Make fragment loop not depend on spin the event loop

Author: annevk

source

@@ -83699,8 +83699,9 @@ new PaymentRequest(…); // Allowed to use
 
   <p>Some of the sections below, to which the above algorithm defers in certain cases, require the
   user agent to <dfn>update the session history with the new page</dfn>. When a user agent is
-  required to do this, it must <span>queue a task</span> (associated with the <code>Document</code>
-  object of the <span>current entry</span>, not the new one) to run the following steps:</p>
+  required to do this, it must <span>queue a task</span> on the <span>networking task source</span>,
+  associated with the <code>Document</code> object of the <span>current entry</span> (not the new
+  one), to run the following steps:</p>
 
   <ol>
 
@@ -83791,24 +83792,32 @@ new PaymentRequest(&hellip;); // Allowed to use
    <li><p>The <span data-x="navigate">navigation algorithm</span> has now <dfn
    data-x="concept-navigate-mature">matured</dfn>.</p></li>
 
-   <li><p><i>Fragment loop</i>: <span>Spin the event loop</span> for a user-agent-defined amount of
-   time, as desired by the user agent implementer. (This is intended to allow the user agent to
-   optimize the user experience in the face of performance concerns.)</p></li>
+   <li><p><span>Try to scroll to the fragment</span> for the <code>Document</code>.</p></li>
+  </ol>
 
-   <li><p>If the <code>Document</code> object has no parser, or its parser has <span data-x="stop
-   parsing">stopped parsing</span>, or the user agent has reason to believe the user is no longer
-   interested in scrolling to the <span data-x="concept-url-fragment">fragment</span>, then
-   return.</p></li>
+  <p>To <dfn>try to scroll to the fragment</dfn> for a <code>Document</code> <var>document</var>,
+  perform the following steps <span>in parallel</span>:</p>
 
-   <li><p><span>Scroll to the fragment</span> given in the document's <span
-   data-x="concept-document-url">URL</span>. If this fails to find <span data-x="the indicated part
-   of the document">an indicated part of the document</span>, then return to the <i>fragment
-   loop</i> step.</p></li>
+  <ol>
+   <li><p>Wait for an implementation-defined amount of time. (This is intended to allow the user
+   agent to optimize the user experience in the face of performance concerns.)</p></li>
 
-  </ol>
+   <li>
+    <p><span>Queue a task</span> on the <span>networking task source</span> to run these steps:</p>
 
-  <p>The <span>task source</span> for this <span data-x="concept-task">task</span> is the
-  <span>networking task source</span>.</p>
+    <ol>
+     <li><p>If <var>document</var> has no parser, or its parser has <span data-x="stop
+     parsing">stopped parsing</span>, or the user agent has reason to believe the user is no longer
+     interested in scrolling to the <span data-x="concept-url-fragment">fragment</span>, then abort
+     these steps.</p></li>
+
+     <li><p><span>Scroll to the fragment</span> given in <var>document</var>'s <span
+     data-x="concept-document-url">URL</span>. If this does not find <span data-x="the indicated
+     part of the document">an indicated part of the document</span>, then <span>try to scroll to the
+     fragment</span> for <var>document</var>.</p></li>
+    </ol>
+   </li>
+  </ol>
 
 
   <h4 id="read-html"><dfn data-x="navigate-html">Page load processing model for HTML files</dfn></h4>