Steve,

Sorry for having you run down for a few days.  I just did a few testing
here.
The  latest GCC 2.96.x snapshot from RedHat,
which I assume you are using, per your previous post, breaks OpenDx for
RedHat 6.x compiled with GCC 2.95.x

I have filed this as a bug with RedHat.  The GCC 2.96 will be the base
compiler
on RedHat 7.0, when released this Fall.  I receieved RH 7.0 Beta-2
today, and ran into
the same trouble you are having.  RedHat Beta-2 has GCC 2.96 and it
broke binary compability
for several applications.

I will compile DX 4.1.1 and release a newer DX archive for users who are
using GCC 2.96
or higher snapshots on RedHat.

Suhaib




Steve Ettorre wrote:
> 
> Suhaib-
> 
> Thanks for the tip. Unfortunately, when I tried to implement it here is what
> happened:
> 
> /usr/local/dx/bin_linux
> dad.27% ldd *
> builder:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> dxexec:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> dxui:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> prompter:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> startupui:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> tutor:
> BUG IN DYNAMIC LINKER ld.so: dl-version.c: 210: _dl_check_map_versions:
> Assertion `needed != ((void *)0)' failed!
> 
> Not very promising - is it?
> 
> -S.
> ----------------------------------------------------------------
> 
> "Suhaib M. Siddiqi" wrote:
> 
> > The bst way to find the names of dynamic libraries... cd to dx/bin_linux
> > and type:
> >
> > ldd *
> >
> > This should output the names of all the libraries dynamically linked to
> > executables.  You can check if you have the same dynamic libraries on your
> > system,
> > if not setup symbolic links.  Unfortunately, RedHat uses weird names for
> > libstdc++.so and change them to weird names each time they release a new rpm
> > of libstdc++.  I do not have a valid explaination why RedHat developers
> > love to rename libstdc++ to something soooooooooo weird which causes
> > applications compiled on one version of RedHat to break on other version
> > of RedHat.  It is a nightmare for developers.  It would be a whole lot
> > easier, if they stick to standard libstdc++.so.
> >
> > I am RedHat Beta team for RedHat upcoming
> > release of RH 7.0.  I will raise this question on their beta-testers site.
> > I hope they listen and stop susing the weird libstdc++ blah blah numbers
> > from one RPM release to another.
> >
> > Suhaib
> >
> > > Hi-
> > >
> > > Following the suggestion from Tom Gardiner, I have examined the soft
> > > links for libstdc in /usr/lib. Here is what is currently setup:
> > >
> > >
> > >    0 lrwxrwxrwx   1 root     root           30 May 22 19:07
> > > libstdc++-libc6.1-1.so.2 -> libstdc++-2-libc6.1-1-2.9.0.so
> > >    0 lrwxrwxrwx   1 root     root           31 May 22 19:07
> > > libstdc++-libc6.1-2.so.3 -> libstdc++-3-libc6.1-2-2.10.0.so
> > >
> > > Note, I have the link libstdc++-libc6.1-2.so.3 ->
> > > libstdc++-3-libc6.1-2-2.10.0.so. This differs from what Tom identified
> > > (i.e., libstdc++-libc6.1-2.so.3 -> libstdc++-2-libc6.1-1-2.9.0.so).
> > >
> > > Also note, I am running a very recent snapshot of the gcc suite of
> > > compilers. The rpm packages for the dynamic libraries
> > > libstdc++-2-libc6.1-1-2.9.0.so and ibstdc++-3-libc6.1-2-2.10.0.so are
> > > libstdc++-compat-2.95.3-0.20000323 and libstdc++-2.95.3-0.20000323,
> > > respectively. Does this seem to be ok?
> > >
> > > One last thing, I am using the rpm package for opendx-4.1.0-1. I have
> > > found README files in /usr/local/dx/doc but have not been able to locate
> > > any specific instructions dealing with setting links to libstdc. If
> > > anyone could point towards where I can find this information, I would
> > > appreciate it.
> > >
> > > TIA,
> > > Steve
> > >
> > >
> > >
> > > --
> > > Steve Ettorre
> > > e-mail: [EMAIL PROTECTED]
> > > -------------------------------------------
> > > "...thinking is not consciousness -
> > >  it requires hard work..." - Rush Limbaugh
> > > -------------------------------------------
> > >
> > >
> 
> --
> Steve Ettorre
> e-mail: [EMAIL PROTECTED]
> -------------------------------------------
> "...thinking is not consciousness -
>  it requires hard work..." - Rush Limbaugh
> -------------------------------------------

Reply via email to