Skip to content

CVE-2026-63950

Is CVE-2026-63950 real, exploitable, or a false positive? Here's the community verdict.

signals

public sources

Exploited in wild
Not listed
CISA KEV
Public exploit
None known
Metasploit/EDB/PoC
Base severity
7.8 High
CVSS
Exploitation prob.
0.2%
FIRST EPSS
Weakness
Not classified
CWE

High CVSS base score, but low real-world exploitation probability (EPSS). Likely less urgent than the score implies.

baseline read

auto · not a community verdict

Real, but low real-world risk

A genuine vulnerability on paper, but EPSS shows little real-world exploitation — the base score may overstate urgency. This is not the same as a false positive.

Based on CVSS · FIRST EPSS

Confirm or dispute →

CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H

In the Linux kernel, the following vulnerability has been resolved: mm/rmap: initialize nr_pages to 1 at loop start in try_to_unmap_one Initialize nr_pages to 1 at the start of each loop iteration, like folio_referenced_one() does. Without this, nr_pages computed by a previous folio_unmap_pte_batch() call can be reused on a later iteration that does not run folio_unmap_pte_batch() again. mmap a 64K large folio with MAP_ANONYMOUS | MAP_DROPPABLE, then call madvise(MADV_FREE), then make the last page device-exclusive via HMM_DMIRROR_EXCLUSIVE. Trigger node reclaim through sysfs. Now, in try_to_unmap_one(), we will first clear the first 15 out of 16 entries mapping the lazyfree folio. This will set nr_pages to 15. In the next pvmw walk, this nr_pages gets reused on a device-exclusive pte, thus potentially corrupting folio refcount/mapcount. At the moment, I have a userspace program which can make the kernel spit out a trace, but the blow up is in folio_referenced_one(), because there are existing bugs in the interaction between device-private and rmap (which too I am investigating). I did a one liner kernel change to avoid going into folio_referenced_one(), and the kernel blows up at folio_remove_rmap_ptes in try_to_unmap_one which is what I wanted. Note that the bug is there not since file folio batching but lazyfree folio batching, since device-exclusive only works for anonymous folios. Userspace visible effect is simply kernel crashing somewhere due to refcount/mapcount corruption.

Published

Embed this verdict
TruePositive verdict for CVE-2026-63950
Markdown
[![TruePositive verdict](https://www.truepositive.app/cve/CVE-2026-63950/badge.svg)](https://www.truepositive.app/cve/CVE-2026-63950)
HTML
<a href="https://www.truepositive.app/cve/CVE-2026-63950"><img src="https://www.truepositive.app/cve/CVE-2026-63950/badge.svg" alt="TruePositive verdict for CVE-2026-63950"></a>

Live badge that updates automatically as the community verdict changes.

Community ground truth

Be the first practitioner to weigh in

So far this is only TruePositive's editorial baseline from public sources. Add your real-world verdict below — it becomes the signal the next person triaging this relies on.

🥇 The first 50 practitioners to contribute earn a Founding Contributor badge.

In your experience, is this finding real and exploitable?

awaiting field verdicts
Real, but not a risk here
Not a real issue

Curated baseline: A curated baseline from public sources, shown separately from community verdicts.

No account needed. Anonymous verdicts post as an unverified signal. Log in to make yours verified and earn reputation.

Field notes & remediation

Verdicts are the quick signal. Notes are the evidence and fixes behind them.

  • 0
    Field note · TruePositive EditorialCurated

    This scores high on paper (CVSS 7.8) but the real-world signal is low. It is not in CISA KEV and no public exploit was found, and near-term exploitation looks unlikely (EPSS about 0%). Verify your own exposure before treating it as urgent. It needs local access, so it reads more as a privilege-escalation or post-access risk than a remote one. Not an emergency for most teams, but patch it in your normal cycle and check whether the affected component is actually exposed in your setup.

Add a field note or remediationoptional
Note type

What are you adding?

Markdown supported · minimum 20 characters.