This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: PATCH: XFAIL c99-float-1.c on Solaris
- To: Mark Mitchell <mark at codesourcery dot com>
- Subject: Re: PATCH: XFAIL c99-float-1.c on Solaris
- From: "Joseph S. Myers" <jsm28 at cam dot ac dot uk>
- Date: Mon, 14 May 2001 23:19:15 +0100 (BST)
- cc: <gcc-patches at gcc dot gnu dot org>
On Mon, 14 May 2001, Mark Mitchell wrote:
> Solaris doesn't define FLT_EVAL_METHOD or DECIMAL_DIG. This test
> failed with GCC 2.95.x as well, so this is not a regression.
>
> Tested on sparc-sun-solaris2.8, applied on the branch.
I think some documentation of the problem of systems which don't use GCC's
float.h is needed. I've installed the following in both copies of
c99status.html.
For a proper fix, such systems need either to get their own versions of
float.h in GCC or to have their system float.h fixincluded.
Index: c99status.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/c99status.html,v
retrieving revision 1.24
diff -u -r1.24 c99status.html
--- c99status.html 2001/02/17 10:22:10 1.24
+++ c99status.html 2001/05/14 22:14:30
@@ -309,7 +309,9 @@
<code>FLT_EVAL_METHOD</code> and <code>DECIMAL_DIG</code>) are
included in most versions of <code><float.h></code>, but not yet
<code>float-c4x.h</code>, <code>float-m68k.h</code> or
-<code>float-vax.h</code>.</li>
+<code>float-vax.h</code>. Also, some targets such as Solaris use the
+system <code><float.h></code> but do not fix it to add these
+definitions.</li>
<li>Compiler support is needed for <code>math_errhandling</code>; see
messages <a
--
Joseph S. Myers
jsm28@cam.ac.uk