<?xml version='1.0' encoding='UTF-8'?>
<?xml-stylesheet href="/static/style.xsl" type="text/xsl"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/" version="2.0">
  <channel>
    <title>Most recent entries from all</title>
    <link>https://vulnerability.circl.lu</link>
    <description>Contains only the most 10 recent entries.</description>
    <docs>http://www.rssboard.org/rss-specification</docs>
    <generator>python-feedgen</generator>
    <language>en</language>
    <lastBuildDate>Tue, 29 Sep 2026 17:56:20 +0000</lastBuildDate>
    <item>
      <title>CVE-2026-80539 — drm/amdgpu: disallow multiple FENCE chunks in one submit</title>
      <link>https://vulnerability.circl.lu/vuln/cve-2026-80539</link>
      <description>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; Linux&lt;/p&gt;
&lt;p&gt;In the Linux kernel, the following vulnerability has been resolved:&lt;/p&gt;
&lt;p&gt;drm/amdgpu: disallow multiple FENCE chunks in one submit&lt;/p&gt;
&lt;p&gt;amdgpu_cs_pass1() dispatches on chunk_id once per chunk without
rejecting repeated ids. p-&amp;gt;uf_bo is a single-slot field, so a
submission carrying two AMDGPU_CHUNK_ID_FENCE chunks runs
amdgpu_cs_p1_user_fence() twice, and the second run overwrites
p-&amp;gt;uf_bo with a freshly referenced BO without dropping the reference
taken by the first.&lt;/p&gt;
&lt;p&gt;amdgpu_cs_parser_fini() only unrefs the final p-&amp;gt;uf_bo, so every FENCE
chunk but the last leaks a BO reference. The leaked BO outlives handle
close and process exit.&lt;/p&gt;
&lt;p&gt;Reject duplicate FENCE chunks the same way commit fec5f8e8c6bc
(&amp;#34;drm/amdgpu: disallow multiple BO_HANDLES chunks in one submit&amp;#34;) did
for p-&amp;gt;bo_list.&lt;/p&gt;
&lt;p&gt;(cherry picked from commit 665b1fc2a1845206408f9a2c6da67101789edb82)&lt;/p&gt;</description>
      <content:encoded>&lt;p&gt;&lt;strong&gt;Affected:&lt;/strong&gt; Linux&lt;/p&gt;
&lt;p&gt;In the Linux kernel, the following vulnerability has been resolved:&lt;/p&gt;
&lt;p&gt;drm/amdgpu: disallow multiple FENCE chunks in one submit&lt;/p&gt;
&lt;p&gt;amdgpu_cs_pass1() dispatches on chunk_id once per chunk without
rejecting repeated ids. p-&amp;gt;uf_bo is a single-slot field, so a
submission carrying two AMDGPU_CHUNK_ID_FENCE chunks runs
amdgpu_cs_p1_user_fence() twice, and the second run overwrites
p-&amp;gt;uf_bo with a freshly referenced BO without dropping the reference
taken by the first.&lt;/p&gt;
&lt;p&gt;amdgpu_cs_parser_fini() only unrefs the final p-&amp;gt;uf_bo, so every FENCE
chunk but the last leaks a BO reference. The leaked BO outlives handle
close and process exit.&lt;/p&gt;
&lt;p&gt;Reject duplicate FENCE chunks the same way commit fec5f8e8c6bc
(&amp;#34;drm/amdgpu: disallow multiple BO_HANDLES chunks in one submit&amp;#34;) did
for p-&amp;gt;bo_list.&lt;/p&gt;
&lt;p&gt;(cherry picked from commit 665b1fc2a1845206408f9a2c6da67101789edb82)&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/cve-2026-80539</guid>
    </item>
  </channel>
</rss>
