This is the mail archive of the
mailing list for the GDB project.
Re: [patch] [4/5] Types reference counting [varobj-validation]
- From: Tom Tromey <tromey at redhat dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Thu, 16 Apr 2009 15:54:31 -0600
- Subject: Re: [patch] [4/5] Types reference counting [varobj-validation]
- References: <20090411102215.GE32624@host0.dyn.jankratochvil.net>
- Reply-to: tromey at redhat dot com
>>>>> "Jan" == Jan Kratochvil <firstname.lastname@example.org> writes:
Jan> As varobj_invalidate was unconditionally invalidating any local
Jan> varobjs more often calls to varobj_invalidate could become more
Jan> intrusive to the user so I also fixed varobj_invalidate to care
Jan> only about varobjs bound to the specific objfile.
I am not sure about this implementation.
It seems to me that a varobj expression can span multiple objfiles.
So, don't we have to use the same logic as Paul's recent "display"
change to determine whether a varobj expression is invalid?
Or am I missing something? I ask because it looks like we have
existing bugs here.