This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Variations on inherit.exp
- From: Joel Brobecker <brobecker at adacore dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- Cc: Luis Machado <lgustavo at codesourcery dot com>, "gdb at sourceware dot org" <gdb at sourceware dot org>
- Date: Wed, 5 Jun 2013 15:52:36 +0400
- Subject: Re: Variations on inherit.exp
- References: <51AF1EBF dot 70709 at codesourcery dot com> <20130605113055 dot GA10055 at host2 dot jankratochvil dot net>
> > Should we add one more variation of output to inherit.exp? Should
> > the testcase be less picky about the symbolic information?
>
> It should be easier to just use "set print symbol off".
> But it would be nice to investigate while x86* has not found the name.
As a side note, the symbolic name has been a constant source of
pain in AdaCore's gdb-testsuite, mostly on bare metal targets
were we use customized runtimes. We get slight differences like
these, and require minute adjustments to pass. I've been tempted
many time to use "set print symbol off" as suggested by Jan, but
I have always (almost!) refrained from doing so, even the testcase
has nothing to do with symbolic address printing, because I believe
Pretty strongly in testing the debugger under the same conditions as
our users.
(digressing even further, I did in fact deviate from the above
principle, by enabling "set print frame-arguments all", but this
causes us to test _more_, so that was an acceptable deviation).
--
Joel