<table><tr><td style="">davidedmundson 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/D25185">View Revision</a></tr></table><br /><div><div><p>Certainly doing all this inside updatePaintNode was wrong. So, concept wise ++</p>
<p>I don't understand the mutex, the main point of the patch is so that we only do network stuff on the main thread. During updatePaintNode the GUI thread is explicitly locked.</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>This patch changes it so the QNAM bits happen upon setSource</p></blockquote>
<p>The other option is to do all the resolving of the final QImage inside ::updatePolish()<br />
You get all the advantages of it happening only once per frame and all the advantages of keeping everything simple in the main thread.</p>
<p>It's what we do in Plasma::IconItem Plasma::SvgItem and since we ported to that it made everything cleaner, faster and safer.</p></div></div><br /><div><strong>REPOSITORY</strong><div><div>R169 Kirigami</div></div></div><br /><div><strong>REVISION DETAIL</strong><div><a href="https://phabricator.kde.org/D25185">https://phabricator.kde.org/D25185</a></div></div><br /><div><strong>To: </strong>apol, Kirigami, mart<br /><strong>Cc: </strong>davidedmundson, anthonyfieroni, bcooksley, plasma-devel, fbampaloukas, GB_2, domson, dkardarakos, ngraham, apol, ahiemstra, mart, hein<br /></div>