freesolid-users — list for user discussions

Re: [Freesolid-users] bug: probably found memory leak

[Rodrigo H. <kw...@ae...>]

Ok, I fixed it, thanks, let me know if you find any more.

Cheers!

Andreas wrote:

> hi,
>
> i used freesolid 2.1.1 compiled as static lib with microsoft visual 
> .net 7.1.
> i have detected a memory leak in my project. after tracing i found it 
> was caused by method Object::do_broadphase() in the following lines:
>
>     Scalar *lower_point = new Scalar[3];
>     Scalar *upper_point = new Scalar[3];
>
> the allocated memory is not used and not freed. it should be replaced by:
>
>     Scalar *lower_point;
>     Scalar *upper_point;
>
> i test it and it works fine.
>
> ciao
> andreas
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> _______________________________________________
> Freesolid-users mailing list
> Fre...@li...
> https://lists.sourceforge.net/lists/listinfo/freesolid-users
>
>

[Freesolid-users] bug: probably found memory leak

[Andreas <pi...@cs...>]

hi,

i used freesolid 2.1.1 compiled as static lib with microsoft visual .net 
7.1.
i have detected a memory leak in my project. after tracing i found it 
was caused by method Object::do_broadphase() in the following lines:

	Scalar *lower_point = new Scalar[3];
	Scalar *upper_point = new Scalar[3];

the allocated memory is not used and not freed. it should be replaced by:

	Scalar *lower_point;
	Scalar *upper_point;

i test it and it works fine.

ciao
andreas

Re: [Freesolid-users] from [Freesolid-cvs] Nice work here is some more.

[Visti A. <ta...@hc...>]

On Sun, 17 Apr 2005 17:02:58 -0600
Rodrigo Hernandez <kw...@ae...> wrote:

> Visti Andresen wrote:
> 
> >A real bug? Or did I just miss something in the manuals again?
> >
> >  
> >
> Well, actually, the procedure when compiling fresh from CVS should be:
> 
> $ sh bootstrap.sh
> $ ./configure
> $ make
> $ make install
> 
> Bootsrap is a simple script that runs aclocal, autoheader, automake and 
> autoconf in that order in order to get a fresh configure script, it 
> passes the necesary parameters to each of them,
> so no need to run each manually is required.
> 
> try running it before you run configure, if you have a "dirty" source 
> base from unsuccessfully trying to compile, try:
> 
> $ sh bootstrap.sh
> $ ./configure
> $ make clean
> $ make
> $ make install
> 
> So the library gets build from scratch.
> 
> then let me know if you have any other problems, if you do, then there 
> is a bug in the code :)
> 
After removing automake1.4 (the automake symlink still pointed to automake1.4 after the installation of automake1.9)
And a new checkout of the source (I initially had some problems that seemed like "make clean" wasn't being clean enough for bootstrap so I just did another checkout)

Now everything works just fine.
I will try to run the samples through valgind.

Thank you for your time and help.

Re: [Freesolid-users] from [Freesolid-cvs] Nice work here is some more.

[Rodrigo H. <kw...@ae...>]

Visti Andresen wrote:

>A real bug? Or did I just miss something in the manuals again?
>
>  
>
Well, actually, the procedure when compiling fresh from CVS should be:

$ sh bootstrap.sh
$ ./configure
$ make
$ make install

Bootsrap is a simple script that runs aclocal, autoheader, automake and 
autoconf in that order in order to get a fresh configure script, it 
passes the necesary parameters to each of them,
so no need to run each manually is required.

try running it before you run configure, if you have a "dirty" source 
base from unsuccessfully trying to compile, try:

$ sh bootstrap.sh
$ ./configure
$ make clean
$ make
$ make install

So the library gets build from scratch.

then let me know if you have any other problems, if you do, then there 
is a bug in the code :)

>Thanks for the e-mail informing me about the new mailing list
>
>  
>
No Problem.

Cheers!

[Freesolid-users] from [Freesolid-cvs] Nice work here is some more.

[Visti A. <ta...@hc...>]

A case of RTFM, and perhaps a real issue when compiling the source.

> Hi Visti,
> 
> Thanks for the kind words, I didnt thought anyone was using the library, 

Actually I know at least one other project that uses FreeSOLID http://www.genecys.org/

> I'll keep that
> in mind next time I am about to commit something that may not entirelly 
> compile.
> 
> Anyway, regarding your problem, make sure you have Automake 1.8.2 or 
> higher as well as
> libtool 1.5 or higher installed, check for a current autoheader if you 
> feel like it too :)
> 
> Aparently even though 1.9 automake packages are available for Debian 
> distributions, by default the version installed is 1.4, this may be the 
> issue here.
> 
> Let me know how it goes :)
> 
> Cheers!
> 
>

Did an dist-upgrade of my Debian machine (unstable) and installed automake 1.9. It did't fix the problem however going through the docs once more did :o)


README clearly states:
"If you've checked out a fresh CVS sandbox, the following steps are
required to generate a proper 'configure' file.

    aclocal
    automake -a
    autoconf
    autoheader"

However I apparently lost my head when reading the following in INSTALL and only ran autoconf:
"The file `configure.in' is used to create `configure' by a program
called `autoconf'.  You only need `configure.in' if you want to change
it or regenerate `configure' using a newer version of `autoconf'."

automake still complains but it dos't seem as serious






However when I try to make I get:
"/bin/sh ../libtool --mode=compile g++ -DHAVE_CONFIG_H -I. -I. -I../include   -I../include -I./include -I. -I../libbroad -I../libmoto  -g -O2 -c Polyhedron.cpp
 g++ -DHAVE_CONFIG_H -I. -I. -I../include -I../include -I./include -I. -I../libbroad -I../libmoto -g -O2 -Wp,-MD,.deps/Polyhedron.pp -c Polyhedron.cpp  -fPIC -DPIC -o .libs/Polyhedron.o
Polyhedron.cpp:36:21: qhull_a.h: No such file or directory"

doing 
export CPPFLAGS="-I../qhull/src"
export CXXFLAGS="-I../qhull/src"
export CFLAGS="-I../qhull/src"
before running ./configure fixes the problem

but then another problem occurs later when trying to link a sample:
g++ -I../qhull/src -o .libs/sample sample.o  -L/home/talpa/projects/opensource/freesolid/libsolid /home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID -lm -L/home/talpa/projects/opensource/freesolid/qhull/src -Wl,--rpath -Wl,/usr/local/lib
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `BP_CreateScene'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_qh'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `BP_DeleteProxy'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_qhull'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_pointid'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_memfreeshort'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_init_B'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `BP_SetBBox'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_freeqhull'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_check_output'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_init_A'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_facet3vertex'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `BP_CreateProxy'
/home/talpa/projects/opensource/freesolid/libsolid/.libs/libFreeSOLID: undefined reference to `qh_initflags'
collect2: ld returned 1 exit status


A real bug? Or did I just miss something in the manuals again?



Thanks for the e-mail informing me about the new mailing list