<html>
    <head>
      <base href="https://bugs.freedesktop.org/">
    </head>
    <body>
      <p>
        <div>
            <b><a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - Wrong and unstable image rendering from GLSL fragment shaders"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=99919#c2">Comment # 2</a>
              on <a class="bz_bug_link 
          bz_status_NEW "
   title="NEW - Wrong and unstable image rendering from GLSL fragment shaders"
   href="https://bugs.freedesktop.org/show_bug.cgi?id=99919">bug 99919</a>
              from <span class="vcard"><a class="email" href="mailto:imirkin@alum.mit.edu" title="Ilia Mirkin <imirkin@alum.mit.edu>"> <span class="fn">Ilia Mirkin</span></a>
</span></b>
        <pre>Created <span class=""><a href="attachment.cgi?id=129905" name="attach_129905" title="apitrace of variant shader">attachment 129905</a> <a href="attachment.cgi?id=129905&action=edit" title="apitrace of variant shader">[details]</a></span>
apitrace of variant shader

For convenience, attaching a trace of the variant.

Very surprising that this breaks... the diff between shaders is

--- large-v100-webgl-9803840f5a05935c_inv_variant_77/original.frag     
2017-02-23 05:18:29.000000000 -0500
+++ large-v100-webgl-9803840f5a05935c_inv_variant_77/variant.frag      
2017-02-23 05:18:58.000000000 -0500
@@ -67,12 +70,18 @@
         }
     else
         {
+            for(int i = 0; i < 1; i++) {
             if((s2.x < s1.x || s1.x <= 0.0) && s2.x > 0.0)
                 {
                     hd.c = ca.y;
+                    if(injectionSwitch.x > injectionSwitch.y)
+                        {
+                            break;
+                        }
                     hd.d = s2.x;
                     hd.n = s2.yzw;
                 }
+            }
         }
     return hd;
 }

(and injectionSwitch is set to 0,1 so that should never be true)

Tested on a SKL GT2, so it's not just some weird SNB problem.</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>