<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body><span class="vcard"><a class="email" href="mailto:eero.t.tamminen@intel.com" title="Eero Tamminen <eero.t.tamminen@intel.com>"> <span class="fn">Eero Tamminen</span></a>
</span> changed
          <a class="bz_bug_link 
          bz_status_REOPENED "
   title="REOPENED - [BXT] up to 6% performance drop with "i965: Set "Subslice Hashing Mode" to 16x16 on Apollolake""
   href="https://bugs.freedesktop.org/show_bug.cgi?id=102272">bug 102272</a>
          <br>
             <table border="1" cellspacing="0" cellpadding="8">
          <tr>
            <th>What</th>
            <th>Removed</th>
            <th>Added</th>
          </tr>

         <tr>
           <td style="text-align:right;">Resolution</td>
           <td>WONTFIX
           </td>
           <td>---
           </td>
         </tr>

         <tr>
           <td style="text-align:right;">Status</td>
           <td>RESOLVED
           </td>
           <td>REOPENED
           </td>
         </tr></table>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_REOPENED "
   title="REOPENED - [BXT] up to 6% performance drop with "i965: Set "Subslice Hashing Mode" to 16x16 on Apollolake""
   href="https://bugs.freedesktop.org/show_bug.cgi?id=102272#c3">Comment # 3</a>
              on <a class="bz_bug_link 
          bz_status_REOPENED "
   title="REOPENED - [BXT] up to 6% performance drop with "i965: Set "Subslice Hashing Mode" to 16x16 on Apollolake""
   href="https://bugs.freedesktop.org/show_bug.cgi?id=102272">bug 102272</a>
              from <span class="vcard"><a class="email" href="mailto:eero.t.tamminen@intel.com" title="Eero Tamminen <eero.t.tamminen@intel.com>"> <span class="fn">Eero Tamminen</span></a>
</span></b>
        <pre>There was also following perf drops during the same time frame, which I assume
are due to same change:
- 2% GpuTest v0.7 PixMark Piano
- 1% GpuTest v0.7 PixMark Volplosion

And now that <a class="bz_bug_link 
          bz_status_RESOLVED  bz_closed"
   title="RESOLVED FIXED - Up to 20% perf drop in texture sampling tests with "i965/surface_state: Get the aux usage from the miptree code""
   href="show_bug.cgi?id=102258">bug 102258</a> is fixed, SynMark Anisotropic filtering test
performance is still 6-7% lower than it was before that bug.

Except for terrain tessellation tests, all the dropping test-cases use
anisotropic filter.

So, I think both cases that have a lot of really small triangles, and/or where
usage of anisotropic filtering has clear impact for the performance, suffer
from this change.


<a class="bz_bug_link 
          bz_status_RESOLVED  bz_closed"
   title="RESOLVED FIXED - Up to 20% perf drop in texture sampling tests with "i965/surface_state: Get the aux usage from the miptree code""
   href="show_bug.cgi?id=102258">Bug 102258</a> may have hidden also other impacts of this change, so it makes sense
to re-test the impact of the hash mode on 18 EU BXT where it's more visible. 
The affected test-cases were:
- GLB 2.7 Fill*
- GpuTest v0.7 GiMark, FurMark, Piano, Volplosion & Triangle*
- SynMark v7 TerrainPanTess, TerrainFlyTess, TexFilterAniso, TexFilterTri*,
TexMem128, TexMem512* & ZBuffer*

Potentially also Unigine Valley 1.0.

(*) These were improved, others regressed with 8x8 -> 16x16 change.</pre>
        </div>
      </p>


      <hr>
      <span>You are receiving this mail because:</span>

      <ul>
          <li>You are the assignee for the bug.</li>
          <li>You are the QA Contact for the bug.</li>
      </ul>
    </body>
</html>