This is the mail archive of the
gdb@sources.redhat.com
mailing list for the GDB project.
I/O mechanism in libGDB
- From: "Nimesh D Chanchani" <nimeshc at rediffmail dot com>
- To: gdb at sources dot redhat dot com
- Date: 9 Mar 2003 11:24:17 -0000
- Subject: I/O mechanism in libGDB
- Reply-to: "Nimesh D Chanchani" <nimeshc at rediffmail dot com>
Hi,
I found your id from google groups. Thank god i found it.
well... I am required to upgrade gdb implementation 4.17 for zsp
processor to libGDB implementation for GDB 5.2.
I was going through API specification of libGDB in the one of the
only
2 documents available on the web.
http://www.cs.utah.edu/dept/old/texinfo/libgdb/libgdb_toc.html#SEC4
I had some queries about the i/o mechanism used in libGDB .
what my understanding is that when for example the GDB gives the
information on the terminal about (say) a recently created
break
point, it would have to be parsed by a libGDB API function and
give it
to the GUI in a "STANDARD FORMAT" irrespective of the version of
GDB
being used.
for this the document
http://www.cs.utah.edu/dept/old/texinfo/libgdb/libgdb_toc.html#SEC4
specifies that a structure of function pointers should be used
for
output.
struct gdb_output_vector
{
/* output */
void (*put_string) (struct gdb_output_vector *, char * str);
}
I couldn't quiet understand the need for such a complicated
mechanism ,
more over I couldn't get the significance of both the input
parametres.
actually i didn't get how does one implement the i/o mechanism in
the
mentioned way.
can you please help me out with this......any help will be
greatly
appreciated.
Thanking in advance
Nimesh Chanchani