Following is the list of fp benchmarks that fail. They all pass with -O3, but some fail with -O4. I did the test run.<br><br>Thanks,<br>Reza<br><br>                                  Estimated                       Estimated<br>
                Base     Base       Base        Peak     Peak       Peak<br>Benchmarks      Ref.   Run Time     Ratio       Ref.   Run Time     Ratio<br>-------------- ------  ---------  ---------    ------  ---------  ---------<br>
410.bwaves                                         --    0.0194             RE <br>416.gamess                                         --    0.00135            RE <br>433.milc                                           --   16.5             -- S  <br>
434.zeusmp                                         --    0.00146            RE <br>435.gromacs                                        --    0.00138            RE <br>436.cactusADM                                      --    0.00135            RE <br>
437.leslie3d                                       --    0.00141            RE <br>444.namd                                           --   19.4             -- S  <br>447.dealII                                         --   19.7             -- S  <br>
450.soplex                                         --    0.0380          -- S  <br>453.povray                                         --    2.49            -- S  <br>454.calculix                                       --    0.00135            RE <br>
459.GemsFDTD                                       --    0.00150            RE <br>465.tonto                                          --    2.40            -- S  <br>470.lbm                                            --    6.52            -- S  <br>
481.wrf                                            --    0.00140            RE <br>482.sphinx3                                        --    3.19            -- S  <br><br><div class="gmail_quote">On Thu, Jul 22, 2010 at 4:59 PM, Dale Johannesen <span dir="ltr"><<a href="mailto:dalej@apple.com">dalej@apple.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div class="im"><br>
On Jul 22, 2010, at 4:18 PMPDT, Reza Yazdani wrote:<br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Hi,<br>
<br>
I ran Spec2006 with -O4. All integer benchmarks passed, but only 8 out 17 of  floating point benchmarks passed. Is this normal or I made a mistake in my build?<br>
</blockquote>
<br></div>
Hi Reza.  Somebody on Linux should answer, but I don't think it's normal.  You may have checked out the source at a moment when it had a bug in it; that unfortunately happens frequently.  Which ones are failing?<br>

<br>
</blockquote></div><br>