This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: patch for sparc-tdep.c for NetBSD/sparc.
- To: Michael Snyder <msnyder at cygnus dot com>
- Subject: Re: patch for sparc-tdep.c for NetBSD/sparc.
- From: jtc at redback dot com (J.T. Conklin)
- Date: 14 Jul 2000 14:14:50 -0700
- Cc: matthew green <mrg at cygnus dot com>, David Taylor <taylor at cygnus dot com>,gdb-patches at sourceware dot cygnus dot com
- References: <28566.963550086@cygnus.com> <5mhf9sr3yz.fsf@jtc.redback.com> <396F7CC2.6C99@cygnus.com>
- Reply-To: jtc at redback dot com
>>>>> "Michael" == Michael Snyder <msnyder@cygnus.com> writes:
>> It's better, but I think the suggestions were to add the following
>> macros.
>>
>> #define GDB_GREGSET_T struct reg
>> #define GDB_FPREGSET_T struct fpreg
>>
>> as that is a more direct way of defining the types.
Michael> I really like typedefs better. They are debuggable,
Michael> #defines are not.
I don't see that this is much different than the GDB_{G,FP}REGSET_T
macro definitions in gregset.h. Those definitions are set to types
by default, but they are only used in to define gdb_{g,fp}regset_t,
the types which are actually used throughought the rest of the code.
It seems wrong to not define GDB_{G,FP}REGSET_T explicitly, but rather
to define the {g,fp}regset_t types which are only used to define the
type of gdb_{g,fp}regset_t if GDB_{G,FP}REGSET_T is not defined.
I'm not clear to me what added debuggability the extra layer of
obfucation provides.
--jtc
--
J.T. Conklin
RedBack Networks