writing in the uart is not interruption, is only for
debug...
Today I will try to see more some examples, but the code that I tried
was obtained of examples. Concretely in:
http://hubbard.engr.scu.edu/embedded/arm/armlib/docs/html/index.html
I hope has been clear because my English is not very good.
Regards
Domingos
>What do you mean by 'can't work'? What is or is not happening?
_________________________________________________________
CEAC Cursos de forma\ufffd\ufffdo profissional - pe\ufffda informa\ufffd\ufffdes aqui.:
http://ceac.online.pt/2004-11-05 by Robert Adsett
At 11:10 AM 11/5/04 +0000, you wrote:
>Thanks for you dedication in my problem!!!
You're welcome.
>What wanted saying is that the program doesn't work well!!
>
>What seems happen is the program to do for some time reset, i.e.,
>sometimes makes following configuration as capture the ports including
>enter in the interruption routine without existing and later it ends
>for blocking.
Sometimes you get the interrupt and sometimes you don't, correct?
This leads me back to interrupt entry and exit. If those were incorrect
then you might get this. Have you checked the assembly output of your
interrupt function?
Another possibility to look at is there is a known bug that will cause you
to miss timer capture/compare interrupts if they occur while the status
register is being updated.
One more note you have the following line in your interrupt code,
VICSoftIntClr = (1 << VIC_TIMER);
I don't believe this is actually causing problems, but it is also not doing
anything worthwhile.
>Remember that writing in the uart is not interruption, is only for
>debug...
Thanks, I did manage to figure that out.
>Today I will try to see more some examples, but the code that I tried
>was obtained of examples. Concretely in:
>
>http://hubbard.engr.scu.edu/embedded/arm/armlib/docs/html/index.html
>
>I hope has been clear because my English is not very good.
I've seen a lot worse, I'm not having much trouble. Keep practising and
soon no one will be able to tell.
Robert
" 'Freedom' has no meaning of itself. There are always restrictions,
be they legal, genetic, or physical. If you don't believe me, try to
chew a radio signal. "
Kelvin Throop, III