 | | | | | |
I have a friend who's not yet able to post, so he wanted me to post a question
for him.
Does anyone have an lnpd executable compiled for Windows (cygwin or not)?
Thanks
Steve
| | | | | | | | | | | | | | | |
| |
|
In lugnet.robotics, Steve Hassenplug wrote:
> I have a friend who's not yet able to post, so he wanted me to post a question
> for him.
>
> Does anyone have an lnpd executable compiled for Windows (cygwin or not)?
>
> Thanks
> Steve
I am the "friend". :) My name is Cortland Starrett. I use a modeling
tool to model applications (robots, medical devices, etc.) using UML.
I then run a "model compiler" which converts the UML into C code. The
C code gets compiled into an executable. This allows me to "program" at
a much higher level of abstraction than C... basically in "executable UML".
My next step is to debug at the UML model level. The model compiler can
generate instrumentation into the C code which traces its execution one
UML state or one UML action language step at a time. I have instrumented
the code to trace through LNP. It is cool to see what is happening inside
the robot as a trace on a completely separate (except for IR) computer.
It works on my Linux box just fine. All the rest of my tools are on the
Windows laptop, though. Thus the need for a Windows lnpd.
Thanks for any help or pointers. And thanks Steve for posting for me
while I waited on my registration.
Cort
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.robotics, Cortland Starrett wrote:
>
> I am the "friend". :) My name is Cortland Starrett. I use a modeling
> tool to model applications (robots, medical devices, etc.) using UML.
> I then run a "model compiler" which converts the UML into C code. The
> C code gets compiled into an executable. This allows me to "program" at
> a much higher level of abstraction than C... basically in "executable UML".
>
> My next step is to debug at the UML model level. The model compiler can
> generate instrumentation into the C code which traces its execution one
> UML state or one UML action language step at a time. I have instrumented
> the code to trace through LNP. It is cool to see what is happening inside
> the robot as a trace on a completely separate (except for IR) computer.
>
> It works on my Linux box just fine. All the rest of my tools are on the
> Windows laptop, though. Thus the need for a Windows lnpd.
>
> Thanks for any help or pointers. And thanks Steve for posting for me
> while I waited on my registration.
>
> Cort
Cort,
I know you are looking for an executable but have your tried building the
source?
(http://sourceforge.net/project/showfiles.php?group_id=58151&package_id=141184).
If I get a chance I will check my old computer tomorrow to see if I have a
binary. You might also check with Paolo Masetti. I know he had done some work
with getting lnpd to run under cygwin. I'm not sure if he still reads lugnet;
I've been absent for about three years and just getting back seeing TLG is about
to release NXT.
Also if you try the source but run into problems you can post them here and we
will see what we can do.
Ed
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.robotics, Ed Manlove wrote:
> In lugnet.robotics, Cortland Starrett wrote:
> >
> > I am the "friend". :) My name is Cortland Starrett. I use a modeling • [trim]
> > Thanks for any help or pointers. And thanks Steve for posting for me
> > while I waited on my registration.
> >
> > Cort
>
> Cort,
>
> I know you are looking for an executable but have your tried building the
> source? • [trim]
> Also if you try the source but run into problems you can post them here and we
> will see what we can do.
>
> Ed
Here is the first error. O_ASYNC is a Linux/BSD thing...
10:11 AM-/lnpd-0.9.0$ make
make all-recursive
make[1]: Entering directory `/lnpd-0.9.0'
Making all in lnpd+liblnp
make[2]: Entering directory `/lnpd-0.9.0/lnpd+liblnp'
Making all in liblnp
make[3]: Entering directory `/lnpd-0.9.0/lnpd+liblnp/liblnp'
/bin/sh ../../libtool --mode=compile gcc -DHAVE_CONFIG_H -I. -I. -I../..
-D_GNU_SOURCE -g -O2 -c stub.c
gcc -DHAVE_CONFIG_H -I. -I. -I../.. -D_GNU_SOURCE -g -O2 -c stub.c -DPIC -o
.libs/stub.o
stub.c: In function `lnp_init':
stub.c:381: error: `O_ASYNC' undeclared (first use in this function)
stub.c:381: error: (Each undeclared identifier is reported only once
stub.c:381: error: for each function it appears in.)
stub.c:383: error: `F_SETSIG' undeclared (first use in this function)
make[3]: *** [stub.lo] Error 1
make[3]: Leaving directory `/lnpd-0.9.0/lnpd+liblnp/liblnp'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/lnpd-0.9.0/lnpd+liblnp'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/lnpd-0.9.0'
make: *** [all-recursive-am] Error 2
I have commented out this check in the code, but not getting much farther.
Cort
| | | | | | | | | | | | | | | | | |
| |
|
In lugnet.robotics, Cortland Starrett wrote:
> In lugnet.robotics, Ed Manlove wrote:
> > In lugnet.robotics, Cortland Starrett wrote:
> > >
> > > I am the "friend". :) My name is Cortland Starrett. I use a modeling [trim]
> > > Thanks for any help or pointers. And thanks Steve for posting for me
> > > while I waited on my registration.
> > >
> > > Cort
> >
> > Cort,
> >
> > I know you are looking for an executable but have your tried building the
> > source? [trim]
> > Also if you try the source but run into problems you can post them here and we
> > will see what we can do.
> >
> > Ed
>
> Here is the first error. O_ASYNC is a Linux/BSD thing...
[SNIP]
>
> I have commented out this check in the code, but not getting much farther.
>
> Cort
Cort,
I would venture to guess from the error you are getting that the lnpd code on
the BrickOS website never was quite working under cygwin. I also gather this
from Stephan Höhrmann response on a different and recent thread on lnpd
(http://news.lugnet.com/robotics/rcx/legos/?n=3984).
It does look like Stephan has does work on an alternative he calls lnphost. You
might check out his work, http://lnphost.sourceforge.net/, and see if what you
are looking for in terms of functionality. Otherwise you might have alot of
debuging just to get the code to be "cygwin friendly".
Ed
| | | | | | | | | | | | | | | | |
>
> Cort,
>
> I would venture to guess from the error you are getting that the lnpd code on
> the BrickOS website never was quite working under cygwin. I also gather this
> from Stephan Höhrmann response on a different and recent thread on lnpd
> (http://news.lugnet.com/robotics/rcx/legos/?n=3984).
>
> It does look like Stephan has does work on an alternative he calls lnphost. You
> might check out his work, http://lnphost.sourceforge.net/, and see if what you
> are looking for in terms of functionality. Otherwise you might have alot of
> debuging just to get the code to be "cygwin friendly".
>
> Ed
lnphost is perfect. Thank you for the reference.
Cort
| | | | | | |
