<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body><table border="1" cellspacing="0" cellpadding="8">
        <tr>
          <th>Bug ID</th>
          <td><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - Regression in Mesa 17 on s390x (zSystems)"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=100613">100613</a>
          </td>
        </tr>

        <tr>
          <th>Summary</th>
          <td>Regression in Mesa 17 on s390x (zSystems)
          </td>
        </tr>

        <tr>
          <th>Product</th>
          <td>Mesa
          </td>
        </tr>

        <tr>
          <th>Version</th>
          <td>17.0
          </td>
        </tr>

        <tr>
          <th>Hardware</th>
          <td>Other
          </td>
        </tr>

        <tr>
          <th>OS</th>
          <td>All
          </td>
        </tr>

        <tr>
          <th>Status</th>
          <td>NEW
          </td>
        </tr>

        <tr>
          <th>Severity</th>
          <td>normal
          </td>
        </tr>

        <tr>
          <th>Priority</th>
          <td>medium
          </td>
        </tr>

        <tr>
          <th>Component</th>
          <td>Drivers/Gallium/swr
          </td>
        </tr>

        <tr>
          <th>Assignee</th>
          <td>mesa-dev@lists.freedesktop.org
          </td>
        </tr>

        <tr>
          <th>Reporter</th>
          <td>sndirsch@suse.de
          </td>
        </tr>

        <tr>
          <th>QA Contact</th>
          <td>mesa-dev@lists.freedesktop.org
          </td>
        </tr></table>
      <p>
        <div>
        <pre>I've observed a regression in Mesa 17 on s390x (zSystems). It's hard to
describe. Here is a video on Youtube.

  <a href="https://youtu.be/xaYauE0Kn8E">https://youtu.be/xaYauE0Kn8E</a>

On the left side glxgears is running on x86_64, on the left side it's running
on s390x(zSystems) - a 64bit bigendian platform. gdm and gnome-shell show a
black screen. I haven't screenhoted that though (since it's boring).

I git bisected Mesa and the culprit patch was

>From e827d9175675aaa6cfc0b981e2a80685fb7b3a74 Mon Sep 17 00:00:00 2001
From: Roland Scheidegger <<a href="mailto:sroland@vmware.com">sroland@vmware.com</a>>
Date: Wed, 21 Dec 2016 04:43:07 +0100
Subject: [PATCH] draw: use SoA fetch, not AoS one

Now that there's some SoA fetch which never falls back, we should always get
results which are better or at least not worse (something like rgba32f will
stay the same).

When reverse-applying this patch on Mesa 17.0.3 the regression on s390x gets
fixed.

Mesa has been compiled with

./configure --host=s390x-ibm-linux-gnu --build=s390x-ibm-linux-gnu
--program-prefix= --disable-dependency-tracking --prefix=/usr
--exec-prefix=/usr --bindir=/usr/bin --sbindir=/usr/sbin --sysconfdir=/etc
--datadir=/usr/share --includedir=/usr/include --libdir=/usr/lib64
--libexecdir=/usr/lib --localstatedir=/var --sharedstatedir=/usr/com
--mandir=/usr/share/man --infodir=/usr/share/info --disable-dependency-tracking
--enable-gles1 --enable-gles2 --enable-dri --with-egl-platforms=x11,drm
--enable-shared-glapi --enable-texture-float --enable-osmesa --enable-dri3
--enable-shader-cache --enable-gbm --enable-glx-tls
--with-dri-searchpath=/usr/lib64/dri --enable-gallium-llvm
--enable-llvm-shared-libs --enable-vdpau --enable-va --enable-xvmc
--with-dri-drivers=swrast --with-gallium-drivers=swrast
'CFLAGS=-fmessage-length=0 -grecord-gcc-switches -fstack-protector -O2 -Wall
-D_FORTIFY_SOURCE=2 -funwind-tables -fasynchronous-unwind-tables -g -DNDEBUG'</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>