This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC 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: [committed] Clear MIPS IV Flush to Zero bit on IRIX


Richard Sandiford <rsandifo@redhat.com> writes:
> This whole thing probably warrants a changes.html entry, to follow shortly.

Here's what I checked in.  Gerald's script seems happy with it.

Richard


Index: gcc-4.0/changes.html
===================================================================
RCS file: /cvs/gcc/wwwdocs/htdocs/gcc-4.0/changes.html,v
retrieving revision 1.26
diff -c -p -F^\([(a-zA-Z0-9_]\|#define\) -r1.26 changes.html
*** gcc-4.0/changes.html	18 Jan 2005 08:10:43 -0000	1.26
--- gcc-4.0/changes.html	26 Jan 2005 22:56:24 -0000
*************** class N::A {
*** 372,377 ****
--- 372,385 ----
          library directories: o32 libraries go into <code>lib/</code>,
          n32 libraries go into <code>lib32/</code> and n64 libraries
          go into <code>lib64/</code>.</li>
+     <li>GCC-linked IRIX applications will now clear the MIPS IV
+         FCSR.FS (flush to zero) bit by default.  This may make the
+         applications run slower on MIPS IV systems than they would
+         do otherwise, but means that they will correctly handle
+         subnormal floating-point values on both MIPS III and MIPS IV
+         systems.  You can override this behavior by linking with either
+         <code>-ffast-math</code> or
+         <code>-funsafe-math-optimizations</code>.</li>
    </ul>
  
  <h3>S/390 and zSeries</h3>


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