RE: [cdt-dev] CDT 2.1 RC1/2: Problems with Remote Debugging

[Neugebauer Manfred <manfred.neugebauer@xxxxxxxxxxx>]

Hallo,

Dave (see last entry in the history of this bug 77643) received already a
copy of my executable. His analysis was that the problem is related to the
fact, that my executable is linked relocatable (linked with -r). From the
last mail I understand that IBinaryExecutable is currently not able to
interprete this type of executable (although gdb can handle this as input to
a debug session).

Now I'm interested in the next steps:
- are there plans to correct this problem in CDT 2.1 (or later) to get the
old state back 
- hints that I can make the correction (and feed it back to the community)
- it will never work again

Thanks. Manfred.


-----Original Message-----
From: cdt-dev-admin@xxxxxxxxxxx [mailto:cdt-dev-admin@xxxxxxxxxxx]On
Behalf Of Alain Magloire
Sent: Mittwoch, 17. November 2004 16:48
To: cdt-dev@xxxxxxxxxxx
Subject: Re: [cdt-dev] CDT 2.1 RC1/2: Problems with Remote Debugging


> 
> 
> Hallo,
> 

a` l'eau ?

> Since one year we and our customers are using Eclipse / CDT successfully
for
> developing and debugging C / C++ applications running on the Siemens
> realtime kernel RMOS. With CDT 2.1 RC1 we recognized a problem when
> launching a (remote) debug session. As it turned out the problem is that
the
> new CDT 2.1 update cannot handle executables which are linked relocatable
> (see bug report 77643). This executable configuration is required that gdb
> and the RMOS loading system work together. This type of launching was
> possible up to CDT 2.0.2 and I don't know why this has changed now. This
> problem will prevent us to use CDT 2.1 which is not a very good option.
> 
> Can somebody check if it is possible to get this feature back into the CDT
> 2.1 release chain? If any support is needed we should talk about this.
> 

The launch was change, it requires now a IBinaryExecutable.
And it looks like the BinaryParser(Elf, GNU/Elf ?) do not recognize it
your file as a valid excutable, that would be my first guess.

Can you attach the offending file to the PR?

_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
http://dev.eclipse.org/mailman/listinfo/cdt-dev