CVE-2026-72485 is a memory corruption in the Linux kernel Coresight subsystem where nr_outconns is incremented before devm_krealloc_array() succeeds, but coresight_release_platform_data() unconditionally dereferences entries up to the counter value on cleanup—meaning if allocation fails, the release path walks off the end of a corrupted array and panics. The identical bug appears in both coresight_add_out_conn() and coresight_add_in_conn(), and the fix in both cases is trivial: move the counter increment after the allocation check. But the trivial fix obscures a deeper structural failure that defenders need to internalize. The duplication tells you this wasn't careless line-by-line coding—it was a mental model error. The developer assumed 'claim your slot early' was good practice, while the cleanup code assumed 'counter equals valid pointer count' was always true. These are opposite assumptions living in the same codebase, and nobody caught the mismatch because allocation logic and cleanup assumptions are reviewed separately. What you should do: First, verify the patch is applied (the increment moves from before the allocation to after the NULL check). Second, audit other Coresight connection management functions for the same pattern—if there's a nr_inconns or similar counter, check its ordering. Third, and this is the part most defenders skip: examine coresight_release_platform_data() itself. The unconditional iteration for (i = 0; i < nr_outconns; i++) is a landmine independent of this CVE. Any future code that corrupts the counter—through this bug, through copy-paste, through drift—triggers the same panic. The proper fix there is either bounds-checking before dereference or switching to an iteration pattern that only touches non-NULL pointers. Finally, recognize this as a class of bug, not an instance. The pattern—counter incremented before allocation, followed by unconditional dereference on cleanup—has appeared in the kernel at least a dozen times under different variable names (nr_pages, count, nr_entries). Each gets its own CVE, each gets its own correct-but-incomplete patch. The systemic exposure isn't just this one vulnerability; it's that the cleanup path itself encodes an invariant ('counter equals valid entries') that nobody states, validates, or checks. Until that invariant becomes machine-checkable, this class will recur. Check your subsystems for the same implicit contract in release paths.
CVE-2026-72485
Official description Straight from the sourceThe vendor's or NVD's own wording, published unedited. Authoritative, but often terse — it says what broke, rarely what to do.
NVD · uneditedIn the Linux kernel, the following vulnerability has been resolved: coresight: platform: defer connection counter increment until alloc succeeds coresight_add_out_conn() increments nr_outconns before calling devm_krealloc_array() and again before devm_kmalloc(). If either allocation fails, the counter is already bumped while the corresponding array entry is NULL or uninitialized garbage. coresight_add_in_conn() has the same problem with nr_inconns and devm_krealloc_array(). In both cases the probe returns -ENOMEM, which causes coresight_get_platform_data() to call coresight_release_platform_data() for cleanup. That function iterates up to nr_outconns (or nr_inconns) entries and dereferences each pointer unconditionally, hitting the NULL or garbage entry and panicking instead of failing gracefully. Fix by moving the counter increments to after all allocations succeed, so the struct is always consistent on any error path.
Technical summary Written by usOur analysis, written from the advisory, the CVSS vector and the affected-version data. It adds context the advisory leaves out, and never invents facts that are not in the source.
dbcve analysisA detailed technical summary for this CVE is being prepared.
CVSS breakdown How the score is builtThe industry scoring standard. It rates how the flaw is reached, what it takes to exploit, and what an attacker gains — the score is derived from those, not the other way round.
From the vector- Attack vector
- Local
- Complexity
- Low
- Privileges
- Low
- User interaction
- None
- Scope
- Unchanged
- Confidentiality
- High
- Integrity
- High
- Availability
- High
CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Remediation Closing itWhat it takes to close this. Where a vendor fix exists we point at it; where none exists we say so plainly, and can build one. Effort estimates are scoped from the advisory, not from your codebase.
From vendor dataThere is no version to upgrade to and no patch to apply. Every affected install stays exposed until the vendor ships a fix — or somebody else builds one.
Free. We build fixes in the order the community asks for them — and we’ll tell you the moment this one lands.
We develop and verify an original fix where the vendor hasn’t, from $4,900. Deployed to your staging first — never straight to production.
Scope it with usSee what else the community needs solved on the solutions-needed board.
Scan for this in your stack
Free · runs locallyCheck whether your project pulls in CVE-2026-72485 — or any other known-vulnerable package — straight from your lock files. Free and open source; it runs locally and uploads nothing.
References Go to the primary sourcePrimary sources — vendor advisories, patches and trackers. Where our summary and a reference disagree, the reference wins.
Primary sourcesCVE-2026-72485 is a memory corruption in the Linux kernel Coresight subsystem where `nr_outconns` is incremented before `devm_krealloc_array()` succeeds, but `coresight_release_platform_data()` unconditionally dereferences entries up to the counter value on cleanup—meaning if allocation fails, the release path walks off the end of a corrupted array and panics. The identical bug appears in both `coresight_add_out_conn()` and `coresight_add_in_conn()`, and the fix in both cases is trivial: move the counter increment after the allocation check. But the trivial fix obscures a deeper structural failure that defenders need to internalize. The duplication tells you this wasn't careless line-by-line coding—it was a mental model error. The developer assumed 'claim your slot early' was good practice, while the cleanup code assumed 'counter equals valid pointer count' was always true. These are opposite assumptions living in the same codebase, and nobody caught the mismatch because allocation logic and cleanup assumptions are reviewed separately. What you should do: First, verify the patch is applied (the increment moves from before the allocation to after the NULL check). Second, audit other Coresight connection management functions for the same pattern—if there's a `nr_inconns` or similar counter, check its ordering. Third, and this is the part most defenders skip: examine `coresight_release_platform_data()` itself. The unconditional iteration `for (i = 0; i < nr_outconns; i++)` is a landmine independent of this CVE. Any future code that corrupts the counter—through this bug, through copy-paste, through drift—triggers the same panic. The proper fix there is either bounds-checking before dereference or switching to an iteration pattern that only touches non-NULL pointers. Finally, recognize this as a class of bug, not an instance. The pattern—counter incremented before allocation, followed by unconditional dereference on cleanup—has appeared in the kernel at least a dozen times under different variable names (`nr_pages`, `count`, `nr_entries`). Each gets its own CVE, each gets its own correct-but-incomplete patch. The systemic exposure isn't just this one vulnerability; it's that the cleanup path itself encodes an invariant ('counter equals valid entries') that nobody states, validates, or checks. Until that invariant becomes machine-checkable, this class will recur. Check your subsystems for the same implicit contract in release paths.
Practitioner notes
ContributedPeer-ranked notes from engineers who’ve handled CVE-2026-72485 in production — separate from our analysis above.
The advisory tells you what broke. It rarely tells you what actually worked. If you’ve dealt with this one, that detail is what the next engineer is searching for.
- The version that genuinely resolved it — not the one the vendor claimed
- A config change or rule that shut the vector down
- A gotcha in the upgrade path that cost you an afternoon
A place for practitioners to share what actually worked: a mitigation you’ve tested, a configuration change, a version- or environment-specific caveat, or a link to a verified patch. The most useful notes rise to the top as peers upvote them, so the signal stays high.
- Verified mitigations, workarounds, and config changes
- Version or environment caveats, and links to real fixes
- No weaponised exploit code, or anything meant to cause harm
- No spam, self-promotion, credentials, or personal data