This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: RFA: test that GDB tolerates bad #inclusion data
- From: Jim Blandy <jimb at redhat dot com>
- To: Michael Elizabeth Chastain <mec at shout dot net>
- Cc: carlton at kealia dot com, gdb-patches at sources dot redhat dot com
- Date: 23 Sep 2003 17:30:11 -0500
- Subject: Re: RFA: test that GDB tolerates bad #inclusion data
- References: <200309231803.h8NI3b2c026041@duracef.shout.net>
Michael Elizabeth Chastain <mec@shout.net> writes:
> Jim, we gotta feed you the gdb_test_multiple kool-aid.
> See gdb1250.exp or float.exp or watchpoint.exp.
Um, why would I want to bow to a charismatic leader's will and commit
an act of ultimate self-destruction?
> Also, could you avoid the dependency on gcc version by using
> preprocessed ".i" files with the bad line numbers already in them?
> Instead of / in addition to the tests with C source.
Let me fool around a bit with #line directives and see if I can do as
well without the -imacro switches.
In the end, though, it's up to the compiler to generate correct
Dwarf 2 information, so it'll always be compiler-dependent.