<?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>Fri, 02 Oct 2026 16:11:45 +0000</lastBuildDate>
    <item>
      <title>CVE-2026-64590 — dma-buf/udmabuf: skip redundant cpu sync to fix cacheline EEXIST warning</title>
      <link>https://vulnerability.circl.lu/vuln/cve-2026-64590</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;dma-buf/udmabuf: skip redundant cpu sync to fix cacheline EEXIST warning&lt;/p&gt;
&lt;p&gt;When CONFIG_DMA_API_DEBUG_SG is enabled, importing a udmabuf into a DRM
driver (e.g. amdgpu for video playback in GNOME Videos / Showtime)
triggers a spurious warning:&lt;/p&gt;
&lt;p&gt;DMA-API: amdgpu 0000:03:00.0: cacheline tracking EEXIST, \
      overlapping mappings aren&amp;#39;t supported
  WARNING: kernel/dma/debug.c:619 at add_dma_entry+0x473/0x5f0&lt;/p&gt;
&lt;p&gt;The call chain is:&lt;/p&gt;
&lt;p&gt;amdgpu_cs_ioctl
   -&amp;gt; amdgpu_ttm_backend_bind
    -&amp;gt; dma_buf_map_attachment
     -&amp;gt; [udmabuf] map_udmabuf -&amp;gt; get_sg_table
      -&amp;gt; dma_map_sgtable(dev, sg, direction, 0)  // attrs=0
       -&amp;gt; debug_dma_map_sg -&amp;gt; add_dma_entry -&amp;gt; EEXIST&lt;/p&gt;
&lt;p&gt;This happens because udmabuf builds a per-page scatter-gather list via
sg_set_folio().  When begin_cpu_udmabuf() has already created an sg
table mapped for the misc device, and an importer such as amdgpu maps
the same pages for its own device via map_udmabuf(), the DMA debug
infrastructure sees two active mappings whose physical addresses share
cacheline boundaries and warns about the overlap.&lt;/p&gt;
&lt;p&gt;The DMA_ATTR_SKIP_CPU_SYNC flag suppresses this check in
add_dma_entry() because it signals that no CPU cache maintenance is
performed at map/unmap time, making the cacheline overlap harmless.&lt;/p&gt;
&lt;p&gt;All other major dma-buf exporters already pass this flag:
  - drm_gem_map_dma_buf() passes DMA_ATTR_SKIP_CPU_SYNC
  - amdgpu_dma_buf_map() passes DMA_ATTR_SKIP_CPU_SYN…&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;dma-buf/udmabuf: skip redundant cpu sync to fix cacheline EEXIST warning&lt;/p&gt;
&lt;p&gt;When CONFIG_DMA_API_DEBUG_SG is enabled, importing a udmabuf into a DRM
driver (e.g. amdgpu for video playback in GNOME Videos / Showtime)
triggers a spurious warning:&lt;/p&gt;
&lt;p&gt;DMA-API: amdgpu 0000:03:00.0: cacheline tracking EEXIST, \
      overlapping mappings aren&amp;#39;t supported
  WARNING: kernel/dma/debug.c:619 at add_dma_entry+0x473/0x5f0&lt;/p&gt;
&lt;p&gt;The call chain is:&lt;/p&gt;
&lt;p&gt;amdgpu_cs_ioctl
   -&amp;gt; amdgpu_ttm_backend_bind
    -&amp;gt; dma_buf_map_attachment
     -&amp;gt; [udmabuf] map_udmabuf -&amp;gt; get_sg_table
      -&amp;gt; dma_map_sgtable(dev, sg, direction, 0)  // attrs=0
       -&amp;gt; debug_dma_map_sg -&amp;gt; add_dma_entry -&amp;gt; EEXIST&lt;/p&gt;
&lt;p&gt;This happens because udmabuf builds a per-page scatter-gather list via
sg_set_folio().  When begin_cpu_udmabuf() has already created an sg
table mapped for the misc device, and an importer such as amdgpu maps
the same pages for its own device via map_udmabuf(), the DMA debug
infrastructure sees two active mappings whose physical addresses share
cacheline boundaries and warns about the overlap.&lt;/p&gt;
&lt;p&gt;The DMA_ATTR_SKIP_CPU_SYNC flag suppresses this check in
add_dma_entry() because it signals that no CPU cache maintenance is
performed at map/unmap time, making the cacheline overlap harmless.&lt;/p&gt;
&lt;p&gt;All other major dma-buf exporters already pass this flag:
  - drm_gem_map_dma_buf() passes DMA_ATTR_SKIP_CPU_SYNC
  - amdgpu_dma_buf_map() passes DMA_ATTR_SKIP_CPU_SYN…&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/cve-2026-64590</guid>
    </item>
  </channel>
</rss>
