<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugs.webkit.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.4.1"
          urlbase="https://bugs.webkit.org/"
          
          maintainer="admin@webkit.org"
>

    <bug>
          <bug_id>302808</bug_id>
          
          <creation_ts>2025-11-19 11:21:01 -0800</creation_ts>
          <short_desc>[libpas] Implement Previous-Tag-Exclusion for MTE allocations</short_desc>
          <delta_ts>2026-04-07 11:08:49 -0700</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>1</classification_id>
          <classification>Unclassified</classification>
          <product>WebKit</product>
          <component>New Bugs</component>
          <version>WebKit Nightly Build</version>
          <rep_platform>Unspecified</rep_platform>
          <op_sys>Unspecified</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>FIXED</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard></status_whiteboard>
          <keywords>InRadar</keywords>
          <priority>P2</priority>
          <bug_severity>Normal</bug_severity>
          <target_milestone>---</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Marcus Plutowski">marcus_plutowski</reporter>
          <assigned_to name="Nobody">webkit-unassigned</assigned_to>
          <cc>webkit-bug-importer</cc>
          

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>2159998</commentid>
    <comment_count>0</comment_count>
    <who name="Marcus Plutowski">marcus_plutowski</who>
    <bug_when>2025-11-19 11:21:01 -0800</bug_when>
    <thetext>rdar://152167632

In a similar vein to Adjacent-Tag-Exclusion, we can impair the ability of attackers to &apos;get lucky&apos; with MTE tags by ensuring that &apos;local&apos; (for ATE spatially local, for PTE temporally local) allocations have different tags. In this case that means that when we retag an object in a slot that previously had some tag A, we make sure to use a tag other than A. This does not provide as hard of a guarantee as ATE does, as ATE deterministically ensures that adjacent allocations will not have the same tag -- whereas in theory heap grooming could be used to re-reallocate on a given slot, thus dodging the prior tag. This is however still an added layer of difficulty and worth considering as a hardening option.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>2160023</commentid>
    <comment_count>1</comment_count>
    <who name="Marcus Plutowski">marcus_plutowski</who>
    <bug_when>2025-11-19 12:09:17 -0800</bug_when>
    <thetext>Pull request: https://github.com/WebKit/WebKit/pull/54200</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>2196026</commentid>
    <comment_count>2</comment_count>
    <who name="Marcus Plutowski">marcus_plutowski</who>
    <bug_when>2026-04-01 15:01:23 -0700</bug_when>
    <thetext>Pull request: https://github.com/WebKit/WebKit/pull/61853</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>2197698</commentid>
    <comment_count>3</comment_count>
    <who name="EWS">ews-feeder</who>
    <bug_when>2026-04-07 11:08:47 -0700</bug_when>
    <thetext>Committed 310726@main (569d28c26a9b): &lt;https://commits.webkit.org/310726@main&gt;

Reviewed commits have been landed. Closing PR #61853 and removing active labels.</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>