<table><tr><td style="">rjvbb added a comment.
</td><a style="text-decoration: none; padding: 4px 8px; margin: 0 8px 8px; float: right; color: #464C5C; font-weight: bold; border-radius: 3px; background-color: #F7F7F9; background-image: linear-gradient(to bottom,#fff,#f1f0f1); display: inline-block; border: 1px solid rgba(71,87,120,.2);" href="https://phabricator.kde.org/D6376" rel="noreferrer">View Revision</a></tr></table><br /><div><div><blockquote style="border-left: 3px solid #8C98B8;
color: #6B748C;
font-style: italic;
margin: 4px 0 12px 0;
padding: 8px 12px;
background-color: #F8F9FC;">
<div style="font-style: normal;
padding-bottom: 4px;">In <a href="https://phabricator.kde.org/D6376#119376" style="background-color: #e7e7e7;
border-color: #e7e7e7;
border-radius: 3px;
padding: 0 4px;
font-weight: bold;
color: black;text-decoration: none;" rel="noreferrer">D6376#119376</a>, <a href="https://phabricator.kde.org/p/cullmann/" style="
border-color: #f1f7ff;
color: #19558d;
background-color: #f1f7ff;
border: 1px solid transparent;
border-radius: 3px;
font-weight: bold;
padding: 0 4px;" rel="noreferrer">@cullmann</a> wrote:</div>
<div style="margin: 0;
padding: 0;
border: 0;
color: rgb(107, 116, 140);"><p>Next try ;=)<br />
Question is, can really such a race happen, that finishNotification triggers duplicate insertion?</p></div>
</blockquote>
<p>Do you see any other explanation why a crash could occur under qDeleteAll, one that is due to a bug in KNotifications and not in Phonon itself?</p>
<p>I'd say the source for a duplicate insertion (and thus the location of the race condition) would be a concurrent execution of <tt style="background: #ebebeb; font-size: 13px;">m_reusablePhonons.takeFirst()</tt>, where more than 1 thread gets hold of the same item.</p>
<blockquote style="border-left: 3px solid #a7b5bf; color: #464c5c; font-style: italic; margin: 4px 0 12px 0; padding: 4px 12px; background-color: #f8f9fc;"><p>At least with a hash set, this won't lead to later issues.</p></blockquote>
<p>Why not use a barrier or lock to ensure that the <tt style="background: #ebebeb; font-size: 13px;">takeFirst()</tt> requests are executed one by one?</p>
<p>FWIW, a race condition wouldn't only lead to duplicate entries in the list (something I'd hope Qt knows how to handle) but also to more than 1 thread using the same Phonon object at the same time. Who's to tell what state that object will be in?</p></div></div><br /><div><strong>REPOSITORY</strong><div><div>R289 KNotifications</div></div></div><br /><div><strong>REVISION DETAIL</strong><div><a href="https://phabricator.kde.org/D6376" rel="noreferrer">https://phabricator.kde.org/D6376</a></div></div><br /><div><strong>To: </strong>cullmann, Frameworks<br /><strong>Cc: </strong>mpyne, rjvbb<br /></div>