This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: miscellaneous nits in NetBSD configuration
- To: richard dot earnshaw at arm dot com
- Subject: Re: PATCH: miscellaneous nits in NetBSD configuration
- From: Jason Merrill <jason at cygnus dot com>
- Date: 03 Aug 1998 11:59:31 -0700
- Cc: egcs at cygnus dot com, tv at pobox dot com, rearnsha at sun52 dot arm dot com
- References: <199808031338.OAA07866@sun52.NIS.cambridge>
>>>>> Richard Earnshaw <rearnsha@arm.com> writes:
> jason@cygnus.com said:
>> The problem, as someone pointed out, is that defaults.h will just
>> #define it again if you #undef it. Methinks the change will be
>> counterproductive, and should be backed out.
> Not quite. It will only get defined if it is not already *and*
> INCOMING_RETURN_ADDR_RTX is defined. Code in final.c then checks whether
> DWARF2_UNWIND_INFO is defined or not, it does not check its value. For an
> a.out machine using stabs (eg netbsd/arm32), defining this to 0 or 1
> causes dwarf information to be sucked into the build.
Yes, but it doesn't get used unless you say -fno-sjlj-exceptions. I don't
think it's a problem for the compiler to contain dwarf code that you don't
use.
Jason