<?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>Wed, 30 Sep 2026 17:55:17 +0000</lastBuildDate>
    <item>
      <title>CVE-2026-53172 — accel/ethosu: fix IFM region index out-of-bounds in command stream parser</title>
      <link>https://vulnerability.circl.lu/vuln/cve-2026-53172</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;accel/ethosu: fix IFM region index out-of-bounds in command stream parser&lt;/p&gt;
&lt;p&gt;NPU_SET_IFM_REGION extracts the region index with param &amp;amp; 0x7f, giving
a maximum value of 127. However region_size[] and output_region[] in
struct ethosu_validated_cmdstream_info are both sized to
NPU_BASEP_REGION_MAX (8), giving valid indices [0..7].&lt;/p&gt;
&lt;p&gt;Every other region assignment in the same switch uses param &amp;amp; 0x7:
  NPU_SET_OFM_REGION:  st.ofm.region  = param &amp;amp; 0x7;
  NPU_SET_IFM2_REGION: st.ifm2.region = param &amp;amp; 0x7;
  NPU_SET_WEIGHT_REGION: st.weight[0].region = param &amp;amp; 0x7;
  NPU_SET_SCALE_REGION:  st.scale[0].region  = param &amp;amp; 0x7;&lt;/p&gt;
&lt;p&gt;The 0x7f mask on IFM is inconsistent and appears to be a typo.&lt;/p&gt;
&lt;p&gt;feat_matrix_length() and calc_sizes() use the region index directly
as an array subscript into the kzalloc&amp;#39;d info struct:
  info-&amp;gt;region_size[fm-&amp;gt;region] = max(...);&lt;/p&gt;
&lt;p&gt;A userspace caller supplying NPU_SET_IFM_REGION with param &amp;gt; 7 causes
a write up to 127*8 = 1016 bytes past the start of region_size[],
corrupting adjacent kernel heap data.&lt;/p&gt;
&lt;p&gt;Fix by applying the same &amp;amp; 0x7 mask used by all other region
assignments.&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;accel/ethosu: fix IFM region index out-of-bounds in command stream parser&lt;/p&gt;
&lt;p&gt;NPU_SET_IFM_REGION extracts the region index with param &amp;amp; 0x7f, giving
a maximum value of 127. However region_size[] and output_region[] in
struct ethosu_validated_cmdstream_info are both sized to
NPU_BASEP_REGION_MAX (8), giving valid indices [0..7].&lt;/p&gt;
&lt;p&gt;Every other region assignment in the same switch uses param &amp;amp; 0x7:
  NPU_SET_OFM_REGION:  st.ofm.region  = param &amp;amp; 0x7;
  NPU_SET_IFM2_REGION: st.ifm2.region = param &amp;amp; 0x7;
  NPU_SET_WEIGHT_REGION: st.weight[0].region = param &amp;amp; 0x7;
  NPU_SET_SCALE_REGION:  st.scale[0].region  = param &amp;amp; 0x7;&lt;/p&gt;
&lt;p&gt;The 0x7f mask on IFM is inconsistent and appears to be a typo.&lt;/p&gt;
&lt;p&gt;feat_matrix_length() and calc_sizes() use the region index directly
as an array subscript into the kzalloc&amp;#39;d info struct:
  info-&amp;gt;region_size[fm-&amp;gt;region] = max(...);&lt;/p&gt;
&lt;p&gt;A userspace caller supplying NPU_SET_IFM_REGION with param &amp;gt; 7 causes
a write up to 127*8 = 1016 bytes past the start of region_size[],
corrupting adjacent kernel heap data.&lt;/p&gt;
&lt;p&gt;Fix by applying the same &amp;amp; 0x7 mask used by all other region
assignments.&lt;/p&gt;</content:encoded>
      <guid isPermaLink="false">https://vulnerability.circl.lu/vuln/cve-2026-53172</guid>
    </item>
  </channel>
</rss>
