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]

Re: [patch 2/2] typedef-checking for CU relative vs. absolute offsets [Re: RFC: problem with DW_OP_GNU_deref_type and dwarf's get_base_type callback]

On Wed, 07 Mar 2012 19:58:17 +0100, Doug Evans wrote:
> The problem stems from "offset" being ambiguous.
> I'd rather just pick a better (clearer) name and be consistent.

I have to strongly disagree, without automatic checking this bug will creep in
again. Even with testcases it does not get found as they rarely excersise
multiple CUs.

Plus these advanced DWARF features are used only for -O2 -g builds and so bugs
there have low chance to be reported - everyone expects -O2 -g debug info is
not good.


I find the other possibility some static checker instead, to have just some:
typedef unsigned int cu_offset; typedef unsigned int sect_offset;
No idea which static checker can find it.


Thanks,
Jan
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]