This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
"Keith" == Keith Seitz<keiths@redhat.com> writes:
Keith> Yeah, I wasn't 100% sure whether this needed a cleanup, Keith> either. Nonetheless, we have two options, catch any errors (which I Keith> think was suggested in another thread for a different problem) or add Keith> a cleanup. A little paranoia can't hurt that much. If you have a Keith> preference, I'll implement that.
One of us can handle it in a followup patch.
Can this call to dwarf2_physname exhibit the problem that this patch is trying to circumvent?
Keith> Unfortunately, it very well could. I guess it would be almost as Keith> useful if this called dwarf2_fullname instead. What do you think?
Yeah. I think it is important for the error case not to crash. Printing something useful is good, maybe even the CU+DIE would be enough.
class a { private: class b { public: b () { } };
std::vector<b> list; };
ChangeLog (for the obstack leak patch) 2010-05-11 Keith Seitz <keiths@redhat.com>
* dwarf2read.c (read_structure_type): Check if the current DIE's type was already completed after dwarf2_full_name was called.
Attachment:
delayed_physnames-2.patch
Description: Text document
Attachment:
obstack-leak.patch
Description: Text document
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |