<?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://db.gcve.eu</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>Fri, 20 Nov 2026 21:22:01 +0000</lastBuildDate>
    <item>
      <title>CVE-2026-99621 — drm/amdgpu: unlock the VM and BO when attaching a KFD BO fails</title>
      <link>https://db.gcve.eu/vuln/cve-2026-99621</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: unlock the VM and BO when attaching a KFD BO fails&lt;/p&gt;
&lt;p&gt;kfd_mem_attach() locks the page directory and the BO with a drm_exec
transaction on its stack, but leaves through the unwind label without
drm_exec_fini() when locking fails, e.g. when MAP_MEMORY_TO_GPU gets a
signal while the page directory or the BO is busy. Reservations already
acquired then remain locked with their ww_acquire_ctx pointing into the
expired stack frame.&lt;/p&gt;
&lt;p&gt;With one thread doing GEM_VA map/unmap on the KFD VM and
another one mapping KFD BOs while getting SIGUSR1 every 20 us:&lt;/p&gt;
&lt;p&gt;BUG: KASAN: stack-out-of-bounds in __ww_mutex_lock.constprop.0+0x34f5/0x3da0
   ww_mutex_lock_interruptible+0x2f/0x150
   drm_exec_lock_obj+0x107/0x2d0 [drm_exec]
   kfd_mem_attach+0x4dd/0x1b60 [amdgpu]&lt;/p&gt;
&lt;p&gt;and the process ends up in D state.&lt;/p&gt;
&lt;p&gt;Finish the transaction before unwinding.&lt;/p&gt;
&lt;p&gt;(cherry picked from commit 7b885fa27e4b9fa5cbdc64e66fc35872aee296f5)&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: unlock the VM and BO when attaching a KFD BO fails&lt;/p&gt;
&lt;p&gt;kfd_mem_attach() locks the page directory and the BO with a drm_exec
transaction on its stack, but leaves through the unwind label without
drm_exec_fini() when locking fails, e.g. when MAP_MEMORY_TO_GPU gets a
signal while the page directory or the BO is busy. Reservations already
acquired then remain locked with their ww_acquire_ctx pointing into the
expired stack frame.&lt;/p&gt;
&lt;p&gt;With one thread doing GEM_VA map/unmap on the KFD VM and
another one mapping KFD BOs while getting SIGUSR1 every 20 us:&lt;/p&gt;
&lt;p&gt;BUG: KASAN: stack-out-of-bounds in __ww_mutex_lock.constprop.0+0x34f5/0x3da0
   ww_mutex_lock_interruptible+0x2f/0x150
   drm_exec_lock_obj+0x107/0x2d0 [drm_exec]
   kfd_mem_attach+0x4dd/0x1b60 [amdgpu]&lt;/p&gt;
&lt;p&gt;and the process ends up in D state.&lt;/p&gt;
&lt;p&gt;Finish the transaction before unwinding.&lt;/p&gt;
&lt;p&gt;(cherry picked from commit 7b885fa27e4b9fa5cbdc64e66fc35872aee296f5)&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://db.gcve.eu/vuln/cve-2026-99621</guid>
    </item>
  </channel>
</rss>
