This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: [patch, libgfortran] PR32611 Print sign of negative zero
- From: Jerry DeLisle <jvdelisle at verizon dot net>
- To: Brooks Moses <brooks dot moses at codesourcery dot com>
- Cc: fortran at gcc dot gnu dot org, gcc-patches at gcc dot gnu dot org
- Date: Mon, 09 Jul 2007 19:57:44 -0700
- Subject: Re: [patch, libgfortran] PR32611 Print sign of negative zero
- References: <4691C8D6.7050703@verizon.net> <4691F8BD.1090200@codesourcery.com>
Brooks Moses wrote:
Jerry DeLisle wrote:
The attached patch provides a solution to this PR.
The patch uses the signbit function to determine if a zero value is
negative. If so, the sign is emitted. If -std=legacy is specified at
compile time, the negative sign for zero values is not emitted.
This patch requires -std=legacy to be used to pass NIST. This is
because F77 explicitly prohibits printing the negative sign for zero
values.
[...]
OK for trunk?
This needs documenting, of course.
If you aren't sure of the best place to document it, feel free to create
a documentation PR about it and assign it to me, and I'll get to it
sometime in August when I have time to think again. :)
In particular, note that this changes the meaning of the -std= option.
Currently, it only affects what language features are accepted,
rejected, or warned about, but any program that compiles will have the
same behavior regardless of which -std= option was used to compile it.
Your patch changes this, which is IMHO a fairly significant little change.
- Brooks
Do you think it better to have a separate compiler option like -fsign-zero?
Others please chime in.
Jerry